Introduction

Recent years have witnessed a fast development of visible light mediated photoredox catalysis1,2,3,4,5, which can allow a series of thermodynamically inaccessible or kinetically inert reactions to occur. Although remarkable progress has been already made, stoichiometric terminal redox agents are always required and undesired byproducts are generated, both of which complicate the photochemical reactions. On the other hand, electrochemistry provides an intriguing approach to this issue by utilizing electricity as traceless redox agent6,7,8. The simultaneous occurrence of anodic oxidation and cathodic reduction enables a vast variety of mild radical-mediated transformations, obviating the need for stoichiometric electron or proton acceptors. Considering the advantages and common features of photoredox catalysis and electrochemistry, the merging of these two synthetic techniques has significantly enriched the catalytic strategies for the exploration of more challenging chemical transformations9,10,11,12. Pioneering work of photoelectrocatalysis by Xu13,14,15,16,17,18,19, Lambert20,21,22,23,24,25,26,27,28, Lin29,30, Wickens31, Lei32,33, and others34,35,36,37,38,39,40,41,42 has demonstrated the successful combination of photocatalysis and electrochemistry to unlock novel reactivity in organic synthesis. From a mechanistic point of view, photoelectrocatalysis can be generally divided into two types of reaction process: anodic oxidation and cathodic reduction (Fig. 1a), and the two systems of photocatalysis and electrolysis remain completely independent and can be modified at will.

Fig. 1: Photoelectrochemical C(sp3)–H borylation.
figure 1

a Photoelectrochemical SET oxidation and reduction. b Current strategies for C(sp3)-H borylation. c This work.

Full size image

Over the past few decades, C(sp3)−H functionalization of hydrocarbons has emerged as a vibrant research field in organic synthesis43,44,45,46,47,48. Selective functionalization of unreactive C(sp3)−H bonds, which are widespread in abundant and inexpensive chemical feedstock, to produce high value-added chemicals is not only highly rewarding in organic chemistry but also yet to be fully explored. Considering the essential role of organoboron compounds as key building blocks in organic synthesis49,50,51,52,53, the selective installation of boryl substitutions into the structure of hydrocarbons continuously attracts chemists’ attention54. Not surprisingly, the recent emergence of elegant research work on C(sp3)−H borylation reaction has provided new strategies for the access to organoborons55,56,57,58,59,60,61,62,63,64,65,66,67,68. For instance, the Hartwig group developed an undirected C(sp3)−H borylation method enabled by iridium catalyst and ligand63. Aggarwal and co-workers recently demonstrated a metal-free approach for selectively installing a boryl group on C(sp3)−H bond of hydrocarbons via photoinduced hydrogen atom transfer (HAT) approach64. Our group also reported an iron-catalyzed C(sp3)−H borylation protocol enabled by the photoinduced ligand-to-metal charge transfer (LMCT) process (Fig. 1b)65.

Considering the significance of organoboron motifs in chemistry, we envision that the exploration of a general and sustainable platform for undirected C(sp3)−H borylation reaction via photoelectrocatalysis would be highly rewarding. Inspired by Xu’s research work14, the proposed photoelectrochemical method, which involves the generation of key chlorine radicals (Cl·) by anodic oxidation of Cl- as well as the following homolytic photolysis, can potentially provide an ideal platform for C(sp3)−H borylation via radical-mediated HAT process (Fig. 1c)69,70,71,72,73,74,75,76.

In this work, we report an oxidative photoelectrochemical strategy that allows the C(sp3)–H borylation reaction of hydrocarbons to smoothly occur. This synthetic protocol is applicable not only to simple alkanes, but also to a wide range of hydrocarbons including halides, silanes, ketones, esters, and nitriles. Importantly, unconventional regioselectivity of C(sp3)−H borylation is herein achieved, with the occurrence of C(sp3)−H borylation preferentially at distal methyl position.

