Volume 16, Issue 17 e202300685
Research Article
Open Access

Electrochemical Aldehyde Oxidation at Gold Electrodes: gem-Diol, non-Hydrated Aldehyde, and Diolate as Electroactive Species

Dr. Christoph J. Bondue

Corresponding Author

Dr. Christoph J. Bondue

Chair of Analytical Chemistry II, Faculty of Chemistry and Biochemistry, Ruhr University Bochum, Bochum, 44801 Germany

Search for more papers by this author
Marius Spallek

Marius Spallek

Chair of Analytical Chemistry II, Faculty of Chemistry and Biochemistry, Ruhr University Bochum, Bochum, 44801 Germany

Search for more papers by this author
Lennart Sobota

Lennart Sobota

Chair of Analytical Chemistry II, Faculty of Chemistry and Biochemistry, Ruhr University Bochum, Bochum, 44801 Germany

Search for more papers by this author
Prof. Dr. Kristina Tschulik

Corresponding Author

Prof. Dr. Kristina Tschulik

Chair of Analytical Chemistry II, Faculty of Chemistry and Biochemistry, Ruhr University Bochum, Bochum, 44801 Germany

Max-Planck-Institut für Eisenforschung GmbH, 40237 Düsseldorf, Germany

Search for more papers by this author
First published: 21 July 2023

Graphical Abstract

In aqueous electrolytes, aldehydes exist in three forms: The non-hydrated aldehyde (red), the diol (green), and the diolate (blue). We show here that all three species are electroactive. Accordingly, strategies to mitigate OH-induced aldehyde decomposition during electrochemical aldehyde oxidation can be developed.

Abstract

To date the electroactive species of selective aldehyde oxidation to carboxylates at gold electrodes is usually assumed to be the diolate. It forms with high concentration only in very alkaline electrolytes, when OH binds to the carbonyl carbon atom. Accordingly, the electrochemical upgrading of biomass-derived aldehydes to carboxylates is believed to be limited to very alkaline electrolytes at many electrode materials. However, OH-induced aldehyde decomposition in these electrolytes prevents application of electrochemical aldehyde oxidation for the sustainable upgrading of biomass to value-added chemicals at industrial scale. Here, we demonstrate the successful oxidation of aliphatic aldehydes at a rotating gold electrode at pH 12, where only 1 % of the aldehyde resides as the diolate species. This concentration is too small to account for the observed current, which shows that also other aldehyde species (i. e., the geminal diol and the non-hydrated aldehyde) are electroactive. This insight allows developing strategies to omit aldehyde decomposition while achieving high current densities for the selective aldehyde oxidation, making its future industrial application viable.

Introduction

Electrode materials such as copper,1, 2 silver,1, 2 gold,1-5 nickel,2 mercury2, 6 oxidize aldehydes selectively to carboxylic acids. This sets them apart from electrode materials such as platinum,7-9 palladium8 and rhodium7 which at least partly cleave the C−C bond between carbonyl carbon and α-carbon and do not yield carboxylic acids selectively. However, it is widely believed that the electrochemical active species of aldehyde oxidation is the diolate anion at gold, mercury and transition metal oxide electrodes.1-6, 10 The latter forms either when OH binds to the carbonyl carbon or when the gem-diol, which forms upon hydration of the carbonyl functional group, is deprotonated (Scheme 1A).11, 12 Accordingly, high equilibrium concentrations of the diolate exist only in very alkaline electrolytes. Hence, the assumption that the diolate is the only electroactive species leads to the believe that efficient electrochemical aldehyde oxidation is limited to very alkaline conditions. However, under these conditions aldehydes also undergo rapid decomposition.13-15 This is highly relevant for the upgrading of bio-mass as a renewable carbon feedstock: the oxidation of bio-derived aldehydes such as sugars or 5-(hydroxymethyl)furfural to the corresponding carboxylic acid is desired to produce value-added compounds that can substitute petrochemicals, which is necessary to develop a sustainable chemical industry.16 Yet, if it is necessary to form the diolate in solution prior to the electrochemical reaction, then it would be impossible to avoid OH induced aldehyde decomposition as an undesired side-reaction. This would therefore strongly limit the economic relevance of this route towards sustainable production of chemicals. Hence, it is worthwhile to examine whether other electroactive species exist and if so, which strategies can be developed that allow decoupling of electrochemical aldehyde upgrading from their electrolyte induced decomposition.

