Soluble viologen polymers as carbohydrate oxidation catalysts for alkaline carbohydrate fuel cells

Soluble viologen polymers as carbohydrate oxidation catalysts for alkaline carbohydrate fuel cells

Accepted Manuscript Soluble viologen polymers as carbohydrate oxidation catalysts for alkaline carbohydrate fuel cells Connor R. Rigby, Haesook Han, ...

645KB Sizes 0 Downloads 44 Views

Accepted Manuscript Soluble viologen polymers as carbohydrate oxidation catalysts for alkaline carbohydrate fuel cells

Connor R. Rigby, Haesook Han, Pradip K. Bhowmik, Meisam Bahari, John N. Harb, Randy S. Lewis, Gerald D. Watt PII: DOI: Reference:

S1572-6657(18)30362-X doi:10.1016/j.jelechem.2018.05.016 JEAC 4072

To appear in:

Journal of Electroanalytical Chemistry

Received date: Revised date: Accepted date:

4 January 2018 11 May 2018 14 May 2018

Please cite this article as: Connor R. Rigby, Haesook Han, Pradip K. Bhowmik, Meisam Bahari, John N. Harb, Randy S. Lewis, Gerald D. Watt , Soluble viologen polymers as carbohydrate oxidation catalysts for alkaline carbohydrate fuel cells. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Jeac(2017), doi:10.1016/j.jelechem.2018.05.016

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Soluble Viologen Polymers as Carbohydrate Oxidation Catalysts for Alkaline Carbohydrate Fuel Cells.

IP

CR

S. Lewis1 and Gerald D. Watt3*

T

Connor R. Rigby1, Haesook Han2, Pradip K. Bhowmik2,Meisam Bahari1, John N. Harb1, Randy

Department of Chemical Engineering, Brigham Young University, Provo, UT 84602,

2

Department of Chemistry, University of Nevada at Las Vegas, 4505 S. Maryland Parkway, Box

AN

US

1

454003, Las Vegas, NV 89154 and 3Department of Chemistry and Biochemistry, Brigham Young

ED

M

University, Provo, UT 84602.

PT

*Send inquiries to GDW. Tel 801 369-6338; FAX 801 422-0158; email: [email protected]

CE

Abbreviations: MV and MMV are dimethyl and monomethyl viologen, respectively. Po and Pr are the oxidized and reduced states of viologen polymers.

AC

Keywords: Viologen polymers; carbohydrate fuel cell; kinetics; fuel cell catalysts.

Acknowledgements. Funding by the College of Physical and Mathematical Science at Brigham Young University and the National Science Foundation (CBET 1540537) is gratefully acknowledged.

1

ACCEPTED MANUSCRIPT Abstract Monomeric dialkyl and monoalkyl viologens rapidly catalyze O2-oxidation of carbohydrates in alkaline solution to form carbonate and formate at pH >11. Optimization of these catalytic reactions is important for alkaline carbohydrate fuel cell applications. Use of these monomeric

T

viologens is not optimal because they exit the fuel cell with the reaction products and would need

IP

to be recovered and recycled for continued use. Polymeric or immobilized viologens would

CR

reduce or avoid these recovery problems. Here, we report the evaluation of water-soluble polymeric viologens as potential fuel-cell-catalysts and show that they rapidly catalyze O2-

US

oxidation of carbohydrates to carbonate and formate at high efficiency. The rates of reduction of

AN

polymeric viologens by carbohydrates were rapid and found to be first order in carbohydrate and one-half order in polymer concentration. An unusual feature of the reduction reaction is that the

M

rate of polymer reduction is two-fold faster if carbohydrate is equilibrated with the buffer prior to

ED

addition of the catalyst rather than adding the carbohydrate in the presence of the catalyst.

AC

CE

PT

Advantages and disadvantages of polymeric viologens as fuel cell catalysts are discussed.

