Surface Layer Evolution on Graphite During Electrochemical Sodium

Mar 24, 2017 - The research leading to these results has also received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) u...
0 downloads 8 Views 3MB Size
Subscriber access provided by University of Newcastle, Australia

Article

Surface layer evolution on graphite during electrochemical sodium-tetraglyme co-intercalation Julia Maibach, Fabian Jeschull, Daniel Brandell, Kristina Edstrom, and Mario Valvo ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b16536 • Publication Date (Web): 24 Mar 2017 Downloaded from http://pubs.acs.org on March 26, 2017

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.

ACS Applied Materials & Interfaces 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 37

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

Surface Layer Evolution on Graphite During Electrochemical Sodium-tetraglyme Co-intercalation Julia Maibach*, Fabian Jeschull, Daniel Brandell, Kristina Edström, and Mario Valvo Department of Chemistry – Ångström Laboratory, Uppsala University, Box 538, 75121 Uppsala, Sweden KEYWORDS: Na-ion batteries, Photoelectron Spectroscopy, Graphite, Solid Electrolyte Interphase, Ether-based electrolytes, TEG-DME, Polyacrylic Acid, NaFSI

ABSTRACT

One obstacle in sodium ion batteries is the lack of suitable anode materials. As recently shown, the most common anode material of the state of the art lithium ion batteries, graphite, can be used for sodium ion storage as well, if ether-based electrolyte solvents are used. These solvents cointercalate with the sodium ions leading to the highly reversible formation of ternary graphite intercalation compounds (t-GIC). In order for the solvent co-intercalation to work efficiently, it is expected that only a very thin surface layer forms during electrochemical cycling. In this article, we therefore present the first dedicated study of the surface layer evolution on t-GICs using soft X-ray photoelectron spectroscopy. This technique with its inherent high surface sensitivity and low probing depth is an ideal tool to study the underlying interfacial reactions during the sodiation and de-sodiation of graphite. In this report, we apply this approach to graphite 1 ACS Paragon Plus Environment

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

composite

electrodes

cycled

in

Na

half

cells

Page 2 of 37

with

a

1

M

sodium

bis(fluorosulfonyl)imide/tetraethylene glycol dimethyl ether (NaFSI/TEG-DME) electrolyte. We have found a surface layer on the cycled electrodes, mainly composed of salt decomposition products and hydrocarbons, in line with irreversible capacity losses observed in the electrochemical cycling. While this surface layer does not seem to block co-intercalation completely, it seems to affect its efficiency resulting in a low Coulombic efficiency of the studied battery system.

1. Introduction Sodium ion batteries (SIBs) are currently receiving a renewed attention as a cost-effective complement to Li-ion batteries (LIBs) for reversible electrochemical energy storage, because sodium is much more abundant than lithium 1-2 and also because its feedstock and distribution are not restricted by geographical factors 3. These alternative battery systems can be envisioned for large-scale (e.g. stationary) electrical energy storage applications 4-5, where the specifications for energy densities and specific gravimetric/volumetric capacities are not as stringent as those required in mobile applications (e.g. portable electronics, hybrid electric, and electric vehicles). Moreover, the overall reduction in costs in energy storage could grant a wider access to electricity produced via renewable sources and promote their penetration into the energy market. In fact, their spread will largely depend on the ultimate price associated to the entire chain of conversion, distribution and storage of the electrical energy. In this scenario SIBs are clearly attractive, although their technological development is far from that reached by LIBs 3. The sensible differences in both size and mass between Li+ and Na+ ions 3, 6 make the latter less convenient for usage in reversible electrochemical intercalation processes, due to a series of 2 ACS Paragon Plus Environment

Page 3 of 37

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

possible structural and kinetic limitations. A crucial example of the impact of the size of Na+ on the associated electrochemical behavior is represented by graphite

7-8

: commercial LIBs rely

heavily on graphite since its early application 9 as suitable negative electrode to attain stable and reversible Li+ storage via topotactic reactions. Lithiation of graphite leads to the formation of LixC6 (0 10 µm), fine KS6-type graphite (Timcal – particle size < 7 µm), carbon black (CB – Super P, Timcal) and polyacrylic acid (PAA – Aldrich, Mw = 450,000, 35 wt.% aqueous solution) were mixed together to prepare a slurry for the 5 ACS Paragon Plus Environment

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 6 of 37

electrode coatings. The weight ratio of these materials in the slurry was 85 wt.% for coarse graphite, 3 wt.% for KS6 graphite, 2 wt.% for CB and 10 wt.% for the polymer binder. The latter was prepared from an original 35 wt.% PAA precursor which was deprotonated by means of 4 M NaOH in water to yield a final 15 wt.% PAA-Na binder solution. PAA-Na was chosen as a binder for this study because it facilitates PES analysis of the surface layers since its molecular structure contains only C and O and thus minimizes the number of overlapping signals with typical salt components such as F, S, and N. Additionally, PAA-Na shows good adhesion properties and can prevent graphite exfoliation as shown for lithium intercalation

42-43

. The binder and the other

components were dispersed in water and their mixture was ball-milled for 2 h. The resulting slurry was casted on a copper foil (Goodfellow, average thickness ≈22 µm) by a dedicated coating equipment (KR – K Control Coater) and dried at ambient conditions to slowly evaporate the water in the obtained layer. Circular coated electrodes having a diameter of 20 mm were cut by a precision perforator (Hohsen) and subsequently dried in a tubular vacuum oven (Büchi) at 80 °C for 12 hours prior to cell assembly. The thickness of the composite coatings was measured by a digital caliper (Mitutoyo), yielding a typical value of approximately 50-60 µm. A dedicated electrolyte was prepared in a Ar-filled glove-box (M-Braun) by first drying a sodium-bis-(fluorosulphonyl)imide (i.e. NaFSI) salt at 110 °C in vacuum for 12 hours in a tubular oven and then dissolving it in tetraethylene glycol dimethyl ether (TEG-DME - Sigma Aldrich, purity >99%, used as received) to give a 1 M solution. The electrolyte was used without any further treatment or additives.