Results

Optimization studies

Our study began by the investigation of the reaction between cyclohexane (1) and bis(catecholato)diboron (B2(cat)2, 2), in a simple undivided cell equipped with a graphite rod as anode and a platinum plate as cathode (Table 1). After an extensive evaluation of the reaction conditions, the corresponding borylated product (3) was afforded in 72% yield (entry 1), employing tetraethylammonium chloride as electrolyte and chlorine source, HCl as the additive in acetonitrile under 390 nm purple LEDs irradiation as well as 2 mA constant current electrolysis for 12 h at room temperature. As shown in entries 2–4, unsatisfying results were obtained in the absence of HCl, and the use of FeCl3 or CuCl2 instead of HCl showed little reactivity. Moreover, replacing the platinum plate cathode with graphite rod cathode, or replacing the graphite rod anode with a platinum plate anode led to relatively lower efficiency (entries 5–6). The experimental results showed that decreasing or increasing the constant current is unfavorable to the reaction (entries 7-8). Furthermore, the use of other solvents, such as DMF, CF3OH, CF3CH2OH, and DMSO, was explored but provided no improvement over CH3CN (entries 9–12), and the reaction using nBu4NCl as electrolyte instead of Et4NCl resulted in only trace of product 3 (entry 13). In addition, the yield of alkyl boronate 3 slightly decreased when the reaction was performed using 5.0 equiv. of cyclohexane 1 (entry 14). When using trifluoroacetic acid (TFA) or trifluoromethanesulfonic acid (TfOH) instead of HCl for the reaction, the target product cannot be obtained at all (entry 15). As anticipated, control experiments revealed that electric current and light irradiation were both critical for the desired transformation (entries 16-17).

Table 1 Optimization Studies a
Full size table

Exploration of substrate scope

Under the existing optimization conditions, we began to study the reaction of various C(sp3)−H species with B2(cat)2 under photoelectrochemical conditions to explore the range of viable substrates. As shown in Fig. 2, in addition to cyclohexane, other cycloalkanes, including cyclopentane (4), cycloheptane (5), cyclooctane (6) and cyclododecane (7), are all found to be suitable C(sp3)−H substrates, which were smoothly converted into the corresponding boronate ester products in 50–67% yields. It is noteworthy that adamantane (8) reacts with B2(cat)2 in a unconventionally regioselective manner. The rigid caged structure of adamantane have led to a series of C–H functionalization protocols to proceed at the slightly stronger 3 °C–H bond (BDE of 3 °C–H bond: 99 kcal/mol; BDE of 2 °C–H bond: 96 kcal/mol)77,78. In contrast, our photoelectrochemical C(sp3)–H borylation reaction preferentially delivered the 2 °C–H borylated product 8 in moderate yield (47%) and 86:14 regioselectivity (2°:3°). Several acyclic alkanes were tested and were found to be well compatible under standard conditions, furnishing the desired alkyl boronate esters 9–12 in moderate yields. Interestingly, the site selectivity matched the Aggarwal’s results64, exclusively occurring at the distal methyl position rather than weaker 3 °C(sp3)–H bonds, which indicates that steric hindrance plays a key role in affecting the regioselectivity of C–H borylation. Furthermore, borylation of n-pentane gave 13 in a combined yield of 63%. The reaction favored the distal methyl group over the methylene groups, and resulted in 54:33:13 selectivity (α:β:γ), which is also similar to Aggarwal’s results. Another interesting example is norbornane, which could react with B2(cat)2 to give borylated product 14 in highly regioselective and stereoselective manner. Chlorine-containing molecules were found to be viable substrates (15–17), with the observation of minor dehalogenation product (usually <10%). In addition, isopropyl-substituted benzene were borylated with complete site selectivity for the methyl groups over both aromatic and benzylic C–H bonds (18–19).

Fig. 2: Substrate scope.
figure 2

Reaction conditions: undivided cell, graphite rod anode (φ = 3 mm), Pt plate cathode (10 × 10 × 1 mm), substrate (2.0 mmol, 10.0 equiv.), B2(cat)2 2 (0.2 mmol, 1.0 equiv.), Et4NCl (0.1 mmol, 0.017 M), concentrated HCl (0.8 mmol, 4.0 equiv.), CH3CN (6.0 mL, technical grade), 390 nm LEDs (10 W), constant current = 2 mA under N2 atmosphere at room temperature for 12 h (4.48 F/mol); then pinacol (0.6 mmol, 3.0 equiv.) and triethylamine (0.84 mL) in DCM (1.0 mL), 1 h. a Yields of isolated products. b Regioselectivity (r.r.) and diastereomeric ratio (d.r.) were determined by 1H NMR analysis. c r.r. and d.r. were determined by GC analysis. d Using HCl in 1,4-dioxane (4.0 M) instead of concentrated HCl, constant current = 1.7 mA, and using anhydrous CH3CN instead of technical grade CH3CN. e Large-scale reaction: undivided cell, graphite rod anode (φ = 6 mm), Pt plate cathode (10 × 10 × 1 mm), 390 nm LEDs (30 W), constant current = 5 mA for 24 h.