Details are in the caption following the image

A) Equilibrium between non-hydrated aldehyde, gem-diol and diolate. B) Schematic representation on how the fraction of aldehyde residing as the diolate species evolves as a function of the electrolyte pH and aldehyde pKa. If the only electroactive species of aldehyde oxidation is the diolate species, then the limiting oxidation current describes the same sigmoidal curve as in (B) in the current/pH domain. This was observed by Bover and Zuman for the oxidation of benzaldehyde derivatives at mercury.6

Only for mercury electrodes there are strong arguments that the selective oxidation of aldehydes to carboxylic acids proceeds via the diolate species:2, 6 Bover and Zuman showed in the 70 s6 that the current density due to the oxidation of benzaldehyde derivatives increases with electrolyte pH. The observed current/pH profile closely follows the same sigmoidal curve, that also describes the fraction of aldehyde that resides as diolate in the electrolyte. This fraction is shown as urn:x-wiley:18645631:media:cssc202300685:cssc202300685-math-0001 in Scheme 1B, where [HA]0 is the overall amount of added aldehyde and [A] the equilibrium concentration of the diolate.[6] This strongly suggests that the diolate is the electroactive species of aldehyde oxidation at mercury electrodes.2, 6 Furthermore, Bover and Zuman observed that electron withdrawing or donating substituents at the phenyl ring of the investigated benzaldehyde derivatives shift the sigmoidal curve along the pH-axis.6 The presence of these substituents have the same effect on the curves as the pKa-value of the gem-diol in Scheme 1B. This makes sense because electron donating substituent decrease the pKa-value and electron withdrawing substituent increase its pKa-value of the gem-diol.6 Hence, the influence of intramolecular electronic effects on the current are further evidence that the diolate is the electroactive species of aldehyde oxidation at mercury electrodes.

However, it cannot be taken for granted that the diolate is also the electroactive species of aldehyde oxidation at other electrode materials. Indeed, van Effen and Evans noted that nickel oxides can convert aldehydes to the corresponding carboxylic acid also in neutral solution, where the equilibrium concentration of diolate is close to zero.2 Moreover, gold electrodes exhibit some activity for aldehyde oxidation in acidic electrolytes.17 Although this activity is small, it cannot be understood if the aldehyde reacts exclusively via the diolate species. Furthermore, it is not obvious that the mechanism of aldehyde oxidation observed at mercury electrodes should be transferrable to coinage metal electrodes: That is, the rate limiting step of aldehyde oxidation at gold1, 10 is the cleavage of the C−H bond between carbonyl carbon and formyl hydrogen, which leads to the formation of adsorbed hydrogen.1, 3, 18 It is unlikely that a similar step occurs during aldehyde oxidation at mercury electrodes. In light of the large overpotentials required for hydrogen evolution at mercury electrodes,19 it appears unfavorable to form adsorbed hydrogen on mercury during the C−H bond cleavage. This suggests that the mechanism of aldehyde oxidation at gold and mercury electrodes differs from each other.

It would be beneficial if the oxidation of aldehydes at gold electrodes was not limited to the diolate species. It is therefore also worthwhile to reexamine whether the diolate is really the only electroactive species of aldehyde oxidation when gold electrodes are employed. For this reason, we study here the electrochemical oxidation of acetaldehyde (CH3−CHO) and other aliphatic aldehydes (R−CHO) at a rotating gold disc electrode at pH 12. At this pH only 1 % of acetaldehyde reside as the diolate in solution. When comparing the theoretically expected diffusion limited current with the experimentally observed current, we find that the diolate concentration at this pH is too low to account for the observed current. This clearly indicates that also the gem-diol (R−CH(OH)2 and the non-hydrated aldehyde (R−CHO)) undergo the electrochemical reaction. This result is highly relevant because it shows that aldehyde oxidation does not require high diolate concentrations in the electrolyte. Accordingly, strategies can be developed that allow the oxidation of aldehydes with high current densities, while avoiding their OH induced decomposition.