2.2 Cell assembly and electrochemical measurements

6 ACS Paragon Plus Environment

Page 7 of 37

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

Polymer-laminated pouch cells were mounted in an Ar-filled glove box (M-Braun) having oxygen and moisture levels below 1 ppm. The battery pouches contained graphite as working electrode, sodium metal as simultaneous reference- and counter-electrode and 1 M NaFSI in TEG-DME solution as electrolyte. A porous polyethylene separator (Solupor, Lyndall Performance Materials) was soaked with a few droplets of electrolyte solution and used in the cells, which had nickel tabs as electrical contacts. Galvanostatic measurements were performed using a VMP2 (Bio-Logic) equipment by applying a constant current to obtain an initial C-rate of approximately C/10 (i.e. the use of full capacity in 10 h) for the first discharge of each cell. The applied current was calibrated on the basis of the mass of the active material coated on each electrode. Typical values for the applied currents ranged approximately from 110 to 130 µA, given the typical graphite loading of 2.6 - 2.7 mg cm-2. The lowest cut-off voltage set during discharge (sodiation of graphite) was 0.05 V vs. Na+/Na, whereas the highest potential reached upon charge (desodiation) was limited to 2.00 V vs. Na+/Na. In the following discussion, these points are also referred to as ‘end of discharge’ (EOD) and ‘end of charge’ (EOC) and all the potentials are measured and reported vs. Na+/Na, unless otherwise stated. Different states of charge (SOC) within the first two complete discharge/charge cycles were investigated by SoXPES, as indicated in Figure 1. An additional sample at open circuit voltage (OCV) was also considered and this corresponded to an assembled battery with a graphite electrode left in an uncycled state. A sample overview indicating the SOC and respective sample nomenclature is given in Table 1.

7 ACS Paragon Plus Environment

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 8 of 37

Table 1. Overview of the electrode samples left in different states of charge during their two initial galvanostatic cycles. Sample

Potential

Discharge/charge

series

(V vs. Na+/Na)

(electrode state)

Pristine

-

-

OCV*

2.62

uncycled

I

0.90

1st discharge

II

0.70

1st discharge

III

0.50

1st discharge

IV

0.05

end of 1st discharge

V

2.00

end of 1st charge

VI

0.90

2nd discharge

VII

0.60

2nd discharge

VIII

0.05

end of 2nd discharge

IX

2.00

end of 2nd charge

*Open circuit voltage

2.3 Characterization via soft X-ray photoelectron spectroscopy All batteries were transported to the synchrotron facility in their assembled state, protecting the metal contacts to avoid unintentional short circuiting. Prior to the measurements, the batteries were dissembled in an Ar-filled glove box to remove the respective graphite electrodes and prepare them for the SoXPES analyses. These electrodes were then rinsed three consecutive times with dimethyl carbonate (DMC) to remove any excess of electrolyte. The samples were mounted on dedicated sample holders using a conductive Cu tape and subsequently transferred to the spectrometer without any contact to ambient air, as described previously

44

. SoXPES 8

ACS Paragon Plus Environment

Page 9 of 37

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

characterization of the graphite electrodes was performed at the soft X-ray beamline I411 at the MAX IV laboratory, Lund, Sweden. Each elemental line was recorded tuning the incoming photon energy so that the emitted photoelectron had a kinetic energy of 140 eV. Additionally, all elemental lines were recorded using 835 eV. A Scienta R4000 WAL hemispherical analyzer was operated at a pass energy of 200 eV in constant analyzer energy mode. The base pressure of the analysis chamber was in the range of 10-8 to 10-7 mbar. All spectra were calibrated in binding energy with respect to the hydrocarbon peak at 285 eV (recorded at a photon energy of 835 eV).

3. Results and discussion 3.1 Electrochemical analysis The electrochemical behavior of graphite:Na half cells in a sodium salt:TEG-DME electrolyte was studied in this section. Reversible cell cycling has been demonstrated repeatedly in several studies

28-29, 31

. Cell cycling was performed confirm that the co-intercalation process takes place

in the chosen system but the main objective was sample preparation for the following SoXPES surface analysis. The characteristic voltage profiles of galvanostatic discharge/charge of a composite graphite electrode cycled in a Na half cell having 1 M NaFSI in TEG-DME electrolyte are presented in Figure 1.

9 ACS Paragon Plus Environment

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 10 of 37

Figure 1. (a) Individual voltage profiles for the first two cycles of galvanostatic discharge/charge of a composite PAA-Na/graphite electrode tested in a Na half cell with 1 M NaFSI in TEG-DME between 2.00 and 0.05 V vs. Na+/Na. Note the similar gravimetric capacities obtained at the end of the respective cut-off voltage points for each discharge and charge. (b) Chronopotentiogram obtained for the same Na half cell schematically showing the various voltage stages sampled in different cells for surface analysis via soft X-ray photoelectron spectroscopy (SoXPES). It can be noticed that in the first discharge the potential dropped sharply from the initial open circuit voltage (OCV) value (i.e. ≈2.60 V vs. Na+/Na) to about 0.80 V to give a first neat plateau, while displaying a very weak inflection around 1.00 V. The flat feature at 0.80 V was followed

10 ACS Paragon Plus Environment

Page 11 of 37

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

by a slight inflection of the curve leading to a second sloping plateau located around 0.60 V and extending to about 0.50 V. Below this potential, the final part of the discharge curve became steeper and evolved in a linear segment down to the cut-off point at 0.05 V. The first discharge (i.e. sodiation) capacity amounted to 163 mAhg-1 for the electrode, which underwent a reduction process in ≈10.5 hours, i.e. with a cycling rate close to C/10. The first charging curve started with an evident step of the voltage to 0.50 V, indicating that a non-negligible overpotential existed to initiate the de-intercalation process for both Na+ and the solvent molecules. Another similar step-like feature was observed for the potential upon charging around 0.85 V. The latter was followed by a more sloping plateau beginning around 1.20 V. The final part of the charge curve displayed a steeper slope at voltages higher than 1.50 V until the upper cut-off point was reached at 2.00 V. Overall, the voltage profiles of these discharge/charge curves approximately matched those of similar previous studies

30-32

, thus confirming that it is

indeed possible to insert Na+ in graphite in this specific battery configuration. Additional confirmation of the Na-glyme co-intercalation process for this specific electrode/electrolyte combination can be obtained from the cyclic voltammetry curves (Figure S1) which match the general structure observed in previous reports28-29. It is worth noticing that the shape of the subsequent discharge and charge profiles in the second cycle closely resembled the curves obtained in the first cycle. In fact, only a more slant feature was observed for the first plateau during the second discharge, beginning at a higher voltage of 1.15 V. The subsequent sloping plateau in the second discharge appeared to be similar to that of the first cycle, apart from the fact that its average value was constantly shifted towards a slightly higher potential, thus giving a displacement of about 100 mV. The origin of such behavior is not entirely understood; it could be due to a deformation of the host lattice structure and/or other

