Decoupled Ion Transport in a Protein-Based Solid Ion Conductor - The

Oct 14, 2016 - School of Mechanical and Materials Engineering, Washington State University, Pullman, Washington 99164, United States. J. Phys. Chem...
0 downloads 0 Views 1MB Size
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Letter

Decoupled Ion Transport in a Protein-Based Solid Ion Conductor Xuewei Fu, Yead Jewel, Yu Wang, Jin Liu, and Wei-Hong (Katie) Zhong J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.6b02071 • Publication Date (Web): 14 Oct 2016 Downloaded from http://pubs.acs.org on October 14, 2016

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 free 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 accessible to all readers and 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.

The Journal of Physical Chemistry Letters 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 21

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

The Journal of Physical Chemistry Letters

Decoupled Ion Transport in a Protein-based Solid Ion Conductor Xuewei Fu,† Yead Jewel,† Yu Wang,†,* Jin Liu†,* and Wei-Hong Zhong†,* †

School of Mechanical and Materials Engineering, Washington State University, Pullman, WA

99164 USA * Corresponding Authors [email protected], [email protected] [email protected]

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry Letters

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 21

ABSTRACT. Simultaneous achievement of good electrochemical and mechanical properties is crucial for practical applications of solid ion conductors. Conventional polymer conductors suffer from low conductivity, low transference number and deteriorated mechanical properties with the enhancement of conductivity, resulting from the coupling between ion transport and polymer movement. Here we present a successful fabrication and fundamental understanding of a high performance soy protein-based solid conductor. The conductor shows ionic conductivity of ~ 10-5 S/cm, transference number of 0.94 and modulus of 1 GPa at room temperature, and still remains flexible and easily processable. Molecular simulations indicate that this is due to appropriate manipulation of the protein structures for effective exploitation of protein functional groups. A decoupled transport mechanism, which is able to explain all results, is proposed. The new insights can be utilized to provide guidelines for design, optimization and fabrication of high performance bio-solid conductors.

TOC GRAPHICS

Keywords: Solid ion conductor, Soy protein, Protein-ion complex, Decoupled ion transport, Denaturation

ACS Paragon Plus Environment

2

Page 3 of 21

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

The Journal of Physical Chemistry Letters

Solid-state ion conductors are becoming one of the most important components in electrochemical devices due to the superior stability and safety properties.1-3 Good electrochemical and mechanical properties are required for practical applications of a solid ion conductor. However, in conventional polymeric conductors or solid polymer electrolytes (SPEs) 4-7

the ion transport is dominated by a coupled process that is responsible for the low transference

number and deteriorated modulus with the enhancement of ionic conductivity. In a SPE, both cations and anions of the salts are mobile, and the motion of the cations is highly coordinated by the polymer chains. As a result, the ion transport is strongly coupled with the local segmental motion of polymer chains in amorphous state above the glass transition temperature. Some strategies, such as addition of plasticizers8-11 and blending with other polymers,12-14 have been employed to increase the mobility of the polymer chains, and thus to improve the ionic conductivity. However, the corresponding mechanical properties were significantly reduced because of the increase in molecule mobility that leads to the improvement in ionic conductivity.15-16 Another important issue with the traditional SPEs is the low transference number (~ 0.4) caused by the unconstrained motion of the anions.5, 17 Proteins are natural polymers made out of a series of amino acids covalently linked through peptide bonds. Compared with the synthetic polymers, proteins in general are compositionally and structurally much more complex. Therefore, utilizing protein as the polymer matrix and exploiting its functional groups through proper manipulation of its structures to achieve high performance solid state ion conductors are not only scientifically intriguing but also technically challenging. Recently, some protein (such as gelatin18-20) has been reported as the matrix for polymer ion conductors. Plasticizers have been added to improve ionic conductivity, therefore the issues associated with coupled transport mentioned above still exist, most likely due to a lack