2

ACCEPTED MANUSCRIPT 1.0. Introduction. The potential for electricity production via viologen-catalyzed alkaline carbohydrate fuel cells has recently been explored [1-7] and shows promise of increased efficiencies over those possible by conversion of the initial carbohydrate energy into ethanol or other liquid fuels for use in internal combustion engines [8-9]. Aside from the higher

T

efficiency of converting glucose energy into electricity is the promise that electricity can be

IP

produced from a wider variety of carbohydrate monomers obtained from both the cellulose

CR

and hemicellulose content of biomass [10-12]. Preliminary reports [1, 7] and a more recently detailed kinetic study [13] have demonstrated that viologens rapidly catalyze extensive

US

oxidation of a variety of carbohydrates at rates commensurate with promising power

AN

generation from alkaline fuel cells.

Previous fuel cell experiments were conducted with homogeneous viologen catalysts [1-

M

7]. However, to enhance efficiency polymeric viologen catalysts offer advantages over their

ED

monomeric viologen counterparts [3, 7, 13]. Here, we report the results of experiments designed to further evaluate the reactivity of viologen polymers as catalysts for alkaline-carbohydrate fuel

PT

cells. Polymeric viologens were found to catalyze extensive oxidation of carbohydrates to

CE

carbonate and formate at >80% efficiency. Rates similar to those found for monomeric viologens were also observed [7]. The results show that viologen polymers can be effective alkaline fuel

AC

cell catalysts at pH 11-12 and 50 C but become unstable at higher pH values and temperatures. 2.0. Materials and Methods The viologen polymers listed in Table 1 were prepared as previously reported [14-16].

3

ACCEPTED MANUSCRIPT

Chemical Structures

Codes

OTs

A

N CH2

H2C N OTs

OTs

N

N CH2

O O S O

CH2

IP

B

OTs =

T

OTs

OTs

OTs

N CH2

OTs

N

M ED

OTs

OTs N n

CE

PT

E

n

N CH2

N

D

CH2

AN

OTs

n

US

N

C

CH2

CR

n

OTs

OTs

O

N

N

n

AC

F

OTs

G

OTs N

N

4

ACCEPTED MANUSCRIPT Table 1. Chemical Structures of Viologen Polymers, Related Model Compounds and Their Codes used in Figures 1 and 2. O2/carbohydrate molar ratios for polymer-catalyzed carbohydrate oxidation were measured by two methods. The vial method (1) consisted of placing 1.0 mL of 0.50 M phosphate

T

buffer pH 11, 10-50 µL of 0.50 M carbohydrate and 100-500 µL of 0.50 M viologen in a 15

IP

mL vial containing air and sealed with a serum cap. The MV/carbohydrate ratio was

CR

typically 10. The reaction was rapidly stirred at 50 C to facilitate air oxidation of reduced MV produced during MV oxidation of the carbohydrate. When the reaction was concluded

US

(2-3 hrs), the vial was cooled to room temperature, weighed and the septum was punctured

AN

under water with an 18 gauge needle and water filled the vacuum produced by O2-uptake during carbohydrate oxidation. The vial was reweighed and the grams of water-uptake

M

(equivalent to the mL of O2 consumed) was used to determine the O2/carbohydrate Identical reactions were conducted using

ED

stoichiometric ratio for each carbohydrate studied.

a MicroLab pressure sensor (Bozeman, Mt) to directly measure the pressure change as O2 is

PT

consumed by the catalyzed reaction [1]. Both gave identical results but the vial method

CE

allowed larger numbers of reactions to be conducted simultaneously. 13C NMR (Varian Nova 300) analysis was used to determine the principal products of polymer-catalyzed

AC

carbohydrate oxidation. Control reactions with the polymers were conducted aerobically in the absence of carbohydrate to evaluate polymer stability and appropriate corrections were applied for the small amounts of polymer decomposition that was observed. These controls established that the carbonate and formate resulted from the catalyzed reaction and not from polymer decomposition. For those slightly unstable polymers, their decomposition products did not produce formate or carbonate, establishing that these products resulted from polymer