11 ACS Paragon Plus Environment

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 12 of 37

phenomena caused by the solvent co-intercalation. In this respect, it has been shown clearly by HR-TEM that electrochemical Na storage in natural graphite causes the formation of disordered planes having expanded lattice distances

29

. This phenomenon could influence the energy

required for subsequent Na+ and solvent co-intercalation, since the pristine graphite lattice is affected by such structural deformation during the first cycle. It is particularly interesting to observe that such a potential shift occurred only upon discharge, i.e. sodiation, whereas charging (de-sodiation) yielded almost congruent voltage profiles displaying the same staircase-like features. Both in the first and second cycle, a significant irreversible capacity of approximately 60 mAh g-1 was observed indicating SEI formation at low potentials during the discharge process. Kim et al. reported approximately 10 mAh g-1 irreversible capacity for a 1 M NaPF6/TEG-DME electrolyte

30

and Zhu et al. reported a similar value for 1 M NaOTf in TEG-DME

31

. This

suggests that the choice of the electrolyte salt might have a significant influence on the capacity losses in the first cycles as the solvent used in this study is the same as in the cited reports. The relatively high irreversible capacity in this study leads to rather low Coulombic efficiencies for both reported cycles: 61.4% in the first cycle which increase to 63.9% in the second, indicating poor reversibility and cell failure after only a few cycles (see Figure S2). The deviations in irreversible capacity as well as in Coulombic efficiency as compared to literature values suggest that the efficiency of the electrochemical formation of ternary graphite intercalation products is highly dependent on the overall chemical system of the battery. The system investigated in this study was adjusted with respect to efficient surface analysis. The used graphite particles were approx. 1-10 µm (for KS6) and >10µm (for the material provided by Leclanché) in size, while the natural graphite particles used in Ref.

29

and Ref.

28

had particle sizes in the order of 40-140

µm. The lower particle size leads to a higher fraction of surface area and thus increased surface 12 ACS Paragon Plus Environment

Page 13 of 37

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

layer formation. The low Coulombic efficiency could also be due to the choice of binder. In LIBs, PAA has proven beneficial, as it provides good particle coverage and thus supports efficient passivation towards electrolyte degradation

43, 45

, which should also be avoided in case

of the co-intercalation process. On the other hand, in a recent work by our group using a graphite:PAA-Na electrode in an ether-based electrolyte (DME:DOL, v/v = 1:1) for Li-S battery applications we showed that sufficient passivation is achieved only in presence the SEI forming additive LiNO342. We therefore suspect, that PAA-Na itself is sufficiently permeable to ethers (but not to carbonates) and thus is not restricting the co-intercalation process in our system. In general, incomplete de-sodiation cannot be ruled out as a possible cause for the poor reversibility observed in this system. The choice of binder in this study was motivated by the prevention of overlapping PES signals with the electrolyte salt. We tentatively rule out the choice of solvent as the main cause for the low Coulombic efficiency, since for example Zhu et al. reported around 6000 cycles with Coulombic efficiencies close to 100% for Na-TEG-DME cointercalation in graphite

31

. The electrolyte salt NaFSI has so far not been used in literature for

Na-glyme co-intercalation but can be obtained at higher purity than NaPF6, which is advantageous for the surface sensitive PES analysis. Its influence will be subject of further discussion in the following section.

3.2 SoXPES analyses and results The main purpose of the present investigation is to probe and clarify the surface processes taking place at this particular graphite/TEG-DME electrolyte interface. It is of utmost importance to evaluate what the major obstacles are that might block the way towards an efficient operation of such graphite electrodes in upcoming SIBs, especially considering the key role that this

13 ACS Paragon Plus Environment

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 14 of 37

material has played in the development of LIBs. For this reason, different representative points have been sampled at various states of charge in a series of Na half cells and carefully analysed by an advanced SoXPES technique, as schematically indicated in Figure 1b (see also Table 1). In Figure 2, the O1s and C1s spectra recorded at 835 eV excitation energy, as well as 690 eV for O1s and 430 eV for C1s are shown. Generally, the O1s spectra show two distinct features at approximately 533.2 eV and 531.5 eV. The peaks are very broad due to many possible oxygen contributions, which makes a detailed evaluation difficult. From reference measurements (see Figure S3) on the binder PAA-Na, the lower binding energy (BE) component at 531.5 eV is attributed to the carboxyl (COO) group, since the sodiation of PAA leads to an additional negative charge which is delocalized on the same group. Therefore, essentially only one oxygen species with a low binding energy is observed for the binder. The higher BE contribution is largely attributed to the electrolyte components NaFSI and TEG-DME. The curve fit shown in Figure 2 is therefore most likely an oversimplification, but nevertheless allows us to follow important trends also observed in the C1s line. In the C1s spectra, four components can be distinguished: i) a graphite peak at 284.2 eV, ii) a hydrocarbon peak (CH) at 285.0 eV, iii) a peak at 286.8 eV attributed to C-O-C environments as present in TEG-DME, and iv) a peak at 288.2 eV from the COO group from PAA. For the OCV sample, the oxygen spectrum is already dominated by the TEG-DME component, although the electrolyte species seem to be mainly located at the surface of the electrode. This is seen from the intensity of the high binding energy component which increases relatively to the bulk-associated PAA contribution when going from a higher probing depth at 835 eV to the highest surface sensitivity at 680 eV. As far as carbon is concerned, the probing depths for the two excitation energies vary more strongly, so that more pronounced spectral changes are noticed when going from 835 eV to 430 eV. While the C1s