ACS Paragon Plus Environment

3

The Journal of Physical Chemistry Letters

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 4 of 21

of fundamental understanding of the protein-ion interactions and transport mechanisms in protein-based conductors. In this letter, we present a systematic investigation of a solid state soy protein-based ion conductor. A conductor with both excellent electrochemical and mechanical properties is achieved. Simulations have been performed to uncover the detailed molecular interactions and ionic transport mechanisms. The primary matrix material in our ion conductor is denatured soy proteins (d-SP) from soybean. It has been reported that soy proteins contain nearly all types of amino acids21 (Table S1 of Supporting Information). In aqueous environment, the soy proteins are kept at a compact native state with several levels of structures through H-bonding, salt-bridges and hydrophobic interactions. To effectively exploit the interactions from the amino acids on side chains, the soy protein needs to be denatured. In our experiments, the acetic acid/water mixture and heat (95 oC) were utilized to disrupt those interactions and break the protein native structures. As illustrated in Figure 1, a semi-transparent yellow SP solution, i.e. the well-denatured SP was obtained. The denaturation effect was further investigated through the characterization of the SP particle sizes (see SEM and TEM images in Figure S1 of Supporting Information). Because of denaturation, the particle size was significantly reduced from ~ 50 µm to ~ 30 nm (see particle size distribution of d-SP in Figure S1c, Supporting Information). After the denaturation, certain amount of lithium salt (LiClO4 in this study) was added into the solution creating a protein-ion complex. Finally, the protein-based ion conductor (PIC) was obtained by properly tuning the evaporation temperature and salt concentration. It is noted that PIC made from pure SP is quite brittle because of the high rigidity of the protein chains, therefore a small amount of high molecular weight poly(ethylene oxide) (PEO) (15 wt%) was added to make the solid conductor more

ACS Paragon Plus Environment

4

Page 5 of 21

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

The Journal of Physical Chemistry Letters

flexible and easier to process. (More details about the fabrication process can be found in the Methods section in Supporting Information).

Figure 1. Fabrication process of protein-based ion conductor. The soy protein derived from soybean is first denatured by acetic acid/water mixture solvent at high temperature to open the chain structures. Then, lithium salt (LiClO4) is added to the denatured soy protein to form protein-ion complex. Afterwards, the solution is evaporated at different temperatures and finally a protein-based ion conductor is obtained. The ionic conductivity of the PIC is highly sensitive to the loading of lithium salt and the evaporation temperature during drying process (protein-ion complex formation temperature). As illustrated in Figure 2a, by simply controlling the protein/lithium salt ratio and the evaporation temperature, the protein undergoes a significant structural reorganization creating specific pathways for ionic conduction. As shown in Figure 2b, for the PIC with 30 wt% loading of lithium salt, simply increasing the complex formation temperature from 25 oC to 40 oC can significantly improve the ionic conductivity by four orders of magnitude from ~ 10-9 to 10-5 S/cm. Meanwhile the PIC shows excellent mechanical flexibility as demonstrated by the inset of Figure 2b. In contrast, an extremely low ionic conductivity of ~ 10-11 S/cm, possibly comes from the salts dissolved in the original soy protein, is observed for the samples without lithium salt. The ionic conductivity is independent of the evaporation temperatures and the film is brittle (see Figure S2a in Supporting Information).

ACS Paragon Plus Environment

5

The Journal of Physical Chemistry Letters

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 6 of 21

Figure 2. Ionic conductivity of protein-based ion conductor (PIC) at room temperature. (a) Schematic of the control of a protein-ion complex by adjusting protein/ion ratio and complex formation temperature. (b) AC ionic conductivity of PIC with 30 wt% of LiClO4 prepared at different complex formation temperatures (Inset shows that the PIC is also mechanically flexible). (c) The effects of complex formation temperature and lithium salt loading on the ionic conductivity. Figure 2c summarizes the effects of temperature and lithium salt loading on the room temperature ionic conductivity (also see Figure S2(b-e) in Supporting Information for more details). As shown, the protein-ion complex formation temperature plays a critical role on the ionic conductivity. Increasing the evaporation temperature from 25 °C to 40 °C significantly improves the ionic conductivity, and this effect becomes stronger with the increasing lithium salt loading up to 30 wt%. This effect on ionic conductivity saturates at a certain evaporation

