Novel Deep-Eutectic-Solvent-Infused Carbon Nanofiber Networks as

Subscriber access provided by Kaohsiung Medical University

Energy, Environmental, and Catalysis Applications

Novel Deep-Eutectic-Solvent-Infused Carbon Nanofiber Networks as High Power Density Green Battery Cathodes Koki Kawase, Jyunichiro Abe, Mizuki Tenjimbayashi, Yuta Kobayashi, Keisuke Takahashi, and Seimei Shiratori ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b03099 • Publication Date (Web): 24 Apr 2018 Downloaded from http://pubs.acs.org on April 25, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Novel Deep-Eutectic-Solvent-Infused Carbon Nanofiber Networks as High Power Density Green Battery Cathodes Koki Kawase, † Jyunichiro Abe, † Mizuki Tenjimbayashi, † Yuta Kobayashi, † Keisuke Takahashi, † Seimei Shiratori †* †

Center for Material Design Science, School of Integrated Design Engineering, Keio University,

3-14-1 Hiyoshi, Yokohama, 223-8522, Japan. *

Corresponding author, email: [email protected]

KEYWORDS: redox flow battery; carbon nanofiber; deep eutectic solvent; wettability; electrospinning

ABSTRACT

Redox flow batteries (RFBs) have emerged as a promising candidate for large-scale energy storage because of the flexible design for high energy, power, and safety. In this study, FeCl3·6H2O/urea composite deep eutectic catholyte (FeU-DEC)-infused self-standing carbon nanofiber (CNF) was synthesized for green and high power density RFB through industrially

ACS Paragon Plus Environment

1

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 27

available processes. FeU-DEC-infused CNF displayed an extremely high power density (874 mW/g) as well as high capacity (27.28 mAh/g) derived from high theoretical capacity of FeUDEC (89.24 mAh/g) in addition to the advantages of the FeU-DEC characteristics (e.g., nonflammable, biodegradable, facile preparation). This is because of the large electroactive area derived from the high surface area of CNF and superlyophilicity of FeU-DEC on CNFs. Furthermore, we compared the wettability of CNF with other electrodes, as well as the chemical stability and electrode performance, based on topological wetting analysis using parameters of fiber radius, fiber interval, the equilibrium contact angle of FeU-DEC on electrodes and surface tension of FeU-DEC, giving wetting threshold for FeU-DEC on fibrous electrodes. The wetting analysis are not only applied for FeU-DEC, but also for a wide-range of other DECs and deep eutectic anolyte. This work contributes to the further development of green and highperformance RFBs.

INTRODUCTION

Energy storage technology is one of the most noteworthy advancing fields in the 21st century.1,2 The supply of energy by using renewable energy sources such as wind power and solar energy is highly required to solve the energy problem such as diminishing fossil fuels.3–5 However, it is difficult to constantly generate renewable energy because renewable energy depends on the natural environment such as weather.6 Therefore, large-scale energy storage with low cost and safety is important for constant energy supply.7–9


2

Page 3 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Redox flow batteries (RFBs), which can be designed to have either high energy or high power depending on the required application and have a long cycle life and are safe, are now emerging as a promising candidate for large-scale energy storage.10–12 In RFBs, the power density is mainly determined by the electrode, which provides the electrochemically active area by conducting electricity to the active species in the electrolyte.13 Conversely, the energy density is mainly determined by the electrolyte, which dissolves the electroactive species. Owing to the features, RFBs have some advantages compared with other batteries:14–16 First, RFBs can store a large amount of energy and high power by the appropriate design of the electrolyte and electrode, respectively. Second, RFBs have a long cycle life because the simple redox reactions do not impart physical and chemical degradation such as destruction of the electrode caused by a change in the morphology, composition, and crystallinity of the electrode and decomposition of the electrolyte. Vanadium RFBs have been the most common redox flow battery because of simple reactions of vanadium, safety, long cycle life and flexible design. However, there are still challenges that vanadium is scarce, expensive and toxic. Electrolytes have also challenges. Aqueous media has been mainly used as an electrolyte for RFBs.16 Although aqueous electrolytes have been used owing to their high ion conductivity and nonflammability, there are still challenges such as low operation voltage limited by water electrolysis and low energy density owing to the limitation of electroactive species that can be dissolved in an aqueous electrolyte.11,17 Although organic electrolytes are still in the early stage of development, they are attractive because RFBs containing organic solvents can have a higher operation voltage and high energy density by the appropriate choice of the electroactive species.18,19 The practical use of organic solvents is inhibited by their low ion conductivity, flammability, and toxicity to the environment.20