5

ACCEPTED MANUSCRIPT catalyzed carbohydrate oxidation. Reactions were not conducted above pH 12 because polymer decomposition slowly occurs above this pH. Kinetic studies were conducted at 50 C and at pH 11 in 0.50 M phosphate buffer by measuring the increase of absorbance at 520-540 nm (ε ~ 8300 M-1cm-1 for all reduced

T

polymers) as a function of time during polymer reduction by various carbohydrates. Molar

IP

absorptivities of the singly reduced monomer units comprising the reduced polymer Pr were

CR

determined independently by anaerobic titration with standard sodium dithionite. Polymers reduced by sodium dithionite or by various carbohydrates were rapidly and reversibly

US

oxidized when exposed to O2 at rates similar to their monomeric counterparts.

AN

The sequence of addition of the various reagents and incubation times of the reagents prior to initiating the reaction were both found to influence the reaction kinetics.

M

Consequently, two different procedures were followed. In the first, 2.0 mL of anaerobic

ED

buffer was equilibrated at 50 C in an anaerobic optical cell and then 0.50 M carbohydrate was added and the reagents incubated an additional 2-5 min. The reaction was then initiated by

PT

addition of anaerobic 0.02 M polymer (buffer + carbohydrate → polymer) under vigorously

CE

stirred conditions to minimize any mass transport problems with the polymers. The second procedure reversed the process by equilibrating buffer and the polymer solution, incubating

AC

for 2-10 min and then initiating the reaction by adding the carbohydrate solution (buffer + polymer → carbohydrate). For both procedures, the polymer and carbohydrate concentrations were varied 10-fold to determine the reaction orders. Cyclic voltammetry measurements were conducted in 5 ml of 0.1 M KCl, at pH 9 using a gold electrode with 0.196 cm2 area with an Ag/AgCl as reference electrode. Identical measurements were conducted using a Pt electrode. The electrolyte solution was sparged

6

ACCEPTED MANUSCRIPT with nitrogen for 30 minutes and the solution was kept under a gentle flow of nitrogen to maintain anaerobic conditions. A minimum of 10 scans were recorded with the eighth scan reported in Figure 4 below. Electrochemical oxidation of DHA and other carbohydrates catalyzed by viologen

T

polymers was conducted in an electrochemical cell consisting of a platinum gauze working

IP

electrode and an Ag/AgCl reference electrode contained in an argon-filled Vacuum

CR

Atmospheres glove box. The electrochemical cell contained 5.0 mL of 0.50 M phosphate buffer pH 11 and 5 mM polymer monomeric unit. The cell was set at 0 mv and when the

US

current stabilized near zero, DHA or other carbohydrate was added to 2.5 mM. The resulting

AN

current-time curve was then integrated to determine the number of electrons removed from the carbohydrate by polymer catalyzed electrochemical oxidation.

M

3.0. Results.

ED

3.1.O2/Carbohydrate Mole Ratios

Reaction 1 represents oxidation of glucose conducted by the oxidized form of each viologen

PT

polymeric unit (Po) to produce the reduced form (Pr). Reaction 2 is the rapid oxidation of Pr

CE

by O2 to regenerate Po. Reactions 1 and 2 are the idealized component reactions that combine to produce the overall catalytic oxidation of glucose (Reaction 3) to form carbonate and water

AC

catalyzed by the polymer.

C6H12O6 (Glucose) + 24 Po + 36 OH- = 24 Pr + 6CO32-+ 24 H2O (1) 24 Pr + 6O2 +12 H2O = 24 Po + 24 OH-

(2)

C6 H12O6 + 6O2 + 12OH- = 6 CO32- + 12H2O (3) Oxidation of other carbohydrates follow a similar reaction sequence but the number of carbons ranges from 6-3 with the O2 requirement also varying with the number of carbon atoms. 7

ACCEPTED MANUSCRIPT Table 2 contains measured O2/carbohydrate mole ratios for the indicated polymeric viologens catalyzing the oxidation of representative 3, 5 and 6 carbon carbohydrates at pH 11 and 50 C.