ACS Paragon Plus Environment

6

Page 7 of 21

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

The Journal of Physical Chemistry Letters

temperature (~ 40 °C) and lithium salt loading (~30 wt%). A further increase of the temperature or salt loading to be above 30 wt% did not further improve or even deteriorate the performance, which is probably due to the saturation of lithium salt in the protein matrix (see the optical images in Figure S3 in Supporting Information). The ionic conductivity of the PIC is adjustable in a wide range from 10-12 to 10-5 S/cm through simply controlling the evaporation temperature and lithium salt loading. Here the PIC sample with 30 wt% loading of lithium salt (around the maximum solubility of LiClO4 in soy protein) and evaporated at 40 °C exhibits the highest ionic conductivity of ~ 10-5 S/cm at room temperature. This value reaches the maximum level that has been realized in PEO-based solid electrolytes, including the ones with pure PEO and the ones modified with various nanoparticles6, 22-23. The samples with 30 wt% of lithium salt and evaporated at 40 °C have been investigated further for electrochemical and mechanical properties. Figure 3a shows the Arrhenius plot of temperature-dependent conductivities () from 25 to 70 °C. The data for sample with 30 wt% of lithium salt and evaporated at 25 °C are also included for comparison. In both cases log  is linearly dependent on 1⁄ indicating that there is no phase transition over the temperature range. The conductivity is exponentially dependent on temperature as:  =  exp− ⁄, in which  is the pre-exponential factor and  is the activation energy. By fitting the data in Figure 3a we can calculate the activation energies for both cases. As shown, the activation energy for PIC evaporated at 40 °C (0.22 eV or 21.2 kJ/mol) is significantly lower than the case with 25 °C (0.36 eV or 34.7 kJ/mol). This value is even close to ceramic superionic conductors, e.g. 0.18-0.22 eV24-25. The Li+ transference number was determined by d.c. polarization method (see Methods section in Supporting Information). The value is 0.94 ± 0.03 (see Nyquist plot of EIS curves and chronoamperometry curve of Li/Conductor/Li cell before and after d.c.

ACS Paragon Plus Environment

7

The Journal of Physical Chemistry Letters

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 8 of 21

polarization in Figure S4 in Supporting Information), which is significantly higher than that of conventional PEO-based solid conductors (usually less than 0.4) as compared in Figure 3b. A similar transference number (0.95-0.98) has recently been obtained in a gelatin protein based solid electrolyte19. The high transference number indicates that the ionic transport in PIC is predominated by Li+ ions. In addition, we also measure the modulus of the sample. As shown, the modulus of the PIC evaporated at both 40 and 25 oC is 1 ± 0.016 GPa. This modulus value of is about two orders of magnitude higher than that of the conventional PEO-based solid electrolytes, which is around 0.02 GPa as shown in Figure 3b. The advantage in mechanical properties for the protein ion conductor as compared with conventional PEO-based electrolyte is contributed by mainly two factors. (1) The difference in flexibility for the polymer chain. It is well-known that PEO is a type of flexible polymer due to the C-O bonds in the backbones. However, soy protein is a type of rigid polymer because of the peptide bonds. (2) The difference in molecular interactions (cohesion). PEO is a simple polymer with simple interactions among different chains, while soy protein processes complicated and strong interactions among different chains. Due to these two reasons, the modulus which is related to the flexibility and cohesion of the chains for the protein ion conductor is much higher than PEO-based systems. If one compares this value with that of ceramic conductors (usually in the range of 10 to 100 GPa26-27), our PIC actually fills a significant gap between conventional polymeric and ceramic solid conductors in terms of mechanical properties.