14 ACS Paragon Plus Environment

Page 15 of 37

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

spectrum is dominated by the bulk-graphite emission in the former case, the surface hydrocarbon, along with the electrolyte components, are more pronounced in the latter one. When going from OCV to 0.9 V in the first discharge (sodiation), an increase in the intensity of surface- and electrolyte-attributed contributions relative to the bulk-assigned components, such as PAA and graphite, can be observed in the O1s and C1s spectra for both probing depths. The spectra recorded after cycling the electrode to 0.7 V (i.e. after the first plateau depicted in Figure 1) show a clear increase in the TEG-DME components. Especially noteworthy is that the component associated with TEG-DME in the C1s spectrum for sample II is more intense than the hydrocarbon contribution at 835 eV, but this does not occur at 430 eV. This can be interpreted as a direct observation of the TEG-DME co-intercalation. At the same time, a surface film is formed which can be seen from that the bulk components of graphite and PAA are not visible in the C1s spectrum at 430 eV and appear strongly reduced in the 835 eV spectra. The trend of increased intensity for the emissions associated with TEG-DME relative to other surface components continues as the electrodes are cycled to even lower potentials (0.5 V, sample III). Since the graphite and PAA components are no longer visible even for the more bulk-sensitive spectrum (i.e., for 835 eV excitation energy), the SEI layer thickness must have increased beyond the probing depth. At the end of the first discharge (sample IV), a clear graphite signal accompanied by a PAA peak both in the C1s and O1s can be observed. At the same time, increased salt decomposition starts (see Figure 3). The presence of the graphite and PAA signal indicate either that the surface layer gets thinner or that the surface layer breaks up due to the large increase in graphite interlayer distance (i.e. ≈255 % for tetraglyme

32

) during the formation of ternary

graphite intercalation products, thus exposing fresh, uncovered surface or new access pathways to the bulk.

15 ACS Paragon Plus Environment

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 16 of 37

At the end of the first charge (desodiation, sample V) the TEG-DME signal intensity in the bulk O1s spectra (recorded with 835 eV) is again lower compared to that of the PAA binder. This can be interpreted as removal of solvent from the graphite structure. A similar trend can be observed in the C1s spectra, where the PAA signal is again visible and the relative intensity of the TEG-DME is reduced compared to the one ascribed to hydrocarbon. Interestingly, no graphite peak can be observed, which indicates that a surface layer with a thickness exceeding the probing depth had been formed on the cycled electrodes. In the beginning of the second sodiation (sample VI), a graphite component is observed in the C1s spectrum recorded with a 835 eV excitation energy. This can be interpreted as a reduced surface layer thickness or breaking of the previously formed SEI during co-intercalation of the solvent and sodium ions. During the second sodiation, similar trends as during the first sodiation can be observed, i.e. the relative increase of the TEG-DME signal compared to the one of PAA. However, also the hydrocarbon signal increases in intensity, which could indicate continuous surface layer formation. The end of both the second discharge and charge (samples VIII and IX) follows in principle the same trends as the first cycle, but the overall contribution of the surface and SEI components is increased. This is in line with the cycling performance shown in Figure 1a, where a significant irreversible capacity loss could be observed also in the second cycle.

16 ACS Paragon Plus Environment

Page 17 of 37

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

Figure 2. O1s and C1s spectra of the various electrode samples recorded for two different photon energies yielding different probing depths. The spectra are calibrated in binding energy with respect to the hydrocarbon peak at 285 eV.

The changes in the graphite peak intensity in the C1s spectra can be used to estimate the SEI thickness as it is a peak characteristic for the bulk electrode and its intensity will be reduced by a growing surface layer. As the graphite peak can still be observed at many stages during the first two cycles using 835 eV excitation energy but not in the surface sensitive measurements, the SEI thickness is larger than the probing depth for 430 eV excitation energy, which is estimated to 17 ACS Paragon Plus Environment

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 18 of 37

around 3 nm, but smaller than 7 nm, which corresponds to the approximate probing depth at 835 eV excitation energy. These estimations of the probing depth (equal to three times the inelastic mean free path of the electrons at a given kinetic energy, IMFP) and thus the SEI thickness are based on calculations of the IMFP using the polyethylene model of Ashley et al.46 in the NIST IMFP database 47. From the overall trends in the C1s spectra, it can be concluded that the SEI contains hydrocarbon species as their intensity increases steadily during cycling. These could originate either from solvent or binder decomposition. In order to evaluate the SEI composition further, the salt-related spectra for F1s, N1s, S2p, and Na2s (recorded at 835 eV photon energy) are shown in Figure 3. In the F1s spectrum of the OCV sample, a clear peak corresponding to the salt anion FSI at 688 eV can be observed. Additionally, a minor peak is seen at around 684.5 eV most likely due to NaF. In the N1s and S2p spectra, the main peaks correspond to the expected FSIcomponents in binding energy (marekd by the dashed line in Figure 3), although minor additional peaks (or peak assymetries) on the low binding energy side of the FSI-peaks are observed. The latter together with the formation of NaF likely indicate salt instability even at the OCV stage or during X-ray exposure as previously observed for the Li analogue

48

and very recently also for

NaFSI in carbonate electrolytes 49. During the SEI formation and at the beginning of the sodiation (samples I-III), the F1s, N1s and S2p emissions do not show significant differences in binding energy or relative intensity variations between the different components. At the end of the first discharge (sample IV), the NaF signal as well as the low binding energy components in the N1s and S2p emissions rise in relative intensity (shaded purple in Figure 3), clearly indicating salt decomposition. During the first charge and the early stages of the second discharge, the F1s, N1s and S2p spectra again show very little changes. However, at the end of the second discharge, further deposition of NaF and other salt degradation products at the sample surface can be 18 ACS Paragon Plus Environment

Page 19 of 37

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

observed. For example, the intensity of the NaF component in Figure 3 has risen relatively to the FSI peak and even appears higher. This intensity ratio is reversed again for the sample at the end of the second charge, however, the NaF spectral contribution is still significantly higher than that recorded after the end of the first charge, thus indicating a clear accumulation of NaF during cycling. The Na2s emissions are included in Figure 3 for completeness. However, a detailed evaluation is not attempted since both the binder and the electrolyte salt contribute with a Na2s signal and the binding energies for these ionized Na species overlapp with intercalated sodium. Additionally, due to an incidental beam loss at the synchrotron facility, the Na2s signal for the sample IX (second desodiation) had to be remeasured after an injection and thus cannot be evaluated reliably. Any evaluation of the Na-to-TEG-DME ratio in the presented experiment is therefore not attempted.

19 ACS Paragon Plus Environment

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 20 of 37

Figure 3. F1s, N1s, S2p, and Na2s spectra of the various electrode samples recorded at 835 eV photon energy. The spectra are calibrated in binding energy with respect to the hydrocarbon peak at 285 eV.