Table 2. O2/carbohydrate Mole Ratios for Viologen Polymer-Catalyzed Oxidation of

IP

T

Selected Carbohydrates by Air at pH 11 and 50 C.

A

B

C

D

DHA (3) Xylose (5) Arabinose (5) Ribose (5) Glucose (6)

2.7d 3.8 4.2 3.4 4.8

2.9 4.1 4.4 3.8 5.6

2.6 -3.9 4.2 --

3.5 -4.0 4.9 --

Fc

G

Theory

3.0 3.5 4.5 6.5 5.2

3.1 3.9 2.9 4.5 4.7

3.4 4.8 4.4 4.1 3.9

3.0 5.0 5.0 5.0 6.0

AN

M

ED

See Table 1 for structures of A-G. A and G are monomers not polymers. The number in parentheses indicates the number of carbon atoms in the carbohydrate. The polymer is only slightly soluble in water and was run as a suspension. The uncertainty in the values is typically  0.20

PT

a. b. c. d.

Ec

US

Carbohydrateb

CR

Viologen Polymera

The last column is the expected results for O2-oxidation of carbohydrates to carbonate and

CE

water and assumes that each O2 requires four electrons for reduction. Glucose, xylose and

AC

arabinose were studied in most detail because they are the most abundant carbohydrates found in the cellulosic and hemicellulosic constituents of biomass and represent the most abundant carbohydrates for fuel cell use. For the carbohydrates shown, O2/carbohydrate values of ~80% of the theoretical oxidation stoichiometry were typically observed, indicating the viologen polymers effectively catalyze extensive carbohydrate oxidation. This was verified by control reactions showing that non-catalyzed carbohydrate oxidation in the absence of polymer represented < 5% of the measured O2 uptake for the catalyzed reactions. 8

ACCEPTED MANUSCRIPT The lower than theoretical oxidation conversion of the carbohydrates results from incomplete carbohydrate oxidation, forming a mixture mostly of carbonate, formate and with small amounts of organic acids. Polymer D and monomer G gave values ~20% larger than the theoretical for DHA

T

oxidation and attempts were made to understand this behavior. Control reactions in the

IP

absence of carbohydrate were small and when applied still gave results larger than

CR

theoretical. For the oxidation conditions (air), H2O2 could form instead of H2O and give higher O2/DHA values but direct measurement for H2O2 after the 20 min DHA reaction

US

period showed <1% formation because, as shown in independent experiments, H2O2 rapidly

AN

oxidizes Pr. O2 uptake by the polymers with formation of N-oxides or other oxidized species is possible but no evidence for this was observed. The reduced form of the polymer might be

M

more susceptible to O2 modification than the oxidized form used for the control reactions and

ED

could undergo reaction with O2 (possibly O2-, superoxide ion) in a destructive manner. But

remains unexplained.

PT

no evidence for this was observed. The anomalous O2 uptake for these two DHA reaction

CE

3.2. Kinetics for Viologen Oxidation of Carbohydrates The anaerobic reaction of arabinose with polymer B follows a reaction similar to that

AC

shown in Figure 1 for glucose and represents complete reduction of polymer B by arabinose. Exposure to air causes Pr to undergo rapid oxidation according to Reaction 2 to regenerate Po. Arabinose was studied because it along with xylose are the most abundant carbohydrates found in hemicellulose and represent the fastest reacting carbohydrates studied but other carbohydrates react similarly (see Figures 2 and 3 below). Figure 1 was obtained by first equilibrating arabinose with buffer for 5.0 min and then starting the reaction by adding

9

ACCEPTED MANUSCRIPT polymer B (buffer + arabinose → polymer B). The same behavior was observed for polymers C and D. The reaction rapidly begins with a lag phase of ~30 s and gives a linear initial rate profile. From variation of the initial rates with variation of polymer and carbohydrate concentration, a one-half order reaction in polymer and a first order reaction in

