PDF file - Facultatea de Chimie şi Inginerie Chimică

R /(mol m -2 s -1 ) 

0.06 

0.05 

0.04 

0.03 

0.02 

0.01 

0.00 

-0.01 

5.66% C2H6-airmixture p0 =100 kPa; Tw =487K 

induction period 

kinetic control 

kinetic to diffusion transition 

0 2 4 6 8 10 

time/s 

diffusion control 

Concentration (ng/ml) 

CHEMIA 

14 

12 

10 

3/2009 

8 

6 

4 

2 

0 


10.0 

1.0 

0 4 8 12 

0.1 

Time (h) 

0 3 6 9 12 

Time (h)

ANUL LIV 2009 

S T U D I A 

UNIVERSITATIS BABEŞ-BOLYAI 

CHEMIA 

3 

Desktop Editing Office: 51 ST B.P. Hasdeu Street, Cluj-Napoca, Romania, Phone + 40 264-405352 

CUPRINS – CONTENT – SOMMAIRE – INHALT 

LIANA MUREŞAN, ALEXANDRA CSAVDÁRI, Professor Ioan Bâldea 

at his 70’s Anniversary ..............................................................................5 

MARCELA ACHIM, DANA MUNTEAN, LAURIAN VLASE, IOAN BÂLDEA, 

DAN MIHU, SORIN E. LEUCUŢA, New LC/MS/MS Method for the 

Quantification of Phenytoin in Human Plasma ...................................... 7 

ECATERINA BICA, LAURA ELENA MUREŞAN, LUCIAN BARBU- 

TUDORAN, EMIL INDREA, IONEL CĂTĂLIN POPESCU, 

ELISABETH-JEANNE POPOVICI, Studies on WO3 Thin Films 

Prepared by Dip-Coating Method ........................................................ 15 

ADRIAN-IONUŢ CADIŞ, ADRIAN RAUL TOMŞA, ECATERINA BICA, 

LUCIAN BARBU-TUDORAN, LUMINIŢA SILAGHI-DUMITRESCU, 

ELISABETH-JEANNE POPOVICI, Preparation and Characterization 

of Manganese Doped Zinc Sulphide Nanocrystalline Powders 

With Luminescent Properties............................................................... 23 

COSMIN CĂŢĂNAŞ, MIHAI MOGOŞ, DANIEL HORVAT, JAKAB 

ENDRE, ELEONORA MARIA RUS, IULIU OVIDIU MARIAN, 

Electrical Characteristics of a Biobattery with Staphylococcus Aureus ..... 31

ANA-MARIA CORMOS, JOZSEF GASPAR, ANAMARIA PADUREAN, 

Modeling and Simulation of Carbon Dioxide Absorption in 

Monoethanolamine in Packed Absorption Columns............................ 37 

EUGEN DARVASI, LADISLAU KÉKEDY-NAGY, Red Pepper Powder 

Color Measurement by Using an Integrating Sphere and Digital 

Image Processing................................................................................ 49 

VALENTINA R. DEJEU, BARABÁS RÉKA, POP ALEXANDRU, 

BOGYA ERZSÉBET SÁRA, PAUL-ŞERBAN AGACHI, Mathematical 

Modeling for the Crystallization Process of Hydroxyapatite 

Obtained by Precipitation in Aqueous Solution ................................... 61 

SILVIA LENUŢA DUNCA, MONICA KULCSAR, ANCA SILVESTRU, 

CRISTIAN SILVESTRU, COSTEL SÂRBU, Study of The 

Chromatographic Retention of Some New Organoselenium and 

Organotellurium Compounds Containing Intramolecular Interactions 

by HPTLC............................................................................................ 71 

NATHAN FLEURET, SEBASTIAN PAIC, GABRIELA NEMES, 

RALUCA SEPTELEAN, PETRONELA PETRAR, IOAN SILAGHI- 

DUMITRESCU, Lower Rim Silyl Substituted Calix[8]Arenes .............. 81 

OSSI HOROVITZ, MARIA TOMOAIA-COTIŞEL, CSABA RACZ, 

GHEORGHE TOMOAIA, LIVIU-DOREL BOBOŞ, AURORA 

MOCANU, The Interaction of Silver Nanoparticles with Lipoic Acid ....... 89 

FLORICA IMRE-LUCACI, SORIN-AUREL DORNEANU, PETRU ILEA, 

Optimisation of Copper Removal from Diluted Solutions .................... 97 

MELINDA-HAYDEE KOVACS, DUMITRU RISTOIU, SIDONIA VANCEA, 

LUMINITA SILAGHI-DUMITRESCU, Volatile Organic Disinfection 

by Products Determination in Distribution System from Cluj Napoca ... 107 

ANDRADA MĂICĂNEANU, COSMIN COTEŢ, VIRGINIA DANCIU, 

MARIA STANCA, Iron Doped Carbon Aerogel as Catalyst for 

Phenol Total Oxidation ...................................................................... 117 

ANDRADA MĂICĂNEANU, HOREA BEDELEAN, SILVIA BURCĂ , MARIA 

STANCA, Heavy Metal Ions Removal from Model Wastewaters 

Using Oraşul Nou (Transilvania, Romania) Bentonite Sample.......... 127 

CRISTINA MIHALI, GABRIELA OPREA, ELENA CICAL, PVC Matrx 

Ionic - Surfactant Selective Electrodes Based on the Ionic Pair 

Tetra Alkyl-Ammonium –Laurylsulphate............................................ 141 

DAN MIHU, LAURIAN VLASE, SILVIA IMRE, CARMEN M. MIHU, 

MARCELA ACHIM, DANIELA LUCIA MUNTEAN, New LC/MS 

Method for Determination of Progesterone in Human Plasma for 

Therapeutic Drug Monitoring in Pregnancy and Gynecological 

Disorders ........................................................................................... 151

ANCA PETER, MONICA BAIA, FELICIA TODERAS, MIHAELA LAZAR, 

LUCIAN BARBU TUDORAN, VIRGINIA DANCIU, Photo-Catalysts 

Based on Gold - Titania Composites................................................. 161 

TÍMEA PERNYESZI, KRISZTINA HONFI, BORBALA BOROS, KATALIN 

TÁLOS, FERENC KILÁR, CORNELIA MAJDIK, Biosorption of Phenol 

from Aqueous Solutions by Fungal Biomass of Phanerochaete 

Chrysosporium .................................................................................. 173 

ANDREI ROTARU, MIHAI GOŞA, EUGEN SEGAL, Isoconversional 

Linear Integral Kinetics of the Non-Isothermal Evaporation of 

4-[(4-Chlorobenzyl)Oxy]-4’-Trifluoromethyl-Azobenzene .................. 185 

OCTAVIAN STAICU, VALENTIN MUNTEANU, DUMITRU OANCEA, 

Overall Kinetics for the Catalytic Ignition of Ethane-Air Mixtures 

on Platinum........................................................................................ 193 

MARIA ŞTEFAN, IOAN BÂLDEA, RODICA GRECU , EMIL INDREA, 

ELISABETH-JEANNE POPOVICI, Growth and Characterisation of 

Zinc-Cadmium Sulphide Thin Films with Special Optical Properties .... 203 

MIHAELA-CLAUDIA TERTIŞ, FLORINA IONESCU, MARIA JITARU, 

Equilibrium Study on Adsorption Processes of 4-Nitrophenol and 

2, 6-Dinitrophenol Onto Granular Activated Carbon.......................... 213 

CAMELIA VARGA, MONICA MARIAN, ANCA PETER, DELIA BOLTEA, 

LEONARD MIHALY-COZMUTA, EUGEN NOUR, Strategies of 

Heavy Metal Uptake by Phaseolus Vulgaris Seeds Growing in 

Metalliferous and Non-Metalliferous Areas........................................ 223 

SIMONA VARVARA, MARIA POPA, LIANA MARIA MURESAN, 

Corrosion Inhibition of Bronze by Amino Acids in Aqueous Acidic 

Solutions............................................................................................ 235 

LIDIA VARVARI, SORIN-AUREL DORNEANU, IONEL CĂTĂLIN 

POPESCU, Potassium-Selective Electrode Based on a CALIX[6] 

Arenic Ester (C6ES6) ........................................................................ 247 

CODRUTA VARODI, DELIA GLIGOR, LEVENTE ABODI, LIANA 

MARIA MURESAN, Comparative Study of Carbon Paste 

Electrodes Modified with Methylene Blue and Methylene Green 

Adsorbed on Zeolite as Amperometric Sensors for H2O2 Detection ..... 255 

LAURIAN VLASE, DANA MUNTEAN, ADINA POPA, MARIA NEAG, 

IOAN BÂLDEA, MARCELA ACHIM, SORIN E. LEUCUŢA, 

Pharmacokinetic Interaction between Ivabradine and Ciprofloxacine 

in Healthy Volunteers ........................................................................ 265

Studia Universitatis Babes-Bolyai Chemia has been selected for coverage 

in Thomson Reuters products and custom information services. Beginning 

with V. 53 (1) 2008, this publication is indexed and abstracted in the following: 

• Science Citation Index Expanded (also known as SciSearch®) 

• Chemistry Citation Index® 

• Journal Citation Reports/Science Edition

Professor Ioan Bâldea at his 70’s Anniversary 

Brilliant teacher and warm colleague, 

Professor Ioan Bâldea was born on the 21 st 

of October 1939. Still active, he dedicates 

himself to science and teaching after 

graduating Babeş-Bolyai University as a 

chemist in 1962. Well-known in the academic 

community as one of the top kineticists 

of the country, his 70’s anniversary 

finds him with unaltered optimism and 

creativeness. 

Beginning immediately after graduation 

as a teaching assistant, his academic 

carrier unfolded rigorously step by step, while 

obtaining his PhD in 1969. As an appointed 

lecturer (1971), associated professor (1990) 

and full professor (1993), his work continuously 

contributed to the development and growth of the Department of Physical 

Chemistry. 

Besides over 120 various scientific papers, Professor Ioan Bâldea also 

published 4 books as a single author and contributed to 2 others. These 

summarize his prodigious experience in chemical kinetics of homogeneous 

redox and organic reactions, characterization of short-life chemical complexes, 

modeling of complex processes, design and reactor engineering as well as 

kinetic methods of analysis. This valuable and diverse knowledge was shared 

with the young scientists he tutored during their PhD scholarships. It also 

contributed both to solving of over 50 contracts and putting forward of 4 patents 

for the Romanian chemical industry. A production line for ethyl acetate was 

based on his designs and functioned till the early 90’s at the plant in 

Craiova. The 12 research Grants he coordinated (among these, one financed 

by the World Bank) substantially contributed to the development of Chemical 

Kinetics and General Physical Chemistry at the Babeş-Bolyai University. 

While always endeavoring on sharing and passing over his knowledge 

to the young generations, Professor Bâldea wrote or collaborated to the 

publication of 8 manuals of Physical as well as General Chemistry. The students 

highly appreciated the funny and nonconformist explanations; the examples 

illustrating daily life made his lectures colorful and easy to understand. The 

approximately 100 graduate and postgraduate thesis coordinated by the 

Professor stand proof for this fact.

6 

PROFESSOR IOAN BÂLDEA AT HIS 70’S ANNIVERSARY 

Also beloved and highly regarded by his colleagues, Professor Ioan 

Bâldea was elected for nearly 10 years as Head of the Department of Physical 

Chemistry and represented for 23 years the same Department in the Faculty’s 

Council. He was also the local coordinator of the CEEPUS academic exchange 

Program for 8 years. The vast experience allowed the Professor to manage the 

Program of didactic personnel improvement for 8 years and to be a referee 

for various scientific publications, grants and events. 

This issue of Studia Universitas Babeş-Bolyai, Seria Chemia reflects 

the tight professional relationships Professor Ioan Bâldea has build within 

the academic community of Romania. Teachers and researchers from the 

University of Bucharest, Baia-Mare or Iuliu HaŃieganu University of Medicine 

from Cluj-Napoca also contributed to this volume. 

Now, at his 70’s anniversary, we – the colleagues of the Department 

of Physical Chemistry – as well as all the colleagues and researchers of the 

Faculty of Chemistry of the Babeş-Bolyai University, along with the Editorial 

Board, wish Professor Ioan Bâldea health in the many years to come, as well 

as the spice of current valuable scientific activity. 

Volum Editors 

Liana Mureşan 

Alexandra Csavdári

STUDIA UNIVERSITATIS BABEŞ-BOLYAI, CHEMIA, LIV, 3, 2009 

NEW LC/MS/MS METHOD FOR THE QUANTIFICATION 

OF PHENYTOIN IN HUMAN PLASMA 

MARCELA ACHIM a , DANA MUNTEAN a , LAURIAN VLASE a , 

IOAN BÂLDEA b , DAN MIHU c , SORIN E. LEUCUŢA a 

ABSTRACT. A simple, reversed-phase high performance liquid chromatography 

method with mass spectrometric detection (HPLC-MS/MS) was developed 

for determination of an antiepileptic drug, phenytoin, in human plasma. The 

procedure involves a simple extraction step by mixing 0.2 ml plasma with 

0.6 ml methanol. After centrifugation, 1 µl of the supernatant was injected onto 

a Zorbax SB-C18 100 mm x 3 mm, 3.5 µm column, and eluted with a mobile 

phase consisting in a mixture of water containing 2 mM ammonium acetate 

and methanol 50:50 (v/v). Detection was in MRM mode, using an electrospray 

positive ionization. The ion transition monitored was 253.1→(182.1+225.1). The 

method was evaluated in terms of linearity (between 2.0 µg/ml to 80.0 µg/ml), 

accuracy, precision, recovery, sensitivity. The lower limit of quantification 

was established at 2.0 µg/ml. The simple extraction procedure and short 

chromatographic runtime make the method suitable for therapeutic drug 

monitoring studies. 

Keywords: phenytoin, HPLC-MS/MS, human plasma 

INTRODUCTION 

Phenytoin (Figure 1) is widely used in the treatment of epilepsy and is 

effective against all types of seisures. No drug has greater need for therapeutic 

drug concentration monitoring and individualized dosing than phenytoin. A good 

correlation usually is observed between the total concentration of phanytoin 

in plasma and the clinical effect. Therapeutic concentration of phenytoin is 

above 10 µg/ml [1]. Enzyme induction by phenytoin is well documented, even 

auto-induction by phenytoin should be considered during the treatment with 

phenytoin [2]. 

a 

University of Medicine and Pharmacy “Iuliu Haţieganu”, Faculty of Pharmacy, Emil Isac 13, 

RO-400023 Cluj-Napoca, Romania, dana@tbs.ubbcluj.ro 

b 

Babeş-Bolyai” University, Faculty of Chemistry and Chemical Engineering, Arany Janos 11, 

RO-400028 Cluj-Napoca, Romania 

c 

University of Medicine and Pharmacy “Iuliu Haţieganu”, Faculty of Medicine, Emil Isac 13, 

RO-400023 Cluj-Napoca, Romania

MARCELA ACHIM, DANA MUNTEAN, LAURIAN VLASE, IOAN BÂLDEA, DAN MIHU, SORIN E. LEUCUŢA 

Plasma concentration monitoring is widely used for the clinical management 

of epileptic patients receiving phenytoin [3]. To minimize toxicity, monitoring of 

plasma anticonvulsant levels is a part of the routine management of patients in 

many clinics. To the best of our knowledge, almost all of the methods that 

were applied for determination of antieplileptic drugs in biological media are 

often chromatography, electrophoresis and immunoassay techniques [4,5]. 

The aim of the present study was to develop a fast LC-MS/MS method, 

able to quantify phenytoin in human plasma after a simple sample preparation 

by protein precipitation. The proposed method proved to be accurate and despite 

of very simple sample preparation, showed high sensitivity. 

RESULTS AND DISCUSSION 

8 

C6H5 

C6H5 

O 

H 

N 

NH 

Figure 1. Molecular structure of phenytoin 

Figure 2 shows representative chromatograms of drug-free (blank) 

human plasma and a sample containing 2.0 µg/ml phenytoin (LOQ). No 

significant interference at the retention time of phenytoin (1.6 min) was 

observed, due to the specificity of the selected signal (Figure 3). 

Intens. 

6000 

4000 

2000 

0 

6000 

4000 

2000 

0 

CAL2__01.D: EIC 182.1; 225.1 ±MS2(253), Smoothed (0.3,1, GA) 

CAL2__02.D: EIC 182.1; 225.1 ±MS2(253), Smoothed (0.3,1, GA) 

0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Time [min] 

Figure 2. Chromatograms of a drug-free plasma sample (up) and LOQ plasma 

standard with 2.0 µg/ml phenytoin (down) 

O

NEW LC/MS/MS METHOD FOR THE QUANTIFICATION OF PHENYTOIN IN HUMAN PLASMA 

Intens. 

x105 

0.6 

0.4 

0.2 

0.0 

x104 

2.0 

1.5 

1.0 

0.5 

0.0 

182.1 

sp____02.d: +MS2(253.2), 540.8min (#2) 

225.1 

253.1 

sp____03.d: +MS2(253.2), 541.5min (#2) 

160 180 200 220 240 260 280 m/z 

Figure 3. Mass spectra of phenytoin. Full-scan spectrum (up) and 

MS/MS spectrum (down). The sum of ions with m/z 182.1 and 225.1 

was used for quantification 

The analyte detection was made in MRM mode, the ion transition 

monitored was m/z 253.1→(m/z 182.1 + m/z 225.1). This way, the detection is 

more sensitive than in case based only on ion m/z 253.1 (Figure 3). 

The analyte carryover was verified using a blank injection made right 

after an injection of the most elevated concentration level from calibration 

curve. No interference at retention time of analyte due to carryover was 

observed. 

The mean calibration curve y = a(±S.D.)x + b(±S.D.), with S.D. standard 

deviation, was: y = (59590.4±1971.9)x – (10490.5±2501.6), N = 8 calibration 

points, n = 5 determinations for each calibration point. The residuals had no 

tendency of variation with concentration. The applied calibration curve model 

proved to be accurate over the concentration range 2 – 80 µg/ml, with a 

correlation coefficient grater than 0.998 (Figure 4). 

The intra-day precision was determined from replicate analysis of 

samples containing phanytoin at four different concentrations covering the low, 

medium and higher ranges of calibration curve (Table 1), in good agreement to 

international regulations regarding bioanalytical methods validation [6-8]. 

The intra-day precision and accuracy ranged from 1.9% to 9.5% and 3.7% 

to 10% respectively. 

9


10 

Figure 4. Calibration curve for phenytoin 

Table 1. Intra-day precision, accuracy and recovery for phenytoin (n = 5) 

cnominal 

(µg/ml) 

Mean cfound 

(µg/ml) (± S.D.) 

C.V. % Bias % 

Recovery % 

(± S.D.) 

2.0 2.07 (0.2) 9.5 3.7 106.5 (10.5) 

5.0 5.50 (0.10) 1.9 10.0 100.3 (1.9) 

32.0 34.87 (1.94) 5.6 9.0 102.4 (5.7) 

64.0 67.60 (3.74) 5.5 5.6 101.1 (5.6) 

The intra-day precision and accuracy were determined by analyzing 

in five different days samples that have the same concentration, at lower, 

medium and higher levels from calibration curve. The precision ranged from 

3.5% to 10.1% and accuracy from 1.5% to 7.2 % (Table 2): 

Table 2. Inter-day precision, accuracy and recovery for phenytoin (n = 5) 

cnominal 

(µg/ml) 

Mean cfound 

(µg/ml) (± S.D.) 

C.V. % Bias % 

Recovery % 

(± S.D.) 

2.0 2.03 (0.20) 10.1 1.5 121.1 (35.3) 

5.0 5.36 (0.21) 4.0 7.2 97.3 (9.1) 

32.0 33.10 (1.16) 3.5 3.4 101.4 (5.6) 

64.0 66.32 (2.87) 4.3 3.6 99.1 (4.6)


Under the experimental conditions used, the lower limit of quantification 

(LOQ) was of 2 µg/ml phenytoin. LOQ is the lowest amount of analyte 

which can be measured with accuracy and precision less than 20%. 

CONCLUSIONS 

A simple, sensitive, accurate and precise HPLC/MS/MS method for 

determination of phenytoin in human plasma using a simple single-step 

extraction procedure is reported. Another advantage of the method is the 

short chromatographic runtime of only 2.1 min. The method is suitable for 

therapeutic drug monitoring studies and can also be used for pharmacokinetic 

studies conducted on healthy volunteers [9-13]. 

EXPERIMENTAL SECTION 

Reagents 

Phenytoin, methanol, ammonium acetate, were purchased from Merck 

(Merck KgaA, Darmstadt, Germany). Solvents used were HPLC grade and 

all other chemicals were of analytical grade. Distilled, deionised water was 

produced by a Direct Q-5 Millipore (Millipore SA, Molsheim, France) water 

system. The drug-free human plasma was supplied by the Local Bleeding 

Centre Cluj-Napoca, Romania. 

Preparation of standard solutions 

A stock solution containing 10 mg/ml phenytoin was prepared in 

methanol. A working solution of 200 µg/ml was prepared by diluting the 

appropriate volume of stock solution with plasma. Than this was used to 

spike different volumes of drug-free plasma, providing finally eight plasma 

standards with concentration between 2.0 and 80.0 µg/ml. Quality control 

samples (QC) of 5.0, 32.0, and 64.0 µg/ml were prepared by diluting specific 

volumes of working solution with plasma and were used to evaluate precision 

and accuracy of the method. 

Chromatographic and mass spectrometry systems and conditions 

The HPLC system was an 1100 series model (Agilent Technologies) 

consisted in a binary pump, an in-line degasser, an autosampler, a column 

thermostat and an Ion Trap VL mass spectrometer detector (Bruckner Daltonics 

GmbH, Germany). Chromatograms were processed using QuantAnalysis 

Software. The detection of phenytoin was in MRM (MS/MS) mode, using an 

electrospray positive ionization (ESI positive). The ion transitions monitored 

was: m/z 253.1→(m/z 182.1 + m/z 225.1). Chromatographic separation was 

performed at 45ºC on a Zorbax SB-C18 100 mm x 3 mm, 3.5 µm column 

(Agilent Technologies), protected by an inline filter. 

11


Mobile phase 

The mobile phase consisted in a mixture of water containing 2 mM 

ammonium acetate and methanol (50:50 v/v). It was always freshly prepared 

and was degassed before elution for 10 min in am Elma Transsonic 700/H 

(Singen, Germany) ultrasonic bath. The pump delivered the mobile phase at a 

flow rate of 1 ml/min. 

Sample preparation 

Plasma samples were prepared as follows in order to be 

chromatographically analyzed: in an Eppendorf tube (max 1.5 ml), 0.2 ml 

plasma and 0.6 ml methanol were added. The tube was vortex-mixed for 10 s 

(Vortex Genie 2, Scientific Industries) and centrifuged for 6 min at 5000 rpm 

(2-16 Sartorius centrifuge, Osterode am Harz, Germany). The supernatant was 

transferred to an autosampler vial and 1 µl was injected into the HPLC system. 

Validation 

As a first step of method validation [6-8], specificity was verified 

using six different plasma blanks obtained from healthy volunteers who had 

not previously taken any medication. 

The concentration of the analyte was determined automatically by the 

instrument data system. The calibration curve model was y = ax + b, weight 

1/y linear response, where y-peak area and x-concentration. Distribution of the 

residuals (% difference of the back-calculated concentration from the nominal 

concentration) was investigated. The calibration model was accepted, if the 

residuals were within ±20% at the lower limit of quantification (LOQ) and 

within ±15% at all other calibration levels and at least 2/3 of the standards 

meet this criterion, including highest and lowest calibration levels. 

The intra-day and inter-day precision (expressed as coefficient of 

variation, CV%) and accuracy (expressed as relative difference between 

obtained and theoretical concentration, bias%) of the assay procedure were 

determined by analysing on the same day five different samples at each of 

the lower (5.0 µg/ml), medium (32.0 µg/ml) and higher (64.0 µg/ml) levels of 

the considered concentration range and one different sample of each at five 

different occasions, respectively. 

The recovery of phenytoin was analyzed at each of the three 

concentration levels mentioned above, e.g. lower, medium and higher level, 

and also at the quantification limit, by comparing the peack area response 

of spiked plasma samples with the response of standards prepared in water 

with the same concentration of ivabradine as the plasma samples, all these 

prepared as stated in section “Sample preparation”. 

ACKNOWLEDGMENTS 

This work was supported by Grant CEEX-ET code 121, contract no. 

5860/2006, financed by CNCSIS Romania. 

12


REFERENCES 

1. Z. Rezaei, B. Hemmateenejab, S. Khabnadideh, M. Gorgin, Talanta, 2005, 65, 21. 

2. M. Cheety, R. Miller, M. A. Seymour, Ther. Drug Monit., 1998, 20, 60. 

3. K. M. Patil, S. L. Bodhankar, J. Pharm. Biomed. Analysis, 2005, 39, 181. 

4. D. J. Speed, S. J. Dickson, E. R. Cairus, N. D. Kim, J. Anal. Toxicol., 2000, 24, 685. 

5. M. E. Queiroz, S. M. Silvia, D. Carralho, F. M. Lancas, J. Chroma. Sci., 2002, 

40, 219. 

6. The European Agency for the Evaluation of Medicinal Products. Note for Guidance 

on the Investigation of Bioavailability and Bioequivalence, London, UK, 2001 

(CPMP/EWP/QWP/1401/98). 

7. U. S. Department of Health and Human Services, Food and Drug Administration, 

Center for Drug Evaluation and Research. Guidance for Industry. Bioavailability 

and Bioequivalence Studies for Orally Administrated Drug Products – General 

Considerations, Rockville, USA, 2003, http://www.fda.gov/cder/guidance/index.htm. 


Guidance for Industry – Bioanalytical Method Validation, 2001. 

9. L. Vlase L, S. E. Leucuta, S. Imre, Talanta, 2008, 75, 1104. 

10. L. Vlase, A. Leucuta, D. Farcau D, M. Nanulescu, Biopharma. Drug Dispos., 2006, 

27, 285. 

11. L. Vlase, S. Imre, D. Muntean, S. E. Leucuta, J.Pharma. Biomed. Analysis, 2007, 

44(3), 652. 

12. L. Vlase, D. Muntean, S. E. Leucuta, I. Baldea, Studia Univ. Babes-Bolyai Chemia, 

2009, 54, 43. 

13. A. Butnariu, D. S. Popa, L. Vlase, M. Andreica, D. Muntean, S. E. Leucuta, Revista 

Romana De Medicina De Laborator, 2009, 15, 7. 

13


STUDIES ON WO3 THIN FILMS PREPARED BY 

DIP-COATING METHOD 

ECATERINA BICA a,b , LAURA ELENA MUREŞAN a , LUCIAN BARBU- 

TUDORAN c , EMIL INDREA d , IONEL CĂTĂLIN POPESCU b AND 

ELISABETH-JEANNE POPOVICI a 

ABSTRACT. WO3 films obtained by dip-coating technique were investigated 

to evidence the properties of tungsten trioxide films for water splitting applications. 

The deposition solution containing peroxo-tungstic acid was prepared by 

sol-gel method. The films were deposited on conductive glass substrates and 

were annealed at 250-550 0 C. The properties of WO3 films were investigated 

by UV-Vis Spectroscopy, X-Ray diffraction (XRD) and Scanning Electron 

Microscopy (SEM). 

Keywords: Tungsten trioxide films; ITO support; Dip-coating. 

INTRODUCTION 

Tungsten oxide (WO3) is a wide – band gap semiconductor of great 

interest because of its applications in optoelectronics, catalysis and environmental 

engineering [1-3]. On the other hand, it was demonstrated that WO3 thin films 

exhibits chemical sensing properties such as H2S, NOx, [4-7]. Moreover, WO3 

thin films electrodes are reversible and have fast electrochromic properties [8]. 

Tungsten oxide films can be synthesized by several physical and 

chemical routes such as sputtering [9], acid precipitation method [10] and 

sol-gel processing [11-14]. 

Application of tungsten trioxide (WO3) thin films strongly depends on 

morpho-structural characteristics that are regulated during the synthesis. 

The aim of this study is to obtain high quality WO3 thin films for photocatalysis 

and water splitting applications. The performed study presents the 

influence of some preparative conditions on the morpho-structural characteristics 

a 

Babeş-Bolyai University, “Raluca Ripan” Institute for Research in Chemistry, Fantanele 30, 

Cluj-Napoca, Romania, ebica@chem.ubbcluj.ro 

b 

Babeş-Bolyai University, Faculty of Chemistry and Chemical Engineering, Arany Janos 11, 

Cluj-Napoca, Romania 

c 

Babeş-Bolyai University, Electronic Microscopy Centre, Clinicilor 5-7, Cluj-Napoca, Romania 

d 

National Institute for R&D of Isotopic and Molecular Technologies, Donath 30, Cluj-Napoca, 

Romania

16 

E. BICA, L.E. MUREŞAN, L. BARBU-TUDORAN, E. INDREA, I.C. POPESCU, E.-J. POPOVICI 

of WO3 films, and put in evidence that the quality of the dip-coating solution 

and the thermal treatment play an important role on the properties of conductive 

glass supported WO3 films. 

RESULTS AND DISCUSSIONS 

WO3/ITO/Glass/WO3 heterostructures containing tungsten oxide thin 

films were obtained using the dip-coating method, from peroxo-tungstic acid 

sol. The multilayer technique was used to prepare films with variable thickness 

whereas the thermal treatment was performed at 250 - 550°C (Table 1). 

Film thickness varies between 35 and 135 nm, in parallel with the number of 

dip coating deposition cycles; it seems that the thickness is not influenced 

by the annealing regime. 

Table 1. Synthesis conditions of WO3 thin films prepared from 

peroxo-tungstic acid sol (dip-coating method) 

Sample code Thermal 

treatment 

( 0 C) 

Number of 

layers 

WO3 weight 

(g) 

Film thickness 

(nm) 

R4 I1 350 1 0.55 x 10 -3 

35 

R4 I2 350 2 0.94 x 10 -3 60 

R4 I3 350 3 1.62 x 10 -3 95 

R4 I4 350 4 1.97 x 10 -3 125 

R4 I5 350 5 2.20x 10 -3 135 

R3.1 I2 250 1 0.30 x 10 -3 20 

R3.1 I3 350 1 0.56 x 10 -3 35 

R3.1 I4 550 1 0.50 x 10 -3 35 

In order to establish the optimal thermal treatment regime for WO3 thin 

films, the peroxo-tungstic acid (PTA) precursor was investigated by thermal 

analysis. 

Figure 1. TGA and DTG curves of 

PTA precursor. 

Figure 2. FT-IR spectrum of 

PTA precursor.

STUDIES ON WO3 THIN FILMS PREPARED BY DIP-COATING METHOD 

The TGA and DTG curves of PTA precursor indicate three most 

important weight loss steps i.e. (1) -14.92% at 20-200°C (removal of physical 

adsorbed water and alcohol); (2) -3.35% at 235-345 °C (removal of H2O2, i.e. 

peroxo-tungstic acid decomposition) and (3) -1.05% at 345-405 °C (removal of 

chemically bonded water i.e. tungstic acid decomposition) (Fig.1).The weight 

loss steps are accompanied by weak endo- and exo- thermal effects. 

The FT-IR spectrum of the precursor powder (that corresponds to 

the as- deposited WO3 film), illustrates the hydrated and the hydroxilated 

nature of the WO3 deposit (Figure 2). Water presence is signalled by the 

3418 cm -1 {ν(OH)} and 1617 cm -1 {δ(HOH)} bands. Because the ν(OH) appears 

as a single featureless band, it is difficult to isolate the independent contributions 

from structural water, hydroxyl groups, hydrogen bonded and adsorbed species. 

Despite the complexity of the W-O stretching bands region (400-1000cm -1 ), it 

gives important information about the precursor. Besides the specific ν(W-Ointra-W) 

and ν(W-Ointer-W) bridging stretches ( 863-823 cm -1 and 687 -625 cm -1 ), the 

stretching vibrations of W(O2) and W-O could be noticed (959 cm -1 and 

553 cm -1 ), thus indicating the formation of [(O2)2W(O).O.W(O)(O2)2] 2- complex 

associated with the peroxo-tungstic acid [15-17]. 

The optical properties of WO3 films were evaluated from UV-Vis 

transmission (Figure 3) and reflection (Figure 4) spectra. 

a. b. 

Figure 3. Transmission spectra of WO3 films obtained in different conditions: 

a) multilayer films treated at 350°C; b) monolayer films annealed at 250 – 550°C. 

The monolayer WO3 film (R4I1) shows an almost constant transmittance 

of about 80% on the entire visible domain. As expected, the multilayer films 

R4I2, R4I3, R4I4 and R4I5 have a lower transmittance (50 -80 %) as compared 

with the monolayer heterostructures. The absorption edge shifts from 300 nm 

to 315 nm as the film thickness increases (Figure 3a). 

17

18 


The thermal treatment determines the decreases of the transmittance, 

in parallel with the temperature increase (Figure 3b). The thermal treatment 

also produces the color change of transparent films from colorless to yellowpale, 

thus suggesting some morpho-structural variation. 

Figure 4. Reflectance spectra for the multilayer WO3 film (R4I5) 

measured on ITO face. 

The specular reflectance (8 0 -0 0 ) spectrum is obtained by the difference 

between the total reflectance (8 0 ) and diffuse reflectance (0 0 ) spectra. The 

reflection maximum is situated in the blue range of the spectral domain. 

In order to determine the optical energy band gap of WO3 films, the 

Bardeen equation [18] was used: 

r 

( α hν ) = A( 

hν 

− Eg 

) 

(1) 

where: α is the absorption coefficient, Eg is the energy band gap of the 

semiconductor, h is the Plank’s constant, A is a parameter that depends 

on the transition probability and r is a number that characterises the transition 

process. Depending on the semiconductor type, r values could be: r =2 and 

2/3 for direct allowed and forbidden transitions, respectively, and r=1/2 and 

1/3 for indirect allowed and forbidden transitions, respectively [18]. The 

absorption coefficient α was calculated using the formula (2): 

exp( − α d ) = T 

(2) 

where: 

d is the film thickness (see table 1), and T is the measured transmittance [19].


From the transmittance spectra of WO3 films (calculated without 

substrate), the band gap energy (Eg) was evaluated using the Tauc plot’s, 

by extrapolation of the straight line in the plot (αhν) 1/2 vs hν (Figure 5). 

The determined band gap of dip-coated WO3 films varies between 

2.9 and 3.2eV, in agreement with the literature data [20, 21]. The plot feature 

illustrates that the as obtained WO3 films behaves as an indirect semiconductor 

between about 3.3 and 4.0 eV [20]. 

Figure 5. Plot of (αhν) 1/2 vs hν for WO3 films. 

The X-ray diffraction indicates that peroxotungstic acid precursor 

is amorphous, whereas the WO3 films deposited on conductive glass are 

crystallized (Figure 6). 

Figure 6. XRD patterns for PTA precursor and the corresponding 

heterostructure WO3/ITO/Glass/WO3 (film R4I5). 

19

20 


The XRD pattern contains the characteristic diffraction lines of the 

conductive substrate identified as being cubic SnO2 (JCPDS 33-1374) and 

the diffraction lines of monoclinic WO3 (JCPDS 72-0677). One can be noted 

that, due to the crystalline structure of the substrate, the growth of WO3 films 

seams to be oriented alongside the (200) reflection plane. 

The SEM images illustrate that WO3 film consists on nano-metric 

particles that creates a homogeneous surface (Figure 7). A small tendency 

toward the increase of cracks number with the number of layers could be 

noticed. More than that, the increase of the annealing temperature from 

350 0 C to 550 0 C, leads to the formation of larger crakes in the WO3 film. 

a b 

c 

Figure 7. SEM images of WO3 film surface: (a) one layer (350°C) 

(b) five layers (350 0 C) and (c) one layer (550 0 C). 

CONCLUSIONS 

Homogeneous and adherent WO3 thin films were obtained by dip 

coating technique, on conductive glass substrates from aqueous solution of 

peroxotungstic acid obtained by dissolving fresh prepared tungstic acid into 

hydrogen peroxide solution. Film thickness increases from ~35 to 135 nm 

as the number of dip coating deposition cycles increases.


Thermal analysis and FT-IR spectroscopy suggest that the precursor 

isolated from the colloidal dip-coating solution, corresponds to peroxotungstic 

acid. 

The optical transmittance of WO3 films is influenced by the number 

of layers. In this respect, the WO3/ITO/Glass/WO3 monolayer has a good 

transmittance between 450-1000 nm and it decreases as the film thickness 

increases. The thermal treatment deteriorates the transmission of WO3 films. 

The calculated values of the optical band gap energy (Eg) vary between 2.9 and 

3.2eV, in accordance with the literature data. The reflectance of the as obtained 

WO3 films is dominant in the blue domain of visible spectra. 

SEM images put in evidence that WO3 film morphology depends on 

the layer number as well as the thermal treatment that both determine the 

number of cracks and their size. 


Preparation. In order to obtain WO3 thin films, the sol-gel solution 

was prepared starting from an 0.5M aqueous solution of sodium tungstate 

(Na2WO4 ⋅ 2H2O – Aldrich), which was passed through a cationic exchange 

resin (~2 ml/min) to yield a yellow pale solution of H2WO4. The freshly prepared 

tungstic acid was dissolved in hydrogen peroxide (H2O2, Merck) and the as 

obtained peroxo-tungstic acid (PTA) was stabilized with ethanol addition. In the 

meantime, the conductive glass substrates (30x30x1mm, Optical Filters Ltd.) 

were cleaned in acidic bath and alcohol, and dried. From this peroxo-tungstic 

acid sol, WO3 films were deposited on the conductive support (notated ITO), 

by dip-coating method, using an withdrawal speed of 4cm/min. The films were 

dried at 110 0 C, and annealed at 350-550 0 C for 30 minutes, in air. Several 

dipping-drying cycles were used to consolidate the WO3 structure. 

Sample characterization. The PTA precursor (dried at ~70°C) was 

investigated by thermal analysis (Mettler Toledo TGA/SDTA851; heating 

rate 5 0 C/min; nitrogen atmosphere) and FT-IR Spectroscopy (JASCO 610 

Spectrometer; KBr pellets technique). 

UV-Vis spectroscopy (UNICAM Spectrometer UV4, with RSA-UC-40 

integrating sphere accessory), X-ray diffraction (DRON 3M Diffractometer, 

CoKα radiation) and scanning electronic microscopy (JEOL-JSM 5510LV 

Microscope Au coated samples) were used to characterize the WO3 thin films. 

The films thickness was estimated by micro-weighing method (Saltec Balance). 

ACKNOWLEDGEMENTS 

This work was supported by the Romanian Ministry of Education, 

Research and Innovation (Project: 71-047). 

21

22 


REFERENCES 

1. J. Luo, M. Hepel, Electrochim. Acta, 2001, 46, 2913. 

2. S. Wang, X. Shi, G. Shao, S. Duan, H. Yang, T. Wang, J. Phys. Chem. Solid., 

2008, 69, 2396. 

3. J.-C. Yang, P. K. Dutta, Sensors and Actuators, 2008, 136, 523. 

4. A. K. Chawla, S. Singhal, H. O. Gupta, R. Chandra, Thin Solid Films, 2008, 517, 

1042 

5. C. Santato, M. Odziemkowski, M. Ullman, J. Augustinski, J. Am. Chem. Soc., 

2001, 123, 10639. 

6. A. I. Gavrilyuk, Electrochim. Acta, 1999, 44, 3027. 

7. P. M. S. Monk, R. D. Partridge, R. Janes, C. G. Granqvist, Solar Energ. Mater. 

Solar Cell., 2000, 60, 201-262. 

8. G. Leftheriotis, P. Yianoulis, Solid State Ionics, 2008, 179, 2192. 

9. S. Supothina, P. Seeharaj, S.Yoriya, M. Sriyudthsak, Ceramics International, 

2007, 33, 931. 

10. M. Deepa, R. Sharma, A. Basu, S. A. Agnihotry, Electrochim. Acta, 2005, 50, 3545. 

11. Y. Suda, H. Kawasaki, T. Ohshima, Y. Yagyuu, Thin Solid Films, 2008, 516, 4397. 

12. B. Yang, P. R. F. Barnes, W. Bertram, V. Luca, J. Mater. Chem., 2007, 17, 2722. 

13. L. Muresan, E. J. Popovici, A. R. Tomsa, L. Silaghi-Dumitrescu, L. Barbu-Tudoran, 

E. Indrea, J. Optoelec. Adv. Mater., 2008, 10, 2261. 

14. K. Huang, J. Jia, Q. Pan, F. Yang, D. He, Physica B, 2007, 396, 164. 

15. B. Pecquenard, H. Lecacheux, J. Livage, C. Julien, J. Solid State Chem., 1998, 

135, 159. 

16. A. Novinrooz, M. Sharbatdaran, H. Noorkojouri, Central European J. Phys, 2005, 

3, 456. 

17. M. F. Daniel, B. Desbat, J. Solid State Chem., 1992, 67, 235. 

18. M. G. Hutchins, O. Abu-Alkhair, M. M. El-Nahass, K. Abd El-Hady, Mater.Chem. 

Phys, 2006, 98, 401. 

19. P. Sharma, V. Sharma, S. C. Katyal, Chalcogenide Letters, 2006, 3, 73. 

20. K. J. Lethy, D. Beena, R. V. Kumar, V. P. Mahadevan Pillai, V. Ganesan, V. Sathe, 

Applied Surface Sci., 2008, 254, 2369. 

21. M. Deepa, A. K. Srivastava, M. Kar, S. A. Agnitory, J. Phys. D: Applied Phys., 

2006, 39, 1885.


PREPARATION AND CHARACTERIZATION OF MANGANESE 

DOPED ZINC SULPHIDE NANOCRYSTALLINE POWDERS 

WITH LUMINESCENT PROPERTIES 

ADRIAN-IONUŢ CADIŞ a,b , ADRIAN RAUL TOMŞA a , ECATERINA BICA a , 

LUCIAN BARBU-TUDORAN c , LUMINIŢA SILAGHI-DUMITRESCU b , 


ABSTRACT. Manganese-doped zinc sulphide nanocrystalline powders have 

been synthesized from zinc-manganese acetate and sodium sulphide, in 

aqueous solution containing methacrylic acid. Precipitation was performed at 

low temperature, using the sequential reagent addition technique. Different Mn 2+ 

concentrations have been used to control the optical properties of ZnS:Mn 2+ 

nanoparticles. All samples were characterized by thermal analysis (TGA), infrared 

absorption spectroscopy (FT-IR), photoluminescence spectroscopy (PL), 

scanning (SEM) and transmission electron microscopy (TEM). A correlation 

between the preparation conditions and optical and morphological characteristics 

of ZnS:Mn 2+ powders was established. 

Keywords: Zinc sulphide, Mn-doped nanoparticles, Photoluminescence 

INTRODUCTION 

Nanocrystalline materials have been of interest for more than 25 years 

[1-3]. The main cause is in their unusual properties based on the high 

concentration of atoms at interfacial structure and the relative simple ways of 

their preparation. Recently, nanoparticles of zinc sulphide have become the 

subject of intense investigations due to their potential applications in catalysis, 

sensors, nonlinear optics and molecular electronics [4-5]. At the same time, 

the synthesis of zinc sulphide (ZnS) particles with uniform morphology and 

narrow size distribution is still in the future and need to be further studied. 

a 

”Raluca Ripan” Institute for Research in Chemistry, “Babes- Bolyai” University, 30 Fântânele, 

RO-400294 Cluj-Napoca, Romania, cadisadi@chem.ubbcluj.ro 

b 

Faculty of Chemistry and Chemical Engineering, “Babes-Bolyai” University, 11 Arany Janos, 

RO-400028 Cluj-Napoca, Romania 

c 

Electronic Microscopy Centre, “Babes-Bolyai” University, 5-7 Clinicilor, RO-400006 Cluj-Napoca, 

Romania

A.-I. CADIŞ, A.R. TOMŞA, E. BICA, L. BARBU-TUDORAN, L. SILAGHI-DUMITRESCU, E.-J. POPOVICI 

The increasing interest in these materials has lead to the development of 

a variety of chemical routes to prepare nanoparticles, including sputtering [6], 

ultrasound irradiation [7], co-evaporation [8], sol–gel method [9], solid-state 

reaction [10], gas-phase condensation [11], liquid-phase chemical precipitation [12], 

ion complex transformation [13], microwave irradiation [14] and biological 

synthesis [15]. From all these works, it has been found that particle size 

and luminescent properties of ZnS powders depend strongly on the specific 

preparation method and the applied experimental conditions. 

This paper is the first in the series dedicated to the comparative 

investigation of luminescent and morphostructural properties of nanocrystalline 

ZnS:Mn powders obtained by different synthesis routes. In this respect, attempts 

to obtain manganese doped ZnS nanoparticles are performed by precipitation, 

using the sequential reagent addition technique (SeqAdd). In order to control the 

particle morphology and size, methacrylic acid is used as passivating agent. A 

systematic investigation on the influence of Mn-concentration on the photoluminescence 

properties of nanocrystalline ZnS:Mn 2+ is reported. A variety 

of methods including scanning electron microscopy (SEM), Fourier transform 

infrared spectroscopy (FTIR), and photoluminescence spectroscopy (PL) were 

used to characterise the ZnS:Mn nanoparticles. 


The goal of our study was to obtain luminescent manganese doped 

zinc sulphide ZnS:Mn 2+ nanoparticles with controlled particle dimensions. In 

this purpose, attempts were made to prepare ZnS:Mn 2+ powders, starting from 

Zn-Mn acetate mixture and sodium sulphide, in presence of methacrylic acid as 

particle size regulating agent. Precipitation was performed at low temperature, 

using SeqAdd technique. Zn-Mn acetate mixtures with variable compositions 

were used to obtain ZnS nanoparticles with variable Mn-doping level. The as 

obtained precipitation product can be considered as Zn-Mn double sulphide 

and the chemical process can be described by the overall equation: 

In this manner, a series of 6 samples was prepared from mixture 

containing the following Mn/(Zn+Mn) ratios: 0 mol% (CAI17), 5.6 mol% (CAI19), 

11.0 mol% (CAI73), 16.4 mol% (CAI21), 21.8 mol% (CAI20) and 25.0 mol% 

(CAI77). Mention has to be made that, the amount of incorporated Mn is 

about 0.1% of its concentration, as illustrated by the ICP measurements. 

24

PREPARATION AND CHARACTERIZATION OF MANGANESE DOPED ZINC SULPHIDE … 

The as prepared ZnS powders show a high capacity to absorb anionic 

impurities from the precipitation medium, as exemplified by thermal analysis 

and FTIR spectroscopy. 

The thermal behaviour of ZnS:Mn 2+ powders was investigated in the 

25-1100ºC range, under N2 flow. Thermogravimetric (TGA) and differential 

thermogravimetric (DTG) curves of ZnS powder are depicted in Figure 1. 

There are four major weight loss steps i.e. (1) - 7.2% at 20-188°C; (2) -9.9% 

at 188-300°C; (3) - 9.9 % at 300-850°C and (4) over 850°C that could be 

associated with: (1) removal of physically adsorbed water; (2) thermal dissociation 

of the acetate species adsorbed on the particle surface; (3) thermal dissociation 

and removal of organic compound and (4) sublimation of zinc sulphide, 

facilitated by the N2 flow. 

Figure 1. TGA and DTG curves for 

ZnS:Mn powder (CAI21) 

Figure 2. FT-IR spectrum of ZnS:Mn 2+ 

sample (CAI21) 

The FT-IR spectrum of ZnS:Mn 2+ powder show the characteristic 

vibration bands for metal acetate, water and methacrylic acid, ionic or molecular 

compounds adsorbed from the precipitation medium (Fig.2.). The most 

important absorption bands are assigned as follows: hydroxyl groups (O-H) 

(between 3000 and 3600 cm -1 ), CH3 bending modes (950-1100 and 2800- 

3000 cm -1 ), H2O and COO group (between1300 and 1600 cm -1 ) and C=C 

and =CH groups (1000-1200 cm -1 ) [16]. 

Mention has to be made of the large surface area of powders 

determines a high amount of anionic species to be adsorbed from the 

precipitation medium. In spite of the fact that all samples were carefully 

washed and centrifuged, the contaminants removal can not be completely 

done, due to the ZnS hydrolysis. 

The as prepared manganese doped zinc sulphide powders show 

photoluminescence properties under ultraviolet excitation. Figure 3 shows 

the emission spectra (λexc = 335 nm) for samples synthesised from acetate 

25


mixtures with different Mn 2+ concentrations. Nanocrystalline ZnS:Mn 2+ shows a 

weak blue emission (430 nm) and an orange emission (600 nm) under 335 nm 

excitation. The blue emission can be assigned to a defect-related emission of 

the ZnS host-lattice whereas the orange emission can be attributed to the 

4 T1- 6 A1 transition of the Mn 2+ ion. 

The sample without Mn 2+ shows only the ZnS-related blue emission. 

As soon as Mn 2+ is incorporated in the ZnS nanoparticles, the intensity of 

the blue emission decreases and the Mn 2+ emission comes up, since the 

energy transfer between ZnS host and Mn 2+ impurity is very efficient. With an 

increasing Mn 2+ concentration, the characteristic 600 nm emission becomes 

stronger. 

Figure 3. Emission spectra of ZnS:Mn 2+ 

powders obtained from acetate mixtures 

with various Mn 2+ concentration. 

26 

Figure 4. Relative intensity of the orange 

and blue bands versus Mn concentration 

(blue band is 5 times multiplied) 

In figure 4, the luminescence intensity of the orange and blue bands 

is plotted as a function of the Mn/(Zn+Mn) ratio from acetate solution. The 

orange band intensity increases with Mn 2+ concentration, in parallel with 

the increase of the emission centres numbers. At higher Mn-amounts, the 

luminescence intensity starts to decrease due to the mutual interaction 

between the doping (activator) ions (concentration quenching). The maximum 

intensity is reached at about 11% Mn 2+ and corresponds to the optimum 

ratio between the number of the emission centres and the quenching ones. 

The blue band intensity decreases continuously with increasing Mn 2+ amount, 

due to the decrease of the numbers of self activated centres related with 

the lattice defects of the zinc sulphide. 

Usually the PL properties of ZnS powders are connected with a high 

temperature firing stage. In this case, the nanostructure state of ZnS powders 

determines the unexpected intense luminescence.


The morphology and particle dimensions were put in evidence by 

scanning (SEM) and transmission (TEM) electron microscopy investigations. 

The SEM image of CAI21 sample shows that ZnS:Mn 2+ powder consists of 

1-3 µm aggregates of tightly packed particles with sizes under 20 nm (Fig. 5). 

The TEM image of the same sample illustrates that in fact, the ZnS:Mn 2+ 

powder consists from very small particles (quantum dots) with diameters less 

than 3 nm. Due to the high surface area, these nanoparticles show a strong 

tendency toward agglomeration to much larger particles. 

Figure 5. SEM image of ZnS:Mn 2+ 

sample (inset scale bar = 1 μm) 

CONCLUSIONS 

Figure 6. TEM image of ZnS:Mn 2+ 

(scale bar=50 nm) 

Manganese doped ZnS nanoparticles were obtained by precipitation, 

using the sequential reagent addition technique (SeqAdd). In order to 

control the particle morphology and size, methacrylic acid was used as particle 

regulating agent. The influence of Mn concentration on the luminescence 

properties of ZnS:Mn 2+ nanocrystals was investigated. Photoluminescence 

measurements show that the as obtained ZnS:Mn nanopowders exhibit a 

strong orange emission centred at 600 nm that is related with the Mn emission 

centres. The maximum intensity is reached at about 11% Mn 2+ in the Zn-Mn 

acetate mixture and corresponds to the optimum ratio between the number 

of the emission centres and the quenching ones. The strong PL of the unannealed 

ZnS:Mn 2+ powder could be associated with the particle nanodimension. 

Although ZnS:Mn 2+ powders are formed from nano-sized crystallites 

(about 3 nm), they are tightly packed into larger and irregular shaped particles. 

The large surface area explains the high absorption capacity of the ZnS 

powder, as illustrated by TGA and FTIR investigations. 

Supplementary work has to be done in order to improve the ZnS:Mn 2+ 

powder dispersability. 

27



ZnS powders were prepared by precipitation, using SeqAdd technique, 

from Zn-Mn acetate and sodium sulphide in aqueous medium, at low 

temperatures (5°C). For this purpose, Zn-Mn acetate mixture containing 

5.6, 11.0, 16.4, 21.8 and 25.0 mol Mn/100 mol (Zn+Mn) was prepared from 

1M Zn(CH3COO)2 and 1M Mn(CH3COO)2 solutions and it was diluted with 

deionised water containing α-methacrylic acid (MA). The aqueous solution 

of Na2S was added to the above mixture solution and vigorously stirred for 

30 min. The resulting powder was washed and centrifuged, and finally dried 

at 80°C under vacuum. The wash process was performed with deionised 

water and isopropyl alcohol. As a comparison, pure ZnS nanoparticles were 

also prepared using the above method. 

A METTLER-TOLEDO TGA/SDTA851 thermogravimeter was used for 

thermal and differential thermal gravimetry (TGA-DTG). The measurements 

were performed in alumina crucibles, in nitrogen flow (20 mL/min), with a 

heating rate of 5°C/min. Infrared absorption spectra (FTIR) of the samples 

prepared in KBr pellets were registered on a NICOLET 6700 FT-IR Spectrometer. 

Photoluminescence (PL) spectra were registered with JASCO FP-6500 Spectrofluorimeter 

Wavel equiped with photomultiplier PMT R928 (Farbglasfilter WG 

320-ReichmannFeinoptik) and were normalized to the maximum intensity of 

the best sample (CAI73). The scanning electron microscopy (SEM) images 

were obtained with a JEOL–JSM 5510LV electron microscope using Au-coated 

powders. The accelerating voltage was 20 kV. The transmission electron 

microscopy (TEM) was performed with JEM JEOL 1010 microscope. The 

accelerating voltage was 20 kV. 


Financial support for this study was provided by the Romanian Ministry 

of Education, Research and Innovation (Project ID-2488). 

28 

REFERENCES 

1. A. Henglein, Berichte der Bunsengesellschaft für Physikalische Chemie, 1982, 

86, 301. 

2. H. Gleiter, Progress in Materials Science, 1989, 33, 223. 

3. L. F. Chen, Y. H. Shang, J. Xu, H. L. Liu, Y. Hu, Journal of Dispersion Science 

and Technology, 2006, 27, 839.


4. J. Wen, G. L. Wilkes, Chemistry of Materials, 1996, 8, 1667. 

5. M. Miyake, T. Torimoto, M. Nishizawa, T. Sakata, H. Mori, H. Yoneyama, Langmuir, 

1999, 15, 2714. 

6. S. K. Mandal, S. Chaudhuri, A. K. Pal, Thin Solid Films, 1992, 350, 209. 

7. A. R. Tomsa, E. J. Popovici, A. I. Cadis, M. Stefan, L. Barbu-Tudoran, S. Astilean, 

Journal of Optoelectronics and Advanced Materials, 2008, 10, 2342. 

8. R. Thielsch, T. Bohme, H. Bottcher, Physica Status Solidi A, 1996, 155, 157. 

9. B. Bhattacharjee, D. Ganguli, S. Chaudhuri, A. K. Pal, Thin Solid Films, 2002, 

422, 98. 

10. P. Balaz, E. Boldizarova, E. Godocıkova, J. Briancin, Materials Letters, 2003, 57, 

1585. 

11. J. C. Sanchez-Lopez, A. Fernandez, Thin Solid Films, 1998, 317, 497. 

12. J. F. Suyver, S. F. Wuister, J. J. Kelly, A. Meijerink, Nano Letters, 2001, 8, 429. 

13. W. B. Sang, Y. B. Qian, J. H. Min, M. D. Li, L. L. Wang, W. M. Shi, Y. F. Liu, 

Solid State Communications, 2002, 121, 475. 

14. Y. Jiang, Y. J. Zhu, Chemistry Letters, 2004, 33, 1390. 

15. H. J. Bai, Z. M. Zhang, J. Gong, Biotechnology Letters, 2006, 28, 1135. 

16. N. B. Colthup, L. H. Daly, S. E. Wiberley, “Introduction to Infrared and Raman 

Spectroscopy”, Academic Press, New York, 1964. 

29


ELECTRICAL CHARACTERISTICS OF A BIOBATTERY 

WITH STAPHYLOCOCCUS AUREUS 

COSMIN CĂŢĂNAŞ a , MIHAI MOGOŞ a , DANIEL HORVAT a , JAKAB ENDRE b , 

ELEONORA MARIA RUS a , IULIU OVIDIU MARIAN a 

ABSTRACT. A homemade biobattery was studied from electrical point of view. 

The open circuit voltage for a 0.2 mA short circuit current was of 0.6 V. The 

maximum power for an internal resistance of 1.1 kΩ and a concentration of 

10 9 cells/cm 3 of Staphylococcus aureus was of 17.57 μW. 

Keywords: biobattery, electrical power, bacteria, Staphylococcus aureus. 

INTRODUCTION 

In the last years, microbial fuel cells (MFCs) have received a great deal 

of attention as a promising solution for renewable energy generation and waste 

disposal [1-6]. A MFC converts energy, available in a bio-convertible substrate, 

directly into electricity through the catalytic activities of microorganisms. 

It can be said that the MFCs are a hybrid of biological and electrochemical 

reactors. By this dualistic nature, MFCs offer the advantage of utilizing 

a wide range of organic compounds as fuel, and exploit the value of electrochemical 

cells by direct generation of electricity [7]. 

In a MFC, electricity is being generated in a direct way from biowastes 

and organic matter. This implies that the overall conversion efficiencies that can 

be reached are potentially higher for MFCs compared to other biofuel processes. 

Parameters influencing the overpotentials at the anode are the 

electrode surface, the electrochemical characteristics of the electrode, the 

electrode potential, and the kinetics together with the mechanism of the 

electron transfer and the current of the MFC. 

Mediators are important in MFC cells that use microorganisms such 

as Escherichia coli, Pseudomonas, Proteus, and Bacillus species that are 

unable to effectively transfer electrons derived from central metabolism to the 

a Babeş-Bolyai University, Faculty of Chemistry and Chemical Engineering, Romania, Cluj- 

Napoca, Arany Ianos, 11 iomar@chem.ubbcluj.ro 

b Babeş-Bolyai University, Center for Molecular Biology, Romania, Cluj-Napoca, Treboniu 

Laurian, 42

32 

C. CĂŢĂNAŞ, M. MOGOŞ, D. HORVAT, E. JAKAB, E. M. RUS, I. O. MARIAN 

outside of the cell. Common electron shuttles include thionine, benzylviologen, 

2,6-dichlorophenolindophenol, 2-hydroxy-1,4-naphtho-quinone and various 

phenazines, phenothiazines, phenoxoazines, iron chelates and neutral red [8]. 

These electron shuttles are typically capable to cross the cell membranes, 

accepting electrons from one or more electron carriers within the cell, exiting 

the cell in the reduced form and then transferring electrons onto the electrode 

surface. The critical issue with mediated electron transfer is the diffusion of 

the shuttle out of the biofilm or the bacterial environment [9]. 

There is a report on the bacteria, Rhodoferax ferrireducens that can 

be used in microbial fuel cells effectively without a mediator [10]. 

Oxygen is generally used as the electron acceptor for the cathodic 

reaction in MFCs. Graphite is a commonly used electrode material. To improve 

catalytic activity, graphite can be modified with platinum [11]. 

Various metals (copper, gold, palladium/cobalt, molybdenum, tungsten, 

and manganese) and their complexes have been investigated as replacements 

for expensive platinum in the cathode [11-13]. 

Because the power output of MFCs is low relative to other types of fuel 

cells, reducing their cost is essential, if power generation using this technology 

is to be an economical method of energy production. Further research is 

required to enhance the power production. 

In this study, we investigate the performance of a biobattery with 

Staphylococcus aureus bacteria without mediator in the original arrangement 

of electrode materials at anode compartment because the biobatteries are 

less investigated at this moment. 


The purpose of the experiments was to measure the open circuit 

voltage, the short circuit current and to plot the power curve of the biobattery. 

The bacteria used in the experiment were Staphylococcus aureus and the 

substrate was the sodium acetate. 

FeCl3 was used as a cathode mediator in order to improve oxygen 

reduction kinetics [14-15] and acetate was used as substrate in the anodic 

compartment. The purpose of the experiment was to measure the voltage 

between the two compartments at 2 different concentrations of acetate in 

the absence and presence of the bacteria. The activation of the biobattery 

has been achieved by adding the bacterial solution. After a few seconds it 

was observed the bubbling process on the anode rods electrode material 

due to the carbon dioxide gas resulted from the bacterial activity. 

The pH of first solution was 8.6 and for the second solution 8.9.

ELECTRICAL CHARACTERISTICS OF A BIOBATTERY WITH STAPHYLOCOCCUS AUREUS 

Figure 1. Biobattery representation 

In the first experiment by monitoring the cell voltage for 5 hours, a drift 

of OCV (open circuit voltage) appears from 0.46V to 0.42 V before activation. 

A maximum peak of 0.6 V was measured after biobattery activation. The short 

circuit current was of 0.2 mA. (Figure.2a). 

a b 

Figure 2. Voltage-time dependency. 

a) for 0.05M sodium acetate solution b) for 0.1M sodium acetate solution 

In the second experiment before activation the same drift appear 

(about 40mV) and after activation we measured a maximum cell voltage of 

0.3 V and the short circuit current of 0.2mA (Figure 2b). In both cases the 

open circuit voltage remains at the constant values of 0.56V respectively 

0.26 V for one month. Current-time curve and power curve, was recorded 

for the biobattery with increased OCV. 

33

34 


Figure 3. Short circuit current-time dependency 

0.05M sodium acetate solution 

Initially, after the biobattery activation, a peak current of about 200 µA 

was registered. After 20 minutes, the limit current of about 20 µA was reached 

in agreement with the last point of the power curve (Figure 4). 

Figure 4. Power curve 0.05M sodium acetate solution

ELECTRICAL CHARACTERISTICS OF A BIOBATTERY WITH STAPHYLOCOCCUS AUREUS 

As we reduce the external resistance, a decrease of the voltage 

was recorded. In this way, we looked for having the smallest possible drop 

in voltage as the current is increased in order to attain the maximum power 

production over the investigated range of interest. 

The internal resistance of the battery and the maximum power 

generated were calculated by using the Mathworks MATLAB® environment 

for processing the experimental data. Spline functions have been used to 

approximate the power curve. (Figure 4). 

All of the equipment used was sterilized with hydrogen peroxide 

after the experiments. 

CONCLUSIONS 

The maximum power of the biobattery was of 17.57 µW, achieved at the 

internal resistance of 1.1 kΩ and for the external resistance of about 15 kΩ 

the minimum power reached was 5.45 µW, in the case of 0.05M substrate 

concentration in anode compartment. 

The value of the OCV remained at the constant value of about 0.56 V 

for a month. The difference which appears in OCV in both experiments can be 

explained by decreasing of the bacterial activity whit pH increase. 

The biobatteries represent a viable method of generating electricity 

from chemical energy. Further research must be done with the purpose of 

reducing the dimensions of the biobattery for possible applications in medicine 

and other fields. 


An electrochemical cell has been built with two chambers separated by 

a Nafion proton exchange membrane (Figure 1). Nine graphite rod electrodes 

of 6 mm in diameter and10 cm in length have been used for each chamber. 

For the anode compartment a 300 ml sodium acetate solution 0.05 M and 

0.1 M respectively, has been combined with 2 ml of Staphylococcus aureus 

(10 

35 

9 cell/ml concentration). For the cathode compartment, 300 ml of 1 M FeCl3 

solution was used. 

The circuits which contain a biobattery and multimeters were connected 

to a computer with continuous data acquisition software. 

The used bacteria were Staphylococcus aureus. The bacteria posed 

no threat to the working environment because they were non sporogenous. 

Bacterial strains. Both used bacterial strains, the UCLA 8076 (University 

of California, Los Angeles, USA) [16] and the 1190R (Semmelweis University, 

Budapest, Hungary) [17], were heterogeneous MRSA (methicillin-resistant 

Staphylococcus aureus)strains. These strains are preserved in glycerol (25% 

final concentration) at -80°C and prepared as follows: 5ml of overnight bacteria 

culture at 37°C in LB broth (Gibco BRL, Life Technologies, Paisley, Scotland)


were centrifuged in 15ml centrifuge tubes in a Sigma 1-18K centrifuge (Sigma 

Laborzentrifugen, Osterode am Harz, Germany –5,500 rpm, 10 min., room 

temperature) and the pellet was resuspended in 500μl sterile 50% glycerol. 

The thawed bacteria were cultured overnight in 5ml of Mueller Hinton 

broth (Fluka, Buchs, Switzerland) into a Certomat BS-T incubation shaker 

(Sartorius Stedim Biotech, Aubagne, France) at 37°C, 150 rpm until the 

culture reached an OD600 of 0.8-1.0 (Spekol UV VIS 3.02, Analytic Jena, Jena, 

Germany). A loop of bacterial suspension was passed on Mueller Hinton agar 

(Fluka, Buchs, Switzerland) to maintain the culture for further analysis. The 

bacteria were cultured also on blood agar plates (Fluka, Buchs, Switzerland) 

and confirmed using colony morphology method. 

REFERENCES 

1. K.J. Chae, M. Choi, F.F. Ajayi, W. Park, I.S. Chang and I.S. Kim, Energy & Fuels, 

2008, 22, 169. 

