Biosensors and Bioelectronics 74 (2015) 142–149
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
An“ON–OFF” switchable power output of enzymatic biofuel cell controlled by thermal-sensitive polymer Yun Chen a,1, Panpan Gai a,1, Jingjing Xue a, Jian-Rong Zhang a,b,n, Jun-Jie Zhu a,nn a
State Key Laboratory of Analytical Chemistry for Life Science and Collaborative Innovation Center of Chemistry for Life Sciences, School of Chemistry & Chemical Engineering, Nanjing University, Nanjing 210093, PR China b School of Chemistry and Life Science, Nanjing University Jinling College, Nanjing 210089, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 10 February 2015 Received in revised form 12 June 2015 Accepted 14 June 2015 Available online 20 June 2015
A novel “ON–OFF” switchable enzymatic biofuel cell (EBFC), controlled by in situ thermal-stimuli signal, has been consciously designed. Poly (N-isopropylacrylamide) (PNIPAm) chains were used to act as “ON” and “OFF” channels. Consecutively switching of temperature below and above the lower critical solution temperature (LCST), the reversible conformation changing of the PNIPAm chains between superhydrophilicity and superhydrophobicity was achieved, which constructed the “ON” and “OFF” channel for the transfer of the electrochemical probe to the underlying electrode correspondingly. Gold nanoparticles (AuNPs) protected glucose oxidase and laccase were successfully entrapped into the intelligent thermal-sensitive PNIPAm chains, and performed as the catalysts for the oxidation of glucose and the reduction of O2, respectively. Below the LCST, the fuels and the mediator could access to the catalytic centers of enzymes (set as “ON” state); while above the LCST, the reaction was impeded because the process of reactant transmission was blocked (set as “OFF” state). By switching the “valve” of mass transfer, the fabricated EBFC displayed the obvious “ON–OFF” controllable behavior. At the “ON” state, the open circuit voltage (Ecellocv) and maximal power output density (Pmax) could reach to 0.70 V and 20.52 μW cm 2, respectively; while at the “OFF” state, the Ecellocv and Pmax were only 0.30 V and 3.28 μW cm 2 correspondingly. The switchable process was repeatable, and the response time was only several minutes. & 2015 Elsevier B.V. All rights reserved.
Keywords: “ON–OFF” switchable power output Enzymatic biofuel cell Poly (N-isopropylacrylamide) Lower critical solution temperature Gold nanoparticles
1. Introduction Enzymatic biofuel cell (EBFC) is the promising power device, (Calabrese Barton et al., 2004; Schröder, 2012) because a wide variety of fuels for EBFCs, especially glucose and O2, are available in living beings (Cooney et al., 2008). Recently, the EBFC designed with reversible switchable power, which is controlled by the exoteric stimulation, is the potential power device in biomedical field. With the external physical or chemical stimuli changes (input signal), the switchable EBFC with the active or inactive highly specific catalytic interface could display the corresponding specific recognition reactions (output response). Katz's group initiated the amazing field of applying logic gates to the design of EBFC devices. n Corresponding author at: State Key Laboratory of Analytical Chemistry for Life Science and Collaborative Innovation Center of Chemistry for Life Sciences, School of Chemistry & Chemical Engineering, Nanjing University, Nanjing 210093, PR China. Fax: þ 86 25 83317761. nn Corresponding author. 1 These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.bios.2015.06.028 0956-5663/& 2015 Elsevier B.V. All rights reserved.
They presented the first prototype of the EBFC controlled by complicated biochemical reactions, and it could deliver power ondemand responding to the logical calculation based on the input physiological signals (Amir et al., 2009). It is their smart designs of using logic gate controlled EBFC for medical diagnostics. For example, based on the EBFC controlled by Boolean Logic calculation, a drug-release technology for diagnose and therapy is developed (Zhou et al., 2012). At present, the design of switchable EBFC always introduces biochemical substances or modules, such as aptamer, (Zhou et al., 2010) immunereaction (Tam et al., 2009) and enzyme logic systems (Amir et al., 2009). Among them, the integration of enzymes into a “smart” polymer could initiate the artificial models of biological redox systems in EBFCs. It is reported that polymer brushes tethered to electrode surfaces can effectively control the interfacial properties, being switchable between shrunken and swollen states, which is motivated by the external pH value (Tam et al., 2009). However, the main defect of the switchable EBFC based on the pH sensitive polymer-brush-modified electrodes is that the corresponding response was not fast enough due to the slow
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
143
Scheme 1. The scheme of the switchable EBFC for controlling the power release by temperature signals.
change of the pH in solution. It is discovered that poly (N-isopropylacrylamide) (PNIPAm), a thermal-sensitive polymer, could reversible switching between superhydrophilicity and superhydrophobicity in a narrow temperature range of about 10 °C, (Sun et al., 2004) and this polymer exhibits a lower critical solution temperature (LCST, above which the polymer becomes insoluble) in water at about 33 °C (Huber et al., 2003). Furthermore, because of its biocompatibility (Klis et al., 2009) and the sharp property change in response to the moderate thermal stimulus near physiological temperature, (Zhao et al., 2012) PNIPAm must be a good material to bridge the gap between biological machines and multifunctional actuators. In this paper, an EBFC with a switchable power release controlled by thermal signals has been designed (Scheme 1). Transitions between the shrunken and swollen states can be controlled by changing the temperature of the solution, yielding the swollen (“ON” state) and shrunken (“OFF” state) of the polymer brush. Upon swelling of the polymer brush (20 °C), the electrode surface became exposed to the solution, and the polymer-bound redox centers can catalyze redox reaction of fuels; while the polymer brush was at the shrunken state (45 °C), the electrode surface for communicating with fuels and redox mediator was effectively blocked. The switchable process of the EBFC was highly repeatable, and the response time was only several minutes, which was obviously faster than the EBFC which was controlled by pH-sensitive polymer. Due to the support of AuNPs for enzymes, the maximal power output of the EBFC at the "ON" state could keep stable. It should be noted that the study introduces a proof of concept to utilize thermal-sensitive polymer to control the power output of EBFC.