In Figure 4, the relative intensity ratios for each element are shown. This representation gives an indication for the sodium intercallation as the relative Na2s intensity increases from roughly 23% relative intensity for samples at potentials above the sodiation thershold to about 5-6 % for sample IV (end of 1st sodiation) and sample VII (end of 2nd sodiation). For the OCV sample, the relative intensity of all the salt components combined together gives a contribution of only about

20 ACS Paragon Plus Environment

Page 21 of 37

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

15%. During the first charge, this increases to about 25%, whereas at the end of the first charge, the salt contibutions rise to ≈35 % and afterwards remain at a high level of about 30 %.

Figure 4. Histograms of the relative intensities (i.e. total peak areas) associated to different elements detected in the SoXPES measurements of the various electrode samples. The peak areas were obtained from the spectra recorded using 835 eV photon energy.

Summarizing the findings on the SEI composition, the salt decomposition seems to occur at potentials below 0.5 V vs. Na+/Na and its by-products seem to accumulate during cycling, contributing to the SEI build-up in line with a recent publication where NaFSI was used in a carbonate-based electrolyte together with organic electrode materials 49. Generally, the observed salt decomposition is somewhat surprising as NaFSI is often considered a very stable salt and can be obtained with higher purity than for example NaPF6

50-51

. One possible explanation for the

increased salt decomposition could be the absence of a protective SEI formed from solvent 21 ACS Paragon Plus Environment

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 22 of 37

decomposition products in our study. As the traditionally used carbonate solvents usually decompose at higher potentials (e.g. around 1 V vs. Li+/Li in case of ethylene carbonate), they form a mostly organic passivation layer on the active material thus preventing - or at least mimizing - salt decomposition, which would otherwise occur at lower potentials. Thus, our findings indicate that salt stability might need to be re-evaluated for t-GIC and novel anode materials (such as those presented in Ref. 49), where very limited and/or different SEI formation and chemistry is observed. Compared to the SEI found on hard carbon materials cycled in classical carbonate-based electrolytes, the here found SEI composition differes significantly, as in the former case the SEI is very comparable to that found on Li ion battery graphite anodes, which consists mainly of solvent decomposition products such as carbonates and ester-containing species 33. From the presented data, it cannot be ruled out that the TEG-DME solvent is involved in the SEI formation as well. The increase in the hydrocarbon contribution during cycling clearly indicates that either the binder or the solvent decompose, although TEG-DME was reported stable down to very low potentials vs. Li+/Li 39. According to Jache et al. the complexation of the Na ion changes with chain length of the glyme, causing the intercalation potential to shift. This makes the stability of glymes a function of the chain length as well. The authors reported higher Coulombic efficiencies for 1G and 2G glymes, whereas 3G and 4G glymes showed ‘minor side reactions’

32

. Since no further experimental data on the product of these side reactions is

provided, it can be speculated if theses side reactions in the Na/4G system contribute to the increased CH intensity observed in this study.

The formation of polyethers as solvent

decomposition products cannot be excluded based on our data, as their characteristic C1s and O1s binding energies would coincide with the signals obtained from TEG-DME. Because of the similar structure of polyethers (compared to glyme solvents) and their ability to swell readily in 22 ACS Paragon Plus Environment

Page 23 of 37

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

the used electrolytes

52

, a polyether surface layer would most likely not hamper the solvent-co-

intercalation. Solvent and binder decomposition might be induced by the decomposition of the salt, such as reactive fluorine intermediates, and lead to the increased hydrocarbon intensity observed. Seh et al. recently studied the SEI layer properties of surface films formed on Na metal in various glyme solvents and sodium salts53. The authors found a denser, more uniform layer when NaPF6 is used, consisting mostly inorganic compounds (Na2O and NaF). In the same study, Na plating and stripping using NaFSI in diglyme was tested with significantly decreased Coulombic efficiency compared to NaPF6. These findings are comparable to our results and further support the idea that the salt plays a very crucial role in systems with no or limited solvent decomposition during SEI formation.

4. Conclusions In this study, the surface of graphite composite electrodes during the electrochemical cointercalation of sodium-TEG-DME was characterized using SoXPES. The PES results confirm the formation of a surface layer, which was indicated by irreversible capacity losses during the electrochemical characterization. The formed SEI seems to be fairly thin since bulk components such as graphite can be observed with PES even after electrochemical cycling, which means that the SEI thickness is in the range of the probing depth of 3-8 nm. In addition, the behavior of the SEI is dynamic and seems to break up during the large volume expansion upon formation of sodium ternary graphite intercalation compounds. Decomposition of the electrolyte salt NaFSI takes place at potentials below 0.5 V vs. Na+/Na and seems to be a key factor in the formation of the surface layer. However, the SEI formed in this specific battery system does not hinder the 23 ACS Paragon Plus Environment

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 24 of 37

solvent-co-intercalation, and solvent insertion and de-insertion could be directly observed from changes in the relative intensity of the C1s and O1s spectra. This first dedicated study of the surface film evolution during the formation of t-Na-GICs can be viewed as a baseline for future work on electrochemically optimized and highly reversible systems. Without the here presented adjustments to the electrochemical system with respect to the choice of the electrolyte salt and binder, important findings such as the identification of salt decomposition would have been severely complicated due to overlapping signals. Therefore, these preliminary findings represent valuable guidelines for future work to focus on electrolyte salt contributions to the SEI formation when solvent decomposition is very limited. Furthermore, our results indicate two potential pathways to further optimize Na-tGICs for SIB applications: - By limiting the cycling potential, salt decomposition can be avoided (in case of NaFSI to 0.5 V vs. Na+/Na), albeit this will also limit the accessible capacity. Reactive species formed in the salt decomposition process can cause decomposition of the solvent and binder, leading to increased surface layer formation, which can limit the reversibility of the tGIC formation. - The choice of binder for composite graphite electrodes differ for SIBs compared to LIBs. The high degree of adhesion of PAA-Na found beneficial in LIBs, rather seems disadvantageous for the formation of Na-tGICs.