Experimental

Acetaldehyde (99.5 %, Sigma-Aldrich), propanal (98 %, Sigma-Aldrich) and butanal (99.5 %, Sigma-Aldrich) were distilled under inert gas from NaHCO3 to reduce the acid content. All aldehydes were stored at −27 °C in a round bottom flask to avoid excessive aerobic oxidation.

Typically, electrolytes were prepared from NaClO4 (99.99 %, Sigma-Aldrich), NaOH (99.99 %, Sigma-Aldrich), Na2HPO4 (99.99 %, Sigma-Aldrich), B(OH)3 (99.8 %, Roth) and NaHCO3 (99.5 %, Grüsig). However, NaOH (98 %, Sigma-Aldrich) was used to prepare electrolytes featuring 1.0 M and 3.0 M NaOH.

After each experiment the pH of the electrolyte was determined with a pH meter. The used pH-electrode (LE410, Mettler Toledo) can measure pH-values accurately even in alkaline electrolytes up to pH 14. Since the electrolytes featuring high NaOH concentration of 1.0 M and 3.0 M are expected to feature pH-values that come close to or go beyond the measurement range of our pH-electrode, we used the pH-values available in literature.20

The rotating gold disc electrode was of 99.999 % purity and was obtained along with the rotator from IPS Elektroniklabor. RDE measurements were limited to a rotation frequency of 60 rpm as otherwise diffusion limited currents would become too large and compliance voltage issues would emerge. The use of higher rotation speeds is affordable when the acetaldehyde concentration (up to 5 mM) and/or the pH value of the electrolyte (10–12) is kept low. However, no effect of the rotation speed was observed beyond the effect expected from the Levich equation. Cyclic voltammograms were recorded with a PGSTAT 128 (Autolab) in staircase mode. Capacity measurements were conducted with a single frequency of 10 Hz and an amplitude of 10 mV using a SP-300 potentiostat (BioLogic).

Rotating-disc electrode (RDE) measurements were conducted with an electrolyte volume of at least 300 mL in an RDE cell, which featured a glass frit that separates the compartments of working and counter electrode. Thus, metal ions dissolving at the counter electrode (titanium) do not contaminate the working electrode easily. Furthermore, the compartments of the working electrode and the reference electrode were separated from each other through a glass tap, which inhibits contamination of the working electrolyte with chloride ions from the employed reference electrode (saturated Ag/AgCl). The latter was protected from the alkaline electrolyte through a double junction filled with an electrolyte of 1.0 M NaClO4. All potentials mentioned in the text, if not stated otherwise, are referenced against Ag/AgCl in a 3 M KCl solution.

Glass ware was cleaned prior to use by storing it in an acidic solution of KMnO4 (5 mM KMnO4 in 0.5 M H2SO4) for at least 8 h. The glass ware was than rinsed with MilliQ water (conductivity at 24.2 °C not more than 0.055 μS cm−1). MnO2 adhering to the glass ware was dissolved with an acidic solution of H2O2 followed by rinsing with MilliQ water to remove any trace of H2O2. If the glass ware was not used right away it was stored under MilliQ water.

Results and Discussion

The curves in Figure 1A show the current due to acetaldehyde oxidation obtained in the positive going sweep at a rotating gold electrode (ϖ=60 rpm) in an electrolyte of 50 mM NaClO4, 12 mM NaOH (pH≈12) and the indicated concentrations of acetaldehyde. Depending on the acetaldehyde concentration, the oxidation current rises above the background current between -0.6 V and -0.45 V and reaches a limiting value at potentials below 0 V (i. e. between −0.3 V and 0 V). Note that the exact potential at which the limiting current is reached depends on the aldehyde concentration. At more positive potentials (i. e. between 0.4 V and 0.7 V), the current due to aldehyde oxidation drops to zero again. Also, here the potential at which the oxidation stops depends on the aldehyde concentration in the electrolyte. However, for low aldehyde concentrations, the potential at which the oxidation stops coincides well with the oxidation of the gold surface in the blank electrolyte, suggesting that gold oxide does not catalyze the electrochemical oxidation of acetaldehyde. One might suggest that it is not the formation of gold oxide but the formation of a layer of chemisorbed carboxylate that inhibits the oxidation of the aldehyde. However, the adsorption of carboxylic acids occurs already several hundred mV negative of the potential of gold oxide formation.21 Accordingly, aldehyde oxidation would stop at much lower potentials if the adsorption of carboxylates inhibited the electrochemical reaction.