2. Experimental 2.1. Chemicals N-isopropylacrylamide (NIPAm), N,N′-methylenebisacryl-amide (BIS), Na2S2O8, 2,2-azinobis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS), and laccase from Trametes versicolor (EC 1.10.3.2, 4 20 units mg 1) were purchased from SigmaAldrich. Glucose oxidase (GOD) from Aspergillus niger (EC 1.1.3.4, 294 units mg 1) was purchased from Sanland, and both of the enzymes were used as received without further purification. HAuCl4 4H2O was purchased from Shanghai Chemical Reagent Co. Ltd. (Shanghai, China). Glucose and NaNO3 were obtained from Sinopharm. The glucose stock solution (1 M) was prepared at least 24 h before use. 0.2 M acetic acid buffer solution (pH 5.0) was made from acetic acid and sodium acetate anhydrous. Aqueous solutions were prepared with ultrapure water from an Elix 5 Pure Water System (4 18 MΩ cm). 2.2. Instrumentation The Au substrates (0.5 cm2) were purchased from the 55th Institute of China Electronic Group (Nanjing, China). Before using, the Au substrates were carefully scraped to a mirror finish using pledget, then, they were rinsed and sonicated by ethanol and ultrapure water, respectively, and dried under nitrogen flow. The morphology of the AuNPs and PNIPAm were characterized by Atomic force microscope (AFM, Agilent 5500) using tapping mode. For measuring the topography of AuNPs, the mica substrate was used to make sure the surface of the background was smooth enough. For measuring the topography of PNIPAm, the Au
144
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
substrate was used. For testing the morphology of the AuNPs/GOD hybrid and AuNPs/laccase hybrid, a 200 kV FEI Tecnai F20 TEM in Nanjing University Center for Electron Microscopy was used. Electrochemical measurements were performed using a workstation (CHI 660D). Cyclic voltammetric measurements were performed with a traditional three-electrode system including a Pt wire electrode as the counter electrode, an SCE as the reference electrode, and the Au substrate modified electrode as the working electrode. The open circuit potentials of the working electrodes were tested with a two-electrode configuration (SCE as the reference electrode). 2.3. Fabrication of AuNPs, bioanode and biocathode AuNPs was fabricated according to a previously published procedure (Yan et al., 2008). The PNIPAm thin film was prepared by an electrochemical-induced radical polymerization method based on the reference with mild modification (Zhou et al., 2007). In a typical process, NIPAm (1.25 M), BIS (0.05 M), NaNO3 (0.25 M), and Na2S2O8 (0.0125 M) were dissolved in 4 mL of the ultrapure water in sequence, and then the solution was degassed by nitrogen for 60 min at 20 °C. After that, the cyclic voltammetry was performed for electrochemical polymerization of PNIPAm at Au substrate electrode in a sweep potential range from 0.4 V to 1.4 V for 120 cycles at 50 mV s 1. During the polymerization, for the first cycle, a reduction peak at about 0.9 V was observed (Fig. S1 in Electronic Supplementary Information, ESI), which should be attributed to the reduction of Na2S2O8 (Song and Hu, 2010). With increase in the number of scanning cycles, the peak current of Na2S2O8 reduction decreased gradually, which indicated the diffusion of Na2S2O8 to Au substrate electrode was impeded, and electrochemically induced free radical polymerization of NIPAm on the electrode surface was successfully realized. After the polymerization of NIPAm finished, the PNIPAm electrode was washed with ultrapure water to get rid of the monomers. For fabrication of the PNIPAm-AuNPs/GOD bioanode or PNIPAmAuNPs/laccase biocathode, the conditions for the polymerization of NIPAm were the same as the method described above but 500 μL AuNPs solution including 10 mg GOD or 10 mg laccase was added into the reaction solution. The enzymes were entrapped into the gel structure during the polymerization process, which was controlled by the polymerization time. 2.4. Design and test of the EBFC The design of the EBFC model was displayed in Scheme 1 and Fig. S2. The perfluorosulfonic acid/PTFE copolymer membrane s (DuPontTM Nafion PFSA NRE-211), with thickness of 25.4 μm, was
used to separate the anodic and cathodic compartments. The bioanode was PNIPAm-AuNPs/GOD modified Au electrode, and anolyte was 30 mL 0.2 M acetic acid buffer solution (pH 5.0) containing 10 mM of glucose and before using, the electrolyte was degassed with high purity nitrogen for at least 1 h. The biocathode was PNIPAm-AuNPs/laccase modified Au electrode, and the oxygen-saturated 30 mL 0.2 M acetic acid buffer solution (pH 5.0) containing 0.5 mM of ABTS was used as the catholyte, which was bubbled with high purity O2 at lease 1 h. After that, the electrodes and electrolytes were put into the corresponding chambers, and the EBFC model was sealed with PTFE tape, which was shown in Fig. S2. Ecellocv of the EBFC was measured using a CHI-660D electrochemical station. In order to test the polarization and power output density curves, after a stable Ecellocv of the EBFC was observed, the variable external load ranged from 100 Ω to 100 kΩ was connected in series between anode and cathode. Then the power outputs were obtained with a precision digital multimeter. The polarization and power output density curves were obtained at both 20 and 45 °C.
3. Results and discussion 3.1. The characterization of AuNPs, PNIPAm, AuNPs/GOD hybrid and AuNPs/laccase hybrid AFM was firstly used to observe the surface topography of the AuNPs and PNIPAm. To assured the accuracy of the experiment, the initial surface of the substrate was detected and observed very smooth (Fig. S3A). AFM testing (Fig. S3B) revealed that the morphology of the prepared AuNPs were with the size of around 3– 4 nm in diameter, which was also corresponding to TEM testing (Fig. S4). AuNPs serve as excellent biocompatible surfaces for the immobilization of enzymes since the interaction between amino and cysteine groups of proteins with AuNPs is as strong as that of the commonly used thiols (Ansari and Husain, 2012). The morphology of the AuNPs/GOD hybrid and AuNPs/laccase hybrid was monitored by HRTEM. Fig. 1A and B showed that the AuNPs with the size around 3–4 nm incorporated into the enzymes, although the single enzyme unit was not observed due to the aggregation. It unambiguously demonstrated the formation of the AuNPs/enzymes hybrids. AFM image in Fig. S5 displayed that the surface of the PNIPAm hydrogel modified electrode was in an uniform porous structure. Because of its porous structure and amide group in each constitutional unit, PNIPAm hydrogel is highly affinitive to enzyme with carboxy group (Zhao et al., 2012). AuNPs/GOD hybrid and AuNPs/laccase hybrid should be entraped into PNIPAm hydrogel during the polymerization of NIPAm and immobilized onto Au substrate electrode (Klis et al., 2009).
Fig. 1. HRTEM images of (A) AuNPs/GOD hybrid and (B) AuNPs/laccase hybrid.
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
145
Fig. 2. (A) CVs of the PNIPAm modified Au electrode at variable temperatures from 20 to 50 °C, (B) (e)–(g) with enlargement. (C) CVs of the same electrode in the reversed temperatures range. (D) The relationship of reduction peak currents and temperatures. Data were recorded in 1 mM Fe(CN)63 /4 plus 0.5 M KNO3 aqueous solution. Scan rate: 50 mV s 1.
3.2. The thermally sensitive “ON–OFF” characters of the PNIPAm interface In order to investigate the thermoresponsive behavior of the PNIPAm, Fe(CN)63 /4 was selected as the electrochemical redox probe to analyze the state of the interface. Fig. 2A and B displays the cyclic voltammograms (CV) of Fe(CN)63 /4 probe at PNIPAm modified electrode in a temperatures range from 20 °C to 50 °C. When the temperature of the electrolyte was controlled at 20 °C, Fe(CN)63 /4 showed a well-defined and nearly reversible redox peak with the formal potential at about 0.20 V. However, the CV signals were greatly suppressed when the temperature increases to 35 °C. Compared to the CV signals at 20 °C, the CV signals at 35 °C reduced about 7 times. This was because the PNIPAm had a reversible lower critical solution temperature (LCST¼35 °C) in aqueous solution. When the temperature is below LCST, because there was the strong hydrogen bonding between the amide groups in PNIPAm and water molecules in electrolyte, which caused water molecules into the polymer and expansion,(Sun et al., 2004; Zhou et al., 2007) the PNIPAm chains tended to taking on a swollen coil structure in solution,(Sheeney-Haj-Ichia et al., 2002) and performed the hydrophilic nature of the interface. The Fe(CN)63 /4 probe could easily diffuse to Au substrate electrode through the PNIPAm matrix, leading to the larger CV response;(Zhou et al., 2007) When the temperature was more than LCST, the hydrogen bond interaction between PNIPAm and water was damaged, and water was squeezed out from the polymer and the intramolecular hydrogen bond in the polymer was formed, which caused the polymer structure collapse and shrinkage. Thus, the PNIPAm became hydrophobic and the insulative (Sun et al., 2004; Zhou et al., 2007). The contracted structure would retard the Fe(CN)63 /4 probe to diffuse through the PNIPAm and reach the surface of the
Au substrate electrode, resulting in the very small CV signal of Fe (CN)63 /4 . Fig. 2C also showed the CV results of Fe(CN)63 /4 probe at the same electrode from 50 °C to 20 °C, and they were almost the same as the results shown in Fig. 2A, which was also confirmed by the relationship of reduction peak currents and temperature in Fig. 2D. Definitely, there is no hysteresis effect, and the equilibration time from one temperature to another in both directions was 5 min. The experiments demonstrated that the electrochemical activity and inertia of the PNIPAm modified electrode could be controlled by temperature, which indicated thermally responsive “ON/OFF” switching characteristics of the EBFC would be achieved. In addition, the polymer is attractive in biological applications because the temperature lies between room temperature and body temperature. 3.3. The characters of the temperature sensitive bioanode When the PNIPAm polymer expanded or contracted with temperature, did the reactivity of the entrapped GOD in the polymer change? It could be illustrated by measurement of the open circuit potential of bioanode (Ea°cp), which was performed in 0.2 M acetic acid buffer solution (pH 5.0) containing 10 mM glucose. Eaocp was recorded after the temperature was stabilized, and the result was shown in Fig. 3A. It showed that the onset of Eaocp was –0.12 V at 45 °C (“OFF” state), then, the Eaocp dropped to the GOD formal potential of –0.32 V when the temperature of the electrolyte decreased to 20 °C (“ON” state), and control experiment showed that Eaocp only displayed the slight response to the temperature, which demonstrated that the entrapped AuNPs/GOD hybrid in PNIPAm polymer still maintained its catalytic activity for glucose as shown in Fig. 3D. The repeatable switch from “ON” to “OFF” state could be realized within 400 s. Fig. 3B showed the Eaocp