ACS Paragon Plus Environment

8

Page 9 of 21

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

The Journal of Physical Chemistry Letters

Figure 3. Electrochemical and mechanical properties of the protein-based ion conductors (PICs) prepared with 30 wt% of LiClO4 loading. (a) Arrhenius plot of the ionic conductivity vs. testing temperature for samples evaporated at 25 and 40 oC. (b) Modulus and Li+-transference number for samples evaporated at 40 °C. For comparison, the values of the conventional PEO-based solid conductors are shown as well. To explore the detailed molecular interactions and investigate the ionic transport mechanism, a series of all-atom molecular dynamics (MD) simulations have been performed (see Methods section in Supporting Information). Soy protein is a mixture of proteins containing 2S, 7S, 11S and 15S. Among them, the 7S (or -conglycinin) is one of the main soy protein components21, 28. The crystal structure of a 7S homotrimer formed by three   subunits (PDB ID 1UIK) is shown in Figure S5(a) in Supporting Information. Due to the computational limitation and structural similarity, we only simulate a portion of 7S as indicated by purple color, which is composed of a core barrel domain formed by -sheets and an extended loop domain containing several  helices. The simulation system containing the protein, water, acetic acid and lithium salts, is illustrated in Figure S5(b) in Supporting Information.

ACS Paragon Plus Environment

9

The Journal of Physical Chemistry Letters

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 10 of 21

Simulations were first performed to study the denaturation process of the soy protein under different conditions. Three cases were created according to the experiments: water at 95 oC, acetic acid-water solution at room temperature and acetic acid-water solution at 95 oC, to show the effects from temperature and acetic acids. The ratio of acetic acid and water molecules was adjusted according to the experimental conditions. The time evolutions of the root mean square deviation (RMSD) and the radius of gyration ( ) of the protein during 200 ns simulations for all three cases has been studied (see Figure S6 in Supporting Information). As clearly shown, the soy protein structure is quite stable at a compact state (small RMSD and  values) in water at 95 oC. In acetic-water solution, the formation of hydrogen bonds between acetic acid molecules and soy protein causes the disruption of the protein secondary structures (-helices and -sheets) and the expansion of the protein (increasing RMSD and  values). Increasing the temperature to 95 oC significantly accelerates the disruption process and cause a quick denaturation of the soy protein, resulting in a flexible and mostly random coil structure as illustrated in Figure S6(c) in Supporting Information. Protein-salt interactions play crucial roles to the final ionic transport in PIC. To elucidate the salt-protein interactions and the effects of evaporation temperature, 30 wt% of LiClO! salts were added into the system and the temperature of the system was controlled at 25, 40, 50 and 60 oC, respectively. From our simulations, as shown in Figure 4 (a)-(b) we observed a clear accumulation of ClO!" ions around the protein surface in all cases. Careful inspection of the simulations indicates that it is due to the strong electrostatic interactions between positively charged residues (Lysine and Arginine as indicated in red color) on the side chain and ClO" ! ions. At 25 oC, we observed a rather scattered distribution of the ClO!" ions. Interestingly as the temperature was increased, the promoted protein random motion and strong ion-protein

ACS Paragon Plus Environment

10

Page 11 of 21

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

The Journal of Physical Chemistry Letters

interactions cause a clear clustering of ClO!" ions and protein reorganization as illustrated in Figure 4(b)-(c). The clustering of ClO!" ions causes an increase of the number of ClO" ! interacting with the protein (N$%&(' ) (Figure 4d) and further strengthens the ionic interaction as manifested by residence time distribution in Figure 4e. As shown, the mean residence time for ions in the cluster is one order of magnitude longer than the ions on protein without cluster. Figure 4f shows the averaged N$%&(' (averaged from 100 to 200 ns from Supplementary Figure S7(a)) at different temperatures. The averaged N$%&(' first increases with temperature but then saturates above 50 oC. The trend is qualitatively consistent with the experimental measurements of temperature effect on the ionic conductivity as shown in Figure 4f. Moreover, we also studied the effect of salt concentration by controlling salt weight percentage to 10, 20, 30, 40 and 50% while keeping the temperature at 50 oC. As shown in Figure 4g, the averaged N$%&(' (averaged from 100 to 200 ns from Supplementary Figure 6b) first increase and then saturates above 30 wt%, which is also consistent with the experimental measurement of the salt concentration effect on ionic conductivities. Both the experiments and simulation results indicate that there exist an optimal evaporation temperature and lithium salt loading, over which N$%&(' and the ionic conductivity saturate. Since the attraction of the denatured soy protein to the ClO" ! ions comes from specific side groups (Lysine and Arginine), the maximum number of ClO!" ions attached to the protein must be limited from the number of those groups contained in the protein. For temperature effect, certain amount of heat promotes the flexibility of the protein and facilitates the formation of the cluster of ClO" ! ions. Once the clusters are formed, further increasing the temperature will enhance the kinetic energy of the ClO" ! ions in the cluster and reduce their residence time. The salt loading and evaporation temperature also affect the micro-structures (morphologies) of the samples, as