T

arabinose was determined. The inset shows the analysis of the complete reaction profile and

IP

confirms the one-half order polymer dependence obtained from the initial rates. A 5.0 min

CR

incubation of arabinose with buffer gave maximal results because longer incubation initiated nonspecific arabinose O2-oxidation, whereas shorter times was not sufficient for arabinose to

US

become “activated”. Under these conditions, rate constants of 0.0088 and 0.0097 M1/2s-1were

AN

determined for arabinose and xylose reacting with polymer B. When the sequence of reagent addition is reversed (buffer + polymer B → arabinose) a rapid initial rate was also observed

Simply reversing the sequence of reagent addition causes the rate to

change by a factor of ~2.

ED

~1/2 that in Figure 1.

M

with one-half order dependence in polymer and first order in arabinose but the rate was only

PT

Figure 2 shows the rate of reaction of mannose and glucose (the slowest reacting

CE

carbohydrates) with polymer F and shows that the reaction does not begin immediately but each reaction has a lag period of ~100 s before reaction begins. The more slowly reacting mannose

AC

has a slightly longer lag period than that of glucose but the lag period for both can be eliminated at higher polymer and carbohydrate concentrations or at the same concentration by an increase in temperature.

10

IP

T

ACCEPTED MANUSCRIPT

US

CR

Figure 1. The reaction of polymer B with arabinose at pH 11.0 and 50 C. The inset is a plot of the square root of the absorbance difference at 540 nm against time, confirming the one-half order dependence in polymer concentration. Once the lag period is over, reaction begins abruptly with little or no transition phase,

AN

forming a near linear rate of polymer F reduction by the two carbohydrates. The initial part of the progress curve was used for determining the reaction orders of the reactants and the overall

M

rate of polymer reduction by the carbohydrate. The lag phase and the effect of reversing reagent

ED

addition seems to be chemical in nature rather than mass transfer problems associated with

AC

CE

PT

polymer reactivity.

Figure 2. Reaction of glucose and mannose with polymer F at pH 11 and 50 C.

11

ACCEPTED MANUSCRIPT

Measuring the initial slopes of the reactions as both the polymer and carbohydrate concentrations were varied over a factor of 4-10 yields rate law (4) that is first order in carbohydrate and ½-order in polymer as was the case for arabinose above. Rate constants of

IP

Rate = k [carbohydrate] [polymer]1/2 (4)

T

0.0054 and 0.0031 M-1/2s-1 for glucose and mannose were determined.

CR

Figure 3 show the reactivity of the other polymers with mannose and glucose under the same conditions. The reactions all give nearly linear rate profiles at 0.50 mM polymer monomeric unit

US

concentration and 25 mM carbohydrate concentration from which rate constants ranging from

AN

0.0023-0.0058 M-1/2 s-1 were obtained, indicating that all carbohydrates derived from cellulose and hemicellulose undergo rapid polymer-catalyzed reactions at very similar rates.

M

A and G are monomers not polymers but their rates of reaction with the various carbohydrates

AC

CE

PT

ED

are nearly identical with their monomeric units contained in polymers.

Figure 3. The Reaction of the Viologen Compounds and Polymers in Table 1 with Glucose and Mannose at pH 11 and 50 C. The lag phase of ~50 s for glucose reacting with polymer B was set to zero in order to compare the lag phases of the other polymers. 3.3.Electrochemical Oxidation of Carbohydrates 12

ACCEPTED MANUSCRIPT Figure 4 shows cyclic voltamograms using a gold electrode of selected polymers from Table 1 and shows that the reduction potentials are near -400 to -500 mV. Similar results were found using a platinum electrode. These results suggest that the reduced viologen polymer

