Enzyme ADH (E.C. 1.1.1.1), from Saccharomyces cerevisiae lyophilized powder (331 Units mg^-1); enzyme Laccase (E.C. 1.10.3.2.), from Trametes Versicolor lyophilized powder (21 Units mg^-1); coenzyme nicotinamide adenine dinucleotide hydrate (NAD⁺); 2,2´-azinobis(3-ethyl-2,3-dihydrobenzothazole-6-sulfonic acid (ABTS); pyrrole (PYR); and polyamidoamine generation 4 dendrimer (PAMAM) were purchased from Sigma-Aldrich in reactant grade. All reagents and enzymes were used without purification. Enzyme solutions were freshly prepared and rapidly immobilized on the carbon platform.

Multiwalled carbon nanotubes (MWCNTs) were acquired from Cheap Tubes Inc. (diameter of 8.0 nm, length of 10 to 30 mm, and > 95% purity). All solutions were prepared with high-purity water from a Millipore Milli-Q system. Solution pH was measured with a pH electrode coupled to a Qualxtron model 8010 pHmeter.

Electrochemical investigations were conducted on a Potentiostat/Galvanostat AUTOLAB PGSTAT 30 (EcoChemie, Netherlands) at room temperature, (25 ± 1) ^oC. M_lessEBFC experiments were carried out in a single-compartment cell (10 mL) by using 200 mmol L^-1 phosphate buffer solution (PBS) or acetate buffer solution (ABS) at pH 6.5, 1.9 mmol L^-1 NAD⁺, and 100 mmol L^-1 EtOH as electrolyte.

ADH bioanodes were prepared on carbon cloth platform (HT 1400 w, Elat CDL-Basf), to which MWCNTs and PAMAM were added. Further preparation details can be obtained elsewhere²⁰.To this end, polymethylene green (poly-MG) was used as immobilized electrocatalyst and was obtained by performing twelve successive voltammetric cycles (−0.3 to 1.3 V vs. Ag/AgCl). After that, a PAMAM anchoring matrix was directly deposited onto the electrode surface, obtained by pippeting 50 mL of 0.025 mol L^-1 commercial PAMAM solution. Then, 50 mL of MWCNT ink (1 mg of commercial MWCNTs dissolved in EtOH (400 mL) and 100 mmol L^-1 PBS (600 mL; pH 7.4) and sonicated for at least 4 h) was deposited. After drying, 50 mL of a freshly prepared enzyme solution containing ADH (300 mL; 3.37 mg L^-1) + NAD⁺ (100 mL; 7.50 mg L^-1) + PAMAM (100 mL; 0.025 mol L^-1) in PBS (100 mL; 100 mmol L^-1; pH 7.4) was cast on the electrode surface. The casting sequence carbon cloth/poly-MG/PAMAM/MWCNTs/ADH was named MWCNTs-ADH configuration.

The osmium [Os(bpy)₂Cl₂] and ruthenium complex [Ru(bpy)₂Cl₂] catalysts for laccase biocathode was synthesized according to the literature^22,23. [Os(bpy)₂Cl₂] or [Ru(bpy)₂Cl₂] entrapment in polyPYR films (1.5% v/v PYR, 8 mg mL^-1 [Os(bpy)₂Cl₂] or [Ru(bpy)₂Cl₂] mediator in 100 mmol L^-1 NaNO₃) was accomplished by chronoamperometry at a fixed potential (0.7 V vs. SCE; 10 min). The film containing polyPYR and [Os(bpy)₂Cl₂] or [Ru(bpy)₂Cl₂] mediator was named polyPYR-Os and polyPYR-Ru, respectively.

Laccase biocathodes, which employed [Os(bpy)₂Cl₂] or [Ru(bpy)₂Cl₂] as nanocatalyst and PAMAM dendrimer as enzymatic anchoring matrix, was prepared according to Cardoso and co-workers¹⁵. To this end, 50 mL of a stock solution, prepared by mixing laccase (300 mL; 3.32 mg mL^-1), PAMAM dendrimer (100 mL; 0.025 mol L^-1), and acetate buffer solution (ABS; 200 mL; 100 mmol L^-1; pH 4), was deposited on the previously modified carbon platform.

Figure 1 displays the representative structures of the bioelectrodes prepared on carbon cloth platforms.

Figure 1.
Bioelectrode side schematic representations: A) MWCNTs-ADH Bioanode; B) polyPYR-Os-laccase or polyPYR-Ru-laccase Biocathode.

Bioelectrodes were kept in their appropriate buffer solution (PBS (pH 7.4) for MWCNTs-ADH bioanode or ABS (pH 4.5) for polyPYR-Os-laccase or polyPYR-Os-laccase biocathodes) for at least 24 h before use. After electrochemical experiments, the respective bioelectrodes were kept refrigerated in buffer solution at 4 °C for at least 24 h before a further test was conducted.

pH influence on enzymatic kinetics was determined by assaying laccase and ADH activities at various pH values ranging between 3.5 and 10. To this end, the following 0.1 mol L⁻¹ buffer solutions were employed: acetate buffer (NaAc/HAc) for pH 3.5-5, phosphate buffer (NaH₂PO₄/Na₂HPO₄) for pH 6-7, and tris(hydroxymethyl)aminomethane–HCl (Tris+) buffer for pH 8–9. Reaction was initiated by adding substrate to the immobilized protein, depending on the study that was being performed.