Details are in the caption following the image

Results of an RDE experiment in an electrolyte of 50 mM NaClO4 and 12 mM NaOH containing the indicated concentrations of acetaldehyde. A) Current as a function of the applied potential (for clarity reasons a full cycle is only shown for an aldehyde concentration of 2 mM). B) Measured current (black squares) and expected diffusion-limited current of acetaldehyde oxidation (red line: acetaldehyde as diffusion-limited species, blue line: OH as diffusion limited species) as a function of the acetaldehyde concentration. Considering that the electrolyte pH was determined as 11.99, a OH concentration of 10 mM was used to calculate the blue curve. Working electrode: Au with a geometric area of 0.501 cm2; rotation rate: 60 rpm; sweep rate: 20 mV s−1.

Furthermore, in the backwards going sweep of the black curve of Figure 1 we observe that the activity of the electrode is recovered at the potential in which gold oxide is reduced to gold (c.f. grey curve in Figure 1). This highlights that the activity of the electrode depends on the presence or absence of a surface oxide.

To determine whether the observed limiting current is due to mass transport effects, we have calculated the expected diffusion limited current (IL) for acetaldehyde oxidation using the Levich equation (Eq. 1),
urn:x-wiley:18645631:media:cssc202300685:cssc202300685-math-0002(1)

where F is the Faraday constant in C/mol, A is the geometric surface area of the electrode in cm2 and c is the concentration of the reacting species in mol/l. In our calculation we have considered a rotation speed of ϖ=60 rpm, a diffusion coefficient of acetaldehyde of D=1.2 ⋅ 10−9 m2 s−1 (in water at 20 °C, determined from the parameters given in Ref. [22]), a kinematic viscosity of water at 20 °C ν=1.0034 mm2 s−1 and z =2 as the number of transferred electrons per reacting molecule (i. e., selective oxidation of acetaldehyde to acetate,2 Scheme 2).23 The calculated values of the diffusion limited current are plotted in Figure 1B as a function of the aldehyde concentration (red curve). Since the experimentally observed limiting current (black squares in Figure 1B) is described well by the red line, it is clear that the oxidation of acetaldehyde is mass transport limited in acetaldehyde at 0.00 V and in the concentration regime between 0 and 10 mmol. Furthermore, the good match confirms that two electrons per reacting acetaldehyde molecule are transferred. This confirms previous results of van Effen and Evans, reporting that aldehydes are oxidized at gold to the corresponding carboxylate.2

Details are in the caption following the image

Oxidation of acetaldehyde to acetic acid in alkaline electrolyte.

However, once the aldehyde concentration exceeds 10 mM, the limiting current does not increase further and follows the blue line in Figure 1B, which represents the current that is expected when the oxidation of acetaldehyde is limited by mass transport of OH to the electrode surface. The latter was determined via Equation (1) using the diffusion coefficient of OH at infinite dilution (D=6.7 ⋅ 10−9 m2 s−1)24 and z=2/3. That is, the oxidation of acetaldehyde to acetate consumes 3 OH and releases 2 electrons per reacting molecule (c.f. Scheme 2), so that 2/3 electrons are transferred per OH that arrives at the electrode surface.

