Accepted Manuscript Title: Enzyme Adsorption, Precipitation and Crosslinking of Glucose Oxidase and Laccase on Polyaniline Nanofibers for Highly Stable Enzymatic Biofuel Cells Author: Ryang Eun Kim Sung-Gil Hong Su Ha Jungbae Kim PII: DOI: Reference:
S0141-0229(14)00138-0 http://dx.doi.org/doi:10.1016/j.enzmictec.2014.08.001 EMT 8666
To appear in:
Enzyme and Microbial Technology
Received date: Accepted date:
4-8-2014 6-8-2014
Please cite this article as: Kim RE, Hong S-G, Ha S, Kim J, Enzyme Adsorption, Precipitation and Crosslinking of Glucose Oxidase and Laccase on Polyaniline Nanofibers for Highly Stable Enzymatic Biofuel Cells, Enzyme and Microbial Technology (2014), http://dx.doi.org/10.1016/j.enzmictec.2014.08.001 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.
Highlights (for review)
Highlight Enzyme adsorption, precipitation and crosslinking (EAPC) approach offered high loading and stability of enzymes. Enzymatic biofuel cells were successfully fabricated and operated using enzyme anode
ip t
(glucose oxidase) and cathode (laccase).
cr
Enzymatic biofuel cells using EAPC-based electrodes improved both power density output
Ac
ce pt
ed
M
an
us
and performance stability.
Page 1 of 26
*Manuscript
1
Enzyme Adsorption, Precipitation and Crosslinking of Glucose
3
Oxidase and Laccase on Polyaniline Nanofibers for Highly Stable
4
Enzymatic Biofuel Cells
cr
5
Ryang Eun Kim1,†, Sung-Gil Hong 1,†, Su Ha2,*, Jungbae Kim1,**
us
6
8
1
9
Republic of Korea
an
7
M
Department of Chemical and Biological Engineering, Korea University, Seoul 136-701,
10 11
2
12
WA 99164, USA
ed
School of Chemical Engineering and Bioengineering, Washington State University, Pullman,
These authors contributed equally to this work.
Ac
†
ce pt
13 14
ip t
2
1
Page 2 of 26
Corresponding Authors
16
* Prof. Su Ha
17
The Gene and Linda Voiland School of Chemical Engineering and Bioengineering,
18
Washington State University
19
Pullman, WA 99164, USA
20
Tel.: 509 335 3786
21
Fax: 509 335 4806
22
E-mail address: [email protected]
us
cr
ip t
15
an
23
** Prof. Jungbae Kim
25
Department of Chemical and Biological Engineering,
26
Korea University
27
Seoul 136-701, Republic of Korea
28
Tel.:+82 2 958 4850
29
Fax: +82 2 926 6102
30
E-mail address: [email protected]
ed
ce pt
Ac
31
M
24
2
Page 3 of 26
32
Abstract Enzymatic biofuel cells have many great features as a small power source for medical,
34
environmental and military applications. Both glucose oxidase (GOx) and laccase (LAC) are
35
widely used anode and cathode enzymes for enzymatic biofuel cells, respectively. In this paper,
36
we employed three different approaches to immobilize GOx and LAC on polyaniline nanofibers
37
(PANFs): enzyme adsorption (EA), enzyme adsorption and crosslinking (EAC) and enzyme
38
adsorption, precipitation and crosslinking (EAPC) approaches. The activity of EAPC-LAC was
39
32 and 25 times higher than that of EA-LAC and EAC-LAC, respectively. The half-life of
40
EAPC-LAC was 53 days, while those of EA-LAC and EAC-LAC were 6 and 21 days,
41
respectively. Similar to LAC, EAPC-GOx also showed higher activity and stability than EA-
42
GOx and EAC-GOx. For the biofuel cell application, EAPC-GOx and EAPC-LAC were applied
43
over the carbon papers to form enzyme anode and cathode, respectively. In order to improve the
44
power density output of enzymatic biofuel cell, 1,4-benzoquinone (BQ) and 2,2′-azino-bis(3-
45
ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) were introduced as the electron
46
transfer mediators on the enzyme anode and enzyme cathode, respectively. BQ- and ABTS-
47
mediated enzymatic biofuel cells fabricated by EAPC-GOx and EAPC-LAC showed the
48
maximum power density output of 37.4 μW/cm2, while the power density output of 3.1 μW/cm2
49
was shown without mediators. Under room temperature and 4 °C for 28 days, enzymatic biofuel
50
cells maintained 54 and 70 % of its initial power density, respectively.
Ac
ce pt
ed
M
an
us
cr
ip t
33
51 52
Keywords: Enzyme adsorption, precipitation and crosslinking (EAPC); Polyaniline nanofibers;
53
Glucose oxidase; Laccase; Enzymatic biofuel cells
3
Page 4 of 26
54
1.
Introduction Enzymatic biofuel cells are energy conversion devices that could efficiently convert the
56
chemical energy of biofuels into electrical energy using enzymes as biocatalysts [1]. They can
57
operate under mild condition such as a neutral pH and an ambient temperature [1,2]. Enzymatic
58
biofuel cells have a great potential to be used as a portable and uninterrupted power source for
59
the various medical, environmental and military applications by using the fuels such as glucose,
60
which are commonly available to biological and environmental systems [3-7]. However, despite
61
of promising application of enzymatic biofuel cells, their low power density and short lifetime,
62
both of which linked to low loading and poor stability of enzymes, have been identified as two
63
critical issues that need to be addressed [8]. As a potential solution, nanobiocatalytic approaches,
64
in which enzyme are incorporated into nanostructured materials, have been employed to provide
65
enhanced the loading and stability of enzymes [9]. In particular, polyaniline nanofibers (PANFs)
66
is a very interesting supporting material because they can offer a large surface area with
67
nanofiber matrices as well as high electron conductive property [10]. Moreover, PANFs can be
68
easily and economically synthesized when compared to other nanostructured materials such as
69
electrospun nanofibers, nanoparticles, carbon nanotubes and mesoporous materials. Because of
70
these promising properties, PANFs have been employed to immobilize and stabilize various
71
enzymes on PANFs [11-13].
Ac
ce pt
ed
M
an
us
cr
ip t
55
72
In the present study, we immobilized glucose oxidase (GOx) and laccase (LAC) on PANFs
73
via the enzyme adsorption, precipitation and crosslinking (EAPC) approach, together with the
74
enzyme adsorption (EA) and enzyme adsorption and crosslinking (EAC) approaches as controls,
75
to fabricate enzymatic biofuel cells. The anode is consisted of GOx immobilized in the form of
76
EAPC (i.e., EAPC-GOx), while the cathode is consisted of LAC immobilized in the form of 4
Page 5 of 26
EAPC (i.e., EAPC-LAC). We investigated the effect of mediators on each electrode, and
78
evaluated their biofuel cell performances in terms of power density and long-term stability by
79
using glucose as the fuel. Based on our knowledge, it is first time to fabricate and successfully
80
operate enzymatic biofuel cells by utilizing both the enzyme anode and enzyme cathode in the
81
form of EAPC.
ip t
77
83
2.
84
2.1. Materials
us
Materials and methods
Laccase (LAC) from Trametes versicolor, glucose oxidase (GOx) from Aspergillus niger, syringaldazine,
87
tetramethylbenzidine (TMB), glutaraldehyde solution (GA, 25%), ammonium sulfate, 2,2’-
88
azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS), 1,4-benzoquinone
89
(BQ), Nafion® solution (5 wt%), aniline and ammonium persulfate were purchased from Sigma
90
(St. Louis, MO, USA). Carbon papers (CPs) and Nafion® 117 membrane were purchase from
91
Fuel Cell Store (Boulder, CO, USA).
horseradish
peroxidase
(HRP),
3,3’,5,5’-
ed
ce pt
2.2. Synthesis of polyaniline nanofibers
Ac
93
β-D-glucose,
M
86
92
methanol,
an
85
cr
82
94
Polyaniline nanofiber was synthesized by initiating polymerization of aniline in acidic
95
condition using ammonium persulfate as an initiator [10]. First, 9 M aniline monomer solution
96
and 0.1 M ammonium persulfate solution were prepared in 1 M HCl. Both aniline and
97
ammonium persulfate solutions in HCl were mixed and shaken using 200 rpm at room
98
temperature for 24 hrs. After a completion of the polymerization reaction, PANFs were
5
Page 6 of 26
99 100
centrifuged down, washed using DI water excessively for 3 times, suspended in DI water and stored at 4 ˚C until use.