M

AN

US

CR

IP

T

formed by carbohydrate oxidation will undergo oxidation at a metallic

ED

Figure 4. Cyclic voltamograms of (13 mM) polymers B (blue), D (red) and E (green) at pH 9.0

PT

using a gold electrode. Scan rate: 50 mV/s, room temperature

CE

electrode poised at a potential more positive than -400 mV and will set up conditions to electrocatalytically oxidize the carbohydrate. Figure 5 shows that addition of DHA to an

AC

electrochemical cell containing polymer B poised at 0 mV immediately forms reduced polymer with increased current flow. This result shows both

13

IP

T

ACCEPTED MANUSCRIPT

CR

Figure 5. Electrochemical Catalytic Oxidation of DHA by Polymer B. The reaction was conducted in 5.0 mL of 0.50 M phosphate buffer pH 11.0 containing 5.0 mM polymer B to which DHA was added to of 2.5 mM. The integrated curve corresponds to 9.2% DHA oxidation.

US

simultaneous polymer reduction by carbohydrate and immediate electrochemical oxidation of the

AN

newly formed reduced polymer leading to catalytic oxidation of DHA. Addition of DHA to the electrochemical cell containing no polymer produced no current, demonstrating that polymer is

M

essential for catalytic electrochemical DHA oxidation. Integration of the current-time curve in

ED

yielded only a 9.2 % (1.1 electron) electrochemical oxidation of the 12 electrons available in DHA. Inspection of the electrochemical cell showed its contents were dark red and upon

PT

standing produced a dark red precipitate of reduced polymer, which was not electroactive. The

CE

insolubility of the reduced polymer can be understood because the oxidized form of each monomeric viologen unit of the polymer has a 2+ charge but upon a one electron reduction the

AC

charge decreases to 1+, decreasing overall polymer solubility. The results establish the important result that polymers can serve as catalysts for catalytic electrochemical oxidation of carbohydrates but that the low solubility of the reduced forms is an obstacle for large-scale carbohydrate oxidation. 4.0. Discussion.

14

ACCEPTED MANUSCRIPT The feasibility of using viologen catalysts for fuel cell operation has been demonstrated using monomeric [1-5] and polymeric [3] viologens as carbohydrate oxidation catalysts but monomeric viologens are not ideally suited for fuel cell use because they would flow from of the anodic chamber and would have to be isolated and reintroduced into the fuel cell. The insoluble

T

polymers previously studied were active catalysts and were 25% more efficient than their

IP

monomeric counterparts but their heterogeneous nature in solution made proper catalytic

CR

assessment difficult. The soluble polymeric viologen catalysts reported here are as effective as the insoluble polymers and are also ~25% more efficient than monomers in catalyzing extensive

US

O2-oxidation of carbohydrates at pH values of 11 or 12 at 50 C. The soluble viologen polymers,

AN

in contrast to their insoluble counter parts, are redox active at metallic electrodes, where the reduced viologen polymers at low concentration readily transfer electrons to and from electrodes.

M

Polymer mediated electrochemical oxidation of carbohydrates is an important property and

ED

represents the reaction that would occur in the anodic compartment of a fuel cell. The results in Figures 4 and 5 establish that polymeric viologens present the opportunity to function as

PT

catalysts in an alkaline fuel cell to generate electrical energy from a wide range of carbohydrates

CE

but the insolubility of the reduced form is a serious limitation. In their present form, the soluble polymeric viologens offer promising advantages for use as catalysts for alkaline fuel cells but

AC

they also present disadvantages that must be overcome. The first is that the polymeric viologens have limited stability at pH > 12 and hydrolytically decompose to poorly defined products that are only weakly catalytically active. Similar behavior also occurs for monomeric dialkyl viologens but in contrast to the polymers, the mono alkyl viologens formed by hydrolysis have increased pH stability >12 and remain catalytically active. A greater voltage output from an alkaline fuel cell increases with pH, which

15

ACCEPTED MANUSCRIPT in turn increases the power output, but this advantage will not accrue using polymeric viologens of the type that we have studied because of their instability above pH 12. Stabilizing viologen polymers against pH degradation is an important research area to enhance catalysts for alkaline fuel cell development.