3


Page 4 of 27

Deep eutectic solvents (DESs), which can be categorized into deep eutectic anolytes and deep eutectic catholytes (DECs), are one of the most promising electrolytes containing electroactive species because DESs are usually nonflammable, biodegradable, and easy to prepare.21–23. Although DESs have similar properties to ionic liquids, DESs are much cheaper than ionic liquids.23 However, owing to the limitation of the amount of electroactive material that can be dissolved, the energy density of DESs is still low. Recently, Wang et al. proposed FeCl3·6H2O/urea composite deep eutectic catholyte (FeUDEC).24 The proposed battery (positive: FeU-DEC with a carbon-pasted titanium mesh (Ti mesh) / negative: organic electrolyte with lithium metal) had an excellent energy density of 330 Wh/kg (based on FeU-DEC), which, to the best of our knowledge, is the highest in DEC-based batteries.25–27 The high energy density was attributed to the high potential of Fe3+/Fe2+ (0.77 V vs. SHE) and high concentration of active materials (Fe3+/Fe2+). Furthermore, the FeU-DEC was composed of materials that are abundant, nonflammable, and easy to prepare, which is important to realize green, large-scale, and low-cost battery. Thus, the FeU-DEC showed the potential of designing green and high energy density RFBs. Therefore, the development of FeU-DEC-based RFB materials is strongly required. In particular, FeU-DEC-based RFBs with high power density as well as low cost and scalable fabrication has not been achieved. First, that is because, it was reported that the insufficient lyophilicity of the FeU-DEC on the carbon-pasted Ti mesh electrode caused low mass transfer and high charge transfer resistance.24 Second, Ti is a relatively expensive metal, which is typically used as an electrode because it is chemically stable to FeU-DEC, which is a strong acid. Third, the high weight density of Ti decreases the overall energy density of the cathode.


4

Page 5 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Here, we synthesized FeU-DEC-infused electrospun self-standing carbon nanofiber (CNF) networks as advanced RFB cathodes. The RFB using FeU-DEC-infused CNF was stable and had a high power density owing to the large electroactive space derived from the superlyophilicity of FeU-DEC on CNF and the high surface area of CNF. Moreover, because of the self-standing feature of the CNFs, expensive and heavy metal collectors (e.g. Ti) were not required.28,29 The cathodes were cheap and light and the electrospinning fabrication method was facile and scalable.30–37 Therefore, FeU-DEC-infused electrospun self-standing CNFs enable high power density in addition to the high energy density owing to FeU-DEC and the choice of materials that are non-flammable, abundant and inexpensive and scalable fabrication process is suitable for industrial applications. In this work, first, we compared the CNF performance with other electrodes in terms of chemical stability and wettability. Then, we analyzed the required conditions for achieving superlyophilicity of FeU-DEC on electrodes based on the topological wetting analysis using robustness factors, which provided an insight for the selection of efficient electrodes based on their wettability. To achieve green and safe battery, the wetting analysis is not only useful for DEC but also for deep eutectic anolyte by changing parameters of surface tension and weight density of liquid, though an extensive future development of DEA is required. Finally, we showed that the battery performance of CNF-based RFBs was superior to RFBs based on other electrodes. This work also contributes to the further development of green and high-performance RFBs by investigating the superwetting condition of FeU-DEC on the electrodes.