101
2.3. Immobilization of LAC and GOx on PANFs
ip t
102
PANFs were used for the immobilization of LAC and GOx in three different enzyme
104
immobilization methods: EA, EAC, and EAPC. PANFs were washed with 100 mM phosphate
105
buffer (PB) solution (pH 7.0) for 3 times prior to the immobilization processes. Immobilized
106
LAC in the form of enzyme adsorption on PANF (i.e., EA-LAC) was prepared by mixing the 2
107
mg of PANF with the LAC solution (10 mg/mL) in 100 mM PB (pH 6.5) under the shaking
108
condition at 150 rpm for 1 hr. For the preparation of immobilized LAC in the form of enzyme
109
adsorption and crosslinking on PANF (i.e., EAC-LAC), the glutaraldehyde (GA) as the chemical
110
crosslinking agent was introduced to make a final concentration of 0.5% (w/v) to the EA-LAC
111
sample under the shaking condition at 50 rpm and 4 ˚C for 17 hrs. To prepare the immobilized
112
LAC in the form of enzyme adsorption, precipitation and crosslinking on PANF (i.e., EAPC-
113
LAC), the ammonium sulfate solution was introduced into the 100 mM PB solution (pH 6.5)
114
containing both LAC and PANF to make a concentration of 50% (w/v). In the presence of the
115
ammonium sulfate salt, the free LAC (i.e., LAC that is not adsorbed over PANF surface) was
116
precipitated out to form the enzyme aggregates. After shaking at 200 rpm for 30 mins, the GA
117
solution was added into the mixture to make a concentration of 0.5% (w/v) to chemically
118
crosslink the precipitated LAC aggregates over the surface of PANF at 4 ˚C for 17 hrs. To cap
119
un-reacted aldehyde groups, the samples were shaken at 200 rpm in 100 mM Tris-HCl buffer
120
(pH 7.4) solution for 30 min and the samples were excessively washed for 3 times with the 100
121
mM PB solution (pH 6.5). EA-LAC, EAC-LAC and EAPC-LAC were stored in 100 mM PB
Ac
ce pt
ed
M
an
us
cr
103
6
Page 7 of 26
122
solution (pH 6.5) at 4 ˚C until use. The EA-GOx, EAC-GOx and EAPC-GOx were also prepared
123
by following the same protocols that were used for the immobilization of LAC on PANF as
124
described above.
126
ip t
125
2.4. Activity and stability measurement of immobilized LAC and GOx on PANFs
The activity was calculated from the time-dependent change of absorbance, and the stabilities
128
of samples were checked by measuring the residual activity time-dependently after incubation in
129
buffer solution at room temperature. The measurement of LAC activity was based on the
130
oxidation of syringaldazine [14]. Syringaldazine (7.8 mg) dissolved in methanol (10 ml) with a
131
final concentration of 0.216 mM. 100 μL of the solution containing the immobilized LAC on
132
PANFs (0.1 mg/ml) was mixed with 800 μL of 100 mM PB solution (pH 6.5) and the mixtures
133
were heated at 30 ˚C for 10 mins. This heated mixture was added with 100 μL of syringaldazine
134
solution (0.216 mM) and the absorbance at 530 nm (A530) was measured by using UV
135
spectrophotometer (UV-1800, Shimadzu, Kyoto, Japan).
ce pt
ed
M
an
us
cr
127
The activity of immobilized GOx on PANFs was measured by GOx assay [15]. The
137
measurement of activity needs a reaction cocktail containing TMB and glucose solution.
138
Reaction cocktail was made of 12 ml of TMB solution (0.576 mg/ml) and 2.5 ml glucose
139
solution (110.1 mg/ml). To measure activity of immobilized GOx, 890 μL of reaction cocktail
140
was mixed with 10 μL HRP solution (3.798 mg/ml). Then, 100 μL of the solution containing the
141
immobilized GOx on PANFs (1 μg/ml) was added to 900 μL of the mixed solution. The
142
absorbance of immobilized GOx was measured at 655 nm (A655) by using UV spectrophotometer.
Ac
136
143 144
2.5. Preparation of enzyme electrodes 7
Page 8 of 26
Carbon papers (CPs, thickness of 370 μm, 0.44 g/cm3) were treated with acid before use. In
146
a typical preparation, 2cm × 2cm squares of CPs was added to an acid solution composed of
147
H2SO4 (98%, 30 ml) and HNO3 (70%, 10 ml) for overnight at room temperature under a stirring
148
condition. Then, acid-treated CPs were washed with distilled water, dried at vacuum condition
149
and stored at room temperature until use. To prepare the GOx-based anode, the immobilized
150
GOx on PANF sample was mixed with Nafion® solution (final conc. 0.5 wt %) and this mixture
151
was stored at 4 ˚C for 1 hr. Acid-treated CPs (0.332 cm2) was soaked into the mixture for 10
152
mins, followed by drying at ambient conditions. After drying, the prepared GOx-based enzyme
153
anode was stored in 100 mM PB solution (pH 7.0) at 4 ˚C. For the LAC-based cathode, ABTS
154
(30 mM, 3.3 mg) was added to the mixture of Nafion® and immobilized LAC on PANF sample.
155
When the LAC-based cathode containing ABTS was dried, it was stored in 100 mM PB solution
156
(pH 7.0) at 4 ˚C. Since ABTS has high solubility in aqueous solution, washing was not carried
157
out [16].
ed
M
an
us
cr
ip t
145
159
ce pt
158
2.6. Biofuel cell operation measurement
The electrochemical measurements were performed by using Bio-Logic SP-150 (Knoxville,
161
TN, USA). The performance of enzymatic biofuel cells was measured by circulating 200 mM
162
glucose solution with and without 10 mM BQ in 100 mM PB solution (pH 7.0) at the flow rate of
163
0.4 ml/min within the GOx-based anode. For the LAC-based cathode, the air-breathing structure
164
was used to utilize the ambient air as its oxygen source. The Bio-Logic SP-150 was used to
165
measure the current and voltage outputs of biofuel cell by 3 minute interval under the various
166
load conditions. The power density (μW/cm2) was calculated by multiplying current and voltage
167
and then divided by surface of electrode (0.332 cm2).
Ac
160
8
Page 9 of 26
168
3.
169
3.1. Immobilization of LAC and GOx on PANFs
Results and discussion
Figure 1 shows schematic illustrations for the immobilization of enzymes (LAC and GOx)
171
in three different enzyme immobilization methods: EA, EAC, and EAPC. The scanning electron
172
microscope (SEM) images of PANFs, EA, EAC and EAPC were shown in Figure 2 for both
173
GOx and LAC samples. The nanofiber morphology of EA and EAC samples was fairly similar
174
with pristine PANFs (SEM image of pristine PANFs is not shown), whereas EAPC showed
175
remarkably thicker nanofibers revealing the enzyme coating layer over the surface of PANFs. By
176
checking twenty samples of nanofiber images, the average thicknesses of EA-LAC, EAC-LAC
177
and EAPC-LAC were estimated to be 61 ± 6, 82 ± 7 and 115 ± 6 nm, respectively, while those of
178
EA-GOx, EAC-GOx and EAPC-GOx were 75 ± 5, 91 ± 7 and 142 ± 15 nm, respectively. The
179
thickness of the enzyme coating layer for EAPC samples increased significantly than those of EA
180
and EAC samples due to their improved enzyme loading induced by the ammonium sulfate
181
assisted enzyme precipitation step to form the enzyme aggregates and its subsequent chemical
182
crosslinking step. Since the only difference between EAC and EAPC samples was the addition of
183
the enzyme precipitation process for EAPC sample, the SEM data clearly indicates that the
184
enzyme precipitation process is a critical step in order to form the thick enzyme-coating layer
185
over the supporting material. Furthermore, it is interesting to note that EAPC samples with GOx
186
offer a much thicker enzyme coating layer than that of EAPC samples with LAC. GOx can be
187
crosslinked more rigorously than LAC because the number of lysine residues per each GOx
188
molecule is 15 [17], while each LAC molecule has only 8 lysine residues [18].
Ac
ce pt
ed
M
an
us
cr
ip t
170
189 190
3.2. Activity and stability of immobilized LAC and GOx on PANFs 9
Page 10 of 26
Activity of immobilized LAC and GOx on PANFs is shown in Figure 3. The activities of
192
EA-LAC, EAC-LAC and EAPC-LAC samples were 1.9, 2.4 and 61.4 A530/min per mg of PANFs,
193
respectively (Fig. 3a). The activity of EAPC-LAC sample was 32 and 25 times higher than that
194
of EA-LAC and EAC-LAC samples, respectively. The activities of immobilized GOx on PANFs
195
have similar tendencies (Fig. 3b). The activities of EA-GOx, EAC-GOx and EAPC-GOx were
196
35.9, 124.6 and 5930 A655/min per mg of PANFs. EAPC-GOx approach offers 165 and 47 times
197
higher activity than that of EA-GOx and EAC-GOx approaches, respectively. The higher activity
198
of EAPC than that of EA and EAC can be interpreted in terms of enhanced enzyme loading by
199
the combination of the precipitation and crosslinking processes. These results matched well with
200
morphological changes of EA, EAC and EAPC samples shown in their corresponding SEM
201
images (Figure 2).