ACS Paragon Plus Environment

11

The Journal of Physical Chemistry Letters

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 12 of 21

illustrated by SEM images of the pore structures (see Figures S8-10 in Supporting Information) and the distributions of pore size and number density (see Figure S11 in Supporting Information). From these SEM images and statistic data, the sample with a low loading of lithium salt (e.g. 20 wt%) and evaporated at a low temperature (e.g. 25 °C) shows pore structures with a much higher number density (4.07x105 pores/mm2) and much bigger pore size as compared with the samples with moderate loading of lithium salt (e.g. 30 wt%) and evaporation temperature (40 °C). However, excessive amount of lithium salt (e.g. 35 wt%) will crystalize during evaporation process and cause more pore structure as shown in Figure S8 in Supporting Information. The trend in morphology is consistent with the ionic conductivity measurements shown in Figure 2 and Figure S2. Therefore, one must be careful in tuning the amount of salt added and the evaporation temperature for an optimized PIC.

ACS Paragon Plus Environment

12

Page 13 of 21

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

The Journal of Physical Chemistry Letters

Figure 4. Molecular simulations of protein-ion interactions. (a)–(b) Snapshot of the protein-ClO!" complex at 25 °C (a) and 50 °C (b). The ClO!" ions interacting with the protein are shown as gold spheres. The amino acids with positively charged side chains (ARG for Arginine and LYS for Lysine) are shown in red color. (c) Schematic illustration of ClO" ! cluster formation at high temperature. (d) Time evolution of the number of ClO!" interacting with the protein (N$%&(' ) at different temperatures. (e) The distribution of residence time (interaction strength) for ClO" ! ions o interacting with the protein at 25 oC (green color) and ClO" ! ions in the cluster at 50 C (red

color). The formation of the cluster at high temperature causes one order of magnitude increase on the mean residence time. Variation of N$%&(' as a function of different evaporation

ACS Paragon Plus Environment

13

The Journal of Physical Chemistry Letters

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 14 of 21

temperature with 30 wt% loading of LiClO4 (f) and as a function of different lithium salt loading at 50 oC (g). The experimental measurements of the ionic conductivities are shown by green circles for comparison purpose. In protein-lithium salt solution, the ClO!" ions are attracted to the protein chain by the positively charged side groups. At a low loading of lithium salt (e.g. lower than 20 wt%) and low evaporation temperature (lower than 40 °C), as shown in Figure 5a, the ClO!" ions are weakly bonded to the protein surface with a scattered distribution. During the evaporation, it is likely that the strong protein-protein interactions can cause the segregation of proteins and the ClO!" ions, leading to the formation of protein particles and non-uniform ionic distribution as illustrated in Figure 5a. This is consistent with the SEM image of the fracture surface of the sample as shown, in which one can observe numerous grain boundaries and voids that can hinder the ionic transport. While, with high salt loading and at high evaporation temperature, as shown in Figure 5b, the ClO!" ions on protein surface tends to form clusters due to the elevated flexibility of the protein. The formation of the clusters increases the number of the ClO!" ions attached on the protein, and more importantly, significantly strengthen the ion-protein interactions (shown in Figure 4) and stabilize the protein-ion complex. Presumably in this situation the ion-protein interactions are strong enough such that the ClO!" ions are tightly locked inside the protein during the evaporation process, which eventually leads to a much smoother morphology and more uniform ionic distribution within the solid electrolyte as confirmed by the SEM image and schematic in Figure 5b.