Supporting Information. The following files are available free of charge. Cyclic voltammetry and additional galvanostatic cycles of PAA-Na graphite electrodes cycled with a 1 M NaFSI TEG-DME electrolyte. Reference spectra of a pristine graphite electrode and the binder PAA-Na. (PDF) 24 ACS Paragon Plus Environment

Page 25 of 37

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

Corresponding Author *Dr. Julia Maibach, [email protected] Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. ACKNOWLEDGMENT The authors would like to acknowledge H. Eriksson, R. Younesi and T. Nordh for technical assistance. M. Lacey is gratefully acknowledged for valuable suggestions. M. Valvo acknowledges funding by the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (Formas) via the personal grant no. 245-2014-668. Financial support has also been received from the Swedish Research Council, grant no. 2012-3837. The research leading to these results has also received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 608575.

References

(1) Wadia, C.; Albertus, P.; Srinivasan, V. Resource Constraints on the Battery Energy Storage Potential for Grid and Transportation Applications. J. Power Sources 2011, 196, 1593-1598. (2) Tarascon, J. M. Key Challenges in Future Li-Battery Research. Philos. Trans. R. Soc., A 2010, 368, 3227-3241. (3) Yabuuchi, N.; Kubota, K.; Dahbi, M.; Komaba, S. Research Development on Sodium-Ion Batteries. Chem. Rev. 2014, 114, 11636-11682.

25 ACS Paragon Plus Environment

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 26 of 37

(4) Palomares, V.; Serras, P.; Villaluenga, I.; Hueso, K. B.; Carretero-González, J.; Rojo, T. NaIon Batteries, Recent Advances and Present Challenges to Become Low Cost Energy Storage Systems. Energy Environ. Sci. 2012, 5, 5884-5901. (5) Ellis, B. L.; Nazar, L. F. Sodium and Sodium-Ion Energy Storage Batteries. Curr. Opin. Solid State Mater. Sci. 2012, 16, 168-177. (6) Hong, S. Y.; Kim, Y.; Park, Y.; Choi, A.; Choi, N. S.; Lee, K. T. Charge Carriers in Rechargeable Batteries: Na Ions Vs. Li Ions. Energy Environ. Sci. 2013, 6, 2067-2081. (7) Ge, P.; Fouletier, M. Electrochemical Intercalation of Sodium in Graphite. Solid State Ionics 1988, 28-30, 1172-1175. (8) Stevens, D. A.; Dahn, J. R. The Mechanisms of Lithium and Sodium Insertion in Carbon Materials. J. Electrochem. Soc. 2001, 148, A803-A811. (9) Yazami, R.; Touzain, P. A Reversible Graphite-Lithium Negative Electrode for Electrochemical Generators. J. Power Sources 1983, 9, 365-371. (10)

Doeff, M. M.; Ma, Y.; Visco, S. J.; De jonghe, L. C. Electrochemical Insertion of Sodium

into Carbon. J. Electrochem. Soc. 1993, 140, L169-L170. (11)

Nobuhara, K.; Nakayama, H.; Nose, M.; Nakanishi, S.; Iba, H. First-Principles Study of

Alkali Metal-Graphite Intercalation Compounds. J. Power Sources 2013, 243, 585-587. (12)

Sangster, J. C-Na (Carbon-Sodium) System. J. Phase Equilib. Diffus. 2007, 28, 571-579.

26 ACS Paragon Plus Environment

Page 27 of 37

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

(13)

Stevens, D. A.; Dahn, J. R. In Situ Small-Angle X-Ray Scattering Study of Sodium

Insertion into a Nanoporous Carbon Anode Material within an Operating Electrochemical Cell. J. Electrochem. Soc. 2000, 147, 4428-4431. (14)

Stevens, D. A.; Dahn, J. R. High Capacity Anode Materials for Rechargeable Sodium-Ion

Batteries. J. Electrochem. Soc. 2000, 147, 1271-1273. (15)

Dahbi, M.; Nakano, T.; Yabuuchi, N.; Ishikawa, T.; Kubota, K.; Fukunishi, M.;

Shibahara, S.; Son, J. Y.; Cui, Y. T.; Oji, H.; Komaba, S. Sodium Carboxymethyl Cellulose as a Potential Binder for Hard-Carbon Negative Electrodes in Sodium-Ion Batteries. Electrochem. Commun. 2014, 44, 66-69. (16)

Alcántara, R.; Jiménez-Mateos, J. M.; Lavela, P.; Tirado, J. L. Carbon Black: A

Promising Electrode Material for Sodium-Ion Batteries. Electrochem. Commun. 2001, 3, 639642. (17)

Chevrier, V. L.; Ceder, G. Challenges for Na-Ion Negative Electrodes. J. Electrochem.

Soc. 2011, 158, A1011-A1014. (18)

Baggetto, L.; Allcorn, E.; Manthiram, A.; Veith, G. M. Cu2Sb Thin Films as Anode for

Na-Ion Batteries. Electrochem. Commun. 2013, 27, 168-171. (19)

Darwiche, A.; Marino, C.; Sougrati, M. T.; Fraisse, B.; Stievano, L.; Monconduit, L.

Better Cycling Performances of Bulk Sb in Na-Ion Batteries Compared to Li-Ion Systems: An Unexpected Electrochemical Mechanism. J. Am. Chem. Soc. 2012, 134, 20805-20811.

27 ACS Paragon Plus Environment

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

(20)

Page 28 of 37

Darwiche, A.; Sougrati, M. T.; Fraisse, B.; Stievano, L.; Monconduit, L. Facile Synthesis

and Long Cycle Life of SnSb as Negative Electrode Material for Na-Ion Batteries. Electrochem. Commun. 2013, 32, 18-21. (21)

Komaba, S.; Matsuura, Y.; Ishikawa, T.; Yabuuchi, N.; Murata, W.; Kuze, S. Redox

Reaction of Sn-Polyacrylate Electrodes in Aprotic Na Cell. Electrochem. Commun. 2012, 21, 65-68. (22)

Fullenwarth, J.; Darwiche, A.; Soares, A.; Donnadieu, B.; Monconduit, L. Nip3: A

Promising Negative Electrode for Li- and Na-Ion Batteries. J. Mater. Chem. A 2014, 2, 20502059. (23)

Valvo, M.; Lindgren, F.; Lafont, U.; Björefors, F.; Edström, K. Towards More

Sustainable Negative Electrodes in Na-Ion Batteries Via Nanostructured Iron Oxide. J. Power Sources 2014, 245, 967-978. (24)

Wang, J.; Luo, C.; Gao, T.; Langrock, A.; Mignerey, A. C.; Wang, C. An Advanced

MoS2/Carbon Anode for High-Performance Sodium-Ion Batteries. Small 2015, 11, 473-481. (25)

Cui, Z.; Li, C.; Yu, P.; Yang, M.; Guo, X.; Yin, C. Reaction Pathway and Wiring

Network Dependent Li/Na Storage of Micro-Sized Conversion Anode with Mesoporosity and Metallic Conductivity. J. Mater. Chem. A 2015, 3, 509-514. (26)

Wen, Y.; He, K.; Zhu, Y.; Han, F.; Xu, Y.; Matsuda, I.; Ishii, Y.; Cumings, J.; Wang, C.