5


Page 6 of 27

EXPERIMENTAL SECTION

Materials. Polyacrylonitrile (PAN, average Mw ~150,000) was purchased from Sigma-Aldrich (St. Louis, MO, USA). N,N-dimethylformamide (DMF; 99.5%) was purchased from Tokyo Chemical Industry Co. Inc. (Tokyo, Japan). Plastic syringes and needles (21G ½) were purchased from Terumo (Tokyo, Japan). LiPF6 (1 M) in ethylene carbonate (EC)/dimethyl carbonate (DMC) (1:1 v/v%) solution was purchased from Kishida Chemical (Osaka, Japan). Lithium conductive glass plate composed of Li1+x+y(Al,Y)x(Ti,Ge)2-xSiyP3-yO12 (LATP plate) was purchased from Ohara. Inc.(Japan). Carbon paper (CP) was purchased from Hohsen Co. (Osaka, Japan). Titanium mesh (Ti mesh), aluminum (Al), copper (Cu), and stainless (SUS) was purchased from Nilaco Co. (Tokyo, Japan). Graphite (particle size: < 20 µm) was purchased from Sigma-Aldrich.

Preparation of CNF. The procedure to prepare CNF is shown in Figure 1(a). (I) PAN nanofiber was fabricated using electrospinning: PAN (10 wt. %) was dissolved in DMF with stirring for at least 24 h at 60 °C. The solution was loaded into a plastic syringe and Al substrates were mounted on a metal collector. The applied voltage was set to 10 kV, the distance between the needle tip and the collector was set to 15 cm, the solution flow rate was 1.0 mL/h, the time was 10 h, and the humidity was maintained at 30–40%. Fabricated PAN nanofibers were removed from Al foil. (II) The electrospun PAN nanofiber was stabilized in air at 280 °C for 2 h (heating rate: 1 °C/min). (III) The stabilized nanofiber was carbonized in a N2 atmosphere at 1000°C for 1 h (heating rate: 5 °C/min) and self-standing electrospun CNF was obtained.

Synthesis of FeU-DEC. The FeU-DEC was prepared by stirring FeCl3·6H2O and urea in a molar ratio of 2:1 on a hot plate (55°C), as shown in Figure 1(b). The melting point of the mixture


6

Page 7 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


gradually decreased and became a dark-brown liquid. Finally, a homogeneous FeU-DEC was obtained.

Design of a battery based on FeU-DEC-infused CNF. A schematic diagram of the design of the battery is shown in Figure 1(c). The CNF was fully soaked in FeU-DEC and then employed as a cathode. The CNF was sandwiched between metal plates and connected to metal wire. Lithium metal was employed as an anode in EC/ DMC (1:1 v/v%) solution containing 1 M of LiPF6. The cathode and anode were separated by a LATP plate. For comparison of the CNF, Carbon pasted Ti mesh (C-Ti) was prepared by pressing the mixture of the graphite and PVDF-HFP in a weight

Figure 1. Experimental procedures. (a) Schematic models for preparing CNF including electrospinning of polyacrylonitrile in DMF (I) and two annealing steps (stabilization and carbonization) (II) and (III). (b) Photographs of the preparation of FeU-DEC by mixing FeCl3·6H2O and urea in a molar ratio of 2:1. The mixtures became a dark-brown liquid after stirring. (c) Schematic models of the design of a FeU-DECbased battery with CNF. The battery was composed of FeU-DEC, CNF (cathode), LATP, lithium, and LiPF6 in EC/DMC (Anode).

ratio of 1:1 on a circular Ti mesh. Commercial carbon paper (CP) was also prepared.