146
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
Fig. 3. (A) The temperature sensitive switching for Eaocp (black) and control experiment (blue) between 20 °C and 45 °C in the electrolyte containing 10 mM glucose. (B) Eaocp recorded in above mentioned electrolyte at (a) 45 °C and (b) 20 °C, control experiment at (d) 45 °C and (c) 20 °C. (C) CVs of the PNIPAm modified Au electrode and the bioanode at 20 °C and 45 °C. (D) CVs of the bioanode in the electrolyte with glucose concentration of (a) 0 mM and (b) 1 mM at 20 °C. The scan rate for the CVs was 50 mV s 1, and the electrolyte was 0.2 M acetic acid buffer solution (pH 5.0) saturated with N2. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
of the as-prepared bioanode at constant temperature. It demonstrated that Eaocp kept stable at the value of –0.12 V in the “OFF” state, and at the value of –0.32 V in the “ON” state, while without loading of GOD, the Eaocp was at the value of –0.05 V in the “OFF” state, and at the value of –0.09 V in the “ON” state, which were in consistent with those shown in Fig. 3A. The catalytic activity of the prepared bioanode for glucose was also illustrated by cyclic voltammetry. Fig. 3C showed the CVs of the PNIPAm electrode and the PNIPAm-AuNPs/GOD bioanode at 20 and 45 °C, respectively. At 20 °C, the bioanode showed a couple of well-defined redox peaks with the redox peak current of 2.16 μA at –0.30 V and –0.34 V (the formal potential of –0.32 V), which can be ascribed to the direct redox of GOD. It demonstrated that GOD was successfully entrapped into the PNIPAm chains. When temperature was at 45 °C, the redox peak current of the bioanode increased to 5.09 μA, which was because a collapsed and shrunken PNIPAm chains caused more bound GOD molecules accessed to the surface of the electrode. Compared to the PNIPAm modified electrode, because of no GOD in the electrode, there was no obvious difference in cyclic voltammograms when temperature changed from 20 to 45 °C. It also indicated that the redox peaks were attributed to the active center of GOD molecules. Fig. 3D showed the oxidation peak current of the bioanode increased from 2.16 to 6.92 μA after 1 mM glucose was added into the electrolyte solution at 20 °C, while the reduction peak current decreased to 1.35 μA, which was indicative of a typical catalytic process for glucose oxidation. In addition, the oxidation of glucose began at about –0.32 V, which demonstrated that the Eaocp should be close to –0.32 V (Holland et al., 2011). The current–time curve technique was also used to identify the catalytic behavior of the bioanode for glucose oxidation. Fig. S6 displays the typical amperometric response of the bioanode with successive injections of glucose to a stirring testing solution at an applied potential of –
0.3 V. It was clear that the amperometric response successively increased with the addition of glucose, and the steady-state current platform was reached rapidly, which indicated that the bioanode possessed a quick response to the target molecule. However, when the temperature was more than LCST, the steadystate current platform was difficult to reach and the steady-state current obviously reduced. The reason could be explained as follows: when the temperatures was below LCST, due to the expansion and hydrophilicity of the PNIPAm polymer, the permeation of glucose in the solution was permitted to arrive at the active center of GOD, and there was no electron transfer barrier between the glucose and the electrode; however, when the temperatures was more than LCST, the collapsed conformation of PNIPAm chains exhibited hydrophobicity, as a result, the mass transfer of glucose molecule was blocked by PNIPAm. All in all, the reversible activation/inactivation of the bioanode for the biocatalytic oxidation of glucose can be controlled by temperature stimulation, and the “ON–OFF” response was sensitive, and repeatable. 3.4. The characters of the temperature sensitive biocathode When the PNIPAm polymer expanded or contracted with temperature, the test for the reactivity of the entrapped laccase in the polymer was similar to the bioanode. The glucose in the electrolyte was replaced by the saturated O2 and 0.5 mM ABTS. Ecocp was recorded as shown in Fig. 4A. It displayed that the onset of Ecocp was about 0.40 V at 20 °C (“ON” state). The Ecocp changed to near 0.20 V when the temperature of the electrolyte increased to 45 °C (“OFF” state). The repeatable switch from “ON” state to “OFF” state could be realized within 1000 s under the optimal conditions. However, control experiment showed that Ecocp only displayed the weak response to the temperature, which demonstrated the entrapped AuNPs/laccase hybrid in PNIPAm polymer
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
147
Fig. 4. (A) The temperature sensitive switching for Ecocp (black) and control experiment (blue) between 20 °C and 45 °C in the electrolyte containing 0.5 mM ABTS saturated with O2. (B) Ecocp recorded in above mentioned electrolyte at (a) 45 °C and (b) 20 °C, and control experiment at (c) 45 °C and (d) 20 °C. (C) CVs of the PNIPAm modified electrode and the biocathode in the electrolyte saturated with N2 at 20 °C and 45 °C. (D) CVs of the biocathode in the electrolyte containing 0.5 mM ABTS, (a) saturated with N2 at 20 °C, (b) saturated with O2 at 45 °C, and (c) saturated with O2 at 20 °C. The scan rate for the CVs was 50 mV s 1, and the electrolyte was 0.2 M acetic acid buffer solution (pH 5.0). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
still maintained its catalytic activity for O2 as shown in Fig. 4D. Fig. 4B displayed the Ecocp of the as-prepared biocathode at the constant temperature. It demonstrated that Ecocp kept stable at the value of 0.40 V in the “ON” state, and at the value of 0.20 V in the “OFF” state, while without loading of laccase, the Ecocp was at the value of 0.28 V in the “OFF” state, and at the value of 0.30 V in the “ON” state, which were in consistent with those shown in Fig. 4A as well. Also, the catalytic activity of the prepared biocathode for O2 was illustrated by cyclic voltammetry. Fig. 4C showed the CVs of the PNIPAm electrode and the PNIPAm-AuNPs/laccase biocathode at 20 and 45 °C, respectively. At 20 °C, the bioanode showed two couple of redox peaks, which could be ascribed to laccase (Klis et al., 2009; Pita et al., 2006; Shleev et al., 2006). It demonstrated that laccase was successfully entrapped into the PNIPAm chains. When temperature was at 45 °C, the redox peak (at higher potential) current of laccase increased from 2.37 to 4.95 μA, which was because a collapsed and shrunken PNIPAm chains caused more bound laccase molecules accessed to the surface of the Au electrode. Compared to the PNIPAm modified electrode, because of no laccase in the electrode, there was no obvious difference in cyclic voltammograms when temperature changed from 20 to 45 °C. It also indicated that the redox peaks were attributed to the bound laccase. ABTS was found to be the best electron mediator in the cathode of biofuel cells due to having a relatively high formal potential (Liu et al., 2005). However, Fig. 4D showed that after the electrolyte solution was saturated with O2 at 20 °C, the reduction peak current of the biocathode corresponding to direct electron transfer towards the T2 site of laccase increased from 1 to 64 μA, while the oxidation peak current disappeared. This result was indicative of a non-mediated catalytic process for O2 reduction to peroxide production (Pita et al., 2006; Shleev et al., 2006).
However, when the temperature was more than LCST, the O2 reduction current obviously reduced. The reason was similar to glucose oxidation in bioanode. Although the shrinkage of the PNIPAm chains can obviously decrease the efficiency of the catalytic reaction, it does not lead to the denaturation of laccase. Therefore, the reversible activation/inactivation of the biocathode for the catalytic oxidation of O2 can be also controlled by temperature stimulation. 3.5. The characters of the temperature sensitive EBFC The EBFC was constructed by the bioanode and biocathode described above and the temperature-dependent “ON–OFF” behavior could be characterized by its open circuit voltage (Ecellocv) curve at two typical temperatures, 20 and 45 °C. The measurement for Ecellocv was shown in Fig. 5A. As expected, the Ecellocv reached to 0.70 V when temperature was below LCST, while the Ecellocv droped to around 0.30 V when temperature was more than the LCST. The values of Ecellocv were almost equal to the difference between Ecocp and Eaocp at 20 and 45 °C, respectively. This thermo-sensitive “ON– OFF” behavior was quite reversible. By successively switching the temperature of the testing solution between 20 and 45 °C, the Ecellocv could be repeatedly cycled between a high value (0.70 V) and relative low value (0.30 V). However, control experiment showed that Ecellocv only showed the weak response to the temperature. Fig. 5B displayed the Ecellocv of the EBFC at the constant temperature, which indicated that Ecellocv kept stable for a long time at a value of 0.70 V in the “ON” state, and at a value of 0.30 V in the “OFF” state. While without loading of enzymes, the Ecellocv was at the value of 0.46 V in the “OFF” state, and at the value of 0.42 V in the “ON” state. It demonstrated the Ecellocv of the EBFC could be switched by changing the environmental temperature.