2. B.H. Kim, I.S. Chang, G.M. Gadd, Appl. Microbiol. Biotechnol., 2007, 76, 485. 

3. B.E. Logan, Water Science & Technology, 2005, 52, 31. 

4. B.E. Rittmann et all., Environ. Sci. Tech., 2006, 15, 1097. 

5. D.R. Lovley, Microbiology, 2006, 4, 497. 

6. A.K. Shukla, P. Suresh, S. Berchmans, A. Rajendran, Current Science, 2004, 87, 

455. 

7. A.K. Marcus, C.I. Torres, B.E. Rittmann, Biotechnology and Bioengineering, 

2007, 98, 1171. 

8. K. Rabaey, W. Verstraete, TRENDS in Biotechnology, 2005, 23, 291. 

9. S.E. Childers, S. Ciufo, D.R. Lovley, Nature, 2002, 416, 767. 

10. A.K. Shukla, P. Suresh, S. Berchmans and A. Rajendran, Current Science, 2004, 

87, 455. 

11. T.H. Pham, J.K. Jang, I.S. Chang, and B.H. Kim, J. Microbiol. Biotechnol., 2004, 

14, 324. 

12. K.H. Kang, J.K. Jang, T.H. Pham, H.I. Moon, S.B. Chang, H. Kim, Biotechnol. Lett., 

2003, 25, 1357. 

13. I.S. Chang, J.K. Jang, G.C. Gil, M. Kim, H.J. Kim, B.W. Cho, Biosens. Bioelectron., 

2004, 19, 607. 

14. S. Cheng, H. Liu, B.E. Logan, Environ. Sci.Technol., 2006, 40, 364. 

15. B.H. Kim, H.J. Kim, M.S. Hyun, D.H. Park, J. Microbiol.Biotechnol., 1999, 9, 127. 

16. J.M. Swenson, F.C. Tenover, J. Clinical Microbiology, Cefoxitin Disk Study Group, 

2005, 43(8), 3818. 

17. L. Majoros, E. Papp Falusi, I. Andriko, F. Rozgonyi, Acta Microbiologica et 

Immunologica Hungarica, (1996) 43(2-3):160. 

36


MODELING AND SIMULATION OF CARBON DIOXIDE 

ABSORPTION IN MONOETHANOLAMINE IN PACKED 

ABSORPTION COLUMNS 

ANA-MARIA CORMOS a , JOZSEF GASPAR a , 

ANAMARIA PADUREAN a 

ABSTRACT. Computer-based process simulations are today recognized as 

an essential tool to be applied in chemical process industries. For this paper a 

mathematical model is needed to describe closer the real physical – chemical 

processes that take place in a reactor (packed absorption column) used for 

carbon dioxide absorption in mono-ethanolamine aqueous solution. 

The mathematical model was developed for analyzing absorption rate 

and understanding of micro level interaction of various processes taking place 

inside de absorption column. The modeling includes transfer processes: mass 

and heat to study the coupled effect of temperature and concentration on the 

rate of absorption. The reaction kinetics and the vapor-liquid equilibrium (VLE) 

are other important parts of the mathematical model. The developed equations 

include both ordinary differential equations and partial differential equation. 

The discretization was used to transform partial differential equations in ordinary 

ones. 

The evolutions (in time and space) of the processes parameters (liquid and 

gaseous flows, composition of the streams, temperatures etc.) were studied 

for carbon dioxide absorption process. 

Keywords: Chemical absorption, Packed column, Stationary-dynamic model 

INTRODUCTION 

The protection of the environment is a significant problem of modern 

human society for a sustainable development. For environmental protection 

and climate change mitigation, the reports of Intergovernmental Panel on 

Climate Change established on scientific basis that climate change and rising of 

global temperature levels noticed in the past 50 years are undoubteful linked 

with human activity and greenhouse gas emissions (mainly carbon dioxide) [1]. 

At the same time with realizing the danger there had been adopted strict 

laws to control the emission of polluting substances. From different greenhouse 

gases, the most important from the volume of emissions is carbon dioxide. 

a Babes-Bolyai University, Faculty of Chemistry and Chemical Engineering, Kogalniceanu Street, 

No. 1, RO-400084 Cluj-Napoca, Romania, cani@chem.ubbcluj.ro

38 

ANA-MARIA CORMOS, JOZSEF GASPAR, ANAMARIA PADUREAN 

For the limitation of climate changes, the goal is to reduce the emission of 

carbon dioxide by capture and storage techniques. For carbon dioxide capture 

there are multiple technological options: capture from combusted gases resulted 

after the burning of fuels (post-combustion capture), capture before the burning 

of fuels (pre-combustion capture), or the oxy-combustion technique where the 

fuels are burned in oxygen, not in air, and after the condensation of the water 

from the burnt gases carbon dioxide can be captured and stored because it 

is the main component. Once carbon dioxide captured, has to be stored in 

safety under special conditions for a long period of time [2,3]. 

Nowadays the absorption of carbon dioxide in alkanolamines is the 

most common technology and one of the few technological options viable 

for large CO2 removal process (to be applied for instance in power generation 

sector). Economic studies show that this method will be competitive also in the 

future. To extensive laboratory work, modeling and simulation of absorption 

process are needed in order to evaluate the process. 

This paper aims to develop a dynamic mathematical model for carbon 

dioxide absorption in aqueous solution of mono-ethanolamine in packed 

absorption columns. The main purpose of the paper is to develop and than 

to validate a model which can be then used in various operation condition 

for optimization of techno indicators of carbon dioxide absorption process. 


Mathematical model of absorption column 

Carbon dioxide absorption using alkanolamines falls in the general 

category of chemical gas – liquid absorption [4,5]. In Figure 1, a simplified 

flowsheet of a carbon dioxide capture using mono-ethanolamine (MEA) is 

presented. The setup consist of 3 major sections [5]: (1) reactor-absorption 

section (tower column called), (2) the liquid low section consisting the peristaltic 

pump and (3) gas collection section with the gas flow composition and pressure 

measurement apparatus. In absorption column with stuffing, the liquid and a gas 

flowing in opposite directions, enable a chemical component to be transferred 

from the gas phase to the liquid phase. The gas enters at the base of the 

absorption column, it flow rates being determined by variable area flow meters. 

The liquid to the top of the column is similarly metered and falls through the 

packing where it is contacted with the rising gas. 

A mathematical model has been developed for prediction the rate 

absorption of carbon dioxide into aqueous solution of mono-ethanolamine (MEA). 

The basic assumption considered for the CO2 absorption by mono-ethanolamine 

solution in an absorption column with Raschig rings are presented below: 

• Model parameters are constant in the radial cross section of the column 

(piston-type flow); 

• Both gas and solids velocities are considered constant;

MODELING AND SIMULATION OF CARBON DIOXIDE ABSORPTION IN MONOETHANOLAMINE … 

• Heat transfer by conduction and radiation are negligible in the axial 

direction; 

• Carbon dioxide solubility in liquid phase is in accordance with the 

Henry's law; 

• Phase liquid is not volatile in the process temperature; 

• Chemical reaction takes place only in the film of liquid; 

• The absorption columns is operated adiabatic; 

• Both phases (liquid and gas) are considered ideal mixtures. 

Figure 1. Gas – liquid packed absorption column 

Chemical absorption of carbon dioxide can be done using base chemicals 

like alkaline hydroxides, ammonia or organic compounds (e.g. alkanolamine). 

The absorption of CO2 into alkanolamine solution involves a chemical reaction 

of a weak base (alkanolamine) with a weak acid (CO2). The chemical equilibrium 

takes place in the liquid phase, when carbon dioxide is absorbed in an 

aqueous solution of alkanolamines. 

The overall chemical reaction occurring in the liquid phase between 

CO2 and mono-ethanolamine (MEA) may be expresses as follow: 

CO2 + 2 RNH2 ↔ RNH3 + + RNHCOO - 

where R indicates HOCH2CH2-. This chemical reaction is second order, 

respectively is first order with respect to CO2 and MEA separately [5], and 

39


the reaction rate (r), can be defined in function of the molar concentration of 

reactants, as follows: r=k*CA*CB [5], where k is the reaction rate constant The 

temperature dependence of the reaction rate constant is presented bellow [3]: 

8 ⎝ 

k = 4. 

4⋅10 

⋅e 

⎛ 5400⎞ 

⎜− 

⎟ 

T ⎠ 

3 

m / mol⋅ 

s 

The model equations include both partial differential equations and 

algebraic equations. The model contains basically mass and heat conservation 

equations presented below [3,6,7]. A list of abbreviations used is presented 

at the end of the paper. 

The mass balance for liquid phase is described by the equations of 

consuming both reactants (mono-ethanolamine and carbon dioxide): 

∂C 

∂t 

∂C 

∂t 

A 

B 

∂CA 

= −vL 

⋅ −k 

⋅CA 

⋅CB 

+ E⋅K 

G ⋅au 

⋅ 

∂z 

∂CB 

= −vG 

⋅ −b⋅k 

⋅CA 

⋅CB 

∂z 

( C −H 

⋅C 

) 

The heat balance equation for the gas phase is presented in the 

following equation: 

40 

AG 

CO2 

A 

(1) 

(2, 3) 

The effect of a chemical reaction is given by the enhancement factor, E, 

defined as the ratio of the absorption rate of a gas into a reacting liquid to that if 

there was no reaction [7]. The enhancement factor can be approximated [6,7]: 

2 , k ⋅ DCO G ⋅CB 

E = (4) 

k 

The heat balance equation for the liquid phase is presented in the 

following equation (the chemical reaction between carbon dioxide and monoethanolamine 

is exothermic): 

∂T 

∂t 

∂T 

L 

Δ H ⋅ k ⋅ C ⋅C 

h ⋅ a 

( T − ) 

L L R 

A B 

u 

= −vL 

⋅ − 

+ ⋅ G TL 

∂z 

ρsol 

⋅ cp 

ρsol 

⋅ c 

sol 

psol 

The heat transfer coefficient in the gas phase was found by using the 

Chilton-Colburn analogy. The value of the heat transfer coefficient depends 

of the gas density, diffusivity, heat capacity and thermal conductivity. 

The mass balance for gas phase is described by the equation of 

consuming carbon dioxide: 

∂C 

∂t 

∂C 

∂z 

( C − H ⋅ ) 

AG AG = −vG 

⋅ − E ⋅ KG 

⋅ au 

⋅ AG CO C 2 A 

(5) 

(6)


Absorption column parameters: 

Height 

Diameter 

Specific area 

Input flows: 

Gas flow 

Liquid flow 

Feed composition: 

Gas 

MEA solution 

∂T ∂T h⋅a ∂ ∂ ⋅ 

H 

CO2 

( ) 

G G u 

=−v G⋅ − ⋅ TG −TL 

t z cp ρ 

G G 

Table 1. Parameters of the model 

H = 1.4 m 

D = 0.075 m 

au= 440 m 2 /m 3 

FG = 2*10 -3 m 3 /s 

FL = 3.42*10 -6 m 3 /s 

YA = 0.11 (molar); 

CB = 30 wt. % 

Input feed temperatures TL=293 K; TG=298 K 

Density: 

CO2 (at 0 °C and 1 atm) [8] 

Gas mixture [9] 

Aqueous alkanolamine solution [10] 

CO2 solubility according with Henry’s law [10] 

Diffusion coefficient of CO2 in water [10] 

Diffusion coefficient of CO2 in amine solution [10,11] 

Diffusion coefficient of CO2 in gas [9] 

Viscosity: 

ρCO2=1.963 kg/m 3 

ρgas =Σ(yi⋅ρi) kg/m 3 

D 

= 10 

CO2 

, w 

MEA solution [9] ( ) 

Thermal conductivity: 

CO2 [8,9]* 

Air [8,9]* 

Specific heat capacity (J/kgK): Cp= b⋅T + cT -2 +a 

MEA [8] 

H2O [9] 

CO2 [8] 

1140 

(5.3−0.035 ⋅CB0 − ) 

TL 

= 

2. 

35 ⋅10 

−6 ⋅ 

e 

⎛ μH 

⋅ ⎜ 

⎝ μ 

kg/m 3 

atm⋅m 3 /kmol 

⎛ 2119 ⎞ 

⎜ − 

⎟ 

⎝ TL 

⎠ 

γ 

m 2 /s 

DCO 

Am sol = D 

2 , . CO2 

, w 

O ⎞ 2 

⎟ 

MEA ⎠ 

m 2 /s 

3/ 2 

−5 ⎛ po ⋅T 

⎞ 

DCO 

= 1.38⋅10 ⋅⎜ 

m 

2 

⎟ 

⎝ p ⋅To 

⎠ 

2 /s 

ln 

298.3 

0.16 19.1 L T 

μMEA 

− 

= ⋅CMEA − ⋅e Pa⋅s 

λCO2=8⋅10 -5 ⋅TG-0.0071 W/m⋅K 

λair=7⋅10 -5 ⋅TG+0.005 W/m⋅K 

a b c 1411.264 4.7151 0 

4185 0 0 

983.24 0.2605 -1.86⋅10 -7 

*the equation has been obtained via regression of data taken from literature 

24.23 

T 

ρ = 2.45⋅ C + 919.13⋅ e L 

BO 

(7) 

41

42 


The partial differential equations of the mathematical model were 

transformed in ordinary ones, by discretization. The mathematical model was 

solved using MATLAB software package (version 2006) to evaluate the rate 

of CO2 absorption changes with respect to the MEA concentration decrease 

in the aqueous solution. 

For modeling and simulation of the carbon dioxide absorption in packed 

columns, the parameters presented in the Table 1 were used. 

Model simulation results 

The simulation of the mathematical model of CO2 absorption into 

aqueous mono-ethanolamine solution show the evolutions (both in time and 

space) of temperature, and concentration in the liquid and gas phase, in the 

gas – liquid absorption column. Some of the most representative simulation 

results are presented in the figures below. 

Figure 2. CO2 concentration (in gas phase) along the column height 

Figure 3. MEA concentration profile along the column height


Figure 4. CO2 concentration (in liquid phase) profile along the column height 

In Figure 3 an increasing in the concentration of MEA in liquid phase 

respectively a decrease of CO2 in gas and liquid phase against column height 

is observed. Because the rate of reaction is directly proportional to carbon 

dioxide concentration in the liquid phase and the concentration of MEA, the 

rate of reaction increases versus the column height. Figure 2 shows the 

decreasing of the CO2 concentration in the gas phase due to the chemical 

absorption [12, 13]. 

The temperature profile along the column is presented in Figure 5 

(chemical absorption of carbon dioxide is an exothermic process). 

Figure 5. Temperature profile along the column height 

43

44 


The simulation results were compared with data collected from 

literature [3,12,14] and data obtained (by authors) from simulation of the CO2 

absorption process using ChemCAD, in the same operational conditions, in 

order to validate the application developed for modeling and simulation of 

the absorption process. 

Table 2. Comparison of data obtained with developed model and 

ChemCAD software, FL=3,42⋅10 -6 m 3 /s, FG=2⋅10 -3 m 3 /s. 

Parameter MATLAB CHEMCAD 

Input Concentration of MEA 

in liquid phase [wt%] 

30.00 30.00 

Output Concentration of MEA 

in liquid phase [wt%] 

3.11 2.64 

Input molar fraction of CO2 in 

gas phase [-] 

0.110 0.110 

Output molar fraction of CO2 

in gas phase [-] 

0.0168 0.0100 

Input temperature of the liquid 

flow [K] 

293.00 293.00 

Output temperature of the 

liquid flow [K] 

323.40 324.78 

Input temperature of the gas 

flow [K] 

298 298 

Output temperature of the 

gas flow [K] 

296.20 296.76 

The good correlation between the results of developed model and 

ChemCAD model data indicates the accuracy of developed chemical 

absorption model described above in predicting actual gas-absorption 

performance of absorption packed column. 

The dynamic behavior of the absorption column was also investigated 

in the presence of the typical process disturbances (e.g. gas and liquid flow 

variation). Knowledge of the behavior in time to the occurrence of disturbance 

is important for the process control design and process optimization [9]. 

Some of the most representative results are presented in the Figures 6 to 8. 

Figure 6 shows the effect of the liquid flow increasing on liquid flow’s 

temperature when appear a perturbation. The temperature of the liquid flow 

at input, at the top of the column, is 293 K and it is increasing to 319.9 K 

at output, at the bottom, because of exothermic reaction (reaction heat is 

65000 kJ/kmole).


Figure 6. Variation of liquid’s temperature (FL = FL + 20 %) 

Figure 7. Variation of CO2 molar fraction in gas phase (FG=FG + 20 %) 

Figure 8. Variation of MEA height concentration (FL = FL + 20 %) 

45

46 


Figure 7 shows the effect of the gas flow increasing on CO2 concentration 

in gas phase. The presence of process perturbations determine a significant 

increasing of CO2 output concentration. Figure 8 shows the MEA concentration 

evolution in time and column’s height if the liquid flow increasing. All these 

stationary and dynamic parameters profiles along the packed absorption 

column are valuable information for process optimization and to develop a 

control strategy for the process. 

The developed mathematical model will be tuned against experimental 

data for adjusting the parameters and then it can be used for pre-screening 

and designing practical CO2 absorption process in various conditions 

(laboratory, pilot and industrial size, different solvents and column internals, 

various operational conditions etc.). 

CONCLUSION 

A mathematical model was developed for analyzing carbon dioxide 

absorption rate and understanding of micro level interaction of various processes 

going inside de reactor (packed absorption column). Simulation of carbon 

dioxide gas – liquid absorption process in mono-ethanolamine (MEA) was 

done using Matlab software. 

The evolutions of the process parameters (liquid and gaseous flows, 

composition of the streams, temperatures) were studied during the carbon 

dioxide absorption process. For verification of the developed model, the 

simulation results were compared with ChemCAD model simulation. 

The mathematical model and the simulation results proved to be a 

reliable tool for process analyzing and are useful for making preliminary 

calculations of experimental parameters for the column absorption. Also, the 

developed model can be used to determine optimal operation conditions 

and to design the control system. 

NOTATION 

au specific area (m 2 /m 3 ) 

A gaseous reactant (CO2) 

b stoichiometric coefficient of B 

B liquid reactant (MEA) 

CA concentration of dissolved gas A (kmol/m 3 ) 

CAG concentration of A in gas (kmol/m 3 ) 

CB concentration of reactant B (kmol/m 3 ) 

CB0 initial concentration of reactant B (kmol/m 3 ) 

cp,CO2 specific heat capacity of CO2 (J/mol*K) 

cp,G specific heat capacity of gas (J/mol*K) 

cpsol specific heat capacity of solution (J/mol*K)


D column’s diameter (m) 

DCO2,i diffusion coefficient of CO2 in substance i (m2/s) 

E enhancment factor (-) 

FG flow rate of gas (m 3 /s) 

FL flow rate of solution (m 3 /s) 

h heat transfer coefficient in gas phase (J/s*m 2 *K) 

H column’s height (m) 

HCO2 Henry’s constant [-] 

k reaction rate constant (m 3 /mol⋅s) 

kL liquid film mass transfer coefficient (m/s) 

KG overall mass transfer coefficient (m/s) 

po normal pressure (1 atm) 

t time (s) 

TG gas phase temperature (K) 

TL liquid phase temperature (K) 

To normal temperature (273 K) 

vL axial velocity of liquid film (m/s) 

vG velocity of gas film (m/s) 

yCO2 molar fraction of CO2 in gas (-) 

z axial coordinate (m) 

∆rH heat of reaction (J/kmol) 

ρg density of gas (kg/m 3 ) 

ρsol density of solution (kg/m 3 ) 

μi viscosity of component i (Pa⋅s) 


This work has been supported by Romanian National University 

Research Council (CNCSIS) through grant no. 2455: “Innovative systems for 

poly-generation of energy vectors with carbon dioxide capture and storage 

based on co-gasification processes of coal and renewable energy sources 

(biomass) or solid waste”. 

REFERENCES 

1. Intergovernmental Panel on Climate Change (IPCC), 4-th Assessments Report, 

Climate Change, 2007, www.ipcc.ch. 

2. International Energy Agency (IEA), Greenhouse Gas Program (GHG), “Oxycombustion 

process for CO2 capture from power plants”, Report 2005/9, 2005. 

3. F.A. Tobiesen, “Modeling and Experimental study of Carbon Dioxide Absorption 

and Desorption”, Ph. Thesis, 2007, chapter 1, 3 and 8. 

47

48 


4. G. Bozga, O. Muntean, “Reactoare chimice – Reactoare eterogene”, vol. II, Editura 

Tehnică, Bucureşti, 2000. 

5. Akanksha, K.K. Pant, V.K. Srivastava, Chemical Engineering Journal, 2007, 

133, 229. 

6. J. Gabrielsen,“CO2 Capture from Coal Fired Power Plants”, Ph. Thesis, 2007, 

chapter 3. 

7. H. Dang, G.T. Rochelle, National Conference on Carbon Sequestration, Washington 

DC, May 14-17, 2001. 

8. R.H. Perry, D.W. Green, “Perry’s Chemical Engineers’ Handbook”, 1999. 

9. C.F. Pavlov, P.G. Romankov, “Procese şi aparate în ingineria chimică”, Editura 

Tehnică, Bucureşti, 1981. 

10. R. Maceiras, E. Alvarez, M.A. Cancela, Chemical Engineering Journal, 2008, 

137, 422. 

11. G.F. Verstege, L.A.J. Van Dijck, W.P.M. Van Swaaij, Chemical Engineering 

Journal, 1996, 144, 113. 

12. A. Aboudheir, P. Tontiwachwuthikul, R. Idem, Chemindix, 2007, CCU/09. 

13. D.R. Olander, A.l.Ch.E. Journal, 1960, 6, 223. 

14. F. A. Tobiesen, H. F. Svendsen, A.l.Ch.E. Journal, 2007, 53 (4), 846.


RED PEPPER POWDER COLOR MEASUREMENT BY USING AN 

INTEGRATING SPHERE AND DIGITAL IMAGE PROCESSING 

EUGEN DARVASI AND LADISLAU KÉKEDY-NAGY a 

INTRODUCTION 

ABSTRACT. A home made integrating sphere and a commercial digital 

camera were used to calculate the color characteristics of 16 pepper powder 

samples. Similar data for the samples were calculated from reflectance 

spectrum data measurements too, as reference ones. The color parameters 

calculated with the two methods agree within the experimental errors. 

Keywords: red pepper, color, integrating sphere, CIELa*b*, RGB, digital 

image 

Different varieties of pepper (Capsicum annuum L. ) are largely grown 

plants for human alimentary consumption. The fruit of the plant is consumed 

as fine grounded powder, called paprika, added to different foods. The young 

fruits are green, during the ripening their color gets red due to the deterioration 

of the chlorophyll and formation of red, orange and yellow colored pigments. 

About thirty different pigments contribute to the paprika color, mainly from the 

class of carotenes (β-carotene, α-carotene, lycopin) and oxygenated carotenes, 

also known as xanthophylls (capsanthin, capsorubin, zeaxanthin, violaxanthin, 

β-cryptoxanthin). The predominant carotenoid in paprika is capsanthin, exclusive 

to Capsicum species [1,2]. Paprika powder as well as the concentrated paprika 

extract, called oleoresin, is used mainly as natural colorants added to the 

foodstuffs, in cosmetics and less as spice or vegetable [3,12]. Since paprika is 

used as a coloring agent, its market value depends partly on the red color, 

the color being considered the main quality factor of it. More intensive the 

red color more valuable the powder. 

The paprika red color and tone depend on many factors such as: the 

pepper brand [4], the growing conditions (soil, temperature, light and humidity), 

the processing and storage conditions (harvest, milling, dehydration, heat, 

light, packing etc. )[5,6,10]. On the other hand, during processing and storage, 

carotenoids are susceptible to degradation, oxidation and isomerization, due 

to the influence of light, temperature, heat, oxygen, enzymes, metals, etc. fact, 

which led to the color change, mainly to the redness loss of the powder [7- 

9,11,12]. 

a Babeş-Bolyai University, Faculty of Chemistry and Chemical Engineering, Department of Analytical 

Chemistry, Str. Arany János 11, 400028 Cluj-Napoca, Romania

50 

EUGEN DARVASI, LADISLAU KÉKEDY-NAGY 

The color of the peppers can be evaluated from three different 

aspects: surface color, extractable color and carotenoids profiles, the first 

two being the standard quality evaluation in the spice industry. The surface 

color measurements are used to specify colors perceived by the human eye. 

The visual evaluation of the color depends to a great extent on the individual’s 

eye color perception and training level, the evaluator momentary state of mind, 

the physical conditions of the observation (light source, illumination and 

background conditions, etc. ). So, each person for the same sample and 

observation condition will interpret the described color a little differently, the 

verbal descriptions could be difficult, imprecise and sometimes confusing. 

In order to avoid the subjective, peculiar color assessment several optical 

instruments, systems and techniques were developed for objective color 

determinations of different objects, including pepper, either fruit or grinded 

powder. The early instruments measured the reflectance of the uniformly 

illuminated samples (at three different wavelengths or in the whole visible 

spectral domain) and the RGB and CIELa*b* values were calculated from 

the reflectance data [13-15]. The data allow the precise calculation of the color 

parameters of the sample, but only the average ones. Meanwhile many 

attempts were made in order to replace the classic illumination and detection 

system. These consists in the replacement of the integrating sphere with a 

light source which allows diffuse, uniform illumination (as new type of discharge 

tubes, halogen tungsten lamps with diffuser and multidirectional illumination 

etc. ) as well as the replacement of the vacuum photo detectors with calibrated 

CCD cameras [16-21]. The use of such camera has several advantages over 

the vacuum tube detectors, as: eliminate the use of the monochromators or 

filters, allows the determination of color parameters of each individual point 

of the surface, the digital image could be easily saved and processed by 

proper PC software programs. 

In principle other digital optical devices, as webcams or still cameras, 

could be used as photo detectors, their use are not come yet into general use. 

To our best knowledge the use of commercially available digital camera for 

determination of the color parameters of foods, including pepper powder, was 

not reported. The aim of our work is to evaluate the features of an integrating 

sphere and digital camera as photo detector system for the determination 

of color characteristics of grounded pepper powder. 

INSTRUMENTATION 

The experiments were carried out with a home-made integrating 

sphere by casting of gypsum cubic block, with a spherical inner hole having 

a diameter of 26 cm (Fig. 1).

RED PEPPER POWDER COLOR MEASUREMENT BY USING AN INTEGRATING SPHERE … 

Figure 1. The schematic drawing of the integrating sphere 

The inner wall was covered with BaSO4 powder (used as white 

reference material for reflectance measurement). The sphere was provided 

with 3 ports, of diameter of 5. 9 cm, each perpendicular to other. The port 

on the upper side served as observation hole for camera. In the horizontally 

disposed entrance port was fixed the box of the light source with a 20 W / 

12V halogen tungsten lamp provided from a Specol 20 spectrophotometer 

(Carl Zeiss, Germany), used as stabilized power source too. The third port, on 

the lower side, and face to face with the first, served for sample introduction. 

The samples were put on a plastic tray, and lifted upwards till they reach a 

tangential position to the sphere. Below the tray a light trap was mounted (a 

black painted plastic pipe, its lower end covered with black velvet). The tray 

has three vertical holes, which communicate with the light trap disposed 

beneath the block, serving as the black reference. The photo detector was 

a commercially available digital camera SONY DSC-H1. The shots were 

carried out in the automatic adjustment mode of the shutter speed and 

aperture. The digital image, of 5. 3 megapixels, was saved as . jpg file and 

transferred to the Pentium III PC for processing. As white reference surface 

a BaSO4 tablet was photographed together with each pepper sample. 

The SFA-01 VIS spectrophotometric module (ICIA, Bucharest, Romania) 

with a concave diffraction grating and a CCD detector was used for the plot 

of reflectance spectra of the samples in the spectral domain of 400-700 nm, 

and 45/0 o geometry. The spectrum was recorded in 512 points and saved as a 

. txt file. The color parameters were calculated from these data and served 

as reference values for comparison. 

51

52 


CONVERSION OF COLOR SPACES 

From the reflectance (R) data the CIE XYZ tristimulus value, the CIELa*b* 

color coordinates and C, h* color attributes have been calculated [22]. The 

RGB values of each camera pixel were converted into device-independent 

CIE XYZ tristimulus values, using a conversion matrix as shown in Equation1. 

p 

p p 

q q q 

r r r 

Using a set of paired RGB and CIE XYZ values obtained with the 

digital camera and reflectance measurements, a cost function was defined 

and then minimized to obtain an optimized conversion matrix for our system 

(Fig. 2) [23]. 

Figure 2. The optimization scheme of the conversion matrix 

The cost function for p : 

(1) 

(2)


The solution, which minimizes this function, is: 

Where: 

Similarly, we calculated the optimized second and third row vectors, 

(q1 q2 q3) and (r1 r2 r3), respectively. 

SAMPLING AND SAMPLE HANDLING 

Thirteen different blend and origin of packed red pepper powders 

were purchased from the commerce and three samples from individual farmers. 

The powders were used as purchased, about 0. 5 g was introduced into a small 

plastic cup, gently pressed in order to obtain a compact smooth surface. The 

sample cups were introduced in the spectrophotometer and the spectrum 

has been drown, than it was placed in the integrating sphere and the digital 

photo was taken. The brand (as indicates the label on the original package) 

and the appropriate code used in this paper are presented in the Table 1. 

Table 1. The pepper samples brand and codes 

Nr. Brand / origin Code 

1 Kalocsai - Csemege PS1 

2 Szegedi - Rózsa PS2 

3 Házi arany - Édesnemes PS3 

4 Fuchs – Édesnemes PS4 

5 Fuchs – Boia de ardei iute PS5 

6 Fuchs – Boia de ardei dulce PS6 

7 Galeo - Boia de ardei iute PS7 

8 Kotányi – Boia de ardei dulce PS8 

9 Cosmin – Boia de ardei dulce PS9 

10 Clever – Boia ardei dulce PS10 

11 Apahida – Boia de ardei dulce PS11 

12 Echom – Boia de ardei dulce PS12 

13 

Home made-1 – Boia de ardei dulce 

(Gheorgheni-area) 

PS13 

14 


(Războieni- area) 

PS14 

15 


PS15 

(Cluj-Napoca area ) 

16 Kamis Chili – Boia de ardei iute PS16 

17 Idem 1 PS17 

18 Idem 1 PS18 

(3) 

53

54 


PROCEDURE 

First the white, the black and the color values of the digital image was 

corrected by using the ColorPilot 4. 80. 01v software package (Two Pilots, 

USA, Germany, Russia). The corrected image was then processed with the 

ImageJ 1. 37v software program (Wayne Rasband, National Institute of Health, 

USA), the RGB parameters of the selected area were determined. Then the 

tristimulus values of XYZ and the CIELa*b* chromatic coordinates and of C, h* 

color attributes have been calculated. 


DETERMINATION OF THE PEPPER POWDER COLOR 

PARAMETERS BASED ON THE REFLECTANCE SPECTRA 

The typical reflectance spectra of five selected pepper samples are 

represented in the Figure 3. 

R (%) 

50 

45 

40 

35 

30 

25 

20 

15 

10 

5 

0 

PS2 

PS3 

PS6 

PS8 

PS12 

PS15 

400 450 500 550 600 650 700 

Wavelength (nm) 

Figure 3. The reflectance spectra of five selected pepper samples


The samples reflect the light only over 500 nm. A group of spectrum 

(samples PS3 and PS6) exhibits a continuous, monotone increase of reflectance 

versus the wavelengths. For the other group of spectra (PS2, PS8, PS12 and 

PS15) the presence of local maxima and minimum are the characteristic 

feature of the increasing reflectance values versus the wavelength. 

The local maxima appear at ~ 600 nm and at ~ 640 nm, respectively, 

the minimum appears at the 670 nm, despite of the pepper brand. In the 

later cases, the spectra look like a superposed molecular spectrum of three 

different colorants in the 530 – 700 nm spectral domain. This fact suggests 

the presence of three, yellow and red, colorants in the pepper powder in 

higher concentrations. 

Based on the reflectance data the calculated color characteristics of 

the samples are summarized in the Table 2. 

Table 2. The CIE color attributes based on the reflectance data 

Sample 

X Y Z 

Color characteristics 

L a b C h 

PS1 11. 665 9. 401 2. 442 36. 74 21. 36 34. 27 40. 38 58. 07 

PS2 14. 440 10. 738 2. 515 39. 14 29. 39 37. 84 47. 91 52. 16 

PS3 14. 137 11. 309 2. 767 40. 10 23. 36 37. 64 44. 30 58. 17 

PS4 7. 357 4. 998 1. 509 26. 73 29. 11 25. 41 38. 64 41. 11 

PS5 8. 210 5. 874 1. 554 29. 09 26. 87 29. 01 39. 54 47. 19 

PS6 15. 929 11. 953 2. 635 41. 14 29. 62 40. 40 50. 10 53. 75 

PS7 12. 665 9. 326 2. 039 36. 60 28. 88 37. 33 47. 20 52. 27 

PS8 10. 462 7. 041 1. 732 31. 90 33. 39 32. 05 46. 28 43. 83 

PS9 8. 608 6. 095 1. 751 29. 65 27. 98 27. 99 39. 58 45. 01 

PS10 14. 911 11. 032 2. 564 39. 63 30. 11 38. 32 48. 74 51. 84 

PS11 14. 660 11. 020 3. 660 39. 61 28. 68 31. 04 42. 26 47. 26 

PS12 12. 586 9. 300 2. 135 36. 56 28. 55 36. 43 46. 28 51. 91 

PS13 16. 938 11. 932 2. 288 41. 11 35. 48 43. 01 55. 75 50. 48 

PS14 17. 550 11. 921 2. 310 41. 09 38. 90 42. 81 57. 84 47. 74 

PS15 12. 839 9. 859 2. 305 37. 59 25. 80 36. 81 44. 95 54. 97 

PS16 15. 970 12. 397 2. 861 41. 84 26. 85 39. 99 48. 17 56. 12 

PS17 13. 322 10. 801 2. 893 39. 24 21. 85 35. 29 41. 50 58. 23 

PS18 12. 480 10. 078 2. 628 37. 98 21. 69 35. 00 41. 18 58. 21 

DETERMINATION OF THE PEPPER POWDER COLOR 

PARAMETERS BASED ON THE DIGITAL IMAGE 

In this case, only the color corrected pictures were taken into the 

consideration, the pixels representing the sample surface points on each photo 

was processed further. RGB values of each camera pixel were converted 

into device-independent CIE XYZ tristimulus values using a conversion matrix 

as shown in Equation1. The calculated color characteristics of the samples 

based on the digital image are summarized in the Table 3. 

55

56 

EUGEN DARVASI, LADISLAU KÉKEDY-NAGY


The CIELa*b* characteristics and C, h* color attributes (Fig. 4) show 

that the data are close to that obtained from the reflectance spectrum, as 

reference data. The differences are within the error of determinations. We 

can conclude, that the use of integrating sphere and digital still camera is 

suitable system for the precise determination of the color characteristics, 

and based on it, the quality assessment of different pepper powder brands. 

Figure 4. The calculated Lightness, Chroma and Hue values of the 

pepper samples based on the reflectance spectra and the digital image 

57

58 


CONCLUSIONS 

The pepper samples exhibit different color and tone which could be 

emphasize by measuring the color characteristics. The reflectance spectra 

reveal the presence of there different colorants in some pepper powders. 

The differences in the reflectance spectrum appear as a tone change on 

the digital photo, which appears clearly as a difference in color characteristics. 

The color parameters calculated with the two methods agree within the 

experimental errors. As final conclusion, we can reveal, that the use of 

integrating sphere and digital still camera is suitable system for the precise 

determination of the color characteristics, and based on it, the quality 

assessment of different pepper powder brands. 

REFERENCES 

1. J. Deli, Gy. Tóth, Z. Lebensm. Unters. Forsch., 1997, 205, 388. 

2. M. Jarén-Galán, M. I. Mínguez-M., J. Agric. Food Chem., 1999, 47, 4379. 

3. Z. Niu, J. Fu, Y. Gao, F. Liu, Intern. J. Poultry Sci., 2008, 7, 887. 

4. R. Gomez, J. E. Pardo, F. Navarro, R. Varon, J. Sci. Food Agric., 1998, 77, 268. 

5. L. O. Vračar, A. N. Tepić, B. L. Vujičić, S. Šolaja, Acta Periodica Technologica, 

2007, 38, 53. 

6. M. Krajayklang, A. Klieber, P. R. Dry, Postharvest Biology and Technology 2000, 

20, 269. 

7. A. N. Tepić, B. L. Vujičić, Acta Periodica Technologica, 2004, 35, 59. 

8. B. Dobrzański, jr., R. Rybczyński, Int. Agrophys., 2002,16, 261. 

9. M. Alvarez-Ortí, R. Gómez, J. E. Pardo, J. Food Agric. Environ., 2009, 7, 16. 

10. N. Staack, L. Ahrné, E. Borch, D. Knorr, Proceedings-RELPOWFLO IV, 2008, 68. 

11. B. Dobrzański, jr., R. Rybczyński, Res. Agr. Eng., 2008, 54, 97. 

12. R. Gómez-Ladrón de Guevara, J. E. Pardo-González, R. Varón-Castellanos, 

F. Navarro-Albaladejo, J. Agric. Food Chem., 1996, 44, 2049. 

13. T. Huszka, M. Halász-F., Gy. Lukács, Hungarian Sci. Instruments,1985, 60, 43. 

14. M. Drdak, G. Greif, P. Kusy, Nahrung, 1989, 33, 737. 

15. M. Drdak, L. Sorman, M. Zemkova, A. Schaller, Confructa, 1990, 25, 141. 

16. A. Szepes, PhD Thesis, 2004, Budapesti Közgazdaságtudományi és Államigazgatási 

Egyetem. 

17. L. Baranyai, PhD Thesis, 2001, Szent István Egyetem, Budapest. 

18. Zs. H. Horváth, C. Hodúr, Int. Agrophys., 2007, 21, 67. 

19. Zs. H. HORVATH ; Acta alimentaria, 2007, 36, 75. 

20. K. León, D. Mery, F. Pedreschi, Food Res. Int., 2006, 39, 1084.


21. D. Bicanic, E. Westra, J. Seters, S. van Houten, D. Huberts, I. Colic-Baric, 

J. Cozijnsen and H. Boshoven, J. Phys. IV France, 2005, 125, 807. 

22. Gy. Lukács, Szinmérés, Műszaki Könyvkiadó, Budapest, 1982. 

23. Chang-Seok Kim, Moon Ki Kim, Byungjo Jung, Bernard Choi, Wim Verkruysse, 

Myung-Yung Jeong, J. Stuart Nelson, Lasers in Surgery and Medicine 2005, 37, 

138. 

59


MATHEMATICAL MODELING FOR THE CRYSTALLIZATION 

PROCESS OF HYDROXYAPATITE OBTAINED BY 

PRECIPITATION IN AQUEOUS SOLUTION 

VALENTINA R. DEJEU a , BARABÁS RÉKA a , POP ALEXANDRU a , 

BOGYA ERZSÉBET SÁRA a , PAUL-ŞERBAN AGACHI a 

ABSTRACT. In the paper are presented the theoretical and experimental 

results concerning the influence of pH and temperature over the rate of the 

transformation: beta-whitlockite (beta-tricalcium phosphate) in hydroxyapatite 

(HAP). From the data of the kinetic studies were calculated the values of 

the rate constants and activation energies at pH = 8.5 – 12. Based on the 

obtained values for the activation energies it has been established that the 

process of transformation of beta-whitlockite in HAP could be described 

using a combined macrokinetic mechanism: transfer – mass transformation. 

The mathematical of the process has been elaborated and the numerical values 

of the constants from the mathematical model were determined. Based on 

the proposed model process simulations were made and the results show 

that the obtained values fit well with the experimental data, which confirms 

the model validation. 

Keywords: hydroxyapatite, macrokinetic mechanism, mathematical model 

INTRODUCTION 

Due to its excellent properties like high biocompatibility, atoxicity and 

bone integration, HAP was accepted as biomaterial for biomedical purposes 

since 1920 and nowadays has many applications in medicine [1, 2, 3, 4]. 

Factors that affect the rate of adsorption of HAP in the human body 

are mostly determined by its chemical reactivity which depends on many 

different factors like: composition, particle size, defects, porosity, surface 

area and crystallinity [5]. On the other hand, these properties are greatly 

influenced on the route and conditions under which HAP is produced [6, 7]. 

HAP powder can be prepared in a variety of ways. The reactions could 

be classified in two categories: 

a - solid phase reaction 

b – precipitation 

a Universitatea Babeş-Bolyai, Facultatea de Chimie şi Inginerie Chimică, Str. Kogălniceanu 

Nr. 1, RO-400084 Cluj-Napoca, Romania, vdejeu@chem.ubbcluj.ro

62 

V.R. DEJEU, R. BARABÁS, AL. POP, E.S. BOGYA, P.-Ş. AGACHI 

Although many methods have been developed to synthesize apatites, 

the most prevalent method in preparation is still precipitation because of its 

ease in operation and tailoring composition, as well as in scaling up for mass 

production [8]. 

Among the precipitation methods, one of the most widely used is the 

precipitated method, uses as reactants calcium nitrate, bi - ammonium 

phosphate and ammonia solution. 

The chemical reaction, which describes the process, is: 

10Ca(NO 3) 2⋅ 4H2O+ 6(NH 4) 2HPO4 + 8NH4OH→ (1) 

→ Ca 10(PO 4 ) 6(OH) 2 + 20NH4NO3 + 6H2O In the literature are no quantitative data concerning how the main 

parameters of the synthesis (pH of the reaction environment and temperature) 

influence the rate of transformation of beta-whitlokite in hydroxyapatite. In 

this paper are presented the experimental results regarding the influence of 

the reaction environment, pH and temperature on the rate of transformation 

of beta-whitlokite in hydroxyapatite. These results obtained from the kinetic 

studies are the base for the mathematical model of the transformation of 

beta-whitlokite in hydroxyapatite. 

MATHEMATICAL MODELING OF THE PREPARATION PROCESS OF 

HYDROXYAPATITE 

The duration of the HAP obtaining procedure doesn’t identify with the 

duration of the chemical reaction (1), because the chemical reaction between 

the calcium nitrate and bi - ammonia phosphate is an ionic reaction with a high 

rate. On the other hand, because the solubility of the reaction product is very 

low, the critical supersaturation will be quickly achieved which makes the 

separated solid phase (beta-whitlockite) be a metastabile one and in time 

will be converted in HAP. 

In the second stage, after the chemical reaction is finished, the forming 

process of the new phase nuclei (HAP) will take place. That means that the 

obtaining process of HAP takes place in two stages, the total duration of the 

process being the sum of each stage. 

The scheme for the obtaining process of HAP by precipitation can be 

presented as in Figure 1. 

Figure 1. The scheme for the forming process of hydroxyapatite

MATHEMATICAL MODELING FOR THE CRYSTALLIZATION PROCESS OF HYDROXYAPATITE … 

This scheme suggests that the process of obtaining HAP could be 

realized after one of the following macrokinetic mechanisms: 

a. mass transformation (chemical reaction – nuclei formation and 

growing) 

b. mass transfer 

c. combined model (mass transfer - mass transformation) 

Because the rate of the chemical reaction is very high, the most probable 

macrokinetic mechanism for the obtaining process of HAP is a combined one, 

which includes not only processes of mass transformation but also processes 

of mass transfer (a, b). This combined mechanism could be presented directly 

using the balance equation (2) of the final reaction product (HAP): 

p 

n = n + n ⋅N 

(2) 

' ' ' 

A1 A1 [ ]l A1 

In the first stage of the reaction beta-whitlockite will be formed with 

metastable character, which will be transformed in time in HAP, in constant 

conditions of supersaturation. 

In these conditions the balance equation (2) becomes: 

p 

n = n ⋅N 

(3) 

' ' 

1 1 

A A 

p 

p 

p p 

dn = n ⋅ dN + N dn (4) 

A A p p A 

' ' ' 

1 1 1 

p 

dn ' ' 

A dN dn 

1 p p 

A1 

= n ' ⋅ + N ⋅ 

A 

p 

1 VT ⋅dτ VT ⋅dτ VT ⋅d τ 

(5) 

Using the relationships which express the number of moles of HAP 

p 

from one particle n ' and the total number of moles of HAP ( n ' ) from the 

A1 

A1 

total volume of the new formed phase ( V []n): 

m ' 

p A1 

n ' = =ρ ' ⋅ v p ; A1 A1 M ' 

A1 

equation (5) will be: 

n ' 

A1 =ρ ' ⋅V 

A []n 

1 

(6) 

dV[]n dNp dvp 

ρ ' ⋅ =ρ ' ⋅ + N ⋅ρ ' ⋅ 

A1 A p 

1 A1 

V ⋅dτ V ⋅dτ V ⋅d τ 

(7) 

T T T 

If we note the volume of the new phase with V []n and we considered 

that the total volume V T is the sum of the two phase volumes: 

VT = V[ ]m + V[ ]n = V[ ]m + v[ ]n ⋅V 

T and = []n V 

v[]n 

V : 

T 

63

we obtain: 

64 


V 

T 

V 

= 

1−v []m 

[]n 

dv[]n dNp dvp 

= (1−v []n) ⋅v p ⋅ + (1−v []n) ⋅Np⋅ dτ V ⋅dτ V ⋅d τ 

(9) 

[]m []m 

Equation (9) represents the nuclei forming and growing combined 

macrokinetic model, where the first right term expresses the rate of the 

nuclei formation and the second one the growing rate. 

When the process is developing after this macrokinetic model and 

because the reaction rate is higher than the germs formation rate, solutions 

with high supersaturation will be formed. 

MATHEMATICAL DESCRIPTION OF THE PROCESS 

For the mathematical description of the process according to the 

macrokinetic model presented in equation (9) we must refer to the geometrical 

shape of the macro particles and to the τ signification ( τ - the time to reach 

the final value of the volume fraction v []n in case of the new phase). 

Considering that the macroparticles are spherical, equation (9) can 

be written as follows [9]: 

dv[]n 4 dNp N 

3 p 2 dr 

= πr ⋅ + ⋅4πr ⋅ 

(1−v ) ⋅dτ 3 V ⋅dτ V d τ 

(10) 

[]n []n []m 

dNp 

where: w3 

= 

V ⋅d τ 

and = dr 

w2 

d τ 

. 

[]n 

Because the process is developing at constant supersaturating, it is 

proposed that not only the rate of the germs formation but also the rate of 

transformation-growing are constant. Based on this hypothesis results that: 

and equation (10) becomes: 

r 

Np = w3 ⋅V []n ⋅τ; 

( w 2 = ) (11) 

τ 

dv 16 

= π⋅w2⋅w3 ⋅τ 

(1−v ) ⋅dτ 3 

[]n 

[]n 3 3 

(8) 

(12)


where w3 expresses the germs forming rate and w2 is their linear growing. 

When w3 and w2 are constant, equation (12) can be written as: 

dv[]n 

3 

= K3 

⋅τ 

(1−v []n) 

⋅dτ The integration of equation (13) in the limits: 

(13) 

τ= 0; V[]n = 0; τ=τ ; V[]n = V []n 

Leads to the equation (14): 

4 

−K3 ⋅τ 

η= 1−e (14) 

In conditions of forming and growing needle like HAP macroparticles: 

l 

vp = l; w 2 = , equation (10) becomes: 

τ 

dv[]n 

= 2⋅w2 ⋅w3 ⋅τ 

(1−v []n) 

⋅dτ which by integration: 

(15) 

2 

−K2 ⋅τ 

η= 1−e (16) 

In the case that the particles don’t grow any more, w2=0, equation (10) 

becomes: 

dv[]n 

= vg⋅w3 ( 1−v[]n) ⋅dτ Integration of equation (17) leads to: 

(17) 

− 1⋅τ 

η= − K 

1 e (18) 

Equations (13), (15) and (17) allow the calculation of the time for the 

fraction v []n≡η 

of the new formed phase (HAP) to reach a certain value 

(within 0÷1). To use these equations experimental measurements are needed, 

which allow the validation of the kinetic mechanism and also the numerical 

values for K1, K2 and K3. 


The results presented in Table 1 describe the evolution in time of 

the transformation of beta-whitlockite in hydroxyapatite. Data form Table 1 

were used for the determination of K1, K2 and K3 constants from the (14), (16) 

and (18) equations. For this purpose these three equations were plotted: 

−ln(1 −η ) = f( τ) 

, Figure 2 ( a – c). 

65

66 

ln(1-η) 

ln(1-η) 

ln(1-η) 

14 

12 

10 

7 

6 

5 

4 

3 

2 

1 

0 


20 C 

0 0.5 1 1.5 2 2.5 3 3.5 

x 10 4 

-1 

time [sec] 

8 

6 

4 

2 

0 1 2 3 4 5 6 7 8 9 

x 10 8 

0 

time [sec 2 ] 

7 

6 

5 

4 

3 

2 

1 

20 C 

20 C 

0 

0 1000 2000 3000 4000 5000 6000 7000 

time [sec 4 ] x10 -13 

Figure 2. a: -ln(1- η) −τ 

; b: 

ln(1-η) 

ln(1-η) 

ln(1-η) 

10 

8 

6 

4 

2 

0 

50 C 

0 0.5 1 1.5 2 2.5 3 3.5 

x 10 4 

-2 

time [sec] 

a 

b 

14 

12 

10 

8 

6 

4 

2 

50 C 

0 1 2 3 4 5 6 7 8 9 

x 10 8 

0 

time [sec 2 ] 

10 

9 

8 

7 

6 

5 

4 

3 

2 

1 

50 C 

0 

0 1000 2000 3000 4000 5000 6000 7000 

time [sec 4 ] x10 -13 

c 

-ln(1- η) −τ 

; c: 

2 

4 

-ln(1- η) 

−τ


Table 1. The experimental results of the transformation grade of beta-whitlockite 

in time and at 20 0 C and 50 0 C for different pH values 

pH 

Time 

[s] 

η 

0 3600 7200 10800 14400 18000 21600 25200 28800 32400 

8.5 

20ºC 

50ºC 

0 

0 

0.003 

0.005 

0.009 

0.025 

0.021 

0.052 

0.042 

0.090 

0.063 

0.142 

0.092 

0.190 

0.120 

0.250 

0.158 

0.328 

0.191 

0.390 

9.1 

20ºC 

50ºC 

0 

0 

0.055 

0.092 

0.215 

0.330 

0.415 

0.590 

0.620 

0.800 

0.775 

0.910 

0.880 

0.971 

0.945 

0.990 

0.981 

1.000 

0.991 

- 

9.7 

20ºC 

50ºC 

0 

0 

0.068 

0.102 

0.240 

0.340 

0.470 

0.610 

0.670 

0.815 

0.820 

0.920 

0.920 

0.977 

0.962 

0.990 

0.980 

1.000 

1.000 

- 

10.2 

20ºC 

50ºC 

0 

0 

0.101 

0.136 

0.350 

0.450 

0.620 

0.740 

0.820 

0.910 

0.940 

0.978 

0.981 

0.996 

0.996 

1.000 

1.000 

- 

- 

- 

11.3 

20ºC 

50ºC 

0 

0 

0.110 

0.157 

0.370 

0.490 

0.650 

0.790 

0.840 

0.931 

0.950 

0.982 

0.986 

0.997 

0.996 

1.000 

1.000 

- 

- 

- 

12 

20ºC 

50ºC 

0 

0 

0.119 

0.187 

0.400 

0.560 

0.680 

0.840 

0.860 

0.960 

0.960 

0.995 

0.989 

1.000 

1.000 

- 

- 

- 

- 

- 

From the slope were determined the numerical values of the K 

constants, presented in Table 2. 

Tablel 2. The values of K constant 

20 0 C 50 0 pH 

C 

K1 K2 K3 K1 K2 K3 

8.5 6.583E-06 2.032E-10 1.920E-19 1.513E-05 4.671E-10 4.412E-19 

9.1 1.510E-04 4.649E-09 4.392E-18 2.480E-04 7.655E-09 7.232E-18 

9.7 1.740E-04 5.372E-09 5.075E-18 2.620E-04 8.089E-09 7.641E-18 

10.2 2.410E-04 8.356E-09 9.915E-18 3.340E-04 1.158E-08 1.374E-17 

11.3 2.600E-04 9.036E-09 1.072E-17 3.830E-04 1.329E-08 1.577E-17 

12 2.450E-04 9.728E-09 1.494E-17 3.980E-04 1.579E-08 2.425E-17 

In order to identify the mathematical model which describes the 

process the process simulation was needed. For this purpose the numerical 

values of the K1, K2 and K3 from Table 2 were used and different times were 

calculated relating to the transformation grade of beta-whitlockite in HAP and 

compared with the experimental results. The results obtained at 20 0 C and 

50 0 C for pH=9.7 are presented in Figure 3 (the simulations were made for all 

pH values, but in this article are presented only the relevant simulation results). 

The analysis of the results show a very good fit of the experimental 

results with those calculated for equation (16), where the errors for the majority 

of the results don’t exceed 2%. In the mathematical models expressed by 

equations (14) and (18) the deviations from the experimental values are high. 

67


At pH = 8.5 the proposed model fits well with experimental data only until 

20 % conversion (at 20 0 C) and until 40 % (at 50 0 C). 

η 

1 

0.9 

0.8 

0.7 

0.6 

0.5 

0.4 

0.3 

0.2 

0.1 

exp20 

teo20 

exp50 

teo50 

pH=9,7 

0 0.5 1 1.5 2 2.5 3 

x 10 4 

0 

time [s] 

Figure 3. The variation of conversion in time at 20 0 C and 50 0 C at pH = 9.7 

CONCLUSIONS 

Based on the kinetic study regarding the transformation process of 

beta-whitlockite in HAP the mathematical model of this process was 

established. The implementation of the model for three particular cases led 

to equations, which were used to calculate the numerical values of K1, K2 and 

K3 constants, based on experimental data. The process simulation using the 

mathematical model indicate that this could be described using the equation: 

− ⋅τ 2 

K2 

η =1-e , which corresponds to a combined macrokinetic mechanism: 

2+ 

3- 

transfer (transfer of Ca and PO 4 ions towards the HAP crystal surface) – 

mass transformation (formation and growing of HAP nuclei) 


For the synthesis of HAP and the kinetic study for the transformation 

of beta-whitlockite in hydroxyapatite, reactants of analytical purity were used: 

calcium nitrate tetra hydrate, bi-ammonia phosphate and ammonia solution 

28 %.The concentration of the calcium nitrate solution was 0.5 mol/L and 

by adding ammonia solution the pH was turned to the value of 8.5. The 

concentration of the bi-ammonia phosphate solution was 0.3 mol/L and to 

this solution was added the necessary amount of ammonia to reach the pH 

values presented in Table 1. In all the experiments the phosphate solution 

68


was added in drops over the calcium nitrate solution. The ammonia losses were 

avoided by fixing a closed ascending refrigerator and a hydraulic closing to the 

reaction vessel. The temperature was measured and maintained constant 

with a thermostat. From time to time, during the reaction, samples were 

taken, filtered, washed with distillated water and dried at 105 0 C until constant 

weight. The dried samples were crushed into fine powders with d p

NOMENCLATURE 

70 


I 1 integrated intensity of pure phase in mixture 

0 

I 1 integrated intensity of pure phase 

m mass of HAP (kg) 

' 

A1 

μ μ 

1 2 

, mass absorption coefficients 

w mass fraction of phase 

1 

n mols of HAP formed in reaction (mol) 

A1 

n mols of HAP formed in the liquid phase (mol) 

A1[] 

l 

n mols of HAP in one macroparticle (mol) 

p 

A1 

N number of macroparticles (-) 

p 

2 

r particle radius (m) 

l particle length (m) 

w 3 rate of germs forming (m/s) 

w 2 rate of germs growing (m/s) 

K rate constant (-) 

V []n the volume fraction of the new phase (m 3 ) 

V []m the volume fraction of the new phase (m 3 ) 

V T total volume of the new formed phase (m 3 ) 

Greek symbols 

ρ ' 

A 1 

molar density (kmol/ m3 ) 

τ time (s) 

v p stoechiometric coefficient of macroparticles (-) 

v stoechiometric coefficient of the new phase (-) 

[]n


STUDY OF THE CHROMATOGRAPHIC RETENTION OF SOME 

NEW ORGANOSELENIUM AND ORGANOTELLURIUM 

COMPOUNDS CONTAINING INTRAMOLECULAR 

INTERACTIONS BY HPTLC 

SILVIA LENUŢA DUNCA a , MONICA KULCSAR a , ANCA SILVESTRU a , 

CRISTIAN SILVESTRU a , COSTEL SÂRBU a 

ABSTRACT. The chromatographic behaviour of some new metal complexes 

of organoselenium and organotellurium compounds containing intramolecular 

interactions, were investigated by means of different HPTLC systems with 

polyamide, cellulose, normal and modified silica gel thin-layers and various 

organic solvents of relatively high polarity. Recommendable phase system for 

the separation of metal-complexes is a combination of fluorescent polyamide 

thin-layer with a methanol-water (8:2 v/v) mixture when well-defined compact 

spots come out and migrate. Dark zones appeared on fluorescent green 

background under UV lamp (λ = 254 nm). RF values were determined by 

using the one-dimensional ascending technique and modelling by using different 

molecular descriptors calculated by using efficient software. It has been 

concluded that a successful analysis will be executable for the compounds 

studied with the possibility of modelling the chromatographic retention. 

Keywords: organoselenium and organotellurium compounds, lipophilicity, 

TLC, QSRR, MLR, PCA, PCR 

INTRODUCTION 

Quantitative structure-activity relationships (QSAR) describe how the 

molecular structure, in terms of descriptors – lipophilic, electronic and steric – 

affects the biological activity of a compound [1-4]. Similarly, quantitative structureretention 

relationships (QSRR) relate these descriptors to chromatographic 

retention. Finally, the quantitative retention-activity relationships (QRAR) imply 

that conclusions concerning biological activity can be based on chromatographic 

experiments [5-11]. In this regard of QRAR it is considering that the same basic 

intermolecular interactions determine the behaviour of chemical compounds in 

both biological and chromatographic environments. As a consequence, the 

a Babeş-Bolyai University, Faculty of Chemistry and Chemical Engineering, Arany Janos 11, 

400028 Cluj-Napoca, Romania; E-mail: costelsrb@yahoo.co.uk

SILVIA LENUŢA DUNCA, MONICA KULCSAR, ANCA SILVESTRU, CRISTIAN SILVESTRU, COSTEL SÂRBU 

chromatographic approach has been quite successful, for example, in duplicating 

Log P data derived by traditional “shake-flask” technique or other procedures. 

The relationships themselves are usually based on correlation analysis. 

Another form of computational analysis used for the correlation of 

chemical or biological activity and chromatographic retention with different 

molecular descriptors are Multiple Linear Regression (MLR) [11-13], Principal 

Component Analysis (PCA) [14-16], Partial Least Squares (PLS) [17-19], or 

Artificial Neural Networks (ANN) [20-22]. In the case of PCA and PLS, for 

example, starting from a multidimensional space described by different variables, 

a quantitative model is derived that transforms the axes of the hypersystem. 

The first principal component (PC1) defines as much of the variation in the data 

as possible. The second principal component (PC2) describes the maximum 

amount of residual variation after the first PC has been taken into consideration, 

and so on. By using only a limited number of PCs, the dimensionality of the 

data space is reduced, thereby simplifying further analysis. 

In this paper we discuss and apply three multivariate regression 

methods to develop comparative studies and to provide a QSAR-QSRR model 

for the characterization and classification of some new organo-selenium and 

organotellurium compounds with potential applications for asymmetric synthesis 

in organic and organometallic chemistry, catalytic antioxidant activity, enzyme 

mimics and chemotherapeutic agents. 

PRINCIPAL COMPONENT ANALYSIS 

Principal components analysis (PCA) is also known as eigenvector 

analysis, eigenvector decomposition or Karhunen-Loéve expansion. Many 

problems from chemistry and other scientific fields are strongly related to PCA. 

The main purpose of PCA is to represent in an economic way the location 

of the samples in a reduced coordinate system where instead of m-axes 

(corresponding to m characteristics) only p (p < m) can usually be used to 

describe the data set with maximum possible information. 

Principal component analysis practically transforms the original data 

matrix (Xnxm) into a product of two matrices, one of which contains the 

information about the objects (Snxm) and the other about the variables (Vmxm). 

The S matrix contains the scores of the n objects on m principal components 

(the scores are the projection of the objects on principal components). The V 

matrix is a square matrix and contains the loadings of the original variables 

on the principal components (the loadings are the weights of the original 

variables in each principal component). 

Moreover, it may well turn out that usually two or three principal 

components provide a good summary of all the original variables. Loading 

and respectively score plots are very useful as a display tool for examining 

the relationships between characteristics and between compounds, looking 

for trends, grouping or outliers. 

72

STUDY OF THE CHROMATOGRAPHIC RETENTION OF SOME NEW ORGANOSELENIUM … 

MULTIPLE LINEAR REGRESSION 

Multiple linear regression (MLR) is an extension of simple linear 

regression consisting of two or more independent variables (e.g. chemical 

descriptors or properties) and a numeric dependent variable (e.g. chromatographic 

retention index). MLR attempts to model the relationship between the independent 

variables and a response variable (R) by fitting a linear equation to observed 

data in the following equation: 

k 

R = ao + ∑ ai 

xi 

i= 

1 

, (1) 

where ao, ai are the estimated regression parameters. 

PRINCIPAL COMPONENT REGRESSION 

Principal component regression (PCR) is a two-step multivariate 

calibration method: in the first step, a principal component analysis of the data 

matrix X is performed. The measured or calculated variables (e.g. descriptors) 

are converted into new ones (scores on latent variables). This is followed by a 

multiple linear regression step, MLR, between the scores obtained in the 

PCA step and the characteristic R to be modelled. 


By reducing the number of features from 9 original descriptors (including 

retention indices Rf and Rm) to three principal components (latent variables), 

the information preserved is enough to permit a primary examination of the 

similarities and differences between descriptors and organoselenium and 

organotellurium compounds. The contribution of the first component represents 

42.81% of the total variance and a two components model accounts for 

68.52% of the total variance. The first three components reproduce 

approximately 81% of the total variance and the first six even 99.31%, and 

the eigenvalues become negligible after the seventh component. 

All the statements above are well supported by the 2D- and 3Drepresentations 

of the loadings (Figures 1 and 2). The projection of the 3Drepresentation 

gives a more complete pattern: it is clear, for example, that 

the majority of the descriptors considered in this study form two close clusters: 

the first one includes M, MW, V and MR, the second one encompasses 

CLogP, H and GS; Rf /(Rm) and GS appear more or less as outliers. 

The scatter plot of scores onto the plane defined by PC1 and PC2 

(Figure 3) and in the space described by PC1, PC2 and PC3 (Figure 4) shows 

interesting results. Two clusters appear to be well defined and in a good 

agreement to the structure of compounds: one of them corresponds to the 

73


compounds 3, 4, 5, 6, 8, 12, 13, 14, 17, 18 and 19 (the largest molecules in 

the series) in the above right part of the graph, the second include the group of 

fluorine derivatives (9, 10, 15, 16), with the exception of compound 1, 2, 7 

and 11, is located in the middle-bottom of the graph. 

74 

PC2 : 19.42% 

1.0 

0.5 

0.0 

-0.5 

-1.0 

MR MW M 

V 

GS 

R f 

CLogP 

H 

AS 

R m 

-1.0 -0.5 0.0 

PC1 : 49.10% 

0.5 1.0 

Figure 1. PC1 and PC2 loading plot of the autoscaled data in Table 2 

R f 

MR MW MV 

GS 

AS 

CLogP 

H 

Figure 2. PC1, PC2 and PC3 loading plot of the autoscaled data in Table 2 

R m


PC2 

3 

2 

1 

0 

-1 

-2 

-3 

2 

6 3 

7 10 9 5 

13 8 

4 

12 14 17 8 

4 18 19 

12 14 17 

18 19 

11 

-4 

-5 -4 -3 -2 -1 0 1 2 3 4 5 

Figure 3. PC1and PC2 score plot of the autoscaled data in Table 2 

7 

9 

15 

PC1 

14 

5 6 3 

12 17 

19 

4 18 

Figure 4. PC1, PC2 and PC3 score plot of the autoscaled data in Table 2 

1 

8 

16 

75


In order to describe the relationship between the chromatographic 

retention indices of the tested compounds (RMo and b, intercept and slope 

values, in equation 7) and the calculated structural descriptors, a multivariate 

regression analysis was performed. By forward stepwise multiple regression 

analysis, the following high-quality regression equations were obtained: 

RMo = 0.119 - 0.404MR + 0.264CLogP + 0.006M (2) 

(r = 0.9153, n = 19, F = 26, p < 0.0000, s = 0.225) 

Rf = 0.500 + 0.173RM - 0.105CLogP - 0.003M (3) 

(r = 0.8894, n = 19, F = 19, p < 0.0000, s = 0.105) 

where n is the number of compounds, r the correlation coefficient, F the F-test 

value, p is the significance level of the all equation and s is standard error of 

estimates. The F and p values of equation (2) and (3) show that the multiple 

regression equations are very significant having high correlation coefficient 

and relatively small s values. 

The results suggest also that the molecular refractivity and partition 

coefficient seem to be dominant in the retention mechanism and, as a 

consequence, control the lipophilicity of the investigated compounds. 

