Thermodynamically Spontaneous Reactions and Use of Acetaldehyde as Fuel in Direct Ethanol Fuel Cells

A set of fuel cells was fabricated using a cation exchange membrane (CEM), and ethanol and acetaldehyde were used as fuels for the cells. Spontaneous reactions in the fuel cells were studied under various conditions such as excess oxygen at the cathode and at higher temperatures without application of an external voltage and without forcing an external current. When acetaldehyde was used as a fuel, the current in the circuit and the concentration of acetic acid (product of the fuel cells) were not influenced by the experimental conditions. On the other hand, when ethanol was used as a fuel, the current in the circuit and the concentration of acetic acid were influenced by the experimental conditions. It was thus concluded that ethanol is oxidized directly to acetic acid without oxidation to acetaldehyde and then to acetic acid. The calculated ΔG for possible half reactions in direct ethanol fuel cells (DEFCs) with CEM and AEM (anion exchange membrane) suggests that a reliable AEM should be developed for the further development of DEFCs.

Fuel cells are important energy sources for generating electricity, especially in an environmentally friendly manner.Hydrogen is the most important fuel for fuel cells, and hydrogen fuel cells have thus been well studied. 1 The problems associated with hydrogen fuel cells are the use of platinum (high cost) and how to prepare H 2 .Ethanol is another possible fuel and can be produced by the fermentation of biomass; therefore, ethanol production is non-problematic.Direct ethanol fuel cells (DEFCs) are thus also attracting some interest.
Unlike hydrogen fuel cells, many reactions are possible in DEFCs before ethanol is completely oxidized into CO 2 and H 2 O.We know from experience that ethanol is oxidized in nature to acetaldehyde, which is then oxidized to acetic acid and then to CO 2 and H 2 O.It is important to understand how ethanol could be oxidized in fuel cells.Antoniassi et al. 2 reported that there is a possibility of ethanol molecules being directly oxidized to acetic acid in DEFCs.One of the purposes of this study is to confirm whether this is true or not.
Reactions such as ethanol oxidation to acetaldehyde in fuel cells occur at the anode and oxygen reduction reaction occurs at the cathode.The anodic and cathodic reactions are different and depend on the type of electrolyte membrane.Are all these half reactions spontaneous?A literature survey of studies on DEFCs reveals that in the majority of works on DEFCs, [2][3][4][5][6][7][8] researchers are conducting experiments under an applied voltage or forced current flow.However, fuel cells could be operated without application of a voltage.In this case, we need to understand the spontaneous reactions in DEFCs.By "spontaneous" reaction, we mean "allowed-to-occur" reaction.In thermodynamics, a spontaneous reaction is a reaction whose ΔG is negative and which occurs without the input of energy.Catalyst does not change the value of ΔG, but it speeds up or slows down the reaction.The second purpose of this study is thus to determine which reactions are spontaneous.