Expanded Graphite as Superior Anode for Sodium-Ion Batteries. Nat. Commun. 2014, 5, 4033.

28 ACS Paragon Plus Environment

Page 29 of 37

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

(27)

Cao, Y.; Xiao, L.; Sushko, M. L.; Wang, W.; Schwenzer, B.; Xiao, J.; Nie, Z.; Saraf, L.

V.; Yang, Z.; Liu, J. Sodium Ion Insertion in Hollow Carbon Nanowires for Battery Applications. Nano Lett. 2012, 12, 3783-3787. (28)

Jache, B.; Adelhelm, P. Use of Graphite as a Highly Reversible Electrode with Superior

Cycle Life for Sodium-Ion Batteries by Making Use of Co-Intercalation Phenomena. Angew. Chem. Int. Ed. 2014, 53, 10169-10173. (29)

Kim, H.; Hong, J.; Park, Y.-U.; Kim, J.; Hwang, I.; Kang, K. Sodium Storage Behavior in

Natural Graphite Using Ether-Based Electrolyte Systems. Adv. Funct. Mater. 2015, 25, 534541. (30)

Kim, H.; Hong, J.; Yoon, G.; Kim, H.; Park, K. Y.; Park, M. S.; Yoon, W. S.; Kang, K.

Sodium Intercalation Chemistry in Graphite. Energy Environ. Sci. 2015, 8, 2963-2969. (31)

Zhu, Z. Q.; Cheng, F. Y.; Hu, Z.; Niu, Z. Q.; Chen, J. Highly Stable and Ultrafast

Electrode Reaction of Graphite for Sodium Ion Batteries. J. Power Sources 2015, 293, 626634. (32)

Jache, B.; Binder, J. O.; Abe, T.; Adelhelm, P. A Comparative Study on the Impact of

Different Glymes and Their Derivatives as Electrolyte Solvents for Graphite Co-Intercalation Electrodes in Lithium-Ion and Sodium-Ion Batteries. PCCP 2016, 18, 14299-14316. (33)

Komaba, S.; Murata, W.; Ishikawa, T.; Yabuuchi, N.; Ozeki, T.; Nakayama, T.; Ogata,

A.; Gotoh, K.; Fujiwara, K. Electrochemical Na Insertion and Solid Electrolyte Interphase for Hard-Carbon Electrodes and Application to Na-Ion Batteries. Adv. Funct. Mater. 2011, 21, 3859-3867.

29 ACS Paragon Plus Environment

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

(34)

Page 30 of 37

Ponrouch, A.; Dedryvère, R.; Monti, D.; Demet, A. E.; Ateba Mba, J. M.; Croguennec,

L.; Masquelier, C.; Johansson, P.; Palacín, M. R. Towards High Energy Density Sodium Ion Batteries through Electrolyte Optimization. Energy Environ. Sci. 2013, 6, 2361-2369. (35)

Philippe, B.; Valvo, M.; Lindgren, F.; Rensmo, H.; Edström, K. Investigation of the

Electrode/Electrolyte Interface of Fe2O3 Composite Electrodes: Li Vs Na Batteries. Chem. Mater. 2014, 26, 5028-5041. (36)

Vogt, L. O.; El Kazzi, M.; Jämstorp Berg, E.; Pérez Villar, S.; Novák, P.; Villevieille, C.

Understanding the Interaction of the Carbonates and Binder in Na-Ion Batteries: A Combined Bulk and Surface Study. Chem. Mater. 2015, 27, 1210-1216. (37)

Bodenes, L.; Darwiche, A.; Monconduit, L.; Martinez, H. The Solid Electrolyte

Interphase a Key Parameter of the High Performance of Sb in Sodium-Ion Batteries: Comparative X-Ray Photoelectron Spectroscopy Study of Sb/Na-Ion and Sb/Li-Ion Batteries. J. Power Sources 2014, 273, 14-24. (38)

Peled, E. Electrochemical Behavior of Alkali and Alkaline Earth Metals in Nonaqueous

Battery Systems - the Solid Electrolyte Interphase Model. J. Electrochem. Soc. 1979, 126, 2047-2051. (39)

Tobishima, S.; Morimoto, H.; Aoki, M.; Saito, Y.; Inose, T.; Fukumoto, T.; Kuryu, T.

Glyme-Based Nonaqueous Electrolytes for Rechargeable Lithium Cells. Electrochim. Acta 2004, 49, 979-987.

30 ACS Paragon Plus Environment

Page 31 of 37

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

(40)

Aurbach, D.; Granot, E. The Study of Electrolyte Solutions Based on Solvents from the

"Glyme" Family (Linear Polyethers) for Secondary Li Battery Systems. Electrochim. Acta 1997, 42, 697-718. (41)

Winter, M.; Besenhard, J. O.; Spahr, M. E.; Novák, P. Insertion Electrode Materials for

Rechargeable Lithium Batteries. Adv. Mater. 1998, 10, 725-763. (42)

Jeschull, F.; Brandell, D.; Edstrom, K.; Lacey, M. J. A Stable Graphite Negative

Electrode for the Lithium-Sulfur Battery. Chem. Commun. 2015, 51, 17100-17103. (43)

Jeschull, F.; Lacey, M. J.; Brandell, D. Functional Binders as Graphite Exfoliation

Suppressants in Aggressive Electrolytes for Lithium-Ion Batteries. Electrochim. Acta 2015, 175, 141-150. (44)

Malmgren, S.; Ciosek, K.; Lindblad, R.; Plogmaker, S.; Kuhn, J.; Rensmo, H.; Edstrom,

K.; Hahlin, M. Consequences of Air Exposure on the Lithiated Graphite Sei. Electrochim. Acta 2013, 105, 83-91. (45)

Komaba, S.; Yabuuchi, N.; Ozeki, T.; Okushi, K.; Yui, H.; Konno, K.; Katayama, Y.;

Miura, T. Functional Binders for Reversible Lithium Intercalation into Graphite in Propylene Carbonate and Ionic Liquid Media. J. Power Sources 2010, 195, 6069-6074. (46)

Painter, L. R.; Arakawa, E. T.; Williams, M. W.; Ashley, J. C. Optical-Properties of

Polyethylene - Measurement and Applications. Radiat. Res. 1980, 83, 1-18. (47)

Powell, C. J.; Jablonski, A. Nist Electron Inelastic-Mean-Free-Path Database, Version

1.2, Srd 71. National Insitute of Standards and Technology: Gaithersburg, MD, 2010.