148
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
Fig. 5. (A) The temperature sensitive switching for Ecellocv (black) and control experiment (red) between 20 °C and 45 °C. High voltage was at 20 °C, and low voltage was at 45 °C. (B) The Ecellocv recorded at (a) 20 °C and (b) 45 °C, and control experiment at (d) 20 °C and (c) 45 °C. (C) Polarization curve (a) and power density curve (b) of the EBFC at 20 °C, and polarization curve (c) and power density curve (d) of the EBFC at 45 °C. (D) Relationship between the power density and external loading resistance at (a) 20 °C and (b) 45 °C. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 5C showed the polarization curve and the power density curve of the EBFC at “ON” and “OFF” state, respectively. At “ON” state, the power density as a function of the cell current density for the EBFC presented the typical bellshaped curve (Zebda et al., 2011) as shown from curve b in Fig. 5C. Thus, the maximum power output, Pmax, were estimated to be 20.52 μW cm 2 (relative to the geometric area of the Au substrate) for the model of EBFC; at “OFF” state, Pmax were estimated to be 3.28 μW cm 2 from curve d in Fig. 5C. The ratio of Pmax-ON and Pmax-OFF was more than 6-folds. When the EBFC reached to the maximum power output, the external load was equal to rohm. Therefore, rohm was also evaluated to be about 8000 Ω at “ON” state, and about 20,000 Ω at “OFF” state from Fig. 5D, respectively. 3.6. The stability of the EBFC As the energy device, reasonable lifetime for portable applications(Calabrese Barton et al., 2004) and low capacity loss under open circuit conditions(Chaudhuri and Lovley, 2003) are of great importance. EBFCs suffer from a very prominent disadvantage for long-term operation, due to loss in enzyme activity.(Fishilevich et al., 2009; Liu et al., 2010) The Ecellocv of the EBFC at “ON” state (20 °C) was continuously recorded over 40 days. Both glucose and O2 were renewed every two days during the 40 days in order to exclude the effect of the substrate. The result was shown in Fig. S7A, which revealed that the Ecellocv could maintain 92% of the maximal Ecellocv. The relationship between Pmax-ON of the EBFC and operation time was also tested, which was as shown in Fig. S7B. It indicated that a slight loss of the enzymatic activity occurred during the operation of the EBFC, and demonstrated that both the AuNPs and the PNIPAm could provide a suitable microenvironment to maintain the activity and structure of the enzymes.
4. Conclusions In summary, the AuNPs/GOD hybrid and AuNPs/laccase hybrid functionalized PNIPAm were successfully fabricated on the surface of Au substrate electrodes by a one-step electrochemical polymerization, respectively. Upon swelling of the polymer brush (20 °C), the electrode surface became exposed to the solution, and the polymer-bound redox centers can catalyze redox reaction of fuels; while the polymer brush was at the shrunken state (45 °C), the electrode surface for communicating with external soluble fuels and redox mediator was effectively blocked. Based on this unique characterization, a novel thermal-sensitive EBFC was designed, which displayed the sensitive, repeatable “ON–OFF” character when the temperature changed from 20 °C to 45 °C. At “ON” state, the EBFC was with Ecellocv 0.70 V and Pmax 20.52 μW cm 2, respectively; while at “OFF” state, the corresponding Ecellocv and Pmax were only 0.30 V and 3.28 μW cm 2, respectively. The repeatable switch could be realized within several minutes under the optimal conditions.
Acknowledgments We gratefully appreciate the support from the National Natural Science Foundation of China (21175065, 21121091) and the National Basic Research Program (2011CB933502) of China. We also greatly thank Professor Peng Wang and Ph. D. Candidate Si Gao in Nanjing University Center for Electron Microscopy for the HRTEM testing. Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at doi:10.1016/j.bios.2015.06.028.
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
References Amir, L., Tam, T.K., Pita, M., Meijler, M.M., Alfonta, L., Katz, E., 2009. J. Am. Chem. Soc 131 (2), 826–832. Ansari, S.A., Husain, Q., 2012. Biotechnol. Adv. 30, 512–523. Calabrese Barton, S., Gallaway, J., Atanassov, P., 2004. Chem. Rev. 104 (10), 4867–4886. Chaudhuri, S.K., Lovley, D.R., 2003. Nat. Biotechnol. 21 (10), 1229–1232. Cooney, M.J., Svoboda, V., Lau, C., Martin, G., Minteer, S.D., 2008. Energy Environ. Sci. 1 (3), 320–337. Fishilevich, S., Amir, L., Fridman, Y., Aharoni, A., Alfonta, L., 2009. J. Am. Chem. Soc. 131 (34), 12052–12053. Holland, J.T., Lau, C., Brozik, S., Atanassov, P., Banta, S., 2011. J. Am. Chem. Soc. 133 (48), 19262–19265. Huber, D.L., Manginell, R.P., Samara, M.A., Kim, B.I., Bunker, B.C., 2003. Science 301 (5631), 352–354. Klis, M., Karbarz, M., Stojek, Z., Rogalski, J., Bilewicz, R., 2009. J. Phys. Chem. B 113 (17), 6062–6067. Liu, C., Alwarappan, S., Chen, Z.F., Kong, X.X., Li, C.Z., 2010. Biosens. Bioelectron. 25 (7), 1829–1833. Liu, Y., Wang, M.K., Zhao, F., Liu, B.F., Dong, S.J., 2005. Chem. Eur. J. 11 (17), 4970–4974.
149
Pita, M., Shleev, S., Ruzgas, T., Fernandez, V.M., Yaropolov, A.I., Gorton, L., 2006. Electrochem. Commun. 8 (5), 747–753. Schröder, U., 2012. Angew. Chem. Int. Ed. 51 (30), 7370–7372. Sheeney-Haj-Ichia, L., Sharabi, G., Willner, I., 2002. Adv. Funct. Mater. 12 (1), 27–32. Shleev, S., Pita, M., Yaropolov, A.I., Ruzgas, T., Gorton, L., 2006. Electroanalysis 18 (19-20), 1901–1908. Song, S.L., Hu, N.F., 2010. J. Phys. Chem. B 114 (17), 5940–5945. Sun, T.L., Wang, G.J., Feng, L., Liu, B.Q., Ma, Y.M., Jiang, L., Zhu, D.B., 2004. Angew. Chem. Int. Ed. 43 (3), 357–360. Tam, T.K., Strack, G., Pita, M., Katz, E., 2009. J. Am. Chem. Soc. 131 (33), 11670–11671. Yan, W., Feng, X., Chen, X., Hou, W., Zhu, J.J., 2008. Biosens. Bioelectron. 23 (7), 925–931. Zebda, A., Gondran, C., Le Goff, A., Holzinger, M., Cinquin, P., Cosnier, S., 2011. Nat. Commun. 2, 370. Zhao, X., Liu, Y., Lu, J., Zhou, J., Li, J., 2012. Chem. Eur. J. 18 (12), 3687–3694. Zhou, J.H., Liu, J., Wang, G., Lu, X.B., Wen, Z.H., Li, J.H., 2007. Adv. Funct. Mater. 17 (16), 3377–3382. Zhou, M., Du, Y., Chen, C.G., Li, B.L., Wen, D., Dong, S.J., Wang, E.K., 2010. J. Am. Chem. Soc. 132 (7), 2172–2174. Zhou, M., Zhou, N.D., Kuralay, F., Windmiller, J.R., Parkhomovsky, S., Valdes-Ramirez, G., Katz, E., Wang, J., 2012. Angew. Chem. Int. Ed. 51 (11), 2686–2689.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
An“ON–OFF” switchable power output of enzymatic biofuel cell controlled by thermal-sensitive polymer Yun Chen a,1, Panpan Gai a,1, Jingjing Xue a, Jian-Rong Zhang a,b,n, Jun-Jie Zhu a,nn a
State Key Laboratory of Analytical Chemistry for Life Science and Collaborative Innovation Center of Chemistry for Life Sciences, School of Chemistry & Chemical Engineering, Nanjing University, Nanjing 210093, PR China b School of Chemistry and Life Science, Nanjing University Jinling College, Nanjing 210089, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 10 February 2015 Received in revised form 12 June 2015 Accepted 14 June 2015 Available online 20 June 2015
A novel “ON–OFF” switchable enzymatic biofuel cell (EBFC), controlled by in situ thermal-stimuli signal, has been consciously designed. Poly (N-isopropylacrylamide) (PNIPAm) chains were used to act as “ON” and “OFF” channels. Consecutively switching of temperature below and above the lower critical solution temperature (LCST), the reversible conformation changing of the PNIPAm chains between superhydrophilicity and superhydrophobicity was achieved, which constructed the “ON” and “OFF” channel for the transfer of the electrochemical probe to the underlying electrode correspondingly. Gold nanoparticles (AuNPs) protected glucose oxidase and laccase were successfully entrapped into the intelligent thermal-sensitive PNIPAm chains, and performed as the catalysts for the oxidation of glucose and the reduction of O2, respectively. Below the LCST, the fuels and the mediator could access to the catalytic centers of enzymes (set as “ON” state); while above the LCST, the reaction was impeded because the process of reactant transmission was blocked (set as “OFF” state). By switching the “valve” of mass transfer, the fabricated EBFC displayed the obvious “ON–OFF” controllable behavior. At the “ON” state, the open circuit voltage (Ecellocv) and maximal power output density (Pmax) could reach to 0.70 V and 20.52 μW cm 2, respectively; while at the “OFF” state, the Ecellocv and Pmax were only 0.30 V and 3.28 μW cm 2 correspondingly. The switchable process was repeatable, and the response time was only several minutes. & 2015 Elsevier B.V. All rights reserved.