For the PCR method, the original 7 descriptors were used for the 

selection of the optimum number of factors (principal components) by using 

also the statistics discussed above. The obtained multiple regression equations 

are also highly significant: 

CONCLUSIONS 

Rm = -0.247 - 0.774PC6 + 0.091PC1 + 0.184PC3 + 0.226PC4 + 

0.857PC7 + 40.179PC8 - 0.111PC5 (4) 

(r = 0.9360, n = 19, F = 11, p < 0.0003, s = 0.229) 

Rf = 0.599 + 0.353PC6 - 0.103PC4 - 0.033PC1 - 0.337PC7 - 

0.054PC3 - 23.392PC8 + 0.038PC5 (5) 

(r = 0.9557, n = 19, F = 17, p < 0.0000, s = 0.079) 

Correlation obtained between chromatographic retention indices 

and structure descriptors for organoselenium and organotellurium compounds 

are high significant and might be used to predict the retention behaviour 

and, as a consequence, the lipophilicity of other members of the series. By 

comparing the multivariate regression methods used in this study, the forward 

stepwise MLR appeared to be the most effective in predicting retention indices 

for the investigated compounds. The molecular refractivity and the partition 

coefficient seem to be dominant in the retention mechanism and hence 

these descriptors control the lipophilicity. 

76



The chromatographic behaviour of the compounds were investigated 

by means of different HPTLC systems with polyamide, cellulose, normal and 

modified silica gel thin-layers and various organic solvents of relatively high 

polarity. Recommendable phase system for the separation of metal complexes 

is a combination of fluorescent polyamide thin-layer with a methanol-water 

(8:2 v/v) mixture when well-defined compact spots come out and migrate. Dark 

zones appeared on fluorescent green background under UV lamp (λ = 254 nm). 

Glass HPTLC plates (20 x 20 cm) were obtained from Macherey-Nagel (Düren, 

Germany) and methanol for chromatography was supplied from Reactivul 

(Bucharest, Romania). Solutions in chloroform of each compound (Table 1) 

Table 1. Chemical structure of the investigated organometallic compounds 

NMe 2 

Se 

1 

NMe2 

Se 

Me2N Se S PR2 

S 

R = Ph (4), OPr i (5) 

8 

NMe2 

SeCl 

Te S P(OPr i )2 

12 

S 

Te S PMe 2 

R 

R = Me (16), H (17) 

S 

N 

Me 

NMe2 Me2N N 

Se 

Se 

NMe2 Me2N 2 

NMe2 

Se S CNMe2 

Se 

6 

9 

E E 

S 

Se N 

E = Te (13), Se (14) 

E S PPh2 

S 

E = Te (18), Se (19) 

N 

Me 

Me 

NMe 2 

NMe2 

Se CH 2C(OH)Me 2 

3 

Se S PMe2=NPPh2=S 

7 

NMe2 

Te S PR2 

S 

R = Ph (10), OPr i (11) 

Te 

15 

Te 

Me 

77


were prepared at a concentration of approximative 1 mg mL -1 . Chromatograms 

were developed by ascending technique at room temperature (∼20 0 C); the 

developing distance being 10 cm. After being developed, the dried plates were 

examined under UV lamp (λ = 254 nm). The RM values of each compound 

were obtained by using the following well-known equation 

RM = log (1/Rf -1) (6) 

The RM values are measured at several compositions of binary mobile 

phase systems and linearly extrapolated (interpolated) on the basis of the 

relationship between the RM and the mobile phase organic modifier as was 

described by the TLC adapted Soczewiński-Wachtmeister equation: 

RM = RM0 + bC (7) 

where RM0 indicates the extrapolated value to the pure water as mobile phase 

and it is the HPTLC descriptor most frequently used into QSAR analysis. b is 

frequently associated to the specific surface area of the stationary phase, while 

C represents the volume fraction of the organic modifier in the mobile phase. 

The specific surface area is considered an alternative descriptor of lipophilicity. 

Table 2. The descriptors and retention indices computed for the organo-selenium 

and organotellurium compounds investigated in this paper 

78 

Nr. M H CLogP MR AS GS V Rf Rm 

1 428.03 0.97 4.71 10.93 69046 160.38 584.13 0.726 -0.423 

2 542.14 1.44 4.38 14.45 3680 377.84 650.25 0.973 -1.562 

3 287.08 0.97 2.36 7.563 155 324.16 486.52 0.873 -0.839 

4 463.01 0.97 8 12.95 21 353.86 603.18 0.455 0.079 

5 427.03 0.97 6.38 11.02 3281 362.82 573.15 0.647 -0.264 

6 334.01 0.97 3.51 9.013 187 330.52 506.56 0.652 -0.273 

7 538.03 0.97 9.51 15.19 186 415.57 712.31 0.729 -0.430 

8 248.98 0.97 1.76 6.046 164 261.12 381.15 0.737 -0.447 

9 538.11 0.97 5.54 14.1 129 352.33 588.69 0.799 -0.598 

10 638.09 0.97 3.68 14.84 685 353.81 591.81 0.925 -1.089 

11 477.02 0.97 5.45 11.39 115180 353.89 556.98 0.650 -0.269 

12 419.96 0.97 5.61 9.628 2918 320.02 487.19 0.554 -0.094 

13 413.89 0.97 3.19 8.154 123 307.05 471.81 0.390 0.194 

14 313.91 0.97 5.04 7.412 125 299.23 462.66 0.436 0.112 

15 441.92 1.28 4.18 9.081 164 335.87 526.59 0.316 0.335 

16 345.93 1.28 3.94 7.93 224 298.46 447.89 0.334 0.300 

17 331.91 0.97 3.44 7.466 119 284.98 414.53 0.403 0.171 

18 455.94 0.97 7.24 11.56 4 331.76 544.93 0.392 0.191 

19 405.95 0.97 8.16 11.19 1 326.12 541.16 0.385 0.203


DESCRIPTION OF ORGANOSELENIUM AND ORGANOTELLURIUM 

COMPOUNDS 

The investigated organoselenium and organotellurium compounds 

were synthesized by procedures described earlier [24-26]. The molecular 

structure of the organoselenium and organotellurium compounds studied in 

this paper is depicted in Table 1. 

In order to define the character of the compound structure, the 

following descriptors available in the ChemDraw Pro program were taken into 

consideration and used as independent variables: exact mass (M), partition 

coefficient (CLogP), molar refractivity (MR), Henry’s law constant (H), surface 

area (AS), surface area (Grind)(GS), volume (V), molecular polarizability (MP). 

The obtained values are presented in Table 2. We have been computing only 

so few descriptors because the conventional software does not recognize Se 

and Te. Also, in the case of polarizability, the software does not have the 

ability to calculate the dithiophosphinates ligands. 


This work was supported by the Romanian Ministery of Education 

and Research (PNCDI-CERES Program, grant: 4-62/2004). 

REFERENCES 

1. A. Leo, C. Hansch, D. Elkins, Chem. Rev., 1971, 71, 525. 

2. R. F. Rekker, R. Mannhold, Calculation of Drug Lipophilicity, VCH, Weinheim,1992. 

3. M. Karelson, Molecular Descriptors in QSAR/QSPR, Wiley & Sons, New York, 

2000. 

4. A. Chiriac, D. Ciubotariu, Z. Simon, QSAR-Quantitative Relationships, Mirton 

Publishing House, Timişoara, 1995. 

5. R. Kaliszan, Anal. Chem., 1992, 64, 619A. 

6. A. Pyka, M. Miszczyk, Chromatographia, 2005, 61, 37. 

7. R. Kaliszan, Trends Anal. Chem., 1999, 18, 401. 

8. Q. S. Wang, L. Zhang, J. Liq. Chrom. & Rel. Technol., 1999, 22, 1. 

9. M. C. Garcia Alvarez-Coque, J. R. Torres Lapasio, Trends Anal. Chem., 1999, 

18, 533. 

10. T. L. J. Djaković-Sekulić, C. Sârbu, N. U. Perišić-Janjić, J. Planar Chromatogr., 

2005, 18, 432. 

11. J. Dai, L. Jin, S. Yao, L. Wang, Chemosphere, 2001, 42, 899. 

79


12. A. Detroyer, Y. V. Heyden, S. Carda-Broch, M. C. Garcia-Alvarez-Coque, D. L. 

Massart, J. Chromatogr. A, 2001, 912, 211. 

13. C. Sârbu, D. Casoni, M. Dărăbanţu, C. Maiereanu, J. Pharm. Biomed. Anal., 

2004, 35, 213. 

14. C. Sârbu, S. Todor, J. Chromatogr. A, 1998, 822, 263. 

15. C. Sârbu, K. Kuhajda, S. Kevresan, J. Chromatogr. A, 2001, 917, 361. 

16. A. Detroyer, V. F. Schoonjans, F. Questier, Y. V. Heyden, A. P. Borosy, Q. Guo, 

D. L. Massart, J. Chromatogr. A, 2000, 897, 23. 

17. H. Martens, T. Naes, Multivariate Calibration, Wiley, Chichester, 1991. 

18. Y. Zhou, L. Xu, Y. Wu, B. Liu, Chemometr. Intell. Lab. Syst., 1999, 45, 95. 

19. T. Li, H. Mei, P. Cong, Chemometr. Intell. Lab. Syst., 1999, 45, 177. 

20. R. Zhang, A. Yan, M. Liu, H. Liu, Z. Hu, Chemometr. Intell. Lab. Syst., 1999, 

45, 113. 

21. R. H. Zhao, B. F. Yue, J. Y. Ni, H. F. Zhou, Y. K. Zhang, Chemometr. Intell. 

Lab. Syst., 1999, 45, 163. 

22. Y. Chen, D. Chen, C. He, S. Hu, Chemometr. Intell. Lab. Syst., 1999, 45, 267. 

23. C. Sârbu, H. F. Pop, Fuzzy Soft-Computing Methods and Their Applications in 

Chemistry. In Reviews in Computational Chemistry. K. B. Lipkowitz, D. B. Boyd, 

T. R. Cundari, Eds., Wiley-VCH, 2004, 249-332. 

24. M. Kulcsar, A. Silvestru, C. Silvestru, J. E. Drake, C. L. B. Macdonald, M. B. 

Hursthouse, M. E. Light, J. Organomet. Chem., 2005, 690, 3217. 

25. J. E. Drake, M. B. Hursthouse, M. Kulcsar, M. E. Light, A. Silvestru, J. Organomet. 

Chem., 2001, 623, 153. 

26. C. Deleanu, J. E. Drake, M. B. Hursthouse, M. Kulcsar, M. E. Light, A. Silvestru, 

Appl. Organometal. Chem., 2002, 16, 727. 

80


LOWER RIM SILYL SUBSTITUTED CALIX[8]ARENES 

NATHAN FLEURET a,b , SEBASTIAN PAIC a , GABRIELA NEMES a , 

RALUCA SEPTELEAN a , PETRONELA PETRAR a , 

IOAN SILAGHI-DUMITRESCU *a 

ABSTRACT. New lower rim silyl substituted calix[8]arenes have been 

synthesized by reaction of the parent p-tert-butylcalix[8]arene with n-BuLi at 

low temperature followed by addition of trimethylchlorosilane/(3-chloropropyl)trimethoxysilane 

when octakis-(trimethylsilyl)calix[8]arene (3)/octakis-(n-propyltrimetoxisilan)calix[8]arene 

(4) is obtained. The new calixarene derivative 4 

has been characterized by NMR and IR spectroscopy. 

INTRODUCTION 

Keywords: calix[8]arene, octakis-(n-propyl-trimetoxisilan)calix[8]arene. 

Functionalization of calix[n]arenes is an important way in getting 

useful products for many areas of research and applications [1, 2, 3]. 

The most accessed positions for functionalizations of the parent 

calixarenes are the phenolic -OH groups of the lower rim or the p- position 

of the aromatic rings (upper rim) [4,5]. Tert-butylcalix[n]arenes (n = 4,6,8) 

can be functionalized at the upper rim either by removal of the tert-butyl 

groups thus forming calix[n]arenes which can be then halogenated (see 

references [6] and [7] for bromination), or directly by reactions of nitration or 

sulfonation [6,7]. The first step in all the processes is however the protection 

of -OH groups by o-methylation or acetylation [8,9]. 

Larger calixarenes have been successfully modified at the lower rim 

with ethers or esters formation leading to new host molecules by the 

introduction of additional functional groups [10]. These functionalizations 

allow a degree of control on the calixarene conformations and also on the 

hindrance of ring inversion since voluminous groups attached to the oxygen 

atoms increase the barrier to conformational inversion in calix[8]arenes 

compared with calix[4]arenes [11]. 

a Faculty of Chemistry, “Babes-Bolyai” University, Cluj-Napoca, RO-400028, Romania 

b IUT de Rouen, Universite de Rouen, 76821 Rouen, France

82 

N. FLEURET, S. PAIC, G. NEMES, R. SEPTELEAN, P. PETRAR, I. SILAGHI-DUMITRESCU 

In view of their potential to bind to inorganic substrates, calixarenes 

functionalized with organosilyl groups are promising precursors for new 

hybrid organic/inorganic materials. 

Literature data show that trimethylsilylated-tert-butylcalix[n]arenes 

are obtained from the brominated derivative of the corresponding calixarene, 

by lithiation with an excess of tert-butyllithium followed by silylation with excess 

chlorotrimethylsilane [12]. In case of calix[4]arenes, a tetra-trimethylsilylated 

product was obtained by this method, however only tri-trimethylsilylated 

and di-trimethylsilylated derivatives were reported for calix[6]arene and 

calix[8]arene respectively [12]. A fully substituted tert-octylcalix[8]arene octakis- 

(trimethylsilyl) ether was obtained by reaction of p-tert-octyl-calix[8]arene 

with N-O-bis[(trimethylsilyl)oxy]acetamide in solution of acetonitrile [12]. 

Herein we present a new method based on the lithiation of calix[8]arenes 

for the synthesis of octakis-(trimethylsilyl) ether of tert-butylcalix[8]arene. 

This protocol has been also applied for obtaining the new silyl substituted 

octakis(n-propyl-trimethoxisilan)calix[8]arene. 


The parent p-tert-butylcalix[8]arene has been obtained by the method 

described in literature [13] (Scheme 1). 

Scheme 1 

1 H NMR spectrum of 1 shows a specific signal at 9.66 ppm due to 

the presence of the –OH groups. Two nonequivalent doublet signals in the 

3 – 4,5 ppm range are assigned to the two nonequivalent (endo and exo) 

protons of the methylene bridges [14,15]. 

The new silyls substituted derivatives 3 and 4 were obtained via the 

lithium intermediate 2 which is stable under inert atmosphere until 0 o C 

(Scheme 2). The reaction is complete after 2 hours of stirring at a temperature 

ranging from -50 to -20 o C.


Scheme 2 

Lithium derivative 2 is added at -50 o C to a solution (THF) containing 

excess trimethylchlorosilane or (3-chloropropyl)trimethoxysilane and the 

corresponding 3 or 4 lower rim silyl substituted calix[8]arenes are formed 

(Scheme 3). 

Scheme 3 

Calix[8]arenes 3 and 4 were characterized by multinuclear NMR, 

and IR spectroscopy. 

In the 1 H NMR spectra of 3 and 4, the singlet signal corresponding to -OH 

groups of the parent calixarene is missing, proving the full substitution at the 

oxygen atoms. Furthermore, in the aliphatic range of spectrum, the characteristic 

signals for O-SiMe3 (2.19 ppm) and -O-(CH2)3-Si(OMe)3 (1.30 ppm for 

-CH2-Si(OMe)3,1.90 ppm for -CH2-CH2-CH2-, 3.53 ppm for -O-CH2-, and 

3.57 ppm for Si-(O-CH3)3 ) protons were observed. 3 and 4 derivatives show a 

83

84 


broadened signal for the bridge methylene protons, as opposed to the sharp 

AB system observed for the parent compound. This has been evidenced in 

the case of other similar substituted compounds as well [12] and it is no doubt 

due to a greater flexibility of the substituted calixarene basket. A PM3 energy 

profile calculation with Spartan06 [16] for a monosubstituted calixarene[8]arene 

(Figure 1) shows a barrier for the inversion of the methylene group, accompanied 

by a corresponding rotation of the arene ring, of less then 15 kcal/mol, comparable 

with the experimentally determined barriers of interconversion of various 

calix[4,8] arenes [10]. The calculated overall barrier is even smaller when 

two vicinal oxygen atoms are substituted with SiMe3 groups (Figure 2), so the 

observed pattern of the bridging methylene groups in the NMR spectra of 3 

and 4 has also some theoretical support. 

Figure 1. A plot of the PM3 calculated energies against the HC…CH (marked) 

dihedral angle for the monosubstituted calix[8]arene. 

The complete substitution at the lower rim has also been evidenced 

by the IR spectra of 3 and 4 where the band corresponding to the –OH 

stretching vibration at 3280 cm -1 [17, 18] is no longer present. 

Furthermore, in the 3000-2800 cm -1 range, the characteristic C-H 

stretching vibrations for the -Me groups of the -SiMe3 fragment appear as 

sharp bands. In addition, an intense band corresponding to the -O-Si- 

stretching vibration evidenced at 1148 cm -1 for 3 and 1013 cm -1 for 4 [18]. 

Most probably the substituents in 3 are oriented alternatively up and down 

relative to the mean plane of 1, (Figure 3) an arrangement which minimizes 

the repusions between the SiMe3 groups and also maximizes the π...HC 

interactions between the arene fragments and the silyl substituents.


Figure 2. A plot of the PM3 calculated energies against the HC…CH (marked) 

dihedral angle for a vicinal disubstituted calix[8]arene. 

Figure 3 shows also a linear dependence of enthalpy of formation 

with the number of SiMe3 substituents. 

Figure 3. Variation of the PM3 calculated enthalpy of formation of SiMe3 substituted 

calix[8]arenes. Only structures of the monosubstituted and fully substituted 

calixarenes are displayed. 

CONCLUSIONS 

Lithiation of the parent calix[8]arene followed by addition of the appropriate 

halogeno-alkoxisilane proves to be a convenient way to prepare lower rim fully 

substituted calix[8]arenes. 

85

86 



All experiments were carried out in flame-dried glassware under 

argon atmosphere by using high-vacuum-line techniques. Solvents were dried 

and freshly purified with SPS-5MB system. NMR spectra were recorded (with 

CDCl3 as solvent) with a Bruker Avance 300 spectrometer at the following 

frequencies: 1 H, 300.13 MHz; 13 C, 75.47 MHz (reference TMS). IR spectra 

were recorded with a Vector 22 Bruker spectrometer by direct introduction 

method and a Jasco FT/IR Specord 600 spectometer in KBr pills. Melting points 

were determined with a Wild Leitz-Biomed apparatus. Me3SiCl, Cl-(CH2)3- 

Si(OMe3)3, and BuLi were purchased from Merck. 

p-tert-butyl-calix[8]arene 1 

p-tert-butylcalix[8]arene was obtained according to the literature data [13]. 

p-tert-butylphenol (50 g; 0,33 mol), paraformaldehyde (17.5 g, 0,58 mol), 

sodium hydroxide (2,5 ml, 10 N) were dissolved in a mixture of xylenes (300 ml) 

at room temperature. This suspension was stirred for 4 h at reflux temperature 

and the water was eliminated through a Dean Stark trap. Solvents and volatile 

products were removed in vacuo, and the white precipitate formed was filtered 

and washed with water and acetone. Recrystallization from chloroform (100 ml) 

afforded pure 1. 

Yield: 46.1 g (85 %). M.p. 405–410 °C. 

1 H NMR: δ = 9.66 ppm (s, 1H, OH), 7.17 ppm (s, 2H, H-5, ArH), 3.53 ppm 

(d, 2H, CH2), 1.25 ppm (s, 9H, C(CH3)3). 

Octakis-(trimethylsilyl)-p-tert-butyl-calix[8]arene 3 

4.24 ml of a solution of nBuLi (1.6 M in hexane, 6.78 mmol) mmol) 

was added dropwise, at –80 °C, to a solution of p-tert-butyl-calix[8]arene (1 g, 

0.77 mmol) in THF (40 ml). The solution turned yellow and was stirred at this 

temperature for an additional hour. The lithium compound was then transferred 

to Me3SiCl (0.9 ml, 7.04 mmol, d = 0.85 g/ml) in THF (20 ml) cooled at –80 °C. 

After 2 hours of stirring the mixture was allowed to warm to room temperature. 

Solvents and volatile products were removed in vacuo, and the residue was 

dissolved in pentane (20 ml) to filter out the lithium salts. Recrystallization 

from chloroform (100 ml) afforded pure 3 

Yield: 0.81 g (62 %). M.p. 316 - 322 °C. 

1 H NMR: δ = 1.10, 1.27 ppm (m, 9H, C(CH3)3), 2.19 ppm (s, 9H -O-SiMe3) 

3.85 ppm (broad m, 2 JHH = 12,85 Hz, 2H, CH2), 6.92 ppm (broad s, 2H, H arom.). 

IR: 2904-2869 cm -1 (νa SiMe3), 1148 cm -1 (νa -O-Si-) 

Octakis-(n-propyl-trimethoxysilane)-p-tert-butyl-calix[8]arene 4 

4.5 ml of a solution of nBuLi (1.6 M in hexane, 7.2 mmol) was added 

dropwise, at –80 °C, to a solution of p-tert-butyl-calix[8]arene (1 g, 0.77 mmol) 

in THF (40 ml). The solution turned yellow and was stirred at this temperature


for an additional hour. The lithium compound was then transferred to a solution 

of 3-chloropropyl-trimetoxysilane (1.30 ml, 7.09 mmol, d = 1.09 g/ml) in THF 

(20 ml) cooled at –40 °C. The orange solution was a llowed to warm to room 

temperature. Solvents and volatile products were removed in vacuo, and the 

residue was dissolved into pentane (20 ml). Lithium salts were filtered out; 

compound 4 crystallized as a brick-colored solid after 4 – 5 h at – 4 °C under 

argon atmosphere. 

Yield: 1.13 g (56 %). M.p. 422 °C (decomposition). 

1 H NMR: δ = 0.80 ppm (dd, 2H, -CH2-Si(OMe)3), 1.30 ppm (m, 9H, t-Bu), 

1.90 ppm (q, 2H, -CH2-CH2-CH2-), 3.53 ppm (t, 2H, -O-CH2-), 3.57 ppm (s, 9H, 

-Si-(O-CH3)3), 7.15 ppm (m, 2H, H arom). 

13 C RMN: δ = 6.63 ppm (s, -CH2-Si(OMe)3), 8.03 ppm (s, -CH2-CH2-CH2-), 

26.21 ppm (s, -C-(CH3)3), 31.66 ppm (s, -C-(CH3)3), 33.77 ppm (s, -Ph-CH2-Ph-), 

47.14 ppm (s, -Ar-O-CH2-) 50.44 ppm (s, -(O-CH3)3), 122-126 ppm (m, meta C), 

126-132 ppm (m, ipso C), 135-145 ppm (m, para C). 

IR: 1013 cm -1 (νa Si-OMe) 


The financial support from CNMP under the project PNII-71-062 is 

gratefully acknowledged. 

REFERENCES 

1. C. D. Gutsche, J. A. Levine, J.Am.Chem.Soc., 1982, 104, 2652. 

2. Z. Asfari, V. Bohmer, J. Harrowfield, J. Vicens, Calixarenes 2001, Kluwer Academic 

Publishers, 2001, chapter 5. 

3. P. Jose, S. Menon, Bioinorg. Chem. Appl., 2007, ID 65815, 1. 

4. J. W. Cornforth, E. D. Morgan, K. T. Potts, R. J. W. Rees, Tetrahedron, 1973, 29, 

1659. 

5. D. Gutsche, Calixarenes: An introduction, 2nd edition, RSC, Thomas Graham House, 

Science Park, Milton Road, Cambridge, UK, 2008, chapter 2. 

6. S. Seiji, K. Hirosuke, A. Takashi, M. Tsutomu, S. Hirosi, M. Osamu, J. Chem. Soc. 

Perkin Trans. 1, 1989, 5, 1073. 

7. A. Arduini, A. Casnati, Macrocycle Synth., 1996, 145. 

8. C. D. Gutsche, L. G. Lin, Tetrahedron, 1986, 42, 1633. 

9. C. D. Gutsche, B. Dhawan, K. H. No, R. Muthukrishnam, J. Am. Chem. Soc., 1981, 

103, 3782. 

10. C.D.Gutsche, L. J. Bauer, J. Am. Chem. Soc., 1985, 107, 6052. 

11. C. D. Gutsche, L. J. Bauer, J. Am. Chem. Soc.,1985, 107, 6059. 

87

88 


12. F. Billo, R. M. Musau, A. Whiting, ARKIVOC, 2006, 10, 199. 

13. J. H. Munch, C. D. Gutsche, Org. Synth., 1990, 68, 243. 

14. G. Ferguson, J. F. Gallagher, M. A. McKervey, E. Madigan, J. Chem. Soc. Perkin 

Trans 1, 1996, 599. 

15. C. D. Gutsche, Calixarenes Revisited, The Royal Society of Chemistry, Thomas 

Graham House, UK, 1998, chapter 3. 

16. Spartan’06 Wavefunction, Inc. Irvine, CA, Except for molecular mechanics and 

semi-empirical models, the calculation methods used in Spartan have been 

documented in: Y. Shao, L.F. Molnar, Y. Jung, J. Kussmann, C. Ochsenfeld, 

S.T. Brown, A.T.B. Gilbert, L.V. Slipchenko, S.V. Levchenko, D.P. O’Neill, R.A. 

DiStasio Jr., R.C. Lochan, T. Wang, G.J.O. Beran, N.A. Besley, J.M. Herbert, 

C.Y. Lin, T. Van Voorhis, S.H. Chien, A. Sodt, R.P. Steele, V.A. Rassolov, P.E. 

Maslen, P.P. Korambath, R.D. Adamson, B. Austin, J. Baker, E.F.C. Byrd, 

H. Dachsel, R.J. Doerksen, A. Dreuw, B.D. Dunietz, A.D. Dutoi, T.R. Furlani, 

S.R. Gwaltney, A. Heyden, S. Hirata, C-P. Hsu, G. Kedziora, R.Z. Khalliulin, 

P. Klunzinger, A.M. Lee, M.S. Lee, W.Z. Liang, I. Lotan, N. Nair, B. Peters, 

E.I. Proynov, P.A. Pieniazek, Y.M. Rhee, J. Ritchie, E. Rosta, C.D. Sherrill, 

A.C. Simmonett, J.E. Subotnik, H.L. Woodcock III, W. Zhang, A.T. Bell, A.K. 

Chakraborty, D.M. Chipman, F.J. Keil, A.Warshel, W.J. Hehre, H.F. Schaefer, 

J. Kong, A.I. Krylov, P.M.W. Gill and M. Head-Gordon, Phys. Chem. Chem. Phys., 

2006, 8, 3172. 

17. G. Ferguson, J. F. Gallagher, M. A. Mckervey, E.Madigan, J. Chem. Soc. Perkin 

Trans 1, 1996, 599. 

18. P. J. Launer, Infrared Analysis of Organosilicon Compounds: Spectra-Structure 

Correlation, B. Arkles, G. L. Larson (eds), Petrarch Systems (section 8), 1987.


THE INTERACTION OF SILVER NANOPARTICLES 

WITH LIPOIC ACID 

OSSI HOROVITZ a , MARIA TOMOAIA-COTIŞEL a , CSABA RACZ a , 

GHEORGHE TOMOAIA b , LIVIU-DOREL BOBOŞ a , AURORA MOCANU a 

ABSTRACT. Silver nanoparticles of about 6 nm average size were prepared 

by reduction of an aqueous silver nitrate solution with sodium citrate and 

characterized by UV-Vis spectroscopy and transmission electron microscopy 

(TEM). The interaction of this aqueous colloidal silver solution with the αlipoic 

acid solution in ethanol was investigated. The changes in the optical 

spectra and the TEM images suggest the adsorption of lipoic acid molecules 

onto the silver nanoparticles and the self assembly of the particles in three 

dimensional aggregates. 

Keywords: silver nanoparticles, α-lipoic acid, UV-Vis spectra, TEM, self 

aggregation 

INTRODUCTION 

Silver is well known for its special optical properties, conferring him its 

role in photography, and for its bactericidal properties. Silver nanoparticles 

are of great interest for their various applications in surface enhanced Raman 

scattering (SERS) [1, 2], photonics [3] and photocatalysis [4], microelectronics 

[5-6], optics [7] and as antibacterial and antimicrobial agent [8, 9]. 

Thus it is not surprising that a great number of methods were devised 

for the synthesis of silver nanoparticles, both in aqueous and in biphasic systems. 

The silver source was mostly silver nitrate and the reducing agents were sodium 

citrate [10-12], sodium borohydride [13, 14], ethanol or methanol [15], polyols 

such as ethylene glycol, in the presence of poly(vinylpyrrolidone) as a polymeric 

capping reagent [16,17], or glycerol in the presence of o-phenylenediamine [18], 

formamide [13] and N,N-dimethylformamide [19-21], N-hexadecylethylenediamine 

[22], sodium acrylate [23] and even a cell filtrate from a fungus [24]. 

a Babeş-Bolyai University of Cluj-Napoca, Faculty of Chemistry and Chemical Engineering, 

Department of Physical Chemistry, 11 Arany J. Str., 400028 Cluj-Napoca 

b Iuliu Haţieganu University of Medicine and Pharmacy, Department of Orthopedic Surgery, 

47 Moşoiu T. Str., 400132 Cluj-Napoca, mcotisel.chem.ubbcluj.ro@gmail.com

90 

O. HOROVITZ, M. TOMOAIA-COTIŞEL, C. RACZ, GH. TOMOAIA, L.-D. BOBOŞ, A. MOCANU 

Since colloidal silver presents a high sensitivity toward oxygen, at times 

synthesis was achieved in an inert atmosphere [23]. The formation of silver 

nanoparticles was also favored by γ radiation [12], laser irradiation [25] or 

UV illumination [26]. 

Bioconjugates of the hemoproteins, myoglobin, and hemoglobin have 

been synthesized by their adsorption on spherical silver nanoparticles [27]. 

Silver nanoparticle-oligonucleotide conjugates were prepared, based upon 

DNA with cyclic disulfide-anchoring groups [28]. When silver nanoparticles 

functionalized with complementary DNA sequences are combined, they 

assemble to form DNA-linked nanoparticles networks. The interaction between 

silver nanoparticles and various DNA bases (adenine, guanine, cytosine, and 

thymine) was described [13]. Silver nanoparticles were adsorbed on the surface 

of natural wool, as a result of the interaction of silver with sulfur moieties 

related to the cysteine group [29]. The adsorption of aliphatic-nonpolar amino 

acids represented by L-methionine on silver nanoparticles was studied by the 

FT-SERS method [30] and a physical and chemical adsorption by -NH3 + , 

COO - , S, was identified 

In previous works, we synthesized gold nanoparticles in aqueous 

solutions and investigated their functionalization and self aggregation with 

various biomolecules, such as proteins [31] and amino acids [32-35]. 

The goal of the present investigation is to obtain silver nanoparticles 

in colloidal aqueous solution and to study their interaction with α-lipoic acid. 


The colloidal aqueous solution containing silver nanoparticles is yellow. 

The UV-visible absorption spectrum of this solution presents a well-defined 

absorption band with a maximum at the wavelength λmax = 399 nm (Fig. 1). 

This value is characteristic for plasmon absorbance for nanometric Ag 

particles. The colloidal solution is rather stable; after one month, only slight 

modifications in the spectrum were observed. 

As reported in literature, the position of λmax for silver nanoparticles is 

highly dependent on particle shape and can range from around 400 nm for 

spherical particles to near 800 nm for sharp-edged triangles. For aggregated 

particles, a red shift is observed [36]. Wavelength and shape of this band 

are also affected by various adsorbed solutes [27]. 

The size of the colloidal silver particles has been measured by TEM 

imaging (for example, Fig. 2). The particles present mostly spherical or ovoid 

shape. From the sizes of a great number of particles, measured on the TEM 

images, the following characteristics were calculated: average size (diameter): 

5.9 nm; standard deviation: 2.7 nm (extreme values 1.4 and 13.7 nm); average 

mass of a particle (considered spherical): 1.1·10 -18 g; average number of 

silver atoms in a particle: 6·10 3 .

THE INTERACTION OF SILVER NANOPARTICLES WITH LIPOIC ACID 

Absorbance, a.u. 

0,08 

0,06 

0,04 

0,02 

0,00 

-0,02 

-0,04 

399 nm 

300 400 500 600 700 800 

Wavelength, nm 

Figure 1. Optical spectrum of the colloidal silver solution 

Figure 2. TEM images for silver nanoparticles 

Figure 3. Histogram of size distribution for Ag particles 

91

92 


A histogram providing the size distribution of silver nanoparticles, 

obtained from TEM pictures, is given in Fig. 3. The predominant fraction is 

that of particles with about 7 nm diameter, but there are also important 

contributions from the particles of about 3 and 5 nm size. 

The α-lipoic acid or thioctic acid, IUPAC name: 5-(dithiolan-3-yl)pentanoic 

acid (Scheme 1) is an important biomolecule, known for its antioxidant properties, 

and is used in the treatment of various diseases [37-39]. The presence of 

the disulfide group (the ditholane ring) suggests a potential strong interaction 

with silver nanoparticles. Lipoic acid is practically insoluble in water, so its 

ethanol solution had to be used to investigate its interaction with the colloidal 

silver solution. 

Scheme 1. 

In the UV-Vis spectrum of α-lipoic acid, an absorption band with the 

maximum at 336 nm is observed (Fig. 4). Adding a small amount of the αlipoic 

acid 0.01 M solution in ethanol to the aqueous silver colloidal solution 

(volume ratio 1/7.5) strongly modifies the spectra of both solutions. The 

absorption peak of α-lipoic acid disappears, being replaced by a shoulder at 

low wavelength, and the absorption peak of the silver nanoparticles vanishes 

Absorbance, a.u. 

1,0 

0,5 

0,0 

334 nm 

399 nm 

300 400 500 

Wavelength, nm 

Ag 

Lipoic acid 0,01 M 

Ag + lipoic acid 7.5:1 

Figure 4. Optical spectra of the aqueous colloidal silver solution, of the α-lipoic acid 

solution in ethanol and of the mixture of the two solutions in the volume ratio 7.5/1


also (Fig. 4). The yellow color of the solutions fades away. These changes 

in the spectra suggest a strong interaction between silver nanoparticles and 

α-lipoic acid. 

The TEM images of the mixtures of silver nanoparticles and α-lipoic 

acid, given in Fig. 5 for different magnifications, confirm this interaction. Large 

aggregates of silver nanoparticles are seen, resulted by their self assembly 

mediated by the α-lipoic acid molecules. The adsorption of the biomolecules 

proceeds probably by means of the sulfur containing groups. The process 

should be analogous to the adsorption of n-octadecyl disulfide onto colloidal 

silver nanoparticles [40], where the formation of three-dimensional selfassembled 

monolayers was observed. 

On the other part, by introducing the lipoic acid solution in ethanol in 

the aqueous colloidal solution, an emulsion of lipoic acid is generated, and 

the silver nanoparticles seem to be trapped in the nanodrops of the organic 

phase (Fig. 5d). 

a. b. 

c. d. 

Figure 5. TEM images of silver nanoparticles with α-lipoic acid. The images bars 

correspond respectively to 5 μm (a), 2 μm (b), 200 nm (c), and 100 nm (d) 

93


CONCLUSIONS 

A stable silver colloidal solution was prepared and characterized by 

UV-Vis spectroscopy and TEM imaging. By means of these techniques the 

self-aggregation of silver nanoparticles is evidenced, the process being induced 

by α-lipoic acid. Functionalization of silver nanoparticles through biomolecules 

is important for the development of new biomaterials with implications in 

nanoscience and nanotechnology. 


The colloidal silver solution was prepared by a method adapted from 

[11]. 125 mL of 0.001 M solution of silver nitrate in water was heated until 

boiling under continuous magnetic stirring. Then 5 mL of 1% sodium citrate 

solution was added and the heating continued until the color was pale yellow. 

The solution was then rapidly cooled on ice to room temperature. The solution 

of colloidal silver particles was stored in brown bottles and kept at 4 o C. The 

silver content of the solution is 104 mg/L. 

AgNO3 was purchased from Merck (high purity above 99.5 %). The 

trisodium citrate dihydrate was obtained from Sigma Aldrich (high purity above 

99%). α-Lipoic acid (extrapure) was purchased from Jiangsu Chemical 

Company – China, and its 0.01 M solution in ethanol was prepared. All aqueous 

solutions were prepared using deionized water with resistivity of 18 MΩ·cm, 

obtained from an Elgastat water purification system. 

The UV/Vis absorption spectrum of the solutions was studied using 

a Jasco UV/Vis V-530 spectrophotometer, with 10 mm path length quartz 

cuvettes in the 190 – 900 nm wavelengths range. 

The silver nanoparticles suspension (7 μL), in the absence and in the 

presence of lipoic acid, was deposited for 30 s on the carbon coated specimen 

grid and observed with a transmission electron microscope (TEM: JEOL – JEM 

1010). TEM images have been recorded with a JEOL standard software 


This research had financial support from PN2 grant no.41-050. We 

thank chemist Ancuţa Ureche for her contribution to the experimental work. 

REFERENCES 

1. Z. Wang, S. Pan, T.D. Krauss, H. Du, L. J. Rothberg, Proceedings of the National 

Academy of Sciences of the USA, 2003, 100, 8638. 

2. P. Matejka, B. Vlckova, J. Vohidal, P. Pancoska, V.J. Baumrunk, Journal of 

Physical Chemistry, 1992, 96,1361. 

94


3. M. P. Plieni, A. Taleb, C. J. Petit, Journal of Dispersion Science and Technology, 

1998, 19 (2 and 3), 185. 

4. Z. Zhang, R. C. Patel, R. Kothari, C. P. Johnson, S. E. Friberg, P. A. Aikens, 

Journal of Physical Chemistry B, 2000, 104, 1176. 

5. R. B. Bright, M. D. Musick, M. J. Natan, Langmuir, 1998, 14, 5695. 

6. R. C. Doty, H. Yu, C. K. Shih, B. A. Korgel, Journal of Physical Chemistry B, 

2001, 105, 8291. 

7. M. Kerker, Journal of Colloid and Interface Science, 1985, 105, 297. 

8. Q. L. Feng, J. Wu, G. Q. Chen, F. Z. Cui, T. N. Kim, J. Kim, Journal of Biomedical 

Materials Research, 2000, 52, 662. 

9. I. Sondi, B. Salopek-Sondi, Journal of Colloid and Interface Science, 2004, 275, 

177. 

10. Z. S. Pillai, P. V. Kamat, Journal of Physical Chemistry B, 2004, 108, 945. 

11. P. V. Kamat, M. Flumiani, G. V. Hartland, Journal of Physical Chemistry B, 

1998, 102, 3123. 

12. A. Henglein, M. Giersig, Journal of Physical Chemistry B, 1999, 103, 9533. 

13. S. Basu, S. Jana, S. Pande, T. Pal, Journal of Colloid and Interface Science, 

2008, 321, 288. 

14. Y.-H. Wang, J. Zhou, T. Wang, Chinese Journal of Inorganic Chemistry, 2007, 

23, 1485. 

15. P.-Y. Silvert, R. Herrera-Urbina, N. Duvauchelle, V. Vijayakrishnan, K. J. Tekaia- 

Elhissen, Journal of Materials Chemistry, 1996, 6, 573. 

16. Y. Sun, Y. Xia, Science, 2002, 298, 2176. 

17. C. Chen, L. Wang, G. Jiang, H. Yu, Reviews on Advanced Materials Science, 

2006, 11, 1. 

18. M. H. Ullah, I. Kim, C.-S. Ha, Journal of Nanoscience and Nanotechnology, 

2006, 6, 777. 

19. I. Pastoriza-Santos, L. M. Liz-Marzán, Langmuir, 1999, 15, 948. 

20. I. Pastoriza-Santos, C. Serra-Rodríguez, L. M. Liz-Marzán, Journal of Colloid 

and Interface Science, 2000, 221, 236. 

21. I. Pastoriza-Santos, L. M. Liz-Marzán, Pure and Applied Chemistry, 2000, 72, 83. 

22. A. Manna, T. Imae, M. Iida, N, Hisamatsu, Langmuir, 2001, 17, 6000. 

23. I. Hussain, M. Brust, A .J. Papworth, A. I. Cooper, Langmuir, 2003, 19, 4831. 

24. K. C. Bhainsa, S. F. D'Souza, Colloids and Surfaces B: Biointerfaces, 2006, 47, 

160. 

25. J. P. Abid, A. W. Wark, P .F. Breevt, H. H. Girault, Chemical Communications 

(Cambridge), 2002 (7), 792. 

26. A. Henglein, Chemistry of Materials, 1998, 10, 444. 

27. R. T. Tom, A. K. Samal, T. S. Sreeprasad, T. Pradeep, Langmuir, 2007, 23, 1320. 

95

96 


28. J.-S. Lee, A. K. R. Lytton-Jean, S. J. Hurst, C. A. Mirkin, Nano Letters, 2007, 7, 

2112. 

29. L. Hadad, N. Perkas, Y. Gofer, J. Calderon-Moreno, A. Hule, A. Gedanken, 

Journal of Applied Polymer Science, 2007, 104, 1732. 

30. S. Li, G.-M. Zhou, D.-C. Yang, D.-N. H.-J. Yu,. X.-J. Peng, N. W. Wu, 

Spectroscopy and Spectral Analysis, 2007, 27, 711. 

31. O. Horovitz, Gh. Tomoaia, A. Mocanu, T. Yupsanis, M. Tomoaia-Cotisel, Gold 

Bulletin, 2007, 40 (3), 213. 

32. O. Horovitz, A. Mocanu, Gh. Tomoaia, L. Boboş, D. Dubert, I. Dăian, T. Yupsanis, 

M. Tomoaia-Cotisel, Studia Universitatis Babes-Bolyai, Chemia, 2007, 52 (1), 97. 

33. I. Petean, Gh. Tomoaia, O. Horovitz, A. Mocanu, M. Tomoaia-Cotisel, Journal 

of Optoelectronics and Advanced Materials, 2008, 10, 2289. 

34. L. Barbu-Tudoran, Gh. Tomoaia, O. Horovitz, A. Mocanu, M. Tomoaia-Cotisel, 

Journal of Optoelectronics and Advanced Materials, 2008, 10, 2293. 

35. A. Mocanu, I. Cernica, Gh. Tomoaia, L. D. Bobos, O. Horovitz, M. Tomoaia-Cotisel, 

Colloids and Surfaces A, Physical and Engineering Aspects, 2009, 338, 93. 

36. W. P. Hall, G. Hartland, R. Van Duyne, H. Petrova, Nanoscape, 2005, 2, 35. 

37. G. P. Biewenga, G. R. M. M. Haenen, A. Bast, General Pharmacology: The Vascular 

System, 1997, 29(3), 315. 

38. P. J. Randle, Diabetes / Metabolism Reviews, 1998, 14(4) 263. 

39. G. P. Biewenga, G. R. M. M. Haenen, A. Bast, Drug Metabolism Reviews, 1997, 

29(4) 1025. 

40. L. A. Porter, Jr., D. Ji, S. L. Westcott, M. Graupe, R. S. Czernuszewicz, N. J. Halas, 

T. R. Lee, Langmuir, 1998, 14, 7378.


OPTIMISATION OF COPPER REMOVAL FROM 

DILUTED SOLUTIONS 

FLORICA IMRE-LUCACI a , SORIN-AUREL DORNEANU, PETRU ILEA 

ABSTRACT. An experimental study concerning the copper removal from a 

simulated wastewater containing 10 ppm of Cu 2+ ions was performed in a 

continuous flow electrochemical reactor equipped with a rectangular threedimensional 

(3D) cathode made of reticulated vitreous carbon (RVC) with 

100 ppi porosity. The influence of low supporting electrolyte concentrations 

and of the electrode polarisation potential was investigated. The reactor 

performance was evaluated considering the residual copper concentration 

and the specific energy consumption. After 90 min of electrolysis, the copper 

ion concentration was reduced to less than 0.1 ppm, permitting the discharge 

of the treated solutions to the drain system. With an operating cell voltage 

around 1.8 V, specific energy consumptions close to 0.18 kWh per cubic 

meter of treated wastewater was calculated. 

Keywords: removal, copper, reticulated vitreous carbon electrode, waste 

waters 

INTRODUCTION 

Wastewaters containing heavy metal ions (HMI), such as cadmium, 

chromium, copper, gold, lead, nickel, silver, tin and zinc, are generated in 

large quantities during electroplating, manufacturing of microelectronic parts, 

mining and processing of photographic materials. For environmental and 

economic reasons, these toxic metals should be removed from wastewater 

before discharge [1, 2]. 

The maximum admitted concentrations (MAC) of heavy metals is strictly 

regulated in EU, imposing a rigorous control and treatment of wastewaters. 

These limits are in the domain 0.05 – 1 ppm for HMI [3]. In order to respect 

the MAC, the removal by cathodic deposition presents a clear, versatile and 

efficient method. 

Depending on the heavy metals concentrations, the removal process 

can be done in different manners. For high concentration (grams per litre), 

two-dimensional cathodes can be used and the content of heavy metals can 

be reduced with one order magnitude. The resulting effluent can be reused in the 

process or it can be introduced in a new stage of chemical or electrochemical 

decontamination [2]. 

a Department of Physical Chemistry, “Babes-Bolyai” University, 11 Arany Janos, 400028 

Cluj-Napoca, Romania; * fimre@chem.ubbcluj.ro

FLORICA IMRE-LUCACI, SORIN-AUREL DORNEANU, PETRU ILEA 

For low concentrations, under hundreds milligrams per litre, threedimensional 

(3D) electrodes must be used and the content of heavy metals can 

be reduced to levels that allow the discharge of the effluents in environment 

[4-11]. 

Preliminary tests concerning the removal of Cu (RCu) from diluted 

solutions of nitric acid lead to very low current efficiencies due to a very intense 

hydrogen evolution reaction (HER) but also due to the cathodic reduction of 

the nitrate and dissolved oxygen. In this context, an alternative method of 

RCu, able to minimize the specific energy consumption, was searched. 

In this work, we present our results concerning the optimization of the 

copper removal and recovery process from synthetic diluted solutions containing 

~ 10 mg/L (ppm) of Cu 2+ and Na2SO4 or NaCl at low concentrations (< 15 

mM) as supporting electrolyte. The prepared solutions were electrolyzed, 

potentiostatically and galvanostatically, in a continuous flow electrochemical 

reactor (ER), equipped with a 3D RVC electrode. The reactor performance was 

evaluated based on the final residual copper concentration (CR) and the global 

specific energy consumption (WS) for metric cube (m 3 ) of treated solution. The 

concentrations of HMI were measured by atomic adsorption spectroscopy (AAS). 


Removal studies 

The studies concerning the electrochemical copper elimination from 

dilute solutions were performed using electrolyte volumes of 250 mL. During 

the whole period of the electrolysis experiments (90 minutes), electrolyte 

samples of ~ 5 mL were taken out every 5 minutes for copper concentration 

evaluation. The working electrode current (IW.E.) and the mean cell voltage (EC) 

were used to estimate WS and the current efficiency (CE) for cooper recovery. 

Potentiostatic removal of Cu from diluted chloride solutions 

The influence of cathodic potential 

The measurements’ results concerning the influence of the cathodic 

potential (εW.E.) on EC, WS, CE and CR are presented in Table 1. These 

experiments were completed using 10 mM NaCl as supporting electrolyte 

and a volume flow rate (VF) of 50 mL/min. 

For a working electrode potential of –100 mV/RE, the system is 

swinging and it has a poor performance (low CE and high WS, respectively). 

Moreover, the oxygen evolved at the anode is reduced on the cathode, 

generating a significant parasitic current even in the absence of an intense 

copper electrodeposition process. 

The increase of the cathodic polarization potential to –200 mV/RE 

induces the enhancement of CE and, obviously, a decrease of WS. Furthermore, 

the formation of a compact cathodic deposit of Cu inhibits the oxygen reduction 

reaction (RRO) and allows achieving the desired MAC level (< 0.1 ppm Cu 

98 

2+ ).

OPTIMISATION OF COPPER REMOVAL FROM DILUTED SOLUTIONS 

Table 1. Global electrolysis parameters obtained at different cathodic polarization 

potential (t = 90 min, CNaCl = 10 mM; VF = 50 mL/min). 

εW.E. [V/ER] EC [V] WS [kWh/m 3 ] CE [%] CR [ppm] 

–0.100 2.17 0.81 2.57 0.626 

–0.200 1.79 0.18 9.18 0.060 

–0.300 2.16 0.52 3.81 0.059 

A cathodic potential of –300 mV/RE is already excessive for RCu, 

producing a porous Cu deposit and a significant increase of WS due to the 

RRO evolved on the anodes’ surface. 

In these conditions, it was decided to continue the experiments at a 

cathode polarization potential of –200 mV/RE. 

Concerning the CR, we observed that, after 90 min of electrolysis at 

εW.E. of –100 mV, only 94 % of Cu is removed, without reaching concentrations 

below 0.1 ppm. Contrarily, at εW.E. of –200 mV and –300 mV, around 99.4 % 

from the initial amount of Cu was removed, reaching CR below 0.1 ppm. 

The effect of the chloride concentration 

Due to the low salinity of the used electrolyte solutions, the corresponding 

electric conductivities were very small and determine high values of WS. In order 

to evaluate the influence of supporting electrolyte concentration, measurements 

have been made at different concentrations of NaCl, the corresponding results 

being presented in Table 2. 

Table 2. Global electrolysis parameters at different CNaCl values 

(t = 90 min, εW.E. = –200 mV/ER; VF = 50 mL/min). 

CNaCl [mM] EC [V] WS [kWh/m 3 ] CE [%] CR [ppm] 

5 2.55 0.73 4.24 0.050 

10 1.79 0.18 9.18 0.060 

15 1.53 0.12 11.37 0.138 

The increase of the NaCl concentration in the electrolyte increases 

the CE and decreases WS. Unfortunately, CR doesn't reach the desired level 

because, in the presence of high concentration of Cl - ions, the formation of 

Cu + ions complexes, [CuClx] -(x-1) , and also the low solubility of CuCl compound 

inhibit RCu [12]. As a result, the presences of chloride ions have complex 

effects and a moderate concentration (~ 10 mM NaCl) represents an optimal 

compromise that includes a reasonable salinity value. 

99

100 


The influence of volume flow rate 

In order to evaluate the positive effect of the mass transport intensification, 

several measurements were accomplished at different volume flow rates. In 

this context, the results concerning the influence of VF on RCu are presented 

in Table 3. The evolutions in time of the copper concentration at different VF 

values are also presented in Figure 1. 

Table 3. Global electrolysis parameters at different volume flow rate 

(t = 90 min, CNaCl = 10 mM, εW.E. = –200 mV/ER). 

VF [mL/min] EC [V] WS [kWh/m 3 ] CE [%] CR [ppm] 

25 1.97 0.35 5.56 0.682 

50 1.79 0.18 9.18 0.060 

75 1.90 0.29 6.05 0.069 

100 1.97 0.49 4.37 0.053 

Log ([Cu] / ppm) 

1.0 

0.5 

0.0 

-0.5 

-1.0 

V F , mL/min 

25 

50 

75 

100 

0 10 20 30 40 50 60 70 80 90 

Time / min. 

Figure 1. The evolution in time of the copper concentrations for different volume 

flow rates (εW.E. = –200 mV/RE; CNaCl = 10 mM). 

At very low VF values, the rate of RCu is also low and the Cu deposit is 

insufficient to inhibit RRO. 

The increase of the volume flow rate intensifies Cu nucleation and a 

fast grows of the deposit, which inhibits RRO. 

Anyway, the implicit increase of de oxygen quantity developed at 

the anode and its faster transport to the cathode produces a decrease of 

the CE and an increase of WS.


For volume flow rates higher than 25 mL/min, time values less than 

90 min are required in order to attain 0.1 ppm Cu residual concentration 

(see Figure 1). 

In these conditions, a flow rate of 50 mL/min is recommended. 

Potentiostatic removal of copper from sulphate solutions 

The influence of cathodic potential 

The measurements’ results concerning the influence of the cathodic 

potential on EC, WS, CE and CR are presented in Table 4. Based on preliminary 

tests (see next results), these experiments were completed using 7.5 mM 

Na2SO4 as supporting electrolyte and a volume flow rate of 50 mL/min. 

Table 4. Global electrolysis parameters at different cathodic polarization 

(VF = 50 mL/min; 7.5 mM Na2SO4). 

εW.E. [V/ER] EC [V] WS [kWh/m 3 ] CE [%] CR [ppm] 

–0.100 1.51 0.15 8.84 0.858 

–0.200 1.69 0.19 7.85 0.093 

–0.300 2.05 0.33 5.37 0.079 

–0.350 2.76 1.32 1.91 0.089 

In sulphate solutions, the increase of the cathodic polarisation to more 

negative values induces an increase of the mean value of the recorded 

current and a decrease of CE. 

At εW.E. values of –100 mV, –200 mV and –300 mV, the Cu deposit 

can inhibit RRO. For εW.E. of –350 mV, the electrochemical system becomes 

lightly unstable (swinging) and a high amount of oxygen developed on anode 

is reduced at cathode, inducing the increase of the parasitic current and, 

consequently, the decrease of CE. 

In these conditions, we decided to continue the experiments at a 

cathodic polarization potential of –200 mV/RE because, at this value, the 

concentration of the Cu decrease bellow 0.1 ppm and reasonable energy 

consumption can be obtained. 

The effect of Na2SO4 concentration 

Because Na2SO4 is more toxic for environment that NaCl, we try to use 

the minimum amount of added sulphate. In order to evaluate the influence of 

Na2SO4 on RCu, several measurements have been made at low concentrations 

of supporting electrolyte, the corresponding results being presented in Table 5. 

At low concentration of Na2SO4 (1.0 mM), the solution have a very low 

electric conductivity and, consequently, the Ws is very highly. The increase 

of Na2SO4 concentration induces a decrease of EC, but the electrochemical 

system remains unstable due to the evolution of parasitic processes. Because, 

101

102 


in the 7.5 mM Na2SO4 solution, the CR decrease under 0.1 ppm and Ws is low, 

we decide to use this concentration for further experiments. 

Table 5. Global electrolysis parameters at different CNa2SO4 values 

(t = 90 min; εW.E. = –200 mV/RE; VF = 50 mL/min) 

CNa2SO4 [mM] EC [V] WS [kWh/m 3 ] CE [%] CR [ppm] 

1.0 5.20 2.04 2.14 0.365 

2.5 1.88 0.19 8.35 0.164 

5.0 1.74 0.35 5.19 0.350 

7.5 1,69 0.18 7.85 0.093 

10.0 1.69 0.21 6.66 0.133 

The influence of volume flow rate 

In order to evaluate influence of electrolyte volume flow rate on RCu, 

measurements were accomplished at VF values of 25, 50, 75, 100 mL/min, 

the obtained results being presented in Table 6. The evolutions in time of the 

copper concentration at different VF values are also presented in Figure 2. 

Table 6. Global electrolysis parameters at different volume flow rate 

(t = 90 min, εW.E. = –200 mV/ER; 7.5 mM Na2SO4). 

VF [mL/min] EC [V] WS [kWh/m 3 ] CE [%] CR [ppm] 

25 1.42 0.11 11.78 0.727 

50 1.69 0.19 7.85 0.093 

75 1.72 0.19 7.82 0.264 

100 1.87 0.29 5.79 0.062 

Log ([Cu] / ppm) 

1.2 

0.8 

0.4 

0.0 

-0.4 

-0.8 

-1.2 

V F , mL/min 

25 

50 

75 

100 

0 10 20 30 40 50 60 70 80 90 

Time / min. 

Figure 2. The evolution in time of the copper concentration for different volume 

flow rate (εW.E. = –200 mV/RE; 7.5 mM Na2SO4)


At very low volume flow rates, RRO concurs with RCu and the Cu 

deposit is insufficient to inhibit RRO. 

Based on the increase of VF, the intensification of the mass transport 

allows Cu nucleation and a fast grow of the deposit, with positive effect on 

the RRO inhibition. Unfortunately, the implicit grow of de oxygen quantity 

evolved at the anode and its faster transport to the cathode induces a 

decrease of CE and an increase of WS. 

Based on these observations, we concluded that a volume flow rate 

of 50 mL/min represents the optimal compromise for RCu. 

Galvanostatic removal of copper from sulphate solutions 

Based on the promising results obtained in potentiostatic conditions, 

we also decided to test RCu in galvanostatic mode. 

The results concerning the influence of the imposed current on the 

RCu parameters are presented in Table 7. The measurements were done 

at current values of 25, 30 and 40 mA, at VF = 50 mL/min and using 7.5 mM 

Na2SO4 as supporting electrolyte. 

Table 7. Global electrolysis parameters evaluated for RCu in galvanostatic mode 

(VF = 50 mL/min; 7.5 mM Na2SO4). 

IW.E. [mA] EC [V] WS [kWh/m 3 ] CE [%] CR [ppm] 

25 2.22 0.44 4.24 0.164 

30 2.24 0.47 4.03 0.136 

40 2.71 0.72 3.14 0.157 

From the point of view of EC, WS and CE, a current intensity of 25 mA is 

favourable to the RCu process but it isn’t for CR. 

Using the same experiment time (90 min.), we observe that a 

volume flow rate of 50 mL/min is insufficient to reach the desired level of 

concentration (< 0.1 ppm). 

CONCLUSIONS 

The results of our researches concerning the removal of copper from 

diluted solutions allow us to take the following conclusions: 

• For the potentiostatic RCu, a cathodic potential of –200 mV/ ER, a 

volume flow rate of 50 mL/minute and concentrations of 10 mM 

NaCl or 7.5 mM Na2SO4, respectively, represent the optimal values. 

• Using the optimised parameters, MAC of Cu can be attained in the 

treated solutions. 

103

104 


• The evolution in time of the currents recorded in potentiostatic 

mode encouraged us to also test RCu in galvanostatic mode; 

• The galvanostatic tests show us that increased volume flow rates 

and/or electrolyse time are required to attain the MAC. The optimisation 

of galvanostatic operating parameters involves additional experiments. 

These promising obtained results prove the feasibility of copper removal 

from diluted solutions simulating wastewaters. 

The validation of the obtained results requires additional measurements 

at scaling-up level, using pilot installation and real wastewaters samples. 


Reagents 

The simulated solutions were prepared starting from analytical grade 

reagents (CuSO4, NaCl and Na2SO4, all from Chimopar, Romania) and doubledistilled 

water. The tested solution contained 10 mg/L Cu 2+ and different 

concentrations of NaCl or Na2SO4. 

Experimental setups 

A Plexiglas home made electrochemical reactor (ER) previously 

described [2], a Reglo-Digital peristaltic pump (Ismatec, Switzerland) and a 

HP72 potentiostat (Wenking, Germany) were used for RCu tests. 

A parallelepiped (L x W x H = 45 mm x 25mm x 25 mm) of 100 ppi 

RVC was used as working electrode. Four graphite cylindrical bars (Ф = 10 mm, 

L = 25 mm) were used as counter electrodes. Two Ag/AgCl/KClSAT reference 

electrodes (RE) were used to record the cathode’s and anode’s potentials. 

The LabView 6.1 software and a PCI 6024 E data acquisition board 

(National Instruments, USA) were used for process control. 

The AAS measurements of Cu concentration in the electrolyte samples 

were performed with an Avanta PM Spectrometer (GBC, Australia). 


The financial support within the CNCSIS Project no. 495 / 2464 / 2009 

is gratefully acknowledged. 

REFERENCES 

1. A. Al.-Shammari, S.U. Rahman, D.-T. Chin, J. Appl. Electrochem., 2004, 34, 447. 

2. S.A. Dorneanu, F.-L. Beke, P. Ilea, Studia Univ. Babes-Bolyai, Chemia, 2008, 

53, 97.


3. Romanian Government Decision no. 352 from 11.05.2005. 

4. P. Ilea, S.A. Dorneanu, Tratarea apelor reziduale cu conţinut de ioni de metale 

grele, A X Ediţie a Zilelor academice timişene, 2007. 

5. C. Ponce de Leoon, D. Pletcher, Electrochim. Acta, 1996, 41, 533. 

6. J.Y. Choi, D.S. Kim, J. Hazard. Mater., 2003, B99, 147. 

7. A. Dutra, A. Espiinola, P. Borges, Minerals Eng., 2000, 13, 1139. 

8. E. Munoz, S. Palmero, M. A. Garcia-Garcia, Electroanalysis, 2000, 10, 12. 

9. G.W. Reade, A.H. Nahle, P. Bond, J.M. Friedrich, F.C. Walsh, J. Chem. Techn. 

Biotech., 2004, 79, 935. 

10. G. W. Reade, P. Bond, C. Ponce de Leon, F.C. Walsh, J. Chem. Techn. 

Biotech., 2004, 79, 946. 

11. F.A. Lemos, L.G.S. Sobral, A.J.B. Dutra, Minerals Eng., 2006, 19, 388. 

12. S.C. Varvara, L.M. Muresan, Metode electrochimice de investigare a electrodepunerii 

metalelor, Casa Cărţii de Ştiinţă, Cluj-Napoca, 2008, 31. 

105


VOLATILE ORGANIC DISINFECTION BY PRODUCTS 

DETERMINATION IN DISTRIBUTION SYSTEM FROM 

CLUJ NAPOCA 

MELINDA-HAYDEE KOVACS a , DUMITRU RISTOIU a , SIDONIA VANCEA b , 

LUMINITA SILAGHI-DUMITRESCU c 

ABSTRACT. Chlorine is one of the most used water disinfectant agent 

used on the world. The use of chlorine in the treatment of drinking water 

has virtually eliminated waterborne diseases, because chlorine can kill or 

inactivate most microorganisms commonly found in water. The majority of 

drinking water treatment plants in Romania use some form of chlorine to 

disinfect drinking water: to treat the water directly in the treatment plant and/or 

to maintain a chlorine residual in the distribution system to prevent bacterial 

regrowth. Unfortunately it’s used results in formation of some disinfection by 

products (DBPs) that are suspected from harmful effects on humans. Such 

of disinfection byproducts are trihalomethanes (THMs). Trihalomethanes 

are a group of volatile organic compounds that can form when the chlorine 

used to disinfect drinking water reacts with naturally occurring organic matter 

(e.g., decaying leaves and vegetation). The preliminary results presented in this 

paper shown that the THMs levels from Gilau Water treatment Plants and 

Cluj-Napoca distribution system. The results have higher values in the summer 

period relative to other seasons. 

Keywords: Volatile disinfection by-products, Trihalomethanes, Chlorine, Water 

Treatment Plant. 

INTRODUCTION 

Chlorine is one of the most common disinfectant agents used in Water 

Treatment Plant (WTP) from Romania. The drinking water disinfection process 

with chlorine has been carried out since the dawn of the 20 th century to 

eradicate and inactivate the pathogens from water. In addition to inactivating 

pathogens in the source water, chlorine are also used as oxidants in drinking 

water treatment to: remove taste and odors, oxidize iron and manganese, 

a 

Babes Bolyai University of Cluj-Napoca, Faculty of Environmental Science, Str. P-ta Stefan 

cel Mare, no. 4, 400084, Cluj-Napoca, email: dristoiu@enviro.ubbcluj.ro 

b 

Garda de Mediu, Comisariatul Judetean Cluj, str. G-ral T. Mosoiu, nr. 49, Cluj-Napoca, Romania 

c 

Universitatea Babeş-Bolyai, Facultatea de Chimie, str. Arany Janos, nr. 11, 400068, Cluj-Napoca, 

Cluj-Napoca.

MELINDA-HAYDEE KOVACS, DUMITRU RISTOIU, SIDONIA VANCEA, LUMINITA SILAGHI-DUMITRESCU 

maintain a residual to prevent biological regrowth in the distribution system, 

improve coagulation and filtration efficiency and prevent algal growth in 

sedimentation basins and filters [1]. Chlorine’s popularity is not only due to 

lower cost, but also to its higher oxidizing potential, which provides a minimum 

level of chlorine residual throughout the distribution system and protects against 

microbial recontamination [2, 3]. However, in 1974, it was discovered that the 

chlorination of water resulted in the production of trihalomethanes (THMs) due 

to reaction of chlorine with natural organic matter (NOM) present in all type 

of water, referred as precursor of THMs formation [4,5]. Since 1980’ THMs have 

raised significant concern due to evidence of their adverse human health 

effects, in special cancer and reproductive disorders [6,7]. 

Because of this unwanted effects of chlorination process USEPA in 

November 29, 1978, has promulgated the first legislation (interim total THMs 

standard) to limit the concentration of total THMs (TTHMs) in drinking water. 

In Romania the regulation and monitoring of THM has became a 

current issue in connection with Romania’s entry to the UE and the fulfillment 

of the required drinking water standards. In order to minimize cancer risk, the 

Romanian has adopted the maximal permissible value fixed in the EU drinking 

water directive has been adopted by the Romanian legislation in 2002, granting 

the water companies a transition time of 10 years to meet the requested 

standards and accepting in the first 5 years a TTHM value of 150 μg/l. In 

terms of monitoring, the Romanian water law stipulates a minimal number of 

samples per year depending on the magnitude of the treatment plant. The 

local health authority carries out the monitoring. 

Hence engineers are required to minimize the concentration of THMs 

in water in the distribution system. 