Power density measurements were accomplished in an EtOH,O₂ M_lessEBFC described in the Instrumentation Section. First, the EtOH,O₂ M_lessEBFC open circuit voltage (OCV) was measured at least 1 h before the cell test. After that, polarization curves at a scan rate of 1 mV s^-1 were registered in triplicate. PD values for all M_lessEBFCs were obtained by multiplying cell voltage (E_cell) by current density (J_cell) (PD = E_cell × J_cell).

To obtain maximum M_lessEBFC performance, it is important to investigate bioelectrode enzymatic behavior as a function of pH because hydrogen ion concentration affects enzymatic activity: enzyme spatial conformation depends on pH values and on the presence of protonated/deprotonated groups in the enzyme catalytic site, which can modify the enzyme tertiary structure. pH also influences intrinsic/extrinsic electron transfer reaction strongly. The individual pH behavior of the immobilized enzymes employed here has previously been investigated in detail^15,20. The optimum pH range for ADH/NAD⁺ bioanode is between 7.0 and 8.0, achieved by employing PBS as buffer²⁴. Laccase works best in more acidic medium (pH 4.5), in ABS buffer solution¹⁵. Therefore, besides operating in different pH ranges, bioanode and biocathode also use distinct buffer solutions.

Direct correlation between enzymatic kinetics and pH is important to obtain maximum bioelectrode performance. Nevertheless, in a M_lessEBFC both bioelectrodes seldom operate at their optimum pH. To find the best pH for M_lessEBFC operation, individual pH curves of each enzyme were plotted together (Fig. 2). On the basis of Fig. 2, pH curves intersect at pH 6.5, which was further employed with M_lessEBFCs. Even though this pH value is close to physiological conditions, other complications may arise and diminish EtOH,O₂ M_lessEBFC PD and OCV values as compared to separated biofuel cells. Other factors may also be associated with this behavior such as problems with the enzyme-mediator-electrode electron transfer enzyme, which hinders the redox process underlying EtOH oxidation by ADH and O₂ reduction to H₂O by multicopper oxidase enzymes (laccase).

Figure 2.
pH influence on the enzymatic activity of bioelectrodes containing immobilized ADH (−■−) in PBS and immobilized laccase (−●−) in ABS.

Eliminating proton exchange membrane (PEM) has several advantages. During the reaction process, PEM is subjected to membrane channel obstruction by ions present in the supporting electrolyte, which dries or floods membrane parts, and fuel crossover. To minimize the aforementioned problems, one strategy is to remove the membrane when biocatalyst specificity can be maintained at M_lessEBFC. However, each bioelectrode must be evaluated for its electron transfer activity and enzymatic selectivity. EtOH,O₂ M_lessEBFC performance was assessed by analyzing OCV and power density curves obtained from polarization curves (results not shown). To investigate how PBS and ABS buffers influenced M_lessEBFC activity, EtOH,O₂ M_lessEBFC performance was measured at pH 6.5 in both buffers (ABS and PBS; buffer concentration = 200 mmol L^-1). This “precaution” was necessary because PEM removal resulted in each enzyme facing a medium that was different from its ideal operation condition (ABS (pH 4.5) for laccase and PBS (pH 7.4) for ADH).

Figure 3 shows the results for M_lessEBFCs containing a MWCNTs-ADH anode and one of the following cathode configurations: polyPYR-Os-laccase or polyPYR-Ru-laccase; either ABS or PBS at pH 6.5 was employed. According to Fig. 3 M_lessEBFC in PBS buffer operated at higher power density (PD) and current (J_cell(max)) as compared to M_lessEBFC in ABS buffer. In PBS buffer (Fig. 3A), PD and J_cell(max) respectively were 21.0 ± 0.2 mW cm^-2 and 0.15 ± 0.07 mA cm^-2 for MWCNTs-ADH, polyPYR-Os-laccase, and 13.5 ± 0.2 mW cm^-2 and 0.11 ± 0.05 mA cm^-2 for MWCNTs-ADH,polyPYR-Ru-laccase. When these systems were switched to ABS buffer, both PD and J_cell(max) decreased significantly (Fig. 3B). For this reason, PBS was selected as buffer for further EtOH,O₂ M_lessEBFC investigations.

Figure 3.
Power density curves for EtOH,O₂ MlessEBFCs employing MWCNTs-ADH as bioanode and (−♦−) polyPYR-Os-laccase or polyPYR-Ru-laccase (−○−) as biocathode in (A) PBS and (B) ABS medium (200 mmol L^-1; pH 6.5, 1.9 mmol L^-1 NAD⁺, 100 mmol L^-1 EtOH, n = 3).