It is interesting to consider the results presented in Figure 1 in the context of Figure 2, which shows the fraction of acetaldehyde that resides in aqueous solution as non-hydrated aldehyde (red), gem-diol (green) and diolate (blue) as a function of the electrolyte pH. At pH 12 the equilibrium concentration of the diolate species is low (ca. 1 % of the overall aldehyde content). Furthermore, as the electrochemical oxidation of aldehyde consumes OH, the local pH at the electrode surface decreases. Therefore, mass transport limitation of hydroxide during aldehyde oxidation indicates a near neutral pH in the vicinity of the electrode, which should come along with a local equilibrium concentration of the diolate species that is essentially zero. Considering the low equilibrium concentration of the diolate species in the bulk electrolyte – and more importantly in the vicinity of the electrode, it appears implausible that only the diolate species is electroactive. If only the diolate would be electroactive, then only 1 % of the acetaldehyde species arriving at the electrode would undergo a reaction and only 1 % of the theoretically expected diffusion limited current would be observed experimentally. This is clearly not the case as significantly larger currents are observed (see Figure 1), revealing that also the gem-diol and/or non-hydrated aldehyde are electroactive species. This estimation implies that the kinetics of diolate formation are slow on the time scale of the diffusion process. We can assume that this is the case, as otherwise the behavior outlined in the introduction and in Scheme 1 would not have been observed by Bover and Zuman at mercury electrodes,6 which is partly reproduced in Figure S1 in the Supporting Information. That is, if the kinetics of diolate formation were fast, then also at mercury electrodes the rate of aldehyde oxidation should not be affected by the electrolyte pH. Hence, the results presented in Figure 1 show that also gem-diol and the non-hydrated aldehyde are active species of acetaldehyde oxidation at gold electrodes in the potential range in which diffusion limitation is achieved. However, it stands to reason that gem-diol and non-hydrated aldehyde are also active species of aldehyde oxidation in the potential region where the current is controlled by kinetics. That is, if the oxidation of the acetaldehyde occurred only via the reaction of the diolate close to the onset potential, we would expect to observe a diffusion limited current, that takes a value that is approximately 1 % of the diffusion limited current at 0 V. Such behavior is not observed in Figure 1, proving that the diolate is not the only electroactive species in aldehyde oxidation on gold.

Details are in the caption following the image

Fraction of aldehyde that resides in aqueous solution as non-hydrated aldehyde (red), diol (green), or diolate (blue) as a function of the electrolyte pH. The curves were determined from the relevant thermodynamic data for gem-diol formation11 and the pKa-value of the gem-diol.12 In doing so we neglected the presence of the enol and enolate species. Grey line: pH of the experiment of Figure 1.

The same results as for acetaldehyde shown in Figure 1 were obtained for propanal and butanal as shown in Figures S2 and S3 in the Supporting Information. The fact that the oxidation of propanal and butanal shows qualitatively the same behavior suggests that aldehyde oxidation at gold is not limited to the oxidation of the diolate as the only electroactive species. That is, opposed to the general assumption in literature it is not necessary to create a high diolate concentration in the electrolyte to achieve the oxidation of aldehydes. For gold electrodes we clearly establish here that also the gem-diol and the non-hydrated aldehyde can be oxidized electrochemically. This statement does not exclude that also the diolate undergoes an electrochemical oxidation reaction. In fact, we show in the Supporting Information (c.f. Figure S4 and related discussion) that diolate is also active. The experimentally observed diffusion limited current of acetaldehyde oxidation continues to match the theoretically expected value when we increase the diolate concentration in the electrolyte by increasing the electrolyte pH. If the diolate was not electroactive, then by raising the pH we would reduce the fraction of aldehyde in the electrolyte that is electroactive. In this case we would expect that the experimentally observed diffusion limited current deviates increasingly from the theoretically expected value with increasing pH. Since we do not observe such a deviation, also the diolate is electroactive during aldehyde oxidation at gold electrodes.

In Scheme 3 we present a mechanism for electrochemical aldehyde oxidation at gold electrodes. This mechanism accounts for the first time coherently for important experimental findings that are presented in literature and for the fact that also gem-diol and non-hydrated aldehyde participate in the electrochemical reaction. In steps 2a, 1b and 1c, respectively, the C−H bond between carbonyl carbon and formyl hydrogen is cleaved, which leads to the carbon bound adsorbates. We derive this from the finding that aldehyde oxidation leads to the formation of adsorbed hydrogen.1, 3, 18 This phenomenon cannot be explained when aldehyde adsorption through the oxygen atom is assumed. Furthermore, it is very unlikely that the carbonyl oxygen should bind readily to the oxophobic gold surface at potentials at which the oxygen atom of water avoids to interact with the gold surface.25

Details are in the caption following the image

Proposed mechanism of electrochemical acetaldehyde oxidation at gold. Following different pathways (red: non-hydrated aldehyde; green: the gem-diol; blue: diolate) all three aldehyde species form the same adsorbed intermediate. Upon adsorption (steps 2a, 1b and 1c) the C−H-bond between the carbonyl carbon and the formyl hydrogen is cleaved. The adsorbates formed from the diol and the non-hydrated aldehyde undergo a reaction with OH (step 2b and 2c) on the electrode surface to form the same adsorbate as the diolate. This adsorbate undergoes oxidative desorption to form acetic acid in step 3.