Experimental
The anode was prepared by the addition of SiO 2 nanoparticles (10 wt%) to a Pt-Ru/carbon catalyst powder (90 wt%).We added SiO 2 nanoparticles because Si-OH group on SiO 2 surface acted as the remover of acetaldehyde. 9SiO 2 was purchased from Nippon Aerosil, Japan and was in the form of 7 nm diameter nanoparticles.The Pt-Ru/carbon catalyst powder (TEC61E54DM, Tanaka Kikinzoku Kogyo, Japan) was a mixture of 29.5 wt% Pt + 22.9 wt % Ru and carbon.Nafion solution, propanol and water were added to the Pt-Ru/carbon catalyst to prepare the electrode paste.In the SiO 2 mixing process, acetone was added as an additional solvent to prepare a uniform dispersion of the silica nanoparticles. 9The mixture was then ball-milled for 24 h.After ball milling was completed, acetone was removed by vacuum distillation to obtain a paste with a suitable viscosity.0.12 g of the paste was painted onto carbon paper (28 × 28 mm) to form the anode electrode.The cathode electrode (28 × 28 mm) was prepared with Pt/carbon catalyst powder (TEC10E70TPM, Tanaka Kikinzoku Kogyo, Japan) and was a mixture of 66.7 wt% Pt and carbon.The anode, Nafion film (electrolyte), and cathode were placed together and hotpressed at 200 °C to form a membrane electrode assembly (MEA).
The fuel cells used in this experiment were made of two cells that were electrically connected in series.Fuel was supplied to the inlet of a central common tank located between the two cells.The cathodes (air electrodes) were outside both cells and were exposed to air.A Perista ® pump was used to circulate 300 ml of 3% acetaldehyde aqueous solution or a 3% ethanol aqueous solution in a reservoir at a flow rate of ca.5.5 ml min −1 .
The output currents of the DEFCs were measured by a previously reported method 10,11 that is schematically shown in Fig. 1.No external voltage was applied to the electrode and no current was forced to flow into the cells so as to determine the spontaneous reactions in the cells.Approximately 5 ml of fuel was sampled every 30 min of operation to determine the concentration of acetaldehyde and acetic acid generated in the fuel solution.Although the cathodes were exposed to air, excess oxygen gas was supplied from a gas cylinder (99.5%) through two funnels set at the cathode side of the DEFC at several different oxygen gas flow rates of 0, 0.2, 0.4 and 0.6 l min −1 .In another series of experiments, the cells were operated at different temperatures of 26.5, 40, 45 and 48 °C.
During the 5 ml sampling process, continuous circulation of the fuel solution was maintained and an attempt was made to avoid air bubbles coming into the aqueous fuel solution.Although only one cell is shown in Fig. 1, two cells were used in the experiments with a central common fuel tank.
Possible reaction products when acetaldehyde is used as a fuel are acetic acid and CO 2 .When ethanol is used as a fuel, the products may be acetaldehyde, acetic acid and CO 2 .The concentration of acetaldehyde in the 3% ethanol solution was measured by gas chromatography.As a first step, standard solutions of acetaldehyde (0-1000 ppm acetaldehyde in 3% ethanol aqueous solution) were z E-mail: ikuma@kaitchem.onmicrosoft.comECS Advances, 2023 2 044501 prepared.4 ml of a standard solution was mixed with 4 ml of chloroform for 3 min.During this process, acetaldehyde was extracted into the chloroform.Acetaldehyde in the chloroform was then analyzed using gas chromatography (GC-4000, GL Sciences, Japan).The results were used to prepare a calibration curve that was then used to determine the concentration of acetaldehyde in samples of unknown concentration.
The concentration of acetic acid in the 3% ethanol solution or 3% acetaldehyde solution was measured with a pH meter.Standard solutions of acetic acid (0-700 ppm acetic acid in 3% ethanol solution or 3% acetaldehyde solution) were prepared, and the pH of these solutions was measured to prepare a calibration curve.
CO 2 is one of the products that could be present in ethanol or acetaldehyde solutions when they are oxidized.However, it was reported 12 that CO 2 was not a major product.We also have found that the concentration of CO 2 in the ethanol solution did not increase significantly after fuel cell operation.Therefore, we did not measure the concentration of CO 2 except for several cases in which CO 2 was measured to make sure that its concentration in the circulating fuel did not increase.In those several cases, we took sample gas of 2 ml from air in fuel reservoir (Fig. 1).Then we analyzed it by gas chromatograph.Before the fuel cell operation, the concentration of CO 2 in the air was 560 ppm.After 180 min operation of fuel cell, it increased to 2270 ppm.Using the volume of 300 ml for fuel and the volume of 200 ml for air in the fuel reservoir, the amount of CO 2 in the fuel was calculated under the assumption that CO 2 obey Henry's law (Henry's law constant for CO 2 = 3.39 × 10 −1 mol m −3 kPa −1 ).There was 2.1 × 10 −7 mol of CO 2 in the fuel and 1.8 × 10 −7 mol of CO 2 in the air.This was smaller than the amount of acetic acid in the fuel at 180 min of operation which was 6.7 × 10 −3 mol (400 ppm from Fig. 3 was used).
An important point in this study is that all the fuel cells measurements were accomplished with one pair of cells for both acetaldehyde fuel and ethanol fuel.This procedure ensures that differences in the results are not due to differences in the fabrication conditions for the cells.After a measurement under one experimental condition, cells must be returned to the original state.Therefore, the cells were cleaned by the following process.As a first step, fresh 3% acetaldehyde solution or 3% ethanol solution was placed into the reservoir and circulated through the fuel cells for 15 min.Deionized water was then placed into the reservoir and circulated for 10 min.These processes were performed to clean the MEA.All other parts such as tubing, O-rings, boards of cells, spacers and the MEA were dismantled and washed with water by hand until there was no odor from chemicals such as acetic acid.Every part of the fuel cells was then dried before being assembled for the next measurement.

Anodic and Cathodic Reactions in DEFCs
Ethanol is oxidized through several routes.These reactions are: If a cation exchange membrane (CEM) is used as the electrolyte, then the cathodic half reaction is: where H + is transported from the anode to the cathode through the CEM, and e -is transported from the anode to the cathode through the outer circuit.Corresponding anode half reactions should produce H + and e − : If an anion exchange membrane (AEM) is used as the electrolyte, then the cathodic half reaction is: where OH − is transported from the cathode to the anode through the AEM, and e − is transported from the anode to the cathode through