ABTS is one of the most common oxygen reduction mediators when laccase is employed in EBFCs. Figure 4 illustrates how ABTS influences EtOH,O₂ M_lessEBFC performance in PBS medium (pH 6.5) for both anode configurations investigated here: polyPYR-Os-laccase and polyPYR-Ru-laccase.

Figure 4.
Power density curves for MWCNTs-ADH,polyPYR-Os-laccase (−♦−) and MWCNTs-ADH,polyPYR-Ru-laccase (−○−) in the presence of 1 mmol L^-1 ABTS (200 mmol L^-1 PBS (pH 6.5), 1.9 mmol L^-1 NAD⁺, and 100 mmol L^-1 EtOH; n = 3).

Homogeneous ABTS introduction diminishes Ru-mediated cathode cell PD by over 80% as compared to Os-mediated cathode cell. Indeed, PD decreased from 15.3 ± 0.2 mW cm^-2 to 2.7 ± 0.3 mW cm^-2 just by changing the Os complex to the Ru complex. Also, when results obtained in the absence (Fig. 3A) and in the presence (Fig. 4) of ABTS are compared for Os-complex in PBS, PD and J_cell(max) was approximately 27.5 and 23.8%, respectively. This decrease could be explained by competition between ABTS and mediators incorporated into the polyPYR matrix and the enzymatic redox sites. The best results were achieved for EtOH,O₂ M_lessEBFCs based on the MWCNT-ADH,polyPYR-Os-laccase system. These results agreed with literature data²⁵ claiming that [Os(bpy)₂Cl₂] can bind to laccase copper hydrophobic T1 active site and establish a strong electrostatic interaction, which entraps the Os-complex in polyPYR, shifts the polyPYR-Os oxidation potential (E_oxi), and facilitates electron transfer. Our results showed that polyPYR-Ru-laccase did not interact in the same way as the Os-mediator.

Table 1 summarizes all experimental parameters obtained for EtOH,O₂ M_lessEBFCs as a function of the different redox mediators entrapped in polyPYR, in the absence or presence of 1 mmol L^-1 ABTS.

Table 1.
Parameters obtained for different EtOH/O₂ M_lessEBFCs.

On the basis of the results above (Table 1), the best EtOH,O₂ M_lessEBFCs was MWCNTs-ADH,polyPYR-Os-laccase in 200 mmol L^-1 PBS (pH 6.5) in the absence of ABTS. Figure 5 illustrates the selected operation system.

Figure 5.
Schematic diagram of mediated electron transfer in EtOH,O₂ M_lessEBFCs based on MWCNTs-ADH,polyPYR-Os-laccase.

We have investigated and reported half-cell data for these electrodes configuration before^15,21. For the biocathode half-cell¹⁵ a gas diffusion membrane (ELAT) consisting of 40% metal in C (Pt0.66Ru0.34, E-TEK commercial mixture) hot pressed in a Nafion NRE-212 membrane was employed as the anode. This configuration furnished at least five times higher power densities values than the value reported for the membraneless fuel cell. For the half-bioanode²¹ Pt was used as cathode, also separated by a Nafion membrane. This configuration furnished power density values as high as 0.25 mW cm^-2. Table 1 shows that the results for EtOH,O₂ M_lessEBFCs are much lower than the separated compartment cell indicating that besides pH effect must be a mutual influence of the fuel and O₂ in the performance of the enzymatic systems. Nevertheless, despite differences with respect to enzyme immobilization method, the values measured herein are in the same order of magnitude (mW cm^-2) with some data reported in the literature data^5,8,27,28. Considering the high efficiency in ethanol/acetaldehyde conversion and concomitant O₂ reduction to H₂O demonstrated by enzymatic systems future application of biofuel cells in miniaturized systems must be solved preparing microfluidic devices that operate with streams of liquid electrolytes²⁹. In this configuration it is possible to operate with different electrolytes for anodic and cathodic compartments without any problem. This is our future goal to increase the power density.

Table 2 lists the MWCNTs-ADH,polyPYR-Os-laccase cell storage lifetime under optimum conditions (200 mmol L^-1 PBS (pH 6.5), 1.9 mmol L^-1 NAD⁺, and 100 mmol L^-1 EtOH).

Table 2.
EtOH,O₂ M_lessEBFCs storage lifetime for polyPYR-Os-Laccase and MWCNTs-ADH bioelectrodes.

After five months, PD and J_cell(max) decreased by approximately 38% (13 ± 4) mW cm^-2 and 0.09 ± 0.02 mA cm^-2, respectively) as compared to freshly prepared electrodes. These values dropped slowly and reached 62% and 48% (8 ± 2 mW cm^-2 and 0.09 ± 0.02 mA cm^-2, respectively) of the initial values. These results attested that immobilization of the enzymes employed here provided a relatively stable medium for long-term storage tests. This result may be important to apply these devices in new types of nanofluid cells to enhance the power harvest from the ethanol molecule³⁰.

ardandra@ffclrp.usp.br