7


Page 8 of 27

Characterization of electrodes. The flammability of the FeU-DEC and organic electrolyte (LiPF6 (1 M) in EC/ DMC (1:1 v/v%) solution) were confirmed by keeping a flame close to them using a commercial lighter. The chemical stability of electrodes was confirmed by casting FeUDEC on CNF, CP, Al, Cu, SUS, and Ti. Pictures of the electrodes were taken immediately after the FeU-DEC was cast. After three days, the electrodes were rinsed with ethanol and analyzed by photography and SEM. The contact angles of the FeU-DEC on the electrodes were measured by casting 5 µL of FeU-DEC on the electrodes and analyzing the images obtained with a digital camera with Image J software (Wayne Rasband). The surface tension of the liquids was measured using a surface tensiometer (CBVP-Z, Kyowa Interface Science Co., Saitama, Japan). The surface morphology of the electrodes was determined with a field emission scanning electron microscope (SEM, S-4700, Hitachi, Tokyo, Japan) and analyzed with Image J software.

Electrochemical measurements. In all the electrochemical measurements, the battery was kept in a constant temperature oven. The battery was discharged and charged in the voltage range of 2.0 to 4.5 V using an ECstat-301 (EC FRONTIER Co., Ltd, Kyoto, Japan). The rate performance was measured at 25°C by changing the current densities from 10 mA/g to 300 mA/g based on the weight of only electrodes every 10 min. The electrochemical impedance spectroscopy (EIS) measurements were performed from 1.0 Hz to 200,000 Hz using an impedance analyzer (IM 353-01, HIOKI E.E. Co., Nagano, Japan). The AC components was 0.1 V at open circuit voltage (3.8 V). The full discharge was measured at 40°C with 300 mA/g based on the electrode weight of CNF and CP, and 50 mA/g based on the carbon weight of C-Ti. The solid phase formed in the discharge was observed by a digital microscope (VHX-1000, KEYENCE, Osaka, Japan).


8

Page 9 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


RESULTS AND DISCUSSION Characterization of FeU-DEC and electrodes


9


Page 10 of 27

The obtained FeU-DEC was homogeneous liquid and any peaks derived from FeCl3·6H2O and urea was not seen (Figure S1). The FeU-DEC was nonflammable, whereas the organic electrolyte composed of EC and DEC was flammable, as shown in Figure 2(a)-(b). The degradation temperature of FeU-DEC was tested on a thermo-gravimetric analysis (Figure S2). Dehydration can be occurred within 200°C and the decomposition of urea and chloride composites can be occurred at about 350°C. In this research, all of the cathode materials were nonflammable, which is significant for the safety of the battery. To choose suitable electrodes for FeU-DEC-based batteries, the chemical stability of the materials making contact with FeU-DEC were investigated by casting FeU-DEC on CNF, CP, Al, Cu, SUS, and Ti (Figure 2(a)). Al, Cu, SUS reacted and were dissolved by FeU-DEC, whereas CNF, CP, and Ti were not dissolved. The SEM images of the CNF before and three days after FeU-DEC was cast showed the fiber structure was not broken (Figure S3). The difference in the stability of the electrodes was also


10

Page 11 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


examined by open circuit voltage (OCV) of the FeU-DEC-based batteries (Figure S4). The FeUDEC-based batteries with CNF, CP, and C-Ti showed an OCV of approximately 3.8 V, which

Figure 2. Properties of FeU-DEC and the electrodes. Photographs showing (a) the nonflammability of FeU-DEC and (b) flammability of the organic electrolyte (LiPF6 in EC/DMC). (c) Chemical stability of the electrodes against FeU-DEC. The photograph were taken immediately after (upper) and 3 days after (lower) DEC was cast on the electrodes.

suits the theoretical voltage of Fe2+/Fe3+.24 Conversely, the FeU-DEC-based battery with SUS, which was dissolved by FeU-DEC, showed an OCV of 2.8 V, which indicated the reaction between FeU-DEC and SUS changed the voltage. These results indicated that the electrodes that can be used in FeU-DEC were limited

to carbon materials or expensive metals such as Ti.

Therefore, CNF, CP, and C-Ti were selected as electrode materials for the next selection experiments.