Chlorine reacts with a wide variety of organics in water to give rise to 

haloform reaction and produce THMs. THMs are organohalogen compounds 

and they are named as derivates of the compound methane. THMs are formed 

when three of the four hydrogen atoms attached to the carbon atom in the 

methane compounds are replaced with atoms of chlorine, bromine and/or 

iodine [8]. THMs include chloroform – CHCl3, dichlorobromomethane – 

CHCl2Br, dibromochloromethane – CHClBr2 and bromoform – CHBr3. This 

complex reaction mechanism of THMs is controlled by parameters such as: 

concentration and type of precursors, concentration of chlorine, temperature, 

pH and time. The THMs formation process may be described by the following 

equation: 

NOM (Precursors) + HOCl → Chlorinated organic intermediates + 

HOCl → CHCl3 (1) 

108 

Natural bromide in water 

(Br - ) + HOCl → Br2 + Cl - → CHCl2Br/CHClBr2 /CHBr3 (2)

VOLATILE ORGANIC DISINFECTION BY PRODUCTS DETERMINATION IN DISTRIBUTION SYSTEM 


THM measurements in Gilau WTP from Cluj: During the THMs analysis 

the main founded THMs species detected in the water sample was the CHCl3. 

The CHCl3 concentration differed from month to month during the years. 

Usually higher CHCl3 concentration was found in the wormer season, than in 

the winter season. In Gilau WTP and distribution system the highest CHCl3 

concentration was detected in August 2007 – 81.1 μg/L. The measurement 

shows that the CHCl3 concentration increased in the distribution system. The 

chloroform concentration range in the WTP at the exit of reservoir sampling 

point was in the range 8 μg/L (March 2007) and18.75 μg/L (November 2007). 

That could be explain by the fact in the winter season in the water are present 

lower concentration of NOM than in summer season due to low temperature 

and light. From that reason less chlorine concentration are required for water 

disinfection process. During the measurements was observed also that the 

chloroform concentration increased with distance. Started from the Exit of 

the reservoir from the WTP (considered 0 point) the chloroform concentration 

increased with 30 % at the enter of the city (located at 18 km from the Gilau 

WTP) and in the city center (located at 25 – 30 km from the Gilau WTP) the 

chloroform concentration was double or almost much with three times. That 

shows the THMs concentration increased with distance – see table 1. 

Table 1. CHCl3 concentration measured in 2006 – 2008 at the Gilau WTP and Cluj. 

* Obs: Ex.R. – represent exit of reservoir, En.C. – enter of city, Cen. – center of city. 

CHCl3 concentration (μg/L) 

2007 

Month Ex.R. En.C. Cen. 

January 25.4 31.1 40.4 

February 12.0 31.6 48.6 

March 8.0 18.7 21.6 

April 8.2 29.2 33.8 

May 14.7 40.6 50.2 

June 19.2 60.2 66.3 

July 21.0 65.2 69.0 

August 28.0 69.0 81.1 

September 24.5 39.58 52.4 

October 20.1 32.8 47.9 

November 18.75 24.9 35.9 

December 9.32 12.78 21.7 

109


Factors affecting THM formation: Manny research showed that are 

several factors affecting the formation potential of THMs. Previous research 

studies have shown that the major variables that affect THM formation are: 

chlorine dose and residual, concentration and nature of NOM (mainly humic 

substances), contact time, pH, temperature of water, and the presence of 

inorganic ions like bromide [9-13]. In general, higher THM concentrations 

are expected at higher levels of the above mentioned parameters [2]. 

Increase of chlorine dose has been reported to have positive influence 

of DBPs yield. The same is true for increased concentrations of natural organic 

matter and increased temperature. The presence of bromide ion shifts the 

speciation of DBPs to more brominated analogues, while increased pH can 

enhance the formation of some categories of DBPs, e.g. THM, and inhibit the 

formation of some others, e.g. haloacetonitriles and haloketones [10-11]. 

The type of raw water also affects the THM levels. Generally, ground 

waters are naturally protected from runoff NOM, while the difference in 

occurrence of DBP precursors in river and lakes depends on geological, 

physical and environmental factors (trophic stage, watershed soil characteristics 

and land use, lake size, river flow rate, etc.) [2]. 

Chlorine dose: During the studies it was observed that one of the 

main important factor that affect the THMs formation in the distribution system 

in the four WTP studied was the chlorine dose that was applied in the WTP 

in water disinfection purposes. 

In Gilau Water Treatment Plant was added different chlorine dose 

during the year, the chlorine dose could differed day-by-day as the water matrix 

was changed due to seasons, temperature, pH and the NOM concentration 

presented in the raw water. So in Gilau WTP the company has set 2 different 

chlorine dose range, as function of season: for summer season the chlorine 

dose range that was added to water for disinfection was set between 0.7 – 

0.9 mg/ L and for winter season the chlorine dose set was in the range of 

0.5 – 0.7 mg/L. After several measurement has shown that in the period 

when higher chlorine dose was used the CHCl3 concentration increased – 

as shown in figure 1. 

Nature and Concentration of NOM: Properties of NOM play an important 

role, since activated aromatic content of NOM increases THM formation [12, 

13]. Singer (1999) in his researches shows that the THMs formation is relatively 

higher for the humic acid fraction, presumably because of the greater aromatic 

carbon content of the fraction [14]. Manny researcher works showed a linear 

relationship between chlorine consumption and the activated aromatic carbon 

content of the various humic and fulvic acid. In addition, NOM contains 

hydrophobic and hydrophilic materials, the nature and distribution of which 

may vary with different types of vegetation in the watershed and different 

species of algae in water. This results in varying influence of NOM on DBP 

formation [15]. 

110


CHCl3 (ug/L) 

90 

80 

70 

60 

50 

40 

30 

20 

10 

y = 84.145x - 64.14 

R 2 = 0.8153 

0 

1 1.1 1.2 1.3 1.4 

Cl2 dose (mg/L) 

1.5 1.6 1.7 1.8 

Figure 1. Chlorine dose as function of CHCl3 concentration in Gilau WTP 

In the WTP the presence of the natural organic matter was 

determinate with consumption of KMnO4 mg/L – CCO Mn [mgKMnO4/L] – 

colorimetric method. The presence of the organic matter in the raw water is 

determinate once at the middle of day in every day in Gilau WTP. The 

results obtained showed that in the period when the NOM presence in 

water was higher also the CHCl3 increases – see figure 2. 

CHCl3 (ug/L) 

80 

70 

60 

50 

40 

30 

20 

10 

y = 7.1384x - 24.143 

R 2 = 0.6682 

0 

7 8 9 10 

COD (mg KMnO4/L) 

11 12 13 

Figure 2. Relationship between CHCl3 (μg/L) and presence of 

natural organic mater in water 

111


Temperature/Season: When temperature increases, reactions are 

faster and a higher chlorine dose is required, leading to higher formation of 

THMs. Subsequently, THMs concentrations are expected to be higher in 

summer than in winter [11; 17; 18-21]. 

After THMs analysis in every month, the results show that in the 

winter season the THMs concentration were much lower that in the summer 

season, with almost 50 %. That could be explain by the fact that in some cases 

where the ice cover protects surface raw waters, the THM concentrations 

are lower due to lower water temperature and NOM. In these conditions, the 

chlorine demand is lower, therefore, the chlorine dose required to maintain 

adequate residual in the distribution system is also less. 

In Gilau WTPs which used as disinfectant agent the chlorine the mean 

of total trihalomethanes (TTHM) levels for summer, fall, winter, and spring 

was: 72.17 μg/L, 40.716 μg/L, 30.606 μg/L and 35.23 μg/L respectively; 

This study showed that the highest TTHM concentrations were found 

in the summer and fall seasons, and the lowest TTHM concentrations were 

present in the winter and spring. Since higher doses of chlorine were used in 

the warm summer and fall months to ensure prevention of microbiological 

problems, it was to be expected that this, in combination with warmer water 

temperatures, would lead to higher TTHMs concentration during the summer 

and fall seasons. 

pH: Several studies have been made to investigate the effect of pH 

on THMs concentrations. The studies have shown that THM concentrations 

increases with increasing pH – as shown in figure 3, that could be explain 

by the fact that THMs formation is mainly attributed to the reactions of the 

chlorinated intermediates with hydroxide ion in the presence of a small 

amount of free chlorine. 

Bromide: Recent studies have examined the relationship between 

bromide concentration in a drinking water supply and THMs formation. Based 

on the differences in bromide concentration, it is inferred that substantial 

variations in THM formation (and THM species) can be expected. Studies 

have shown that as the concentration of bromide increases, the concentration 

of TTHMs increases and more brominated THMs forms because there is 

more bromide present in the water source for the organics to react with. In 

the presence of bromide ion (Br-), more brominated and mixed chlorobromo 

derivatives are formed [22-25]. When bromide is present, chlorine in 

the form of hypochlorous acid-hypochlorite ion (HOCl-OCl-) oxidizes bromide 

ion to hypobromous acid-hypobromite ion (HOBr-OBr-). A mixture of HOCl 

and HOBr can lead to the formation of both chlorinated and brominated 

by-products [26]. 

During the analysis of the THMs species in WTPs the brominated 

THMs species were very lower all the time, the total brominated THMs species 

never exceed 10 μg/L. So in the Gilau WTP the only brominated THMs species 

112


that was found was CHCl2Br and its highest concentration was 9.69 μg/L. 

After that results we could concluded that in all WTPs the bromide concentration 

are almost inexistent or in very less concentration. 

CHCl3 concentration (ug/l) 

80 

70 

60 

50 

40 

30 

20 

10 

0 

y = 66.281x - 422.36 

R 2 = 0.4471 

6.8 6.9 7 7.1 7.2 7.3 

pH value 

Figure 3. Relationship between CHCl3 (μg/L) and pH value of water. 

CONCLUSIONS 

This study examined the trihalomethanes formation and they 

concentration in different sampling point take from different sampling points 

in Cluj-Napoca. Following conclusions were drawn based on the results of 

the measurements conducted with water samples of every month in 2007. 

Chloroform was the determinate trihalomethanes species observed 

after the chlorination of reservoir water in all of the sampled months. 

Brominated compounds of trihalomethanes wasn’t observed, just CHCl2Br 

but in very slow concentration. We can suppose for that reason, the bromine 

compounds in the raw water is very slow or almost inexistent. 

The major factor that influences the trihalomethanes formation is the 

temperature that is clearly praise in chloroform concentration measured in 

different months of year. In the months like July, August, September and 

October when the temperature was highest (21 – 24 °C water temperature) 

also the chloroform concentration was high (66.3 – 81.1 μg/L) and when the 

water temperature wasn’t high (2.8 – 3.7 °C) the chloroform concentration was 

between 21.7 – 40.4 μg/L. The major differences in temperature and pH value 

affect trihalomethanes potential formation (THMFP), with more volatile THM 

species being produced at the higher temperature and pH. Also we can 

concluded the season when the water matrix is change from the winter season 

to summer season also have a seriously influence on trihalomethanes formation. 

113


MATERIALS AND METHODS 

Sampling: Several samples were collected from different sampling 

point of the Gilau Water Treatment Plant (WTP) and distribution system from 

Cluj-Napoca – located in Cluj jurisdiction. Gilau WTP has as water sourced 

the Somesul Mic river basin and also Tarnita Storage Lake. All the samples 

were collected in every month of 2007. The sampling contains the following 

samples: 

Gilau WTP: 

- raw water, filtrated water, chlorinated water, exit reservoir – 

sampling points that are located in the WTP; 

- Sapca Verde, Beer Factory, Chemistry Faculty, Environmental 

Faculty and Public Health Institute – sampling points that 

are located in the distribution system. 

All water samples were collected and stored in 40 L vials and closed 

with Teflon lined screw cap and they were preserved with sodium thiosulfate 

(Na2S2O3) at 4 0 C until the analysis. All samples were measured between in 

1 and 7 days after sampling. 

THMs analysis: THMs were carried out by Thermo Finningan U.S.S. 

Trace GC Ultra gas chromatography with electron capture detector (GC-ECD) 

with TriPlus HS auto sampler. The analysis was made using headspace 

technique. 10 ml of sample was filled into 20 ml headspace vials and closed 

with Teflon lined screw cap. After that the samples were equilibrated in an 

oven at 60 °C for 45 minutes, 1 ml of the headspace was then injected into the 

GC (Cyanopropylphenyl Polysiloxane column, 30 m x 53 μm, 3 μm film thickness, 

Thermo Finnigan, USA). The column program was 35 °C (hold time 3 minutes), 

15 °C/minutes to 200 °C (hold time 3 minutes). The inlet was set at 200 °C. 

The standard stock solution containing 2000 μg/l of each Trihalomethanes 

from Restek (Bellefont, U.S.) was used. THM working standard solutions 

(100 mg/l, 4mg/l) were prepared through dilution of the stock solution in 10 ml 

Methanol. The calibration standards were made using that THM stock solution 

and the calibration were prepared for the range 0 – 100 μg/L in mineral water 

Izvorul Alb. 

114 

REFERENCES 

1. USEPA, 1999a, Microbial and Disinfection Byproduct Rules Simultaneous 

Compliance Guidance Manual, United States Environmental Protection 

Agency, EPA 815-R-99-015. 

2. R. Sadiq, M. J. Rodriguez, Science of the Total Environment, 2004, 321, 21. 

3. J. Huang, G. C. Smith, J. of American Water Works Association, 1984, 76, 168.


4. J. J. Rook, J. of American Water Works Association, 1976, 68, 168. 

5. T. A. Bellar, J. J. Lichtenberg, R. Krasner, J. of American Water Works 

Association, 1974, 66, 703. 

6. K. P. Cantor, C. F. Lynch, M. E. Hildesheim, M. Dosemeci, J. Lubin, M. Alavanja, 

G. Craun, Epidemiology, 1988, 9, 21. 

7. C. G. Graves, G. M. Matanoski, R. G. Tardiff, Regul Toxicol Pharmacol, 2002, 

34, 103. 

8. C. Vogt, S. Regli, Journal of American Water Works Association, 1981, 73 (1), 

33. 

9. G. L. Amy, P. A. Chadik, Z. K. Chowdhury, J. of American Water Works 

Association, 1987, 79, 89 . 

10. A. D. Nikolaou, S. K. Golfinopoulos, T. D. Lekkas, M. N. Kostopoulou, 

Environmental Monitoring and Assessment, 2004, 93, 301. 

11. A. D. Nikolaou, T. D. Lekkas, S. K. Golfinopoulos, Chemical Engineering 

Journal, 2004, 100, 139. 

12. W. E. Elshorbagy, H. Abu-Quadis, K. Elsheamy, Water Research, 2000, 

34, 3431. 

13. S. K. Golfinopoulos, G. B. Arhonditsis, Chemosphere, 2002, 47, 1007. 

14. D. A. Reckhow, P. C. Singer, R. L. Malcolm, Environmental Science & 

Technology, 1990, 24, 1655. 

15. G. Harrington, Characteristics of natural organic matter and their Influence on 

alum coagulation, Ph.D. Dissertation, 1997, University of North Caroline. 

16. P.C. Singer, Water Science and Technology, 1999, 40, 25. 

17. P. C. Singer, J. of Environmental Engineering, 1994, 120. 

18. D. T. Williams, G. L. LeBel, F. M. Benoit, Chemosphere, 1997, 34, 299. 

19. G. L. LeBel, F. M. Benoit, D. T. Williams, Chemosphere, 1997, 34, 2301. 

20. W. J. Chen, C. P. Weisel, J. of American Water Works Association, 1998, 90 

151. 

21. H. Arora, M. W. LeChevallier, K. L. Dixon, J. of American Water Works 

Association, 1997, 89, 60. 

22. M. J. Rodriguez, J. B. Serodes, Water Research, 2001, 35, 1572. 

23. S. W. Krasner, W. H. Glaze, H. S. Weinberg, P. A. Daniel, I. N. Najm, J. of 

American Water Works Association, 1993, 85, 73. 

24. L. Heller-Grossman, J. Manka, B. Limoni-Relis, M. Rebhun, Water Research, 

1993, 27, 1323. 

25. H. Pourmoghaddas, A. A. Stevens, Water Research, 1995, 29, 2059. 

26. H. Pourmoghaddas, A. Stevens, R. N. Kinman, R. C. Dressman, L. A. Moore, 

J. C. Ireland, J. of American Water Works Association, 1993, 85, 82. 

115


IRON DOPED CARBON AEROGEL AS CATALYST FOR 

PHENOL TOTAL OXIDATION 

ANDRADA MĂICĂNEANU a , COSMIN COTEŢ a , 

VIRGINIA DANCIU a , MARIA STANCA a 

ABSTRACT. Two types of iron doped carbon aerogels were prepared by 

sol-gel polymerization of potassium salt of 2,4-dihydroxybenzoic acid with 

formaldehyde. This process was followed by an ion-exchange process between 

K + doped wet gel and Fe(II) or Fe(III) ion aqueous solutions. The resulted Fe(II) 

or Fe(III) doped gels were dried in supercritical conditions with liquid CO2 

and then pyrolyzed when two types of iron carbon aerogels were obtained. 

These aerogels were morpho-structural investigated by means of transmission 

electron microscopy (TEM), X-ray diffraction (XRD), specific surface area 

determination using nitrogen adsorption (BET and BJH methods) and elemental 

analysis. Iron doped carbon aerogels were tested as catalytic materials in 

phenol wet air oxidation process (total oxidation). Temperature, air flow, catalyst 

quantity and phenol concentration over the removal efficiency of the organic 

compound was studied. 

Keywords: iron doped carbon aerogel, phenol, catalytic wet air oxidation 

INTRODUCTION 

Wastewaters containing organic compounds (pollutants) from chemical, 

petrochemical or pharmaceutical industries can create many problems in 

choosing an appropriate method to treat them. Many of them, (refractory organic 

pollutants), are difficult to remove by conventional methods (mechanical and 

biological treatment), therefore alternative methods have to be developed [1- 

3]. Because of their toxicity and frequency of their presence in industrial 

wastewaters, phenol and phenolic compounds have gained increased attention 

in the last two decades. Moreover, phenol is considered to be an intermediate 

in the oxidation route of higher molecular weight aromatics and so usually is 

taken as a model compound in research studies [4,5]. The choice of treatment, 

in case of wastewaters polluted with phenol and phenolic compounds, depends 

on the concentration, which can varies from 0.1 to 6800 mg/dm 3 depending on 

the wastewater source (pulp and paper industry, refineries, coking operations, 

coal processing) [5], economics, efficiency, easy control and reliability [6]. 


Nr.1, RO-400084 Cluj-Napoca, România, andrada@chem.ubbcluj.ro

118 

ANDRADA MĂICĂNEANU, COSMIN COTEŢ, VIRGINIA DANCIU, MARIA STANCA 

Several processes for phenol abatement from wastewaters, which can 

be included in two large categories, recuperative methods (separation) and 

destructive methods, were studied. 

In the first category, methods such as: steam distillation [5], liquidliquid 

extraction (n-hexane, cyclohexane, benzene, toluene, etc.) [5], adsorption 

(activated carbon, activated bentonite, activated zeolites, perlite, resins, etc.) 

[5,7,8], adsorption-flocculation [9], membrane pervaporation [5], membranebased 

solvent extraction [5,10,11] are included. 

In the second category the following methods are included: (a) total 

oxidation by air or oxygen (non catalytic wet air oxidation – WAO, supercritical 

water oxidation – SCWO, catalytic wet air oxidation – CWAO, oxidative 

polymerization with oxygen in presence of enzimes), (b) wet oxidation with 

chemical oxidants (ozone and H2O2 – wet peroxide oxidation WPO – with its 

alternative, non catalytic oxidation, catalytic homogeneous and heterogeneous 

Fenton oxidation CWPO, catalytic non iron heterogeneous catalysis, and 

oxidative polymerization in presence of peroxidases) [5,12-16], (c) oxidation 

with chlorine, chlorine dioxide, potassium permanganate, ferrate (VI) ion [5], 

(d) electrochemical treatment (indirect electro-oxidation, direct anodic oxidation) 

[5,17,18], photocatalytic oxidation [5,19,20], supercritical water gasification – 

SCWG [5], electrical discharge (electro-hydraulic discharge, pulsed corona 

discharge, glow discharge electrolysis) [5], sonochemical processes [5,21], 

biodegradation (microbial or fungi species) [5,22,23]. 

Chemical oxidation with all its alternatives is widely used for treatment 

of wastewaters in order to remove organic pollutants, and has as final objective 

total oxidation (mineralization) of the organic contaminants to CO2, H2O and 

inorganics or, at least at their transformation into harmless products [13]. 

Also between the oxidation processes the catalytic ones become useful 

alternative due to the working conditions, which are milder [24]. As catalyst 

used to destroy organic pollutants, including phenol, we can mention: Ru, Rh, 

Pt, Ir, Ni, Ag supported on TiO2, CeO2, Al2O3 [24,25], metal oxides CuO, CoO, 

Cr2O3, NiO, MnO2, Fe2O3, ZnO, CeO2 [6,24,26,27], activated carbon [28], and iron 

and copper immobilised on synthetic zeolites [29,30] or pillared clays [31,32]. 

Due to their controllable and interesting nanostructural properties, such 

as high surface area, low mass density, high conductivity and continuous 

porosity, iron doped carbon aerogels [33-36], could be attractive materials 

for total oxidation of organic compounds. 

In this paper two iron doped, Fe(II) and Fe(III) carbon aerogels were 

prepared, characterized and investigated as catalytic materials in phenol 

total oxidation process (catalytic wet air oxidation). 


Iron doped aerogel morpho-structural characterization 

TEM images of the iron doped carbon aerogels presented in figure 1, 

show metal particles dispersed in the carbon aerogel matrix.

IRON DOPED CARBON AEROGEL AS CATALYST FOR PHENOL TOTAL OXIDATION 

The XRD patterns of Fe (2+)/(3+) -DCA shows two broad bands around 

2θ = 22 o and 43 o , indicating an amorphous structure of the carbon matrix 

(figure 2). The thin peaks correspond to the iron/iron oxide phases of metal 

particles [35]. 

Specific surface area was determined to be 745 m 2 /g for Fe (3+) -DCA 

and 368 m 2 /g for Fe (2+) -DCA. 

The iron content determined using elemental analysis is about 10% (wt) 

for Fe (3+) -DCA and 18% (wt) for Fe (2+) -DCA [35]. Because of its high metal content 

a smaller specific surface area and a higher content of graphitic structures 

are present in Fe (2+) -DCA [35]. A part of these graphitic structures cloud the 

iron/iron oxide particles. 

Figure 1. TEM images of iron doped carbon aerogels. 

Figure 2. XRD patterns of iron doped carbon aerogels. 

Phenol total oxidation results 

The influence of operating temperature over the evolution of the overall 

efficiency, air flow 60 L/h, 0.1 g Fe (3+) -DCA catalyst and Ci = 1000 mg phenol/dm 3 , 

is presented in figure 3(a). As expected, overall efficiency increased with the 

119

120 


increasing of the temperature. In the first 30 minutes, the efficiency increased 

up to 50.65, 55.74 and 66.56% for an operating temperature of 20, 60 and 

80°C respectively. Maximum values for overall efficiency are presented in 

figure 3(b). The highest value obtained was 68.85% at 80°C. 

Efficiency, (%) 

80 

70 

60 

50 

40 

30 

20 

10 

0 

0 50 100 150 

Time, (min) 

80°C 

60°C 

20°C 


70 

60 

50 

40 

30 

20 

10 

0 

53.44 

(a) (b) 

58.03 

20°C 60°C 80°C 

Figure 3. Variation (a) and maximum values (b) of overall efficiency in the phenol 

CWAO process, at different temperatures and constant air flow, 60 L/h, for 

Fe (3+) -DCA catalyst (0.1 g catalyst and Ci = 1000 mg phenol/dm 3 ). 

The influence of air flow over the overall efficiency variation in time, 

for Fe (3+) -DCA catalyst, at constant temperature 80°C, 0.1 g catalyst and 

Ci = 1000 mg phenol/dm 3 , is presented in figure 4(a). Maximum increase of 

the overall efficiency was observed in the first 15 minutes and varies from 

36.07 to 66.56% for 20 and 60 l/h respectively. We observed also that the 

increase of overall efficiency is not as steep as in case of temperature 

variation, suggesting that oxygen diffusion process could be rate determining 

step in the catalytic wet air oxidation process. Maximum values obtained for 

all air flows are presented in figure 4(b). 


80 

70 

60 

50 

40 

30 

20 

10 

0 

0 20 40 60 80 100 120 140 

Timp, (min) 

60 L/h 40 L/h 20 L/h 


70 

60 

50 

40 

30 

20 

10 

0 

39.02 

46.89 

68.85 

20 L/h 40 L/h 60 L/h 

a b 

Figure 4. Variation (a) and maximum values (b) of overall efficiency in the phenol 

CWAO process, at different air flows and constant temperature, 80°C, for 

Fe (3+) -DCA catalyst (0.1 g catalyst and Ci = 1000 mg phenol/dm 3 ). 

68.85


We also studied the influence of the catalyst quantity and phenol initial 

concentration over the overall efficiency in the CWAO process. Results obtained 

in these cases are presented in figures 5 and 6. If we used a double catalyst 

quantity (0.2 g instead of 0.1 g) an increase with only 11% of the overall 

efficiency was observed, figure 5. Also a 10 times decrease of the phenol 

solution concentration led to an increase of 28% in overall efficiency value 

(figure 6). 

0.2 g 

0.1 g 

68.85 

In the same working conditions, on the Fe 

121 

(2+) -DCA catalyst we obtained 

overall efficiency values with 20% smaller than in case of Fe (3+) -DCA, fact 

that can be explained by its smaller surface area and partial coverage of 

iron/iron oxide particles by graphitic clouds. The blank carbon aerogel sample 

had no activity in phenol oxidation process. 

76.39 

64 66 68 70 72 74 76 78 


Figure 5. Catalyst quantity influence over the maximum overall efficiencies obtained 

in the phenol CWAO process, at constant air flow and constant temperature, 60 L/h, 

80°C, for Fe (3+) -DCA catalyst (Ci = 1000 mg phenol/dm 3 ). 

100 mg/L 

1 000 mg/L 

68.85 

88.39 

0 20 40 60 80 100 


Figure 6. Phenol initial concentration influence over the overall efficiencies in the 

phenol CWAO process, at constant air flow and constant temperature, 60 L/h, 80°C, 

for Fe (3+) -DCA catalyst (0.1 g catalyst).

122 


CONCLUSIONS 

The influence of catalyst type and air flow rate over the overall process 

efficiency was studied. An increase of the temperature and air flow rate led to 

an increase of the overall efficiency. Total oxidation overall efficiencies up 

to 88.39% were reached for Fe (3+) -DCA catalyst at 80°C and 60 l air/h for 

0.1 g catalyst and Ci = 100 mg phenol/dm 3 . 

Fe (3+) -DCA catalyst proved to be more efficient than Fe (2+) -DCA catalyst, 

probably due to its superior morphological properties (higher surface area, 

iron particles on the DCA surface). 

Further studies will be performed in order to establish the optimum 

working conditions for total oxidation of organic compounds, and catalyst 

reproducibility and lifetime 


Iron doped carbon aerogel preparation 

Carbon aerogels doped with Fe were prepared by sol-gel polymerization 

of potassium salt of 2,4-dihydroxybenzoic acid with formaldehyde, followed 

by an ionic exchange process between K + doped wet gel and Fe(II) or Fe(III) 

ion aqueous solutions [35,37]. The resulted Fe(II) or Fe(III) doped gels were 

dried in supercritical conditions with liquid CO2 and then pyrolyzed in high 

temperature and inert atmosphere. 

K2CO3 was added under vigorous stirring to a 2,4-dihidroxybenzoic 

acid (DHBA) demineralised water suspension (DHBA/K2CO3 = 0.5; DHBA/H2O = 

0.0446 g/cm 3 ), a potassium salt solution resulting. After 30 min, when all the 

acid was neutralized, the solution became clear and after another 30 min, 

37% formaldehyde (F) and then K2CO3 (C) (DHBA/F = 2; DHBA/C = 50) 

were added to the solution. The resulting mixture was placed into tightly closed 

glass moulds (7 cm – length × 1 cm – internal diameter) and cured in two 

time intervals: 1 day at room temperature and 4 days at 70 o C. The resulting 

K + -doped gel rods were cut into 0.5-1 cm pellets and washed with fresh 

acetone for 1 day. The K + -loaded wet gels were then soaked for 3 days in 

0.1M aqueous solutions of Fe(NO3)3·9H2O or Fe(OAc)2 [35]. Iron solutions 

were renewed daily. Finally, samples were washed once more with fresh 

acetone and then were subsequently dried with CO2 in supercritical conditions. 

The resulting iron doped organic aerogels were pyrolysed at 750 o C for 3h in 

an Ar atmosphere, obtaining iron and iron oxide particles-doped carbon 

aerogel. The iron doped carbon aerogels prepared using Fe (II) and Fe (III) 

salts were termed Fe (2+) -DCA and Fe (3+) -DCA, respectively. By drying and 

pyrolysis of K + -doped gels the blank carbon aerogel sample (K-DCA) was 

obtained.


Iron doped aerogel morpho-structural investigations 

Transmission electron microscopy (TEM) of the metal doped carbon 

aerogels was performed with a Hitachi H-7000 microscope operating at 125 

keV. 

X-ray diffraction patterns were recorded in a θ–2θ Bragg–Bretano 

geometry with a Siemens D5000 powder diffractometer with Cu-Kα incident 

radiation (λ = 1.5406 Ǻ) and a graphite monochromator. 

Specific surface area determinations were performed using Brunauer- 

Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) methods using an 

ASAP 2000 surface area analyzer (Micrometrics Instruments Corp.). Prior to 

determination, samples of approximately 0.03 g were heated to 130 o C under 

vacuum (10 -5 Torr) for at least 18 h to remove all adsorbed species. 

Elemental analyses were performed with an inductively coupled 

plasma-mass spectroscope (ICP-MS). 

Phenol total oxidation – working conditions 

Phenol total oxidation or catalytic wet air oxidation (CWAO), was carried 

out in a thermostated stirred batch reactor (magnetic stirrer) at atmospheric 

pressure, using different temperatures of 20, 40 and 60°C, air flows (20, 40 and 

60 L/h), catalyst quantities (0.1 and 0.2 g) and phenol initial concentrations 

(100 and 1000 mg/dm 3 ). The Fe (3+) -DCA catalyst, brought at a grain size of 

d < 250 μm using an appropriate sieve, was contacted with 100 cm 3 phenol 

solution. During the experiment, determination of the organic compounds in 

solution was carried out every 15 minutes, (first two determinations), and 

then every 30 minutes. Taking in account the fact that in this stage of the 

research we were interested to see how this type of material acts as catalyst 

for total oxidation of organic compounds we used KMnO4 chemical oxygen 

demand, CCO-Mn, method in order to establish the final concentration of the 

organics in solution. This determination is currently used in environmental 

laboratories for wastewaters characterization (STAS 3002/85, SR ISO 6060/96) 

according to Romanian legislation [38]. The experiment was carried out for 

120 minutes, until no modifications were observed in the organic compound 

final concentration. We also tested a Fe (2+) -DCA catalyst and a blank carbon 

aerogel sample (K-DCA). 

The evolution of phenol oxidation process was followed by means of 

overall efficiency (calculated using chemical oxygen demand values as CCO-Mn 

at a moment t and the initial CCO-Mn value), eq. (1). 

Ci 

− Ct 

X = ⋅ 100 

(1) 

C 

where, 

Ci is the CCO-Mn initial value, in mg KMnO4/dm 3 

i 

Ct is the CCO-Mn value at moment t, in mg KMnO4/dm 3 . 

123

124 



Authors would like to thank EU Marie Curie Training Site Grant HPMT- 

CT-2000-0006, Romanian National University Research Council Grant CNCSIS 

Td 402/2006-2007, and also to research group of Prof. Elies Molins and 

Dr. Anna Roig from Institut de Ciencia de Materials de Barcelona (Spain). 

REFERENCES 

1. A. Santos, P. Yustos, B. Durban, F. Garcia-Ochoa, Industrial Engineering Chemical 

Research, 2001, 40, 2773. 

2. A. Santos, P. Yustos, B. Durban, F. Garcia-Ochoa, Catalysis Today, 2001, 66, 511. 

3. C. Catrinescu, C. Teodosiu, M. Macoveanu, J. Miehe-Brendle, R. Le Dred, Water 

Research, 2003, 37, 1154. 

4. A. Santos, P. Yustos, A. Quintanilla, F. Garcia-Ochoa, J. A. Casas, J. J. Rodriguez, 

Environmental Science and Technology, 2004, 38, 133. 

5. G. Busca, S. Berardinelli, C. Resini, L. Arrighi, Journal of Hazardous Materials, 

2008, 160, 265. 

6. A. Alejandre, F. Medina, A. Fortuny, P. Salagre, J. E. Sueiras, Applied Catalysis B: 

Environmental, 1998, 16, 53. 

7. S. Al-Asheh, F. Banat, L. Abu-Aitah, Separation and Purification Technology, 2003, 

33, 1. 

8. E. Fernandez, D. Hugi-Cleary, M. V. Lopez-Ramon, F. Stoeckli, Langmuir, 2003, 

19, 9719. 

9. Y-H. Shen, Water Research, 2002, 36, 1107. 

10. Z. Lazarova, S. Boyadzhieva, Chemical Engineering Journal, 2004, 100, 129. 

11. W. Kujawski, A. Warszawski, W. Ratajczak, T. Porebski, W. Capala, I. Ostrowska, 

Desalination, 2004, 163, 287. 

12. F. Luck, Catalysis Today, 1999, 53, 81. 

13. R. Andreozzi, V. Caprio, A. Insola, R. Marrota, Catalysis Today, 1999, 53, 51. 

14. B. Legube, N. Karpel Vel Leitner, Catalysis Today, 1999, 53, 61. 

15. S. Zrncevic, Z. Gomzi, Industrial Engineering Chemical Research, 2005, 44, 6110. 

16. I. V. Perez, S. Rogak, R. Branion, Journal of Supercritical Fluids, 2004, 30, 71. 

17. D. Fino, C. Carlasi Jara, G. Saracco, V. Specchia, P. Spinelli, Journal of Applied 

Electrochemistry, 2005, 35, 405. 

18. A. M. Polcaro, A. Vacca, S. Palmas, M. Marcia, Journal of Applied Electrochemistry, 

2003, 33, 885. 

19. J-M. Herrmann, Catalysi Today, 1999, 53, 115.


20. U. Bali, E. C. Catalkaya, F. Sengul, Journal of Environmental Science and Health, 

2003, A38, 2259. 

21. Y. G. Adewuyi, Industrial Engineering Chemical Research, 2001, 40, 4681. 

22. G. Collins, C. Foy, S. McHugh, T. Mahony, V. O’Flaherty, Water Reserach, 2005, 

39, 1614. 

23. A. Nuhoglu, B. Yalciu, Proceses in Biochemistry, 2005, 40, 1233. 

24. Y.I. Matattov-Meytal, M. Sheintuch, Industrial Engineering Chemical Research, 

1999, 37, 309. 

25. A. Cybulski, J. Trazynski, Applied Catalysis B: Environmental, 2004, 47, 1. 

26. L. Chang, I-P. Chen, S-S. Lin, Chemosphere, 2005, 58, 485. 

27. P. M. Alvarez, D. Mclurgh, P. Plucinski, Industrial Engineering Chemical Research, 

2001, 41, 2153. 

28. A. Santos, P. Yustos, S. Gomis, G. Ruiz, F. Garcia-Ochoa, Industrial Engineering 

Chemical Research, 2005, 44, 3869. 

29. M. Noorjahan, V.D. Kumari, M. Subrahmanyam, L. Panda, Applied Catalysis B: 


30. N.H. Phu, T.T.K. Hoa, N.V. Tan, H.V. Thang, P.L. Ha, Applied Catalysis B: 


31. J. Barrault, J-M. Tatibouet, N. Papayannakos, C. R. Academie des Science Paris, 

Serie IIc Chimie, 2000, 3, 777. 

32. J. Barrault, C. Bouchoule, K. Echachoui, N. Frini-Srasra, M. Trabelsi, F. Bergaya, 

Applied Catalysis B: Environmental, 1998, 15, 269. 

33. C. Moreno-Castilla, F. J. Maldonado-Hodar, Carbon, 2005, 43, 455. 

34. A. Martinez, L. C. Vallribera, M. Cotet, L. Popovici, A. Martin, M. Roig, E. Moreno- 

Manas, E. Molins, New Journal of Chemistry, 2005, 29, 1342. 

35. L. C. Cotet, M. Gich, A. Roig, I. C. Popescu, V. Cosoveanu, E. Molins, V. Danciu, 

Journal of Non-Crystalline Solids, 2006, 352, 2772. 

36. S. Cacchi, C. L. Cotet, G. Fabrizi, G. Forte, A. Goggiamani, L. Martín, S. Martínez, 

E. Molins, M. Moreno-Mañas, F. Petrucci, A. Roig, A. Vallribera, Tetrahedron, 

2007, 63, 2519. 

37. T. F. Baumann, G. A. Fox, J. H. Satcher, N. Yoshizawa, R. Fu, M. S. Dresselhaus, 

Langmuir, 2002,18, 7073. 

38. Law 458/2002, 311/2004, 107/1996, 311/2004, 112/2006, HG 188/2002, HG 

352/2005. 

125


HEAVY METAL IONS REMOVAL FROM MODEL 

WASTEWATERS USING ORAŞUL NOU (TRANSILVANIA, 

ROMANIA) BENTONITE SAMPLE 

ANDRADA MĂICĂNEANU a , HOREA BEDELEAN b , 

SILVIA BURCĂ a , MARIA STANCA a 

ABSTRACT. A bentonite sample from Oraşul Nou deposit, (Transilvania, 

Romania), was used to remove heavy metal ions (Zn 2+ , Pb 2+ , Cd 2+ ) from model 

monocomponent wastewaters. A representative sample of bentonite (ON) 

was characterised using, wet chemical analyses, XRD, BET and FTIR. The 

bentonite deposit from Oraşul Nou formed by alteration of rhyolites and perlites. 

Mineralogically, they contain clay minerals (montmorillonite, and subordinately 

kaolinite, illite), cristobalite, carbonates, zeolites (clinoptilolite), iron oxi-hydroxides 

and relics of primary minerals such as quartz and feldspar.The bentonite sample 

was used as powder, (d < 0.2 mm), without any chemical treatment. We 

studied the influence of the working regime, static and dynamic, concentration, 

and solid : liquid ratio over the process efficiency. We used monocomponent 

synthetic wastewaters containing zinc, lead and cadmium ions. The bentonite 

sample proved to be efficient for the removal of the considered heavy metal 

ions, removal efficiencies up to 100% (lead and zinc removal) were reached. 

First-order, pseudo- second-order and Elovich models were used to study the 

adsorption kinetic of zinc ions on the bentonite sample. 

Keywords: bentonite, montmorillonite, zinc, lead, cadmium, removal 

INTRODUCTION 

Heavy metals such as Cd, Pb and Zn are usually found in Earth’s 

crust mainly as minerals and are mobilised by soil erosion, volcanic activities 

and forest fires (natural sources). Heavy metals concentration in the environment 

increased drastically due to intense human activities (anthropogenic sources). 

Cadmium, lead and zinc can be found in air (eroded particles), water 

(in ionic form, from wastewaters insufficient treated or mobilised from polluted 

soils) and soil (adsorbed on soil particles) in different combinations. Heavy 

metals exposure can take place by inhalation (contaminated air, cigarette 


Nr.1, RO-400084 Cluj-Napoca, România, andrada@chem.ubbcluj.ro 

b Universitatea Babeş-Bolyai, Facultatea de Biologie şi Geologie, Str. Kogălniceanu Nr. 1, 

RO-400084 Cluj-Napoca, România, bedelean@bioge.ubbcluj.ro

128 

ANDRADA MĂICĂNEANU, HOREA BEDELEAN, SILVIA BURCĂ , MARIA STANCA 

smoke, working exposure), ingestion (water and contaminated food) and dermal 

(contact with contaminated soil, heavy metal compounds) exposures. With 

exception of Zn, which in small quantities is nutritive elements, heavy metals 

are toxic to humans, animals and plants. Heavy metal ions have the tendency 

to accumulate in living organisms causing a variety of disorders [1-3]. 

Heavy metal ions can be removed from the environment using a variety 

of methods such as precipitation, solvent extraction, vacuum evaporation, 

membrane technologies, adsorption or ionic exchange, the selection of a 

method depending mainly upon the heavy metal ion concentration. Adsorption 

is used when concentration of heavy metal ions is small, but is higher that 

the allowable concentration according to environmental legislation. 

Low-cost adsorbents such as natural zeolites, clay minerals (kaolinite, 

montmorillonite), chitosan, peat moss, fly ash, coal, natural oxides (aluminium 

oxide, ferric oxide), industrial waste (waste slurry, iron (III) hydroxide, lignin, 

furnace slag, sawdust, activated red mud, bagasse fly ash, brewery waste 

biomass), rice husk, coconut shell, agricultural waste, etc. were all considered 

as adsorbents for heavy metal ions (cadmium, chromium, cobalt, copper, iron, 

lead, manganese, nickel, zinc) removal from wastewaters [4,5]. Due to their 

availability and low costs, natural materials such as zeolites and clay minerals 

became object of many research papers in the last years. 

In Romania, most studies related to the identification of new materials 

used in environmental protection, concerned natural zeolites (zeolitic volcanic 

tuffs), which were used in various applications based on their properties [6]. 

Clays, especially bentonite clays, are present in large amounts in Romania. 

Bentonite collected from Transilvanian deposits, Petreşti, Oraşul Nou and 

Valea Chioarului were considered for heavy metal ions (Petreşti), ammonium 

ions (Petreşti, Valea Chioarului, Oraşul Nou) and organics (Petreşti, Oraşul 

Nou) removal from wastewaters [7-9]. 

Clay minerals are aluminium hydrosilicates, crystallized in monoclinic 

system, characterized by planar reticular structures. Stratified structure of 

mineral clays is determined by the combination in one reticular plan (structural 

unit) of two cationic layers, one layer in which silicon is coordinated tetrahedric 

with O⋅OH (tetrahedric level Te) and one layer in which aluminium is coordinated 

octahedric with O⋅OH (octahedric level Oc). These layers are bonded between 

them with van der Waals bonds forming the structural unit. According to 

the numbers of Te and Oc layers, clay minerals are classified in 1Te:1Oc 

(e.g. kaolinite), 2Te:1Oc (e.g. montmorillonite) and 2Te:2Oc (e.g. chlorite). 

The layered structure of the clay minerals and presence of isomorphous 

replacement of Si 4+ with Al 3+ (negative charge compensated by Na + , K + , 

Ca 2+ and/or Mg 2+ ) determines their main properties: swelling, water and 

organic compounds adsorption and ionic exchange capacity [4,10].

HEAVY METAL IONS REMOVAL FROM MODEL WASTEWATERS USING ORAŞUL NOU 

Bentonites are clay rocks formed by devitrification and chemical alteration 

of a glassy igneous material, usually a tuff or volcanic ash. Bentonites are 

composed mainly of smectite group minerals. This group includes dioctahedral 

minerals such as montmorillonite, beidellite and nontronite and trioctahedral 

minerals such as hectorite and saponite. The main mineral from smectite 

group is montmorillonite, a hydrated sodium calcium aluminum magnesium 

silicate hydroxide. 

The purpose of this study was to characterise the bentonite from 

Oraşul Nou (ON) deposit, Satu Mare County, Romania and to evaluate its 

heavy metal ions (Zn 2+ , Pb 2+ , Cd 2+ ) removal capacity. Also, first, pseudo second 

and Elovich kinetic models were used to study the adsorption kinetics of zinc 

ions on the betonite sample. 


Bentonite characterization 

The bentonite deposit from Oraşul Nou (Satu Mare County) is located 

about 30 km north-west of the city of Baia Mare, in the Oaş Basin, Maramureş 

Depression. The area consists of sedimentary rocks and volcanic rocks. 

Sedimentary formations are of marine origin (marls, fine sandstones) of Neogene 

age. Volcanic rocks outcrops are up to 300 m in thickness and consist of rhyolite 

and rhyolitic tuffs, andesitic tuffs, perlites, dacite and pyroclastic rocks. The 

bentonite deposit from Oraşul Nou formed by alteration of rhyolites and perlites. 

Bentonite bodies are lens-like shaped, ranging between 3 and 8 m 

in thickness, 100 to 400 m in length and 50 to 250 m in width. 

In Mujdeni area, from where we collected the samples used for our 

experiments, a rhyolite basement is covered by altered perlites gradually 

passing into bentonites. The bentonite layers are followed by unaltered perlites, 

clay and the soil. The bentonitization of perlites and rhyolites is a process 

connected to solutions’ complex circulation favoured by the porosity of the 

rhyolites in the basement and by the small, concentric fissures within the 

perlites. 

As a rule, bentonites are white in colour, fine grained or compact, 

the perlitic structure being more or less preserved. 

Mineralogically, they contain clay minerals (montmorillonite, and 

subordinately kaolinite, illite), cristobalite, carbonates, zeolites (clinoptilolite), 

iron oxi-hydroxides and relics of primary minerals such as quartz and feldspar. 

The ratio between the montmorillonite and kaolinite-group minerals varies 

according to the degree of transformation of the original rock. Thus, the 

minerals from the kaolinite group are clearly dominant as compared to 

montmorillonite in the areas where bentonitization was not completed, while 

in the intervals with bentonite montmorillonite represents the major phase. 

129

130 


The chemical composition of the bentonites from Oraşul Nou is presented 

in the table 1. It is worthy to note the relatively high Al 2O 3 and relatively low 

Na2O contents. 

Semi-quantitative mineralogical composition of clay raw material 

realised by means of X-ray diffraction analysis, figure 1, indicated that in all 

bentonite samples, smectite (montmorillonite) is the most abundant clay 

mineral. Identified minerals were: smectite (montmorillonite), cristobalite, quartz, 

feldspars. The basal space was recorded at 15 Å, indicating the precence of 

Ca-montmorillonite [11]. 

The montmorillonite amount varies between 20-85%, according to 

the alteration degree of the rock; locally, it may even reach values of 95%. 

Oxides, 

[%] 

Oraşul 

Nou 

Table 1. Chemical composition of the bentonite samples from 

Oraşul Nou (Satu-Mare County). 

SiO 2 Al 2 O 3 Fe 2 O 3 CaO MgO Na 2 O K 2 O TiO 2 L.O.I. 

60.62 

-71.46 

15.40 

-23.55 

0.84 

-2.34 

1.18 

-1.86 

0.54 

-2.05 

0.00 

-0.30 

0.25 

-2.16 

0.15 

-0.36 

Figure 1. Powder X-ray diffractogram of the Oraşul Nou bentonite sample; 

Mm- montomorillonite, Cr-cristobalite, Q-quartz, Fp-feldspar. 

8.69 

-12.4 

Examination of IR spectra of the bentonite sample (figure 2), indicates 

the presence of specific 2:1 clay mineral peaks (montmorillonite) with peaks 

in 600-700 cm -1 region that could be attributed to illite, clinoptilolite or opal 

CT accompanying minerals [12-19]. The identified peaks can be attributed


as follows: 3631.30 cm -1 – stretching vibrations of isolate hydroxyls and OH 

less firmly bond to the tetrahedrical outer layer, 3448.10 cm -1 – stretching 

vibrations of the structural OH and also of the hydration water, 1639.20 cm -1 – 

angular deformation of hydroxyls from adsorbed water molecules (held in 

interlayers), 1095.37 and 1047.53 cm -1 – main bands corresponding to the 

stretching vibrations of Si,Al-O, 794.53 and a shoulder at 620.97cm -1 – weak 

bands corresponding to the stretching vibrations of Si,Al-O (there are not specific 

peaks for montmorillonite and could be attributed to the accompanying minerals), 

and 474.40 and 524.54 cm -1 bending vibrations of Si-O-M bonds (M can be 

Mg, Al, Fe). 

Absorbance 

1.8 

1.6 

1.4 

1.2 

1 

0.8 

0.6 

0.4 

0.2 

474.40 

794.53 

1095.37 

1400.07 

1639.20 

3448.10 

3631.30 

0 

400 900 1400 1900 2400 2900 3400 3900 

Wavenumber, [cm -1 ] 

Figure 2. FTIR spectrum of the bentonite sample from Oraşul Nou deposit. 

Heavy metal ions removal results 

The results obtained in case of zinc removal from monocomponent 

model solutions are presented and discussed in terms of removal efficiency 

and adsorption capacity. 

In figure 3, evolution of zinc removal efficiency in time is presented 

for the experiment conducted in static regime (Ci = 125 mg Zn 

131 

2+ /L, 2 g 

bentonite, 20 ml zinc solution). A closer inspection of this evolution led to 

the conclusion that in the first 24 hours, concentration of zinc ions drops 

significantly from the initial value to 1.68 mg Zn 2+ /L, leading to high removal 

efficiency (98.71%). Equilibrium was reached in 48 hours, when concentration 

of zinc ions in solution drops to 0 (100% efficiency) and adsorption capacity 

increased up to 1.3003 mg Zn 2+ /g bentonite. 

The experiments realised in dynamic regime (3D shaker) were conducted 

with the modification of the initial concentration of zinc in model solutions, 

and bentonite quantity. Results are presented in figures 4 to 7 in terms of 

removal efficiency and adsorption capacity.

132 


Zinc concentration, (mg/L) 

140 

120 

100 

80 

60 

40 

20 

0 

0 20 40 60 80 100 

Time, (h) 

Figure 3. Removal efficiency evolution in time during zinc removal process on 

Oraşul Nou bentonite sample in static regime (Ci = 125 mg Zn 2+ /L, 2 g bentonite). 

From figure 4, where we presented the influence of the initial zinc 

concentration over the evolution of the removal efficiency it can be concluded 

that an increase of the initial concentration led to a decrease of the removal 

efficiency. Also in terms of time evolution of the removal efficiency, we observed 

that trends are similar. Equilibrium was reached after around 45 minutes. 

Maximum removal efficiencies were as follows 100, 83.48, and 71.65 for 125, 

250, and 525 mg Zn 2+ /L initial concentrations respectively. Adsorption 

capacities calculated for the three experiments are presented in figure 5. It is 

easy to observe, that with an increase in the initial concentration, zinc quantities 

retained on the bentonite sample increase. The highest adsorption capacity 

was calculated to be 3.8171 mg Zn 2+ /g bentonite in case of the highest 

initial concentration (525 mg Zn 2+ /L). 

Influence of the bentonite quantity over the zinc removal efficiency 

is presented in figure 6 as evolution in time and in figure 7 as maximum 

values. If we follow the evolution of the removal efficiency in time we can 

observe that as the bentonite quantity increase the removal efficiency has 

higher values and equilibrium is reached faster. When we used 3 g of bentonite 

the equilibrium was reached in 15 minutes, with a maximum removal efficiency 

of 100%. In case of 2 g, equilibrium was reached in 45 minutes with 100% 

removal efficiency, while in case of 1 g, equilibrium was reached in 60 minutes 

with a removal efficiency of 98.71%. 

At a closer inspection of the results obtained in case of the bentonite 

quantity influence we can conclude that an increase in the adsorbent quantity 

has to be implemented in an industrial scale process only after the realisation 

of the economic calculations, taking in account that an increase from 1 to 2 g 

of bentonite led to an increase of just 1.29 points (from 98.71 to 100%) of 

the removal efficiency, but with a decrease from 2.5671 to 1.3003 mg of the 

Zn 2+ retained on the adsorbent unit (g).



100 

90 

80 

70 

60 

50 

40 

30 

20 

10 

0 

0 15 30 45 60 75 

Time, (min) 

125 mg/L 250 mg/L 525 mg/L 

Figure 4. Influence of the initial zinc concentration over the removal efficiency evolution 

in time, during zinc removal process on 2 grams of Oraşul Nou bentonite sample in 

dynamic regime (3D shaker). 

525 mg/L 

250 mg/L 

125 mg/L 

1.3003 

Figure 6. Influence of the bentonite quantity over the removal efficiency evolution in 

time, during zinc removal process on Oraşul Nou bentonite sample in dynamic regime 

(3D shaker), Ci = 125 mg Zn 

133 

2+ /g. 

2.1141 

0 0.5 1 1.5 2 

q, (mg/g) 

2.5 3 3.5 4 

Figure 5. Influence of the initial zinc concentration over the adsorption capacity, in zinc 

removal process on 2 grams of Oraşul Nou bentonite sample in dynamic regime 

(3D shaker). 


100 

90 

80 

70 

60 

50 

40 

30 

20 

10 

0 

0 15 30 45 60 75 

Time, (min) 

1g 2g 3g 

3.8171

134 


3g 

2g 

1g 

98.71 

98 98.5 99 


99.5 100 

Figure 7. Influence of the bentonite quantity over the maximum removal efficiency, in 

zinc removal process on Oraşul Nou bentonite sample in dynamic regime (3D shaker), 

Ci = 125 mg Zn 2+ /g. 

In order to design and optimise a wastewater treatment process is 

also important to know the adsorption kinetic models, which can correlate 

the adsorbate up-take rate with the bulk concentration of the adsorbate. 

First-order (Lagergren), pseudo-second-order and Elovich models were used 

to study the adsorption kinetic of zinc ions on the bentonite sample [20]. Linear 

regression was used to determine the best fitting kinetic rate equation (correlation 

coefficients, R 2 ) [21]. 

Lagergren suggested a first-order equation for the adsorption of liquid/ 

solid system based on solid capacity, which can be expressed as follows: 

dqt 

= k1( 

qe 

− qt 

) 

(1) 

dt 

Integrating eq. (1) from the boundary conditions t = 0 to t = t and qt = 0 

to qt = qt, gives: 

ln( qe 

− qt 

) = lnqe 

− k1t 

(2) 

where, 

qe and qt are the amounts of zinc adsorbed (mg/g) at equilibrium 

and time t, respectively 

k1 is the rate constant of first-order adsorption (1/min). 

In order to determine the rate constant and equilibrium zinc uptake, 

the straight line plots of ln(qe-qt) against t, eq. (2), were made at four 

different initial zinc concentrations. Correlation coefficients between 0.8080 

and 0.9500 were obtained (figure not shown), therefore zinc adsorption on 

bentonite cannot be classified as first-order. 

The pseudo-second-order kinetic model is derived on the basis of 

the adsorption capacity of the solid phase, expresses as: 

dqt 

2 

= k2 

( qe 

− qt 

) 

(3) 

dt 

100 

100




1 1 

= + k2t 

(4) 

( qe 

− qt 

) qe 

where, 

qe and qt are the amounts of zinc adsorbed (mg/g) at equilibrium 

and time t, respectively 

k2 is the rate constant of first-order adsorption (g/mg⋅min). 

Equation (4) can be rearranged in linear form, as follows: 

t 1 t 

= + 2 

(5) 

qt 

k2qe 

qe 

In order to determine the rate constant and equilibrium zinc uptake, 

the straight line plots of t/qt against t, eq. (5), were made at four different initial 

zinc concentrations. Correlation coefficients between 0.9867 and 1.0000 were 

obtained (figure 8 and table 1), therefore zinc adsorption on bentonite can 

be classified as pseudo-second-order. 

Elovich equation that is widely used to describe the kinetics of 

chemisorption of gas and solids can also be applied to liquid/solid systems and 

can give information about the possibility that chemical adsorption or chemical 

adsorption involving valence forces through sharing or exchange of electrons 

between adsorbent and adsorbate to be rate determining step [20]. The Elovich 

equation was derived from the Elovich kinetic equation: 

dqt −βqt 

= αe 

(6) 

dt 



1 1 

qt = ln( αβ) 

+ ln( t + t0 

) 

(7) 

β β 

where, 

α and β are the parameters of the equations, and 

t0 = 1/(αβ) 

α represents the rate of chemisorption at 0 coverage, (mg/g⋅min), 

β is related to the extent of surface coverage and activation energy 

for chemisoprtion, (g/mg) [20]. 

When t0


In order to determine equations parameters, the straight line plots of 

qt against ln(t), eq. (8), were made at four different initial zinc concentrations. 

Correlation coefficients between 0.8562 and 0.9188 were obtained (figure 

not shown), therefore zinc adsorption on bentonite cannot be classified as 

chemisorption. 

t/qt 

70 

60 

50 

40 

30 

20 

y = 0.6271x + 2.708 

R 2 = 0.9867 

y = 0.766x + 0.1955 

R 2 = 1 

y = 0.4659x + 0.4483 

R 2 = 0.9999 

y = 0.2536x + 0.5736 

R 2 10 

0 

= 0.9997 

0 20 40 

t, [min] 

60 80 

125 mg/L 

170 mg/L 

250 mg/L 

525 mg/L 

Linear (525 mg/L) 




Figure 8. Plots of the second-order model, at different initial zinc concentrations 

(dynamic regime, 2 g bentonite, 20 ml zinc solution). 

Table 1. Second order adsorption rate constants, and calculated and experimental 

qe values for zinc adsorption using different initial concentrations. 

C, 

(mg Zn 2+ /L) 

qe (exp), 

(mg Zn 2+ /g) 

k2, 

(g/mg⋅min) 

qe (calc), 

(mg Zn 2+ /g) 

R 2 

125 1.3003 3.0013 1.3005 1.0000 

170 1.5772 0.1452 1.5946 0.9867 

250 2.1141 0.4842 2.1464 0.9999 

525 3.8171 0.1121 3.9432 0.9997 

The results obtained in case of lead and cadmium removal from 

monocomponent model solutions are also presented and discussed in terms 

of removal efficiency and adsorption capacity. 

Lead removal process was realised in static and dynamic (3D shaker) 

regimes using 2 g of bentonite and 20 ml solution of 54 mg Pb 

136 

2+ /L. In both 

static and dynamic regime the initial concentrations drops significantly in the 

first 24 h and 15 min, from the initial 54 to 3.73 and 0.01 mg Pb 2+ /L, respectively, 

indicating a favourable diffusion process in case of dynamic conditions. Maximum 

efficiency reached 99.99% in both cases, while the adsorption capacity was 

calculated to be 0.5390 mg Pb 2+ /g bentonite. Removal efficiency evolution in 

time during lead removal process on Oraşul Nou bentonite sample in dynamic 

regime is presented in figure 9. Equilibrium was reached in 72 h in static regime 

and 60 minutes in dynamic regime.


Cadmium removal process was realised in static regime using a 

2150 mg Cd 2+ /L model solution, 20 ml, and 2 g of bentonite. Maximum removal 

efficiency value was calculated to be 48.08% corresponding to an adsorption 

capacity of 10.3418 mg Cd 2+ /g. Equilibrium was reached in 48 h. Evolution 

of removal efficiency in dynamic regime (Ci = 25 mg Cd 2+ /L, 2 g bentonite, 

20 ml solution) is presented in figure 10. Removal efficiency increases slowly 

in the first 60 minutes, until 62.37%, value obtained at equilibrium (60 min). 

At equilibrium, adsorption capacity was calculated to be 0.1589 mg Cd 2+ /g. 

The adsorption capacity values obtained, ranging between 0.1589 and 

10.3418 mg Cd 2+ /g indicates that our bentonite sample has high adsorption 

capacities, therefore it can be used to remove cadmium ions from diluted 

as well as more concentrated solutions. 

75 min 

60 min 

45 min 

30 min 

15 min 

99.9700 99.9750 99.9800 99.9850 99.9900 99.9950 


Figure 9. Removal efficiency evolution in time during lead removal process on 

Oraşul Nou bentonite sample in dynamic regime (3D shaker), 

Ci = 54 mg Pb 2+ /L, 2 g bentonite, 20 ml solution. 


70 

60 

50 

40 

30 

20 

10 

0 

0 10 20 30 40 

Time, (min) 

50 60 70 80 

Figure 10. Removal efficiency evolution in time during cadmium removal process 

on Oraşul Nou bentonite sample in dynamic regime (3D shaker), 

Ci = 25 mg Cd 

137 

2+ /L, 2 g bentonite, 20 ml solution.

138 


CONCLUSIONS 

A bentonite sample from Oraşul Nou deposit, (Transilvania, Romania), 

was used to remove heavy metal ions (Zn 2+ , Pb 2+ , Cd 2+ ) from model 

monocomponent wastewaters. 

Physico-chemical and mineralogical analyses of the bentonite from 

Oraşul Nou, Satu Mare County, Romania indicated that our sample is a 

smectitic clay mineral, containing Ca-montmorillonite (up to 85%). 

The considered bentonite sample proved to be efficient in heavy 

metal ions removal process. Efficiencies up to 100% in case of Zn 2+ 

(Ci = 125 Zn 2+ mg/L), 99.99% in case of Pb 2+ (Ci = 54 mg Pb 2+ /L) and 62.37% 

in case of Cd 2+ (Ci = 25 Cd 2+ mg/L) in dynamic regime, were obtained. 

First-order, pseudo-second-order and Elovich models were used to 

study the adsorption kinetic of zinc ions on the bentonite sample. We conclude 

that zinc adsorption on Oraşul Nou bentonite sample can be classified as 

pseudo-second-order. 


Bentonite compositional investigations 

Representative bulk rock samples of bentonite were collected from 

open bentonite pits in Oraşul Nou area (Satu Mare County, Romania). 

The analyses of whole-rock chemistry were performed at ICEI (Physico- 

Chemical Analises Center) Cluj-Napoca using usual analytical methods for 

silicate materials (wet chemistry). 

X-ray diffraction analyses on random powders were performed using 

a Siemens Bruker unit with Cu Kα anticathode. The diffractograms were 

recorded from 10° to 70° 2θ. The analytic conditions are 40 A, 40 kV, step of 

2 degrees. A semi-quantitative X-ray diffraction method to determine mineral 

composition was used. 

Specific surface area of the bentonite sample was determined using 

Brunauer-Emmett-Teller (BET) method using a home made apparatus. BET 

specific surface area was determined to be 72 m 2 /g. 

FTIR spectrum of the bentonite sample was recorded using a Jasco 

615 spectrophotometer, 400-4000 cm -1 , resolution 2 cm −1 . 

Heavy metal ions removal procedure 

Heavy metal ions (Zn 2+ , Pb 2+ , Cd 2+ ) removal was realised on a 

representative bentonite sample from Oraşul Nou (ON) deposit, Satu Mare 

County, Romania. The bentonite sample was used as powder, particles with 

grain size smaller than 0.2 mm, without any chemical treatment.


For the heavy metal ions removal study we used synthetic 

monocomponent solutions containing zinc ions (125-525 mg Zn 2+ /L), lead 

ions (54 mg Pb 2+ /L) and cadmium ions (25-2150 mg Cd 2+ /L) prepared from 

ZnSO4⋅7H2O, (CH3OO)2Pb⋅3H2O and Cd(NO3)2⋅4H2O salts (analytical purity 

reagents). Cadmium and lead ions in solution were determined using an ion 

selective electrode and a pH meter (Jenway 3330), while in case of zinc ions 

we used a spectrophotometric method (potassium ferrocyanure, λ = 420 nm, 

Jenway 6305 spectrophotometer). Experiments were carried out without 

any modification of the temperature (the experiments were realised at room 

temperature, 20°C) and pH of the synthetic solutions. 

Heavy metal ions removal process was realised in a batch reactor in 

static and dynamic (3D shaker) regimes, using 1, 2 and 3 grams of bentonite 

in contact with 20 ml heavy metal ion solution. In order to determine the 

exact concentration of heavy metal ions, water samples were taken every 

24 hours in static regime and every 15 minutes in dynamic regime, until the 

equilibrium was reached. 

We studied the influence of the working regime, static and dynamic 

(3D shaker), initial concentration and bentonite quantity (zinc) over the 

process efficiency and adsorption capacity. 

Removal efficiencies (%) and adsorption capacities (mg M n+ /g) were 

calculated in order to establish the effectiveness of the considered bentonite 

sample in the heavy metal ions removal process (the calculated values of 

removal efficiencies and adsorption capacities should be regarded according 

to the precision of the determination methods we used). 

First-order, pseudo-second-order and Elovich models were used to 

study the adsorption kinetic of zinc ions on the bentonite sample. 


This work was realized with financial support from Romanian National 

University Research Council, Grant CNCSIS A 1334. 

REFERENCES 

1. H. Abadin, A. Ashizawa, Y. W. Stevens, F. Llados, G. Diamond, G. Sage, 

M. Citra, A. Quinones, S. J. Bosch, S. G. Swarts, “Toxicological profile for lead”, 

U.S. Department of Health and Human Services, Public Health Service Agency 

for Toxic Substances and Disease Registry, 2007. 

2. J. Taylor, R. DeWoskin, F. K. Ennever, “Toxicological profile for cadmium”, U.S. 

Department of Health and Human Services, Public Health Service, Agency for 

Toxic Substances and Disease Registry, 1999. 

139

140 


3. N. Roney, C. V. Smith, M. Williams, M. Osier, S. J. Paikoff, “Toxicological profile 

for zinc”, U.S. Department of Health and Human Services, Public Health Service, 

Agency for Toxic Substances and Disease Registry, 2005. 

4. K. G. Bhattacharyya, S.S. Gupta, Advances in Colloid and Interface Science, 2008, 

140, 114. 

5. S. Babel, T. A. Kurniawan, Journal of Hazardous Materials, 2003, B97, 219. 

6. A. Măicăneanu, H. Bedelean, M. Stanca, “Zeoliţii naturali. Caracterizare şi aplicaţii 

în protecţia mediului”, Presa Universitară Clujeană, 2008, Cluj-Napoca, chapter 

1-3, 6. 