Full size image

Next, we studied the range of oxidative photoelectrochemical C(sp3)–H borylation of silicon-containing molecules. C(sp3)–H borylation was successful at the α-silyl position of trimethyl(phenyl)silane (20), benzyltrimethylsilane (21), 1,4-bis(trimethylsilyl)benzene (22), 1,1,1,2,2,2-hexamethyldisilane (23), and bis(trimethylsilyl)methane (24), delivering synthetic valuable (borylmethyl)silanes in 44–69% yields. The same trend was also observed for the utilization of halogen-containing silanes (25–27). Notably, the C(sp3)–H borylation of tetraethylsilane (28) revealed that steric effect still dominates the site selectivity, preferentially giving the distal methyl borylated products over the α-borylated ones (α:β = 12:88). The C–H borylation of boryl group-containing compound gave 30 in 46% yield with complete site selectivity for the less-hindered primary C–H sites. Moreover, selective C(sp3)–H borylation of ether compounds, such as 1-(isopentyloxy)−3-methylbutane (31) and 1-propoxypropane (32), was also obtained, albeit in a decreased yield. Unlike conventional C–H functionalization protocols that exhibited high selectivity at the α-position to oxygen79,80, our method allowed C–H borylation to occur preferentially at the primary methyl position, with no borylation of α-C(sp3)–H bond observed. This is probably because the C(sp3)–H bonds adjacent to an oxygen atom can be further oxidized to form the corresponding carbocations under electrochemical conditions81,82.

The ubiquity of ketones has demonstrated the significance of these building blocks in organic chemistry. However, the introduction of a boryl group into ketone motifs has been rarely reported. Different types of carbonyl compounds were selected to test the adaptability of our photoelectrochemical C(sp3)–H borylation method. As expected, the reaction of several simple ketones as substrates gave the corresponding borylated products 33–39 in moderate to good yields and a complete site selectivity for the less-hindered distal methyl sites. The regioselectivity seems to be very sensitive to the carbonyl group, with no borylation observed at its α-position in all the cases. When the use of pentan-2-one (40) and heptan-4-one (41) was investigated under the standard conditions, the corresponding products can be obtained without obvious preference (β:γ = 50:50), probably due to an effect of boron–carbonyl-oxygen interaction83,84. Likewise, the C(sp3)–H borylation of 2-methylpentan-3-one (42) gave no preference to the β- and β’-positions. Moreover, esters also proved to be viable substrates, and again C–H borylation preferentially occurred at distal methyl position (43–46). Althought the site selectivity of substrates such as methyl butyrate (47) showed the same trend as pentan-2-one, steric hindrance still have significant effect on the site selectivity, allowing methyl 2-methylbutanoate (48) to proceed with the γ-C–H borylated product being the most favored. When the isopropyl end was far away from the aliphatic ester functional group, the reaction showed good yields as well as regioselectivity (49–51). In addition, propyl propionate (52), isobutyl propionate (53), and isobutyl isobutyrate (54) were selected as suitable substrates to further investigate the selectivity of ester-containing compounds. Our experimental results revealed the same trend of regioselectivity, affording C(sp3)–H borylation products 52–54 with coupling site preferentially at the distal methyl group on the ester oxygen side. Similar results were observed for the use of phenyl-containing esters as substrates, with the exclusive installation of boryl moiety at the terminal isopropyl group (55–57). In addition, the extension of our photoelectrochemical protocol to nitriles was also successful in affording cyano-containing organoboron product 58 in 42% yield and 50:50 selectivity (δ:γ). The limitation of this photoelectrochemical protocol is that the reaction cannot be well applied to nitrogen-containing substrates, such as amides, amines, or sulfonamides. Late-stage modification of complex organic molecules is the basis for the evaluation of a practical protocol. The direct use of B2(cat)2 with complex hydrocarbons to alkylboron units makes the standard photoelectrochemical conditions applicable for the late-stage modification of biologically valuable natural products and pharmaceuticals. For example, squalane, a saturated oil that has anti-inflammatory property and is often utilized in skin care products as a moisturizer, was successfully borylated at the terminal methyl site to give alkyl boronate product (59) in 43% yield and complete site selectivity.