Keywords: “ON–OFF” switchable power output Enzymatic biofuel cell Poly (N-isopropylacrylamide) Lower critical solution temperature Gold nanoparticles
1. Introduction Enzymatic biofuel cell (EBFC) is the promising power device, (Calabrese Barton et al., 2004; Schröder, 2012) because a wide variety of fuels for EBFCs, especially glucose and O2, are available in living beings (Cooney et al., 2008). Recently, the EBFC designed with reversible switchable power, which is controlled by the exoteric stimulation, is the potential power device in biomedical field. With the external physical or chemical stimuli changes (input signal), the switchable EBFC with the active or inactive highly specific catalytic interface could display the corresponding specific recognition reactions (output response). Katz's group initiated the amazing field of applying logic gates to the design of EBFC devices. n Corresponding author at: State Key Laboratory of Analytical Chemistry for Life Science and Collaborative Innovation Center of Chemistry for Life Sciences, School of Chemistry & Chemical Engineering, Nanjing University, Nanjing 210093, PR China. Fax: þ 86 25 83317761. nn Corresponding author. 1 These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.bios.2015.06.028 0956-5663/& 2015 Elsevier B.V. All rights reserved.
They presented the first prototype of the EBFC controlled by complicated biochemical reactions, and it could deliver power ondemand responding to the logical calculation based on the input physiological signals (Amir et al., 2009). It is their smart designs of using logic gate controlled EBFC for medical diagnostics. For example, based on the EBFC controlled by Boolean Logic calculation, a drug-release technology for diagnose and therapy is developed (Zhou et al., 2012). At present, the design of switchable EBFC always introduces biochemical substances or modules, such as aptamer, (Zhou et al., 2010) immunereaction (Tam et al., 2009) and enzyme logic systems (Amir et al., 2009). Among them, the integration of enzymes into a “smart” polymer could initiate the artificial models of biological redox systems in EBFCs. It is reported that polymer brushes tethered to electrode surfaces can effectively control the interfacial properties, being switchable between shrunken and swollen states, which is motivated by the external pH value (Tam et al., 2009). However, the main defect of the switchable EBFC based on the pH sensitive polymer-brush-modified electrodes is that the corresponding response was not fast enough due to the slow
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
143
Scheme 1. The scheme of the switchable EBFC for controlling the power release by temperature signals.
change of the pH in solution. It is discovered that poly (N-isopropylacrylamide) (PNIPAm), a thermal-sensitive polymer, could reversible switching between superhydrophilicity and superhydrophobicity in a narrow temperature range of about 10 °C, (Sun et al., 2004) and this polymer exhibits a lower critical solution temperature (LCST, above which the polymer becomes insoluble) in water at about 33 °C (Huber et al., 2003). Furthermore, because of its biocompatibility (Klis et al., 2009) and the sharp property change in response to the moderate thermal stimulus near physiological temperature, (Zhao et al., 2012) PNIPAm must be a good material to bridge the gap between biological machines and multifunctional actuators. In this paper, an EBFC with a switchable power release controlled by thermal signals has been designed (Scheme 1). Transitions between the shrunken and swollen states can be controlled by changing the temperature of the solution, yielding the swollen (“ON” state) and shrunken (“OFF” state) of the polymer brush. Upon swelling of the polymer brush (20 °C), the electrode surface became exposed to the solution, and the polymer-bound redox centers can catalyze redox reaction of fuels; while the polymer brush was at the shrunken state (45 °C), the electrode surface for communicating with fuels and redox mediator was effectively blocked. The switchable process of the EBFC was highly repeatable, and the response time was only several minutes, which was obviously faster than the EBFC which was controlled by pH-sensitive polymer. Due to the support of AuNPs for enzymes, the maximal power output of the EBFC at the "ON" state could keep stable. It should be noted that the study introduces a proof of concept to utilize thermal-sensitive polymer to control the power output of EBFC.