7. H. Bedelean, A. Măicăneanu, S. Burcă, M. Stanca, Clay Minerals, 2009, in press. 

8. H. Bedelean, A. Măicăneanu, S. Burcă, M. Stanca, Sesiunea Ştiinţifică anuală 

GEO 2009 a Facultăţii de Geologie şi Geofizică, în parteneriat cu Societatea 

Geologică a României şi Societatea Română de Geofizică, 2009, Bucureşti. 

9. M. Stanca, A. Măicăneanu, S. Burcă, H. Bedelean, International Conference, 

Sustainable Development in the Balkan Area: Vision and Reality, 2007, Alba-Iulia. 

10. D. Rădulescu, N. Anastasiu, “Petrologia rocilor sedimentare”, Editura Didactică şi 

Pedagogică, 1979, Bucureşti, chapter 6. 

11. N. Anastasiu, “Minerale şi roci sedimentare, Determinator, Editura Tehnică, 

1977, Bucureşti, chapter 3. 

12. D. Haffad, A. Chambellan, J. C. Lavalley, Catalysis Letter, 1998, 54, 227. 

13. A. Bakhti, Z. Derriche, A. Iddou, M. Larid, European Journal of Soil Science, 2001, 

52, 683. 

14. W. P. Gates, J. S. Anderson, M. D. Raven, G. J. Churchman, Applied Clay Science, 

2002, 20, 189. 

15. P. Komadel, Clay Minerals, 2003, 38, 127. 

16. K. A. Carrado, (S.M. Auerbach, K.A. Carrado, P.K. Dutta, editors), in Handbook 

of layered materials, Marcel Dekker Inc., 2004, New-York, chapter 1. 

17. R. J. Hu, B. G. Li, Catalysis Letter, 2004, 98, 43. 

18. A. Meunier, Clays, Springer-Verlag, 2005, Berlin, chapter 1. 

19. S. Ozcan, O. Gok, A. Ozcan, Journal of Hazardous Materials, 2009, 161, 499. 

20. C. Namasivayam, D. Sangeetha, Adsorption, 2006, 12, 103. 

21. J. Febrianto, A. N. Kosasih, J. Sunarsao, Y. Ja, N. Indraswati, S. Ismadji, Journal 

of Hazardous Materials, 2009, 162, 616.


PVC MATRX IONIC - SURFACTANT SELECTIVE ELECTRODES 

BASED ON THE IONIC PAIR TETRA ALKYL-AMMONIUM - 

LAURYLSULPHATE 

CRISTINA MIHALI a , GABRIELA OPREA a , ELENA CICAL a 

ABSTRACT. A series of all-solid-state electrodes sensible to the anionic 

surfactants were developed using silver pills coated with PVC membranes. 

We used cetyltrimethylammonium laurylsulfate (CTMA-LS) and tricaprylmethylammonium 

laurylsulfate (TCMA-LS) as electroactive matter (ionophore) 

and different plasticizers. The electrochemical properties of the electrodes were 

studied. The best electrochemical characteristics were obtained with TCMA- 

LS (dioctylsebaccate as plasticizer): nearnerstinan slope: 58,56 mV/pC; 

concentration range of linear response 10 -3 -2x10 -6 M·L -1 of laurylsulfate anion. 

The selected membrane-electrode can be applied as equivalence point indicator 

to the potentiometric titration of anionic surfactants and to the determination of 

anionic surfactants in environmental samples. 

Keywords: anionic surfactant-selective electrode; PVC - matrix membrane 

electrodes, sodium laurylsulfate 

INTRODUCTION 

The surfactants are compounds essential to the modern civilization and 

technology. Determination and monitorization of the surfactants concentration 

is necessary in the production of detergents, in the industrial processes 

where anionic surfactants are used, in quality control of products containing 

added surfactants and in environment surveillance activities, especially 

monitoring water quality. 

In the last decades there have been created and improved numerous 

analytic methods for determination of anionic surfactants [1]. Among these, 

there are the potentiometric methods based on potentiometric sensors 

(electrodes sensible to anionic surfactants) .There are some review articles 

that presents the electrochemical or potentiometric measurement applied 

for the determination of surfactants [2-3]. Potentiometric sensors for anionic 

surfactants are very attractive tools due to their good precision, relatively 

a 

North University of Baia Mare, Sciences Faculty, 76 Victoriei, RO-430072, Baia Mare, Romania, 

cmihali@yahoo.com

CRISTINA MIHALI, GABRIELA OPREA, ELENA CICAL 

simple manufacturing, relatively low cost and their ability to determine the 

surfactants in the samples without previous separation steps [4-10]. Also some 

anionic surfactant ion-sensitive field effect transistors (anionic surfactant FET) 

were proposed [11-12]. 

In order to determine the concentration of the anionic surfactants 

potentiometric sensors with liquid membrane and polymeric membrane 

(especially PVC membrane) have been produced. 

The polymeric membrane is made up of polyvinyl chloride (PVC) in 

most of the cases, with high relative molecular mass (100,000), a substance 

with plasticizer role and the organic ion exchanger (ionophore) that makes 

the electrode component. The plasticizer assures the mobility of the ion 

exchanger into the polymeric membrane; fixes the dielectric constant value 

of the membrane and confers it the corresponding mechanical properties. The 

proportion of the membrane components depends on the polymer type and on 

the ion exchanger used. It is chosen such that the electrode performances 

are optimum. 

The electrodes with polymeric membrane are prepared in the classical 

way with internal reference solution, in the ''coated wire" version [13] or in 

the constructive form of “all solid state” [14]. In the case of the last type of 

surfactant selective electrodes, the polymeric membrane is built on a graphite 

epoxy support. We realized in all solid-state electrode for anionic surfactants 

with the polymeric membrane attached by a metallic pill. These constructive 

forms was not presented in the literature devoted to the anionic surfactant 

polymeric electrodes and has as main advantages the robustness and an 

easier handling. 

This layout of the electrode has the great advantage of a solid internal 

contact because it does not require internal reference electrode and internal 

electrolyte. 

First, we wanted to produce electrodes sensible to anionic surfactants 

adopting this layout, using different compositions of the PVC membrane and 

to establish the functional characteristics of those electrodes. We used as 

ionophores CTMA-LS and TCMA-LS and different plasticizers: tricresylphosphate 

(TCF), ortho-nitrophenyloctylether (NPOE) and dioctylsebaccate (DOS). 

Second, we wanted to use the electrodes with the best analytical 

performances to the determination of the anionic surfactants concentration 

from the environmental samples. 


The response function 

In order to establish the electrode function we have used 10 

142 

-7 - 5×10 -2 M 

sodium laurysulphate solutions. We have worked in thermostatic regime at 25°C 

in magnetically stirred solutions. The sensors have been introduced successively

PVC MATRX IONIC - SURFACTANT SELECTIVE ELECTRODES BASED ON THE IONIC PAIR… 

in laurysulphate solutions with increasing concentrations. The electrode potential 

has been registered after stabilizing its response. The potential values on 

different concentration levels are the average of three determinations. 

We wanted to choose an optimum ionic strength adjustor in order to 

obtain a value of the slope as close as possible to the Nernstian one and 

an extended linear response range. In this purpose we tested the following 

solutions: Na2SO4 0.01 M (J=0.03), Na2SO4 0.1 M (J=0.3), NaCl 0.01M 

(J=0.01), NaCl 0.1 M (J=0.1). 

The tests were performed using the CTMA-LS membrane electrode 

(plasticized with DOS). The results are presented in Table 1. 

The electrode functions for the pure NaLS solutions and for those 

adjusted with Na2SO4 0.1 M and Na2SO4 0.01 M are exposed in Figure 1. 

Table 1. Influence of ionic strength adjustor on the characteristics of 

the laurylsulphate sensible electrodes based on CTMA-LS (DOS plasticizer) 

The nature and concentration of 

Membrane characteristics 

the ionic strength adjustor Slope, mV/conc. decade Linear response range, M 

Without ionic strength adjustor 

(pure NaLS solutions) 

59.33 10 -3 -4.33x10 -6 

Na2SO4 0,1M, J=0.3 56.22 10 -3 -4.29x10 -6 

Na2SO4 0,01M, J=0.03 59.39 10 -3 -3.93x10 -6 

NaCl 0.1M, J=0.1 55.66 10 -3 -2.94x10 -6 

NaCl 0.01M, J=0.01 57.53 10 -3 -3.99x10 -6 

E/ESC (mV) 

150 

100 

50 

0 

-50 

-100 

-150 

0.00000001 0.0000001 0.000001 0.00001 0.0001 0.001 0.01 0.1 

7 5 3 

pC 

NaLS NaLS, Na2SO4 0,01M NaLS, Na2SO4 0,1M 

Figure 1. The influence of the ionic strength on the response functions of 

the laurylsulphate sensible electrode: 

X - pure NaLS solutions; O - NaLS solutions with Na2SO4 0.01 M (J=0.03) 

Δ- NaLS solutions with Na2SO4 0.1 M (J=0.3) 

143

144 


The best results, the slope 59.39 mV/decade and the linear response 

range 10 -3 -3.93×10 -6 M was obtained with Na2SO4 0.01 M (J=0.03) that was 

selected as optimum ionic strength adjustor. 

We can observe that, in the case of pure NaLS solutions, the electrode 

function has a minimum at 8×10 -3 M. This corresponds to the critical micellar 

concentration (CCM). For concentration values greater than CCM takes place 

the association on a large scale of the surfactant molecules and in these 

conditions, the number of the free laurylsulphate anions decreases. The sensor 

senses only the free laurylsulphate anions. CCM value depends on the presence 

and the nature of the ionic strength adjustor. 

In the case of the ionic strength adjustor Na2SO4 0.01M, CCM takes 

the value 5×10 -3 M and 2×10 -3 M in the case of Na2SO4 0.1M respectively. 

When we use NaCl as ionic strength adjustor, CCM is 6×10 -3 M for 

NaCl 0.01M and 2×10 -3 M for NaCl 0.1M. The CCM value determined this 

way corresponds to the value presented in the literature [15]. As the CCM 

value is a property related to industrial applications of the surfactants and it 

can be determined quickly with potentiometric sensors, they can be used to 

establish CCM value for surfactant solutions in presence of different inorganic 

electrolytes [16]. 

Comparing the electrodes prepared from the point of view of the 

linear response range and the slope (Table 2) we can notice that the best 

performances have been obtained with the CTMA-LS ionophore based 

electrode, plasticized with DOS (slope 59.39 mV/concentration decade, 

linear response range 10 -3 -3.93x10 -6 M) and the electrode with TCMA-LS 

ionophore and the same plasticizer (slope: 58.56 mV/concentration decade, 

linear response range 10 -3 -2x10 -6 ). Nearer values have been obtained for 

membrane with TCMA-LS ionophore, plasticized with TCF (slope: 58.89 

mV/decade and linear response range 10 -3 -2.9x10 -6 M). 

Table 2. Influence of polymeric membrane composition on the 

electrode performances 

Membrane characteristics 

Ionophore / Plasticizer Slope, 

mV/conc. decade 

Linear response range, M 

DOS 59.39 10 -3 -3.93x10 -6 

TCF 58.19 10 -3 -4.2x10 -6 

CTMA-LS 

NPOE 56.08 10 -3 -4.88x10 -6 

DOS 58.56 10 -3 -2x10 -6 

TCF 58.87 10 -3 -2.9x10 -6 

TCMA-LS 

NPOE 55.07 10 -3 -3.5x10 -6


The performances of the prepared membranes are close to those of 

the membranes described in literature. Thus, the PVC membrane proposed by 

Gerlache [5] having as ionophore 1,3-didecyl-2-methyl-imidazolium laurylsulphate 

has the slope 58.9 mV/concentration decade and linear response range 

2.5x10 -6 – 5x10 -3 M. 

The pH function 

We determined experimentally the pH functions of the laurylsulphatesensible 

sensors with TCMA-LS membrane, plasticized with DOS, TCF and 

NPOE. The useful pH range where the electrode response is not influenced 

by the pH change depends on the nature of the plasticizer used and on the 

primary ion concentration (LS − ). In Table 3 there are shown the optimum 

pH ranges corresponding to different levels of NaLS concentration. 

The useful pH is enough large allowing the utilization of the sensor 

to the determination of the anionic surfactants from water samples whose 

pH, generally does not exceed this domain. 

Table 3. Influence of plasticizer nature 

on the useful pH range for sensors with TCMA-LS membrane 

NaLS conc., M 

DOS 

Plasticizer / useful pH range 

TCF o-NPOE 

10 -6 3.5-9 4.5-9 4-9 

10 -5 2.5-10 4-10.5 3.5-10 

10 -4 2-11.5 3.5-11 3-11.5 

10 -3 2-12 2.5-12 2.5-12 

Interferences 

The selectivity of the TCMA-LS laurylsulphate-sensible sensors to 

Cl − , Br − , I − , NO3 − , HCO3 − anions was established using the mixed solutions 

method, by maintaining the primary ion (LS − ) concentration constant and 

varying the concentration of the interfering ion. 

The Cl − , Br − , NO3 − and HCO3 − anions do not interfere, as their 

selectivity coefficients are less than 10 -5 , but the I − interferes and its 

response was influenced even for relatively great concentrations of NaLS 

(10 -3 M). The selectivity with respect to other anionic surfactants like 

dodecylbenzenesulphonate (DBS - ) and ethoxylaurylsulphate (Etoxi-LS - ) was 

studied by applying the separate solutions method, by graphically valuing 

the concentrations corresponding to the same potential value. The values of 

selectivity coefficients of the sensors with TCMA-LS membrane are centralized 

in Table 4. 

145

146 

Interfering ion, J - 

I - 

DBS - 

Etoxi-LS - 


Table 4. Influence of plasticizer nature on the selectivity 

of the membrane with TCMA-LS ionophore 

Selectivity coefficients 

K − 

pot 

− 

LS / J 

, Plasticizer 

DOS TCF o-NPOE 

3.2x10 -1 

3.2x10 -1 8x10 -2 

6.9x10 -1 1.86 1.77 

4.7x10 -1 

1.2 5.9 x10 -1 

As a consequence it appears that these electrodes could be used in 

river or wastes waters without any interference from these ions except the I - 

that, if it is present, interferes and must be removed from the sample. 

Potentiometric titrations 

In order to continue the study the electrode based on TCMA-LS 

ionophore and the DOS plasticizer was selected. This electrode has been 

used as indicating electrodes in potentiometric titration of anionic surfactants 

with cationic surfactants titrants. Several cationic surfactants were used, 

monitoring the accuracy of determinations, the potential jump around the 

equivalence point and the reproducibility. 

The potential jump was considered the difference between the potential 

registered when the titrating ratio is 90% and 110% [18]. 

We have used the following cationic detergents: 

- cetylpyridinium bromide (CPY-Br); 

- cetyltrimethylammonium bromide (CTMA-Br); 

- Hyamine 1622. 

In order to select the optimal cationic agent we performed titrations 

of the sodium laurylsulphate (50 mL solution 10 -3 M) with titrating agents 

mentioned above. Each titration was repeated three times after performing 

1-2 titrations for conditioning the electrode with the new titrant. 

The titration curves are shown in figure 3 and the results in Table 5. 

Table 5. Statistical evaluation of potentiometric titrations of NaLS solutions, 

performed with TCMA-LS (DOS plasticizer) membrane electrode 

and different cationic surfactants 

Cationic surfactant used as titrant CTMA-Br Hyamine 1622 CPY-Br 

Equivalence volume, VE, mean value [mL] 

Mean square error sn-1 [mL] 

Relative mean square error, sn-1 [%] 

Potential jump ΔE 90/110%, 

mean value (mV) 

9,97 

0,107 

1,07 

9,93 

0,134 

1,35 

9,84 

0,114 

1,15 

185,33 110 170,33 

Recovery % 99,67 99,27 98,40


E/ESC (mV) 

350 

300 

250 

200 

150 

100 

50 

0 

0 5 10 15 20 

V (mL) 

CTMA-Br Hyamine CPY-Br 

Figure 3. Influence of the titrating agent on the potentiometric titration curve 

(titration of 50 mL NaLS 10 -3 M with different cationic surfactants, 5x10 -3 M) 

Indicating electrode: TCMA-LS membrane plasticized with DOS 

We obtained a good recovery in the case of titration with CTMA-Br 

and Hyamine. The potential jump around the equivalence point was high for 

CTMA-Br and CPY-Br. The best results were obtained for CTMA-Br. 

We have carried out potentiometric titrations of other anionic surfactant 

species: sodium ethoxylated laurylsulphate and sodium dodecylbenzenesulphonate. 

We used CTMA-Br 5×10 -3 M, and the indicating electrode was the one with 

TCMA-LS membrane plasticized with DOS. The titration curves are well 

contoured, the potential jump around the equivalence point is high, having 

values close to the value obtained in titrating of NaLS (203 mV for NaDBS, 

167 mV for NaEtoxiLS, versus 217 mV for NaLS). 

Measurement of anionic surfactants in river and waste waters 

Samples from the river Săsar and from the waste water plant of 

Baia Mare were analyzed by potentiometric titration with CTMA-Br 10 -5 M 

using the TCMA-LS electrode. The results compared with two-phase titration 

are presented in Table 6. A good correlation was obtained. 

Table 6. Results of the determination of of anionic surfactants in 

environmental samples 

Anionic surfactant concentration, cx10 6 M 

Sample Potentiometric titration Two-phase titration 

Sasar River 2.86 (5.2%) * 

2.88 (2.2%) 

Waste water 1 7.60 (1.9%) 7.67 (0.13%) 

Waste water 2 4.43 (2.8%) 4.4 (0.18%) 

* 

RSD (relative standard deviation is indicated in the brackets); 

Means values corresponds to measurements in fourfold. 

147

CONCLUSIONS 

148 


Some series of polymeric membranes sensible to anionic surfactants 

based on plasticized PVC containing CTMA-LS or TCMA-LS ionophores were 

prepared. We elaborated the preparing methods for the ionic association 

compounds used as ionophores and we prepared electrodes with polymeric 

membranes formed directly on an adequate metallic plate, attached to the 

inferior side of the electrode body (internal solid contact). 

We established the main functional characteristics of the electrodes: 

sensitivity (the slope of the electrode function mV/pC) and linear response 

range. 

The sensor was utilized in the determination of anionic surfactants 

from the river and waste water samples. The results obtained results agree 

with the two-phase titration method. 


Reagents and apparatus 

All the reagents used were of analytical grade. Aliquot 336 S (Fluka) and 

sodium laurylsulphate (Na-LS) form Merck were used to obtain the ionophore. 

The main component in Aliquat is tricaprylmethylammonium chloride (TCMA-Cl). 

High molecular weight polyvinyl chloride (PVC) (Fluka) was used as polymeric 

matrix of the sensitive membrane of anionic surfactants electrodes. 

The following plasticizers have been used: tricresylphosphate (TCF) 

(BDH Chemicals), orto-nitrophenyoctylether (NPOE) (Fluka) and dioctylsebaccate 

(DOS) (Merck). 

The pH/mVmeter Consort P 901 (Belgium) was used for the potential 

measurements. The reference electrode was a calomel saturated electrode 

(ESC). The potential measurements were made in stirred solutions at 25 ˚C 

(thermostat) using a magnetical stirrer. For pH measurements a pH-combination 

glass electrode was used. 

Surfactant electrode preparation 

Preparation of the ionophore 

The ionic association compounds are made of a tensioactive anion 

and a bulky cation from a cationic surfactant. They are solid or viscous liquid 

substances. The technology for preparing the ionic association compounds 

is shortly described below: 

We mixed equimolar solutions of anionic and cationic surfactants. The 

precipitation of the ionic association compound takes place. Their separation 

from the reaction mix is done with respect to its state (solid or liquid) by filtering 

or by extracting with organic solvent (chloroform).


We purified the ionic association compound by washing the precipitate 

with small amounts of water on the filter or by washing the extract in the organic 

solvent with distilled water several times (negative reaction with AgNO3). 

The ionic association compound obtained may contain traces of water that 

can be removed by treating with anhydrous sodium sulfate, after dissolving 

in acetone. 

CTMA-LS has been obtained based on the information found in literature 

regarding the preparation of ionic association compounds [17,19]. 

We mixed hot 10 mL of NaLS 0.05 M solution with the same volume 

of CTMA-Br 0.05 M solution, stirring continuously. A white precipitate of 

CTMA-LS was obtained, that was purified and desiccated according to the 

method presented above. 

The ionic association between the laurylsulphate anion and the 

tricaprylmethylammonium cation was obtained by mixing together 50 mL 

NaLS 0.05 M solution with 50 mL TCMA-Cl equimolar solution. 

The mixture of the two solutions was agitated for 15 minutes on 

water bath at 60°C. From the white emulsion obtained was extracted the 

TCMA-LS using chloroform and then it was purified and desiccated. The 

TCMA-LS obtained has a semisolid state. 

Preparation of the membrane 

From the point of view of the composition, the classical procedure 

has been used, with 1% ionophore, 33% PVC and 66% plasticizer [17,19-20]. 

We have used as ionophores: the ionic association compounds: 

CTMA-LS and TCMA-LS. As plasticizers we have used TCF, o-NPOE and DOS. 

The polymeric membrane has been prepared by dissolving its 

components in tetrahydrofurane in the following order: ionophore, PVC and 

plasticizer. We poured a few drops of the resulting solution on the surface 

of the metallic plate of the electrode. 

After the slow evaporation of the solvent (enclosed recipient) on the 

metallic surface remains the polymeric membrane sensible to anions, well 

attached to the PVC body of the electrode. 

In order to raise the calibration curve, the electrodes were conditioned 

by maintaining them in a NaLS 10 -2 M solution for 24 hours. 

When not used, the electrodes should be stored dry, in dark places. 

They should be reconditioned by washing with HCl 5x10 -3 M solution for 30 

minutes and then by magnetic stirring in a NaLS 10 -4 M solution before use. 

Construction of the electrode 

The technology for producing the “all solid state” sensors consists in 

forming the sensible PVC membrane directly on the metallic support, 

attaching it to the PVC electrode body. 

149

150 


The electrode is made of a PVC body with a copper or silver plate 

tightly attached at its bottom, which has attached the copper wire of the 

coaxial cable for coupling to the measuring device. On the plate surface there 

is the polymeric membrane which contains the ionophore and the plasticizer. 

This layout of the electrode has the great advantage of a solid internal 

contact because it does not require internal reference electrode and internal 

electrolyte. 

REFERENCES 

1. D. O. Hummel, Analyse der Tenside, Carl Hansen Verlag, München, Wien, 1995, 

chapters 4-5. 

2. M. Gerlache, J. M. Kauffmann, G. Quarin, J. C. Vire, G. A. Bryant, J. M. Talbot, 

Talanta, 1996, 43, 507. 

3. J. Sanchez, Critical Reviews in Analytical Chemistry, 2005, 35, 15. 

4. P. Săp, D. F. Anghel, C. Luca, Rev. Roum. Chim., 1983, 28, 883. 

5. W. Szczepaniak, M. Ren, Electroanalysis, 1994, 6, 341. 

6. L. Campanella, F. Mazzei, M. Tomassetti, R. Sbrilli, Analyst, 1988, 113, 327. 

7. W. Szczepaniak, Analyst, 1990, 115, 1451. 

8. J. Baró Romá, J. Sánchez, M, del Valle, J. Alonso, J. Bartroli, Sensors and 

Actuators B, 1993, 15-16, 179. 

9. R. Matesic-Puac, B. Sak-Bosnar, M. Bilic, B.S. Grabaric, Sensors and Actuators B, 

2005, 106, 221. 

10. N. M. Mikhaleva, E. G Kulapina,. Electroanalysis, 2006, 13-14, 1389. 

11. J. Sancez, M. del Valle, Talanta, 2001, 54, 893. 

12. T. Masadome, S. Kugoh, M. Ishikawa, E. Kawano, S. Wakida, Sensors and 

Actuators B, 2005, 108, 888. 

13. J. Vessel, K. Tulip, ˝Analysis with ion-selective electrodes˝, Ellis Horwood Ltd., 

Chichester, 1978, chapter 2. 

14. S. Alegret, J. Alonso, J. Bartroli, J. Baró-Romà, J. Sánchez, Analyst, 1994, 119, 

2319. 

15. E. Chifu, 2000, ˝Chimia coloizilor şi a interfeţelor (˝The colloid and interfaces 

chemistry˝), Ed. Presa Universitară Clujeană, Cluj-Napoca, 2000, chapter 3. 

16. D. Mihai, R. von Klitzing, D. F. Anghel, 11 th Physical Chemistry Conference with 

International Participation ROMPHYSCHEM 11, Timişoara, Romania, 2003. 

17. M. Gerlache, Z. Sentürk, J. C. Viré, J. M. Kauffmann, Anal. Chim. Acta, 1997, 349, 

59. 

18. N. Buschmann, H. Strap, Tenside Surf. Det., 1997, 34, 84. 

19. B. Kovács, B. Csóka, G. Nagy, A. Ivaska, Anal. Chim. Acta, 2001, 437, 6. 

20. G. J. Moody, J. D. R. Thomas, ˝Selective Ion Sensitive Electrodes˝, Merrow, Watford, 

1971, chapter 3.


NEW LC/MS METHOD FOR DETERMINATION OF 

PROGESTERONE IN HUMAN PLASMA FOR THERAPEUTIC 

DRUG MONITORING IN PREGNANCY AND 

GYNECOLOGICAL DISORDERS 

DAN MIHU a , LAURIAN VLASE a , SILVIA IMRE b , CARMEN M. MIHU a , 

MARCELA ACHIM a , DANIELA LUCIA MUNTEAN b 

ABSTRACT. A new simple, sensitive and selective liquid chromatography 

coupled with mass spectrometry (LC/MS) method for quantification of 

progesterone in human plasma was validated. The analyte was eluted in 1.9 

minutes on a reversed phase column (Zorbax SB-C18, 100 mm x 3.0 mm I.D., 

3.5 μm) under isocratic conditions using a mobile phase of a 20:80 (v/v) 

mixture of formic acid 0.1% (v/v) and methanol. The flow rate was 1 ml/min at 

the column temperature of 45 ºC. The detection of the analyte was in MS/MS 

mode using an atmospheric pressure chemical ionization source (APCI+, m/z 

315.2 → m/z 279.2). The sample preparation was very simple and rapid and 

consisted in plasma protein precipitation from 0.2 ml plasma using 0.5 ml 

methanol. Calibration curves were generated over the range of 0.8-80 ng/ml 

with values for coefficient of determination greater than 0.995 and by using 

a weighted (1/y 2 ) linear regression. The values of precision (coefficient of 

variation %) and accuracy (relative error %) were less than 9.4% and 14.2%, 

respectively, both for within- and between-run analysis. The mean recovery of 

the analyte was 98.6%. The developed LC/MS/MS method could be applied 

for determination of progesterone in human plasma for therapeutic drug 

monitoring in pregnancy and gynecological disorders. 

Keywords: progesterone, human plasma, LC/MS/MS, method validation 

INTRODUCTION 

Progesterone (PRG) is a steroid, secreted in large amounts by the 

corpus luteum and the placenta. It is an important intermediate in steroid 

biosynthesis in all tissues that secrete steroid hormones and small amounts 

enter the circulation from the adrenal cortex. It plays a key role in the female 

menstrual cycle (mainly produced after ovulation) and during pregnancy, 

when its production causes suppression of further ovulation and provides 

the correct environment for the developing embryo [1]. 

a 

University of Medicine and Pharmacy “Iuliu Hatieganu”, Emil Isac 13, RO-400023, Cluj-Napoca, 

Romania, vlaselaur@yahoo.com 

b 

University of Medicine and Pharmacy Targu-Mures, Gheorghe Marinescu 38, RO-540139, 

Targu-Mures, Romania

152 

D. MIHU, L. VLASE, S. IMRE, C. M. MIHU, M. ACHIM, D. L. MUNTEAN 

The analysis of steroid hormones in biological samples can be 

employed as a diagnostic tool in diseases promoted by disorders in the 

steroids profile. Progesterone is suitable to be monitored during treatment of 

infertility. Exogenous progestogens are administered in hormone replacement 

therapy and the modern, accurate and succesfully treatment involves hormon 

plasma level monitoring. This kind of therapeutic strategy has the problems of 

low concentration of steroids in plasma and the complexity of sample matrix, 

and demands for the development of highly selective and sensitive analysis 

methods. Progress in instrumental analytical chemistry and robust extraction 

techniques have enabled the detection of more compounds at lower 

concentrations, contributing to the success of different kind of therapies. 

Liquid Chromatography coupled with Mass Spectrometry (LC–MS) has 

been commonly selected for the analysis of the steroids due to its advantages 

of high sensitivity and selectivity of MS and allows a very short analysis 

run-time. Many applications have been reported, regarding steroid hormons 

determinations, including, determinations in water [2-9], tissues [10-12], 

food [13], cosmetics [14] and only a few in urine, serum or blood [15-17]. 

In the present study, we attempted to develop a fast HPLC/MS/MS 

method able to quantify PRG in human plasma after a simple protein 

precipitation for both physiological and therapy levels monitoring of PRG. 

Then, the developed method was applied to monitor PRG level in pregnant 

women or with gynecological disorders under or without PRG treatment. 


No significant interference at the retention time of PRG (1.4 min) 

(Figure 1) was observed in different human male plasma blank samples 

chromatograms due to the specificity of selected signals (Figure 2). 

Two ionization sources were tested, atmospheric pressure chemical 

ionization (APCI) and electrospray ionization (ESI), respectively. Finally, the APCI 

source was used due to the absence of pH-ionizable groups on progesterone 

molecule which determined a lower signal when electrospray ionization mode 

was applied. Different organic solvents were used for mobile phase: acetonitrile 

and methanol. At the same retention time for the analyte (obtained by using 

72% acetonitrile with 28% formic acid 0.1% or 80% methanol with 20% formic 

acid 0.1%), the signal intensity of PRG was about 3 times higher in case of 

methanol and for this reason, this was selected for further investigation. 

All the studied literature papers propose m/z 109.1 and/or 97.1 as 

daughter ions for monitoring. Our experiments demonstrated that, in this 

conditions, later eluting compounds interfere the determination (Figure 3, upper 

image), so a supplementary wash period is needed which extends the analysis 

with another six minutes. The selected monitoring ion m/z 279.2 alows a 

specific and sensitivie analysis in a very short run-time of 2 minutes (Figure 2 

and Figure 3 lower image).

NEW LC/MS METHOD FOR DETERMINATION OF PROGESTERONE IN HUMAN PLASMA … 

2.5 4 

Intens. 

x10 

2.0 

1.5 

1.0 

0.5 

0.0 

x10 4 

3 

2 

1 

0 

0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 Time [min] 

1 

EIC 279.2 +MS2(315.2), Smoothed (0.3,1, GA) 


Figure 1. Chromatograms of a blank human male plasma (upper image) and a 

plasma standard sample of 0.8 ng/ml PRG (RT 1.4 min)(lower image) 

Intens. 

x106 

2.0 

1.5 

1.0 

0.5 

0.0 

x105 

2.0 

1.5 

1.0 

0.5 

0.0 

97.2 

109.1 

119.1 145.2 159.1 215.1 239.2 255.1 

173.1 201.1 

+MS2(316.2), 1.1-1.1min #(274-276) 

279.2 

297.2 

315.3 

+MS2(315.2), 1.1-1.1min #(273-277) 

314.2 

100 125 150 175 200 225 250 275 300 m/z 

Figure 2. Mass spectra of progesterone: MS/MS non-reactive spectrum - isolation 

(upper image); MS/MS reactive spectrum - fragmentation (lower image) 

The applied calibration curve model proved to be linear over the 

concentration range 0.8 - 80 ng/ml PRG, with a determination coefficient 

greater than 0.995. The mean calibration curve, y = a (±SD) x + b (±SD) 

with SD standard deviation, was: y = 18097.2 (±1850.7) x – 1018.4 (±538.3), 

N = 8 calibration points, n = 5 determinations for each calibration point. The 

residuals had no tendency of variation with concentration and were between 

±15% values. 

153

154 

Intens. 

x105 

1.5 

1.0 

0.5 

0.0 

x104 

8 

6 

4 

2 

0 


EIC 97.1; 109.1 +MS2(315.2), Smoothed (0.3,1, GA) 


0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 Time [min] 

Figure 3. Chromatograms of the sample plasma with PRG (RT 1.4 min), different 

detection channels: transition 315.2>(97.1 + 109.1) (upper image) and transition 

315.2> 279.2 (lower image) 

The method had within- and between-run accuracy and precision 

(Tables 1 and 2), in agreement to international regulations regarding bioanalytical 

methods validation [18-20]. The lower limit of quantification was established at 

0.8 ng/ml PRG, with accuracy and precision less than 20% (Tables 1 and 2). 

The recovery was consistent and reproducible (Table 1 and 2). 

The analyte proved its stability after sample preparation for at least 

12 hours, the Bias% of found concentration being less than 15%, the maximum 

accepted value for method’s accuracy. 

Table 1. Within-run precision, accuracy and recovery for PRG (n = 5) 

Nominal 

concentration 

ng/ml 

Measured 

concentration ng/ml 

(± D.S.) 

Precision 

% 

Accuracy 

% 

Recovery % 

(± D.S.) 

0.80 0.88 0.04 4.5 10.0 102.0 4.5 

2.40 2.35 0.19 8.0 -2.0 101.6 4.8 

16.00 15.87 1.50 9.4 -0.8 94.0 5.5 

32.00 34.47 1.69 4.9 7.7 94.6 7.4 

Table 2. Between-run precision, accuracy and recovery for PRG (n = 5) 

Nominal 

concentration 

ng/ml 

Measured 

concentration ng/ml 

(± D.S.) 

Precision 

% 

Accuracy 

% 

Recovery % 

(± D.S.) 

0.80 0.92 0.07 7.4 14.2 98.6 8.4 

2.40 2.54 0.05 2.1 5.7 100.8 3.7 

16.00 16.61 0.76 4.5 3.8 98.8 7.6 

32.00 33.62 2.50 7.4 5.1 98.2 7.6


In comparison with the studied chromatographic – mass spectrometry 

methods about PRG determination in human plasma, the sensitivity of the 

proposed method (LLOQ of 0.8 ng/ml PRG in plasma) is better [16,17]. But 

the main advantage, except the short run-time of 2 minutes, is the sample 

preparation by protein precipitation and besides its simplicity, that sample 

treatment allows obtaining a very good recovery of analyte. The current LC/MS 

method does not intend to be a competitor or to replace the immunoassays 

methods, and is focused as an analytical tool for therapeutic drug monitoring 

of progesterone. In Table 3 a critical comparison between determination of 

progesterone by LC/MS and immunoassay, focused on the main advantages 

of the former technique, is given. 

Table 3. Comparison between determination of progesterone by 

LC/MS and immunoassay 

Parameter Current LC/MS Immunoassay methods 

Observations 

method 

EIA, ELISA [18] 

Selectivity Highly selective Less selective, possible LC/MS method is more selective 

overestimation due interference than immunoassays methods 

of compounds with very similar 

structure (cross-reaction) 

Sensitivity 0.8 ng/ml Depends on method/ Although immunoassay methods 

technique, usually 

are more sensitive, the aim of our 

between 0.02-0.2 ng/ml LC/MS method is therapeutic drug 

monitoring, and the expected 

levels of progesterone are in 

this case more than 10 ng/ml 

Upper limit of 80 ng/ml Depends on method / If concentration is above ULOQ, 

quantification 

technique, usually 

sample dilution is required and re- 

(ULOQ) 

between 5-50 ng/ml analysis, increasing the total 

analysis time and costs 

Matrix effects or Less susceptible to Highly susceptible to matrix LC/MS method is more robust 

interferences matrix effects 

effects due inter- and intraindividual 

differences in 

qualitative and quantitative 

serum composition; internal 

control required 

due relatively less matrix effects 

Analysis time / Less than 10 min per Usually 60-120 min per LC/MS is high-throughput 

throughput sample, including sample 

compared with immunoassays 

sample preparation 

and analysis 

methods 

Sample 0.2 ml plasma 0.02-0.2 ml serum Immunoassays methods usually 

volume 

require a smaller sample volume 

required 

in comparison with LC/MS method 

Reagents In house-preparation Special designed kits of LC/MS method is much 

preparation and low cost of reagents, about 10-50 folds cheaper regarding reagents 

and costs reagents 

higher costs as in case of costs compared with 

LC/MS 

immunoassays methods 

Shelf-life of The reagent is prepared Limited shelf-life of reagent kits, No problem with shelf-life of 

reagents in house as needed, typically 3-18 months in case of reagents in case of LC/MS method 

from components with un-opened recipient, less than 

shelf life more than 3 month after opening the 

3 years 

reagent bottle 

155


Regarding method’s applicability, as it can be seen from Table 4, 

the proposed analytical method demonstrated the increasing of the level of 

progesterone in human plasma in the luteal phase of menstrual cycle and 

it’s decreasing in menopause. During the pregnancy and also during the 

treatments with Utrogestan 100 mg PRG, the level of progesterone in human 

plasma is increased. 

Table 4. Levels of PRG in pregnant or in women with gynecological disorders 

with or without PRG treatment 

Subject Diagnostic Observations 

Age 22 years, 

Treatment with PRG 

PRG level 

found 

(ng/ml) 

S01 Pelvic inflamatory disease The luteal phase of teh 

menstrual cycle, day 22 

No 5.69 

S02 Endocervical polyp 

Age 52 years, 

Menopause from 5 years 

No


The proposed method allows evaluation of progesterone in human 

plasma in different phases of the menstrual cycle, menopause and also in the 

response of the progesterone replacement therapy, together to evaluation of 

pregnancies prognosis and the efficiency of Utrogestan therapy. 


Reagents 

Progesterone (PRG) was reference standards from Sigma-Aldrich 

(St. Louis, MO, SUA). Methanol and formic acid were Merck products (Merck 

KgaA, Darmstadt, Germany). Distilled, deionised water was produced by a 

Direct Q-5 Millipore (Millipore SA, Molsheim, France) water system. The human 

blank plasma was obtained from male volunteers. 

Standard solutions 

A stock solution of PRG with concentration of 2 mg/ml was prepared 

by dissolving appropriate quantities of reference substance (weighed on an 

Analytical Plus balance from Ohaus, USA) in 10 ml methanol. A working 

solution of 8000 ng/ml was then obtained by diluting specific volume of stock 

solution with plasma and it was further diluted with plasma to 80 ng/ml. 

Then these working solutions were used to spike different volumes of plasma 

blank, providing finally eight plasma standards with the concentrations ranged 

between 0.8 and 80 ng/ml. Accuracy and precision of the method was verified 

using plasma standards with concentrations 0.8, 2.4, 16 and 32 ng/ml PRG. 

Quality control samples (QC) at 2.4 (QCA), 16 (QCB) and 32 (QCC) ng/ml 

analyte will be used during clinical samples analysis. 

Chromatographic and mass spectrometry systems and conditions 

The HPLC system was an 1100 series model (Agilent Technologies) 

consisted of a binary pump, an in-line degasser, an autosampler, a column 

thermostat, and an Ion Trap SL mass spectrometer detector (Brucker Daltonics 

GmbH, Germany). Chromatograms were processed using QuantAnalysis 

software. The detection of the analyte was in MS/MS mode using an atmospheric 

pressure chemical ionization source (APCI), positive ionization, by monitoring 

the transition m/z 315.2 → m/z 279.2). Other apparatus parameters: capillary 

3500 V, vaporizer temperature 450 °C, nebulizer 60 psi, dry gas temperature 

300 °C, dry gas flow 7.00 L/min. Chromatographic separation was performed at 

45ºC on a Zorbax SB-C18 100 x 3 mm, 3.5 µm column (Agilent Technologies), 

protected by an in-line filter. 

Mobile phase 

The mobile phase consisted of a mixture of formic acid 0.1% (V/V) 

and methanol (20:80 v/v), each component being degassed, before elution, 

for 10 minutes in an Elma Transsonic 700/H (Singen, Germany) ultrasonic bath. 

The pump delivered the mobile phase at 1 ml/min. 

157

158 


Sample preparation 

Standard and test plasma samples were prepared as follows in order 

to be chromatographically analyzed. In an Eppendorf tube, to 0.2 ml plasma, 

0.5 ml methanol was added. The tube was vortex-mixed for 10 seconds and 

then centrifuged for 3 minutes at 10000 rpm. A volume of 150 µl supernatant 

was transferred in an autosampler vial and 30 µl were injected into the HPLC 

system. 

Analytical performance of the method 

The bio-analytical methods for utilization in pharmacokineticsbioavailability 

studies and therapeutic drug monitoring are conducted in 

concordance with FDA, EMEA and laboratory’s SOP regulations [19-26]. 

As a first step for the analytical performance determination of the 

method, specificity was verified using six different plasma blanks obtained 

from healthy human male volunteers. The progesterone level in men plasma 

not exceeds 0.1-0.4 ng/ml, however plasma with no detected PRG was used 

in order to obtain calibrators and quality control samples. 

The concentration of analytes was determined automatically by the 

instrument data system using the external standard method. Calibration was 

performed using singlicate calibration standards on five different occasions. 

The calibration curve model was determined by the least squares analysis. 

The applied calibration model was a linear one: y = ax + b, 1/y 2 weight, 

where y – peak area and x – concentration. Distribution of the residuals (% 

difference of the back-calculated concentration from the nominal concentration) 

was investigated. The calibration model was accepted, if the residuals were 

within ±20% at the lower limit of quantification (LLOQ) and within ±15% at 

all other calibration levels and at least 2/3 of the standards met this criterion, 

including highest and lowest calibration levels. 

The lower limit of quantification was established as the lowest calibration 

standard with an accuracy and precision less than 20%. 

The within- and between-run precision (expressed as coefficient of 

variation, CV%) and accuracy (expressed as relative difference between 

obtained and theoretical concentration, Bias%) of the assay procedure 

were determined by analysis on the same day of five different samples at 

each of the lower (2.4 ng/ml), medium (16 ng/ml), and higher (32 ng/ml) 

levels of the considered concentration range and one different sample of 

each on five different occasions, respectively. 

The relative recoveries at each of the previously three levels of 

concentration and limit of quantification were measured by comparing the 

response of the treated plasma standards with the response of standards in 

solution with the same concentration of analytes as the prepared plasma 

sample.


The post-preparative stability (PPS) in the autosampler of the analytes 

in human plasma was investigated at lower (2.4 ng/ml) and higher concentration 

(32 ng/ml) for 12 hours, the expected longest storage times of the samples 

in autosampler before injection. The requirement for stable analytes was 

that the difference between mean concentrations of the tested samples in 

various conditions and nominal concentrations had to be in ±15% range. 

Clinical application 

The developed method was verified by analyzing different plasma 

samples obtained from pregnant women or with gynecological disorders, 

under treatment with vaginal capsules of PRG (Utrogestan 100 mg vaginal 

capsules) or without treatment. 

REFERENCES 

1. R. C. Tuckey, Placenta, 2005, 26, 273. 

2. H. Chang, S. Wu, J. Hu, M. Asami, S. Kunikane, J. Chromatogr. A., 2008, 1195, 44. 

3. E. P. Kolodziej, J. L. Gray, D. L. Sedlak, Environ. Toxicol. Chem., 2003, 22, 2622. 

4. E. P. Kolodziej, D. L. Sedlak, Environ. Sci. Technol., 2007, 41, 3514. 

5. M. Kuster, M. J. Lopez de Alda, M. D. Hernando, M. Petrovic, J. Martin-Alonso, 

D. Barcelo, J. Hydrol., 2008, 358, 112. 

6. M. J. Lopez de Alda, D. Barcelo, J. Chromatogr. A, 2000, 892, 391. 

7. M. Sole, M. J. Lopez de Alda, M. Castillo, C. Porte, K. Ladegaard-Pedersen, 

D. Barcelo, Environ. Sci. Technol., 2000, 34, 5076. 

8. B. J. Vanderford, R. A. Pearson, D. J. Rexing, S. A. Snyder, Anal. Chem., 2003, 

75, 6265. 

9. A. Yamamoto, N. Kakutani, K. Yamamoto, T. Kamiura, H. Miyakoda, Environ. Sci. 

Technol., 2006, 40, 4132. 

10. D. Caruso, S. Scurati, O. Maschi, L. De Angelis, I. Roglio, S. Giatti, Neurochemistry 

International, 2008, 52, 560. 

11. T. Higashi, A. Nagahama, Y. Mukai, K. Shimada, Biomed. Chromatogr., 2008, 

22, 34. 

12. P. E. Joos, M. V. Ryckeghem, Anal. Chem., 1999, 71, 4701. 

13. Y. Yang, B. Shao, J. Zhang, Y. Wu, J. Ying, J. Chromatogr. B, 2008, 870, 241. 

14. D. De Orsi, M. Pellegrini, S. Pichini, D. Mattioli, E. Marchei, L. Gagliardi, J. Pharm. 

Biomed. Anal., 2008, 48, 641. 

15. B. Alvarez Sanchez, F. Priego Capote, J. Ruiz Jimenez, M. D. Luque de Castro, 

J. Chromatogr. A, 2008, 1207, 46. 

159

160 


16. N. Janzen, S. Sander, M. Terhardt, M. Peter, J. Sander, J. Chromatogr. B, 2008; 

861, 117. 

17. S. M. Zhang, S. R. Mada, S. Sharma, J. Pharm. Biomed. Anal., 2008, 48, 1174. 

18. http://www.biocompare.com/ProductListings/29520/Progesterone-ELISA-EIA- 

Kits.html 

19. The European Agency for the Evaluation of Medicinal Products, Note for 

Guidance on the Investigation of Bioavailability and Bioequivalence, 

CPMP/EWP/QWP/1401/98, 2001, London, UK, http://www.emea.europa.eu 

/pdfs/human/qwp/140198enfin.pdf 



and Bioequivalence studies for orally administered drug products - general 

considerations, 2003, Rockville, USA, 

http://www.fda.gov/cder/guidance/5356fnl.pdf 

21. U.S. Department of Health and Human Services, Food and Drug Administration, 

Guidance for Industry – Bioanalytical Method Validation, 2001, 

http://www.fda.gov/cder/guidance/4252fnl.pdf 

22. L. Vlase L, S. E. Leucuta, S. Imre, Talanta, 2008, 75(4), 1104. 

23. L. Vlase, A. Leucuta, D. Farcau D, M. Nanulescu, Biopharmaceutics & Drug 

Disposition, 2006, 27(6), 285. 

24. L. Vlase, S. Imre, D. Muntean, S. E. Leucuta, Journal Of Pharmaceutical And 

Biomedical Analysis, 2007, 44(3), 652. 

25. L. Vlase, D. Muntean, S. E. Leucuta, I. Baldea, Studia Universitatis Babes-Bolyai 

Chemia, 2009, 54(2), 43. 

26. A. Butnariu, D. S. Popa, L. Vlase, M. Andreica, D. Muntean, S. E. Leucuta, Revista 

Romana de Medicina de Laborator, 2009, 15(2), 7.


PHOTO-CATALYSTS BASED ON GOLD - TITANIA COMPOSITES 

ANCA PETER a , MONICA BAIA b , FELICIA TODERAS b , MIHAELA 

LAZAR c , LUCIAN BARBU TUDORAN d , VIRGINIA DANCIU e 

ABSTRACT. Porous TiO2 aerogels with different Au colloidal particles 

concentrations were synthesized and their functionality to decontaminate the 

water was evaluated using model pollutants. It was showed that the decrease 

of noble metal concentration determines a decrease of the composites pore 

size and specific surface area and of the photo-degradation apparent rate 

constant. The morphological (porosity, surface area, TEM and TGA) particularities 

of the synthesized porous composites were also briefly discussed from the 

perspective of the photo-catalytic results. Since the photodecomposition rate 

depends on the [OH]surface adsorbed on the surface, additional measurements 

have been performed. 

Keywords: Au -TiO2 composite, porosity, TEM microscopy, photo-catalysis, 

salicylic acid 

INTRODUCTION 

TiO2 photo-catalysis has been improved by numerous investigations 

in recent years, particularly owing to its application for the complete 

mineralization of almost all organic contaminants to carbon dioxide, water 

and inorganic constituents [1, 2]. The advantages of using TiO2 are its nontoxic 

nature and stability, but it is also significant that the low rate of 

electron transfer to oxygen and the high recombination rate of electron-hole 

pairs limit the rate of organic compounds photo-oxidation on the catalyst 

surface [3]. Many investigations have reported that the addition transition 

metals to TiO2 are two ways to enhance the photo-catalytic reaction rate [3-9]. 

The noble metals as gold [4, 5], platinum [6, 7] and silver [9] were usually 

a 

North University, Faculty of Science, Department of Chemistry-Biology, 430083, Baia Mare, 

Romania, peteranca@yahoo.com 

b 

Babes-Bolyai University, Faculty of Physics, 400084, Cluj-Napoca, Romania 

c 

National Institute for Research and Development of Isotopic and Molecular Technologies, 

Donath 71 – 103, RO-400293, Cluj-Napoca, Romania 

d 

Faculty of Biology and Geology, Electron Microscopy Center, Babes-Bolyai University, 5-7 

Clinicilor Str., 400006, Cluj-Napoca, Romania 

e 

Babes-Bolyai University, Faculty of Chemistry and Chemical Engineering, 400028, Cluj-Napoca, 

Romania

162 

A. PETER, M. BAIA, F. TODERAS, M. LAZAR, L. B. TUDORAN, V. DANCIU 

used to produce the highest Schottky barriers among the metals, in order to 

facilitate the electron capture. Moreover, the research studies [10, 11] have 

demonstrated that the gold particles interact with TiO2 surface oxygen 

vacancies (note that on an oxygen vacancy site, two neighbouring Ti atoms 

are reduced from Ti 4+ to Ti 3+ ), thus reducing the surface defects. 

Due to their very low density, high surface area, translucency or 

transparency to the visible light and to their microstructure, the TiO2 aerogels 

are the preferred materials as supports for noble metals. On the other hand, 

the photo-catalytic activity of Au-TiO2 composites has been described as 

strongly influenced by morphological and structural parameters, such as their 

structure, surface area, porosity, gold particle size and surface hydroxyl group 

density [12, 13]. 

In this paper, we study the photo-catalytic behaviour of a series of 

Au - TiO2 composites in the aqueous salicylic acid (SA) oxidation. The goal 

of this study was to investigate the influence of the Au loading on the photocatalytic 

activity of the Au –TiO2 composites. 

RESULTS AND DISSCUTION 

Morphology and Structure of the Composites 

The morpho-structural characteristics of the Au – TiO2 composites 

are presented in Table 1. In columns 2 and 3 of the Table 1 are indicated the 

gold colloid volumes used to the TiO2 gels impregnation and gold content, 

respectively. One observes an increase of the Au content as the colloid 

volume increases from 20 to 75 ml. By using a volume higher than 75 ml 

Au colloid, the Au content slowly decreases, probably due to the saturation 

of the TiO2 surface with Au particles (Figure 1). 

The pore size distribution of the obtained composites is presented in 

Figure 2 A (a-e). The Au -TiO2 composites have a porous structure containing 

pores ranging from 8 to 55 Å, while TiO2 aerogel contains pores ranging from 

20 to 80 Å. So, the Au presence in the composites reduces the composite’s 

porosity. This behaviour was attributed to the insertion of a small amount of 

gold nano-particles into the pores. This insertion occurs without damaging the 

pore structure. Similar changes have been reported in the surface structure of 

porous titania with the insertion / deposition of ruthenium and platinum nanoparticles 

by the sono-chemical method [14, 15]. The decrease of porosity by 

Au deposition can be observed also by analyzing the pore volume values 

from Table 1. It is interesting to notice that the porosity decreases with the 

decrease of Au content (Figure 2B). The B1Au and B2Au composites with 

similar Au content (0.395 % and 0.397%, respectively) have the major pore 

size ranging from 35 to 40 Å while B4Au composite with 0.131% Au, which 

contains predominantly 15 A pores. At low Au concentrations, pores with 20 Å 

diameters are dominant, whereas composites with high Au content (samples 

B1Au and B2Au) contain pores with diameters higher than 20 Å.

Au - TiO2 

ID 


Table 1. Morpho-structural particularities and photo-catalytic behavior of the 

studied Au-TiO2 composites 

VAu coll. sol. 

(ml) 

[Au] 

(% mol) 

SBET 

(m 2 /g) 

Vpore 

[OH]surf 

(cm 3 /g) (mmol/g) 

kads x 

10 3 

(min -1 ) 

kphoto 

x 10 3 

(min -1 ) 

B1 Au 150 0.395 477 0.86 0.56 0.23 10.9 63 

B2 Au 75 0.397 450 0.44 0.61 0.24 11 59 

B3 Au 40 0.373 332 0.38 0.52 0.12 10.6 52 

B4 Au 25 0.131 337 0.95 0.57 0.04 8.1 50 

TiO2 - - 340 1.08 0.6 0.09 10.1 60 

Au content [% mol] 

0.40 

0.35 

0.30 

0.25 

0.20 

0.15 

0.10 

20 40 60 80 100 120 140 160 

Au colloid volume [ml] 

Figure 1. Dependence of Au content on the Au colloid volume used 

inTiO2 gel impregnation 

The specific surface area BET of the obtained composites increases 

with Au content (Table 1). These effects have been previously reported for 

gold supported on MCM-41 and MCM-48 zeolites [16] and for gold supported 

on alumina and ceria [17], and are explained through the expansion of the 

porous structure of the support due to of the introduction of gold nanoparticles 

[18]. 

X 

(%) 

163

164 

(a) 

V relative [%] 

(c) 


(e) 


20 

15 

10 

5 

0 

20 

15 

10 

5 

0 

50 

40 

30 

20 

10 

0 


B 1 Au composite 

20 40 60 80 

Pore size [angstrom] 

20 40 60 80 



20 40 60 80 


TiO 2 

(b) 


(d) 


35 

30 

25 

20 

15 

10 

5 

0 

30 

25 

20 

15 

10 

5 

0 


20 40 60 80 


20 40 60 



Figure 2A. Influence of the Au loading on the composites pore structure.


B 1 Au 

B 2 Au 

B 3 Au 

B 4 Au 

Au loading [% mol] 

TiO 2 

10 

0 

40 

30 

20 

50 

major pore predominance [%] 

major pore size [angstrom] 

Figure 2B. Variation of major pore size and major pore predominance 

with Au loadings. 

According to TEM pictures analysis, the composite B1Au (Figure 3) has 

an amorphous TiO2 structure. The Au particles (dark points) are heterogeneously 

dispersed on the TiO2 surface. In some areas, the Au colloidal particles were 

agglomerate into clusters (indicated by the arrow from Figure 3), while in other 

regions individual Au particles with a 15 nm diameter (determined from TEM 

analyses) were detected. The association of Au nano-particles in large clusters 

reduced the TiO2 photo-excitation, due to the fact that all its active centres are 

occupied with Au particles. This behaviour will, subsequently, cause a decrease 

of the SA photo-degradation rate. 

Figure 3. TEM image of the B1Au composite (bar 100 nm). 

165


In Figure 4 presents the Au -TiO2 composites weight decrease after 

heat treatment at temperatures increasing from 25 to 500 0 C. 

The weight loss at t < 200 0 C is due to the first dehydration step, 

resulted from physically adsorbed water, and the elimination of the ethanol 

traces [19]. It can be observed that the mass loss increases with decrease 

of the Au content. This is explained by the fact that the Au nano-particles 

from TiO2 surface inhibit the network shrinkage in thermal treatment. At 

250-300 0 C, a second dehydration step resulted from structural / ligand water 

and thermal decomposition of the un-reacted titania precursors, takes place. 

Again, at this stage, the mass loss increases with decrease of the Au content. 

This is explained by the fact that the higher number of Au nano-particles from 

TiO2 surface stabilize the aerogel network, by forming a metal skeleton around 

the TiO2 structure. Moreover, at temperatures higher than 300 0 C, the aerogel 

begins to become crystalline. By comparing the weight loss of the TiO2 and 

Au -TiO2 composites, one observes that the TiO2 network is more stable 

than the composites network. The weight loss in the case of TiO2 aerogels 

is lower than that observed for the composites. 

Mass decrease [ % ] 

100 

95 

90 

85 

80 

75 

70 

65 

0 100 200 300 400 500 

Temperature [ 0 C ] 

% (B1 Au) 

% (B2 Au) 

% (B3 Au) 

% (B4 Au) 

% (TiO 2 ) 

Figure 4. Influence of the Au loading on the thermal behavior 

of the Au-TiO2 composites. 

Catalytic Activity 

In Table 1, there are presented some catalytic parameters (salicylic 

acid adsorption and photo-degradation apparent rate constants and photodegradation 

efficiency) which are determinated for the SA photo-decomposition 

process. The adsorption constants were calculated from the slope of the plot 

166


ln (Qt) vs. time (Qt – current adsorbed quantity) after applying a linear fit [20]. 

The photodecomposition rate constants were obtained from the slope of the 

plot ln (C0/C) vs. time after applying a linear fit [21, 22]. The photo-degradation 

efficiency was calculated with the following equation: 

X (%) = (C0 – C) / C0 x 100 (1) 

where: C0 – SA initial concentration, C– SA concentration after 150 minutes 

iradiation [23]. 

The more intense adsorption of SA occurs on B1Au and B2Au 

composites with the highest Au content (Table 1 and Figure 5). The adsorption 

rate constants decrease with decrease of the Au loading, in the same manner 

as the specific surface area and pore volume. Moreover, the [OH]surface is 

relatively high for those two composites, except for the B4Au sample which 

has the highest [OH]surface. Even if the pore volume and [OH]surface of B4Au 

composite are highest than those of the other composites, the adsorption rate 

constant is very low. This is explained by the fact that the B4Au composite 

contains, in the major preponderance, pores smaller than 20 Å diameter 

(micro-porous structure) which do not permit the penetration of the SA molecules 

having a higher surface area (35.52 Å 2 ), due to the steric impediment [24]. So, 

in this case SA adsorption and photo-decomposition take place only on geometric 

surface and not in the real surface. This observation is available also for the 

TiO2 aerogel and shows that the adsorption intensity is deeply influenced by 

the specific surface area and pore size. 

The SA photo-degradation rate constants vary with Au loading in the 

same manner as the adsorption constants (Table 1 and Figure 5). The SA 

photo-degradation on B1Au, B2Au and B3Au composites occurs with a higher 

rate, due to the fact that SA adsorption on these composites was more intense. 

The SA photo-degradation apparent rate constants increase with the specific 

surface area, pore size and [OH]surface. In B1Au and B2Au composites, pores of 

about 40 Å are predominant, thus the SA molecules may easily get into the 

composite network, increasing the SA photo-degradation rate. 

By comparing the relationship between Au loading and photo-degradation 

rate constant in the case of B2Au and B4Au composites, one observes that 

whereas the Au content increase by almost three times, the photo-degradation 

rate constant increase by 1.35 times. This indicates a non-linear dependence 

of photo-degradation rate on Au loading on the composites and challenges 

to new researches in order to establish that this dependence is kept up for 

Au loadings higher than 0.397%. 

The photo-degradation efficiency (Table 1) decreases with the decrease 

of the adsorption and photo-degradation rate constants and is influenced 

also by the specific surface area, pore size and [OH]surface, which, subsequently, 

are induced by the Au loading. 

167

168 


k ads x 10 3 [min -1 ] 

0.25 

0.20 

0.15 

0.10 

0.05 

B 4 Au 

photodegradation profile 

adsorption profile 

0.00 

0.10 0.15 0.20 0.25 0.30 0.35 0.40 

Au content [% mol] 

B 3 Au 

B 1 Au 

Figure 5. Influence of Au loading on SA adsorption and photo-degradation rate 

constants (B 1 Au – 0.395% Au, B 3 Au – 0.373% Au, B 4 Au – 0.131% Au) 

CONCLUSIONS 

Au – TiO2 composites with different Au loadings were prepared and 

characterized by nitrogen adsorption-desorption method, TEM microscopy 

and thermo-gravimetric analyses. Additionally, the OH group’s concentration 

was measured. The catalytic activity of the obtained composites was tested 

in the SA photo-degradation process. 

• The pores size varies inversely proportionate with the Au loading. 

• The specific surface area, SA adsorption and photo-degradation rate 

constants vary proportionate with Au loading. 

• The SA adsorption and photo-degradation rates are deeply influenced 

by the pore size and specific surface area of the Au-TiO2 composites. 


Composites preparation 

At first, the TiO2 gels were prepared by sol-gel method. The TiO2 

sols were obtained by mixing titanium-isopropoxide (IV) (Merck, 99.9%), 

with anhydrous ethanol (Fluka, 99.8%), ultra pure water and nitric acid 

reagent (Merck, 65%) as catalyst. The molar ratio of reactants was: 

[Ti(OC3H7)4]:[H2O]:[C2H5OH]:[HNO3] = 1:3.675:21:0.08. 

12 

10 

8 

k photo x 10 3 [min -1 ]


The transparent TiO2 gels were kept for 3 days in a gold colloidal 

solution, obtained by HAuCl4 10 -3 M (Merck, 99.8%) reduction with sodium 

citrate 38,8 x 10 -3 M (Merck). The colloidal solution volumes used for TiO2 gel 

impregnation were between 25-150 ml (Table 1). 

The obtained Au – TiO2 gels were supercritically dried using liquid 

CO2 in a SAMDRI – 790 A (Tousimis) dryer (T = 40 0 C and p = 1400 psi). 

The composite aerogels were transparent, with a violet nuance. The gold 

contents for the obtained composites are presented in Table 1. 

Characterisation Techniques 

The surface area and the pore volume of the as-prepared composites 

were determined by the Brunauer-Emmett-Teller (BET) method. A Sorptomatic, 

Thermo Electron Corporation system coupled with a Flatron L 1718S computer 

system was used. The partial pressure range for surface area calculation 

was 0.05 < P/P0 < 0.3. The nitrogen adsorption was carried out at 77 K. 

Before each measurement, the composites were degassed at p = 0.5 Pa 

and T = 333 K for 2 h. 

A transmission electron microscope (TEM) Jeol JEM 1010 operating 

at an accelerating voltage of 100 kV was employed to obtain bright field 

images. Au particle sizes were measured from images using SIS software 

after calibration. 400 mesh cooper grid covered with a plain carbon film, 

obtained by vacuum evaporation on freshly cleaved mica, were used to 

support a drop of ultrasonic dispersed composite powder in distilled water. 

The TEM images were recorded with a MegaView III CCD camera. 

The thermo-gravimetric analyses were performed using a Thermal 

Analysis System Mettler Toledo with a thermo-gravimetric cell TGA / SDTA 851 

(heating rate 5 0 C / min, in nitrogen flow) and a thermal analysis cell DSC 822. 

The [OH]surface on the TiO2 aerogels was determined by spectrophotometrical 

evaluation of the Fe (III) acetyl-acetonate concentration in 

toluene solution after its partial coupling with [OH]surface (reaction 1) [25]. 

(3-x)(S– OH) + Fe(acac)3 → (S – O)3-xFe(acac)x + y(TiS – OH) + (3-x)Hacac (1) 

where: S - aerogel surface, Hacac- pentane-2,4-dione, solvent – toluene. 

The concentration of gold in all composites was determinated using 

an FAAS 800 Flame Atomic Adsorption Spectrometer. The sample preparation 

consists in the chemical treatment with 5 ml of acid mixture (HNO3 37%: HF = 

4:1 volume proportion). The mixture was homogenized and brought to 25 ml 

volumetric flask with distilled water and than filtered 

Catalytic Activity 

The adsorption properties and the photocatalytic activity of the Au -TiO2 

composites was established from the adsorption and photo-oxidation rate 

of salicylic acid used as a standard pollutant molecule [26]. In all adsorption 

and photo-oxidation processes 0.05 g composite was used. The decrease in 

169

170 


salicylic acid concentration (C0 = 5.25 x 10 -4 M for all investigated composites) 

was monitored by UV-Vis spectroscopy (λ = 297 nm). The composites immersed 

in salicylic acid solution were irradiated with a medium pressure Hg lamp HBO 

OSRAM (500 W). A Teflon photochemical cell with quartz window (S = 12 cm 2 ) 

and a volume of 8 ml was also used. The distance between the photochemical 

cell and the lamp was about 30 cm. The working temperature was of 20-22 0 C 

and the solution pH was 5.3. Before the UV irradiation as well as before the 

UV-Vis measurements, the cell with the SA solution and composite was kept 

in dark for 30 min in order to achieve the equilibrium of the SA adsorptiondesorption 

process [15]. One should emphasize that throughout the 

photodecomposition process no shift of the UV band located at 297 nm was 

observed and no other new absorption band occurred. 


This research was supported by a TD No. 44 / 2006 project. 

REFERENCES 

1. F.B. Li, X.Z. Li, Appl. Catal. A: General , 2002, 228, 15. 

2. M.R. Hoffmann, S.T. Martin, W.Y. Choi, Chem. Rev., 1995, 95, 69. 

3. A. Linsebigler, G. Lu, J.T. Yates, Chem. Rev., 1995, 95, 735. 

4. C.Y. Yang, C.Y. Liu, J. Chen, J. Colloid. Interf. Sci., 1997, 191, 464. 

5. C.Y. Yang, C.Y. Liu, X. Zheng, Colloid. Surf. A, 1998, 131, 271. 

6. A. Sclafani, L. Palmisana, G. Marci, Sol. En. Mater. Sol C, 1998, 51, 203. 

7. J.C. Yang, Y.C. Kim, Y.G. Shul, Appl. Surf. Sci., 1997, 121/122, 525. 

8. W. Choi, A. Termin, M.R. Hoffmann, J. Phys. Chem., 1998, 98, 13669. 

9. P. Falaras, I.M. Arabatzis, T. Stergiopoulos, M.C. Bernard, Intern. J. Photoen., 

2003, 5, 123. 