Gram-scale reaction

In order to further demonstrate the synthetic potential of this photoelectrochemical method, a gram-scale experiment was performed on the batch. When photo-irradiation and electrolysis were carried out with modified light source, electrode materials, and reaction conditions using 10 mmol B2(cat)2 and 100 mmol 2,4-dimethylpentan-3-one as substrates, 1.25 g (52% yield) of the desired β-borylated product 38 was successfully obtained (Fig. 2), which indicates that it represents a promising synthetic method and has the potential to easily transform two readily available starting materials into highly valuable organoboron compounds. Subsequently, we also successfully translated our batch protocol to a scalable continuous-flow fashion, which enables a larger scale of transition between cyclohexane (1) and B2(cat)2 (2) under modified photoelectrochemical conditions (Fig. 3). Although the reaction time may not be greatly shortened (very slow flow rate was applied due to the restriction of our continuous-flow setup), the collected reaction solution smoothly afforded the borylated product (3) in 61% yield on a 20 mmol scale, after workup and purification. By applying a flow rate of 0.1 mL/min in flow, we were also able to prepare alkylboron product 38 in 56% yield and complete distal methyl selectivity.

Fig. 3: Gram-scale reaction in continuous-flow.
figure 3

Reaction conditions: substrate 1 or S38 (200 mmol, 10 equiv.), B2(cat)2 2 (20 mmol, 1.0 equiv.), Et4NCl (10 mmol), concentrated HCl (60 mmol), CH3CN (150 mL, technical grade), 390 nm LEDs (50 W), constant current = 10 mA under N2 atmosphere at room temperature (flow rate: 0.1 mL min−1); then pinacol (60 mmol, 3.0 equiv.) and triethylamine (400 mmol, 20 equiv.) in DCM (20 mL), 1 h.

Full size image

Mechanistic studies

Cyclic voltammetry (CV) was then performed to add more credence to the above results. The battery device was composed of glassy carbon working electrode, Pt coil pair electrode and Hg/HgCl (in saturated KCl solution) reference electrode. All measurements were carried out in dry degassed MeCN, using 0.1 M nBu4NClO4 as the supporting electrolyte, and each measurement consisted of an oxidation scan followed by a reverse reduction scan using a 100 mV/s scan rate. As shown in Fig. 4, it can be seen from the CV diagram that the oxidation potential of HCl is 1.87 V vs. SCE, the oxidation potential of chlorine is 1.62 V, and the oxidation potential of B2(cat)2 (2) is 2.20 V. From these results we can realize that Cl− species is more susceptible to oxidation when compared with B2(cat)2, suggesting that the anodic oxidation process is more likely to initiate with the oxidation of Cl− to generate chlorine radical intermediate, rather than the oxidation of B2(cat)2 (2).

Fig. 4: Cyclic voltammograms.
figure 4

The cyclic voltammogram was measured in MeCN (nBu4NClO4). a HCl (2 mmol L−1); b Et4NCl (2 mmol L−1); c B2(cat)2 2 (10 mmol L−1).

Full size image

To gain more insights into the reaction mechanism, a series of mechanistic experiments were conducted, as shown in Fig. 5. According to Xu’s work14, the C(sp3)–H borylation of cyclohexane (1) with B2(cat)2 (2) in the presence of Cl2, which was generated by mixing NaOCl with HCl, provided the desired alkyl boronate ester 3 in 28% yield under 390 nm light irradiation. However, the formation of 3 was not observed in the absence of LED light (Fig. 5a), which revealed the involvement of Cl2 formation in our photoelectrochemical C(sp3)–H borylation reaction, and the Cl2 is produced through electrochemical process. To explore whether the HAT process is the rate-determining step of the whole photoelectrochemical process, a kinetic isotope effect (KIE) experiment was subsequently carried out, as shown in Fig. 5b. The reaction of cyclohexane (1) and deuterated cyclohexane (1-d12) with B2(cat)2 delivered a mixture of products 3 and isotopically labeled 3-d11 in 64% yield with secondary kinetic isotope effects (kH/kD = 1.0), indicating that the HAT process is not a turnover limiting step for this reaction. Furthermore, the electrolysis of styrene (60) in the absence of boron reagent afforded a chlorine-containing ketone product 61 in 69% yield, indicating the generation and involvement of chlorine radicals during our reaction process (Fig. 5c). Consistent with this finding, the reaction of N,N-diallyltosylamine (62) under the photoelectrochemical system produced the chlorinated ring-closing product 63, which is derived from a chlorine radical addition–5-exo-trig cyclization–hydrogenation sequence85. In order to capture the key radical intermediates, an electron paramagnetic resonance (EPR) experiment was performed by adding a free radical spin trap 5,5-dimethyl-1-pyrroline N-oxide (DMPO). The afforded EPR spectroscopy revealed that carbon-centered radicals are generated via intermolecular HAT process and rapidly captured by DMPO to form relatively stable free radicals (g = 2.007, AN = 14.81 G, and AHβ = 21.38 G)86,87,88,89.