M
an
us
cr
ip t
191
Figure 4 shows the stabilities of EA-LAC, EAC-LAC and EAPC-LAC over the 78 days at
203
room temperature. After 78 days, EAPC-LAC maintained 43% of the initial activity, whereas
204
EA-LAC and EAC-LAC maintained 5% and 12% of their initial activities, respectively. The
205
inactivation profiles of all samples were bi-phasic with faster inactivation followed by the slower
206
inactivation at the later phase. The faster inactivation in the early phase can be explained by the
207
labile form of enzymes after being immobilized. The half-life of each sample in the early phase
208
was estimated from the first-order inactivation kinetics. The estimated half-lives of EA-LAC,
209
EAC-LAC and EAPC-LAC were 6, 21 and 53 days, respectively. The stability of immobilized
210
GOx on PANFs was measured in our previous work [12]. After 56 days, the relative activities of
211
EA-GOx, EAC-GOx and EAPC-GOx samples were 22%, 19% and 91%. EA and EAC
212
approaches resulted in poor enzyme stability due to the denaturation and continuous leaching of
213
enzymes from PANFs. Furthermore, the higher enzyme stability offered by EAPC approach can
Ac
ce pt
ed
202
10
Page 11 of 26
be explained in terms of its effective enzyme precipitation and crosslinking steps on PANFs.
215
Precipitation by ammonium sulfate allows enzyme molecules to be closely packed. This closely
216
packed and large enzyme aggregates can make the formation of multi-point chemical linkages on
217
the surface of each enzyme molecule more effective when they are treated with glutaric
218
dialdehyde (glutaraldehyde, GA) for the crosslinking of enzymes. These multi-point chemical
219
linkages on the enzyme surface can effectively prevent denaturation and leaching of enzymes,
220
thus stabilizing the activity of enzymes over the long operation time as it was demonstrated by
221
our EAPC samples [19,20].
us
cr
ip t
214
223
an
222
3.3. Enzymatic biofuel cells
Figure 5 shows the schematic of enzymatic biofuel cell. Based on the activity and stability
225
tests, both GOx and LAC showed the best performances when they were immobilized in the
226
form of EAPC. Thus, we fabricated the enzyme anode and enzyme cathode by entrapping
227
EAPC-GOx and EAPC-LAC over the carbon paper using Nafion® as a binder, respectively. As
228
indicated in Figure 5, the enzymatic biofuel cell is made of the anode chamber, anode current
229
collector, enzyme anode, proton exchange membrane (Nafion® 117), enzyme cathode, cathode
230
current collector and cathode holder. During the cell operation, the glucose fuel was
231
electrochemically oxidized by GOx to produce two electrons, two protons and by-product (e.g.
232
gluconolactone) over the enzyme anode. Generated electrons from the anode flow to the cathode
233
through the external load circuit to provide the electrical power. At the cathode, both electrons
234
and protons are combined together to form H2O.
Ac
ce pt
ed
M
224
235
Figure 6a and Figure 6b show the voltage-current (V-I) curves and the power density-
236
current (P-I) curves of enzymatic biofuel cell with EAPC-GOx and EAPC-LAC at room 11
Page 12 of 26
temperature, respectively. Without any mediator, the open circuit voltage (OCV) is about 0.34 V
238
and the maximum power density is 3.1 μW/cm2. According to previous work [12], we
239
demonstrated that the power density output depends on the enzyme loading. As the enzyme
240
loading increases per unit weight of PANFs, the enzyme activity also proportionally increases
241
per unit weight of PANFs up to a certain value of enzyme loading. For the anode case, the
242
enzyme activity is closely related with electron generation rate and it represents the maximum
243
amount of electrons that can be generated without considering the charge transfer resistance. We
244
speculate that the large amount of electrons generated per unit time by the high enzyme loading
245
of EAPC-based electrode increases the probability of collecting these electrons at the current
246
collector. However, to significantly increase the power density output of our enzymatic biofuel
247
cell, it is also important to improve the electron transfer rate of its electrodes because only a
248
limited amount of electrons would be collected if the enzyme electrodes have poor electron
249
transfer rates regardless of the enzyme activities.
ed
M
an
us
cr
ip t
237
In order to increase the electron transfer rates of both enzyme electrodes, the mediators have
251
been utilized [2,21,22]. For the present study, BQ and ABTS were used as the electron transfer
252
mediators for the enzyme anode and enzyme cathode, respectively [23-25]. BQ is a liquid
253
mediator, and it is mixed with the 200 mM of glucose fuel, while ABTS is a solid mediator
254
where it is simultaneously entrapped with EAPC-LAC samples using Nafion® binder over the
255
carbon paper surface. The performance of biofuel cells containing EAPC-GOx and EAPC-LAC
256
with mediators is also shown in Figures 6. According to the voltage-current (V-I) curves of
257
enzymatic biofuel cells shown in Figure 6a, when the ABTS is incorporated to the enzyme
258
cathode containing EAPC-LAC, the OCV increases from 0.34 to 0.50 V. However, its current
259
density rapidly drops in a similar manner as the cell without the ABTS-mediated cathode. On the
Ac
ce pt
250
12
Page 13 of 26
other hand, when both the BQ and ABTS are incorporated to the enzyme anode and enzyme
261
cathode, respectively, the enzymatic biofuel cell shows a smaller cell voltage drop as the current
262
density increases compared to that of the cell with just ABTS-mediated cathode. This result
263
suggests that the overall cell performance is mainly limited by the slower electron transfer rate of
264
EAPC-GOx-based anode. Due to this smaller overpotential, the enzymatic biofuel cell with both
265
BQ- and ABTS-mediated electrodes generates a much higher current density output as shown in
266
Figure 6a.
us
cr
ip t
260
Figure 6b shows the power density-current (P-I) curves of enzymatic biofuel cells. The
268
power density output of the biofuel cell with just ABTS-mediated cathode does not improve
269
much compared to the biofuel cells without the mediators. Consequently, the maximum power
270
density outputs of the biofuel cells with no mediators and with just ABTS-mediated cathode are
271
3.1 and 5.7 μW/cm2, respectively. However, the biofuel cells with BQ- and ABTS-mediated
272
electrodes containing EAPC-GOx and EAPC-LAC showed the maximum power density up to
273
37.4 μW/cm2 due to its decreased overpotential mainly offered by the improved electron transfer
274
rate of the anode in the presence of the mediator.
M
ed
ce pt
276
3.4. Performance stability of enzymatic biofuel cells
Ac
275
an
267
277
To confirm the performance stability of enzymatic biofuel cells with EAPC-GOx and
278
EAPC-LAC electrodes, the enzyme electrodes containing EAPC-GOx and EAPC-LAC were
279
incubated in the aqueous buffer under two different temperatures (room temperature and 4 ˚C)
280
over 28 days (Fig. 7). In a 7-day interval, these enzyme electrodes were taken out from the
281
incubation and integrated into our biofuel cell shown in Figure 5 to measure its maximum power
282
density using 200 mM glucose solution at ambient temperature. After the biofuel cell test, both 13
Page 14 of 26
EAPC-GOx and EAPC-LAC electrodes were removed from the cell and washed with 100 mM
284
PB solution (pH 7.0 for the anode and pH 6.5 for the cathode) to remove any residue glucose
285
solution before they were placed back into the incubation. When incubated over 28 days, the
286
maximum power density output of the biofuel cell with the electrodes stored at room temperature
287
maintained 54% of its initial value, while the biofuel cell with the electrodes stored at 4 ˚C
288
maintained 70% of its initial value. As seen in Figure 7, there is almost no difference in the
289
maximum power density of biofuel cell between two different temperatures used for the thermal
290
stability tests within the 7 days. However, as the thermal stability test continues beyond 7 days,
291
the power density output of the biofuel cell with the electrodes stored at 4 ˚C shows a lower
292
deactivation rate than the biofuel cell with the electrodes stored at room temperature. Overall, the
293
biofuel cell with EAPC-GOx and EAPC-LAC electrodes shows a good performance stability,
294
which agrees with their stability results shown in Figure 3.
295
4.
Conclusions
ce pt
296
ed
M
an
us
cr
ip t
283
In this study, we introduced biofuel cells with anode electrode and cathode electrode based
298
on immobilized GOx and LAC on PANFs in the form of enzyme adsorption, precipitation and
299
crosslinking (EAPC). EAPC approach demonstrates that both the loading and stability of GOx
300
and LAC on PANFs can be significantly improved by introducing the enzyme precipitation and
301
crosslinking steps for their immobilization processes. By applying EAPC method to the
302
enzymatic biofuel cell, we have achieved both high power density output and improved
303
performance stability. As shown in successful application to biofuel cells, it is anticipated that
304
EAPC method can be utilized for various types of enzyme-based electrochemical applications
305
such as biosensors and enzyme logic gates.
306
Ac
297
14
Page 15 of 26
307
Acknowledgement
308
This work was supported by the grant from the Agency for Defense Development (ADD -14-70-
309
04-01).
ip t
310
References
312
[1] Barton SC, Gallaway J, Atanassov P. Enzymatic biofuel cells for Implantable and microscale
313
devices. Chem Rev 2004;104:4867-86.