Results and Discussion
In order to observe the nature of spontaneous reactions in DEFCs, fuel cell experiments were performed without application of an external voltage and without forcing an external current into the system.Fuel cell experiments were performed under different conditions.It was previously reported 13 that one way to promote the overall fuel reaction is to enhance the cathodic reaction.For that purpose, excess oxygen was supplied to the cathode of the fuel cells.Chemical reactions occur in the fuel cells and are typically promoted as the temperature increases.Figure 2 shows results of current measurements during fuel cell operation with 3% acetaldehyde (3% AAL, standard) and 3% ethanol (3% EtOH, standard) used as fuel.These experiments were conducted at 26.5 °C with the cathode being exposed to atmosphere without excess oxygen.A series of experiments was conducted at 26.5 °C with excess oxygen blown onto the cathode at 0.2, 0.4 and 0.6 l min −1 .Only the results measured at 0.6 l min −1 are shown in Fig. 2, as indicated by 3% AAL, O2 = 0.6 l min −1 and 3% EtOH, O2 = 0.6 l min −1 .A different series of experiments was performed at 40, 45 and 48 °C without excess oxygen.Only the results at 48 °C are shown in Fig. 2, as indicated as 3% AAL, temp = 48 °C and 3% EtOH, temp = 48 °C.
The results of current measurements are shown for different cases in Fig. 2. For all cases, the current increased quickly within approximately 10 min and then gradually decreased.The current for 3% acetaldehyde was ca. 10 mA.Although this value is not large, acetaldehyde could still be used as a fuel for these fuel cells.For 3% AAL, standard and 3% AAL, O 2 = 0.6 l min −1 , the decrease of the current at time = 10-180 min was not large, which indicates relatively stable operation of the cells under these conditions.Stable operation of 3% AAL, standard and 3% AAL, O 2 = 0.6 l min −1 in Fig. 2 may be related to negative ΔG (the value will be shown later) of the half reaction at anode.We think that the results of the current measurement for 3% AAL, standard and 3% AAL, O 2 = 0.6 l min −1 are almost the same.For 3% AAL, temp = 48 °C, the current was slightly high (about 14 mA) at time = 10-60 min.This could be due to a fact that temperature was higher than temp.= 26.5 °C of 3% AAL, standard, because higher temperature promotes chemical reaction.At longer time (90--180 min), the current dropped to about 10 mA level.Since the flow rate of fuel was 5.5 ml min −1 , it took about 60 min in average to circulate the fresh fuel.After 60 min of operation, fuel is circulating for the second time.Therefore, the products in the fuel could have affected the performance of the cell.When the values at time =150-180 min under three conditions are considered, the current with 3% acetaldehyde as fuel is not influenced by the experimental conditions.On the other hand, different behavior was observed with 3% ethanol as fuel.In all three cases, the current showed a maximum value at 10 min, decreased as time increased, and then became almost constant at 120 min or longer.As excess O 2 was blown onto the cathode, the current increased.An increase in temperature increased the current further.The current with 3% ethanol fuel was influenced by the experimental conditions; O 2 = 0.6 l min −1 and temp = 48 °C were factors that increased the current in the fuel cells.The higher current with 3% ethanol than that with 3% acetaldehyde is due to the larger number of anode half reactions in the former case.As we describe later, ΔG of most of anode half reactions ((1a), (3a), (4a), (5a), and (6a)) are positive.ΔG of only reaction (2a) is negative.That could be the reason why the trend is different from that of acetaldehyde fuel.
We now consider how the reaction products change.Figure 3 shows the concentration of acetic acid as a function of operation time for the 6 different experimental conditions shown in Fig. 2. For 3% acetaldehyde, the acetic acid concentration increased at a constant rate as the operation time increased.Even when the  experimental conditions were changed to provide excess O 2 or a higher temperature, the acetic acid concentration was almost identical to that for the standard conditions.The production of CO 2 is not significantly large; therefore, the independence of the acetic acid concentration for acetaldehyde with respect to the experimental conditions implies that Eq. 2 is not influenced by the experimental conditions.The results also indicate that the temperature dependence of Eq. 2 is very small.However, the situation for 3% ethanol is different.The concentrations of acetic acid and acetaldehyde were measured for 3% ethanol, and the results are shown in Figs. 3 and 4. The concentrations of both acetic acid and acetaldehyde increased at a constant rate as the operation time increased with ethanol as a fuel.The increase in acetaldehyde concentration was slower under the standard conditions and was faster with excess O 2 and at 48 °C.Eq. 1 is the only reaction that produces acetaldehyde; therefore, the results in Fig. 4 indicate that Eq. 1 proceeds faster when excess oxygen is supplied or when the temperature is increased.The concentration of acetic acid with 3% ethanol shown in Fig. 3 indicates that it increases at a slower rate under standard conditions, at a faster rate when excess oxygen is supplied, and at the fastest rate when the temperature is increased.For fuel cells that use 3% acetaldehyde, the rate of Eq. 2 is not influenced by excess oxygen or a higher temperature.Therefore, Eq. 2 was not used to produce acetic acid with 3% ethanol by excess oxygen or a higher temperature, and Eq. 4 is the only possible reaction that enhances the production of acetic acid from ethanol.Therefore, the experimental results in Fig. 3 confirm that ethanol changes directly to acetic acid under special conditions, such as when excess oxygen is provided and the temperature of the cells is increased to 48 °C.Antoniassi et al. 2 reported results that support this conclusion, where the presence of SnO 2 in the electrocatalyst contributed to the direct oxidation of ethanol to acetic acid.
To elucidate the role of oxygen, nitrogen gas (99.9%) was blown onto the cathode with 3% acetaldehyde.The nitrogen gas had a negative effect on both the current and the production of acetic acid, which implies that oxygen is required for Eq. 2 to proceed in acetaldehyde fuel cells.However, too much oxygen does not help the reaction to proceed faster.
A Nafion CEM was used as the electrolyte in this study; therefore, H + is the only cation that migrates from the anode to the cathode.The cathodic half reaction is thus Eq. 7.There are many possible anodic half reactions: Eqs.1a-6a.The Gibbs energy, ΔG, for the reactions was calculated using the Δ f G o values 14 listed in Table I.The values of ΔG in these half reactions are shown in Fig. 5 along with the ΔG values for the total reactions.The values of ΔG for the total reactions shown above the arrows are all negative, which indicates that these reactions are thermodynamically spontaneous.However, most of the values of ΔG for anodic reactions of the CEM shown below the arrows are positive.These anodic reactions with positive ΔG are not spontaneous and only occur because ΔG for their corresponding cathodic reaction (Eq.7) is negative (−237.1 kJ mol −1 ).Only Eq. 2a, the half reaction of Eq. 2, has a negative ΔG; therefore, this is the only anodic reaction that is thermodynamically spontaneous.This may be related to the experimental observation in Fig. 3, where the concentration of acetic acid in acetaldehyde fuel does not change as the experimental conditions change.Equation 4must proceed at a faster rate to have a large concentration of acetic acid in the ethanol fuel as the experimental condition changes.Conditions such as providing extra oxygen to the cathode or increasing the temperature to 48 °C contribute to oxidizing ethanol directly to acetic acid.Among the anodic reactions of the CEM with positive ΔG, Eq. 4a has the smallest positive ΔG value, which is probably the reason why Eq. 4 becomes active with excess oxygen and at higher temperature.
ΔG for anodic reactions (Eqs.1b-6b) when the AEM is used are shown by two lines below the arrows of Fig. 5.These ΔG values are all negative and these anodic half reactions are thus all thermodynamically spontaneous.This is different from the case when the CEM is used, in which most anodic reactions are not thermodynamically favorable.The ΔG values in Fig. 5 indicate that an AEM for the direct ethanol fuel cells is better than a CEM from a thermodynamic point of view because all the anodic half reactions are thermodynamically spontaneous.We performed calculations of ΔG when an AEM was used in direct methanol fuel cells (Supplementary material 1) and hydrogen fuel cells (Supplementary material 2).All possible anodic reactions in these fuel cells were thermodynamically spontaneous.Fujiwara et al. 15 prepared an AEM and made measurements of the fuel cells, and claimed that the cells show better performance than cells with a CEM.Their AEM requires a supply of an alkaline solution such as an aqueous KOH solution to compensate for insufficient OH − conductivity.For the development of DEFCs, it is thus important to produce a reliable AEM.