2. Experimental 2.1. Chemicals N-isopropylacrylamide (NIPAm), N,N′-methylenebisacryl-amide (BIS), Na2S2O8, 2,2-azinobis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS), and laccase from Trametes versicolor (EC 1.10.3.2, 4 20 units mg 1) were purchased from SigmaAldrich. Glucose oxidase (GOD) from Aspergillus niger (EC 1.1.3.4, 294 units mg 1) was purchased from Sanland, and both of the enzymes were used as received without further purification. HAuCl4 4H2O was purchased from Shanghai Chemical Reagent Co. Ltd. (Shanghai, China). Glucose and NaNO3 were obtained from Sinopharm. The glucose stock solution (1 M) was prepared at least 24 h before use. 0.2 M acetic acid buffer solution (pH 5.0) was made from acetic acid and sodium acetate anhydrous. Aqueous solutions were prepared with ultrapure water from an Elix 5 Pure Water System (4 18 MΩ cm). 2.2. Instrumentation The Au substrates (0.5 cm2) were purchased from the 55th Institute of China Electronic Group (Nanjing, China). Before using, the Au substrates were carefully scraped to a mirror finish using pledget, then, they were rinsed and sonicated by ethanol and ultrapure water, respectively, and dried under nitrogen flow. The morphology of the AuNPs and PNIPAm were characterized by Atomic force microscope (AFM, Agilent 5500) using tapping mode. For measuring the topography of AuNPs, the mica substrate was used to make sure the surface of the background was smooth enough. For measuring the topography of PNIPAm, the Au
144
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
substrate was used. For testing the morphology of the AuNPs/GOD hybrid and AuNPs/laccase hybrid, a 200 kV FEI Tecnai F20 TEM in Nanjing University Center for Electron Microscopy was used. Electrochemical measurements were performed using a workstation (CHI 660D). Cyclic voltammetric measurements were performed with a traditional three-electrode system including a Pt wire electrode as the counter electrode, an SCE as the reference electrode, and the Au substrate modified electrode as the working electrode. The open circuit potentials of the working electrodes were tested with a two-electrode configuration (SCE as the reference electrode). 2.3. Fabrication of AuNPs, bioanode and biocathode AuNPs was fabricated according to a previously published procedure (Yan et al., 2008). The PNIPAm thin film was prepared by an electrochemical-induced radical polymerization method based on the reference with mild modification (Zhou et al., 2007). In a typical process, NIPAm (1.25 M), BIS (0.05 M), NaNO3 (0.25 M), and Na2S2O8 (0.0125 M) were dissolved in 4 mL of the ultrapure water in sequence, and then the solution was degassed by nitrogen for 60 min at 20 °C. After that, the cyclic voltammetry was performed for electrochemical polymerization of PNIPAm at Au substrate electrode in a sweep potential range from 0.4 V to 1.4 V for 120 cycles at 50 mV s 1. During the polymerization, for the first cycle, a reduction peak at about 0.9 V was observed (Fig. S1 in Electronic Supplementary Information, ESI), which should be attributed to the reduction of Na2S2O8 (Song and Hu, 2010). With increase in the number of scanning cycles, the peak current of Na2S2O8 reduction decreased gradually, which indicated the diffusion of Na2S2O8 to Au substrate electrode was impeded, and electrochemically induced free radical polymerization of NIPAm on the electrode surface was successfully realized. After the polymerization of NIPAm finished, the PNIPAm electrode was washed with ultrapure water to get rid of the monomers. For fabrication of the PNIPAm-AuNPs/GOD bioanode or PNIPAmAuNPs/laccase biocathode, the conditions for the polymerization of NIPAm were the same as the method described above but 500 μL AuNPs solution including 10 mg GOD or 10 mg laccase was added into the reaction solution. The enzymes were entrapped into the gel structure during the polymerization process, which was controlled by the polymerization time. 2.4. Design and test of the EBFC The design of the EBFC model was displayed in Scheme 1 and Fig. S2. The perfluorosulfonic acid/PTFE copolymer membrane s (DuPontTM Nafion PFSA NRE-211), with thickness of 25.4 μm, was
used to separate the anodic and cathodic compartments. The bioanode was PNIPAm-AuNPs/GOD modified Au electrode, and anolyte was 30 mL 0.2 M acetic acid buffer solution (pH 5.0) containing 10 mM of glucose and before using, the electrolyte was degassed with high purity nitrogen for at least 1 h. The biocathode was PNIPAm-AuNPs/laccase modified Au electrode, and the oxygen-saturated 30 mL 0.2 M acetic acid buffer solution (pH 5.0) containing 0.5 mM of ABTS was used as the catholyte, which was bubbled with high purity O2 at lease 1 h. After that, the electrodes and electrolytes were put into the corresponding chambers, and the EBFC model was sealed with PTFE tape, which was shown in Fig. S2. Ecellocv of the EBFC was measured using a CHI-660D electrochemical station. In order to test the polarization and power output density curves, after a stable Ecellocv of the EBFC was observed, the variable external load ranged from 100 Ω to 100 kΩ was connected in series between anode and cathode. Then the power outputs were obtained with a precision digital multimeter. The polarization and power output density curves were obtained at both 20 and 45 °C.
3. Results and discussion 3.1. The characterization of AuNPs, PNIPAm, AuNPs/GOD hybrid and AuNPs/laccase hybrid AFM was firstly used to observe the surface topography of the AuNPs and PNIPAm. To assured the accuracy of the experiment, the initial surface of the substrate was detected and observed very smooth (Fig. S3A). AFM testing (Fig. S3B) revealed that the morphology of the prepared AuNPs were with the size of around 3– 4 nm in diameter, which was also corresponding to TEM testing (Fig. S4). AuNPs serve as excellent biocompatible surfaces for the immobilization of enzymes since the interaction between amino and cysteine groups of proteins with AuNPs is as strong as that of the commonly used thiols (Ansari and Husain, 2012). The morphology of the AuNPs/GOD hybrid and AuNPs/laccase hybrid was monitored by HRTEM. Fig. 1A and B showed that the AuNPs with the size around 3–4 nm incorporated into the enzymes, although the single enzyme unit was not observed due to the aggregation. It unambiguously demonstrated the formation of the AuNPs/enzymes hybrids. AFM image in Fig. S5 displayed that the surface of the PNIPAm hydrogel modified electrode was in an uniform porous structure. Because of its porous structure and amide group in each constitutional unit, PNIPAm hydrogel is highly affinitive to enzyme with carboxy group (Zhao et al., 2012). AuNPs/GOD hybrid and AuNPs/laccase hybrid should be entraped into PNIPAm hydrogel during the polymerization of NIPAm and immobilized onto Au substrate electrode (Klis et al., 2009).
Fig. 1. HRTEM images of (A) AuNPs/GOD hybrid and (B) AuNPs/laccase hybrid.
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
145
Fig. 2. (A) CVs of the PNIPAm modified Au electrode at variable temperatures from 20 to 50 °C, (B) (e)–(g) with enlargement. (C) CVs of the same electrode in the reversed temperatures range. (D) The relationship of reduction peak currents and temperatures. Data were recorded in 1 mM Fe(CN)63 /4 plus 0.5 M KNO3 aqueous solution. Scan rate: 50 mV s 1.
3.2. The thermally sensitive “ON–OFF” characters of the PNIPAm interface In order to investigate the thermoresponsive behavior of the PNIPAm, Fe(CN)63 /4 was selected as the electrochemical redox probe to analyze the state of the interface. Fig. 2A and B displays the cyclic voltammograms (CV) of Fe(CN)63 /4 probe at PNIPAm modified electrode in a temperatures range from 20 °C to 50 °C. When the temperature of the electrolyte was controlled at 20 °C, Fe(CN)63 /4 showed a well-defined and nearly reversible redox peak with the formal potential at about 0.20 V. However, the CV signals were greatly suppressed when the temperature increases to 35 °C. Compared to the CV signals at 20 °C, the CV signals at 35 °C reduced about 7 times. This was because the PNIPAm had a reversible lower critical solution temperature (LCST¼35 °C) in aqueous solution. When the temperature is below LCST, because there was the strong hydrogen bonding between the amide groups in PNIPAm and water molecules in electrolyte, which caused water molecules into the polymer and expansion,(Sun et al., 2004; Zhou et al., 2007) the PNIPAm chains tended to taking on a swollen coil structure in solution,(Sheeney-Haj-Ichia et al., 2002) and performed the hydrophilic nature of the interface. The Fe(CN)63 /4 probe could easily diffuse to Au substrate electrode through the PNIPAm matrix, leading to the larger CV response;(Zhou et al., 2007) When the temperature was more than LCST, the hydrogen bond interaction between PNIPAm and water was damaged, and water was squeezed out from the polymer and the intramolecular hydrogen bond in the polymer was formed, which caused the polymer structure collapse and shrinkage. Thus, the PNIPAm became hydrophobic and the insulative (Sun et al., 2004; Zhou et al., 2007). The contracted structure would retard the Fe(CN)63 /4 probe to diffuse through the PNIPAm and reach the surface of the
Au substrate electrode, resulting in the very small CV signal of Fe (CN)63 /4 . Fig. 2C also showed the CV results of Fe(CN)63 /4 probe at the same electrode from 50 °C to 20 °C, and they were almost the same as the results shown in Fig. 2A, which was also confirmed by the relationship of reduction peak currents and temperature in Fig. 2D. Definitely, there is no hysteresis effect, and the equilibration time from one temperature to another in both directions was 5 min. The experiments demonstrated that the electrochemical activity and inertia of the PNIPAm modified electrode could be controlled by temperature, which indicated thermally responsive “ON/OFF” switching characteristics of the EBFC would be achieved. In addition, the polymer is attractive in biological applications because the temperature lies between room temperature and body temperature. 3.3. The characters of the temperature sensitive bioanode When the PNIPAm polymer expanded or contracted with temperature, did the reactivity of the entrapped GOD in the polymer change? It could be illustrated by measurement of the open circuit potential of bioanode (Ea°cp), which was performed in 0.2 M acetic acid buffer solution (pH 5.0) containing 10 mM glucose. Eaocp was recorded after the temperature was stabilized, and the result was shown in Fig. 3A. It showed that the onset of Eaocp was –0.12 V at 45 °C (“OFF” state), then, the Eaocp dropped to the GOD formal potential of –0.32 V when the temperature of the electrolyte decreased to 20 °C (“ON” state), and control experiment showed that Eaocp only displayed the slight response to the temperature, which demonstrated that the entrapped AuNPs/GOD hybrid in PNIPAm polymer still maintained its catalytic activity for glucose as shown in Fig. 3D. The repeatable switch from “ON” to “OFF” state could be realized within 400 s. Fig. 3B showed the Eaocp