T

A second problem with using polymers of the type in Table 1 is that they have low

IP

solubility of 5-100 mM, which is a limitation to fuel cell power production. This is especially

CR

the case for polymer E and F and for compound G that all have a high aromatic content, which decreases their solubility. An even more serious problem is encountered when the polymers are

US

reduced because their solubility decreases significantly. Variation of the counter ion and

AN

developing appropriate synthetic procedures should enhance solubility to overcome this limitation.

M

Figure 4 shows that all of the polymers give well-defined cyclic voltamograms and that

ED

reduced viologen polymers produced by carbohydrate reduction will be oxidized by metallic fuel cell electrodes. Attempts to demonstrate this reactivity at higher concentrations than shown in

PT

Figure 4 were only partially successful as shown in Figure 5 because the reduced form of the

CE

polymer precipitated and was not electroactive. The lack of electrochemical activity of the insoluble, reduced polymer is a limitation for polymer use as fuel cell catalysts and ways to

AC

increase its solubility are needed. Another problem is also presented in Figure 4. Carbohydrates at pH 11 produce a potential near 1200 mV [17], whereas the polymers accept electrons at ~400-500 mV. A loss of ~800 mV would occur during fuel cell operation, which lowers the power produced from carbohydrate oxidation. To make polymers more efficient in transferring the chemical energy in carbohydrates to electrical energy, the potential of the polymer must be changed to approach

16

ACCEPTED MANUSCRIPT 1200 mV produced by the carbohydrate. The high aromatic content of the polymers in Table 1 is an important factor responsible for the unfavorable polymer solubility as well as the more positive redox potential. Ways to modify its content or attaching groups to the viologen rings that lower the potential and increase solubility would improve polymer characteristics.

T

Observing the one-half order dependence on polymer concentration during polymer

IP

reduction (carbohydrate oxidation) is surprising and unexpected. A mechanistic understanding

CR

was not found but the one-half order behavior must be a consequence of multiple monomeric viologen units connected in a long polymer chain. This has the advantage of increasing the local

US

concentration of viologen units [3] and is advantageous in oxidizing carbohydrates as extensively

AN

as shown in Table 2. It is possible that the one-half order behavior is a consequence that carbohydrate reduction occurs simultaneously and independently at each end of the long polymer

M

chain, giving a 2/1 carbohydrate/polymer ratio required for a one-half order reaction.

ED

Understanding this kinetic behavior by examining polymeric viologens with different structures

CE

PT

and properties may produce important insights into polymer reactivity for fuel cell use.

5.0.Conclusions

AC

Soluble viologen polymers that were investigated have similar redox potentials and rapidly oxidize carbohydrates forming the reduced polymer and demonstrate their potential utility as effective carbohydrate oxidation catalysts for use in alkaline carbohydrate fuel cells. All carbohydrates found in the cellulose and hemicellulose fractions of biomass react rapidly with the polymers studied providing a wide range of carbohydrate fuel options. While promising results have been established, application of viologen polymers to creating

17

ACCEPTED MANUSCRIPT efficient fuel cells has identified limitations which must be overcome. Developing modified viologen polymers through synthetic means is a promising avenue of research to increase solubility, stability and produce more favorable redox potentials to produce more effective

AC

CE

PT

ED

M

AN

US

CR

IP

T

viologen polymers to function as carbohydrate oxidation catalysts.

18

ACCEPTED MANUSCRIPT 6.0. References 1. D. R. Wheeler, J. Nichols, D. Hansen, M. Andruss, S. Choi, G. D. Watt. Viologen catalyst for a direct carbohydrate fuel cell. J. Electrochem. Soc. 146: B (2009) 146: B1201-7.