10. M.S. Chen, D.W. Goodmann, Topics in Catal., 2007, 44, 1-2, 41. 

11. A. Kolmakov, D.W. Goodmann, Catal. Lett., 2000, 70, 93. 

12. X. Chen, S.S. Mao, Chem. Rev., 2007, 107, 2891. 

13. S. Sakthivel, M.C. Hidalgo, D.W. Bahnemann, S-U Geissen, V. Muguresan, A. 

Vogelpohl, Appl. Catal. B, 2006, 63, 31. 

14. N. Perkas, Z. Zhong, L. Chen, M. Besson, A. Gedanken, Catal. Lett., 2005, 103, 9. 

15. N. Perkas, D. Minh Pham, P. Gallezot, A. Gedanken, M. Bessot, Appl. Catal. 

B, 2005, 59, 121.


16. Z. Kónya, V.F. Puntes, I. Kiricsi, J. Zhu, J.W. Ager, M.K. Ko, H. Frei, P. Alivisatos, 

G.A. Somorjai, Chem. Mater, 2003, 15, 1242. 

17. M.I. Dominguez, M. Sanchez, M.A. Centeno, M. Montes, J.A. Odriozola, Appl. 

Catal. A, 2006, 302, 96. 

18. M.A. Centeno, M.C. Hidalgo, M.I. Dominguez, J.A. Navió, J.A. Odriozola, 

Catal. Lett., 2008, 123, 198. 

19. S.A. Selim, Ch.A.Philip, S. Hanafi, H.P. Boehm, J. Mat. Sci., 1990, 25 (11), 4678. 

20. S.M. Ould-Mame, O. Zahraa, M. Bouchy, Inter. J. Photoen., 2000, 2, 59. 

21. W. Lee, H-S. Shen, K. Dwight, A. Wold, J. Sol. State Chem., 1993, 106 (2), 288. 

22. J. Papp, H. S. Shen, R. Kershaw, K. Dwight, and A. Wold, Chem. Mater., 1993, 

5(3), 284. 

23. N.A. Laoufi, D. Tassalit, F. Bentahar, Global Nest Journal, 2008, 10(3), 404. 

24. http://www.chemspider.com/331#suppinfo. 

25. J.A. Rob van Veen, F.T.G. Veltmaat, G. Jonkers, J. Chem. Soc., Chem.Commun., 

1985, 1656. 

26. M. Tomkiewicz, Catal. Today, 2000, 58, 115. 

171


BIOSORPTION OF PHENOL FROM AQUEOUS SOLUTIONS BY 

FUNGAL BIOMASS OF PHANEROCHAETE CHRYSOSPORIUM 

TÍMEA PERNYESZI a , KRISZTINA HONFI a , BORBALA BOROS a , 

KATALIN TÁLOS a , FERENC KILÁR a , CORNELIA MAJDIK b 

ABSTRACT. The biosorption of phenol from aqueous solution on nonliving 

mycelial pellets of Phanerochaete chrysosporium was studied using 

batch technique with respect to pH, initial concentration and biomass 

dosage. Ph. chrysosporium was grown in a liquid medium with a simple 

constitution. The phenol biosorption studies on fungal biomass was carried 

out at an initial pH of 5. Adsorption kinetics was characterized at an initial 

concentration of 12.5, 25 and 50 mg/L in a suspension concentration of 5.0 g/L. 

The sorption process followed the second-order kinetics. Phenol adsorption 

isotherms were determinated on fungal biomass at biomass concentration 

of 1.0 and 5.0 g/L and initial pH of 5. The adsorption equilibrium of phenol 

from aqueous solutions by mycelial pellets could be well described with 

Freundlich equation. The adsorption capacity of phenol and the Freundlich 

constant decreased with increasing biomass concentration. 

Keywords: phenol, biosorption, Phanerochaete chrysosporium biomass, 

adsorption isotherm, adsorption kinetics, water treatment 

INTRODUCTION 

Wastewaters containing phenolic compounds present a serious problem. 

Wastewaters containing phenol cannot be discharged in surface waters without 

prior treatment due to the phenol toxicity. The toxic and hazardous nature 

of phenols and their associated derivatives, and their increasing amounts 

in industrial wastewaters have been documented. They are known human 

carcinogens. Phenolic compounds are present in the wastewaters generated 

by paint, solvent, petroleum, coal-conversion, pharmaceutical, wood preserving 

chemicals, plastic, rubber-proofing, pesticide, iron-steel, paper and pulp industries. 

Traditionally, adsorption on activated carbon and polymer based 

adsorbents is the most widely used technique for the removal of phenols. The 

high cost of activated carbon and polymer has stimulated interest to use cheaper 

a 

University of Pécs, Faculty of Science, Department of Analytical and Environmetal Chemistry, 6 

Ifjúság, H-7624 Pécs, Hungary 

b 

University Babeş-Bolyai; Faculty of Chemistry and Chemical Engineering, 11 Arany J., RO- 

400293 Cluj-Napoca, Romania

174 

T. PERNYESZI, K. HONFI, B. BOROS, K. TÁLOS, F. KILÁR, C. MAJDIK 

raw environmental-friendly materials as adsorbents. Recently, microorganisms 

have been considered as one of the most promising adsorbents [1-12]. 

However, information on fungal interacting with toxic phenolic compounds is 

still limited. In the concept of biosorption, several chemical processes may 

be involved, such as adsorption, ion exchange, and covalent binding. The 

biosorptive sites on the microorganisms are carboxyl, hydroxyl, sulphuryl, 

amino and phosphate groups [4,5]. Fungal cell walls and their components 

have a major role in biosorption. Aksu and Yener evaluated the biosorption 

of phenol and monochlorinated phenols on the dried activated sludge [6]. 

Ning et al. reported that anaerobic biosorption of 2,4-dichlorophenol was mainly 

a physicochemical process. They studied the equilibrium sorption isotherms 

and sorption kinetics of 2,4-DCP on live and chemically inactivated anaerobic 

biomass [7]. Rao and Viraraghavan have used nonviable pretreated cells of 

Aspergillus niger to remove phenol from an aqueous solution, and observed 

that maximum removal of phenol occurred at an initial pH of 5.1 [8]. Other 

workers investigated the biosorption capacity of dead and live fungal biomass, 

and they found, that better removal was achieved with dead fungal biomass 

than with live one [9-11]. 

Wu and Yu have used Phanerochaete chrysosporium biomass as a 

sorbent material for removal of phenol and chlorophenols [12,13]. They found 

that the sorption capacity on mycelial pellets increased in order: phenol 

< 2-CP < 4-CP < 2,4-CP. The adsorption increased with decreasing water 

solubility and increasing octanol-water partitioning coefficients. The presence 

of 2-CP or 4-CP and the initial concentration of 2-CP and 4-CP had no 

significant effect on the sorption of 2,4-DCP on fungal mycelial pellets. 

These suggests that partitioning was largely involved in biosorption mechnisms, 

and that hydrophobicity might govern the biosorption of phenolic compounds 

by mycelial pellets [4,12,13]. 

The objectives of this study were: 

1. to test the biomass of Phanerochaete chrysosporium grown in a 

medium having simple constitution for phenol biosorption, 

2. to evaluate the influences of different experimental parameters on 

biosorption such as initial pH, sorption time and initial phenol 

concentration using batch technique, 

2 to model phenol biosorption kinetics by mycelial pellets using pseudofirst-order 

and second-order kinetic equations, 

3 to determine adsorption isotherms using batch technique and analyse 

the adsorption equilibrium using Freundlich-equation, 

4 to investigate the effect of biomass concentration on biosorption 

process.

BIOSORPTION OF PHENOL FROM AQUEOUS SOLUTIONS BY FUNGAL BIOMASS ... 


Factors influencing biosorption of phenol by mycelial pellets 

Effect of initial pH on phenol biosorption on Phanerochaete 

chrysosporium in aqueous suspension 

The effect of initial pH on the equilibrium uptake capacity of phenol 

by mycelial pellets of P. chrysosporium at pH values between 3.0 and 10.0 

and 22.5 ± 2 °C is shown in Figure 1. The biosorption of fungi was influenced 

by pH in a range of 3.0 – 10.0. The initial concentration of phenol was 25 mg/L 

and the biomass concentration was 0.5 g/L. The maximal adsorbed phenol 

amount was q max = 0.073 mg/g at pH 4. In the natural state of phenol solution, 

pH 5.0, without pH adjustment, the adsorption was slightly reduced, the 

adsorbed amount of phenol was 0.065 mg/g. The adsorption was reduced 

in an alkaline medium and slightly reduces in an acidic medium. The effect 

of pH on the adsorption of phenol was not significant at pH 3.0 – 6.0, and 

the uptake of phenol in the same pH interval was larger than that observed 

at other pH values. Further biosorption experiments were carried out at the 

natural state of pH 5 in the biomass suspensions. 

n σ(v) (mg/g) 

0.075 

0.07 

0.065 

0.06 

0.055 

0.05 

0.045 

0.04 

0.035 

0.03 

2 4 6 8 10 

pH 

Figure 1. The pH effect over the phenol biosorption on Phanerochaete 

chrysosporium biomass. The biomass concentration is 

0.5 g/L and the initial phenol concentration is 25 mg/L. 

Phenol is weakly acidic, and pH has a significant effect on the 

degree of ionization of phenol and the cell surface properties. The amount 

of adsorbed phenol seemed to be related to the dissociation constant (pKa), 

which is 9.9 for phenol [14]. The ionic fraction of phenolate ion increases 

which increasing pH, and phenol could be expected to become more 

negatively charged as pH increases. The surface charge on fungal biomass 

175

176 


is predominately negative at pH 3.0 – 10.0 [8, 13]. At pH < 3.0, the overall 

surface charge on fungal cells become positive. Thus, phenol can be adsorbed 

to a lesser extent at pH ≥ pKa , due to the repulsive forces prevailing at higher 

pH values. A lower pH value resulted in a higher undissociated fraction of 

phenol and led to a decrease in phenol uptake by the mycelium pellets. 

Adsorption kinetics of phenol on Phanerochaete chrysosporium 

biomass in aqueous suspension 

The phenol adsorption kinetics was investigated on the biomass at a 

suspension concentration of 0.5 g/L. The initial phenol concentrations were 

of 12.5; 25 and 50 mg/L. In the figure 2 the adsorbed phenol amounts are 

presented against the adsorption time. The results show that adsorption 

equilibrium was reached during sixty minutes. The adsorption rate was higher 

in the first thirty minutes, but decreased until the equilibrium was reached. 

Similar trends were found by other workers [2, 12, 13, 15]. It should be noticed 

that the adsorption of phenol increased with an increas of the sorption time. 

In the adsorption equilibrium at initial phenol concentration of 12.5 mg/L 

the maximal adsorbed phenol amount is qmax = 0.09 mg/g, at initial phenol 

concentration of 25.0 mg/L the maximal adsorbed phenol amount is qmax = 

0.10 mg/g, and at initial phenol concentration of 50.0 mg/L the maximal 

adsorbed phenol amount is qmax = 0.19 mg/g. 

To evaluate the biosorption kinetics of phenol, two kinetic models were 

used to fit the experimental data at different initial concentrations at pH 5.0. 

q (mg/g) 

0.25 

0.2 

0.15 

0.1 

0.05 

0 

0 50 100 150 200 

t (min) 

50.0 mg/L 

25.0 mg/L 

12.5 mg/L 

Figure 2. The effect of initial concentration on the sorption kinetics of phenol by 

mycelial pellets of Phanerochaete chrysosporium, initial concentration: 12.5; 25.0; 

50.0 mg/L, temperature: 22.5 o C, biomass concentration: 5 g/L.


Pseud-first-order Lagergren model 

The pseudo first-order rate expression of Lagergren model [16] is 

generally expressed as follows: 

dq 

= k1, 

ad ( qeq 

− q) 

(2) 

dt 

where, 

qeq and q have their usual meanings and 

k1,ad is the rate constant of first-order biosorption (min -1 ). 

The integrated form of equation (2) is: 

t 

log( qeq − q) 

= log qeq 

− k1, 

ad 

(3) 

2. 

303 

However, to fit equation (3) to experimental data, the value of qeq 

(equilibrium sorption capacity) must be pre-estimated by extrapolating the 

experimental data to t = ∞. In addition, in most cases the first-order rate 

equation is usually applicable over the initial 30 − 50 minutes of the sorption 

[13, 17, 18]. The plots of log(qeq – q) as a function of sorption time are shown 

in Figure 3a. The linear relationships were observed only for the initial 60 

minutes of sorption and the experimental data considerably deviated from 

the theoretical ones (not shown in the figure) after this period. The rate 

constants k1,ad and theoretical values of qeq calculated from the slope and 

intercept of the linear plots are summarized in table 1 along with the 

corresponding correlation coefficients. The first-order rate constants k1,ad 

and the equilibrium sorption capacities qeq,cal (qeq,cal = 0.103 mg/g for the 

initial concentration of 12.5 mg/L, qeq,cal = 0.104 mg/g for 50.0 mg/L) have 

almost the same values for both initial concentration of 12.5 and 25.0 mg/L. 

In the case of initial concentration of 50.0 mg/L acceptable calculated 

results were not received using the first-order Lagergren model. 

Pseudo- second-order kinetic model 

If the sorption rate is second-order, the pseudo second-order kinetic 

rate equation is expressed as [18]: 

dq 

2 

= k2 

, ad ( qeq 

− q) 

(4) 

dt 

where, 

k2,ad is the rate constant of second-order biosorption (g/mg min) After 

integration, the following equation is obtained: 

t 

q 

1 

= 2 

k2, 

adqeq 

+ 

t 

q 

eq 

(5) 

177

178 


It should be noticed that for the utilization of this model, the 

experimental value of qeq is not necessary to be pre-estimated. By plotting 

t/q against t for the initial concentrations (12.5, 25.0, 50.0 mg/L), straight 

lines were obtained as shown in Figure 3b. The second-order rate constants 

k2,ad and qeq values are presented in table 1 were determined from the 

slopes and intercepts of the plots. The results show that the second-order 

rate constants k2,ad increased with an increase in initial phenol concentration. 

The biosorption of 2,4-dichlorophenol from aqueous solution on non-living 

pellets of Phanerochaete chrysosporium grown by Kirk et al [20] was also 

studied by Wu and Yu [12,13]. On the basis of their experiments they also 

found that the biosorption process to biomass followed pseudo secondorder 

kinetics. The pseudo-second-order kinetic constants decreased with an 

increase in initial concentration for 2,4-dichlorophenol biosorption [13]. 

The correlation cefficients for the second-order kinetic model were 

close to 1.0, and the theoretical values of qeq also agreed well with the 

experimental data. On the other hand, the correlation coefficients for the 

pseudo-first-order kinetics were lower than those for the pseudo-secondorder 

one. Using pseudo-first-order model the theoretical qeq values did not 

give reasonable values. This can be explained that the sorption of phenol 

on mycelial pellets follow the second-order kinetics. The second-order 

kinetic parameters can be used to determine the equilibrium sorption capacity, 

percent of the removal of phenol, rate constants and initial sorption rate for 

a bioreactor design. 

log(qeq - q) 

-0.5 

-1 

-1.5 

-2 

-2.5 

-3 

-3.5 

0 

0 10 20 30 40 50 60 70 

a 

t (min) 

12.5 mg/L 

25.0 mg/L 

t/q (min g/mg) 

2500 

2000 

1500 

1000 

500 

0 

b 

0 50 100 150 200 

t (min) 

12.5 mg/L 

25.0 mg/L 

50.0 mg/L 

Figure 3. (a) Linearized pseudo-first-order kinetic model for phenol by mycelial pellets 

of Phanerochaete chrysosporium at different initial concentrations, initial concentrations: 

12.5 and 25.0 mg/L, temperature: 22.5 ° C, biomass concentration: 5 g/L. 

(b) Linearized pseudo-second-order kinetic model for phenol by mycelialpellets of 

Phanerochaete chrysosporium at different initial concentrations, initial concentrations: 

12.5, 25.0, 50.0 mg/L, temperature: 22.5 °C, biomass concentration: 5 g/L.


Table 1. The first-order and secon- order adsorption rate constants of phenol 

for different initial concentrations, at pH 5.0, temperature: 22.5 °C, 

biomass concentration: 5 g/L 

c0 

(mg/L) 

k1,ad, 

(min -1 ) 

qeq,cal 

(mg/g) 

R 2 k2,ad 

(g/mg min) 

qeq,cal 

(mg/g) 

R 2 qeq,exp 

(mg/g) 

12.5 0.076 0.104 0.918 1.861 0.090 0.991 0.086 

25.0 0.077 0.103 0.985 3.475 0.105 0.999 0.103 

50.0 - - - 4.172 0.193 0.997 0.189 

Adsorption isotherms of phenol by Phanerochaete chrysosporium 

biomass in aqueous suspension 

The biosorption isotherms of phenol by mycelium pellets were evaluated 

in the initial concentration range of 10 − 100 mg/L by varying biomass dosage. 

The biomass concentrations were 1.0 and 5.0 g/L. It is observed from 

Figure 4a that the uptake of phenol by biomass increases with the decrease of 

the biosorbent dosage and also increases with an increase of the initial 

concentration of phenol in solution. When suspension concentration was 5 g/L 

the maximal adsorbed amount was about 0.3 mg/g, while in the suspension 

of concentration of 1 g/L it was about 0.9 mg/g. 

Analysis of equilibrium is important for developing a model that can be 

used for the design of biosorption systems. Two classical adsorption models, 

Langmuir and Freundlich isotherms, are mostly frequently employed. 

Freundlich isotherm 

The Freundlich equation based on sorption on a heterogeneous surface 

is given below as equation (6): 

1/ 

n 

qeq 

= K FCeq 

(5) 

where, 

KF and n are the Freundlich constants, which are indicators of adsorption 

capacity and adsorption intensity of the sorbents [6, 7, 21]. Equation (5) can 

be linearized in logarithmic form as follows: 

1 

log qeq = log K F + logC 

eq 

(6) 

n 

The values of KF and n can be estimated respectively from the intercept 

and slope of a linear plot of experimental data of log qeq versus log Ceq. 

The linearized Freundlich adsorption isotherms of phenol obtained 

using different biomass dosages are shown in figure 4.b. The values of KF 

and n calculated from the plot are also given in table 2. along with the 

regression correlation coefficients. The parameter KF related to the sorption 

capacity increased with decreasing biomass concentration and thus increasing 

179

180 


maximal adsorbed amount of phenol. In table 2, n is less than unity, indicating 

that phenol is slightly linearly adsorbed by mycelial pellets in the equilibrated 

concentration range of 10 − 90 mg/L at temperature of 22.5 ° C, and then 

the monolayer is saturated. Adsorption equilibriums of organic pollutants, 

such es phenol and chlorophenols, followed the Freundlich isotherm better 

than the Langmuir one [13, 15, 21]. 

q (mg/g) 

1 

0.9 

0.8 

0.7 

0.6 

0.5 

0.4 

0.3 

0.2 

0.1 

0 

a 

1 g/L 

5 g/L 

0 20 40 60 80 100 120 

ce (mg/L) 

log q 

0.5 

0 

-0.5 

-1 

-1.5 

-2 

b 

1 g/L 

5g/L 

1.2 1.4 1.6 1.8 2 

log ce 

Figure 4. (a) Phenol adsorption isotherms by mycelial pellets of Phanerochaete 

chrysosporium from aqueous solutions at the biomass concentrations of 1 g/L 

and 5 g/L in the initial concentration range of 10 – 100 mg/L. 

(b) Linearized adsorption isotherms of Freundlich. 

Table 2. The Freundlich isotherm constants of phenol on micelial pellets 

in the initial concentration range of 10 −100 mg/L, at different biomass 

concentrations, at pH of 5.0 

biomass dosage 

(g/L) 

qeq 

(mg/g) 

KF 

(mg/g)(mg/L) n 

n R 2 

1.0 0.9 0.004 0.78 0.989 

5.0 0.3 0.001 0.76 0.942 

CONCLUSIONS 

The potential of non-living mycelial pellets of Phanerochaete 

chrysosporium grown in a medium of simple constitution to adsorb phenol 

molecules from aqueous solution was demonstrated in this study. The 

sorption capacity increased with an increase in initial phenol concentration 

and decreased with increasing biomass concentration. The biosorption of 

phenol by Ph. chrysosporium followed pseudo-second-order kinetics. The 

second-order kinetic constants increased with increasing initial concentration. 

The Freundlich model exhibited a good fit to the adsorption data of phenol.


In the studied concentration range of 10 – 100 mg/L phenol was adsorbed 

almost linearly by the fungal biomass and then the monolayer coverage 

was reached. Increasing biomass concentration, the maximal adsorbed amount 

of phenol decreased, and thus the calculated Freundlich constant decreased 

as well. 


Microorganism and its growth conditions 

Phanerochaete chrysosporium a white-rot fungus, obtained from the 

Institute of Microbiology, University of Pécs, was maintained by subculturing on 

potato dextrose agar slants. Hyphal suspensions were prepared from 7-day old 

cultures, grown on potato dextrose agar slants at 35 ± 2 °C. Ph. chrysosporium 

was grown in a liquid medium containing (g/L) D-glucose, 10.0; KH2PO4, 2.0; 

MgSO4·7H2O, 0.5; NH4Cl, 0.12; CaCl2·H2O, 0.1; thiamine, 0.001. The medium 

pH was adjusted to 4.5 with 1.0 mol/L HCl and 1.0 mol/L NaOH. The incubation 

was carried out at 39 o C in an orbital shaker incubator at 150 rpm for 5 days [1]. 

Preparation of the biosorbent 

After 5 days, the mycelial pellets were harvested through filtering. 

The biomass was then washed thoroughly with distilled water to remove the 

growth medium adhering on its surface. In order to exclude the possibility of 

biodegradation of phenol by living mycelia, the mycelial pellets used in the all 

adsorption experiments were inactivated at 120 o C and 104 kPa for 20 min. The 

biosorbent used in this study was in the form of mycelium pellets without 

homogenization. Therefore, term particle size reffers to the diameter of the 

mycelial pellet. 

Chemicals 

Phenol (>99 % purity) was purchased from Sigma-Aldrich Ltd (Hungary) 

and was used without further purification. All other inorganic chemicals were of 

analytical grade. Stock solutions were prepared by dissolving 0.1 g of phenol 

in 1.0 L of distilled water. The test solutions containing phenol were prepared 

by diluting 100 mg/L of stock solutions of phenol to the desired concentrations. 

The phenol concentrations of prepared solutions varied between 10 – 100 mg/L 

in the sorption experiments. The pH value of the solutions in this study (2.0 – 

11.0) was adjusted to the required value by using NaOH or HCl solutions. 

All solutions were stored in the dark at 4 °C prior to use. 

Batch experiments 

Biosorption experiments were carried out in batch mode. The biomass 

concentration was 5.0 g/L, (0.25 g dry mycelial pellets mixed with 50 mL of 

solution containing a pre-determined concentration of phenol). Mycelial pellets 

and phenol solution were placed in a test-tube, which was subsequently 

181

182 


covered to prevent photodegradation. All adsorption experiments were 

conducted in the dark to avoid formation of photodegradation products. 

Tubes were agitated on a shaker at 150 rpm and a constant temperature 

(22.5 ± 2 °C). Samples were taken at given time intervals, and then centrifuged 

at 10 000 rpm for 10 min. The supernatant was used for analysis of the 

residual phenol. The amount of phenol adsorbed at equilibrium, q (mg/g), was 

obtained as follows: 

( c0 

− c V 

q = e) 

(1) 

m 

where, 

c0 and ce are the initial and equilibrium liquid phase concentrations (mg/L) 

V is the volume of the solution (L) and 

m is the weight of the dry biomass used (g). 

Analysis 

Phenol concentration in supernatant was determined by HPLC. The 

HPLC system containes a liquid chromatograph (LC-10 ADVP, Shimadzu), a 

micro vacuum degasser (DGU-14 A, Shimadzu), a system controller (SCL-10 

AVP, Shimadzu), a diode array detector (SPD-M 10 AVP, Shimadzu) and an 

injector (7725i, Rheodyne). The LCMS solution software was applied on the 

HPLC system. The measurements were performed on the UV/VIS-photo 

diode array detector with detection at 270 nm. Chromatographic separations 

were performed on a Phenomenex C18 column (150×4.6 mm i.d., 5 μm, 

Phenomenex, USA). For separations the mobile phase A, water and mobile 

phase B, methanol and gradient system (0.03 min 42 % B eluent, 8.00 min 

60 % B eluent, 8.10 min 42 % B eluent and 11.00 min 42 % B eluent) were 

used. Operating conditions were as follows: flow rate 1.0 mL/min, column 

temperature ambient and injection volumes 20 μL of the standard and samples. 

Calibration curve of the standard was made by diluting stock solution of 

standard in water to yield 10-100 mg/L. 

ACKNOWLEDGEMENT 

Timea Pernyeszi, Ferenc Kilár and Cornelia Majdik gratefully acknowledge 

support for this research from the Hungarian-Romanian Intergovernmentals & 

Cooperation Programme between University of Pécs and University of Babes- 

Bolyai for 2008-2009.


REFERENCES 

1. M. Iqbal, A. Saeed, Process Biochemistry, 2007, 42, 1160. 

2. N. Calace, E. Nardi, B.M. Petronio, M. Pietronio, M. Pietroletti, Environ. Pollut., 

2002, 118(3), 315. 

3. G. Bülbül, Z. Aksu, Turkish J. Eng. Envir. Sci., 1997, 21, 175. 

4. A. Denizli, N. Cihangir, A.Y. Rad, M. Taner, G. Alsancak, Process Biochemistry, 

2004, 39, 2025. 

5. H. Volesky, Z.R. Holan, Biotechnol. Prog., 1995, 11, 239. 

6. Z. Aksu, J. Yener, Process Biochem., 1998, 33, 649. 

7. Z. Ning, K.J. Kennedy, L. Fernandes, Water Res., 1996, 30(9), 2039. 

8. I.R. Rao, T. Viraraghavan, Bioresour. Technol., 2002, 85(2), 165. 

9. J.L. Wang, Y. Qian, N. Horan, E. Stentiford, Bioresourcs Technol., 2002, 65, 165. 

10. M. Tsezos, J.P. Bell., Water Res. 1989, 23, 561. 

11. O. Yesilda, K. Fiskin, E. Yesilda, Envir. Techn., 1995, 16, 95. 

12. J. Wu, H-Q. Yu, Process Biochemistry, 2006, 41, 44. 

13. J. Wu, H-Q. Yu, Journal of Haradous Materials B., 2006, 137, 498. 

14. C.G. Silva, J.L.Faria, J. of Molecular Catalysis, 2009, 305, 147. 

15. P. Benoit, E. Barriuso, R. Calvet, Chemosphere, 1998, 37(7), 1271. 

16. S. Lagergren, Zur theorie der sogenannten adsorption geloster stoffe: Kungliga 

Svenska vetenskapsakademiens, Handlingar 1998, 24, 1. 

17. D. Batabyal, A. Sabu, S.K. Chaudhuri, Separations Technol., 1995, 5(4), 179. 

18. Z. Aksu, S. Tezer, Process Biochemistry, 2000, 36(5), 431. 

19. G. Mckay, Y.S. Ho, Process Biochem., 1999, 34, 451. 

20. T.K. Kirk, E. Schultz,cW.J. Connors, Arch Microbiol., 1978, 117, 227. 

21. B. Antizar-Ladislao, N.J. Galil, Water Res., 2004, 38(2), 267. 

183


ISOCONVERSIONAL LINEAR INTEGRAL KINETICS 

OF THE NON-ISOTHERMAL EVAPORATION OF 

4-[(4-chlorobenzyl)oxy]-4’-trifluoromethyl-azobenzene 

ANDREI ROTARU a , MIHAI GOŞA b,c , EUGEN SEGAL d 

ABSTRACT. The activation energy of the linear non-isothermal evaporation 

of 4-[(4-chlorobenzyl)oxy]-4’-trifluoromethyl-azobenzene has been determined 

by various linear integral isoconversional methods. Activation energy may 

be evaluated using several isoconversional methods as the well-known 

Tang et al. method, Generalized KAS method, but also by means of two 

so-called “local linear integral” (Tang & Chen) and ”average linear integral” 

(Ortega) isoconversional methods. A comparison study has been carried 

out in order to understand how the activation energy values are affected 

when using different approaches. 

Keywords: azomonoether dyes, non-isothermal kinetics, linear integral, 

local linear integral and average linear integral isoconversional methods. 

INTRODUCTION 

Dyes from the category of azoic aromatics have been intensively 

studied since they are of large interest from the point of view of possible 

applications [1-4]. Thermal characterizations and stability studies are 

usually required before choosing them to be part of composite materials, 

moreover since they are used in thermal-controlled devices. Kinetic studies 

are required for predicting their behaviour in other conditions than in those 

that are accessible for normal runs [5-9]. 

This paper aims to present a isoconversional kinetic study of the 

linear non-isothermal evaporation of 4-[(4-chlorobenzyl)oxy]-4’-trifluoromethylazobenzene 

liquid crystal that has been previously investigated [7] only by 

KAS (Kissinger-Akahira-Sunose) [10,11] and FWO (Flynn-Wall-Ozawa) [12,13] 

a INFLPR – National Institute for Laser, Plasma and Radiation Physics, Laser Department, 

Bvd. Atomistilor, Nr. 409, PO Box MG-16, RO-077125 Magurele, Bucharest, Romania, 

andrei.rotaru@inflpr.ro 

b University of Craiova, Faculty of Automatics, Computers and Electronics, Bvd. Decebal, Nr. 

107, Craiova, Romania 

c KineTAx, Str. Ştefan cel Mare, nr. 3, bl. L, sc. B, ap. 1, 200137 Craiova, Romania, 

mihai.gosa@kinetax.com 

d University of Bucharest, Faculty of Chemistry, Bvd. Regina Elisabeta Nr. 4-12, Bucharest, 

Romania, esegal@gw-chimie.math.unibuc.ro

186 

ANDREI ROTARU, MIHAI GOŞA, EUGEN SEGAL 

isoconversional methods. Since for our early papers we have used regular 

calculations, we were able to report results only for a few methods and 

conversions. The new software package TKS-SP [14,15] for kinetic analysis 

allows the rapid evaluation of the activation energy by means of isoconversional 

methods (integral and differential), as well as by other more complex procedures. 

With the increasing number of papers dealing with the approximation 

of the temperature integral, its evaluation improved [16-22] and led to the 

general opinion that other methods should be used instead of those consecrated 

ones [10-13]. The method of Tang et al. [16] and the generalized KAS [17, 

18] methods have been used here, together with the new “local linear 

isoconversional” of Tang & Chen [19] and ”average linear integral”of Ortega [20] 

methods as well. 


Thermal behaviour of 4-[(4-chlorobenzyl)oxy]-4’-trifluoromethyl-azobenzene 

was investigated in a previous paper [7], together with a couple of other 

liquid crystals from the same category of aromatic azomonoethers. Before 

evaporating in the temperature range of 190-310 o C (endothermic effect), at 

155 o C the compound melts. The FTIR spectrum of the evolved gases perfectly 

matches to the FTIR spectrum of the solid compound. No influence of the 

gas atmosphere (air or inert flow) was found. 

The complexity of a physical or chemical process can be expressed 

from the activation energy dependence on the conversion degree. Usually, 

“model-free” kinetic methods are the most popular. Such applications require 

the use of isoconversional methods for the evaluation of the activation energy. 

1) Regular linear integral methods (Tang et al. and Generalized KAS) 

Here we made use of several isoconversional methods: The methods 

of Tang et al. [16] (very close to KAS method) and Generalized KAS [17,18]; 

are the isoconversional integral linear ones, based on the following integral 

form of the reaction rate: 

α 

dα 

A α 

g( 

α ) = ∫ = ∫ e 

f ( α ) β 

T 

0 0 

E 

RT 

A 

dT = I( 

E 

β 

where β is the heating rate, R is the universal gas constant, g(α) is the 

integral conversion function and I(Eα ,Tα) represents the temperature integral. 

Substitution of I(Eα,Tα) in the equation (1) by various approximation mathematical 

expressions, provides various equations − therefore a multitude of isoconversional 

methods. 

− 

α 

, T 

α 

) 

(1)

ISOCONVERSIONAL LINEAR INTEGRAL KINETICS OF THE NON-ISOTHERMAL EVAPORATION 

In the case of Tang et al. method [16], for α=const., the plot ln(β/T 1.894661 ) 

vs. (1/T), obtained from the experimental thermogravimetric curves recorded 

for several constant-heating rates, should be a straight line whose slope could be 

used for the activation energy evaluation (E [kJ·mol -1 ]= – slope·R/1001.45033). 

In the case of Generalized KAS method [17,18], for α=const., several plots 

of ln(β/T n+m ) vs. (1/T), obtained from the experimental thermogravimetric 

curves recorded for several constant-heating rates, should gave straight 

lines whose slopes could be used for the activation energy evaluation 

(E [kJ·mol -1 ]= – slope·R/1000). For KAS method, n=2, m=0. 

Figure 1 shows the kinetic results of the non-isothermal evaporation 

of the investigated compound, obtained using Tang et al. and Generalized 

KAS (m=0.5, 1 and 2) methods, as values of the activation energy for various 

conversion degrees from 0.2 to 0.9 with a step of 0.01, using TKS software 

(SP 1.0 and SP2.0) [14,15]. 

Activation energy [kJ/mol] 

200 

180 

160 

140 

120 

100 

80 

60 

40 

20 

Tang et al. 

G-KAS m=0.5 

G-KAS m=1.0 

G-KAS m=2.0 

0 

0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 

Conversion degree 

Figure 1. Isoconversional activation energy of the non-isothermal evaporation of 

4-[(4-chlorobenzyl)oxy]-4’-trifluoromethyl-azobenzene 

(by Tang et al. and G-KAS methods) 

All employed regular linear integral isoconversional methods exhibit 

a similar trend of the activation energy; for all of them, the activation energy 

decreases with about 10% in the conversion range of 0.2-0.9. Table 1 contains 

a statistic of the activation energy values: 

187

188 


Table 1. Statistics of the activation energy values by means of 

regular linear integral isoconversional methods 

Method Max. activation 

energy 

kJ/mol 

Min. activation 

energy 

kJ/mol 

Average 

activation energy 

kJ/mol 

Tang et al. 145.2 129.3 136.2±4.8 

G-KAS (m=0.5) 142.7 126.5 133.5±4.9 

G-KAS (m=1.0) 140.6 124.3 131.3±4.9 

G-KAS (m=2.0) 136.3 119.7 126.9±5.0 

The differences between the results of the employed methods are 

due to the exponents (n and m) of the temperature that are used. With 

increasing the temperature exponent, the activation energy decreases, but 

the shape of the isoconversional evaluation remains the same. Although 

the absolute value is not determined, its trend is the same. 

Correlation coefficients 

1.000 

0.998 

0.996 

0.994 

0.992 

Tang et al. 

G-KAS m=0.5 

G-KAS m=1.0 

G-KAS m=2.0 

0.990 

0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 


Figure 2. Correlation coefficients for the isoconversional activation energy of 

the non-isothermal evaporation of 4-[(4-chlorobenzyl)oxy]-4’-trifluoromethylazobenzene 

(by Tang et al. and G-KAS methods) 

During the evaporation process, the activation energy may be however 

considered to remain practically constant; the accuracy in determining the 

activation energy is very high – correlation coefficients over 0.99650 (Figure 2). 

Since this compound is a liquid crystal, it may be used in specific devices; knowing 

the dependence of the activation energy behaviour with the consumption of the


evaporation, it may help manufacturers when designing their products and 

establish the limits the devices may be used. It can be easily noticed that 

with increasing value of the temperature’s exponent m, the activation 

energy is evaluated with higher accuracy. 

2) Local linear (Tang & Chen) and Average linear (Ortega) integral 

isoconversional methods 

In 2005 Tang & Chen [19] have proposed an integral isoconversional 

procedure that they have called “local linear” method. This method is basically the 

linear version of Vyazovkin’s “advanced integral isoconversional” method [21], 

called by Tang & Chen: “modified integral non-linear isoconversional procedure”. 

Equation 2 was derived by the authors and used for Δα→0; however it 

was concluded that integration over very small segments provides activation 

energy values close to those obtained using Friedmann [23] method. 

β [ 1/( 

1+ 

α )] 

ln = ln Aα 

− ln[ g( 

α + Δα 

) − g( 

α − Δα 

)] − 

T −T 

α + Δα 

α −Δα 

For very small intervals of conversion, this method is no more an 

integral one, but rather a differential isoconversional method. Here we made 

use of Δα=0.2, which is quite a high value, therefore the method being no 

more a differential one (Figure 3). 

Activation energy [kJ/mol] 

200 

180 

160 

140 

120 

100 

80 

60 

40 

20 

Tang & Chen (Δα=0.2) 

Ortega (Δα=0.2) 

0 

0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 


Figure 3. Isoconversional activation energy of the non-isothermal evaporation 

of 4-[(4-chlorobenzyl)oxy]-4’-trifluoromethyl-azobenzene 

(by Tang&Chen and Ortega methods) 

189 

E 

RT 

α 

(2)


The same judgement was used when applying Ortega [20] method 

(recently published in 2008). Ortega method, or Average linear (Ortega) 

integral isoconversional method uses equation 3. This method takes into 

account the history of the process as well. 

β 

ln = ln Aα 

− ln[ g( 

α + Δα 

) − g( 

α − Δα 

)] − 

T 

α −Tα −Δα 

From Figure 3 it can be seen that these two methods provide similar 

results. 

Even if the correlation coefficients (Figure 4) are higher for Tang & Chen 

method, Ortega method permits the evaluation until a higher conversion degree. 

Correlation Coefficients 

0.994 

0.992 

0.990 

0.988 

0.986 

0.984 

0.982 

Tang & Chen (Δα=0.2) 

Ortega (Δα=0.2) 

These two methods that strongly depend on the history and future 

of the process, provide in this case however similar results to the results of 

regular isoconversional ones. In the conversion degree range of 0.25-0.75, for 

Tang & Chen method the activation energy decreases from 145 to 123 kJ·mol 

190 

-1 , 

while in the conversion degree range of 0.25-0.75, for Ortega method the 

activation energy decreases from 148 to 120 kJ·mol -1 . 

E 

RT 

0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 


Figure 4. Correlation coefficients for the isoconversional activation energy of 

the non-isothermal evaporation of 4-[(4-chlorobenzyl)oxy]-4’-trifluoromethylazobenzene 

(by Tang&Chen and Ortega methods) 

α 

(3)


CONCLUSIONS 

The non-isothermal evaporation of 4-[(4-chlorobenzyl)oxy]-4’trifluoromethyl-azobenzene 

has been investigated by several isoconversional 

methods; regular integral methods provided different, but symmetric trends 

of the isoconversional activation energy, that are strongly dependent on the 

temperature exponents (n and m). New methods like Tang & Chen and 

Ortega methods have been derived by using two other concepts and have 

to be still used until they will be completely understood. An imposed range 

of conversion confidence Δα=0.2 should overcome the problem of differential 

approach; however, these methods induce a higher inaccuracy in the evaluation 

of activation energy. It is not clear how this Δα interval should be chosen, 

but it is obvious that the same value of it provides the same activation 

energy values for both methods. 


Aromatic azomonoethers were obtained by the condensation of 

some(phenyl-azo)phenols with chloromethylated derivatives of chlorobenzene 

in alkaline medium (Williamson ether synthesis) [24]. Thermal stability (TG, 

DTG and DSC) measurements of 4-[(4-chlorobenzyl)oxy]-4’-trifluoromethylazobenzene 

were carried out in air flow (150 mL·min -1 ) in a horizontal Diamond 

Differential/Thermogravimetric Analyzer from Perkin-Elmer Instruments [7]. 

Samples from 0.8 to 1 mg, contained in Al2O3 crucibles, were heated from 

room temperature to 800 o C, with the heating rates of: 2, 4, 6 and 8 K·min -1 . 

REFERENCES 

1. A. Rotaru, C. Constantinescu, P. Rotaru, A. Moanta, M. Dumitru, M. Socaciu, 

M. Dinescu, E. Segal, J. Therm. Anal. Cal., 2008, 92, 279. 

2. T.-Y. Chao, H.-L. Chang, W.-C. Su, J.-Y. Wu, R.-J. Jeng, Dyes Pigments, 2008, 

77, 515. 

3. H. Dincalp, F. Toker, J. Durucasu, N. Avcibasi, S. Icli, Dyes Pigments, 2007, 75, 11. 

4. M. Gűr, H. Kocaokutgen, M. Taş, Dyes Pigments, 2007, 72, 101. 

5. A. Rotaru, A. Moanţă, I. Sălăgeanu, P. Budrugeac, E. Segal, J. Therm. Anal. Cal., 

2007, 87, 395. 

6. A. Rotaru, B. Jurca, A. Moanta, I. Salageanu, E. Segal, Rev. Roum. Chim., 2006, 

51, 373. 

7. A. Rotaru, A. Kropidłowska, A. Moanţă, P. Rotaru, E. Segal, J. Therm. Anal. Cal., 

2008, 92, 233. 

191

192 


8. A. Rotaru, A. Moanta, P. Rotaru, E. Segal, J. Therm. Anal. Cal., 2009, 95, 161. 

9. A. Rotaru, G. Bratulescu, P. Rotaru, Thermochim. Acta, 2009, 489, 63. 

10. H. E. Kissinger, Anal. Chem., 1957, 29, 1702. 

11. T. Akahira, T. Sunose, Res. Report Chiba Inst. Technol., 1971, 16, 22. 

12. J. H. Flynn, L. A. Wall, J. Res. Natl. Bur. Stand., A. Phys. Chem., 1966, 70, 487. 

13. T. Ozawa, Bull. Chem. Soc. Jpn., 1965, 38, 1881. 

14. A. Rotaru, M. Gosa, P. Rotaru, J. Therm. Anal. Cal., 2008, 94, 367. 

15. A. Rotaru, M. Gosa, J. Therm. Anal. Cal., 2009, 96, x. 

16. W. Tang, Y. Liu, H. Zhang, C. Wang, Thermochim. Acta, 2003, 408, 39. 

17. H. X. Chen, N. A. Liu, J. Therm. Anal. Cal., 2007, 90, 449. 

18. H. X. Chen, N. A. Liu, J. Therm. Anal. Cal., 2008, 92, 573. 

19. W. Tang, D. Chen, Thermochim. Acta, 2005, 433, 72. 

20. A. Ortega, Thermochim. Acta, 2008, 474, 81. 

21. S. Vyazovkin, J. Comput. Chem., 2001, 22, 178. 

22. P. Budrugeac, J. Therm. Anal. Cal., 2002, 68, 131. 

23. H. L. Friedmann, J. Polym. Sci. C, 1963, 6, 183. 

24. S. Radu, C. Sarpe-Tudoran, A. Jianu, G. Rau, Rev. Roumaine. Chim., 1998, 43, 735


OVERALL KINETICS FOR THE CATALYTIC IGNITION 

OF ETHANE-AIR MIXTURES ON PLATINUM 

OCTAVIAN STAICU a , VALENTIN MUNTEANU a , DUMITRU OANCEA a 

ABSTRACT. The overall kinetics of the catalytic ignition of a stoichiometric 

ethane/air mixture on an isothermally heated platinum wire is studied. The 

isothermal regime, achieved in an original experimental set-up, allows the 

direct measurement of the reaction rate in various operational conditions and 

the determination of the overall activation energy and reaction order on the 

basis of a previously reported kinetic model. The results are discussed and 

compared with other data obtained using different experimental techniques 

and models. 

Keywords: catalytic combustion, ignition, ethane/air mixture, isothermal, kinetics 

INTRODUCTION 

The catalytic combustion of alkanes represents one of the most 

important ways to reduce the pollutant emissions which unavoidably accompany 

the conversion of chemical energy stored in fuels into other forms of energy. 

The high temperature combustion of alkanes in air, within the specific flammable 

ranges, between the lower flammability limit and the stoichiometric concentration, 

promotes the NOx formation with unwanted consequences on the environment. 

The heterogeneous catalytic combustion of the same fuels, occurring at much 

lower temperatures, leads to a drastic reduction of NOx formation. Moreover, 

the catalytic combustion on solid catalysts allows the process to occur for 

fuel concentrations much lower than the lower flammability limits. The catalytic 

combustion has consequently a valuable potential in many applications like the 

radiant heaters, catalytic combustors for power production, removal of hazardous 

air pollutants etc. [1]. The phenomenological description of this process advanced 

considerably during the last decades. Less information is available for the 

kinetics and mechanism of the interconnected steps including mass and heat 

transport, fluid flow and heterogeneous chemical reactions [2]. Additionally, at 

higher temperatures, the simultaneous homogeneous combustion induced by 

a Department of Physical Chemistry, University of Bucharest, 4-12 Bd. Elisabeta, 030018 Bucharest, 

Romania, doan@gw-chimie.math.unibuc.ro

194 

OCTAVIAN STAICU, VALENTIN MUNTEANU, DUMITRU OANCEA 

the heterogeneous catalytic reaction plays a significant role. One of the most 

known difficulties associated with the kinetic studies of the combustion reactions 

is related to its exothermicity, which leads to catalyst heating at temperatures 

significantly higher than the surrounding fluid. The catalyst heating can be 

diminished either by working at very low reactant concentrations or at very 

high fluid recirculation rate. Both procedures add new difficulties for these 

studies. This well-known phenomenon can explain the differences reported by 

various researchers. For platinum wires, frequently utilized in kinetic studies [3- 

5], we recently described an efficient method able to maintain a constant 

temperature during the catalytic reaction of different fuel/air mixtures [7-10]. 

The results obtained for n-butane/air, iso-butane/air and propane/air mixtures 

demonstrated the potentialities of the proposed method. A similar analysis is 

presented in this paper for the stoichiometric ethane/air mixture. A quasi-step 

temperature perturbation of a platinum catalytic wire immersed in a fuel/air 

mixture is applied and the corresponding reaction heat flow rate, dQr/dt = Fr, is 

measured in time until the process becomes stationary. The heterogeneous 

catalytic reaction exhibits an induction period, τi, which can be measured as a 

function of wire temperature, Tw, total pressure, p0 and gas composition. The 

analytical form of this dependence, τi = f(Tw, p0), allows the determination of 

the overall kinetic parameters through linear regression analysis [10]. 


The experimentally recorded reaction heat flow rate, Fr, as a function 

of time can be easily converted in the more relevant catalytic reaction rate, 

rR, in time through: 

rR r 

T 

c 0 

= ( dQ dt) 

( Δ H ⋅S) 

(1) 

c 0 

where Δ HT 

is the standard heat of combustion at the wire temperature, 

c 0 c 0 

and S is the surface of the platinum wire ( Δ HT 

= Δ H 298 =1,428.61 kJ/mole 

and S=1.41×10 -5 m 2 ). A typical result is given in Figure 1. 

Two properties are available from such a plot: the length of the 

induction period, τi, and the diffusion controlled reaction heat flow rate, Fr, 

or the corresponding reaction rate, rR. These properties were measured for 

the stoichiometric ethane-air mixture at various wire temperatures and gas 

pressures. Their physical significances were discussed in several previous 

papers. The induction period is given by [10]: 

i 

−n 

Ea 

RTw 

∗ 

τ = β ⋅ p p ) ⋅e 

(2) 

( 0

OVERALL KINETICS FOR THE CATALYTIC IGNITION OF ETHANE-AIR MIXTURES ON PLATINUM 

where β is a proportionality constant including the pre-exponential factor 

∗ 

and a critical amount, Δ C , of a reactant required to initiate the ignition 

∗ 

∗ 

( rR = ΔC 

Δt 

= ΔC 

τ i ), p ∗ is the standard pressure (≈101 kPa), n and Ea 

are the overall reaction order and activation energy, respectively, and R is 

the universal gas constant. 

On the other hand, the diffusion controlled reaction heat flow rate 

can be rationalized according to an Arrhenius type equation [8]: 

r R /(mol m -2 s -1 ) 

dQ 

0.06 

0.05 

0.04 

0.03 

0.02 

0.01 

0.00 

-0.01 

r 

r 

∗ 

n 

−Ea 

RTw 

dt = F = A0 

⋅ ( p0 

p ) ⋅ e 

(3) 

5.66% C2H6- air mixture 

p0 =100 kPa; Tw = 487 K 

induction period 

kinetic control 

kinetic to diffusion transition 

0 2 4 6 8 10 

time/s 

diffusion control 

Figure 1. Variation of the surface reaction rate in time indicating the existence of 

an induction period and of the transition from kinetic to diffusion control 

Equations (2) and (3) are used to evaluate the parameters n and Ea. It 

can be observed that equation (2) refers to the kinetic control, while equation (3) 

refers to the diffusion control. 

Activation energy evaluation. A typical plot for the kinetic control is 

given in Figure 2 resulted from the linear regression ln(τi) versus 1/Tw of 

equation (2) at constant pressure. Similar plots, ln(Fr) versus 1/Tw, were obtained 

also for the diffusion control from equation (3) at constant pressure. 

195

ln(τ i / s) 


Table 1. Kinetic and diffusion overall activation energies 

1.0 

0.5 

0.0 

-0.5 

-1.0 

-1.5 

-2.0 

-2.5 

-3.0 

p0/kPa 

Ea/(kJ/mol) 

ln(τi) vs 1/Tw 

Ea/(kJ/mol) 

ln(Fr) vs 1/Tw 

100 97.6 ± 3.2 11.5 ± 0.35 

70 84.9 ± 8.5 11.54 ± 0.01 

50 87.1±3.1 10.71 ± 0.37 

30 74.4 ± 6.2 10.61 ± 0.36 

10 75.6 ± 11.5 9.08 ± 2.43 

5.66 % C H - air, p = 100 kPa 

2 6 0 

507 K < T < 585 K 

w 

E = 97.6 ± 3.2 kJ / mol; r = 0.997 

a 

1.70 1.75 1.80 1.85 1.90 1.95 2.00 

1000 K / T w 

Figure 2. Evaluation of the overall activation energy for the kinetic control 

The results obtained at various initial pressures for both kinetic and 

diffusion controls are given in Table 1. Within the limits of experimental errors 

no significant variation of activation energies can be detected using this method. 

The activation energies for the kinetic control are in good agreement with 

other data reported in literature: 114 kJ/mol on platinum wire from ignition 

temperature measurements [11], or on platinum foil in a recirculating batch 

reactor with mass spectrometry detection [12]; 95 kJ/mol on platinum foil in 

a stagnation point flow [13]; 57 kJ/mol on platinum wire from ignition 

temperature measurements [14]; 109 kJ/mol on platinum wires in a flow 

reactor [15]. When the reaction occurs on platinum dispersed on different 

supports the activation energy varies from 29.7 to 143 kJ/mol [16]. 

The activation energies for the diffusion control are within the limits 

accepted for these reactions. 

196


Reaction rate and related properties evaluation. The measurement 

of the reaction rate for the diffusion regime offers the possibility to compare 

the experimental results with several relevant properties like the turnover 

frequency, TOF, the collision frequency of fuel molecules with the catalytic 

surface, νF and the ratio rR/νF. The turnover frequency was calculated as 

TOF=rR/ΓPt, using the literature data for the atomic surface density of platinum 

exposed atoms (ΓPt=2.49 × 10 –5 mol m -2 [12]). The collision frequency of fuel 

molecules with the catalytic surface was calculated for the stoichiometric 

ethane-air mixture as [8]: 

1 2 

ν F = pF ( 2πM 

F RTw) 

(4) 

where pF is the partial pressure of the fuel and MF its molar mass. 

The results are given in Table 2. 

Table 2. Reaction rates and associated properties for diffusion control 

at Tw = 549 K 

p0 

kPa 

rR 

(mol m -2 s -1 ) 

10 -3 ×TOF 

(s -1 ) 

νF 

(mol m -2 s -1 ) 

10 4 ×(rR/νF) 

100 0.0765 3.07 193 3.96 

70 0.0720 2.89 135 5.33 

50 0.0660 2.65 96.5 6.84 

30 0.0591 2.37 57.9 10.2 

10 0.0348 1.40 19.3 18.0 

The TOF values are very large compared with other literature data [8, 

10]. A possible explanation is based on the observation that the platinum wire 

exhibits significant catalytic activity only after heating at temperatures higher 

than 800 K, when the initially smooth surface becomes coarse with an increased 

atomic surface density. The ratio rR/νF has very low values, (4-18) ×10 -4 , 

indicating a reduced efficiency of fuel collisions with the catalyst surface. If these 

values are corrected for the activation energy factor (of the order 9.6×10 -2 ), 

very small sticking coefficients are obtained. The data given in Table 2 indicate 

that the method and the measurements give realistic results for a typical 

heterogeneous catalytic reaction of combustion. 

Reaction order evaluation. The measurements at constant temperature 

of the reaction heat flow rates for a diffusion controlled process and of the 

induction periods for a kinetically controlled process can be used to evaluate 

the overall reaction orders using the linear regression analysis of equations (3) 

and (2). The results are given in Table 3. It can be observed that the overall 

reaction orders have an approximately constant value of 0.34, without a 

significant trend. These figures are in agreement with the frequently reported 

197

198 


order nF=1 for alkane-air catalytic combustion and with the recognized 

inhibitory effect of oxygen in these reactions (nox


It was also assumed that the surface reaction is of first order in 

hydrocarbon and zero order in oxygen concentration (for large excess of 

oxygen). This relationship has been subsequently generalized [2] for a n order 

reaction in the form: 

n 

F 

2 

ign − T = 1 

Ea 

RTign 

C ( T 0) α ⋅ e 

(6) 

Since our experimental technique allows the measurement of the 

ignition temperature at various total pressures, we recently proposed an 

alternative procedure for the determination of the activation energy [10] as 

n n 

long as ∝ p for constant T0 and fuel molar fraction: 

CF 0 

n 

0 

2 

ign − T0 

) = 2 

Ea 

RTign 

p ( T α ⋅ e 

(7) 

where α2 is a proportionality constant. 

A representative result is given in Figure 3. The measured activation 

energy is in good agreement with the results given in Table 1 for the kinetically 

controlled process. 

CONCLUSIONS 

The catalytic combustion of the stoichiometric ethane-air mixture on a 

platinum wire heated in isothermal conditions allows a deeper understanding 

of the kinetics of this heterogeneous reaction. Since the wire temperature is 

constant during the ignition and subsequent combustion, the self-acceleration 

of the reaction rate cannot be explained on the basis of the thermal ignition, 

which assumes a continuous increase of the catalyst temperature. From the 

measurement of the induction periods of the catalytic ignition at various 

temperatures and total pressures, the overall kinetic parameters – activation 

energy and reaction order – can be evaluated. The obtained results are in 

good agreement with those obtained using other experimental approaches. 

The analysis of the ignition temperature variation with the total pressure using 

a new model allows an alternative method for the determination of activation 

energy. All these results validate the proposed method and recommend it 

as a valuable alternative for kinetic studies concerning the heterogeneous 

catalytic combustion reactions. 

199

200 



The experimental measurements were carried out using the equipment 

and procedure described in details previously [7-10] and given schematically 

in Figure 4. A platinum wire of 0.1 mm diameter and 45 mm length (99.99% 

from Aldrich), connected through brass conductors in a heating circuit, is 

immersed in the center of a cylindrical test cell of 9 cm diameter and height, 

containing either air or fuel/air mixture at a prescribed pressure. It is heated 

according to a quasi-rectangular profile, with a rise time of 1 – 2 ms using the 

discharge of a capacitor C followed by a controlled feeding system designed 

to maintain a constant wire temperature. A standard resistor Rstd, connected in 

series with the wire of resistance Rw, forms a Wheatstone bridge with the 

potentiometer P and allows the measurement of the input power, recording 

the variation of the voltage drop Ustd during the test. Any unbalance of the 

bridge is detected by the integrated circuit IC which readjusts the voltage 

applied across Rw through the series transistor T in order to maintain its 

resistance constant. The diagram Ustd versus time contains the information 

necessary for the kinetic study. To evaluate the heat flow rate dQr/dt due 

to the catalytic reaction occurring on the wire it is necessary to eliminate 

the power dissipated through the heat transfer from the hot wire to the 

surroundings. 

The kinetically relevant quantity dQr/dt can be obtained from the 

diagrams recorded in air and in a fuel/air mixture in similar conditions as: 

r 

w 

2 

std 

2 

2 

[ ( U ) − ( U ] 

dQ dt = ( R R ) 

) 

(8) 

std 

air 

std 

mixture 

Figure 4. Schematic representation of the electrical circuit able to heat 

the platinum wire according to a quasi-rectangular profile


A typical recording of Ustd is given in Figure 5. 

U std /V 

5 

4 

3 

2 

1 

0 

0 2 4 6 8 10 

time /s 

5.66% C2H6-air 

p 0 =100 kPa; T w =487 K 

Figure 5. Recorded Ustd versus time diagram 

Using equation (8) and the similar diagram recorded in air (without fuel) 

at the same wire temperature and total pressure, a kinetic diagram given in 

Figure 1 can be obtained. The wire temperature is calculated according to 

literature recommendations as [17]: 

x − 2. 

64 i 

T Di 

⋅[ 

] 

1. 

64 

9 

w = 273. 

15 + D0 

+ ∑ 

i= 

1 

where x is the ratio between the wire resistance at temperature Tw and at 

273.15 K and D0, Dj are constants given in literature [17]. All the other 

details were given elsewhere [7-10]. 


The authors acknowledge the financial support of CNCSIS through 

the Contract nr.38/2007 for the Project ID_1008. 

(9) 

201

202 


REFERENCES 

1. J. Saint-Just, J. der Kinderen, Catalysis Today, 1996, 29, 387. 

2. A. Schwartz, L. Holbrook and H. Wise, Journal of Catalysis, 1971, 21, 199. 

3. C. G. Rader and S. W. Weller, American Institute of Chemical Engineering 

Journal, 1974, 20, 515. 

4. S. W. Weller and C. G. Rader, American Institute of Chemical Engineering 

Journal, 1975, 21, 176. 

5. P. Cho and C.K. Law, Combustion and Flame, 1986, 66, 159. 

6. M. A. A. Cardoso and D. Luss, Chemical Engineering Science, 1969, 24, 1699. 

7. D. Oancea, D. Razus, M. Mitu and S. Constantinescu, Revue Roumaine 

Chimie, 2002, 47, 91. 

8. D. Oancea, O. Staicu, V. Munteanu, and D. Razus, Catalysis Letters, 2008, 

121, 247. 

9. O. Staicu, V. Munteanu, D. Oancea, Catalysis Letters, 2009, 129, 124. 

10. O. Staicu, D. Razus, V. Munteanu, D. Oancea, Central European Journal of 

Chemistry, 2009, 7, 478. 

11. L. Hiam, H. Wise, S. Chaikin, Journal of Catalysis,1968, 9-10, 272. 

12. M. Aryafar and F. Zaera, Catalysis Letters, 1997, 48, 173. 

13. G. Veser and L.D. Schmidt, American Institute of Chemical Engineering Journal, 

1996, 42, 1077. 

14. T. A. Griffin, L. D. Pfefferle, American Institute of Chemical Engineering Journal, 

1990, 36, 861. 

15. Y.-F. Yu Yao, Industrial and Engineering Chemistry, Product Research and 

Development, 1980, 19, 293. 

16. T. F. Garetto, E. Rincon, C. R. Apesteguia, Applied Catalysis B: Environmental, 

2007, 73, 65. 

17. H. Preston-Thomson, Metrologia, 1990, 27, 3.


GROWTH AND CHARACTERISATION OF ZINC-CADMIUM 

SULPHIDE THIN FILMS WITH SPECIAL OPTICAL PROPERTIES 

MARIA ŞTEFAN a, b , IOAN BÂLDEA b , RODICA GRECU a , EMIL INDREA c , 


ABSTRACT. Nanostructured ZnCdS thin films with variable thickness and 

high transparency were deposited under variable growth parameters on optical 

glass platelets, using chemical bath deposition. UV-VIS absorption/reflection 

spectroscopy, X-ray diffraction, fluorescence spectroscopy, electron microscopy 

investigations (SEM, EDX) were used to investigate the effect of growth 

and annealing regime on the films morphology and optical properties. The 

luminescence properties of copper activated ZnCdS thin films were also 

investigated. The deposition conditions (medium pH, reagent molar ratio, 

deposition time) and the presence of the doping ions (copper) influence the 

morpho-structural properties and optical characteristics of the nanocrystalline 

zinc-cadmium sulfide thin films. 

Keywords: thin films; chemical synthesis; optical properties; luminescence 

INTRODUCTION 

Zinc sulfide (ZnS), cadmium sulfide (CdS) and their solid solution 

zinc-cadmium sulfide (Zn1-xCdxS or ZnCdS) are interesting wide band gap 

semiconductors, which attract much attention because of their interesting 

optoelectronic properties [1-4]. 

Due to their special optical and electrical properties, metal chalcogenide 

thin films have been used in various optoelectronic devices, such as solar energy 

absorbers, photodetectors and electroluminescent displays [5-7]. Different 

methods could be used for metal chalcogenide thin films preparation including 

sputtering [8], MOCVD (metal organic chemical vapor deposition) [9], SILAR 

(successive ionic layer adsorbtion and reaction) [10], pulsed laser deposition [11], 

chemical bath deposition (CBD) [12] or electrodeposition [13]. Among them, 

a 

Raluca Ripan“ Institute for Research in Chemistry, “Babes-Bolyai” University, 30 Fantanele, 

400294, Cluj-Napoca, Romania; marialadar@yahoo.com 

b 

Faculty of Chemistry and Chemical Engineering, “Babes-Bolyai” University, 400028 Cluj-Napoca, 

Romania 

c 

National Institute for R & D of Isotopic and Molecular Technology, 71 Donath, 103,400293 

Cluj-Napoca, Romania

204 

MARIA ŞTEFAN, IOAN BÂLDEA, RODICA GRECU ET ALL 

CBD is a simple and inexpensive method to produce uniform and adherent 

films onto a variety of substrates. There are few literature data referring to the 

chemical bath deposition of ZnCdS thin films with optical properties illustrating 

that the characteristic films properties strongly depend on chemical bath 

deposition conditions [14,15]. In order to control the quality of ZnCdS thin 

films, supplementary studies aiming at the optimization of CBD parameters 

are needed. 

The paper presents our results on the growth and characterization 

of some reproducible ZnCdS thin films prepared by chemical bath deposition. 

The effect of cadmium ions addition into the chemical deposition bath on 

the quality of metal chalcogenide thin films is investigated. Moreover, attempts 

were made to prepare ZnCdS thin films with luminescent properties using a 

special doping procedure, not mentioned in the literature. 


The preparation of ZnCdS thin films by CBD was based on the reaction, 

in alkaline medium (NH3), between zinc acetate, Zn(CH3COO)2 and cadmium 

acetate, Cd(CH3COO)2 as metal sources and thiourea (NH2)2CS, as chalcogen 

source. The main chemical process for zinc-cadmium sulphide thin films 

formation is described by the following equation: 

M (CH3COO)2 + (NH2) 2 CS + 2OH - → MS + H2CN2 + 2H2O + 2CH3COO - 

where M=Zn and Cd 

The formation of thin films takes place either in the bulk of the solution 

due to the spontaneous precipitation of ZnS and CdS (homogeneous 

reaction) or at the surfaces of the substrate leading to the film formation (by 

heterogeneous reaction). 

Our previously studies on chemical bath deposited ZnS thin films [16] 

pointed out the importance of sodium citrate [Cyt] as complexing agent. Based 

on our results, the standard chemical deposition bath corresponds to the 

following reagent ratio: [Zn 2+ ]: [Cyt 3- ]: [NH3]: [thiourea] =1:3:20:10. Starting 

from this bath, ZnCdS /glass/ ZnCdS hetero-structures were prepared by 

replacing 5 ÷100 mol % of zinc acetate with cadmium acetate. 

Zinc-cadmium sulfide thin films were grown by multilayer technique. 

Four consecutive layers were deposited on glass platelets to give ZnCdS 

/glass/ ZnCdS heterostructures. For the same deposition time, addition of 

cadmium acetate into the standard chemical bath used for ZnS-film 

deposition does not deteriorate the film quality. The as obtained ZnCdS thin 

films possess good adherence to the substrate, are transparent and show 

colors varying from white-yellowish to orange-yellowish, depending on the 

Cd 2+ - amount from the chemical deposition bath.

GROWTH AND CHARACTERISATION OF ZINC-CADMIUM SULPHIDE THIN FILMS … 

According to the literature data [17], the deposition of ternary materials 

by chemical bath methods is a problem due to the different hydrolytic 

stabilities of the two metals. This is illustrated with the fact that, ZnCdS 

films obtained by CBD show properties that vary non-linearly with the bath 

composition. 

The packing density of ZnCdS thin films varies non-monotonically with 

Cd 2+ - concentration, thus suggesting the existence of more different competitive 

deposition processes with consequences on the film homogeneity (Fig. 1). 