146
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
Fig. 3. (A) The temperature sensitive switching for Eaocp (black) and control experiment (blue) between 20 °C and 45 °C in the electrolyte containing 10 mM glucose. (B) Eaocp recorded in above mentioned electrolyte at (a) 45 °C and (b) 20 °C, control experiment at (d) 45 °C and (c) 20 °C. (C) CVs of the PNIPAm modified Au electrode and the bioanode at 20 °C and 45 °C. (D) CVs of the bioanode in the electrolyte with glucose concentration of (a) 0 mM and (b) 1 mM at 20 °C. The scan rate for the CVs was 50 mV s 1, and the electrolyte was 0.2 M acetic acid buffer solution (pH 5.0) saturated with N2. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
of the as-prepared bioanode at constant temperature. It demonstrated that Eaocp kept stable at the value of –0.12 V in the “OFF” state, and at the value of –0.32 V in the “ON” state, while without loading of GOD, the Eaocp was at the value of –0.05 V in the “OFF” state, and at the value of –0.09 V in the “ON” state, which were in consistent with those shown in Fig. 3A. The catalytic activity of the prepared bioanode for glucose was also illustrated by cyclic voltammetry. Fig. 3C showed the CVs of the PNIPAm electrode and the PNIPAm-AuNPs/GOD bioanode at 20 and 45 °C, respectively. At 20 °C, the bioanode showed a couple of well-defined redox peaks with the redox peak current of 2.16 μA at –0.30 V and –0.34 V (the formal potential of –0.32 V), which can be ascribed to the direct redox of GOD. It demonstrated that GOD was successfully entrapped into the PNIPAm chains. When temperature was at 45 °C, the redox peak current of the bioanode increased to 5.09 μA, which was because a collapsed and shrunken PNIPAm chains caused more bound GOD molecules accessed to the surface of the electrode. Compared to the PNIPAm modified electrode, because of no GOD in the electrode, there was no obvious difference in cyclic voltammograms when temperature changed from 20 to 45 °C. It also indicated that the redox peaks were attributed to the active center of GOD molecules. Fig. 3D showed the oxidation peak current of the bioanode increased from 2.16 to 6.92 μA after 1 mM glucose was added into the electrolyte solution at 20 °C, while the reduction peak current decreased to 1.35 μA, which was indicative of a typical catalytic process for glucose oxidation. In addition, the oxidation of glucose began at about –0.32 V, which demonstrated that the Eaocp should be close to –0.32 V (Holland et al., 2011). The current–time curve technique was also used to identify the catalytic behavior of the bioanode for glucose oxidation. Fig. S6 displays the typical amperometric response of the bioanode with successive injections of glucose to a stirring testing solution at an applied potential of –
0.3 V. It was clear that the amperometric response successively increased with the addition of glucose, and the steady-state current platform was reached rapidly, which indicated that the bioanode possessed a quick response to the target molecule. However, when the temperature was more than LCST, the steadystate current platform was difficult to reach and the steady-state current obviously reduced. The reason could be explained as follows: when the temperatures was below LCST, due to the expansion and hydrophilicity of the PNIPAm polymer, the permeation of glucose in the solution was permitted to arrive at the active center of GOD, and there was no electron transfer barrier between the glucose and the electrode; however, when the temperatures was more than LCST, the collapsed conformation of PNIPAm chains exhibited hydrophobicity, as a result, the mass transfer of glucose molecule was blocked by PNIPAm. All in all, the reversible activation/inactivation of the bioanode for the biocatalytic oxidation of glucose can be controlled by temperature stimulation, and the “ON–OFF” response was sensitive, and repeatable. 3.4. The characters of the temperature sensitive biocathode When the PNIPAm polymer expanded or contracted with temperature, the test for the reactivity of the entrapped laccase in the polymer was similar to the bioanode. The glucose in the electrolyte was replaced by the saturated O2 and 0.5 mM ABTS. Ecocp was recorded as shown in Fig. 4A. It displayed that the onset of Ecocp was about 0.40 V at 20 °C (“ON” state). The Ecocp changed to near 0.20 V when the temperature of the electrolyte increased to 45 °C (“OFF” state). The repeatable switch from “ON” state to “OFF” state could be realized within 1000 s under the optimal conditions. However, control experiment showed that Ecocp only displayed the weak response to the temperature, which demonstrated the entrapped AuNPs/laccase hybrid in PNIPAm polymer
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
147
Fig. 4. (A) The temperature sensitive switching for Ecocp (black) and control experiment (blue) between 20 °C and 45 °C in the electrolyte containing 0.5 mM ABTS saturated with O2. (B) Ecocp recorded in above mentioned electrolyte at (a) 45 °C and (b) 20 °C, and control experiment at (c) 45 °C and (d) 20 °C. (C) CVs of the PNIPAm modified electrode and the biocathode in the electrolyte saturated with N2 at 20 °C and 45 °C. (D) CVs of the biocathode in the electrolyte containing 0.5 mM ABTS, (a) saturated with N2 at 20 °C, (b) saturated with O2 at 45 °C, and (c) saturated with O2 at 20 °C. The scan rate for the CVs was 50 mV s 1, and the electrolyte was 0.2 M acetic acid buffer solution (pH 5.0). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
still maintained its catalytic activity for O2 as shown in Fig. 4D. Fig. 4B displayed the Ecocp of the as-prepared biocathode at the constant temperature. It demonstrated that Ecocp kept stable at the value of 0.40 V in the “ON” state, and at the value of 0.20 V in the “OFF” state, while without loading of laccase, the Ecocp was at the value of 0.28 V in the “OFF” state, and at the value of 0.30 V in the “ON” state, which were in consistent with those shown in Fig. 4A as well. Also, the catalytic activity of the prepared biocathode for O2 was illustrated by cyclic voltammetry. Fig. 4C showed the CVs of the PNIPAm electrode and the PNIPAm-AuNPs/laccase biocathode at 20 and 45 °C, respectively. At 20 °C, the bioanode showed two couple of redox peaks, which could be ascribed to laccase (Klis et al., 2009; Pita et al., 2006; Shleev et al., 2006). It demonstrated that laccase was successfully entrapped into the PNIPAm chains. When temperature was at 45 °C, the redox peak (at higher potential) current of laccase increased from 2.37 to 4.95 μA, which was because a collapsed and shrunken PNIPAm chains caused more bound laccase molecules accessed to the surface of the Au electrode. Compared to the PNIPAm modified electrode, because of no laccase in the electrode, there was no obvious difference in cyclic voltammograms when temperature changed from 20 to 45 °C. It also indicated that the redox peaks were attributed to the bound laccase. ABTS was found to be the best electron mediator in the cathode of biofuel cells due to having a relatively high formal potential (Liu et al., 2005). However, Fig. 4D showed that after the electrolyte solution was saturated with O2 at 20 °C, the reduction peak current of the biocathode corresponding to direct electron transfer towards the T2 site of laccase increased from 1 to 64 μA, while the oxidation peak current disappeared. This result was indicative of a non-mediated catalytic process for O2 reduction to peroxide production (Pita et al., 2006; Shleev et al., 2006).
However, when the temperature was more than LCST, the O2 reduction current obviously reduced. The reason was similar to glucose oxidation in bioanode. Although the shrinkage of the PNIPAm chains can obviously decrease the efficiency of the catalytic reaction, it does not lead to the denaturation of laccase. Therefore, the reversible activation/inactivation of the biocathode for the catalytic oxidation of O2 can be also controlled by temperature stimulation. 3.5. The characters of the temperature sensitive EBFC The EBFC was constructed by the bioanode and biocathode described above and the temperature-dependent “ON–OFF” behavior could be characterized by its open circuit voltage (Ecellocv) curve at two typical temperatures, 20 and 45 °C. The measurement for Ecellocv was shown in Fig. 5A. As expected, the Ecellocv reached to 0.70 V when temperature was below LCST, while the Ecellocv droped to around 0.30 V when temperature was more than the LCST. The values of Ecellocv were almost equal to the difference between Ecocp and Eaocp at 20 and 45 °C, respectively. This thermo-sensitive “ON– OFF” behavior was quite reversible. By successively switching the temperature of the testing solution between 20 and 45 °C, the Ecellocv could be repeatedly cycled between a high value (0.70 V) and relative low value (0.30 V). However, control experiment showed that Ecellocv only showed the weak response to the temperature. Fig. 5B displayed the Ecellocv of the EBFC at the constant temperature, which indicated that Ecellocv kept stable for a long time at a value of 0.70 V in the “ON” state, and at a value of 0.30 V in the “OFF” state. While without loading of enzymes, the Ecellocv was at the value of 0.46 V in the “OFF” state, and at the value of 0.42 V in the “ON” state. It demonstrated the Ecellocv of the EBFC could be switched by changing the environmental temperature.