Conclusions
Spontaneous reactions in DEFCs were investigated under various conditions using acetaldehyde and ethanol as fuels.The oxidation reaction of acetaldehyde to acetic acid in acetaldehyde fuel is not sensitive to changes in the experimental conditions, while the concentration of acetic acid in ethanol fuel increases as oxygen is blown onto the cathode and as the temperature of the cells is increased.Therefore, we conclude that under certain experimental conditions, ethanol is oxidized directly to acetic acid, without oxidation to acetaldehyde and then to acetic acid.Since we are discussing thermodynamically spontaneous reaction, this conclusion is universal for other catalysts.The calculation of ΔG for the possible reactions in DEFCs suggests that a reliable AEM should be developed for the advancement of DEFCs.

Figure 3 .
Figure 3. Concentrations of acetic acid in DEFCs under various experimental conditions.

Figure 4 .
Figure 4. Concentrations of acetaldehyde in DEFCs under various experimental conditions.

Figure 5 .
Figure 5. Values of ΔG during ethanol oxidation in DEFCs with CEM and AEM.Values above the arrows are ΔG for the total reactions.Values below the arrows are ΔG for anodic reactions when CEM and AEM are used.Data shown in TableIwere used for the calculation.

Table I .
Standard Gibbs energy of formation. 14 f G o /kJmol −1