31 ACS Paragon Plus Environment

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

(48)

Page 32 of 37

Philippe, B.; Dedryvere, R.; Gorgoi, M.; Rensmo, H.; Gonbeau, D.; Edstrom, K.

Improved Performances of Nanosilicon Electrodes Using the Salt LiFSI: A Photoelectron Spectroscopy Study. J. Am. Chem. Soc. 2013, 135, 9829-9842. (49)

Oltean, V. A.; Philippe, B.; Renault, S.; Duarte, R. F.; Rensmo, H.; Brandell, D.

Investigating the Interfacial Chemistry of Organic Electrodes in Li- and Na-Ion Batteries. Chem. Mater. 2016, 28, 8742-8751. (50)

Han, H. B.; Zhou, S. S.; Zhang, D. J.; Feng, S. W.; Li, L. F.; Liu, K.; Feng, W. F.; Nie, J.;

Li, H.; Huang, X. J.; Armand, M.; Zhou, Z. B. Lithium Bis(Fluorosulfonyl)Imide (LiFSI) as Conducting Salt

for Nonaqueous

Liquid

Electrolytes for

Lithium-Ion

Batteries:

Physicochemical and Electrochemical Properties. J. Power Sources 2011, 196, 3623-3632. (51)

Otaegui, L.; Goikolea, E.; Aguesse, F.; Armand, M.; Rojo, T.; Singh, G. Effect of the

Electrolytic Solvent and Temperature on Aluminium Current Collector Stability: A Case of Sodium-Ion Battery Cathode. J. Power Sources 2015, 297, 168-173. (52)

Aihara, Y.; Appetecchi, G. B.; Scrosati, B. A New Concept for the Formation of

Homogeneous, Poly(Ethylene Oxide) Based, Gel-Type Polymer Electrolyte. J. Electrochem. Soc. 2002, 149, A849-A854. (53)

Seh, Z. W.; Sun, J.; Sun, Y. M.; Cui, Y. A Highly Reversible Room-Temperature Sodium

Metal Anode. ACS Cent. Sci. 2015, 1, 449-455.

32 ACS Paragon Plus Environment

Voltage / V vs. Na+/Na

Elapsed time / h Page ACS 33 ofApplied 37 Materials & Interfaces 3.0 0 1 2 3 4 5 6 7 8 9 10 11 2.5

(a)

1st cycle 2nd cycle

Voltage / V vs. Na+/Na

1 2 2.0 3 4 1.5 5 6 1.0 7 8 0.5 9 0.0 10 0 25 50 75 100 125 150 175 11 Gravimetric capacity / mAhg-1 12 13 143.0 OCV (b) 15 2.5 16 17 V IX 182.0 19 201.5 21 I VI 221.0 II VII 23 III 240.5 IV VIII 25 ACS Paragon Plus Environment 0.0 26 0 5 10 15 20 25 30 35 27 Elapsed time / h

more surface sensitive hv = 835 eV

OCV

PAA

hv = 835 eV C1s graphite

PAA

Page 34 of 37

hv = 430 eV

C1s

PAA

CH TEG-DME OCV I (0.9 V)

II

II (0.7 V)

III

III (0.5 V)

IV

IV (1st EOD)

V

V (1st EOC)

VI

VI (0.9 V)

VII

VII (0.6 V)

VIII

VIII (2nd EOD)

ACS Paragon Plus Environment 532

528

536

Binding energy / eV

532

528

292

288

284

280

IX (2nd EOC) 292

Binding energy / eV

288 284

280

2nd charge (de-sodiation)

IX 536

2nd discharge (sodiation)

I

1st charge (de-sodiation)

Intensity / arb. units

O1s TEG-DME

hv = 680 eV O1s TEG-DME

1st discharge (sodiation)

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

more surface sensitive ACS Applied Materials & Interfaces

Page 37 F1s35 ofNaFSI 1 OCV

N1s

ACS Applied S2p Materials & Interfaces NaFSI

NaFSI

salt decomp.

NaF

OCV

I (0.9 V) II (0.7 V)

Intensity / arb. units

2 3I 4 5 6 7 II 8 9 10III 11 12 13IV 14 15 16V 17 18 VI 19 20 21VII 22 23 24VIII 25 26 27 28IX 29 30 692 688 684 31 Binding energy / eV 32

Na2s

III (0.5 V)

IV (1st EOD)

V (1st EOC)

VI (0.9 V)

VII (0.6 V)

VIII (2nd EOD)

ACS Paragon Plus Environment 410 405 400 395 Binding energy / eV

175

170 165 160 155 68 64 60 56 Binding energy / eV Binding energy / eV

IX (2nd EOC)

ACS Page 36 of 37 OCVApplied Materials & Interfaces I (0.9 V)

1 II (0.7 V) 2 3 III (0.5 V) 4 5IV (1st EOD) 6 V (1st EOC) 7 8 VI (0.9 V) 9 10 VII (0.6 V) 11 VIII (2nd EOD) 12 13 IX (2nd EOC) ACS Paragon Plus Environment 14 0% 20% 40% 60% 80% 100% 15 Relative intensities 16

C O F N S Na

O1s

C1s

TEG-DME

O1sTEG-DME C1s TEG-DME

graphite

CH

PAA ACS Page Applied 37 of 37Materials & Interfaces 536

532

528

X-ray

292

288

284

536

280

532

528

292

X-ray

e-

288

e-

284

280

1 Na 2ACS Paragon Plus Environment TEG-DME 3 1 sodiation 4 mAh g 5 V vs. Na+/Na

+

st

2.0 1.0 0.0

0

25

50

75

-1

100

125

150

175