Figure 1. Packing density of ZnCdS versus Cd 2+ -concentration from 

the chemical deposition bath 

The optical properties of ZnCdS thin films are strongly influenced by 

the deposition technique and CBD bath composition. The transmittance spectra 

of ZnCdS thin films obtained from chemical bath with different Cd 2+ amounts 

are depicted in figure 2. 

UV-Vis transmittance spectra illustrate the non-monotonous variation 

of film transparency with Cd 2+ content. Films obtained from bath containing 

5-10 mol% Cd 2+ are the most transparent and optical homogeneous. The 

absorption edge (Fig. 3) shifts non-linearly from UV domain (~ 300 nm for 

0 mol% Cd 2+ ) to blue domain (~455 nm for 100 mol% Cd 2+ ). Evaluation is 

based on the wavelength corresponding to 0.1 % transmittance values. 

UV-Vis reflectance spectra (Fig. 4) illustrate that the behavior of 

ZnCdS films is very different in respect with their specular reflectivity. The 

reflection peak shifts non-monotonically from ultraviolet (0 mol% Cd 2+ ) to 

red domain (100 mol% Cd 2+ ). The most light reflecting films are obtained in 

bath containing 5 ÷ 20 mol % Cd 2+ . 

205

206 


Figure 2. Transmittance spectra of ZnCdS thin films obtained from 

chemical bath with different Cd 2+ amounts 

Figure 3. Variation of absorption edge position of ZnCdS thin films with 

Cd 2+ -amount from chemical bath 

The specular reflectance was evaluated as difference between total 

reflectance (measured at 8° incidence) and diffuse reflectance (measured 

at 0° incidence).


Figure 4. Specular reflectance spectra of ZnCdS thin films obtained from 

chemical bath with different Cd 2+ amounts 

Morphology of ZnCdS films was investigated by scanning electronic 

microscopy (Fig. 5). SEM images put in evidence that the surface of films 

deposited from bath with different Cd 2+ amounts, is variable. The most 

structured/particulate film is obtained from bath containing only cadmium 

acetate. 

Figure 5. SEM images of ZnS film (H2.3) and CdS film (H 1.4) 

207

208 


EDX investigations proved that film composition is different of that 

one of the chemical bath, as already suggested by the UV-Vis spectroscopic 

investigations (Table 1). For instance, the composition of film H7.2 obtained 

from bath with 40 mol% Cd 2+ contains about 75 mol% Cd 2+ . One notes also 

that the ZnCdS are usually thinner than the pure ZnS or CdS films). As 

expected, the preferential precipitation of cadmium sulfide is observed and 

Cd-rich ZnCdS films are usually growth in our CBD conditions. 

One can also note that, according to the EDX data, the heterostructure 

contains Zn, Cd and S from the chalcogenide films and Si, O, Ca, Na, Mg 

and Al from the glass substrate (the electron beam penetrates the thin 

ZnCdS film and enters into the glass substrate, see table 1). 

Table 1. Composition of the ZnCdS/glass/ZnCdS films heterostructures obtained from 

chemical bath with different compositions (where others = Na, Mg, Al, K, Ca, O) 

Code 

Zn/Cd 

in bath 

Heterostructure composition (mol %) 

Si Others Zn Cd S 

Zn/Cd 

in film 

Packing 

density 

(mg/cm 2 ) 

H2.1 1:0 10.70 72.51 11.25 0 5.54 1:0 0.26 

H3.2 1:0.25 62.89 29.79 1.23 1.56 4.53 1:1.27 0.13 

H7.2 1:0.66 29.89 67.03 0.29 1.05 1.74 1:3.62 0.12 

H9.2 1:4.00 29.76 62.54 0.19 3.27 4.24 1:17.21 0.25 

H1.5 0:1 29.65 56.00 0 6.53 7.82 0:1 0.30 

The crystalline structure of ZnCdS thin films has been investigated 

by the X-ray diffraction technique (Fig. 6). All films deposited on glass substrate 

are amorphous, as illustrated by the X-ray scattering curves; an organization 

tendency of CdS containing layers in comparison with ZnS films can be observed. 

Figure 6. X-ray scattering curves of some ZnCdS films (MoKα)


The light emitting properties of metal chalcogenide thin films are 

highly sensitive to the annealing regime as well as to the doping conditions [16]. 

In order to develop photoluminescent properties, ZnCdS /glass/ ZnCdS 

heterostructures were activated with copper ions, using indirect doping 

technique [16]. Emission spectra of Cu doped heterostructures obtained 

from chemical bath with different Cd 2+ amounts are presented in figure 7. 

Figure 7. Emission spectra of ZnCdS: Cu films obtained from chemical bath 

with different Cd 2+ amounts 

ZnS:Cu film show intense green luminescence, with maxima at 483 nm 

and 500-510 nm. The addition of cadmium into the bath produces a dramatic 

decrease of the photoluminescent emission and a small shift of the emission 

peak toward higher wavelengths. 

CONCLUSIONS 

High quality zinc-cadmium sulfide thin films with special optical properties 

were prepared by chemical bath deposition onto optical glass platelets using 

zinc acetate- cadmium acetate– sodium citrate - ammonia – thiourea system. 

The deposition from chemical bath with variable cadmium amounts leads to 

ZnCdS /glass/ ZnCdS heterostructures with very different morpho-structural 

and optical (transmittance/reflectance) characteristics. 

Photoluminescence measurements performed on some Cu-doped 

ZnCdS films illustrate that CBD grown metal chalcogenide films can be activated 

with different cations, using the indirect doping technique. Further investigations 

are needed in order to obtain more performing luminescent ZnCdS layers. 

209


210 


ZnCdS thin films have been grown on 3 x 4.5 x 1 cm 3 optical glass 

substrates by CBD using the multilayer technique already employed. The 

deposition of ZnCdS thin films was carried out using the method employed for 

the chemical bath deposition of ZnS and CdS, respectively. The starting point 

was the chemical bath used for ZnS film deposition, as follows: [zinc acetate] = 

0.015 M; [sodium citrate] = 0.045 M; [ammonia] = 0.3 M; [thiourea] = 0.15 M. 

ZnCdS thin films were deposited from chemical bath obtained by replacing 

5 ÷100 mol % of the zinc acetate from the standard chemical bath with cadmium 

acetate. The glass substrates have been previously ultrasonically cleaned in 

acetone-ethanol mixture. During the deposition, the bath temperature was 

maintained at 82-86°C and the solution pH at 9.5-10.5. Luminescent ZnCdS 

thin films were prepared using indirect doping technique. In this purpose, 

ZnCdS /glass/ ZnCdS heterostructures were introduced into a Cu-containing 

doping mixture based on high-purity ZnS powder and annealed at 550°C. The 

details of the experimental technique and conditions have been described in 

our previous works [16, 18-20] 

Chemical bath deposited ZnCdS /glass/ ZnCdS structures have 

been analysed as grown or after annealing. Packing density was determined 

by the micro-weighing method. An UNICAM Spectrometer UV4 has been used 

for optical investigations of ZnCdS thin films and a Perkin Elmer 204 

Fluorescence Spectrophotometer was used to investigate photoluminescence 

characteristics. The structural properties of the films were studied with a standard 

DRON-3M Diffractometer using the filtered Kα emission of molybdenum 

(λ=0.70932 Å). Morphology of the films was studied with a JEOL-JSM 5510LV 

Electron Microscop. Chemical composition of the films was detected by Energy 

Dispersive X-ray Analysis (EDX) in a JSM 5510LV Scanning Electron Microscop 

atached with an Oxford Instruments, Inca 200 analytical system. 


The work was financially supported by the Romanian Ministry of 

Education and Research MEC-CNCSIS Grant Td 8/52- 2007. 

REFERENCES 

1. I. O. Oladeji, L. Chow, Thin Solid Films, 1999, 399, 148. 

2. T. B. Nasr, N. Kamoun, M. Kanzari, R. Benaceur, Thin Solid Films, 2006, 500, 4. 

3. J. Lee, Applied Surface Science, 2005, 252, 1398. 

4. K. Sambhu, C. O. Larry, Thin Solid Films, 2005, 471, 298.


5. Z. Khefaca, M. Mnari, M. Dachraoui, Physical&Chemical News, 2003, 14(1), 77. 

6. T. B. Nasr, N. Kamoun, C. Guasch, Applied Surface Science, 2008, 254, 5039. 

7. R.S. Mane, C.D. Lochande, Materials Chemistry Physics, 2000, 65, 1. 

8. L.X. Shao, K.H. Chang, H.L.Hwang, Applied Surface Science, 2003, 212-213, 305. 

9. C.T. Hsu, Thin Solid Films, 1998, 335, 284. 

10. M. P. Valkonen, S. Lindroos, T. Kanniainen, M. Leskalä, U. Tapper, E. Kauppinen, 

Applied Surface Science, 1997, 120, 58. 

11. S.Yano, R.Schroeder, B.Ullrich, H. Sakai, Thin Solid Films, 2003, 423, 273. 

12. L. Zhou, Y. Xue, J. Li, Journal of Environmental Sciences Supplement, 2009, S76. 

13. M. Ichimura, T. Furukawa, K. Shirai, F. Goto, Materials Letters, 1997, 33, 51. 

14. J. M. Dona, J. Herrero, Thin Solid Films, 1995, 268, 5. 

15. I. O. Oladeji, L. Chow, Thin Solid Films , 2005, 474, 77. 

16. M. Lădar, E.J. Popovici, I. Baldea, R. Grecu, E. Indrea, Journal of Alloys and 

Compounds, 2007, 434-435, 697. 

17. D. S. Boyle, O. Robbe, D. P. Halliday, M. R. Heinrich, A. Bayer, P. O’Brien, 

D. J. Otway, M. D. G. Potter, Journal of Materials Chemistry, 2000,10, 2439. 

18. M. Lădar, E- J. Popovici, L Pascu, R. Grecu, I.C. Popescu, E. Indrea, Studia 

Universitatis Babes-Bolyai, seria Physica, 2003, 2, XLVIII, 469. 

19. R. Grecu, E. J. Popovici, M. Lădar, L. Silaghi-Dumitrescu, E. Indrea, Studia 

Universitatis Babes-Bolyai, seria Physica, 2003, 2, XLVIII, 472. 

20. R. Grecu, E. J. Popovici, M. Lǎdar, L. Pascu, E. Indrea, Journal of Optoelectronics 

and Advanced Materials, 2004, 6, 127. 

211


EQUILIBRIUM STUDY ON ADSORPTION PROCESSES OF 

4-NITROPHENOL AND 2, 6-DINITROPHENOL ONTO 

GRANULAR ACTIVATED CARBON 

MIHAELA-CLAUDIA TERTIŞ a , FLORINA IONESCU a , 

MARIA JITARU a 

ABSTRACT. Adsorption of 4-nitrophenol and 2, 6-dinitrophenol onto granular 

activated carbon has been studied. Adsorption experiments were carried out 

in a batch system and were followed by UV-Visible spectroscopy over a period 

of 120 min. Adsorption isotherms were derived at 25 o C and the isotherm data 

were treated according to Langmuir, Freundlich and Tempkin isotherm equations. 

The fitting of experimental data was tested and the parameters of these 

equations were determined. Based on the correlation coefficients both Langmuir 

and Freundlich models are suitable for the study (the squares of correlation 

coefficients are all > 0.97). Based on the values of normalized percent deviation 

P, the Freundlich model is suitable for 4-NP and the Langmuir model is suitable 

2,6-DNP adsorption onto granular activated carbon type NORIT GAC 1240W 

(value of P less than 5). The calculated adsorption capacity values (qmax) are: 

277.77 mg g -1 for 4-NP, respective 41.15 mg g -1 for 2, 6-DNP. 

Keywords: adsorption, nitrophenols, isotherm, activated carbon 

INTRODUCTION 

Among the environmentally concerned substances, nitrophenols 

represent one of the most common groups of highly toxic water pollutants. 

The monosubstituted 4-nitrophenol is found in wastewaters discharged from 

various industrial activities such as pulp and paper industries, textile mills, 

steel plants, oil refineries, etc. [1]. This compound is also associated with 

agricultural activities as an intermediate for the production of pesticides, 

herbicides and insecticides [2]. 

Phenol and some of its derivatives are considered priority pollutants 

with a permissible limit of 0.1 mg mL -1 in wastewater [3-6] and must be removed 

from industrial effluents before discharge into environment. Particular attention 

has been given to 4-nitrophenol which presents some undesired effects and 

has a permissible limit of 0.06 mg mL -1 [3, 6]. 

a Research Centre LAF-INT-ECOL, Faculty of Chemistry and Chemical Engineering, “Babes- 

Bolyai” University, 11 Arany Janos Street, 400028

214 

MIHAELA-CLAUDIA TERTIŞ, FLORINA IONESCU, MARIA JITARU 

Attempts have been made to remove mononitrophenols from wastewater 

by a number of methods: oxidation with strong oxidizing agents as H2O2 [7], 

biodegradation [8], biosorption [9], photo catalytic degradation [10], etc. 

Adsorption is an efficient, simple and inexpensive method for the 

removal of organic pollutants. As adsorbents, activated carbons, polymeric 

resins, silica, fly ash, zeolites were commonly used. Activated carbons are 

preferred adsorbents in industrial processes even they had poor mechanical 

properties and difficult regeneration processes [11]. 

Phenolic compounds in aqueous solutions can exist as phenolate ions, 

depending on the pH of the solution. Conversion of phenol and nitrophenol 

derivatives to phenolate anions is negligible in water solutions considering 

its small acidic dissociation constants [12, 13]. 


The equilibrium nitrophenols solutions concentrations were determined 

by the aid of the calibration curves equations, obtained from spectrophotometric 

measurements (UV-Visible spectrophotometer UNICAM HELIOS and DR2800 

HACH-LANGE). 

The equilibrium adsorption capacities of the adsorbent were calculated 

using equation (1). 

V ( C0 

− Ce) 

qe 

= (1) 

m 

where qe is the adsorption capacity of adsorbent material (mg nitrophenol/g 

of adsorbent); C0 is the initial concentration of nitrophenol (mg L -1 ); Ce is the 

equilibrium concentration of nitrophenol (mg L -1 ); V is the volume of nitrophenol 

solution (L) and m is the mass of the adsorbent (g). 

Adsorption characteristics and calibration data for nitrophenol 

derivatives 

In water and acid solution, nitrophenols derivatives are in neutral 

form, with inessential amount of ionisation products [12, 13]. The spectral 

and calibration data for the studied compounds are given in Table 1. 

It can be observed that both for 4-NP and 2, 6-DNP the UV-Visible 

spectra present two adsorption bands having significantly different values 

for the wavelength of maximum absorption (λmax: 223 and 317nm in the case 

of 4-nitrophenol respective 222 and 430nm in the case of 2, 6-dinitrophenol). 

Although the calibration data were derived at both λmax values for each 

compound, only one of them was utilized in this experiment due to the higher 

intensities of adsorption bands (ε = 9330 at 317nm for 4-NP respectively 

ε = 6280 at 430nm for 2,6-DNP).

EQUILIBRIUM STUDY ON ADSORPTION PROCESSES OF 4-NITROPHENOL… 

Table 1. Spectral and calibration data for treated nitrophenol derivatives in water 

Nitro phenol 

4-NP 

HO NO2 

HO 

2,6-DNP 

O 2N 

O2N 

λmax 

(nm) 

ε 

(M -1 cm -1 ) 

317 9330 

430 6280 

Equation for 

calibration curve * 

C = 

4−NP 

A 

0. 

2793 

A 

C 2, 

6−DNP 

= 

0.0194 

Standard 

deviation 

for A(%) 

Regression 

coefficient 

(R 2 ) 

± 0.3÷1.4 R 2 = 0.9902 

± 0.4÷13 R 2 = 0.9989 

* 7 calibration points for 4-NP; 14 calibration points for 2, 6-DNP; 3 different 

determinations for each calibration point. 

The use of the data corresponding to the other two wavelengths of 

maximum adsorption, which have smaller values for ε, will affect the quantitative 

determination. On the other hand, due to the difference of about 100 nm 

between λmax the method can be used for determination of both nitro phenols 

derivatives in water. 

Adsorption isotherms 

Adsorption isotherms at 25 o C, derived for 4-nitrophenol and 2, 6dinitrophenol 

in water, are shown in Figures 1(a) and 1(b). All the measurements 

were three times repeated and the values used represent the arithmetical 

mean of the correspondent three values. 

The isotherm data were analyzed according to three well known 

isotherm equations: Langmuir, Freundlich and Tempkin, whose liniarized forms 

are given in equations (2)-(4) respectively [14-16]. 

- The liniarized form for Langmuir isotherm equation: 

Ce Ce 

1 

= + 

q q bq 

e 

max 

max 

- The liniarized form for Freundlich isotherm equation: 

⎛ 1 ⎞ 

ln qe = ln K F + ⎜ ⎟ln Ce 

(3) 

⎝ n ⎠ 

- The liniarized form for F Tempkin isotherm equation: 

q = k k + k ln C 

(4) 

e 

1 ln 2 1 

e 

(2) 

215

216 


where qe is the amount of adsorbate adsorbed per unit mass of adsorbent at 

equilibrium (mg nitrophenol g -1 NORIT GAC 1240W); Ce is the final concentration 

at equilibrium (mg L -1 ); qmax is the maximum adsorption at monolayer coverage 

of surface (mg nitrophenol g -1 NORIT GAC 1240W); b is the adsorption 

equilibrium constant related to the energy of adsorption (L mg -1 ); KF is a 

Freundlich constant representing the adsorption capacity (mg g -1 )(L mg -1 ) 1/n ; 

n is a constant depicting the adsorption intensity; k1 the Tempkin isotherm 

energy constant (L mg -1 ) and k2 the Tempkin isotherm constant. The main 

difference between these three isotherm models is in the variation of heat 

of adsorption with the surface coverage. Langmuir model assumes uniformity, 

Freundlich model assumes logarithmic decrease and Tempkin model assumes 

linear decrease in heat of adsorption with surface coverage. 

qe (mg 4-NP g -1 NORIT GAC 

qe (mg 2,6-DNP g -1 NORIT GAC 

1240W) 

1240W) 

70 

60 

50 

40 

30 

20 

10 

0 

350 

300 

250 

200 

150 

100 

50 

0 

(a) 

0 100 200 300 400 

(b) 

Ce (mg L -1 ) 

0 2 4 6 8 10 

Ce (mg L -1 ) 

Figure 1. The fit of experimental data (♦) to Langmuir (■) and Freundlich (○) 

models at 25 0 C for: (a) 4-nitrophenol and (b) 2, 6-dinitrophenol in water.


The Langmuir equation is valid for monolayer adsorption of adsorbate 

onto adsorbent surface and assumes there are a definite and energetically 

equivalent number of adsorption sites. The bonding to adsorption sites can 

either be chemical or physical, but it must be sufficiently strong to prevent 

displacement of adsorbed molecules along the surface. This model also 

presumes that molecules adsorbed on neighbouring sites do not interact 

each other [11, 17]. 

(a) 

Ce/qe (g L -1 ) 

Ce/qe (g L -1 ) 

1.6 

1.4 

1.2 

1 

0.8 

0.6 

0.4 

0.2 

y = 0.0036x + 0.3212 

R 2 = 0.974 

0 

0 50 100 150 200 250 300 350 

0.3 

0.25 

0.2 

0.15 

0.1 

0.05 

Ce (mg L -1 ) 

(b) 

y = 0.0243x + 0.0382 

R 2 = 0.9932 

0 

0 2 4 6 8 10 

Ce (mg L -1 ) 

Figure 2. Linear form of Langmuir adsorption isotherm for: (a) 4-NP and (b) 2, 6- 

DNP adsorption on granular activated carbon type NORIT GAC1240W, at 25 o C. 

The liniarized forms of Langmuir, Freundlich and Tempkin isotherm 

equations are given in equations (2)-(4), which describe these models. Figures 

2-4 present the obtained results, and the parameters of these equations are 

given in Table 2. 

217

218 


Table 2. The equations parameters of Langmuir, Freundlich and Tempkin 

adsorption isotherm for nitrophenolic compounds at 25 o C. 

Nitro 

phenol q max 

(mg g -1 ) 

Langmuir parameters 

b 

(L mg -1 ) 

Freundlich 

parameters 

K 

F 

1/n 

Tempkin 

parameters 

k 1 

(L mg -1 ) 

(mg g -1 ) 

(L mg -1 ) 1/n 

4-NP 277.77 86.44 2.04 1.52 33.6 0.59 

2,6-DNP 41.15 0.63 2.01 2.30 5.4 36.39 

lnqe 

6 

5 

4 

3 

2 

1 

(a) 

y = 0.7144x + 1.5204 

R 2 = 0.9901 

0 

-1 0 1 2 3 4 5 6 7 

lnCe 

y = 0.6988x + 2.2994 

R 2 = 0.9865 

(b) 

-2 

k 2 

-5 -4 -3 -2 -1 

0 

0 

-1 

1 2 3 

Figure 3. Linear form of Freundlich adsorption isotherm for: (a) 4-NP and (b) 2, 6-DNP 

adsorption on granular activated carbon type NORIT GAC1240W, at 25 o C. 

q e (m g 4-N P g -1 N OR IT GA C 

1240W ) 

250 

200 

150 

100 

50 

(a) 

y = 33.607x - 17.826 

R 2 = 0.8026 

0 

-1 0 

-50 

1 2 3 4 5 6 7 

lnCe 

lnq e 

lnCe 

(b) 

y = 5.4003x + 19.41 

R 2 = 0.8766 

0 

-5 -4 -3 -2 -1 0 1 2 3 

-10 

Figure 4. Linear form of Tempkin adsorption isotherm for: (a) 4-NP and (b) 2, 6-DNP 

adsorption on granular activated carbon type NORIT GAC1240W, at 25 o C. 

q e (m g 2,6-D N P g -1 N OR IT 

GA C 1240W ) 

5 

4 

3 

2 

1 

lnC e 

40 

30 

20 

10


A better criterion to test the correlation between experimental data 

to one of the three isotherm equations (Langmuir, Freundlich and Tempkin), is 

a parameter known as normalized percent deviation [14], or percent relative 

deviation modulus, P, [15, 16], given by the equation (5): 

⎛100 

⎞ ⎛ q 

⎞ 

⎜ e(exp) 

− qe( 

pred ) 

P = ⎜ ⎟∑ 

⎟ 

(5) 

⎝ N ⎠ ⎜ q ⎟ 

⎝ e(exp) 

⎠ 

where qe(exp) is the experimental qe at any Ce; qe(pred) is corresponding 

predicted qe according to the equation under study with best fitted parameters; 

N is the number of measurements (5 in our case). It is generally accepted 

that when the P value is less than 5, the fit is considered to be good [15]. 

The values for percent relative deviation modules, P, calculated for all three 

models and both nitrophenol derivatives are presented in Table 3. It can be 

seen that, in the case of 4-nitrophenol the value of P is less than 5 only for 

Freundlich model (P=0.9), and in the case of 2, 6-dinitrophenols for Langmuir 

model (P=2). 

Table 3. Values of normalized percent deviations (P) for Langmuir, Freundlich 

and Tempkin models for nitrophenolic derivatives, at 25 o C. 

Nitro 

phenol 

P 

Langmuir 

model 

P 

Freundlich 


P 

Tempkin 


4-NP 10.99 0.90 26.60 

2,6-DNP 2.00 26.20 80.00 

The efficiency of adsorption process can be predicted by the 

dimensionless equilibrium parameter RL, which is defined by the equation (6): 

R L 

1 

= (6) 

1+ 

b ⋅C 

0 

where b is the Langmuir constant (L mg -1 ); C0 the initial concentration of 

nitrophenolic compound (mg L -1 ). Isotherm is considered to be unfavourable 

when RL > 1, linear when RL = 1, favourable when 0 < RL < 1 and irreversible 

when RL = 0 [16, 18, 19]. 

The RL calculated values are given in Table 4. 

219

220 


Table 4. Values of equilibrium parameter RL for Langmuir, Freundlich and 

Tempkin models for nitrophenolic derivatives, at 25 o C. 

Nitro 

phenol 

4-NP 

2,6-DNP 

C0 

(mg L -1 ) 

13.91 

69.55 

139.11 

695.55 

1391.10 

1.84 

9.20 

18.41 

92.04 

184.11 

RL 

0.46 

0.15 

0.08 

0.02 

0.01 

8.30 x 10 -4 

1.60 x 10 -4 

8.30 x 10 -5 

1.60 x 10 -5 

8.30 x 10 -6 

Since all the RL values calculated for treated nitrophenols are between 

0 and 1, the adsorption processes are favorable in all cases. 

CONCLUSIONS 

In this study granular activated carbon type NORIT GAC1240W was 

used as adsorbent for 4-nitrophenol and 2, 6-dinitrophenol, to evaluate the 

applicability to remove these nitrophenols from water solutions. The experiments 

were conducted in a batch mode at constant working conditions. The equilibrium 

adsorption data were described by the Langmuir, Freundlich and Tempkin 

mathematical models. The values of the constants and correlation coefficients 

in all three models were calculated. Based on the correlation coefficients both 

Langmuir and Freundlich models are suitable for the study (the squares of 

correlation coefficients are all > 0.97). 

According to the analysis of the results, and the values of normalized 

percent deviation P, the experimental data for 4-NP adsorption onto granular 

activated carbon type NORIT GAC1240W were correlated reasonably by the 

Freundlich model, unlike those for 2, 6-DNP adsorption which were correlated 

by the Langmuir model. 

From the liniarized Langmuir equation, the calculated adsorption 

capacity values (qmax) are: 277.77 mg g -1 for 4-NP, respective 41.15 mg g -1 for 

2, 6-DNP adsorption on granular activated carbon type NORIT GAC 1240W. 

Because the equilibrium parameter RL values are close to 0 (10 -5 ÷10 -6 ) 

for 2, 6-dinitrophenol at higher concentrations (5x10 -4 ÷10 -3 mol L -1 ), it is possible 

to presume that, in this condition, adsorption is irreversible. Future experiments 

are necessary to study the reversibility of the nitrophenol compounds adsorption 

onto activated carbon.



Materials and methods 

4-NP with purity greater than 98% (Merk, Germany), and 2, 6-DNP 

with purity greater than 95% (calculated based on dry substance), moistened 

with 20% H2O (ALDRICH, Switzerland) were used to prepare the solutions, 

with desirable concentration, for the experiments in this study. Distilled water 

was used to prepare the aqueous solutions. All the reagents were analytical 

grade and used without further purification. 

The activated carbon used in the present work was NORIT GAC 1240W, 

obtained from NORIT (Netherlands), having the following characteristics: 

micropores volume: 0.38 cm 3 g -1 ; specific area: 1062 m 2 g -1 ; mesopores 

volume: 0.45 cm 3 g -1 ; apparent density: 495 kg m -3 [20]. 

Before use carbon was dried at 105 0 C for 12 h and stored in desiccators 

at room temperature. 

In adsorption experiments, the concentration of nonadsorbed 4nitrophenol 

and/or 2, 6-dinitrophenol was determined from the absorbance 

of the species obtained spectrophotometrically at a wavelength of 316 nm 

for 4-nitrophenol and 430 nm for 2, 6-dinitrophenol. A Unicam Helios B 

spectrophotometer with the specific software VISION 32, and a quartz vat of 

2 ml, with optical route of 1 cm, and a Direct Reading Spectrophotometer 

type DR/2800 HACH-LANGE with a quartz vat of 2 ml and with optical route of 

1 cm were used to measure the adsorption intensities of the species. 

The amount of the adsorbed nitrophenol was calculated from the 

equations which express the connection between the nitrophenols absorbance 

and concentration. 

Batch adsorption experiments 

Adsorption experiments were carried out using the conventional batch 

technique. Nitrophenol solutions were prepared by dissolving required amount 

of solid nitrophenol in distilled water. The initial concentrations of nitrophenolic 

compounds (between 13.91 and 1391.10 mg L -1 for 4-NP, respective 1.84 and 

184.11 mg L -1 for 2, 6-DNP); the amount of activated carbon (1 g activated 

carbon to 0.2 L of nitrophenolic solution) and the temperature (25±2) 0 C were 

kept constant during the adsorption experiments. 

Isotherm tests 

For adsorption isotherms tests, granular activated carbon was weighed 

and transferred to several glass containers. A known concentration of adsorbate 

solution was added and the containers were sealed and placed onto a 

TERMOMIX GRANT LTD6G thermostat where they were maintained at constant 

temperature of 25 o C for 48 h to ensure that equilibrium was reached. The 

adsorption experiments occurred under mechanical stirring (350 rpm) with 

an AGITUVAR 10W stirrer. 

221

222 


REFERENCES 

1. Spectrum Laboratories, Chemical Fact Sheet – CAS # 100027; 

(http://www.speclab.com). 

2. The International Programme on Chemical Safety (IPCS), Cincise, International 

Chemical Assessment Document 20 (CICAD), Mononitrophenols. 

3. USEPA, Technical support document for water quality based toxics control, 

U.S. Environmental Protection Agency, Washington DC, USA, 1991. 

4. S.E. Manahan, “Fundamentals of Environmental Chemistry”, CRC Press, Boca 

Raton, 2000. 

5. P.M. Alvarez, J.F. Garcia-Araya, F.J. Beltran, F.J. Masa, F. Medina, Journal of 

Colloid and Interface Science, 2005, 283, 503. 

6. B. Pan, X. Chen, W. Zhang, X. Zhang, Q. Zhang, Journal of Hazardous 

Materials, 2006, 137, 1236. 

7. Y.S. Li, Y.H. You, E.T. Lien, Archives of Environmental Contamination and 

Toxicology, 1999, 4, 427. 

8. J-P. Arcangeli, E. Arvin, Water Science and Technology, 1995, 31, 117. 

9. B. Koumanova, Z. Kircheva, Journal of the University of Chemical Technology 

and Metallurgy (Sofia), 2003, 38 (1), 71. 

10. M. Salaices, B. Serrano, H.I. de Lasa, Chemical Engineering Science, 2004, 59, 3. 

11. M. Erdem, E. Yuksel, T. Tay, Y. Cimen, H. Turk, Journal of Colloid Interface 

Science, 2009, article in press, doi: 10.1016/j.jcis.2009.01.014. 

12. E. Ayranci, O. Duman, Journal of Hazardous Materials B, 2005, 124, 125. 

13. J.B. Lambert, H.F. Shurvell, D. Lightner, R.G. Cooks, „Introduction to Organic 

Spectroscopy”, Macmillan Publishers, New York, 1987. 

14. R-S. Juang, R-L. Tseng, F-C. Wu, S-H. Lee, Separation Science and 

Technology, 1991, 26, 661. 

15. C.J. Lamauro, A.S. Bakshi, T.B. Labuza, Lebensm-Wiss Technology, 1985, 18, 111. 

16. E. Ayranci, O. Duman, Journal of Food Engineering, 2005, 70, 83. 

17. W.S. Wan Ngah, S. Fatinathan, Colloids and Surfaces A: Physicochemical and 

Engineering Aspects, 2006, 277, 214. 

18. K. Kadirvelu, C. Namasivayam, Advances in Environmental Research, 2003, 7, 471. 

19. R. Sivaraj, C. Namasivayam, K. Kadirvelu, Waste Management, 2001, 21, 105. 

20. I. M. Gullon, R. Font, Water Research, 2001, 35, 516.


STRATEGIES OF HEAVY METAL UPTAKE BY PHASEOLUS 

VULGARIS SEEDS GROWING IN METALLIFEROUS AND 

NON-METALLIFEROUS AREAS 

CAMELIA VARGA a , MONICA MARIAN a , ANCA PETER a* , DELIA BOLTEA a , 

LEONARD MIHALY-COZMUTA a , EUGEN NOUR b 

ABSTRACT. This study focuses on the accumulation of metal ions (Pb 2+ , 

Zn 2+ , Cu 2+ , Fe 2+ ) in Phaseolus vulgaris seeds (Fabaceae family) collected from 

metalliferrous / non- metalliferrous areas in Maramures County. By “metalliferous 

areas” we will understand following in this research paper only areas displaying 

such pollution, as opposite to the other areas, consequently labeled as “nonmetalliferous”. 

Four different concentrations of metal ions were used and their 

imbibition’s degree into the seeds was investigated. The accumulation of metal 

ions in the seeds of P. vulgaris increases as the initial concentration of metal 

ions is higher. Even more, the seeds growing in non-metalliferous areas display 

a higher absorption of all metal ions, except the Fe 2+ , than the seeds growing 

in metalliferous areas. From an anatomical point of view, we have observed 

that regardless of concentration, the metal ions were able to penetrate the 

seed through the hilum, up to the coat. 

Keywords: Phaseolus vulgaris, lead, copper, iron, zinc, heavy metal 

accumulation, atomic absorption spectrometry, optical microscopy 

INTRODUCTION 

Plants usually show the ability to accumulate large amounts of metals 

without visible changes in their appearance or yield. In many plants, the level 

of metal accumulation can exceed even several hundred times the maximum 

level permissible for human beings, without having any negative effect on 

their growth or yield [1]. Therefore, it seems that plants can endure a level 

of environmental pollution that might be even several times higher than the 

level observed nowadays. 

The persistence of metallic pollutants in soil is the most difficult 

issue. The presence of metals accumulated in soil will span hundreds of years, 

impacting to a lesser or greater extent plants grown on such sites [2]. The plants 

a 

Universitatea de Nord Baia Mare, Faculty of Science, Department of Chemistry-Biology, 430083, 

Baia Mare, Romania 

* 

coresponding author: Anca Peter, e-mail peteranca@yahoo.com 

b 

Inspectorate of Emergency Situation, Maramures, Romania

224 

C. VARGA, M. MARIAN, A. PETER, D. BOLTEA, L. MIHALY-COZMUTA, E. NOUR 

growing in metalliferous habitats probably have the ability to inactivate the 

heavy metal by binding the excess metal ions and/or by changing the chemical 

composition and physical organization of their cell membranes [3]. There are 

numerous investigations [4, 5] on the influence of heavy metals on the plants 

metabolism, but data remains scarce about the accumulation of heavy metals 

in the reproductive organs of plants [6, 7]. Stefanov et. al. [3], which studied the 

accumulation of lead, zinc and cadmium in different plants seeds, showed that 

plants accumulate selectively heavy metal ions in their seeds. Peanut and corn 

seeds accumulate mainly lead, pea seeds accumulate mainly cadmium and 

wheat seeds accumulate mainly zinc. On the other hand, Lane and Martin [8] 

showed that the seed coats of Raphanus sativus were a strong barrier to lead 

and helped prevent contamination of embryos until the seed coat was torn 

apart by the germinating embryonic root. There are reports on the inhibitory 

effect of lead on the germination of seeds of the following species Lupinus 

luteus [9], Oryza sativa [10] and Sinapis alba [11]. The fact that a method of 

measuring the tolerance of plants to metals entails sowing the seeds of tolerant 

and sensitive species on metal-contaminated soil points to a significant effect 

of heavy metals on germination [12]. The above studies, unlike those by Lane 

and Martin [8], point to the significant influence of lead on seed germination. 

In this situation it seems probable that the effect of lead on germination depends 

on interspecies differences in seed structure-in particular, on differences in the 

structure of seed coats. As it is well known, the role of the seed coat is to 

protect the embryo from harmful external factors. But seed coats have a wide 

range of anatomic forms that exist in no other plant organ or tissue [13]. 

Wierzbicka et. al [1] showed that from all plants families tested, the Fabaceae 

family is very susceptibile to the higher concentration of metal ions. 

The aims of this study were to evidence the role playing the seeds 

coat of bean, but also the vulnerability of these seeds to the heavy metals 

presence. The role of the seed coats as heavy metal barriers is often bypassed 

by the high concentration of these chemical species, this fact generating a 

risk of these seeds consumption. We present a selection of research data 

on the accumulation of heavy metals (Pb 2+ , Zn 2+ , Cu 2+ , Fe 2+ ) in P. vulgaris 

seeds growing in metalliferous and non-metalliferous areas in Maramures 

county. Investigating how the initial concentration of metal ions impacts the 

imbibition capacity of seeds and, on the other hand, establishing the degree 

to which those ions have penetrated the seeds were our key research goals. 

We have also researched how the imbibition capacity of seeds varies 

depending on their area of origin (metalliferous or not). 


Figure 1 displays the P. vulgaris seeds, as a whole and split, impregnated 

with different ions in different concentrations and originating from metallic and 

non – metallic areas. All images reveal the imbibition of metal ions into the

STRATEGIES OF HEAVY METAL UPTAKE BY PHASEOLUS VULGARIS SEEDS GROWING … 

seeds as taking place starting with the hilum and going up to the cotyledon, as 

the colored zones on the seeds split demonstrate. Moreover, as substantiated 

in all images by the change in color of the tegument of seeds in whole, the 

penetration of metal ions takes place up to the tegument of the seed. On 

the other hand, the analysis of images of seeds impregnated with the same 

metal ion (Figures 1(a)-(b), 1(c)-(d), 1(e)-(f) and 1(g)-(h) respectively) reveals 

that the intensity of color inside the seed increases as the concentration of 

metal ions is higher (namely green for Fe 2+ , red-brown for Zn 2+ , red for Cu 2+ 

and yellow for Pb 2+ ). The optical microscopic analyses confirm this observation 

(see below). 

The macroscopic views of seeds as illustrated in Figures 1 (h) and 

(i) reveal that seeds impregnated with Pb 2+ ions, at the same concentration 

(10. 000 mg/L), display different colors depending on the source of each seed 

(coming either form a metalliferous area – Ferneziu, in our particular case, 

or from a non–metalliferous one – Oarta de Jos, in our case). The images 

demonstrate that in seeds originating in a metalliferous area (as in Figure 1 (i)) 

Pb 2+ has prevalently accumulated in the tegument and in the superior part of the 

cotyledons, while the accumulation in seeds originating from a non–metalliferous 

area take place in the whole cotyledons and tegument. 

Thus, the absorption capacity of seeds originating from non–metalliferous 

areas exceeds the absorption capacity of those from metalliferous ground. This 

happens as plants growing in metalliferous sites have the ability to inactivate the 

excess of metal ions and/or to alter the chemical composition and/or physical 

layout of membranes of their cells [3]. 

At the lowest concentration in Fe 2+ we have tested (see Figure 2 (a)) no 

presence of the Fe 2+ ions into seed structures was microscopically observed. 

In opposition to this, at 50 mg/L in concentration (see Figure 2 (b)) the 

penetration of Fe 2+ into the cotyledon and going up to the tegument can be 

noticed, to stabilize in the area between tegument and cotyledon. A more or 

less homogeneous distribution of metal ions in the cotyledon is obvious in 

Figure 2(d). The green color of the cotyledon at the highest concentration of 

Fe 2+ , as in Figure 2(d), is clearly more intense than the color displayed at 

lower concentration of Fe 2+ , as in Figure 2(c). The imbibition’s degree trends 

up, as the concentration in Fe 2+ increases. The analysis of Fe 2+ ions in 

terms of their imbibition capacity depending on the origin of the seeds (as in 

Figures 2(a) and (b)), has led us to conclude that the seeds originating from 

a non-metalliferous area have a higher capacity to absorb Fe 2+ ions than those 

originating from a metalliferous area. In particular, Fe 2+ ions penetrate the 

seeds originating from a non-metalliferous area up to the cotyledon, to a 

smaller extent the tegument and agglomerate in the area between the 

tegument and the cotyledon (the green layer). We have remarked no Fe 2+ 

ions penetration at all, however, in seeds originating from a metalliferous 

area. 

225

226 


(a) 

(c) 

(e) 

(g) 

(i) 

(b) 

(d) 

(f) 

(h) 

Figure 1. Macroscopic view of P. vulgaris 

seeds impregnated with different ions, at 

different concentrations, originating from 

metalliferous (F – Ferneziu) and nonmetalliferous 

(O – Oarta de Jos) areas. 

The concentrations units are mg/L.


This behavior is explained by the fact that the iron (known as essential 

trace element for plants [14]) diminish the toxic effect of the other metals by 

compete with these for binding sites on the cell membranes, followed by 

penetration and accumulation into the cells [15, 16, 17]. This ionic exchange 

process is part of the plant strategy to survive on soils strongly polluted with 

heavy metals. 

(a) - Fe 10 F 

(c) - Fe 50 F 

(e) - Cu 10 O 

(b) - Fe 10 O 

(d) - Fe 250 F 

(f) - Cu 20 F 

(g) - Zn 15 F 

(h) - Zn 15 O 

Figure 2. P. vulgaris seed – cross sections as captured in optical microscope 

images at 10 X magnification to detect how the concentration of metal ions 

impacts the imbibition degree. 

227

228 


As we have monitored the process of imbibition of seeds with Cu 2+ 

ions (see Figures 2(e) and (f)) we have noticed the trend of these ions to 

agglomerate on the cellular wall [3, 5, 18] (red lines between cells) and in 

the tegument, regardless of the concentration and origination of seeds. 

We have conducted the microscopic analysis of seeds impregnated 

with Zn 2+ ions (see Figures 2(g) and (h)) and Pb 2+ ions (not shown) exactly 

as we did to research the imbibition process with Cu 2+ ions. We have in this 

case noticed that regardless of concentration and origination of seeds, the 

ions agglomerate on the cellular wall (brown lines between cells, for Zn 2+ ) 

and on the tegument. 

Table 1 gives the range of concentrations we have used for the metal 

ions. We have concluded, as outcome valid for all the metal ions, that the higher 

their initial concentration was (as in Figure 3), the higher the concentration of 

metal in seeds is. 

We have prepared our initial solutions with the pH ranging between 

3,99 and 5. 41 (as revealed in Table 1) and with weak acidity. The process 

by which Pistacia vera L. removes Pb 2+ ions from aqueous solutions takes place 

with maximum efficiency (ca. 93,8–95,1%) at a pH ranging from 3,5 to 5,5, 

as demonstrated by Yetilmezsoy et. al. [19]. The absorption of metal ions 

into seeds subsides at low pH (namely at high concentrations of hydrogen), as 

hydrogen ions compete with those of metals in the attempt to penetrate the 

plant [20]. As the pH grows above 7, a decline in mobility of the metal ions 

in aqueous solutions translates into those ions lower probability to reach 

and penetrate the seed. [21]. 

The influence of P. vulgaris seeds provenience on the Cu content in 

seeds is illustrated in Figure 4. Seeds originating in non-metalliferous area 

(Oarta de Jos) display a higher concentration in copper than seeds in 

metalliferous area (Ferneziu). Zinc and lead displays a behavior comparable 

to copper, as confirmed during our optical microscopic analyses. The ability 

of plants to inactivate the heavy metal by binding the excess and/or by 

changing the chemical composition and physical organization of their cell 

membranes [3] stands as explanation for this outcome. Moreover, the plants 

growing on polluted areas develop, in time, some mechanisms to survive in 

drastically ecologic conditions. The selection of this kind of resistant plants 

constitutes during the time a particularly ecotype. 

Iron makes the exception to the above – mentioned behavior. Figure 5 

reveals how the origin of seeds has an influence on the Fe content. Seeds 

originating in non-metalliferous areas (Oarta de Jos) display a lower Fe 

concentration than the seeds from metalliferous sites (Ferneziu).


Table 1. Particularities of the metal ions investigated in P. vulgaris seeds. 

Sample ID Seed [Metal 

Provenience 

ion] 2+ 

[Metal 

initial/pH 

(mg / L) 

ion] 2+ 

[Metal 

soil 

(mg / Kg) 

ion] 2+ 

RSD 

seed 

(mg/Kg) 

(%) * 

Fe–10 F 10 / 4,46 40,4 1,9 

Fe-50 F 50 / 3. 99 15 287,5 138,2 7,3 

Fe-250 F 250 / 3,51 

979,2 1,8 

Reference - 30,5 4,7 

Cu-10 F 10 / 4,62 36,9 1,6 

Cu-20 F 20 / 4,58 153,2 37,6 0,2 

Cu-25 F 25 / 4,65 

115,5 0,2 

Ferneziu 

Reference - 0 0 

Zn-5 F 5 / 5,07 63,6 6,5 

Zn-10 F 10 / 5,41 93,64 93,3 1,8 

Zn-15 F 15 / 4,7 

150,4 0,8 


Pb-100 F 100 / 4,84 5. 067,4 7,3 

Pb-1. 000 F 1. 000 / 4,68 357,89 13. 793,4 0,3 

Pb-10. 000 F 10. 000 / 4,33 

Reference 

92. 247, 1 1,4 

- 8,1 14,5 

Fe-10 O 10 / 4,46 56,1 3,4 

Fe-50 O 50 / 3. 99 11 952,52 60,5 16,1 

Fe-250 O 250 / 3,51 

454,5 4,0 


Cu-10 O 10 / 4,62 72,4 1,2 

Cu-20 O 20 / 4,58 14,61 104,3 0,3 

Cu-25 O 25 / 4,65 

151,2 0,01 

Reference 

Zn-5 O 

Oarta de 

Jos 

- 

5 / 5,07 

0 

86,9 

0 

3,1 

Zn-10 O 10 / 5,41 40,41 102,5 1,0 

Zn-15 O 15 / 4,7 

172,7 1,3 


Pb-100 O 100 / 4,84 3. 989,4 2,4 

Pb-1. 000 O 1. 000 / 4,68 40,13 18. 396,1 0,9 

Pb-10. 000 O 10. 000 / 4,33 

87. 442,1 0,9 

Reference 

- 8,2 21,3 

* RSD was determined from three parallel measurements. 

229

230 


metal content in seeds (mg/kg) 

100000 

10000 

1000 

100 

10 

1 

reference 

Fe-10 F 

Fe-50 F 

Fe-250 F 

reference 

Cu-10 F 

Cu-20 F 

Cu-25 F 

reference 

Zn-5 F 

Zn-10 F 

Zn-15 F 

reference 

Pb-100 F 

Pb-1.000 F 

Pb-10.000 F 

Figure 3. Influence of the initial concentration on the metal contents in 

P. vulgaris seeds originating in Ferneziu 

Cu content in seeds(mg/kg) 

160 

140 

120 

100 

80 

60 

40 

20 

0 

Ferneziu Oarta de Jos 

reference 10 20 25 

Figure 4. Influence of provenience of the P. vulgaris seeds 

on the Cu content in seeds 

Various levels of concentration in Fe ions could generate on a plant 

substantially different outcomes. In small concentrations, Fe ions play a 

beneficial role, as they are essential to the well-being of the plant. In higher 

concentrations, however, Fe ions become a source of stress, as at sub-cellular 

level the concentration of metabolic compounds that react with oxygen goes 

up [22].


Fe content in seeds(mg/kg) 

1000 

100 

10 

1 

Ferneziu Oarta de Jos 

reference 10 50 250 

Figure 5. Dependence of the Fe content in seeds on the origin of P. vulgaris seeds 

CONCLUSIONS 

The accumulation of metal ions (Pb 2+ , Zn 2+ , Cu 2+ , Fe 2+ ) in Phaseolus 

vulgaris seeds (Fabacea family) collected from metalliferous and non-metalliferous 

areas in Maramures County is the key focus of this study. In this respect, we 

have used four different concentrations of metal ions and we have investigated 

their respective degree of imbibition into the seeds. 

♦ The intensity of color of the seed (green for Fe 2+ , red-brown for Zn 2+ , red 

for Cu 2+ and yellow for Pb 2+ ) grows as the concentration of metal ions 

increases. 

♦ Seeds originating in non-metalliferous areas display a higher adsorption 

capacity of metal ions (except iron) than seeds from metalliferous areas 

♦ The plants growing on polluted areas develop during the time, mechanisms 

to survive in drastically ecologic conditions, following a selection process, 

thus resulting a particularly ecotype. 


Seed Material 

Ferneziu is a neighborhood of Baia Mare and also a metalliferous 

area due to the mining and ore processing activity. The soils originating 

from this area is very polluted, containing heavy metals (Table 1) and toxic 

anions which influence the vegetation and fauna growing and propagation. 

Oarta de Jos is located at aproximatively 50 km away from Baia Mare, in the 

231

232 


south–east of the Maramures County (see the map in Figure 6) and is a 

non-metalliferous area due to the fact that there is no any mining and industrial 

activity. Moreover, the village is at aproximatively 10 km away from the main 

road, which is also considered a source of pollution. The concentrations of 

the reference metal ions in soil are presented in Table 1. 

We have collected P. vulgaris seeds out of the crop harvested in 

Ferneziu and respectively in Oarta de Jos, in the fall of 2008. 

Oarta de Jos 

Figure 6. Map of Maramures County. Green circles identify the locations 

from where we have collected the seeds of P. vulgaris 

for the research interests of this paper. 

Series of fifty seeds, collected in the above-mentioned areas, were 

washed three times with distillated water and immersed in solutions with 

different concentrations of heavy metal ions. The solutions containing the 

mentioned metal ions were prepared using FeSO4, Pb(NO3)2, CuSO4, ZnSO4 

salts. The initial concentrations of those solutions were established having in 

view the same metal ions concentration in soil and are presented in Table 1. 

Analyses on heavy metals 

After 7 days of immersion at 22 0 C in the solution with ions of heavy 

metals, we have washed three times the seeds with distillated water before 

initiating our analyses, conducted in two ways. 

1. We have relied on a German – made Krúss Optronic binocular 

microscope in order to investigate the degree of imbibition by optical 

microscopy. This method has entailed the use of a special reagent to color 

each category of metal ions under scrutiny. Table 2 gives the color assigned 

to each metal ion and corresponding to each reagent.


Table 2. The chemical reagents used to identify the iron, copper, 

zinc and lead ions and the colors obtained 

Heavy metal Reagent Color 

Fe K3[Fe(CN)6] 5% Green 

Cu Ditizone 0,12% in CHCl3 Red 

Zn Ditizone 0,12% in CHCl3 Brown 

Pb KI 5% Yellow 

We have maintained the seeds for 10 minutes immersed in reagent, 

and have dried them up for another 10 minutes in open air. A manual MIC – 

500 – type microtome has allowed us to produce mono-cellular microscopic 

sections. In order to use the microtome efficiently, before sectioning we 

have immersed the seeds in a paraffin bath, to get them fastened. 

2. A Perkin Elmer AAS 800 Spectrometer has allowed us to apply a 

spectrometric method in order to measure the concentration of metals in seeds. 

The drying process, in air at 1200 o C for 2 hours, took place in a US-made 

Binder drying oven. A German – made Retsch RM 100 grinding machine 

has completed the next step in the process. The mineralization of the powder 

of seeds (mpowder = 5 g), thus produced has entailed the chemical treatment 

with 28 ml of acid mixture (HNO3 65%, d=1,4 kg/L (Lach-Ner): HCl 37%, 

d=1,19 kg/L (Merck) = 1 : 3, volume proportion). We have ensured the 

adequate homogeneity of the mixture after a half an hour process and we 

have brought to 100 ml volumetric flask with distilled water the mix that we 

have finally filtered. 


We have conducted our research work and subsequent analyses within 

the framework of the 52144/01. 10. 2008 PNCDI II Project. 

REFERENCES 

1. M. Wierzbicka, J. Obidzińska, Plant Sci., 1998, 137, 155. 

2. A. Piechalak, B. Tomaszewska, D. Baralkiewicz, A. Malecka, Phytochem., 2002, 

60, 153. 

3. K. Stefanov, K. Seizova, N. Yanishlieva, E. Marinova, S. Popov, Food Chem., 

1995, 54, 311. 

4. A. J. M. Baiker, New Phytol., 1987, 106, 93. 

5. W. H. O. Ernst, J. A. C. Verkleij, H. Schat, Acta Bot Neerl, 1992, 41, 229. 

233

234 


6. W. H. O. Ernst, Schwerzmetalpflanzen: Pflanzenokologie und Mineralstoffwechsel, 

1982, 472. 

7. K. B. Searcy, D. C. Mulcahy, Am. J. Bot., 1985, 72, 1695. 

8. S. B. Lane, E. S. Martin, New Phytol., 1977, 79, 281. 

9. A. Woźny, B Zatorska, F. Młodzianowski, Acta. Soc. Bot. Pol., 1982, 51, 345. 

10. R. Mukherji, P. Maitra, Pflanzenphysiol., 1977, 81, 26. 

11. A. Fargašová, Bull. Environ. Contam. Toxicol., 1994, 52, 452. 

12. A. J. M. Baker, P. L. Walker, Bioavailability, 1989, 1, 7. 

13. S. Grzesiuk, K. Kulka, “Pan´stwowe Wydawnictwo Rolnicze i Les´ne”, Warsaw, 

1981, chapter 3. 

14. A. E. Mohamed, M. N. Rashed, A. Mofty, Ecotoxicol. Environ. Safety, 2003, 

55, 251. 

15. C. J. Engleman Jr, W. F. McDiffett, Environmental Pollution, 1996, 94, 67. 

16. S. O. King, C. E. Match, P. L. Brezonik, Environ. Pollut., 1992, 78, 9. 

17. J. W. Moore, “Inorganic Contaminants of Surface Water Research and Monitoring 

Priorities”, Springer-Verlag, New York, USA, 1991. 

18. H. Nishizono, K. Kubota, S. Suzuki, F. Ishii, Plant Cell. Physiol., 1989, 30, 595. 

19. K. Yetilmezsoy, S. Demirel, J. Hazard. Mat., 2008, 153, 1288. 

20. D. Demirel, MSc Thesis, Department of Environmental Science, The Graduate 

School of Natural and Applied Sciences of Harran University, Sanliurfa, 2002. 

21. H. Chen, A. Wang, J. Colloid. Interface Sci., 2007, 307, 309. 

22. S. Sinha, R. Saxena, Chemosphere, 2006, 62, 1340.


CORROSION INHIBITION OF BRONZE BY AMINO ACIDS 

IN AQUEOUS ACIDIC SOLUTIONS 

SIMONA VARVARA a , MARIA POPA a , LIANA MARIA MURESAN b 

ABSTRACT. The electrochemical behavior of bronze in an aerated solution of 

0.2 g/L Na2SO4 + 0.2 g/L NaHCO3 (pH=5), in the absence and in the 

presence of different amino acids was studied using open-circuit potential 

and electrochemical impedance spectroscopy measurements. The investigated 

amino acids were: glutamic acid (Glu), arginine (Arg), histidine (His), methionine 

(Met) and cysteine (Cys). 

The impedance measurements revealed that the bronze interface in the 

presence of the amino acids could be suitably represented by a 3RC equivalent 

electrical circuit. The magnitude of polarization resistance, determined from 

the impedance spectra, and the efficiency of corrosion inhibition were found to 

significantly depend on the structure of amino acids. The protection efficiencies of 

the investigated compounds decrease in the order: Cys > Glu > Met > Arg > His. 

Keywords: bronze, corrosion, amino acids, electrochemical impedance 

spectroscopy 

INTRODUCTION 

One of the most efficient methods for protecting metals from degradation 

is the use of corrosion inhibitors. Various types of organic compounds, 

especially nitrogen, sulphur and oxygen containing substances have been 

widely used as corrosion inhibitors in various aggressive media [1-11]. The 

effectiveness of heterocyclic molecules as corrosion inhibitors is based on 

their ability to adsorb on the metallic surface and to form an organic layer 

which protects the metal from corrosion [5-12]. The adsorption of inhibitors 

takes place through the heteroatoms (nitrogen, oxygen and sulfur), aromatic 

rings or triple bounds. Generally, the inhibition efficiency increases in the 

following order: O < N < S [12]. 

In spite of their effectiveness, most of the heterocyclic compounds 

used as corrosion inhibitors are highly toxic and their replacement by “green” 

inhibitors is desirable [13-14]. 

a 

“1 Decembrie 1918” University, Science Faculty, Str. Nicolae Iorga, Nr. 11-13, RO-510009 

Alba Iulia, Romania, svarvara@uab.ro 

b 

“Babes-Bolyai” University, Faculty of Chemistry and Chemical Engineering, Str. Kogalniceanu, 

Nr. 1, RO-400084 Cluj-Napoca, limur@chem.ubbcluj.ro

236 

SIMONA VARVARA, MARIA POPA, LIANA MARIA MURESAN 

Among the organic compounds tested as environmental friendly 

corrosion inhibitors, amino acids have been reported as promising alternatives 

to the toxic inhibitors. They are innoxious, biodegradable, soluble in aqueous 

media, relatively cheap and easy to produce at high purity [12]. These properties 

would justify the use of amino acids as corrosion inhibitors. 

In the last years, various works focused on the investigation of the 

anticorrosive properties of different amino acids on aluminium [15], steel [16- 

19], vanadium [20], lead [21], and alloys [12] corrosion in various aggressive 

media. 

Inhibition of copper corrosion by amino acids has also aroused interests. 

The investigations revealed that cysteine (Cys) is the most efficient corrosion 

inhibitors of copper in neutral and acidic chloride solutions, due to its physical 

adsorption on the metallic surface via the mercapto- group in its molecular 

structure. 

K. Ismail [13] reported that Cys acts as cathodic-type inhibitor for copper 

corrosion and its maximum inhibition efficiency (84%) was achieved at 

concentrations of about 16 mM in 0.6 M NaCl and at 18 mM in 1 M HCl, 

respectively. The presence of Cu 2+ ions increases the inhibition efficiency of 

Cys up to 90%. The adsorption of Cys on the copper surface in neutral and 

acidic chloride solutions obeys the Langmuir adsorption isotherm. Matos 

et al. [22] found that Cys inhibits the anodic dissolution of copper in sulphuric 

acid media at low polarisation due to formation of the cysteine–Cu (I) 

intermediate. At high overpotentials, Cys has no influence on anodic process. 

Zang et al. [23] studied the inhibiting effect of serine (Ser), threonine 

(Thr) and glutamic acid (Glu) on copper corrosion in an aerated 0.5 M HCl 

solution. They reported that the amino acids act as cathodic inhibitors due to 

their adsorption on the metallic surface through both nitrogen and oxygen atoms, 

which forms a blocking barrier to copper dissolution. The inhibition action of 

the above-mentioned amino acids decrease in the order: Glu > Thr > Ser. 

A limited inhibiting effect on copper corrosion in 0.5 M HCl was also 

noticed in the presence of aspartic acid (Asp), asparagine (Asn), glutamine (Gln) 

and glutamic acid (Glu) [24]. The efficiency of these inhibitors depends on their 

chemical structure and decreases in the following order: Gln > Asn > Glu > Asp. 

The results obtained from the potentiodynamic polarisation indicate that the four 

amino acids are mixed-type inhibitors and their protection efficiency increases 

with increasing their concentration up to 0.1 M. 

Zang et al. [25] also showed that alanine (Ala) and cysteine (Cys) at 

a concentration of 10 -5 M act as anodic inhibitors against copper corrosion in 

0.5 M HCl. The maximum protection efficiency was obtained in the presence 

of Cys (58.7%). The inhibition efficiencies of Ala and Cys were found to be 

higher than the one of the benzotriazole (BTA), which is widely known as the 

most efficient corrosion inhibitor of copper and its alloys.

CORROSION INHIBITION OF BRONZE BY AMINO ACIDS IN AQUEOUS ACIDIC SOLUTIONS 

Methionine (Met) has shown limited inhibiting properties for copper 

corrosion in 0.5 M HCl. The presence of Zn 2+ ions in the corrosive solution 

increases the inhibition efficiency up to 92% [26]. 

Barouni et al. [27] measured the effect of valine (Val), glycine (Gly), 

lysine (Lys), arginine (Arg) and Cys on the copper corrosion in aerated 

nitric acid solution. They reported that some amino acids (Arg, Cys, Lys) 

were able to decrease the dissolution rate, while others, such as, Val and 

Gly actually seems to accelerate the corrosion phenomenon. The inhibition 

efficiencies determined from polarisation measurements vary in the order: 

Val (-15%) < Gly (-4%) < Arg (38%) < Lys (54%) < Cys (61%). 

Furthermore, in a recent paper we have reported the beneficial 

effect exerted by Cys and Ala on bronze corrosion in an aerated electrolyte 

containing Na2SO4 and NaHCO3 at pH=5. The electrochemical investigations 

showed that the two innoxious amino acids have fairly good inhibiting 

properties for bronze corrosion, the best anticorrosive protection being 

obtained in the presence of 0.1 mM Cys (90%) [28]. 

In the present study, four environmentally safe amino acids were 

tested as inhibitors on bronze corrosion in an aerated solution of 0.2 g/L 

Na2SO4 + 0.2 g/L NaHCO3 (pH=5). The compounds examined were: glutamic 

acid (Glu), methionine (Met), histidine (His) and arginine (Arg). Some of our 

previous results obtained using cysteine (Cys) as bronze corrosion inhibitors 

will also be mentioned for comparison sake. 

In order to establish some correlation between the molecular structure 

of the amino acids and their inhibiting efficiency, as well as the optimum 

concentration of each amino acid as bronze corrosion inhibitor, conventional 

electrochemical techniques, such as open-circuit potential (ocp) and 

electrochemical impedance spectroscopy (EIS) measurements were used. 


Open-circuit potential measurements 

The evolution of the open-circuit potential (ocp) for bronze over 60 

minutes immersion in the corrosive solution in the absence and presence of 

different concentration of amino acids is illustrated in Figure 1. 

As it can be seen in Figure 1, the open-circuit potential shows the 

same trend in most of the investigated solutions. Thus, it gets more negative 

with time until it reaches a steady state value. 

Generally, the steady state potential (Ess) was reached within less 

than 30 minutes after the electrode immersion in the corrosive media. 

The value of the steady-state potential of bronze in blank solution 

was + 48.70 mV vs. SCE. 

237

ocp / V vs. SCE 


0.1 

0.0 

-0.1 

Arginine 


-0.2 

0 900 1800 

time / s 

2700 3600 

0.06 

0.05 

0.04 

0.03 

0.02 

0.01 

0.00 

Methionine 

-0.01 

0 900 1800 

time / s 

2700 3600 


-0.2 

0 900 1800 

time / s 

2700 3600 

The value of the shift in the Ess is related to the structure of the amino 

acid and its concentration in the electrolyte. Thus, for Cys and Arg the maximum 

negative shifts occur at concentrations of 1 mM and 10 mM, respectively. 

238 


0.1 

0.0 

-0.1 

0.1 

0.0 

-0.1 

-0.2 

-0.3 

Histidine 

Cysteine 

-0.4 

0 900 1800 

time / s 

2700 3600 

Figure 1. Variation of the open-circuit potential in time for bronze electrode 

after immersion in the corrosive media in the absence and in the presence 

of amino acids at different concentrations of amino acids (mM): 

() 0; (― ―) 0.01; (·····); 0.1; (―·―) 1; (―··―) 10. 

The values of Ess obtained in the presence of different concentrations 

of amino acids are presented in Table 1. 

Table 1. The values of the steady-state potential of bronze in electrolytes 

containing different concentrations of amino acids 

Amino acids 

conc. (mM) 

Glu His 

E ss / mV vs. SCE 

Arg Met Cys 

0.01 +47.11 - +32.71 +40.10 +35.93 

0.1 +45.48 +19.78 +28.54 +22.17 -11.77 

1 +26.72 + 7.56 -73.43 - 5.33 -13.33 

10 - -17.95 -177.51 - -327.60


Addition of Glu, Met and His at concentrations of 1 mM has a less pronounced 

effect on the Ess value. 

Generally, the potential shift can be attributed to the adsorption of the 

amino acids molecules on the active sites and/or the deposition of corrosion 

products on the electrode surface [steel, Pb, carbon steel]. 

A precise categorization of a compound as an anodic or cathodic 

inhibitor requires an Ess displacement of up to 85 mV with respect to the blank 

corrosive solution [16]. The magnitude of the Ess displacements suggests 

that at low concentrations (0.01-0.1 mM) the amino acids simultaneously 

affect the cathodic and anodic reactions, while at higher concentrations (1 and 

10 mM) they predominantly influence the cathodic reduction, probably acting 

as barriers to the diffusion of oxygen molecules from the solution to the 

bronze surface. 

Electrochemical impedance spectroscopy measurements 

In order to provide insight into the characteristics and kinetics of 

bronze corrosion in the presence of amino acids, the electrochemical process 

occurring at the open-circuit potential was examined by electrochemical 

impedance spectroscopy. 

As it is well-know [29], the most important advance of the electrochemical 

impedance spectroscopy is the fact that it enables the fitting of the experimental 

impedance data to theoretical values according to equivalent circuit models 

allowing the understanding of the corrosion inhibition mechanism and the 

suggestion of the suitable electrical model that explains the behaviour of 

the metal under different conditions. 

Nyquist plots collected after 60 minutes immersion of the bronze in 

solutions without and with various concentrations of amino acids are presented 

in Figure 2. 

The impedance spectra obtained in the absence and in the presence 

of amino acids display a capacitive behaviour in the whole frequency domain 

and the low frequency limit of the impedance significantly increases by addition 

of various concentrations of organic compounds in the corrosive solution. 

In order to calculate the numerical values of the parameters that 

describe the electrochemical system and to verify the mechanistic model of 

bronze corrosion in the presence of the amino acids, the impedance spectra 

were appropriately analyzed by fitting the experimental data to the electrical 

equivalent circuits presented in Figure 3. 

We have recently reported and discussed [5, 28] that the impedance 

spectra obtained for bronze corrosion in the absence of any inhibitor could be 

suitable described by two R-C ladder circuits, while in the presence of some 

inhibitors (i.e. Cys and Ala), three capacitive loops, though badly separated 

each other, were necessary for computer fitting of experimental data with 

an electrical equivalent circuit. 