314
[2] Osman MH, Shah AA, Walsh FC. Recent progress and continuing challenges in bio-fuel cells.
315
Part I: Enzymatic cells. Biosens Bioelectron 2011;26:3087-102.
316
[3] MacVittie K, Halamek J, Halamkova L, Southcott M, Jemison WD, Lobel R, et al. From
317
"cyborg" lobsters to a pacemaker powered by implantable biofuel cells. Energy Environ Sci
318
2013;6:81-6.
319
[4] Halamkova L, Halamek J, Bocharova V, Szczupak A, Alfonta L, Katz E. Implanted Biofuel
320
Cell Operating in a Living Snail. J Am Chem Soc 2012;134:5040-3.
321
[5] Szczupak A, Halamek J, Halamkova L, Bocharova V, Alfonta L, Katz E. Living battery -
322
biofuel cells operating in vivo in clams. Energy Environ Sci 2012;5:8891-5.
323
[6] Rasmussen M, Ritzmann RE, Lee I, Pollack AJ, Scherson D. An Implantable Biofuel Cell for
324
a Live Insect. J Am Chem Soc 2012;134:1458-60.
325
[7] Yan YM, Zheng W, Su L, Mao LQ. Carbon-nanotube-based glucose/O-2 biofuel cells. Adv
326
Mater 2006;18:2639-43.
327
[8] Kim J, Jia HF, Wang P. Challenges in biocatalysis for enzyme-based biofuel cells. Biotechnol
328
Adv 2006;24:296-308.
329
[9] Kim JB, Grate JW, Wang P. Nanobiocatalysis and its potential applications. Trends
Ac
ce pt
ed
M
an
us
cr
311
15
Page 16 of 26
Biotechnol 2008;26:639-46.
331
[10] Huang JX. Syntheses and applications of conducting polymer polyaniline nanofibers. Pure
332
Appl Chem 2006;78:15-27.
333
[11] Hong SG, Kim HS, Kim J. Highly Stabilized Lipase in Polyaniline Nanofibers for
334
Surfactant-Mediated Esterification of Ibuprofen. Langmuir 2014;30:911-5.
335
[12] Kim H, Lee I, Kwon Y, Kim BC, Ha S, Lee JH, et al. Immobilization of glucose oxidase
336
into polyaniline nanofiber matrix for biofuel cell applications. Biosens Bioelectron
337
2011;26:3908-13.
338
[13] Lee G, Kim J, Lee JH. Development of magnetically separable polyaniline nanofibers for
339
enzyme immobilization and recovery. Enzyme Microb Technol 2008;42:466-72.
340
[14] Ride JP. The Effect of Induced Lignification on the Resistance of Wheat Cell-Walls to
341
Fungal Degradation. Physiol Plant Pathol 1980;16:187-92.
342
[15] Bergmeyer HU, Gawehn, K., Grassl, M. In: Bergmeyer, H.U. (Ed.), in Methods of
343
Enzymatic Analysis., second ed. Academic Press Inc., New York, NY 1974:457-8.
344
[16] Zebda A, Gondran C, Cinquin P, Cosnier S. Glucose biofuel cell construction based on
345
enzyme, graphite particle and redox mediator compression. Sensors Actuators B: Chem
346
2012;173:760-4.
347
[17] www.ncbi.nlm.nih.gov/, PDB ID: 1CF3, (Accessed August 2014).
348
[18] www.ncbi.nlm.nih.gov/, PDB ID: 1GYC, (Accessed August 2014).
349
[19] Mozhaev VV, Melik-nubarov NS, Sergeeva MV, Šikšnis V, Martinek K. Strategy for
350
Stabilizing Enzymes Part One: Increasing Stability of Enzymes via their Multi-Point Interaction
351
with a Support. Biocatal Biotransform 1990;3:179-87.
352
[20] Martinek K, Klibanov AM, Goldmacher VS, Berezin IV. Principles of Enzyme
Ac
ce pt
ed
M
an
us
cr
ip t
330
16
Page 17 of 26
Stabilization .1. Increase in Thermostability of Enzymes Covalently Bound to a Complementary
354
Surface of a Polymer Support in a Multipoint Fashion. Biochim Biophys Acta 1977;485:1-12.
355
[21] Kuwahara T, Ohta H, Kondo M, Shimomura M. Immobilization of glucose oxidase on
356
carbon paper electrodes modified with conducting polymer and its application to a glucose fuel
357
cell. Bioelectrochemistry 2008;74:66-72.
358
[22] Meredith MT, Minteer SD. Biofuel Cells: Enhanced Enzymatic Bioelectrocatalysis. Annu
359
Rev Anal Chem 2012;5:157-79.
360
[23] Morozova OV, Shumakovich GP, Shleev SV, Yaropolov YI. Laccase-mediator systems and
361
their applications: A review. Appl Biochem Microbiol 2007;43:523-35.
362
[24] Lau KT, de Fortescu SAL, Murphy LJ, Slater JM. Disposable glucose sensors for flow
363
injection analysis using substituted 1,4-benzoquinone mediators. Electroanalysis 2003;15:975-81.
364
[25] Fei JF, Song HK, Palmore GTR. A biopolymer composite that catalyzes the reduction of
365
oxygen to water. Chem Mater 2007;19:1565-70.
Ac
ce pt
ed
M
an
us
cr
ip t
353
17
Page 18 of 26
Figure Captions
367
Figure 1. Schematic illustrations for three different enzyme immobilization methods using
368
PANFs: enzyme adsorption (EA), enzyme adsorption and crosslinking (EAC) and enzyme
369
adsorption, precipitation and crosslinking (EAPC). Magnified figure of EAPC represents
370
crosslinking between enzymes molecules.
371
Figure 2. SEM images of (a) EA-LAC, (b) EAC-LAC, (c) EAPC-LAC, (d) EA-GOx, (e) EAC-
372
GOx and (f) EAPC-GOx. The scale bars in all images correspond to 1 μm.
373
Figure 3. The activities of (a) EA-LAC, EAC-LAC and EAPC-LAC, and (b) EA-GOx, EAC-
374
GOx and EAPC-GOx.
375
Figure 4. Stabilities of EA-LAC, EAC-LAC and EAPC-LAC on PANFs at room temperature.
376
Figure 5. Schematic illustrations of biofuel cells consisting of EAPC-GOx (anode electrode) and
377
EAPC-LAC (cathode electrode).
378
Figure 6. (a) The voltage-current (V-I) curves and (b) the power density-current (P-I) curves of
379
enzymatic biofuel cell containing EAPC-GOx and EAPC-LAC with and without mediators (BQ
380
for the anode mediator and ABTS for the cathode mediator).
381
Figure 7. Relative maximum power density of biofuel cells containing EAPC-GOx (anode) and
382
EAPC-LAC (cathode), which were stored at room temperature and 4 ˚C over 28 days.
Ac
ce pt
ed
M
an
us
cr
ip t
366
18
Page 19 of 26
an
us
cr
ip t
383
M
384
Fig. 1. Schematic illustrations for three different enzyme immobilization methods using PANFs:
386
enzyme adsorption (EA), enzyme adsorption and crosslinking (EAC) and enzyme adsorption,
387
precipitation and crosslinking (EAPC). Magnified figure of EAPC represents crosslinking
388
between enzymes molecules.
Ac
ce pt
ed
385
19
Page 20 of 26
ip t cr us an M ed ce pt Ac
389 390
Fig. 2. SEM images of (a) EA-LAC, (b) EAC-LAC, (c) EAPC-LAC, (d) EA-GOx, (e) EAC-
391
GOx and (f) EAPC-GOx. The scale bars in all images correspond to 1 μm.
20
Page 21 of 26
ce pt
ed
M
an
us
cr
ip t
392
Ac
393 394
Fig. 3. The activities of (a) EA-LAC, EAC-LAC and EAPC-LAC, and (b) EA-GOx, EAC-GOx
395
and EAPC-GOx.
21
Page 22 of 26
ip t cr us
396
ce pt
ed
M
an
Fig. 4. Stabilities of EA-LAC, EAC-LAC and EAPC-LAC on PANFs at room temperature.
Ac
397
22
Page 23 of 26
ip t cr us
398
Fig. 5. Schematic illustrations of biofuel cells consisting of EAPC-GOx (anode electrode) and
400
EAPC-LAC (cathode electrode).
Ac
ce pt
ed
M
an
399
23
Page 24 of 26
ip t cr us an M ed ce pt
401
Fig. 6. (a) The voltage-current (V-I) curves and (b) the power density-current (P-I) curves of
403
enzymatic biofuel cell containing EAPC-GOx and EAPC-LAC with and without mediators (BQ
404
for the anode mediator and ABTS for the cathode mediator)..
Ac
402
24
Page 25 of 26
ip t cr us
405
Fig. 7. Relative maximum power density of biofuel cells containing EAPC-GOx (anode) and
407
EAPC-LAC (cathode), which were stored at room temperature and 4 ˚C over 28 days.