ACS Paragon Plus Environment

14

Page 15 of 21

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

The Journal of Physical Chemistry Letters

Figure 5. Decoupled ion transport mechanism. (a) At low salt loading and low evaporation temperature, ClO!" ions are weakly attached to protein with scattered distribution. Removing of solvents during evaporation causes the segregation of protein and anions, leading to morphology with grain boundaries and voids. (b) At high salt loading and high evaporation temperature, ClO!" ions are clustered and strongly locked within the protein and form a highly ion-conductive protein-ions complex. During evaporation, the anions remain in the protein leading to a smooth morphology (SEM image) and uniform anion distribution. (c) Schematic illustration of the decoupled Li+-transportation process. The hopping of Li+ ions is facilitated by the anions which are strongly locked by the protein through positively charged side chains. The scale bars in SEM images are 1 µm. Based on the experimental findings and simulation results, we propose a decoupled ionic transport mechanism in PIC. As illustrated by Figure 5c, the ClO!" ions are immobile and strongly locked within the protein by specific side groups with positive charges. The migration

ACS Paragon Plus Environment

15

The Journal of Physical Chemistry Letters

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 16 of 21

of Li) ions in such system is realized through the hopping among coordination sites (most likely formed by the protein backbone oxygen atoms) and the “locked” anions, a mechanism similar to the conduction in a ceramic conductor29-31 but completely different from the traditional PEObased solid conductors. The hopping of Li) ions is facilitated by the ClO!" ions that connect those coordination sites, therefore the ionic transport is decoupled with the protein chain motion. Based on this picture, the ionic conductivity can be improved by increasing the number and/or the spatial arrangement of ClO!" ions without sacrificing the mechanical properties as demonstrated in our experiments. In addition, the clustering of the anions locally modifies the energy landscape and effectively lowers the activation energy for Li) transport. Finally, since the ClO!" ions are immobilized by the protein, the ion transport in PIC is dominated by the Li) ions leading to a high transference number as shown by the experiments. In summary, this work presents a successful fabrication and systematic investigation of a high performance solid state protein-based ion conductor. The conductor, made from soy proteins through simply doping with proper amount of lithium salt and evaporating at appropriate temperature, has both excellent electrochemical properties (ionic conductivity of ~10"+ S/cm, transference number of 0.94 and activation energy of 0.22 eV) and mechanical properties (high modulus of 1 GPa, while still flexible and easily processable). Molecular simulation results indicate that the ion transport in the protein ion conductor is facilitated by the anions “locked” by specific functional groups in the protein structure. A decoupled transport mechanism, similar to the conduction in ceramic conductors, is proposed to explain all findings in experiments. We expect the decoupled transport mechanism to be transferable to other combinations of proteins and salts. Although the ion conductivity for our conductor is still relatively low compared with liquid electrolytes for practical battery applications, significant improvements are expected with

ACS Paragon Plus Environment

16

Page 17 of 21

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

The Journal of Physical Chemistry Letters

strategies of adding nanoparticles and more precise controlling of protein structures. Moreover, the proteins are compatible with other biological systems, therefore the conductor may have great potential in biomedical applications in addition to energy storage devices. Finally, considering the complexity of the protein structures and compositions, we anticipate the insights from this work will provide guidelines for design, optimization and fabrication of new proteinbased ion conductors.