T

2. D. Scott, B. Y. Liaw. Harnessing electric power from monosaccharides-a carbohydrate-

G. D. Watt, D. Hansen, D. Dodson, M. Andrus, D. Wheeler. Electrical energy from

CR

3.

IP

air alkaline fuel cell mediated by redox dyes. Energy Environ Sci. (2009) 10-15.

Renew Energy. 36 (2011) 36 1523-1528.

US

carbohydrate oxidation during viologen-catalyzed O2-oxidation: Mechanistic insights.

AN

4. X. Liu, M. Hao, M. Feng, L. Zhang, Y. Zhao, X. Du, G. A. Wang G. One-compartment

Energy. 106 (2013) 176-83.

M

direct glucose alkaline fuel cell with methy viologen as electron mediator. Applied

ED

5. M. Hao, X. Liu, M. Feng, P. Zhang, G. Wang. Generating power from cellulose in an alkaline fuel cell enhanced by methyl viologen as an electron-transfer catalyst. J. Power

PT

Sources. 251 (2014) 222-228.

104.

CE

6. G. D. Watt, G. A new future for carbohydrate fuel cells. Renew. Energy. 72 (2014) 99-

AC

7. G. D. Watt Kinetic evaluation of the viologen-catalyzed carbohydrate oxidation reaction for fuel cell application. Renew Energy. 63 (2014) 370-375. 8. J. E. Campbell, D. B. Lobell, C. B. Field CB. Greater transportation energy and GHG offsets from bioelectricity than ethanol. Science. 314 (2009) 1055-7. 9. J. Ohlrogge, A. Allen, D. Della Penna, Y. Shachar-Hill, S. Styme. Driving on biomass. Science 32 1019-22. 19

ACCEPTED MANUSCRIPT 10. D. Pimentel, T. W. Patzek. Ethanol Production Using Corn, Switchgrass, and Wood; Biodiesel Production Using Soybean and Sunflower. Natural Resources Research 14 (2005) 65-76. 11. M. Balat. Production of bioethanol from lignocellulosic materials via the biochemical

T

pathway: A review. Energy conversion and Management. 52 (2011) 858-875.

IP

12. M. Mascal, E. B. Nikitin. Direct, high-yield conversion of cellulose into biofuel. Angew

CR

Chem Int Ed (2008) 7924-6.

13. C. Rigby, M. Bahari, J. Harb, R. Lewis, G. D. Watt. Viologen-Catalyzed Oxidation of

US

Carbohydrates and Application to Alkaline Fuel Cell Development. To be submitted.

AN

14. P. K. Bhowmik, P.K. Han, H. Nedeltchev, I.K, J. J. Cebe. Room-temperature thermotropic ionic liquid crystals: viologen bis(triflimide) salts. Mol Cryst Liq Cryst 419

M

(2004) 27-46.

ED

15. P. K.Bhowmik, P. K., H. Han, J. J. Cebe, R. A. Burchett, A. M. Sarker. Main-chain viologen polymers with organic counterions exhibiting thermotropic liquid-crystalline

PT

and fluorescent properties. J Polym Sci Part A: Polym Chem. 40 (2002) 659-674.

CE

16. P. K. Bhowmik, R. A. Burchett, H. Han, J. J. Cebe. Synthesis and characterization of poly (pyridinium salt) with organic counter ion exhibiting both lyotropic liquid-

AC

crystalline and light-emitting properties. Macromolecules. 34 (2001) 7579-7581. 17. W. M. Clark. Oxidation-Reduction Potentials of Organic Systems. Baltimore: USA; Williams and Wilkins Co.1960, p. 511.

20

ACCEPTED MANUSCRIPT

Highlights

CE

PT

ED

M

AN

US

CR

IP

T

Polymeric viologens catalyze rapid oxidation of carbohydrates All carbohydrates found in biomass are catalytically oxidized by polymeric viologens Carbohydrate oxidation by polymeric viologens is rapid Polymeric viologens are important catalysts for fuel cell development

AC

   

21