Upon adsorption of the diolate (blue pathway) an adsorbate is formed that essentially constitutes a carbon bound carboxylic acid. The carbon bound adsorbates formed from the non-hydrated aldehyde (red pathway) or the gem-diol (green pathway) need to react first with OH to form the same adsorbed species. Considering that aldehyde oxidation is limited to a potential range in which water fragments are not adsorbed on the gold surface (c.f. Figure 1), this reaction cannot follow a mechanism in which a surface reaction between adsorbed OH and adsorbed aldehyde species occurs. That is, the reaction does not follow a Langmuir-Hinshelwood type mechanism. This statement is further corroborated by Figure S5 and the related discussion. Instead, the adsorbates that are formed from the non-hydrated aldehyde and the gem-diol, respectively, react with dissolved OH in step 2b and 2c. Accordingly, the reaction follows an Eley-Rideal type mechanism, which leads to the carbon bound carboxylic acid adsorbate. The later desorbs oxidatively in step 3 as a carboxylic acid, which is deprotonated under alkaline conditions (not shown).

The important difference between the blue pathway on the one hand and the green and red pathway on the other is the sequence of the reaction steps: In the red and green pathway the aldehyde adsorbs first to the electrode surface and reacts then with OH. In the blue pathway the aldehyde reacts first with OH and adsorbs afterwards to the electrode surface. This sequence is very important for process engineering and the selection of reaction conditions: the reaction sequence of steps 1a, 2a and 3 (blue, diolate) implies a need for electrolytes of high pH to create a high equilibrium concentration of the diolate species. This is unfavorable, because high pH values facilitate undesired side reactions such as the condensation of aldehydes to polymers in the homogeneous phase.13-15 That is, aldehydes featuring an α-H-atom can undergo base catalyzed aldol condensation.26 Also, aromatic aldehydes such as HMF that do not feature an α-H-atom are known to undergo decomposition to undesired humin-like products under alkaline conditions.13 It is therefore important that the oxidation of aldehydes can proceed via the red or the green pathways, where OH is only required as a reactant but not as a means to increase the electrolyte pH. That is, a buffer can provide copious amounts of OH for the electrochemical reaction without raising the electrolyte pH. Hence, the OH catalyzed decomposition of aldehydes in the homogenous phase can be avoided or at least slowed down significantly (c.f. Figure S6 and related discussion).

Indeed, when the reaction is conducted in 1.0 M phosphate buffer (adjusted to pH 12.2; Figure 3) the current due to aldehyde oxidation can exceed 1.6 mA. That is, in the phosphate buffered electrolyte the current due to acetaldehyde oxidation reaches up to 4 mA (for 160 mM acetaldehyde). Note that the current due to acetaldehyde oxidation cannot exceed 1.6 mA in unbuffered solution even if the acetaldehyde concentration is increased dramatically. This is due to the relatively low pH of 12, which limits the mass transport of OH to the electrode surfaces. In a buffered solution no such limitation arises since the OH is constantly replenished at the electrode surface through a chemical reaction. Note that even higher currents of 11.5 mA and 14.5 mA are reached in carbonate and borate buffered electrolytes, respectively (c.f. Figures S7 and S8 in the Supporting Information). At the same time, aldehyde decomposition in the electrolyte is avoided because the buffer affords a relatively low alkalinity at around pH 12. It is worth mentioning here that detailed knowledge of the mechanism of electrochemical aldehyde oxidation and of the involved species allows us to design reaction conditions, which support high current densities while avoiding the consumption of the aldehyde due to undesired condensation reactions.

Details are in the caption following the image

Results of an RDE experiment in an electrolyte of 1.0 M Na2HPO4 (adjusted to pH 12.2 by adding the appropriate amount of NaOH) containing the indicated concentrations of acetaldehyde. A) Current (positive going scan) as a function of the applied potential. B) Measured current (black squares) and expected diffusion-limited current for acetaldehyde oxidation as a function of the acetaldehyde concentration (red line). Grey line: Maximal current that can be achieved in an unbuffered electrolyte of pH 12. Working electrode: Au(pc) with a geometric area of 0.501 cm2; rotation rate: 60 rpm; sweep rate: 20 mV s−1.