ASSOCIATED CONTENT Supporting Information Details of the methods (experimental and simulation), morphology characterization, electrochemical/mechanical property measurements and molecular simulations; additional figures and tables mentioned in the text. AUTHOR INFORMATION Corresponding Authors [email protected] [email protected] [email protected] Notes The authors declare no competing financial interests. ACKNOWLEDGMENT

ACS Paragon Plus Environment

17

The Journal of Physical Chemistry Letters

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 18 of 21

The authors gratefully acknowledge the financial support from USDA NIFA 2015-6702122911, NSF CBET 1250107 and partially from NSF CMMI 1463616. The authors also gratefully acknowledge the support on characterizations from the Franceschi Microscopy & Imaging Center, and the Composite Materials and Engineering Center at Washington State University. Computational resources were provided in part by the Extreme Science and Engineering Discovery Environment (XSEDE) under grant No. MSS150014. REFERENCES (1)

Tarascon, J. M.; Armand, M. Issues and Challenges Facing Rechargeable Lithium

Batteries. Nature 2001, 414, 359-367. (2)

Knauth, P. Inorganic Solid Li Ion Conductors: An Overview. Solid State Ion. 2009, 180,

911-916. (3)

Li, J.; Ma, C.; Chi, M.; Liang, C.; Dudney, N. J. Solid Electrolyte: The Key for High-

Voltage Lithium Batteries. Adv. Energy Mater. 2015, 5, 1401408. (4)

Karmakar, A.; Ghosh, A. Dielectric Permittivity and Electric Modulus of Polyethylene

Oxide (Peo)-Liclo4 Composite Electrolytes. Curr. Appl. Phys. 2012, 12, 539-543. (5)

Hallinan, D. T., Jr.; Balsara, N. P. Polymer Electrolytes. Annu. Rev. Mater. Res. 2013, 43,

503-525. (6)

Xue, Z.; He, D.; Xie, X. Poly(Ethylene Oxide)-Based Electrolytes for Lithium-Ion

Batteries. J. Mater. Chem. A 2015, 3, 19218-19253. (7)

Porcarelli, L.; Gerbaldi, C.; Bella, F.; Nair, J. R. Super Soft All-Ethylene Oxide Polymer

Electrolyte for Safe All-Solid Lithium Batteries. Sci. Rep. 2016, 6, 19892.

ACS Paragon Plus Environment

18

Page 19 of 21

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

The Journal of Physical Chemistry Letters

(8)

Kelly, I.; Owen, J. R.; Steele, B. C. H. Mixed Polyether Lithium-Ion Conductors. J.

Electroanal. Chem. 1984, 168, 467-478. (9)

Frech, R.; Chintapalli, S. Effect of Propylene Carbonate as a Plasticizer in High

Molecular Weight Peo-Licf3so3 Electrolytes. Solid State Ion. 1996, 85, 61-66. (10) Fan, L.-Z.; Maier, J. Composite Effects in Poly(Ethylene Oxide)-Succinonitrile Based All-Solid Electrolytes. Electrochem. Commun. 2006, 8, 1753-1756. (11) Wang, H.; Im, D.; Lee, D. J.; Matsui, M.; Takeda, Y.; Yamamoto, O.; Imanishi, N. A Composite Polymer Electrolyte Protect Layer between Lithium and Water Stable Ceramics for Aqueous Lithium-Air Batteries. J. Electrochem. Soc. 2013, 160, A728-A733. (12) Tsuchida, E.; Ohno, H.; Tsunemi, K.; Kobayashi, N. Lithium Ionic-Conduction in Poly(Methacrylic Acid) Poly(Ethylene Oxide) Complex Containing Lithium Perchlorate. Solid State Ion. 1983, 11, 227-233. (13) Jacob, M. M. E.; Prabaharan, S. R. S.; Radhakrishna, S. Effect of Peo Addition on the Electrolytic and Thermal Properties of Pvdf-Liclo4 Polymer Electrolytes. Solid State Ion. 1997, 104, 267-276. (14) Rocco, A. M.; da Fonseca, C. P.; Pereira, R. P. A Polymeric Solid Electrolyte Based on a Binary Blend of Poly(Ethylene Oxide), Poly(Methyl Vinyl Ether-Maleic Acid) and Liclo4. Polymer 2002, 43, 3601-3609. (15) Stephan, A. M. Review on Gel Polymer Electrolytes for Lithium Batteries. Eur. Polym. J. 2006, 42, 21-42.