Fig. 5: Mechanistic studies.
figure 5

a C(sp3)-H borylation in the presence of Cl2, b Kinetic isotopic effect (KIE) experiment, c Cl radical trapping experiment, d EPR experiment, e H/D scrambling experiment, f Trapping by other radical acceptors, g Divided-cell and control experiment.

Full size image

After demonstrating the key intermediates during the reaction process, we next started to investigate the origin of the generated site selectivity for the C(sp3)–H borylation. In general, C(sp3)–H functionalization reactions via chlorine radical-mediated HAT process preferentially occur at the weaker C–H bonds, leading to the regioselectivities that follow the trend of 3° > 2° > 1°87,90,91,92. Considering that the chlorine radical-mediated HAT step has proved to be not turnover-limiting through our KIE experiment, we hypothesized that chlorine radical performs the HAT process to initially give a 3° carbon-centered radical. However, slower reactivity of more substituted radical with B2(cat)2 may result in faster functionalization of the sterically unhindered 1° radical, which implies that the alkyl radical formation via HAT is fast and reversible. To test our hypothesis, three deuterated compounds (S55-d, S56-d, and S57-d) were synthesized, with moderate to high levels of deuterium incorporation at the tertiary sites (Fig. 5e)93. As expected, upon light irradiation and electrolysis for 12 h moderate H/D scrambling was observed, with decreased D-incorporation at the tertiary sites of the corresponding C–H borylated products (55-d, 56-d, and 57-d). Furthermore, 2,3-dimethylbutane (DMB, S9) was utilized as a standard alkane substrate to probe the site selectivity (Fig. 5f)87,94,95. Under the standard photoelectrochemical conditions, control reaction revealed that the addition of benzylidenemalononitrile (64)96 as a typical radical acceptor greatly inhibited the desired C–H borylation reaction, while generating C–H alkylated product 65 preferentially at the weaker 3 °C–H bond (intrinsic selectivity of 1°/3°: 1:3). Consistent with this finding, the addition of phenyl acrylate (66)79 into the photoelectrochemical system with DMB (S9) and B2(cat)2 (2) produced C–H alkylated product 67 in 65% yield, and again, the alkylation at the 3 °C–H bond is preferred (intrinsic selectivity of 1°/3°: 1:6). These results supported the reversible HAT process to form carbon-centered radicals, and implied that site selectivity is largely affected by the use of radical acceptors65.

In order to realize the HAT species under our photoelectrochemical conditions, more mechanistic experiments were further carried out (see the Supplementary Information for details). The experimental results indicate that free Cl radical and Cl-boron complexes are both likely to be the HAT species during the reaction process97. To further understand the role of HCl in this photoelectrochemical system, several control experiments were carried out, as shown in Fig. 5g. Replacing HCl with other inorganic chloride salts, such as NaCl, CsCl, and LiCl, was found to be ineffective. Although the use of other acids instead of HCl, such as TFA and TfOH, led to complete loss of reaction efficiency (Table 1, entry 15), substituting HCl by H3PO4 successfully gave the desired C–H borylated product 3 in 67% yield, which indicates that HCl mainly plays a crucial role as proton source rather than chlorine source14. At the cathode surface, protons undergo cathodic reduction to generate H2, and protect the substrate B2(cat)2 from single-electron reduction in certain conditions98. Moreover, a divided-cell experiment was carried out (Fig. 5g). Substrates 1 and 2, along with Et4NCl, HCl, and CH3CN, was placed into the anode chamber, while Et4NCl and CH3CN were added into the cathode chamber. As expected, upon light irradiation and electrolysis for 12 h the desired product 3 was detected in the anode chamber with 46% yield, suggesting that the photoelectrochemical C(sp3)–H borylation occurred surrounding the graphite rod anode. Interestingly, a moderate yield (41%) of alkyl boronate 3 was still afforded in the absence of HCl at the anode chamber, probably because the B2(cat)2 reagent in the anode chamber of divided-cell would not be electrochemically reduced, and therefore it does not need the protection of HCl any more. The protons that come from trace amount of H2O in the cathode cell are combined with electrons at the cathode to generate H2.