239

- Imaginary part / kΩ.cm 2 


20 

(A) 

10 

240 

0.1 


10m 

10m 

Glu (mM) 

0 

0.01 

0.1 

1 

0 

0 10 20 

Real part / kΩ.cm 

30 40 

2 

1 

0.1k 

18 

12 

Arg (mM) 

0 

0.1 

1 

10 

0.1 

10m 

(C) 

6 

0.1 

10m 

10m 

0 

0 12 24 


36 

2 

1 

0.1k 



9 

6 

3 

1 

1 

10m 

0.1 

10m 

0.1 

10m 

0.1k 

0 

0 

1 

0.1 

6 12 


18 

2 

(B) 

His (mM) 

0 

1 

10 

18 

12 

6 

10m 

0.1 

0 

0 

1 

9 18 


27 36 

2 

Met (mM) 

0 

(D) 

0.01 

0.1 

10m 

1 

0.1 

1 

0.1k 

Figure 2. Nyquist plots of bronze electrode in 0.2 g/L Na2SO4 + 0.2 g/L NaHCO3 

(pH=5) solution, in the absence and in the presence of different amino acids: 

(A) glutamic acid; (B) histidine; (C) arginine; (D) methionine. The symbol (—+—) 

corresponds to the simulated spectra. Frequencies are expressed in Hz. 

R e 

R f 

R t 

C f, n f 

C d, n d 

R F 

C F, n F 

Figure 3. Equivalent electrical circuit used for computer fitting of experimental data 

In the present work, the (3RC) electrical circuit was also adopted for 

carrying out a non-linear regression calculation of the impedance data obtained 

in the presence of the amino acids (Glu, His, Arg, Met) using a Simplex method. 

The origin of the various variables used in the equivalent circuit from 

Figure 3 were ascribed as follows [5, 28]: Re-electrolyte resistance; Rf - 

resistance representing the ionic leakage through pores of a dielectric thin 

film formed on the surface that is reinforced in the presence of the inhibitors 

and by the ionic conduction through its pore; Cf - capacitance due to the 

dielectric nature of the surface film (corrosion products); Rt - charge transfer 

resistance; Cd - double layer capacitance at the bronze|electrolyte interface;


RF - faradic resistance of the corrosion products layer accumulated at the 

interface; CF - faradic capacitance due to a redox process taking place at 

the electrode surface, probable involving the corrosion products; nd, nf, and nF: 

are coefficients representing the depressed characteristic of the capacitive 

loops in the Nyquist diagrams. 

A capacitive loop was calculated according to the following equation: 

R 

Z = 

(1) 

1+ 

j ⋅ω 

⋅ R ⋅C 

( ) n 

C has the dimension of F cm −2 , and corresponds to the value at the 

frequency of the apex in Nyquist diagram (ωRC = 1) [5]. 

Table 2 summarises the results of the regression calculations with the 

electrical circuits from Figure 3. For comparison, the values of the impedance 

parameters previously obtained in the absence and in the presence of the 

optimum concentration of Cys (0.1 mM) [28] were included in the Table 2, 

as well. 

The fine overlap between the experimental and the calculated data 

(cross symbols) observed in Figure 2 proves that the chosen equivalent 

electrical circuits properly reproduce the experimental data obtained in the 

absence and in the presence of different concentrations of amino acids, 

respectively. 

In presence of oxidation–reduction process at the electrode surface, 

the polarization resistance Rp is the parameter the most closely related to 

the corrosion rate [5]. 

The values of the polarisation resistance, Rp were determined as the 

sum (Rf + Rt + RF) from the resistances values determined by regression 

calculation (Table 2). 

In most cases, the addition of the amino acids in the corrosive 

solution decreases the bronze corrosion rate, as attested by the increase of 

the polarization resistance values. This indicates that amino acids inhibit 

the bronze corrosion process, probably due to their ability to adsorb and to 

form a protective layer on the metallic surface. The highest values of the Rp 

were obtained in the presence of: 0.1 mM Glu; 0.1 mM Arg; 10 mM His and 

0.1 mM Met. 

In order to evaluate the anticorrosive effectiveness of the amino 

acids, their inhibition efficiency (IE) was calculated using the polarization 

resistance values, according to the following equation: 

0 

R p − R p 

IE(%) = 100⋅ 

(2) 

R 

p 

where Rp and R 0 p are the polarisation resistances in electrolytes with and 

without amino acids, respectively. 

241

Amino 

acids 

conc. 

(mM) 

242 


Table 2. The results of non-linear regression of the impedance spectra 

presented in Figure 2 

Re 

(kΩcm 2 ) 

Rf 

(kΩcm 2 ) 

Cf 

(μF/cm 2 ) 

Rt 

(kΩcm 2 ) 

Cd 

(μF/cm 2 ) 

RF 

(kΩcm 2 ) 

CF 

(mF/cm 2 ) 

Rp 

(kΩcm 2 ) 

0 

Glu 

0.46 - - 3.34 36.84 12.66 1.65 16.01 - 

0.01 1.04 0.56 0.13 4.14 62.68 38.22 0.36 42.92 62.70 

0.1 9.84. 3.94 0.14 15.28 85.99 79.23 0.038 98.45 83.74 

1 9.77 0.55 0.18 2.68 108.74 1.63 19.32 4.86 - 

His 

1 0.99 0.16 5.33 10.62 15.06 9.80 5.45 20.58 2.19 

10 

Arg 

1.01 0.34 5.53 17.74 3.05 10.04 3.82 28.12 43.06 

0.01 

0.89 

2.65 

3.28 

4.17 

13.35 

10.49 

1.82 

17.31 

0.1 0.97 5.31 10.73 9.42 61.20 30.10 3.20 44.83 64.28 

1 0.79 0.085 4.21 5.20 19.69 24.00 1.30 29.29 45.34 

10 

Met 

0.82 0.051 2.09 2.21 7.54 3.64 35.75 5.90 - 

0.01 

1.01 

5.71 

9.69 

20.90 

29.87 

15.00 

6.14 

41.61 

IE 

(%) 

7.51 

61.52 

0.1 1.02 3.96 3.83 33.82 1.67 53.89 0.2 91.67 82.54 

1 0.99 3.47 27.50 5.63 57.85 13.68 0.49 22.78 29.73 

Cys 

0.1 

1.08 1.53 0.94 42.55 1.95 126.71 0.031 170.79 90.63 

* Rp=Rf + Rt + RF 

As it can be seen from the last column of table 2, the anticorrosive 

protection offered by the investigated amino acids on bronze is relatively 

weak at low concentrations. As the amino acid concentration increases, 

their inhibition efficiencies increase and reach a maximum value in the 

presence of optimum concentration of the inhibitors. Nevertheless, a further 

increases of the amino acids concentration leads to a decrease of their 

protective effectiveness. This phenomenon could be probably attributed to 

the saturation of the bronze surface with inhibitor molecules at a certain 

concentration [13]. 

As previously mentioned [30], the effectiveness of the corrosion 

organic inhibitors is related to the extent to which they absorb and cover the 

metallic surface. The adsorption process mainly depends on the number of 

adsorption sites in the inhibitors molecule and their charge density, molecular 

size and interaction mode with the metallic surface [24]. 

In general, the amino acid molecule occurs in its protonated form in 

acidic solution according to the following equilibrium [13, 20]:


R CH 

NH2 

Anion form 

( Basic solution) 

COO - OH 

CH 

- H + 

COO - 

R 

+ NH3 Zwitter ion 

( Neutral solution) 

R CH 

+ NH3 Protonated form 

( Acidic solution) 

COOH 

In acidic solutions, the amino acid molecules could be adsorbed on the 

electrodic surface through the nitrogen, the oxygen or sulphur atoms, which 

form a blocking barrier to metallic surface and decrease the corrosion rate [20]. 

In the investigated experimental conditions, the inhibition efficiency 

of the amino acids as bronze corrosion inhibitors decreases in the order: 

Cys > Glu > Met > Arg > His. 

As expected, among the studied amino acids, Cys exhibits the best 

inhibition efficiency compared with the five others, probably because its 

adsorption on bronze as bidentate ligand in which surface coordination is 

taking place through both the amino group and the –S– moiety [5, 13, 17]. 

The relatively good anticorrosive protection of Glu on bronze corrosion 

could be explained if we take into consideration that the molecule has a smaller 

net positive charge of N atom and a more net negative charge of O atoms [23], 

which can contribute to its adsorption on the bronze surface. Consequently, 

the improved inhibition of glutamic acid could be due to the stabilization of 

its adsorption on the metallic surface by the oxygen atoms in its structure. 

In the case of metionine, the electron-donor group (–CH3) attached 

to the S atom could exert a steric hindrance [17] which probably affects the 

adsorption of the organic molecule on bronze surface. Therefore, the inhibition 

efficiency of Met slightly decreases as compared to Glu. 

Although Arg has a radical which contains 3 nitrogen atoms, its 

effectiveness is lower probably due to the existence of a tautomeric and 

steric hindrance at the N atoms [15]. The lower inhibition efficiency of His 

compared to Arg could be due to the fact that His contains a cyclic imidazole 

group which probably give a smaller surface coverage than the straight chain 

structure of the radical in Arg. Moreover, the adsorption of His on the metallic 

surface is probably taking place through only one -N atom in the secondary 

amine group [15]. 

CONCLUSIONS 

Our study reports the effects of several amino acids (glutamic acid, 

arginine, histidine, methionine and cysteine) on bronze corrosion in an aerated 

solution of 0.2 g/L Na2SO4 + 0.2 g/L NaHCO3 at pH 5 using open-circuit 

potential measurements and electrochemical impedance spectroscopy. 

243

244 


The electrochemical measurements showed that all investigated 

amino acids present inhibition properties on bronze corrosion. 

The molecular structure of amino acids significantly influences the 

magnitude of Rp values and, consequently, their inhibition efficiency. 

The order of anticorrosive effectiveness of the inhibitors was Cys > 

Glu > Met > Arg > His. The variation of the inhibition efficiency with the structure 

of the amino acids was interpreted taking into consideration the number of 

adsorption active centres in the molecule, the adsorption mode and the 

molecular size of the compounds. 

In the investigated experimental conditions, the optimum concentration 

of each amino acids was relatively low (0.1 mM), except for the case of His, 

when a concentration of 10 mM was necessary to attain its maximum inhibition 

efficiency. 


Reagents 

The corrosive medium was an aqueous aerated solution of 0.2 g/L 

Na2SO4 + 0.2 g/L NaHCO3, acidified to pH=5 by addition of dilute H2SO4. 

This electrolyte corresponds to an acidic rain in an urban environment. The 

solutions used in this study were prepared using analytical grade reagents 

(Merk, Darmstadt, Germany) and ion-exchanged water. 

The amino acids were dissolved in the electrolyte solution to the 

concentration of 0.01-10 mM. They were purchased from Sigma-Aldrich 

and used as received. 

The amino acids used in the electrochemical investigations are: 

1. Acidic amino acids: glutamic acid. 

2. Basic amino acids: histidine and arginine. 

3. Sulphur-containing amino acids: methionine and cysteine. 

The molecular structures of the amino acids are shown in scheme 1. 

glutamic acid (Glu) arginine (Arg) 

histidine (His) methionine (Met) cysteine (Cys) 

Scheme 1. Molecular structure of the investigated amino acids


Electrochemical measurements 

The investigation of the inhibiting properties of the amino acids on 

bronze corrosion was performed by open-circuit potential measurements 

and electrochemical impedance spectroscopy. 

An electrochemical cell with a three-electrode configuration was used; 

a large platinum grid and a saturated calomel electrode (SCE) were used 

as counter and reference electrodes, respectively. To avoid the electrolyte 

infiltration, the working electrode was made of a bronze cylinder rod which 

was first covered with cataphoretic paint layer (PGG; W742 and P962), and 

cured at 150°C for 15 minutes. Then, the bronze cylinder was embedded into 

an epoxy resin leaving only a circular cross section (0.38 cm 2 ) in contact 

with the corrosive solution. The composition of the working electrode is 

presented in Table 3. 

Table 3. Weight composition (%) of the bronze working electrode 

Cu Sn Pb Zn Sb Ni Fe Mn As S P, Si 

87.975 6.014 4.02 1.172 0.299 0.181 0.11 0.002 0.033 0.19 0.004 

Prior to use, the bronze surface was mechanically polished using 

grit paper of 600 and 1200 and then rinsed thoroughly with distilled water. 

Electrochemical experiments were performed using a PAR model 

2273 potentiostat controlled by a PC computer. 

Electrochemical impedance measurements were carried out at the 

open circuit potential after 60 minutes immersion of the bronze electrode in 

the corrosive medium. The impedance spectra were acquired in the 

frequency range 100 kHz to 10 mHz at 10 points per hertz decade with an 

AC voltage amplitude of ± 10 mV. The impedance data were then analyzed 

with software based on a Simplex parameter regression. 


The financial support from CNCSIS under the project PNCDI II-Idei, 

contract no. 569/2009, CNCSIS code 17 is gratefully acknowledged. 

REFERENCES 

1. H. Otmacic, E. Stupnisek-Lisac, Electrochimica Acta, 2003, 48, 985. 

2. K. Es-Salah, M. Keddam, K. Rahmouni, A. Srhiri, H. Takenouti, Electrochimica 

Acta, 2004, 49, 2771. 

3. A. Dermaj, N. Hajjaji, S. Joiret, K. Rahmouni, A. Srhiri, H. Takenouti, V. Vivier, 

Electrochimica Acta, 2007, 52, 4654. 

4. K. Rahmouni, N. Hajjaji, M. Keddam, A. Srhiri, H. Takenouti, Electrochimica Acta, 

2007, 52 7519. 

245

246 


5. L. Muresan, S. Varvara, E. Stupnisek-Lisac, H. Otmacic, K. Marusic, S. Horvat 

Kurbegovic, L. Robbiola, K. Rahmouni, H. Takenouti, Electrochimica Acta, 2007, 

52, 7770. 

6. H. Ma, S. Chen, L. Niu, S. Zhao, S. Li, D. Li, Journal of Applied Electrochemistry, 

2002, 32, 65. 

7. J.B. Matos, L.P. Pereira, S.M.L. Agostinho, O.E. Barcia, G.G.O. Cordeiro, E. D’ Elia, 

Journal of Electroanalytical Chemistry, 2004, 570, 91. 

8. S.A. Abd El-Maksoud, Electrochimica Acta, 2004, 49, 4205. 

9. K.F. Khaled, Materials Chemistry and Physics, 2008, 112, 104. 

10. D. Zhang, L. Gao, G. Zhou, K.Y. Lee, Journal of Applied Electrochemistry, 2008, 

38, 71. 

11. S.M. Milic, M.M. Antonijevic, Corrosion Science, 2009, 51, 28. 

12. M.A. Kiani, M.F. Mousavi, S. Ghasemi, M. Shamsipur, S.H. Kazemi, Corrosion 

Science, 2008, 50, 1035. 

13. K. M. Ismail, Electrochimica Acta, 2007, 52, 7811. 

14. E. Stupnisek-Lisac, A. Gazivoda, M. Madzarac, Electrochimica Acta, 2002, 47, 4189. 

15. A.A. E.Shafei, M.N.H. Moussa, A.A. ElFar, Journal of Applied Electrochemistry, 

1997, 27 1075. 

16. E. Oguzie, Y. Li, F.H. Wang, Electrochimica Acta, 2007, 53, 909. 

17. M. S. Morad, Journal of Applied Electrochemistry, 2008, 38, 1509. 

18. A.B. Silva, S.M.L. Agostinho, O.E. Barcia, G.G.O. Cordeiro, E.D’Elia, Corrosion 

Science, 2006, 48, 3668. 

19. P. Singh, K. Bhrara, G. Singh, Applied Surface Science, 2008, 254, 5927. 

20. M.M. El-Rabiee, N.H. Helal, Gh.M. A. El-Hafez, W.A. Badawy, Journal of Alloys 

and Compounds, 2008, 459, 466. 

21. N.H. Helal, M.M. El-Rabiee, Gh.M. Abd El-Hafez, W.A. Badawy, Journal of Alloys 

and Compounds, 2008, 456, 372. 

22. J.B. Matos, L.P. Pereira, S.M.L. Agostinho, O.E. Barcia, G.G.O. Cordeiro, E. Delia, 

Journal of Electroanalytical Chemistry, 2004, 570, 91. 

23. D.-Q. Zhang, Q.-R. Cai, L.-X. Gao, K.Y. Lee, Corrosion Science, 2008, 50, 3615. 

24. D.-Q. Zhang, Q.-R. Cai, X.-M. He, L.-X. Gao, G.-D. Zhou, Materials Chemistry 

and Physics, 2008, 112, 353. 

25. D.-Q. Zhang, L.-X. Gao, G.-D. Zhou, Journal of Applied Electrochemistry, 2005 

35, 1081. 

26. D.-Q. Zhang, Q.-R. Cai, X.-M. He, L.-X. Gao, G.S. Kim, Materials Chemistry 

and Physics, 2009, 114, 612. 

27. K. Bauroni, L. Bazzi, R. Saghi, M. Mihit, B. Hammouti, A. Albourine, S. E. Issami, 

Materials Letters, 2008, 62, 3325. 

28. S. Varvara, M. Popa, G. Rustoiu, R. Bostan, L. Muresan, Studia Universitatis Babes- 

Bolyai, Chemia, 2009, in press. 

29. A. J. Bard, L. R. Faulkner, “Electrochemical methods. Fundamentals and Applications” 

(2 nd Edition), John Wiley & Sons Publisher, New York, 2001, chapter 10. 

30. E. Stupnisek-Lisac, A. Gazivoda, M. Madzarac, Electrochimica Acta, 2002, 47, 4189.


POTASSIUM-SELECTIVE ELECTRODE 

BASED ON A CALIX[6]ARENIC ESTER (C6Es6) 

LIDIA VARVARI a , SORIN-AUREL DORNEANU a , 

IONEL CĂTĂLIN POPESCU a * 

ABSTRACT. An ion-selective electrode (ISE) was elaborated based on 

a calyx[6]arenic ester (C6Es6) as ionophore and PVC as polymer matrix. 

Potentiometric measurements were performed in standard solutions of 

Ca 2+ , Mg 2+ , Na + , K + + + + 

, NH4 and Li . The best response was observed for K : 

slope 52 mV/ΔpK; linear range from 0.1 mM up to 0.1 M; detection limit of 

0.02 mM; potentiometric selectivity coefficients of 0.06 and 0.16 for Na + 

+ 

and NH4 , respectively. 

Keywords: calyx[6]arenic ester, PVC-based ISE, potassium ISE 

INTRODUCTION 

Ion-selective electrodes (ISE) are of great importance due to their 

wide applications mainly in clinical, food and environmental chemistry [1]. 

For example, it was estimated that, all over the world, over a billion of clinical 

analyses are performed annually using ISE [2]. 

Figure 1. The structure of C6Es6 ester 

a Department of Physical Chemistry, Babes-Bolyai University, 400028 Cluj-Napoca, Romania, 

*e-mail address: cpopescu@chem.ubbcluj.ro

248 

LIDIA VARVARI, SORIN-AUREL DORNEANU, IONEL CĂTĂLIN POPESCU 

The ion selective membrane of an ISE is usually made of a polymer 

matrix (such as PVC) incorporating an ionophore. Most often, ionophores are 

macrocyclic compounds, which selectively bind different ions by entrapping 

them in their cavity. Consequently, the selectivity of the membrane strongly 

depends on the size match between the ion and the host cavity, but also on 

the ion charge. Many calixarenic compounds have been successfully used 

as ionophores [3- 10]. 

The aim of this paper was to evaluate the ionophore abilities of 

a calyx[6]arenic ester synthesized at ICCRR (Cluj-Napoca) using a new 

chemical route. The structure of this compound, named 4-tert-butylcalix 

[6]arene-hexaacetic acid hexaethyl ester (C6Es6), is shown in Figure 1. The 

study was focused on the determination of the main analytical parameters 

(sensitivity, detection limit, selectivity) of C6Es6-based ISE for various alkaline 

and alkaline-earth cations. 


I. Calibration curves for Ca 2+ , Mg 2+ , Na + , K + , NH4 + and Li + 

As the ionophore properties of the C6Es6, synthesized using a new 

chemical route, were unpredictable, our study started with the evaluation of the 

sensitivity (S) and detection limit (DL) towards several cations of biotechnological 

and medical interest: Ca 2+ , Mg 2+ , Na + , K + , NH4 + and Li + . All measurements were 

performed in “batch”, using the known addition method for the preparation 

of standard solutions. Figure 2 shows the calibration curves obtained in 

presence of K + and Na + , at variable ionic strength. The calculated values of 

the corresponding S and DL parameters are presented in Table 1. All values 

represent the average of two successive measurements, carried out with 

four electrodes, under the same experimental conditions. 

The best S values were obtained for Na + , K + and NH4 + , being quasinernstian 

in the limit of experimental errors. For all other investigated 

cations the S values were significantly under-nernstian. A low interelectrode 

reproducibility was observed for S value calculated in the presence of Ca 2+ . 

DL values around 10 -5 M were observed for K + and NH4 + . All other 

DL values were at least 10 times higher than for K + and NH4 + . 

Table 1. Values of S and DL for C6Es6-based ISE in presence of different cations 

Ion Ca 2+ Mg 2+ Na + K + NH4 + Li + 

S (mV/Δpi) 14-26 12 55 52 56 5 

DL (mM) 2.2*10 -3 5.4*10 -4 1.4*10 -4 2.0*10 -5 8.6*10 -5 1.1*10 -3

E / mV vs. SCE 

POTASSIUM-SELECTIVE ELECTRODE BASED ON A CALIX[6]ARENIC ESTER (C6ES6) 


290 

280 

270 

300 

200 

100 

electrode 1 



-8 -6 -4 -2 0 

log aK + 

-4 -2 0 

log aNa + 

Figure 3 presents an example of potentiometric response obtained 

from the interference study carried out in presence of variable Na 

249 

+ concentration. 

The average values of the potentiometric selectivity coefficients estimated 

for Na + and NH4 + were 0.06 and 0.16, respectively. 

Obviously, the selectivity study can be extended for different concentrations 

of the primary ion. Taking into account that the selectivity coefficients are a 

complex function on the primary and interfering ion concentration, the level 

of the primary ion concentration should be established in direct correlation 

with the specific application of the investigated ISE. 


300 

200 

100 





Figure 2. Calibration curves recorded for C6Es6-based ISE 

in presence of K + and Na + 




-4 -2 0 

log aNa + 

Figure 3. Potentiometric response 

recorded at C6Es6-based ISE in 

presence of constant concentration of 

K + and variable concentrations of Na + 

II. Study of the ionic interference 

Based on the calibration curves 

obtained for the investigated cations, K + 

was chosen as primary ion (best S and 

DL values). During the whole interference 

study the K + concentration was kept at 

10 -2 M for two main reasons: (i) this 

value corresponds to the middle of the 

linear range on the K + calibration curve; 

(ii) the K + concentration in common 

biologic fluids (blood, plasma) is close 

to this value. A rigorous evaluation of the 

potentio-metric selectivity coefficients 

requi-res a quasi-nernstian response for 

the interfering cations, too. For this reason 

the selectivity study was restricted only 

to Na + and NH4 + ions.

250 


III. Study of repeatability 

Based on experimental results obtained from calibration curves recorded 

in presence of Na + and K + , studies of inter-measurement and inter-electrode 

repeatability were also performed. 

As can be seen from Figure 4, an excellent repeatability was observed 

both between two successive measurements and between four similar 

electrodes. The data dispersion observed in the domain of low concentrations 

could be due to the absence of the supporting electrolyte. 


300 

200 

100 

CONCLUSIONS 

(A) 

-6 -4 -2 0 

log aNa + 

-8 -6 -4 -2 0 

The aim of the present study was to evaluate the main analytical 

parameters of ISE based on a PVC membrane containing the ionophore 

C6Es6, synthesized by a new chemical route. Measurements were performed 

in separate solutions of Ca 2+ , Mg 2+ , Na + , K + , NH4 + and Li + . Quasi-nernstian 

slopes were obtained for K + (52 mV), Na + (55 mV) and NH4 + (56 mV). ). For 

all the other ions, the sensor showed undernernstian sensitivities. The lowest 

detection limit was obtained in the case of K + (2.0*10 -5 M). A very good 

repeatability was observed both for inter-measurement and inter-electrode 

tests. 

Based on these main electroanalytical parameters, it can be concluded 

that the newly prepared calix[6]arenic ester may be used as an ionophore for 

obtaining K + -selective electrodes, good enough to replace the consecrated 

but expensive valinomycin. Further studies are in progress in order to optimize 

the PVC membrane composition. 


300 (B) 

200 

100 

log aK + 

Figure 4. Mean values and standard deviations for: (A) two measurements 

performed successively, using the same electrode for Na + ion; (B) measurements 

performed in parallel, using four electrodes for K + ion



I. Materials 

The C6Es6 ionophore was provided by dr. Elisabeth-Jeanne Popovici 

from “Raluca Ripan” Chemistry Research Institute, Cluj-Napoca (Romania). 

All reagents used were of analytical grade. Calcium chloride, lithium 

acetate, lithium chloride, ammonium chloride, 2-nitrophenyloctylether (NPOE), 

high molecular weight polyvinyl chloride (PVC) and tetrahydrofurane (THF) 

were purchased from Fluka (Darmstadt, Germany). Potassium chloride was 

from Riedel-deHaën (Darmstadt, Germany), magnesium chloride was 

purchased from Chimopar (Bucharest, Romania) and sodium chloride was 

from Merck (Darmstadt, Germany). 

II. Membrane preparation 

The ISE membrane was prepared from 1% (w/w) ionophore (C6Es6), 

33% (w/w) polymer matrix (PVC), and 66% (w/w) plasticizer (NPOE); the mixture 

had a total weight of 0.3 g. The components were successively dissolved in 

THF, under stirring, in the following order: ionophore, polymeric matrix, and 

plasticizer. After complete dissolution, the mixture was poured into a glass 

cylinder and covered by a glass recipient, under which a THF-impregnated 

paper was placed, in order to avoid pores formation. When dried, the 

membrane was placed for one day in a dark place, in open air. 

III. Experimental setup 

All the measurements were performed using a PC-controlled setup 

[11]; its scheme is presented in Figure 5. This device is composed of: the 

electrochemical cell; a personal computer; two interfaces for communication 

between the computer and other components; an unit for controlling the 

solution stirring; a magnetic stirrer; a peristaltic pump allowing automatic 

exponential additions of standard solution. 

The system control as well as data acquisition were performed 

using the LabView 5.1 software. Data processing was done by using the 

Origin 5.0 software. 

EIS were prepared by fixing an 8 mm diameter disc membrane at 

the bottom end of a plastic syringe body. As internal reference, a Ag/AgCl 

system was used. The inner electrolytes contained the same ion as the test 

solution, at a concentration of 5 mM. Four similar electrodes were tested 

in parallel, and each measurement was repeated three times under same 

experimental conditions. A double-junction saturated calomel electrode was 

used as external reference. The external liquid junction was filled with 

CH3COOLi 0.1 M. 

251

252 


PCI – 6024 E 

interface 

PC – AO – 2DC 

interface 

Figure 5. The scheme of the computer-controlled device 

used for the ISE potentiometric measurements; 

MS – magnetic stirrer; P – peristaltic pump 

IV. Experimental procedure 

Before use, all electrodes were conditioned for at least 24 hours in 

the solution containing the cation to be determined. 

The experimental procedure consisted in two main parts: in the first 

one, the potentiometric response of the prepared membranes was recorded 

for Ca 2+ , Mg 2+ , Na + , K + , NH4 + and Li + using separate solutions, and the 

corresponding calibration curves were recorded. In the second part, the 

ionic interference between K + and different common cations was examined. 

All potentiometric measurements were performed in batch mode, using the 

standard addition method for the preparation of standard solutions. 

For the interference study, the method of fixed primary ion concentration 

was used: the primary ion concentration was kept constant, while the 

concentrations of the interfering ions were increased. 


Electrode 1 




Ref. electrode 

Electrochemical 

interface 

Controll 

unit 

Electrochemical 

cell 

The authors are grateful to dr. Elisabeth-Jeanne Popovici from the 

“Raluca Ripan” Chemistry Research Institute for providing the compound 

C6Es6 and to ANCS for the financial support. 

MS 

P


REFERENCES 

1. E. Bakker, D. Diamond, A. Lewenstam, E. Pretsch, Anal. Chim. Acta, 1999, 

393, 11. 

2. E. Bakker, P. Buhlmann, E. Pretsch, Chem. Rev., 1997, 97, 3083. 

3. P. Buhlmann, E. Pretsch, E. Bakker, Chem. Rev., 1998, 98, 1593. 

4. V. Arora, H. M. Chawla, S. P. Singh, ARKIVOC, 2007, 2, 172. 

5. R. Ludwig, N. T. K. Dzung, Sensors, 2002, 2, 397. 

6. K. Belhamel, R. Ludwig, M. Benamor, Microchimica Acta, 2005, 149, 145. 

7. T. D. Chung, H. Kim, Journal of Inclusion Phenomena and Molecular 

Recognition in Chemistry, 1998, 32, 179. 

8. V. S. Bhat, V. S. Ijeri, A. K. Srivastava, Sensors and Actuators B, 2004, 99, 98. 

9. V.K. Gupta, R. N. Goyal, M. Al Khayat, P. Kumarc, N. Bachheti, Talanta, 2006, 

69, 1149. 

10. A.K. Jain, V.K. Gupta, L. P. Singh, J. R. Raisoni, Electrochimica Acta, 2006, 

51, 2547. 

11. S. A. Dorneanu, V. Coman, I. C. Popescu, P. Fabry, Sensors and Actuators B, 

2005, 105, 521. 

253


COMPARATIVE STUDY OF CARBON PASTE ELECTRODES 

MODIFIED WITH METHYLENE BLUE AND METHYLENE 

GREEN ADSORBED ON ZEOLITE AS AMPEROMETRIC 

SENSORS FOR H2O2 DETECTION 

CODRUTA VARODI a , DELIA GLIGOR b , LEVENTE ABODI a , 

LIANA MARIA MURESAN a * 

ABSTRACT. Methylene Blue (MB) and Methylene Green (MG) adsorbed on a 

synthetic zeolite (13X) and incorporated in carbon paste resulted in modified 

electrodes (MB-13X-CPEs and MG-13X-CPEs) with electrocatalytic activity for 

H2O2 reduction. Using the treatment proposed by Laviron, the heterogeneous 

electron transfer rate constant (ks) was estimated, and values between 2.9 and 

6.6 s -1 were found. The electrocatalytic ability of the modified electrodes 

towards H2O2 electroreduction depends mainly on the formal potential of 

mediator. The MB-13X-CPEs and MG-13X-CPEs respond to H2O2 in the linear 

range from ~10 -4 to 10 -1 M. The best performances were attained with 

MB-13X-CPEs in phosphate buffer (pH 6): detection limit 0.13 mM, at a signal 

to noise ratio equal to 3; sensitivity 0.94 mA/M, calculated as the ratio Imax/KM. 

Keywords: carbon paste electrodes; Methylene Blue; Methylene Green; 

Zeolite; Hydrogen peroxide. 

INTRODUCTION 

Zeolites ability to immobilize large amounts of redox dyes, with significant 

solubility in specific experimental conditions recommend them as attractive 

matrices for the preparation of composite electrode materials [1-4]. Methylene 

Blue (MB) and Methylene Green (MG) are water-soluble cationic dyes with 

the redox potential close to the optimal potential window for amperometric 

detection of ascorbic acid, NADH or H2O2. Electrodes incorporating zeolites 

modified with MB and MG were realized for amperometric detection of ascorbic 

acid [2,5,6], NADH [5,7] or H2O2 [8-11]. 

Continuing our preoccupation in obtaining modified electrodes by 

immobilization of different dyes onto zeolites [6,7,9-11], in the present paper, 

the electrochemical behavior and the electrocatalytic properties of carbon paste 

a Facultatea de Chimie şi Inginerie Chimică, Str. Kogălniceanu, Nr. 1, RO-400084 Cluj-Napoca 

b Facultatea de Ştiinţa Mediului, Universitatea Babeş-Bolyai, Str. Kogălniceanu, Nr. 1, RO- 

400084 Cluj-Napoca, Romania, limur@chem.ubbcluj.ro

256 

CODRUTA VARODI, DELIA GLIGOR, LEVENTE ABODI, LIANA MARIA MURESAN 

electrodes incorporating a synthetic zeolite (molecular sieves 13X, Aldrich) 

modified with Methylene Blue (MB-13X-CPEs) or Methylene Green (MG-13X- 

CPEs) are assessed comparatively and the influence of some experimental 

parameters (pH, and potential scan rate) on the voltammetric response of 

these electrodes were investigated. 

The electrochemical parameters for the heterogeneous electron 

transfer process corresponding to the surface immobilized mediators were 

determined from voltammetric measurements. All observed differences 

were used to put on evidence the influence of the mediator structure on the 

redox response. The modified electrodes were tested for electrocatalytic 

mediated reduction of H2O2 at MB-13X-CPES and MG-13X-CPEs, using 

cyclic voltammetry (CV) and the calibration of modified electrodes were 

realized by using amperometry. 


Cyclic voltammetric experiments were carried out using MB-13X-CPEs 

and MG-13X-CPEs. In Figures 1A and 1B, a pair of well-defined redox waves in 

the case of both electrodes, can be observed. This peak pair was assigned 

to the oxidation and reduction of MB and MG adsorbed on the 13X zeolite. 

I / μA 

50 A 

0 

-50 

-500 0 

E / mV vs. Ag/AgCl/KClsat 

-400 0 400 

Figure 1. Cyclic voltammograms for MB-13X-CPEs (A) and MG-13X-CPEs (B). 

Experimental conditions: starting potential, -650 mV vs. Ag|AgCl/KClsat (A) 

and -400 mV vs. SCE (B); potential scan rate, 10 mV s -1 ; 

supporting electrolyte, 0.1 M phosphate buffer, pH 7. 

E 0’ value for MG-13X-CPEs is more positive (see Table 1) than the E 0’ 

value for MB-13X-CPEs, suggesting that MG is more difficult to oxidize than 

MB. This behavior could be explained by the presence of a nitro-group at the 

4-position, in the MG structure. The nitro-group is a strong electron acceptor 

group and it enhances the stability of the compound, making its oxidation 

more difficult. 

I / μA 

0 

-10 


B

COMPARATIVE STUDY OF CARBON PASTE ELECTRODES MODIFIED WITH METHYLENE BLUE… 

Table 1. Electrochemical parameters corresponding to MB-13X-CPEs and 

MG-13X-CPEs modified electrodes. Experimental conditions: as in Figure 1. 

* * 0’* 

Electrode Epa Epc E ΔEp Γ (mol cm -2 ) 

MB-13X-CPEs -130 -220 -175 90 1.40·10 -8 

MG-13X-CPEs -35 -81 -58 46 1.03·10 -8 

* mV vs. Ag|AgCl/KClsat 

The peak separation ΔEp (ΔEp = Epa – Epc) for MB-13X-CPEs and 

MG-13X-CPEs was found to be 90 mV and 46 mV, respectively, in phosphate 

buffer, pH 7 (10 mV s -1 ), indicating a quasi-reversible redox process. The peaks 

split are larger than that reported for MG adsorbed on graphite (~ 20 mV) [12], 

indicating a diffusion behavior of the mediator. 

As can be observed from Figure 2, for MG-13X-CPEs, the slope of 

E 0’ vs. pH dependence (pH 4-7) was 0.034 V/pH, indicating a e - /H + ratio 

equal to 2 corresponding to the reaction: 

MGH → MG + + 2e - + H + . 

The E 0’ vs. pH dependence could be mainly explained by the steric 

impediments which hinder the entrance of the MG molecules in the 

channels and pores of the zeolite and make them sensitive to the external 

solution pH change. 

Contrarily, for MB-13X-CPEs, E 0’ does not change significantly with 

pH. This result is remarkable and can be due to the absence of NO2 group 

from MB structure. Due to its smaller size, MB is entrapped in the holes of 

the 13X type zeolite (host matrix) where it is strongly hold by electrostatic 

interactions being not affected by the external solution pH change [11]. 

E 0' / mV vs. Ag/AgCl/KClsat 

400 

200 

0 

-200 

MB-Z-CPE 

MG-Z-CPE 

-400 

0 2 4 6 8 10 

Figure 2. Variation of E 0’ with pH for MB-13X-CPEs and MG-13X-CPEs. Experimental 

conditions: potential scan rate, 50 mV s -1 ; for other conditions, see Figure 1. 

pH 

257


Moreover, the slope of the log I -log v dependence (see Table 2) 

confirms that MB is stronger adsorbed than MG on the zeolite surface. 

Table 2. Parameters of the log-log linear regression corresponding to the peak 

current dependence on the potential scan rate for MB-13X-CPEs and 

MG-13X-CPEs. Experimental conditions: as in Figure 3. 

Electrode pH 

Slope R / N 

oxidation reduction oxidation reduction 

MB-13X-CPEs 3 0.80 0.92 0.992 / 12 0.996 / 12 

5 0.81 0.95 0.999 / 12 0.997 / 12 

6 0.82 0.96 0.999 / 12 0.997 / 12 

7 0.82 0.95 0.998 / 12 0.995 / 12 

8 0.78 1.01 0.999 / 8 0.998 / 8 

9 0.81 0.96 0.999 / 12 0.998 / 12 

MG-13X-CPEs 5 0.67 0.64 0.996 / 11 0.996 / 11 

6 0.62 0.73 0.997 / 12 0.984 / 9 

7 0.57 0.60 0.998 / 10 0.993 / 8 

Using the treatment proposed by Laviron [13], the heterogeneous 

electron transfer rate constant (ks) and the transfer coefficients (α) were 

determined (Figure 3 and Table 3). At pH values greater than 5, MB-13X-CPEs 

present relatively higher ks values than MG-13X-CPEs, but the differences are 

not important. The ks values in the case of MB-13X-CPEs and MG-13X-CPEs 

are in accordance with the better adsorption of MB and MG on the zeolite 

and suggest a faster electron transfer rate. It should be mentioned that the 

values of ks calculated for MG-13X-CPEs are higher than those obtained using a 

NaX type synthetic zeolite from Bayer [10], suggesting a stronger interaction 

between MG and 13X type zeolite than between MG and NaX type zeolite. 

This confirms that the type of adsorbent is very important for the modified 

electrode electrochemical behavior. 

The differences occurring between the transfer coefficients values, 

α, corresponding to the two electrodes suggest a different electron transfer 

mechanism in the two cases and a possible change of the reaction path 

with increasing pH. Probably, in the case of MB-13X-CPEs the charges are 

transferred mainly via a mechanism involving an ion exchange process 

between the immobilized positively charged MB species and the supporting 

electrolyte cations, followed by an electron transfer process occurring between 

the free MB species and the graphite particles from the carbon paste, while 

in the case of a MG-13X-CPEs the charge transfer mechanism is partially 

based on a surface-mediated electron transfer. Moreover, the α values indicate 

that the redox processes are not fully reversible. 

258


(E p - E 0' ) / mV vs. Ag/AgCl/KCl sat 

0 

-200 

-400 

-600 

A 

pH 3 

pH 7 

pH 9 

0.01 0.1 1 

log (v / V s -1 ) 

(Ep - E 0' ) / mV vs. SCE 

600 

400 

200 

0 

-200 

-400 

B 

pH 2 

pH 5 

0.01 0.1 1 

log (v / V s -1 ) 

Figure 3. Experimental dependence of (Ep - E o’ ) vs. logarithm of the scan rate for 

MB-13X-CPEs (A) and MG-13X-CPEs (B). Experimental conditions: as in Figure 1. 

Table 3. Kinetic parameters for the heterogeneous electron transfer at 

MB-13X-CPEs and MG-13X-CPEs modified electrodes. 

Experimental conditions: as in Figure 3. 

Electrode pH ks (s -1 ) α 

MB-13X-CPEs 

MG-13X-CPEs 

R/N 

oxidation reduction 

3 5.6 0.55 0.998 / 5 0.999 / 5 

5 5.4 0.61 0.998 / 5 0.998 / 6 

7 5.5 0.69 0.992 / 5 0.993 / 6 

8 5.5 0.72 0.996 / 5 0.999 / 5 

9 4.9 0.79 0.989 / 5 0.999 / 5 

3 6.6 0.82 0.986 / 5 0.978 / 7 

5 3.5 0.90 0.991 / 5 0.990 / 7 

6 2.9 0.80 0.993 / 6 0.986 / 8 

Hydrogen peroxide electroreduction studies were performed by 

cyclic voltammetry on MB-13X-CPEs and MG-13X-CPEs immersed in H2O2 

solutions of different concentrations (phosphate buffer, pH 7) (see Figure 4 

for MB-13X-CPEs). In the presence of H2O2 an enhancement of the cathodic 

currents and a small peak potential shift towards negative direction with the 

increase of H2O2 concentration were observed. The electrocatalytic efficiency, 

estimated as the ratio, at an applied potential of -400 mV vs. Ag|AgCl/KClsat, in 

phosphate buffer pH 7, was 0.55 for MG-13X-CPEs and 0.10 for MB-13X-CPEs. 

( Ipeak 

) [ H2O2 

] = 3mM 

− 

( Ipeak) 

[ H2O2 

] = 0 

( Ipeak) 

[ H2O2 

] = 0 

259

260 


The slight difference between the electrocatalytic activity of the two 

electrodes suggests that the acceptor group -NO2 existing in the MG molecule 

enhances the electrocatalytic efficiency of this mediator. 

Batch amperometric measurements at constant applied potential 

(-400 mV vs. Ag|AgCl/KClsat) proved that both MB-13X-CPEs and MG-13X-CPEs 

work well as H2O2 amperometric sensors (Figure 5), as suggested by the 

analytical parameters presented in Table 4. 

I / μA 

50 

0 

-50 

-100 

A 

-500 0 

buffer 

1 mM 

3 mM 

5 mM 

E / mV vs. Ag/AgCl/KCl sat 

Figure 4. Cyclic voltammograms obtained 

at MB-13X-CPEs, in the absence and in 

the presence of H 2O 2. Experimental 

conditions: potential scan rate, 

10 mV s -1 ; starting potential, -800 mV 

vs. Ag|AgCl/KCl sat; supporting electrolyte, 

0.1 M phosphate buffer, pH 7.0. 

I / μA 

160 

80 

MB-Z-CPEs 

MG-Z-CPEs 

0 

0 20 40 60 

[H 2 O 2 ] / mM 

Figure 5. Calibration curves for H2O 2 at MB- 

Z-CPEs and MG-13X-CPEs. Experimental 

conditions: applied potential, -400 mV 

vs. Ag|AgCl/KCl sat; supporting electrolyte, 

0.1 M phosphate buffer, pH 6.0. 

Table 4. Electroanalytical parameters corresponding to MB-13X-CPEs and 

MG-13X-CPEs modified electrodes. Experimental conditions: as in Figure 5. 

Electrode pH Detection limit Linear domain Sensitivity* R / N 

(mM) 

(M) (mA/M) 

6 0.13 10 -4 – 3⋅10 -1 MB-13X-CPEs 

1.20 0.991 / 8 

7 0.79 8⋅10 -4 - 10 -1 1.80 0.986 / 15 

6 0.42 4⋅10 -4 – 2⋅10 -1 MG-13X-CPEs 

3.90 0.999 / 17 

7 0.60 6⋅10 -4 - 1 2.00 0.995 / 20 

*calculated as the slope of calibration curve 

The best detection limit (for a signal to noise ratio of 3) was obtained for 

MB-13X-CPEs in phosphate buffer pH 6 and that obtained for MG-13X-CPEs is 

better than that already reported by using a natural zeolitic volcanic tuff (natural 

X type mesoporous clinoptilolyte) [10]. In all cases the response time < 20 s.


The higher sensitivity of MG-13X-CPEs as compared with that of 

MB-13X-CPEs for H2O2 detection is in accordance with the better electrocatalytic 

effect of the MG-13X-CPEs, even if they exhibit smaller ks values than 

MB-13X-CPEs. 

CONCLUSIONS 

Modified electrodes with electrocatalytic activity toward H2O2 reduction 

were obtained by Methylene Blue and Methylene Green adsorption on a synthetic 

zeolite (molecular sieves 13X, Aldrich), followed by their incorporation in carbon 

paste. The characteristics of the voltammetric response of MB-13X-CPEs and 

MG-13X-CPES (ΔEp of 90 and 46 mV, respectively and Ipa/Ipc of ~ 1) pointed 

out to a quasi-reversible, surface confined redox process in both cases. 

The observed differences between the electrochemical behavior of 

MB-13X-CPEs and MG-13X-CPEs (the standard formal potentials, the effect 

of pH on E 0 ’ and the magnitude of the rate constants for heterogeneous 

electron transfer) as well as the electrocatalytic activity for H2O2 reduction 

can be explained in terms of mediator structure. The MG-13X-CPEs present a 

better sensitivity and electrocatalytic efficiency, but its E 0 ’ is not pH independent 

as in the case of MB-13X-CPEs. 

Concluding, both investigated modified electrodes showed moderate 

catalytic efficiency towards H2O2 reduction and a relatively low limit of detection 

(0.13 mM for MB-13X-CPEs and 0.42 mM for MG-13X-CPEs, pH 6). 


Chemicals 

Methylene Blue (MB) and Methylene Green (MG) (Schemes 1 and 2), 

graphite powder and paraffin oil were purchased from Fluka (Buchs, Switzerland). 

The 13X type zeolite, 1Na2O:1Al2O3:2.8±0.2 SiO2 × H2O (particle size, 

3-5μ; pore diameter, 10 Å; specific surface area 548.69 m 2 /g; bulk density 

480.55 kg/m 3 ; Si/Al ratio 1.5) was purchased from Aldrich (Germany). 

Hydrogen peroxide, K2HPO4⋅2H2O and KH2PO4⋅H2O were purchased 

from Merck (Darmstadt, Germany). All other reagents were of analytical grade 

and used as received. 

The supporting electrolyte was a 0.1 M phosphate buffer solution. The 

pH was adjusted in the interval 1-9 using appropriate H3PO4 or NaOH solutions. 

N 

(CH3) 2N 

S 

Cl - 

Scheme 1 

N(CH 3) 2 

261

262 


(CH3)2N 

N 

S 

+ 

Cl - 

Scheme 2 

NO2 

N(CH3)2 

Electrode preparation 

50 ml of a 0.001 % (w/v) MB or MG solution in water were shaken 

(3 days) with 50 mg zeolite. The modified zeolite was filtered, washed and 

dried. 25 mg of the modified zeolite were mixed with 25 mg graphite powder 

and 10 µl paraffin oil in order to obtain the modified carbon paste electrodes 

(MB-13X-CPEs and MG-13X-CPEs). 

The preparation of MB-13X-CPEs and MG-13X-CPEs was reproducible 

when the experimental conditions and variables were maintained constant 

during the preparation period. The current response of the electrodes did not 

change significantly by storing them in air for several months. 

Electrochemical measurements 

Electrochemical experiments were carried out using a typical threeelectrode 

electrochemical cell. The modified carbon paste electrode was used 

as working electrode, a platinum ring as counter electrode and an Ag|AgCl/KClsat 

or SCE as reference electrodes. 

Cyclic voltammetry experiments were performed on a PC-controlled 

electrochemical analyzer (Autolab-PGSTAT 10, EcoChemie, Utrecht, The 

Netherlands). 

Batch amperometric measurements at different H2O2 concentrations 

were carried out at an applied potential of -400 mV vs. Ag|AgCl/KClsat, under 

magnetic stirring, using 0.1 M phosphate buffer solution (pH 7) as supporting 

electrolyte. The current-time data were collected using the above-mentioned 

electrochemical analyzer. 

For each electrode, the surface coverage (Γ, mol cm -2 ) was estimated 

from the under peak areas, recorded during the cyclic voltammetry (CV) 

measurements at low potential scan rate (v ≤ 10 mV s -1 ) [14] and considering 

the number of transferred electrons equal to 2 [15,16]. 

The experimental results are the average of at least 3 identically 

prepared electrodes, if not otherwise mentioned. 


Financial support from CNCSIS (Project ID_512) is gratefully 

acknowledged.


REFERENCES 

1. A. Walcarius, Anal. Chim. Acta, 1999, 384, 1. 

2. M. Arvand, Sh. Sohrabnezhad, M. F. Mousavi, Anal. Chim. Acta, 2003, 491, 193. 

3. A. Walcarius, Electroanalysis, 2008, 20, 711. 

4. I. Svancara, K. Vytras, K. Kalcher, A. Walcarius, J. Wang, Electroanalysis, 2009, 

21, 7. 

5. J. Kulys, G. Gleixner, W. Schuhmann, H.-L. Schmidt, Electroanalysis, 1993, 5, 201. 

6. C. Varodi, D. Gligor, L. M. Muresan, Stud. Univ.Babes-Bolyai, Chemia, 2007, 1, 109. 

7. C. Varodi, D. Gligor, A. Maicaneanu, L. M. Muresan, Rev. Chim., 2007, 58, 890. 

8. H. Yao, N. Li, S. Xu, J.-Z. Xu, J.-J. Zhu, H.-Y. Chen, Biosen. Bioelectron., 2005, 

21, 372. 

9. D. Gligor, L. Muresan, I.C. Popescu, Acta Univ. Cibiniensis, Seria F Chemia, 2004, 

7, 29. 

10. D. Gligor, L. M. Muresan, A. Dumitru, I. C. Popescu, J. Appl. Electrochem., 2007, 

37, 261. 

11. C. Varodi, D. Gligor, L. M. Muresan, Rev. Roum. Chim., 2007, 52, 81. 

12. Q. Chi, S. Dong, Anal. Chim. Acta, 1994, 285, 125. 

13. E. Laviron, J. Electroanal. Chem., 1979, 101, 19. 

14. H. Huck, Phys. Chem. Chem. Phys., 1999, 1, 855. 

15. C. Lei, Z. Zhang, H. Liu, J. Kong, J. Deng, Anal. Chim. Acta, 1996, 332, 73. 

16. B. Liu, Z. Liu, D. Chen, J. Kong, J. Deng, Fresenius J. Anal. Chem., 2000, 367, 

539. 

263


PHARMACOKINETIC INTERACTION BETWEEN IVABRADINE 

AND CIPROFLOXACINE IN HEALTHY VOLUNTEERS 

LAURIAN VLASE a , DANA MUNTEAN a, b , ADINA POPA a , MARIA NEAG a , 

IOAN BÂLDEA b , MARCELA ACHIM a , SORIN E. LEUCUŢA a 

ABSTRACT. The pharmacokinetic interaction of ivabradine with ciprofloxacin 

in healthy volunteers was evaluated. A dose of 5 mg ivabradine in combination 

with 500 mg ciprofloxacin was administered to 18 healthy male volunteers in a 

two treatment study design, separated by 6 days in which the ciprofloxacin 

alone was administrated as a single p.o. dose daily. Plasma concentrations 

of ivabradine were determined during a 12 hours period following drug 

administration. Ivabradine plasma concentrations were determined by a validated 

LC/MS method. Pharmacokinetic parameters of ivabradine were calculated 

using non-compartmental and compartmental analysis. In the two periods of 

treatments, the mean peak plasma concentrations (Cmax) were 8.52 ng/ml 

(ivabradine alone) and 8.40 ng/ml (ivabradine after pre-treatment with 

ciprofloxacin). The times taken to reach Cmax, tmax, were 0.86 hr and 1.52 hr 

respectively and the total areas under the curve (AUC0-∞) were 27.9 ng.hr/ml and 

28.1 ng.hr/ml, respectively. The absorption rate constants (ka) of ivabradine 

were 10.8 hr -1 and 5.27 hr -1 , respectively. Statistically significant differences 

have been observed for both tmax and ka of ivabradine when administered alone or 

with ciprofloxacin, whereas for Cmax and AUC0-∞ the differences were nonsignificant. 

The experimental data demonstrate the pharmacokinetic interaction 

between ciprofloxacin and ivabradine, however, the observed interaction may 

not be clinically significant. 

Keywords: ivabradine, ciprofloxacin, pharmacokinetics, drug interaction 

INTRODUCTION 

Ivabradine (3-(3-{[((7S)-3,4-dimethoxi-bicyclo[4.2.0]octa-1,3,5-trien-7-yl) 

methyl]methylamino}propyl)-1,3,4,5-tetrahydro-7,8-dimethoxy-2H-3-benzazepin- 

2-one, hydrochloride) (Figure 1a) is a novel heart rate–lowering agent that 

selectively and specifically inhibits the depolarizing cardiac pacemaker If current 

in the sinus node. Its activity provides pure heart rate reduction at rest and 

a 

University of Medicine and Pharmacy “Iuliu Haţieganu”, Faculty of Pharmacy, Emil Isac 13, 

RO-400023, Cluj-Napoca, Romania, vlaselaur@yahoo.com 

b 

“Babeş-Bolyai” University, Faculty of Chemistry and Chemical Engineering, Arany Janos 

11, RO-400028, Cluj-Napoca, Romania

L. VLASE, D. MUNTEAN, A. POPA, M. NEAG, I. BÂLDEA, M. ACHIM, S. E. LEUCUŢA 

during exercise, which improves myocardial oxygen balance and increases 

coronary perfusion, without any relevant influence on conduction, contractility, 

ventricular repolarization or blood pressure. The anti-ischemic efficacy and the 

safety of ivabradine have been demonstrated in patients with stable angina 

pectoris [1-5]. 

Despite its therapeutical benefit, ivabradine has some important 

side effects, including bradycardia, AV block, ventricular extrasystoles and 

luminous phenomena [1,3,5]. Due to high potential of ivabradine to give adverse 

reactions on overdosing, but also lack of therapeutic effect on under-dosing, is 

important to know the way in some other substances modify the ivabradine 

pharmacokinetics. 

After oral administration, the metabolic clearance of ivabradine accounts 

for about 80% of its total clearance, with the other 20% corresponding to 

renal clearance. Only CYP3A4 is involved in invabradine’s metabolism, so 

numerous potential interactions can therefore arise with CYP3A4 inhibitors 

and inducers [6-8]. 

Ciprofloxacin (1-cyclopropyl- 6-fluoro- 4-oxo- 7-piperazin- 1-yl- quinoline- 

3-carboxylic acid) (Figure 1b) is a fluoroquinolone used for treating severe 

bacterial infections. Ciprofloxacin is a potent inhibitor of CYP1A2 and medium 

inhibitor of CYP3A4 [9-11], thus it can interfere with metabolism of ivabradine. 

The aim of our study is to determine if a potentially harmful 

pharmacokinetic interaction occurs between ivabradine and ciprofloxacin, 

when administered together. 

H3CO 

H3CO 

N 

O 

N 

CH3 

OCH3 

OCH 3 

(a) (b) 

Figure 1. Molecular structure of ivabradine (a) and ciprofloxacin (b) 


The mean plasma concentrations of ivabradine when administered alone 

or in combination with ciprofloxacin after 5 days treatment with ciprofloxacin 

are shown in Figure 2. The bi-phasic plot describes absorption and elimination 

consecutive processes (see eq. in section Pharmacokinetic analysis). 

The mean pharmacokinetic parameters of ivabradine administered 

alone or in combination with ciprofloxacin, as well as the statistical significance 

following their comparison are given in Table 1. 

266 

HN 

N N 

F 

O 

COOH

PHARMACOKINETIC INTERACTION BETWEEN IVABRADINE AND CIPROFLOXACINE … 


14 

12 

10 

8 

6 

4 

2 

0 

Figure 2. Mean (±SD) plasma levels of ivabradine (5 mg p.o.) given alone 

(continuous line) or in combination with ciprofloxacin (500 mg, p.o.) 

after treatment with ciprofloxacin for 6 days (500 mg p.o.) (dotted line), 

n=18; in insert: semilogaritmic presentation. 

Table1. Pharmacokinetic parameters of ivabradine administered alone or 

after treatment with ciprofloxacin and the result of statistical t test 

used for comparison 

Pharmacokinetic Ivabradine alone Ivabradine + 

p* value, 

parameter (±SD) 

ciprofloxacin t-test 

Cmax (ng/ml) 8.52(4.37) 8.40(4.67) 0.78 

tmax (hr) 0.86(0.41) 1.52(0.52) * 0.0014 

AUC0-∞ (ng.hr/ml) 27.92(13.59) 28.15(13.74) 0.85 

kel (1/hr) 0.35(0.10) 0.37(0.08) 0.36 

ka (1/hr) 10.8(6.8) 5.27(6.71) * 0.025 

t½ (hr) 2.10(0.60) 1.93(0.44) 0.28 

* significance for p

268 


The pharmacokinetic parameters Cmax, tmax and AUC0-∞, were also 

used for bioequivalence evaluation of ivabradine administered in Test and 

Reference period respectively. The parametric 90% confidence interval for 

the ratio Test/Reference period of the mean pharmacokinetic parameters 

Cmax and AUC0-∞ (log transformed) of ivabradine and the significance of the 

difference of tmax are shown in Table 2. 

Table 2. Bioequivalence evaluation of pharmacokinetic parameters of 

ivabradine administered alone or after treatment with ciprofloxacin. 

Pharmacokinetic 

parameter 

90% Confidence 

intervals 

AUC0-∞ (ng.h/ml) 0.93-1.08 (ANOVA, NS) 

Cmax (ng/ml) 0.89-1.07 (ANOVA, NS) 

tmax (hr) χ 2 =3.841 (Friedman, S) 

The 90% confidence intervals for geometric mean of ivabradine in 

Test/Reference individual ratios for Cmax and AUC0-∞ were in the acceptable 

limits of bioequivalence (0.8-1.25). However, the difference between mean 

tmax values of the test and reference formulations was statistically significant. 

The present study shows that the treatment with ciprofloxacin 

influences the pharmacokinetics of ivabradine. No systemic metabolic drugdrug 

interaction was observed, as time the half-life is not changing between 

treatments and the drug exposure (Cmax and AUC0-∞) is about the same. 

However, the ciprofloxacin has a negative effect of absorption rate of 

ivabradine, because the related pharmacokinetic parameters (tmax and ka) 

are significantly changing between treatments. Despite those observed 

differences, since the drug exposure related pharmacokinetic parameters 

(Cmax and AUC0-∞) were in bioequivalence interval, the observed pharmacokinetic 

interaction may not have clinical significance. 

CONCLUSIONS 

A pretreatment with ciprofloxacin until achieving the steady state 

plasma concentrations influences the pharmacokinetics of ivabradine coadministered 

as a single oral dose in healthy volunteers.



Subjects 

Eighteen, non-smoking males, aged 22-27 years took part in the study. 

The study was conducted according to the principles of Declaration of 

Helsinki (1964) and its amendments (Tokyo 1975, Venice 1983, Hong Kong 

1989) and Good Clinical Practice (GCP) rules. The clinical protocol was 

reviewed and approved by the Ethics Committee of the University of Medicine 

and Pharmacy “Iuliu Hatieganu”, Cluj-Napoca, Romania. All volunteers gave 

their written informed consent prior to study inclusion. The volunteers were 

healthy according to history, physical examination and laboratory tests, had 

no history of alcohol or drug abuse and did not take any regular medication. 

Study design 

The study consisted of 2 periods: Period 1 (Reference), when each 

volunteer received a single dose of 5 mg ivabradine and Period 2 (Test), when 

each volunteer received a single dose of 5 mg ivabradine and 500 mg 

ciprofloxacin. Between the two periods, the subjects were treated for 6 days with 

a single daily dose of 500 mg ciprofloxacin. All the drugs were administered in 

the morning, in fasted state. The pharmaceutical products used were Corlentor 

(5 mg tablets, producer Les Laboratoires Servier, France) and Ciprolen (500 mg 

tablets, producer AC Helcor, Romania). Venous blood (5 ml) was drawn into 

heparinized tubes, in the first and in the last day of the study, before drug 

administration as well as at 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, 6, 7, 8, 10 and 12 hours 

after drug administration and the separated plasma was stored frozen (-20°C) 

until analysis. 

Analysis of plasma samples 

Ivabradine plasma concentrations were determined by a validated 

LC/MS method. [12] 

Pharmacokinetic analysis 

The noncompartmental and compartmental pharmacokinetic analysis 

method was employed to determine the pharmacokinetic parameters of 

ivabradine given alone or in combination with ciprofloxacin. The maximal 

plasma concentration (Cmax, ng/ml) and the time to reach the peak concentration 

(tmax, hr) were obtained directly by the visual inspection of each subject’s 

plasma concentration-time profile. The area under the concentration-time 

curve (AUC0-t) has been estimated by integration using trapezoildal rule 

from time zero to the last measurable concentration at time t. The area was 

extrapolated to infinity (AUC0-∞) by addition of Ct/ kel to AUC0-t where Ct is 

the last quantifiable drug concentration and kel is the elimination rate constant. 

269

270 


The elimination rate constant kel was estimated by the least-square regression 

of plasma concentration-time data points lying in the terminal region by using 

semilogarithmic dependence that corresponds to a first-order kinetics. The 

half-life (t½) was calculated as 0.693/kel. The absorption constant rate of 

ivabradine (ka) was calculated using a mono-compartmental pharmacokinetic 

model by fitting the data using the equation: 

Q = D 

k 

a 

ka 

− k 

el 

( e 

−kelt 

− e 

−kat 

where Q is the quantity of ivabradine in the body at time “t” after oral 

administration, D is the doze and ka and kel are the absorption and elimination 

rate constants in hr −1 . 

The pharmacokinetic analysis was performed using Kinetica 4.0.2 

(Thermo Labsystems, U.S.A.) [13]. 

Statistical analysis 

The t-test for paired values was used to compare the calculated 

pharmacokinetic parameters of ivabradine for the two periods. In order to 

evaluate a possible clinical significance of the pharmacokinetic interaction, 

an analysis of variance (ANOVA) was performed on the pharmacokinetic 

parameters Cmax and AUC0-∞ using general linear model procedures, in which 

sources of variation were subject and period. Then the 90% confidence 

intervals of the test/reference period ratios for Cmax and AUC0-∞ (log transformed) 

were determined by the Schuirmann’s two one-sided t test [14]. The 

bioequivalence between ivabradine in Test and Reference period can be 

concluded when the 90% confidence intervals for these pharmacokinetic 

parameters of two products are found within an acceptable range of 0.8- 

1.25 [15-18]. Regarding analysis of tmax, the limit for bioequivalence range 

was expressed as untransformed data, the significance of the difference of 

tmax (Test-Reference) being established by a nonparametric test (Friedman 

test). All the statistical analysis was performed using Kinetica 4.0.2 

software [13]. 


This work was supported by a PNII-IDEI project, code 462, contract 

229/2007 financed by CNCSIS Romania, for which the authors gratefully 

acknowledge. 

)


REFERENCES 

1. U.K. Prasad, D. Gray, H. Purcell, Advances In Therapy, 2009, 26(2), 127. 

2. M.Z. Khawaja, D.M. Walker, International Journal Of Clinical Practice, 2009, 

63(4), 542 

3. J. C. Tardif, P. Ponikowski, T. Kahan, European Heart Journal, 2009, 30(5), 540 

4. P. Milliez, S. Messaoudi, J. Nehme, C. Rodriguez, J. L. Samuel, C. Delcayre, 

American Journal Of Physiology-Heart And Circulatory Physiology, 2009, 

296(2), H435 

5. G. Riccioni, N. Vitulano, N. D'Orazio, Advances In Therapy, 2009, 26(1), 12 

6. A. Portoles, A. Calvo, A. Terleira, L. Laredo, G. Resplandy, C. Gorostiaga, A. 

Moreno, Journal of Clinical Pharmacology, 2006, 46, 1195. 

7. A. Portoles, A. Terleira, A. Calvo, Journal Of Clinical Pharmacology, 2006, 

46(10), 1188 

8. A. Portoles, A. Calvo, A. Terleira, L. Laredo, G. Resplandy, G. Gorostiaga, A. 

Moreno, Journal Of Clinical Pharmacology, 2006, 46(10), 1195 

9. Z. Iqbal, A. Khan, A. Naz, J. A. Khan, G. S. Khan, Clinical Drug Investigation, 

2009, 29(4), 275 

10. H. S. Abou-Auda, A. A. Mustafa, M. S. Al-Humayyd, Biopharmaceutics & Drug 

Disposition, 2008, 29(1), 29 

11. K. Herrlin, M. Segerdahl, L. L. Gustafsson, E. Kalso, Lancet, 2000 356(9247), 

2069. 

12. L. Vlase, D. Muntean, S. E. Leucuta, I. Baldea, Studia Universitatis Babes- 

Bolyai Chemia, 2009, 54(2), 43 

13. http://www.thermo.com/eThermo/CMA/PDFs/Product/productPDF_27347.pdf 

14. D. J. Schuirmann, J Pharmacokinet Biopharm 1987, 15(6), 657 



and Bioequivalence Studies for Orally Administrated Drug Products – General 

Considerations, Rockville, USA, 2003, http://www.fda.gov/cder/guidance/index.htm 

16. The European Agency for the Evaluation of Medicinal Products. Note for 

Guidance on the Investigation of Bioavailability and Bioequivalence, London, 

UK, 2001 (CPMP/EWP/QWP/1401/98). 

17. L. Vlase, A. Leucuta, D. Farcau D, M. Nanulescu, Biopharmaceutics & Drug 

Disposition, 2006, 27(6), 285 

18. L. Vlase, B. Bodiu B, S.E. Leucuta, Arzneimittel-Forschung-Drug Research, 2005, 

55(11), 664 

271

PDF file - Facultatea de Chimie şi Inginerie Chimică

Create successful ePaper yourself

Delete template?

Save as template?