Ac
ce pt
ed
M
an
406
25
Page 26 of 26
S0141-0229(14)00138-0 http://dx.doi.org/doi:10.1016/j.enzmictec.2014.08.001 EMT 8666
To appear in:
Enzyme and Microbial Technology
Received date: Accepted date:
4-8-2014 6-8-2014
Please cite this article as: Kim RE, Hong S-G, Ha S, Kim J, Enzyme Adsorption, Precipitation and Crosslinking of Glucose Oxidase and Laccase on Polyaniline Nanofibers for Highly Stable Enzymatic Biofuel Cells, Enzyme and Microbial Technology (2014), http://dx.doi.org/10.1016/j.enzmictec.2014.08.001 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.
Highlights (for review)
Highlight Enzyme adsorption, precipitation and crosslinking (EAPC) approach offered high loading and stability of enzymes. Enzymatic biofuel cells were successfully fabricated and operated using enzyme anode
ip t
(glucose oxidase) and cathode (laccase).
cr
Enzymatic biofuel cells using EAPC-based electrodes improved both power density output
Ac
ce pt
ed
M
an
us
and performance stability.
Page 1 of 26
*Manuscript
1
Enzyme Adsorption, Precipitation and Crosslinking of Glucose
3
Oxidase and Laccase on Polyaniline Nanofibers for Highly Stable
4
Enzymatic Biofuel Cells
cr
5
Ryang Eun Kim1,†, Sung-Gil Hong 1,†, Su Ha2,*, Jungbae Kim1,**
us
6
8
1
9
Republic of Korea
an
7
M
Department of Chemical and Biological Engineering, Korea University, Seoul 136-701,
10 11
2
12
WA 99164, USA
ed
School of Chemical Engineering and Bioengineering, Washington State University, Pullman,
These authors contributed equally to this work.
Ac
†
ce pt
13 14
ip t
2
1
Page 2 of 26
Corresponding Authors
16
* Prof. Su Ha
17
The Gene and Linda Voiland School of Chemical Engineering and Bioengineering,
18
Washington State University
19
Pullman, WA 99164, USA
20
Tel.: 509 335 3786
21
Fax: 509 335 4806
22
E-mail address: [email protected]
us
cr
ip t
15
an
23
** Prof. Jungbae Kim
25
Department of Chemical and Biological Engineering,
26
Korea University
27
Seoul 136-701, Republic of Korea
28
Tel.:+82 2 958 4850
29
Fax: +82 2 926 6102
30
E-mail address: [email protected]
ed
ce pt
Ac
31
M
24
2
Page 3 of 26
32
Abstract Enzymatic biofuel cells have many great features as a small power source for medical,
34
environmental and military applications. Both glucose oxidase (GOx) and laccase (LAC) are
35
widely used anode and cathode enzymes for enzymatic biofuel cells, respectively. In this paper,
36
we employed three different approaches to immobilize GOx and LAC on polyaniline nanofibers
37
(PANFs): enzyme adsorption (EA), enzyme adsorption and crosslinking (EAC) and enzyme
38
adsorption, precipitation and crosslinking (EAPC) approaches. The activity of EAPC-LAC was
39
32 and 25 times higher than that of EA-LAC and EAC-LAC, respectively. The half-life of
40
EAPC-LAC was 53 days, while those of EA-LAC and EAC-LAC were 6 and 21 days,
41
respectively. Similar to LAC, EAPC-GOx also showed higher activity and stability than EA-
42
GOx and EAC-GOx. For the biofuel cell application, EAPC-GOx and EAPC-LAC were applied
43
over the carbon papers to form enzyme anode and cathode, respectively. In order to improve the
44
power density output of enzymatic biofuel cell, 1,4-benzoquinone (BQ) and 2,2′-azino-bis(3-
45
ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) were introduced as the electron
46
transfer mediators on the enzyme anode and enzyme cathode, respectively. BQ- and ABTS-
47
mediated enzymatic biofuel cells fabricated by EAPC-GOx and EAPC-LAC showed the
48
maximum power density output of 37.4 μW/cm2, while the power density output of 3.1 μW/cm2
49
was shown without mediators. Under room temperature and 4 °C for 28 days, enzymatic biofuel
50
cells maintained 54 and 70 % of its initial power density, respectively.
Ac
ce pt
ed
M
an
us
cr
ip t
33
51 52
Keywords: Enzyme adsorption, precipitation and crosslinking (EAPC); Polyaniline nanofibers;
53
Glucose oxidase; Laccase; Enzymatic biofuel cells
3
Page 4 of 26
54
1.
Introduction Enzymatic biofuel cells are energy conversion devices that could efficiently convert the
56
chemical energy of biofuels into electrical energy using enzymes as biocatalysts [1]. They can
57
operate under mild condition such as a neutral pH and an ambient temperature [1,2]. Enzymatic
58
biofuel cells have a great potential to be used as a portable and uninterrupted power source for
59
the various medical, environmental and military applications by using the fuels such as glucose,
60
which are commonly available to biological and environmental systems [3-7]. However, despite
61
of promising application of enzymatic biofuel cells, their low power density and short lifetime,
62
both of which linked to low loading and poor stability of enzymes, have been identified as two
63
critical issues that need to be addressed [8]. As a potential solution, nanobiocatalytic approaches,
64
in which enzyme are incorporated into nanostructured materials, have been employed to provide
65
enhanced the loading and stability of enzymes [9]. In particular, polyaniline nanofibers (PANFs)
66
is a very interesting supporting material because they can offer a large surface area with
67
nanofiber matrices as well as high electron conductive property [10]. Moreover, PANFs can be
68
easily and economically synthesized when compared to other nanostructured materials such as
69
electrospun nanofibers, nanoparticles, carbon nanotubes and mesoporous materials. Because of
70
these promising properties, PANFs have been employed to immobilize and stabilize various
71
enzymes on PANFs [11-13].
Ac
ce pt
ed
M
an
us
cr
ip t
55
72
In the present study, we immobilized glucose oxidase (GOx) and laccase (LAC) on PANFs
73
via the enzyme adsorption, precipitation and crosslinking (EAPC) approach, together with the
74
enzyme adsorption (EA) and enzyme adsorption and crosslinking (EAC) approaches as controls,
75
to fabricate enzymatic biofuel cells. The anode is consisted of GOx immobilized in the form of
76
EAPC (i.e., EAPC-GOx), while the cathode is consisted of LAC immobilized in the form of 4
Page 5 of 26
EAPC (i.e., EAPC-LAC). We investigated the effect of mediators on each electrode, and
78
evaluated their biofuel cell performances in terms of power density and long-term stability by
79
using glucose as the fuel. Based on our knowledge, it is first time to fabricate and successfully
80
operate enzymatic biofuel cells by utilizing both the enzyme anode and enzyme cathode in the
81
form of EAPC.
ip t
77
83
2.
84
2.1. Materials
us
Materials and methods
Laccase (LAC) from Trametes versicolor, glucose oxidase (GOx) from Aspergillus niger, syringaldazine,
87
tetramethylbenzidine (TMB), glutaraldehyde solution (GA, 25%), ammonium sulfate, 2,2’-
88
azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS), 1,4-benzoquinone
89
(BQ), Nafion® solution (5 wt%), aniline and ammonium persulfate were purchased from Sigma
90
(St. Louis, MO, USA). Carbon papers (CPs) and Nafion® 117 membrane were purchase from
91
Fuel Cell Store (Boulder, CO, USA).
horseradish
peroxidase
(HRP),
3,3’,5,5’-
ed
ce pt
2.2. Synthesis of polyaniline nanofibers
Ac
93
β-D-glucose,
M
86
92
methanol,
an
85
cr
82
94
Polyaniline nanofiber was synthesized by initiating polymerization of aniline in acidic
95
condition using ammonium persulfate as an initiator [10]. First, 9 M aniline monomer solution
96
and 0.1 M ammonium persulfate solution were prepared in 1 M HCl. Both aniline and
97
ammonium persulfate solutions in HCl were mixed and shaken using 200 rpm at room
98
temperature for 24 hrs. After a completion of the polymerization reaction, PANFs were
5
Page 6 of 26
99 100
centrifuged down, washed using DI water excessively for 3 times, suspended in DI water and stored at 4 ˚C until use.