Wettability control of electrodes with a carbon-based surface


11


Page 12 of 27

The wettability of the FeU-DEC on the electrodes was investigated. The wettability of FeU-DEC on the electrodes is a crucial factor for the electrochemical performance because insufficient wetting leads to low mass transfer and an increase in the charge-transfer resistance.24 In particular, the FeU-DEC has a high viscosity and is difficult to fully contact with the electrode. However, the wettability control of the electrodes is important for improving electrochemical performance. Figure 3(a) (and Figure S5) shows the wetting behavior of 5 µL FeU-DEC droplets on CNF, CP, and C-Ti. The FeU-DEC droplets on the CNF absorbed and spread and reached superlyophilicity (contact angle of < 1˚) within 3 s after casting. The FeU-DEC droplets on CP gradually absorbed and an equilibrium contact angle of 52° was obtained after approximately 5 s. The C-Ti showed an equilibrium contact angle of 18° and the contact angle did not change with time. The lyophilicity depends on the rough surface structure of C-Ti (Figure 3(b)), which obeys Wenzel’s rule.38 In the experiment, the samples were all carbon-based electrodes and the differences in wettability could be explained by the surface morphology. CNF and CP have a fibrous structure with different fiber radius and interval, as shown in Figure 3(b)-(c). The average fiber radius and interval of CNF were 148 ± 14 nm and 17.5 ± 10.2 µm, respectively, whereas the average fiber radius and interval of CP were 3.8 ± 0.3 µm and 46.5 ± 16.9 µm, respectively. In general, the wettability was strongly influenced by the solid morphology.33 Figure 3(d) shows schematic models of the wetting behavior between FeU-DEC and fibrous electrodes (i.e., CNF and CP). We discuss the wetting behavior using two parameters: (I) the robustness factor A* for “absorption” (vertical direction); and (II) pressure threshold Phorizontal for “spreading” (horizontal direction), as described in equation (1) and (2), respectively.39–43


12

Page 13 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


∗ =

1 − cos = 1 + 2 sin #$ % & = −

'&( ) *+, 2 -

in which R is the fiber radius, D is the fiber interval, ./0 =

(1)

(2) 2'&( , = 4'56 ⁄ρg is the 3 1 2

capillary length for FeU-DEC ('56 is the surface tension of FeU-DEC, ρ is the density of FeUDEC and g is the acceleration owing to gravity), is the equilibrium contact angle, and f; is the

air fraction. γ56 is 65.2 ± 0.4 mN/m measured with a surface tensiometer and γ=6 is 51.4 mN/m.

44

The theoretical was calculated by the Berthelot-Young equation as shown in equation (3) to

be 39.1°.45

'=6 cos = −1 + 2> '56

(3)

The robustness factor is the ratio of Pbreakthrough and reference pressure, Pref. The reference pressure is close to the minimum pressure, which is applied to the solid surface by liquids. Therefore, the FeU-DEC was absorbed in the fibrous electrodes that had an A* 1 (log10 A*> 0) prevented the absorption of FeUDEC. For spreading of FeU-DEC, Phorizontal < 0 was required for FeU-DEC to spread by capillary force. Therefore, a of 39.1° was enough for the FeU-DEC to spread. The smaller fiber interval gives a stronger capillary force. Therefore, in the electrodes with non-oriented fibers, the FeUDEC spread from the space where the fiber interval was narrow (small D value) to the space where the fiber interval was wide (large D value). A* and Phorizontal include the parameters R and D related to the fibrous morphology. A fibrous morphology leading to log10 A*< 0 and Phorizontal < 0 are required for complete wetting of fibrous electrodes with FeU-DEC.


13


Page 14 of 27

A summary of the distribution of vertical wettability (log10 A*) as a function of R and D is shown in Figure 3(e). For CNF, at R = 148 nm, D ≥ 16.3 µm is required to achieve log10 A*< 0. The 45% of whole fibers satisfied D ≥ 16.3 µm based on the histogram as show in Figure 3(c). In terms of spreading, Phorizontal was negative (

Novel Deep-Eutectic-Solvent-Infused Carbon Nanofiber Networks as

Recommend Documents