However, Figure 3B also highlights that there is still scope for improvements: the measured current at 0.00 V (black squares) remains below the theoretically expected diffusion limited current of acetaldehyde oxidation (red line; determined via Eq. (1)). Evidently, the reaction is limited by kinetics in the phosphate buffered electrolyte. That is, while the phosphate buffer allows us to exceed the maximum current imposed by mass transport limitation of OH in the unbuffered electrolyte, we still cannot reach the current that is limited by mass transport of acetaldehyde. However, for process development in bio-mass upgrading, much higher currents are desired. It is therefore important to understand that the observed kinetic limitation results from the adsorption of phosphate anion on the gold surface (c.f. Figure S9 and related discussion in the Supporting Information). This leads to a competition between phosphate ions and acetaldehyde for adsorption sites. Accordingly, the rate of aldehyde oxidation declines. Also, in carbonate and borate buffered electrolyte diffusion limitation is not reached for high aldehyde concentrations. Evidently also borate and carbonate adsorption hamper the electrochemical aldehyde oxidation, albeit to a lesser degree. Yet, it would be desirable to identify buffer systems that feature less strongly adsorbing anions or electrode materials that oxidize aldehydes via the same mechanism as gold electrodes but that are less affected by the presence of strongly adsorbing anions.

Conclusions

We have studied the electrochemical oxidation of aliphatic aldehydes at a rotating gold electrode in electrolytes of pH 12. For low aldehyde concentrations, the observed diffusion limited current matches the theoretically expected value. However, the diolate concentration at this pH is too low to account for the diffusion limited current alone. Based on this observation we can show that also the gem-diol and the non-hydrated aldehyde are electroactive species at gold electrodes. In doing so, we correct the long-standing assumption that aldehyde oxidation requires high OH concentrations to increase the equilibrium concentration of the diolate. Instead, the function of OH is limited to that of a reactant. This is evident from the fact that aldehyde oxidation becomes mass transport limited in OH when the aldehyde concentration in the electrolyte exceeds a certain threshold value.

The insight that it is not necessary for the aldehyde and OH to form the diolate in the electrolyte prior to adsorption is important: since the aldehyde can adsorb first to the gold surface and react afterwards with OH, it is possible to design reaction conditions in which the OH induced decomposition of the aldehyde is avoided and the current density of the aldehyde oxidation is increased. To this end we provide a large OH reservoir in form of a buffer system, which addresses the issue of mass transport limitation in OH. At the same time the buffer maintains mildly alkaline conditions, at which the aldehyde is sufficiently stable.

However, the presence of strongly adsorbing anions hampers the kinetics of aldehyde oxidation. The effect is pronounced in phosphate buffered electrolytes, but also exists in carbonate and borate buffered systems. Hence, the task ahead is the selection of buffer systems containing anions that adsorb less strongly. The goal is to increase the current density due to selective aldehyde oxidation. This will greatly benefit efforts to develop processes for the electrochemical upgrading of bio-mass derived aldehydes to value-added carboxylates.

Note in passing how the presented mechanistic studies allow us to develop strategies to circumvent problems that impair the use of electrochemical procedures in industrial processes. Similar mechanistic studies for other compound classes are required to develop electrochemistry into a viable tool of synthetic chemistry that can be applied at large scale.

Supporting Information

Additional Information are provided on pH dependence on gold amalgam, oxidation behavior of propanal and butanal, effect of pH on the diffusion limiting current of acetaldehyde oxidation, effect of pH on the onset potential of acetaldehyde oxidation, effect of pH on aldehyde decomposition, acetaldehyde oxidation in borate and carbonate buffered electrolytes and phosphate adsorption

Additional references cited within the Supporting Information.27-32

Acknowledgments

This project received funding from the European Union's Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 801459-FP-RESOMUS and was funded by the Deutsche Forschungsgemeinschaft (DFG) under Germany's Excellence Strategy – EXC 2033-390677874-RESOLV and in the framework of the TRR 247-388390466 (project A09). Open Access funding enabled and organized by Projekt DEAL.

    Conflict of interest

    The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

    Data Availability Statement

    The data that support the findings of this study are available from the corresponding author upon reasonable request.