101
2.3. Immobilization of LAC and GOx on PANFs
ip t
102
PANFs were used for the immobilization of LAC and GOx in three different enzyme
104
immobilization methods: EA, EAC, and EAPC. PANFs were washed with 100 mM phosphate
105
buffer (PB) solution (pH 7.0) for 3 times prior to the immobilization processes. Immobilized
106
LAC in the form of enzyme adsorption on PANF (i.e., EA-LAC) was prepared by mixing the 2
107
mg of PANF with the LAC solution (10 mg/mL) in 100 mM PB (pH 6.5) under the shaking
108
condition at 150 rpm for 1 hr. For the preparation of immobilized LAC in the form of enzyme
109
adsorption and crosslinking on PANF (i.e., EAC-LAC), the glutaraldehyde (GA) as the chemical
110
crosslinking agent was introduced to make a final concentration of 0.5% (w/v) to the EA-LAC
111
sample under the shaking condition at 50 rpm and 4 ˚C for 17 hrs. To prepare the immobilized
112
LAC in the form of enzyme adsorption, precipitation and crosslinking on PANF (i.e., EAPC-
113
LAC), the ammonium sulfate solution was introduced into the 100 mM PB solution (pH 6.5)
114
containing both LAC and PANF to make a concentration of 50% (w/v). In the presence of the
115
ammonium sulfate salt, the free LAC (i.e., LAC that is not adsorbed over PANF surface) was
116
precipitated out to form the enzyme aggregates. After shaking at 200 rpm for 30 mins, the GA
117
solution was added into the mixture to make a concentration of 0.5% (w/v) to chemically
118
crosslink the precipitated LAC aggregates over the surface of PANF at 4 ˚C for 17 hrs. To cap
119
un-reacted aldehyde groups, the samples were shaken at 200 rpm in 100 mM Tris-HCl buffer
120
(pH 7.4) solution for 30 min and the samples were excessively washed for 3 times with the 100
121
mM PB solution (pH 6.5). EA-LAC, EAC-LAC and EAPC-LAC were stored in 100 mM PB
Ac
ce pt
ed
M
an
us
cr
103
6
Page 7 of 26
122
solution (pH 6.5) at 4 ˚C until use. The EA-GOx, EAC-GOx and EAPC-GOx were also prepared
123
by following the same protocols that were used for the immobilization of LAC on PANF as
124
described above.
126
ip t
125
2.4. Activity and stability measurement of immobilized LAC and GOx on PANFs
The activity was calculated from the time-dependent change of absorbance, and the stabilities
128
of samples were checked by measuring the residual activity time-dependently after incubation in
129
buffer solution at room temperature. The measurement of LAC activity was based on the
130
oxidation of syringaldazine [14]. Syringaldazine (7.8 mg) dissolved in methanol (10 ml) with a
131
final concentration of 0.216 mM. 100 μL of the solution containing the immobilized LAC on
132
PANFs (0.1 mg/ml) was mixed with 800 μL of 100 mM PB solution (pH 6.5) and the mixtures
133
were heated at 30 ˚C for 10 mins. This heated mixture was added with 100 μL of syringaldazine
134
solution (0.216 mM) and the absorbance at 530 nm (A530) was measured by using UV
135
spectrophotometer (UV-1800, Shimadzu, Kyoto, Japan).
ce pt
ed
M
an
us
cr
127
The activity of immobilized GOx on PANFs was measured by GOx assay [15]. The
137
measurement of activity needs a reaction cocktail containing TMB and glucose solution.
138
Reaction cocktail was made of 12 ml of TMB solution (0.576 mg/ml) and 2.5 ml glucose
139
solution (110.1 mg/ml). To measure activity of immobilized GOx, 890 μL of reaction cocktail
140
was mixed with 10 μL HRP solution (3.798 mg/ml). Then, 100 μL of the solution containing the
141
immobilized GOx on PANFs (1 μg/ml) was added to 900 μL of the mixed solution. The
142
absorbance of immobilized GOx was measured at 655 nm (A655) by using UV spectrophotometer.
Ac
136
143 144
2.5. Preparation of enzyme electrodes 7
Page 8 of 26
Carbon papers (CPs, thickness of 370 μm, 0.44 g/cm3) were treated with acid before use. In
146
a typical preparation, 2cm × 2cm squares of CPs was added to an acid solution composed of
147
H2SO4 (98%, 30 ml) and HNO3 (70%, 10 ml) for overnight at room temperature under a stirring
148
condition. Then, acid-treated CPs were washed with distilled water, dried at vacuum condition
149
and stored at room temperature until use. To prepare the GOx-based anode, the immobilized
150
GOx on PANF sample was mixed with Nafion® solution (final conc. 0.5 wt %) and this mixture
151
was stored at 4 ˚C for 1 hr. Acid-treated CPs (0.332 cm2) was soaked into the mixture for 10
152
mins, followed by drying at ambient conditions. After drying, the prepared GOx-based enzyme
153
anode was stored in 100 mM PB solution (pH 7.0) at 4 ˚C. For the LAC-based cathode, ABTS
154
(30 mM, 3.3 mg) was added to the mixture of Nafion® and immobilized LAC on PANF sample.
155
When the LAC-based cathode containing ABTS was dried, it was stored in 100 mM PB solution
156
(pH 7.0) at 4 ˚C. Since ABTS has high solubility in aqueous solution, washing was not carried
157
out [16].
ed
M
an
us
cr
ip t
145
159
ce pt
158
2.6. Biofuel cell operation measurement
The electrochemical measurements were performed by using Bio-Logic SP-150 (Knoxville,
161
TN, USA). The performance of enzymatic biofuel cells was measured by circulating 200 mM
162
glucose solution with and without 10 mM BQ in 100 mM PB solution (pH 7.0) at the flow rate of
163
0.4 ml/min within the GOx-based anode. For the LAC-based cathode, the air-breathing structure
164
was used to utilize the ambient air as its oxygen source. The Bio-Logic SP-150 was used to
165
measure the current and voltage outputs of biofuel cell by 3 minute interval under the various
166
load conditions. The power density (μW/cm2) was calculated by multiplying current and voltage
167
and then divided by surface of electrode (0.332 cm2).
Ac
160
8
Page 9 of 26
168
3.
169
3.1. Immobilization of LAC and GOx on PANFs
Results and discussion
Figure 1 shows schematic illustrations for the immobilization of enzymes (LAC and GOx)
171
in three different enzyme immobilization methods: EA, EAC, and EAPC. The scanning electron
172
microscope (SEM) images of PANFs, EA, EAC and EAPC were shown in Figure 2 for both
173
GOx and LAC samples. The nanofiber morphology of EA and EAC samples was fairly similar
174
with pristine PANFs (SEM image of pristine PANFs is not shown), whereas EAPC showed
175
remarkably thicker nanofibers revealing the enzyme coating layer over the surface of PANFs. By
176
checking twenty samples of nanofiber images, the average thicknesses of EA-LAC, EAC-LAC
177
and EAPC-LAC were estimated to be 61 ± 6, 82 ± 7 and 115 ± 6 nm, respectively, while those of
178
EA-GOx, EAC-GOx and EAPC-GOx were 75 ± 5, 91 ± 7 and 142 ± 15 nm, respectively. The
179
thickness of the enzyme coating layer for EAPC samples increased significantly than those of EA
180
and EAC samples due to their improved enzyme loading induced by the ammonium sulfate
181
assisted enzyme precipitation step to form the enzyme aggregates and its subsequent chemical
182
crosslinking step. Since the only difference between EAC and EAPC samples was the addition of
183
the enzyme precipitation process for EAPC sample, the SEM data clearly indicates that the
184
enzyme precipitation process is a critical step in order to form the thick enzyme-coating layer
185
over the supporting material. Furthermore, it is interesting to note that EAPC samples with GOx
186
offer a much thicker enzyme coating layer than that of EAPC samples with LAC. GOx can be
187
crosslinked more rigorously than LAC because the number of lysine residues per each GOx
188
molecule is 15 [17], while each LAC molecule has only 8 lysine residues [18].
Ac
ce pt
ed
M
an
us
cr
ip t
170
189 190
3.2. Activity and stability of immobilized LAC and GOx on PANFs 9
Page 10 of 26
Activity of immobilized LAC and GOx on PANFs is shown in Figure 3. The activities of
192
EA-LAC, EAC-LAC and EAPC-LAC samples were 1.9, 2.4 and 61.4 A530/min per mg of PANFs,
193
respectively (Fig. 3a). The activity of EAPC-LAC sample was 32 and 25 times higher than that
194
of EA-LAC and EAC-LAC samples, respectively. The activities of immobilized GOx on PANFs
195
have similar tendencies (Fig. 3b). The activities of EA-GOx, EAC-GOx and EAPC-GOx were
196
35.9, 124.6 and 5930 A655/min per mg of PANFs. EAPC-GOx approach offers 165 and 47 times
197
higher activity than that of EA-GOx and EAC-GOx approaches, respectively. The higher activity
198
of EAPC than that of EA and EAC can be interpreted in terms of enhanced enzyme loading by
199
the combination of the precipitation and crosslinking processes. These results matched well with
200
morphological changes of EA, EAC and EAPC samples shown in their corresponding SEM
201
images (Figure 2).