148
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
Fig. 5. (A) The temperature sensitive switching for Ecellocv (black) and control experiment (red) between 20 °C and 45 °C. High voltage was at 20 °C, and low voltage was at 45 °C. (B) The Ecellocv recorded at (a) 20 °C and (b) 45 °C, and control experiment at (d) 20 °C and (c) 45 °C. (C) Polarization curve (a) and power density curve (b) of the EBFC at 20 °C, and polarization curve (c) and power density curve (d) of the EBFC at 45 °C. (D) Relationship between the power density and external loading resistance at (a) 20 °C and (b) 45 °C. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 5C showed the polarization curve and the power density curve of the EBFC at “ON” and “OFF” state, respectively. At “ON” state, the power density as a function of the cell current density for the EBFC presented the typical bellshaped curve (Zebda et al., 2011) as shown from curve b in Fig. 5C. Thus, the maximum power output, Pmax, were estimated to be 20.52 μW cm 2 (relative to the geometric area of the Au substrate) for the model of EBFC; at “OFF” state, Pmax were estimated to be 3.28 μW cm 2 from curve d in Fig. 5C. The ratio of Pmax-ON and Pmax-OFF was more than 6-folds. When the EBFC reached to the maximum power output, the external load was equal to rohm. Therefore, rohm was also evaluated to be about 8000 Ω at “ON” state, and about 20,000 Ω at “OFF” state from Fig. 5D, respectively. 3.6. The stability of the EBFC As the energy device, reasonable lifetime for portable applications(Calabrese Barton et al., 2004) and low capacity loss under open circuit conditions(Chaudhuri and Lovley, 2003) are of great importance. EBFCs suffer from a very prominent disadvantage for long-term operation, due to loss in enzyme activity.(Fishilevich et al., 2009; Liu et al., 2010) The Ecellocv of the EBFC at “ON” state (20 °C) was continuously recorded over 40 days. Both glucose and O2 were renewed every two days during the 40 days in order to exclude the effect of the substrate. The result was shown in Fig. S7A, which revealed that the Ecellocv could maintain 92% of the maximal Ecellocv. The relationship between Pmax-ON of the EBFC and operation time was also tested, which was as shown in Fig. S7B. It indicated that a slight loss of the enzymatic activity occurred during the operation of the EBFC, and demonstrated that both the AuNPs and the PNIPAm could provide a suitable microenvironment to maintain the activity and structure of the enzymes.
4. Conclusions In summary, the AuNPs/GOD hybrid and AuNPs/laccase hybrid functionalized PNIPAm were successfully fabricated on the surface of Au substrate electrodes by a one-step electrochemical polymerization, respectively. Upon swelling of the polymer brush (20 °C), the electrode surface became exposed to the solution, and the polymer-bound redox centers can catalyze redox reaction of fuels; while the polymer brush was at the shrunken state (45 °C), the electrode surface for communicating with external soluble fuels and redox mediator was effectively blocked. Based on this unique characterization, a novel thermal-sensitive EBFC was designed, which displayed the sensitive, repeatable “ON–OFF” character when the temperature changed from 20 °C to 45 °C. At “ON” state, the EBFC was with Ecellocv 0.70 V and Pmax 20.52 μW cm 2, respectively; while at “OFF” state, the corresponding Ecellocv and Pmax were only 0.30 V and 3.28 μW cm 2, respectively. The repeatable switch could be realized within several minutes under the optimal conditions.
Acknowledgments We gratefully appreciate the support from the National Natural Science Foundation of China (21175065, 21121091) and the National Basic Research Program (2011CB933502) of China. We also greatly thank Professor Peng Wang and Ph. D. Candidate Si Gao in Nanjing University Center for Electron Microscopy for the HRTEM testing. Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at doi:10.1016/j.bios.2015.06.028.
Y. Chen et al. / Biosensors and Bioelectronics 74 (2015) 142–149
References Amir, L., Tam, T.K., Pita, M., Meijler, M.M., Alfonta, L., Katz, E., 2009. J. Am. Chem. Soc 131 (2), 826–832. Ansari, S.A., Husain, Q., 2012. Biotechnol. Adv. 30, 512–523. Calabrese Barton, S., Gallaway, J., Atanassov, P., 2004. Chem. Rev. 104 (10), 4867–4886. Chaudhuri, S.K., Lovley, D.R., 2003. Nat. Biotechnol. 21 (10), 1229–1232. Cooney, M.J., Svoboda, V., Lau, C., Martin, G., Minteer, S.D., 2008. Energy Environ. Sci. 1 (3), 320–337. Fishilevich, S., Amir, L., Fridman, Y., Aharoni, A., Alfonta, L., 2009. J. Am. Chem. Soc. 131 (34), 12052–12053. Holland, J.T., Lau, C., Brozik, S., Atanassov, P., Banta, S., 2011. J. Am. Chem. Soc. 133 (48), 19262–19265. Huber, D.L., Manginell, R.P., Samara, M.A., Kim, B.I., Bunker, B.C., 2003. Science 301 (5631), 352–354. Klis, M., Karbarz, M., Stojek, Z., Rogalski, J., Bilewicz, R., 2009. J. Phys. Chem. B 113 (17), 6062–6067. Liu, C., Alwarappan, S., Chen, Z.F., Kong, X.X., Li, C.Z., 2010. Biosens. Bioelectron. 25 (7), 1829–1833. Liu, Y., Wang, M.K., Zhao, F., Liu, B.F., Dong, S.J., 2005. Chem. Eur. J. 11 (17), 4970–4974.
149
Pita, M., Shleev, S., Ruzgas, T., Fernandez, V.M., Yaropolov, A.I., Gorton, L., 2006. Electrochem. Commun. 8 (5), 747–753. Schröder, U., 2012. Angew. Chem. Int. Ed. 51 (30), 7370–7372. Sheeney-Haj-Ichia, L., Sharabi, G., Willner, I., 2002. Adv. Funct. Mater. 12 (1), 27–32. Shleev, S., Pita, M., Yaropolov, A.I., Ruzgas, T., Gorton, L., 2006. Electroanalysis 18 (19-20), 1901–1908. Song, S.L., Hu, N.F., 2010. J. Phys. Chem. B 114 (17), 5940–5945. Sun, T.L., Wang, G.J., Feng, L., Liu, B.Q., Ma, Y.M., Jiang, L., Zhu, D.B., 2004. Angew. Chem. Int. Ed. 43 (3), 357–360. Tam, T.K., Strack, G., Pita, M., Katz, E., 2009. J. Am. Chem. Soc. 131 (33), 11670–11671. Yan, W., Feng, X., Chen, X., Hou, W., Zhu, J.J., 2008. Biosens. Bioelectron. 23 (7), 925–931. Zebda, A., Gondran, C., Le Goff, A., Holzinger, M., Cinquin, P., Cosnier, S., 2011. Nat. Commun. 2, 370. Zhao, X., Liu, Y., Lu, J., Zhou, J., Li, J., 2012. Chem. Eur. J. 18 (12), 3687–3694. Zhou, J.H., Liu, J., Wang, G., Lu, X.B., Wen, Z.H., Li, J.H., 2007. Adv. Funct. Mater. 17 (16), 3377–3382. Zhou, M., Du, Y., Chen, C.G., Li, B.L., Wen, D., Dong, S.J., Wang, E.K., 2010. J. Am. Chem. Soc. 132 (7), 2172–2174. Zhou, M., Zhou, N.D., Kuralay, F., Windmiller, J.R., Parkhomovsky, S., Valdes-Ramirez, G., Katz, E., Wang, J., 2012. Angew. Chem. Int. Ed. 51 (11), 2686–2689.