ACS Paragon Plus Environment

19

The Journal of Physical Chemistry Letters

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 20 of 21

(16) Goodenough, J. B.; Kim, Y. Challenges for Rechargeable Li Batteries. Chem. Mater. 2010, 22, 587-603. (17) Bouchet, R., et al. Single-Ion Bab Triblock Copolymers as Highly Efficient Electrolytes for Lithium-Metal Batteries. Nat. Mater. 2013, 12, 452-457. (18) Vieira, D. F.; Avellaneda, C. O.; Pawlicka, A. Conductivity Study of a Gelatin-Based Polymer Electrolyte. Electrochim. Acta 2007, 53, 1404-1408. (19) Basu, T.; Goswami, M. M.; Middya, T. R.; Tarafdar, S. Morphology and IonConductivity of Gelatin-Licio4 Films: Fractional Diffusion Analysis. J. Phys. Chem. B 2012, 116, 11362-11369. (20) Ghadami, A.; Qazvini, N. T.; Nikfarjam, N. Ionic Conductivity in Gelatin-Based Hybrid Solid Electrolytes: The Non-Trivial Role of Nanoclay. J. Mater. Sci. Technol. 2014, 30, 10961102. (21) Nishinari, K.; Fang, Y.; Guo, S.; Phillips, G. O. Soy Proteins: A Review on Composition, Aggregation and Emulsification. Food Hydrocolloid. 2014, 39, 301-318. (22) Christie, A. M.; Lilley, S. J.; Staunton, E.; Andreev, Y. G.; Bruce, P. G. Increasing the Conductivity of Crystalline Polymer Electrolytes. Nature 2005, 433, 50-53. (23) Zhang, C.; Gamble, S.; Ainsworth, D.; Slawin, A. M. Z.; Andreev, Y. G.; Bruce, P. G. Alkali Metal Crystalline Polymer Electrolytes. Nat. Mater. 2009, 8, 580-584. (24) Yamane, H.; Shibata, M.; Shimane, Y.; Junke, T.; Seino, Y.; Adams, S.; Minami, K.; Hayashi, A.; Tatsumisago, M. Crystal Structure of a Superionic Conductor, Li7p3s11. Solid State Ion. 2007, 178, 1163-1167.

ACS Paragon Plus Environment

20

Page 21 of 21

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

The Journal of Physical Chemistry Letters

(25) Seino, Y.; Ota, T.; Takada, K.; Hayashi, A.; Tatsumisago, M. A Sulphide Lithium Super Ion Conductor Is Superior to Liquid Ion Conductors for Use in Rechargeable Batteries. Energy Environ. Sci. 2014, 7, 627-631. (26) Sakuda, A.; Hayashi, A.; Tatsumisago, M. Sulfide Solid Electrolyte with Favorable Mechanical Property for All-Solid-State Lithium Battery. Sci. Rep. 2013, 3. (27) Sakuda, A.; Hayashi, A.; Takigawa, Y.; Higashi, K.; Tatsumisago, M. Evaluation of Elastic Modulus of Li2s-P2s5 Glassy Solid Electrolyte by Ultrasonic Sound Velocity Measurement and Compression Test. J. Ceram. Soc. Jpn. 2013, 121, 946-949. (28) Saio, K.; Kamiya, M.; Watanabe, T. Food Processing Characteristics of Soy Bean-D 11s and 7s Proteins I Effect of Difference of Protein Components among Soy Bean-D Varieties on Formation of Tofu-Gel. Agric. Biol. Chem. 1969, 33, 1301-1308. (29) Kamaya, N., et al. A Lithium Superionic Conductor. Nat. Mater. 2011, 10, 682-686. (30) Wang, Y.; Richards, W. D.; Ong, S. P.; Miara, L. J.; Kim, J. C.; Mo, Y.; Ceder, G. Design Principles for Solid-State Lithium Superionic Conductors. Nat. Mater. 2015, 14, 10261032. (31) Bachman, J. C., et al. Inorganic Solid-State Electrolytes for Lithium Batteries: Mechanisms and Properties Governing Ion Conduction. Chem. Rev. 2016, 116, 140-162.

ACS Paragon Plus Environment

21