M
an
us
cr
ip t
191
Figure 4 shows the stabilities of EA-LAC, EAC-LAC and EAPC-LAC over the 78 days at
203
room temperature. After 78 days, EAPC-LAC maintained 43% of the initial activity, whereas
204
EA-LAC and EAC-LAC maintained 5% and 12% of their initial activities, respectively. The
205
inactivation profiles of all samples were bi-phasic with faster inactivation followed by the slower
206
inactivation at the later phase. The faster inactivation in the early phase can be explained by the
207
labile form of enzymes after being immobilized. The half-life of each sample in the early phase
208
was estimated from the first-order inactivation kinetics. The estimated half-lives of EA-LAC,
209
EAC-LAC and EAPC-LAC were 6, 21 and 53 days, respectively. The stability of immobilized
210
GOx on PANFs was measured in our previous work [12]. After 56 days, the relative activities of
211
EA-GOx, EAC-GOx and EAPC-GOx samples were 22%, 19% and 91%. EA and EAC
212
approaches resulted in poor enzyme stability due to the denaturation and continuous leaching of
213
enzymes from PANFs. Furthermore, the higher enzyme stability offered by EAPC approach can
Ac
ce pt
ed
202
10
Page 11 of 26
be explained in terms of its effective enzyme precipitation and crosslinking steps on PANFs.
215
Precipitation by ammonium sulfate allows enzyme molecules to be closely packed. This closely
216
packed and large enzyme aggregates can make the formation of multi-point chemical linkages on
217
the surface of each enzyme molecule more effective when they are treated with glutaric
218
dialdehyde (glutaraldehyde, GA) for the crosslinking of enzymes. These multi-point chemical
219
linkages on the enzyme surface can effectively prevent denaturation and leaching of enzymes,
220
thus stabilizing the activity of enzymes over the long operation time as it was demonstrated by
221
our EAPC samples [19,20].
us
cr
ip t
214
223
an
222
3.3. Enzymatic biofuel cells
Figure 5 shows the schematic of enzymatic biofuel cell. Based on the activity and stability
225
tests, both GOx and LAC showed the best performances when they were immobilized in the
226
form of EAPC. Thus, we fabricated the enzyme anode and enzyme cathode by entrapping
227
EAPC-GOx and EAPC-LAC over the carbon paper using Nafion® as a binder, respectively. As
228
indicated in Figure 5, the enzymatic biofuel cell is made of the anode chamber, anode current
229
collector, enzyme anode, proton exchange membrane (Nafion® 117), enzyme cathode, cathode
230
current collector and cathode holder. During the cell operation, the glucose fuel was
231
electrochemically oxidized by GOx to produce two electrons, two protons and by-product (e.g.
232
gluconolactone) over the enzyme anode. Generated electrons from the anode flow to the cathode
233
through the external load circuit to provide the electrical power. At the cathode, both electrons
234
and protons are combined together to form H2O.
Ac
ce pt
ed
M
224
235
Figure 6a and Figure 6b show the voltage-current (V-I) curves and the power density-
236
current (P-I) curves of enzymatic biofuel cell with EAPC-GOx and EAPC-LAC at room 11
Page 12 of 26
temperature, respectively. Without any mediator, the open circuit voltage (OCV) is about 0.34 V
238
and the maximum power density is 3.1 μW/cm2. According to previous work [12], we
239
demonstrated that the power density output depends on the enzyme loading. As the enzyme
240
loading increases per unit weight of PANFs, the enzyme activity also proportionally increases
241
per unit weight of PANFs up to a certain value of enzyme loading. For the anode case, the
242
enzyme activity is closely related with electron generation rate and it represents the maximum
243
amount of electrons that can be generated without considering the charge transfer resistance. We
244
speculate that the large amount of electrons generated per unit time by the high enzyme loading
245
of EAPC-based electrode increases the probability of collecting these electrons at the current
246
collector. However, to significantly increase the power density output of our enzymatic biofuel
247
cell, it is also important to improve the electron transfer rate of its electrodes because only a
248
limited amount of electrons would be collected if the enzyme electrodes have poor electron
249
transfer rates regardless of the enzyme activities.
ed
M
an
us
cr
ip t
237
In order to increase the electron transfer rates of both enzyme electrodes, the mediators have
251
been utilized [2,21,22]. For the present study, BQ and ABTS were used as the electron transfer
252
mediators for the enzyme anode and enzyme cathode, respectively [23-25]. BQ is a liquid
253
mediator, and it is mixed with the 200 mM of glucose fuel, while ABTS is a solid mediator
254
where it is simultaneously entrapped with EAPC-LAC samples using Nafion® binder over the
255
carbon paper surface. The performance of biofuel cells containing EAPC-GOx and EAPC-LAC
256
with mediators is also shown in Figures 6. According to the voltage-current (V-I) curves of
257
enzymatic biofuel cells shown in Figure 6a, when the ABTS is incorporated to the enzyme
258
cathode containing EAPC-LAC, the OCV increases from 0.34 to 0.50 V. However, its current
259
density rapidly drops in a similar manner as the cell without the ABTS-mediated cathode. On the
Ac
ce pt
250
12
Page 13 of 26
other hand, when both the BQ and ABTS are incorporated to the enzyme anode and enzyme
261
cathode, respectively, the enzymatic biofuel cell shows a smaller cell voltage drop as the current
262
density increases compared to that of the cell with just ABTS-mediated cathode. This result
263
suggests that the overall cell performance is mainly limited by the slower electron transfer rate of
264
EAPC-GOx-based anode. Due to this smaller overpotential, the enzymatic biofuel cell with both
265
BQ- and ABTS-mediated electrodes generates a much higher current density output as shown in
266
Figure 6a.
us
cr
ip t
260
Figure 6b shows the power density-current (P-I) curves of enzymatic biofuel cells. The
268
power density output of the biofuel cell with just ABTS-mediated cathode does not improve
269
much compared to the biofuel cells without the mediators. Consequently, the maximum power
270
density outputs of the biofuel cells with no mediators and with just ABTS-mediated cathode are
271
3.1 and 5.7 μW/cm2, respectively. However, the biofuel cells with BQ- and ABTS-mediated
272
electrodes containing EAPC-GOx and EAPC-LAC showed the maximum power density up to
273
37.4 μW/cm2 due to its decreased overpotential mainly offered by the improved electron transfer
274
rate of the anode in the presence of the mediator.
M
ed
ce pt
276
3.4. Performance stability of enzymatic biofuel cells
Ac
275
an
267
277
To confirm the performance stability of enzymatic biofuel cells with EAPC-GOx and
278
EAPC-LAC electrodes, the enzyme electrodes containing EAPC-GOx and EAPC-LAC were
279
incubated in the aqueous buffer under two different temperatures (room temperature and 4 ˚C)
280
over 28 days (Fig. 7). In a 7-day interval, these enzyme electrodes were taken out from the
281
incubation and integrated into our biofuel cell shown in Figure 5 to measure its maximum power
282
density using 200 mM glucose solution at ambient temperature. After the biofuel cell test, both 13
Page 14 of 26
EAPC-GOx and EAPC-LAC electrodes were removed from the cell and washed with 100 mM
284
PB solution (pH 7.0 for the anode and pH 6.5 for the cathode) to remove any residue glucose
285
solution before they were placed back into the incubation. When incubated over 28 days, the
286
maximum power density output of the biofuel cell with the electrodes stored at room temperature
287
maintained 54% of its initial value, while the biofuel cell with the electrodes stored at 4 ˚C
288
maintained 70% of its initial value. As seen in Figure 7, there is almost no difference in the
289
maximum power density of biofuel cell between two different temperatures used for the thermal
290
stability tests within the 7 days. However, as the thermal stability test continues beyond 7 days,
291
the power density output of the biofuel cell with the electrodes stored at 4 ˚C shows a lower
292
deactivation rate than the biofuel cell with the electrodes stored at room temperature. Overall, the
293
biofuel cell with EAPC-GOx and EAPC-LAC electrodes shows a good performance stability,
294
which agrees with their stability results shown in Figure 3.
295
4.
Conclusions
ce pt
296
ed
M
an
us
cr
ip t
283
In this study, we introduced biofuel cells with anode electrode and cathode electrode based
298
on immobilized GOx and LAC on PANFs in the form of enzyme adsorption, precipitation and
299
crosslinking (EAPC). EAPC approach demonstrates that both the loading and stability of GOx
300
and LAC on PANFs can be significantly improved by introducing the enzyme precipitation and
301
crosslinking steps for their immobilization processes. By applying EAPC method to the
302
enzymatic biofuel cell, we have achieved both high power density output and improved
303
performance stability. As shown in successful application to biofuel cells, it is anticipated that
304
EAPC method can be utilized for various types of enzyme-based electrochemical applications
305
such as biosensors and enzyme logic gates.
306
Ac
297
14
Page 15 of 26
307
Acknowledgement
308
This work was supported by the grant from the Agency for Defense Development (ADD -14-70-
309
04-01).
ip t
310
References
312
[1] Barton SC, Gallaway J, Atanassov P. Enzymatic biofuel cells for Implantable and microscale
313
devices. Chem Rev 2004;104:4867-86.
314
[2] Osman MH, Shah AA, Walsh FC. Recent progress and continuing challenges in bio-fuel cells.