Based on the above mechanistic studies and relevant literature reports, we propose the plausible mechanism outlined in Fig. 6. First, the Cl- is electrochemically oxidized to Cl2 at the anode surface, and then the chlorine radical species (A) is generated via light-promoted homolytic cleavage of Cl2. According to its inherent selectivity, the chlorine radical undergoes HAT process with C(sp3)–H compounds (taking S9 as an example) to initially release a more substituted carbon-centered radical (B), which could not proceed with the C–H borylation reaction at the tertiary site, probably due to steric hindrance. Since the alkyl radical formation is reversible (or the alkyl radical can “isomerize” by performing the subsequent reversible HAT with other molecules of the substrate) and fast, a sterically unhindered primary radical (D) is generated and trapped by B2(cat)2 to give an alkyl boronate ester (E) as well as the ligated boryl radical (F) (Path A). Alternatively, the HO-Bcat (G) generated by hydrolysis and anodic oxidation in the reaction system can be complexed with Cl radical species (A) to obtain Cl-radical-boron “ate” complex (H), followed by the HAT process to obtain carbon center radical (D) (Path B). Treating intermediate E with pinacol and triethylamine finally delivers the desired product 9. At the cathode surface, protons undergo cathodic reduction to generate H2, obviating the need for sacrificial oxidants.

Fig. 6: Plausible reaction mechanism.
figure 6

Path A HAT via Cl radical, Path B HAT via Cl-boron complex.

Full size image

Discussion

In summary, we have developed an efficient method that is capable of preparing different types of aliphatic boronate ester products from readily available hydrocarbons and B2(cat)2 via photoelectrochemical C(sp3)–H borylation reaction. Without the need of metal catalysts or oxidants, our protocol addresses the limitations of narrow substrate scope for previously reported C(sp3)–H borylation reactions. The scope of our method has been remarkably expanded (>57 examples), and includes the use of simple alkanes, halides, silanes, ketones, esters and nitriles as viable substrates. Importantly, the reaction exhibits unconventional site selectivity, with the occurrence of C(sp3)–H borylation preferentially at distal methyl position. In addition, a gram-scale synthesis of alkylboron product has been performed on a 20 mmol scale in continuous-flow, which greatly highlights the synthetic potential of our method. Detailed mechanistic studies and discussion of the unconventional site selectivity has been also carried out. Given the significance and versatility of organoboron compounds as valuable building blocks in organic synthesis as well as unconventional site selectivity we are reporting, we believe that this study will offer enormous opportunities for selectively installing valuable functional groups of drug molecules at the late stage.

Methods

General procedure A for photoelectrochemical C(sp3)–H borylation (alkanes, halides, and silanes)

To a 10 mL vial equipped with a stir bar was added B2(cat)2 (48 mg, 0.20 mmol, 1.0 equiv.), Et4NCl (0.1 mmol, 0.5 equiv.), and HCl (concentrated, 0.8 mmol, 4.0 equiv.). MeCN (6 mL) was then added followed by the alkane (2.0 mmol, 10 equiv.). The reaction mixture was stirred at 680 rpm, irradiated with 390 nm LED lamps and 2.0 mA electrolysis for 12 h. The reaction temperature was maintained at approximately room temperature by cooling with a desk fan. After irradiation, a solution of pinacol (71 mg, 0.60 mmol, 3.0 equiv.) and Et3N (0.84 mL, 6.0 mmol, 30 equiv.) in CH2Cl2 (1 mL) was added and stirring was continued for 1 h. The reaction mixture was concentrated in vacuo and purified by flash column chromatography to afford the corresponding boronate ester product.

General procedure B for photoelectrochemical C(sp3)–H borylation (ethers, esters, and nitriles)

To a 10 mL vial equipped with a stir bar was added B2(cat)2 (48 mg, 0.20 mmol, 1.0 equiv.), Et4NCl (0.1 mmol, 0.5 equiv.), and HCl in 1,4-dioxane (0.8 mmol, 4.0 equiv.). Dry MeCN (6 mL) was then added followed by the alkane (2.0 mmol, 10 equiv.). The reaction mixture was stirred at 680 rpm, irradiated with 390 nm LED lamps and 1.7 mA electrolysis for 12 h. The reaction temperature was maintained at approximately room temperature by cooling with a desk fan. After irradiation, a solution of pinacol (71 mg, 0.60 mmol, 3.0 equiv.) and Et3N (0.84 mL, 6.0 mmol, 30 equiv.) in CH2Cl2 (1 mL) was added and stirring was continued for 1 h. The reaction mixture was concentrated in vacuo and purified by flash column chromatography to afford the corresponding boronate ester product.