315
Part I: Enzymatic cells. Biosens Bioelectron 2011;26:3087-102.
316
[3] MacVittie K, Halamek J, Halamkova L, Southcott M, Jemison WD, Lobel R, et al. From
317
"cyborg" lobsters to a pacemaker powered by implantable biofuel cells. Energy Environ Sci
318
2013;6:81-6.
319
[4] Halamkova L, Halamek J, Bocharova V, Szczupak A, Alfonta L, Katz E. Implanted Biofuel
320
Cell Operating in a Living Snail. J Am Chem Soc 2012;134:5040-3.
321
[5] Szczupak A, Halamek J, Halamkova L, Bocharova V, Alfonta L, Katz E. Living battery -
322
biofuel cells operating in vivo in clams. Energy Environ Sci 2012;5:8891-5.
323
[6] Rasmussen M, Ritzmann RE, Lee I, Pollack AJ, Scherson D. An Implantable Biofuel Cell for
324
a Live Insect. J Am Chem Soc 2012;134:1458-60.
325
[7] Yan YM, Zheng W, Su L, Mao LQ. Carbon-nanotube-based glucose/O-2 biofuel cells. Adv
326
Mater 2006;18:2639-43.
327
[8] Kim J, Jia HF, Wang P. Challenges in biocatalysis for enzyme-based biofuel cells. Biotechnol
328
Adv 2006;24:296-308.
329
[9] Kim JB, Grate JW, Wang P. Nanobiocatalysis and its potential applications. Trends
Ac
ce pt
ed
M
an
us
cr
311
15
Page 16 of 26
Biotechnol 2008;26:639-46.
331
[10] Huang JX. Syntheses and applications of conducting polymer polyaniline nanofibers. Pure
332
Appl Chem 2006;78:15-27.
333
[11] Hong SG, Kim HS, Kim J. Highly Stabilized Lipase in Polyaniline Nanofibers for
334
Surfactant-Mediated Esterification of Ibuprofen. Langmuir 2014;30:911-5.
335
[12] Kim H, Lee I, Kwon Y, Kim BC, Ha S, Lee JH, et al. Immobilization of glucose oxidase
336
into polyaniline nanofiber matrix for biofuel cell applications. Biosens Bioelectron
337
2011;26:3908-13.
338
[13] Lee G, Kim J, Lee JH. Development of magnetically separable polyaniline nanofibers for
339
enzyme immobilization and recovery. Enzyme Microb Technol 2008;42:466-72.
340
[14] Ride JP. The Effect of Induced Lignification on the Resistance of Wheat Cell-Walls to
341
Fungal Degradation. Physiol Plant Pathol 1980;16:187-92.
342
[15] Bergmeyer HU, Gawehn, K., Grassl, M. In: Bergmeyer, H.U. (Ed.), in Methods of
343
Enzymatic Analysis., second ed. Academic Press Inc., New York, NY 1974:457-8.
344
[16] Zebda A, Gondran C, Cinquin P, Cosnier S. Glucose biofuel cell construction based on
345
enzyme, graphite particle and redox mediator compression. Sensors Actuators B: Chem
346
2012;173:760-4.
347
[17] www.ncbi.nlm.nih.gov/, PDB ID: 1CF3, (Accessed August 2014).
348
[18] www.ncbi.nlm.nih.gov/, PDB ID: 1GYC, (Accessed August 2014).
349
[19] Mozhaev VV, Melik-nubarov NS, Sergeeva MV, Šikšnis V, Martinek K. Strategy for
350
Stabilizing Enzymes Part One: Increasing Stability of Enzymes via their Multi-Point Interaction
351
with a Support. Biocatal Biotransform 1990;3:179-87.
352
[20] Martinek K, Klibanov AM, Goldmacher VS, Berezin IV. Principles of Enzyme
Ac
ce pt
ed
M
an
us
cr
ip t
330
16
Page 17 of 26
Stabilization .1. Increase in Thermostability of Enzymes Covalently Bound to a Complementary
354
Surface of a Polymer Support in a Multipoint Fashion. Biochim Biophys Acta 1977;485:1-12.
355
[21] Kuwahara T, Ohta H, Kondo M, Shimomura M. Immobilization of glucose oxidase on
356
carbon paper electrodes modified with conducting polymer and its application to a glucose fuel
357
cell. Bioelectrochemistry 2008;74:66-72.
358
[22] Meredith MT, Minteer SD. Biofuel Cells: Enhanced Enzymatic Bioelectrocatalysis. Annu
359
Rev Anal Chem 2012;5:157-79.
360
[23] Morozova OV, Shumakovich GP, Shleev SV, Yaropolov YI. Laccase-mediator systems and
361
their applications: A review. Appl Biochem Microbiol 2007;43:523-35.
362
[24] Lau KT, de Fortescu SAL, Murphy LJ, Slater JM. Disposable glucose sensors for flow
363
injection analysis using substituted 1,4-benzoquinone mediators. Electroanalysis 2003;15:975-81.
364
[25] Fei JF, Song HK, Palmore GTR. A biopolymer composite that catalyzes the reduction of
365
oxygen to water. Chem Mater 2007;19:1565-70.
Ac
ce pt
ed
M
an
us
cr
ip t
353
17
Page 18 of 26
Figure Captions
367
Figure 1. Schematic illustrations for three different enzyme immobilization methods using
368
PANFs: enzyme adsorption (EA), enzyme adsorption and crosslinking (EAC) and enzyme
369
adsorption, precipitation and crosslinking (EAPC). Magnified figure of EAPC represents
370
crosslinking between enzymes molecules.
371
Figure 2. SEM images of (a) EA-LAC, (b) EAC-LAC, (c) EAPC-LAC, (d) EA-GOx, (e) EAC-
372
GOx and (f) EAPC-GOx. The scale bars in all images correspond to 1 μm.
373
Figure 3. The activities of (a) EA-LAC, EAC-LAC and EAPC-LAC, and (b) EA-GOx, EAC-
374
GOx and EAPC-GOx.
375
Figure 4. Stabilities of EA-LAC, EAC-LAC and EAPC-LAC on PANFs at room temperature.
376
Figure 5. Schematic illustrations of biofuel cells consisting of EAPC-GOx (anode electrode) and
377
EAPC-LAC (cathode electrode).
378
Figure 6. (a) The voltage-current (V-I) curves and (b) the power density-current (P-I) curves of
379
enzymatic biofuel cell containing EAPC-GOx and EAPC-LAC with and without mediators (BQ
380
for the anode mediator and ABTS for the cathode mediator).
381
Figure 7. Relative maximum power density of biofuel cells containing EAPC-GOx (anode) and
382
EAPC-LAC (cathode), which were stored at room temperature and 4 ˚C over 28 days.
Ac
ce pt
ed
M
an
us
cr
ip t
366
18
Page 19 of 26
an
us
cr
ip t
383
M
384
Fig. 1. Schematic illustrations for three different enzyme immobilization methods using PANFs:
386
enzyme adsorption (EA), enzyme adsorption and crosslinking (EAC) and enzyme adsorption,
387
precipitation and crosslinking (EAPC). Magnified figure of EAPC represents crosslinking
388
between enzymes molecules.
Ac
ce pt
ed
385
19
Page 20 of 26
ip t cr us an M ed ce pt Ac
389 390
Fig. 2. SEM images of (a) EA-LAC, (b) EAC-LAC, (c) EAPC-LAC, (d) EA-GOx, (e) EAC-
391
GOx and (f) EAPC-GOx. The scale bars in all images correspond to 1 μm.
20
Page 21 of 26
ce pt
ed
M
an
us
cr
ip t
392
Ac
393 394
Fig. 3. The activities of (a) EA-LAC, EAC-LAC and EAPC-LAC, and (b) EA-GOx, EAC-GOx
395
and EAPC-GOx.
21
Page 22 of 26
ip t cr us
396
ce pt
ed
M
an
Fig. 4. Stabilities of EA-LAC, EAC-LAC and EAPC-LAC on PANFs at room temperature.
Ac
397
22
Page 23 of 26
ip t cr us
398
Fig. 5. Schematic illustrations of biofuel cells consisting of EAPC-GOx (anode electrode) and
400
EAPC-LAC (cathode electrode).
Ac
ce pt
ed
M
an
399
23
Page 24 of 26
ip t cr us an M ed ce pt
401
Fig. 6. (a) The voltage-current (V-I) curves and (b) the power density-current (P-I) curves of
403
enzymatic biofuel cell containing EAPC-GOx and EAPC-LAC with and without mediators (BQ
404
for the anode mediator and ABTS for the cathode mediator)..
Ac
402
24
Page 25 of 26
ip t cr us
405
Fig. 7. Relative maximum power density of biofuel cells containing EAPC-GOx (anode) and
407
EAPC-LAC (cathode), which were stored at room temperature and 4 ˚C over 28 days.
Ac
ce pt
ed
M
an
406
25
Page 26 of 26
