MoS2 锂离子电池

ARTICLEMoS2/Graphene Composite Paper for Sodium-Ion Battery ElectrodesLamuel David, Romil Bhandavat, and Gurpreet Singh* Mechanical and Nuclear Engineering Department, Kansas State University, Manhattan, Kansas 66506, United StatesABSTRACT We study the synthesis and electrochemical and mechanical performance of layered free-standing papers composed of acid-exfoliated few-layer molybdenum disul de (MoS2) and reduced graphene oxide (rGO) akes for use as a self-standing exible electrode in sodium-ion batteries. Synthesis was achieved through vacuum ltration of homogeneous dispersions consisting of varying weight percent of acid-treated MoS2 akes in GO in DI water, followed by thermal reduction at elevated temperatures. The electrochemical performance of the crumpled composite paper (at 4 mg cmÀ2) was evaluated as counter electrode against pure Na foil in a half-cell con guration. The electrode showed good Na cycling ability with a stable charge capacity of approximately 230 mAh gÀ1 with respect to total weight of the electrode with Coulombic e ciency reaching approximately 99%. In addition, static uniaxial tensile tests performed on crumpled composite papers showed high average strain to failure reaching approximately 2%.KEYWORDS: TMDC . sodium battery . free-standing electrode . graphene . MoS2ithium-ion batteries (LIBs) have been extensively studied for energy-storage applications like portable electronic devices and electric vehicles.1À3 However, concerns over the cost, safety, and availability of Li reserves4 for large-scale applications involving renewable energy integration and electrical grid have to be answered. In this regard, sodium-ion batteries (SIBs) have drawn increasing attention because, in contrast to lithium,5À7 sodium resources are practically inexhaustible and evenly distributed around the world while the ion insertion chemistry is largely identical to that of lithium. Also, from an electrochemical point of view, sodium has a very negative redox potential (À2.71 V vs SHE) and a small electrochemical equivalent (0.86 gA hÀ1), which make it the most advantageous element for battery applications after lithium. However, many challenges remain before SIBs can become commercially competitive with LIBs. For instance, Na ions are about 55% larger in radius than Li ions, which makes it di cult to nd a suitable host material to allow reversible and rapid ion insertion and extraction.8 To this end, researchers have proposed a number of high-capacity sodium host materials (negative electrode) involving either carbon or group IVA and VA elements that form intermetallic compounds with Na.9À13DAVID ET AL.LThe alloying compounds demonstrate high rst cycle Na-storage capacities, such as Na15Sn4 (847 mAh gÀ1), Na15Pb4 (485 mAh gÀ1), Na3Sb (600 mAh gÀ1), and Na3P (2560 mAh gÀ1). However, this comes at the cost of very high volume change upon Na insertion (as much as 500% in some cases), resulting i

n formation of internal cracks, loss of electrical contact, and eventual failure of the electrode (particularly for thick electrodes).14 Novel nanostructured designs that can accommodate large volumetric strains need further exploration.15À18 For carbon-based electrode materials, much of the emphasis has been on hard carbons due to large interlayer spacing and disordered structure.19À25 For example, hard carbon prepared from pyrolyzed glucose, carbon black, and carbon microspheres has been shown to exhibit initial reversible capacities of 300, 200, and 285 mAh gÀ1, respectively, in a Na-ion cell.15À17 More recently, another hard carbon material that could deliver a reversible capacity of more than 200 mAh gÀ1 over 100 cycles has been reported.22,25 However, these studies were conducted on traditional anode architecture (prepared through slurry coating of active material on metallic current collector foil, and the capacities reported were with respect to the active material only), either at low cycling current rates or at elevated temperatures. Overall,VOL. 8’* Address correspondence to gurpreet@ksu.edu. Received for review November 29, 2013 and accepted January 21, 2014. Published online January 21, 2014 10.1021/nn406156bC 2014 American Chemical SocietyNO. 2’1759 – 1770’2014 1759ARTICLEnew electrode design and concepts based on chemistry other than alloying and ion intercalation must also be explored to realize improved performance in Na-ion batteries under normal operating conditions. Studies on Li-ion batteries have shown that 2-D layered nanomaterials such as graphene and transition metal dichalcogenides (TMDCs e.g., MoS2 and WS2) are promising materials for e cient storage and release of Li ions.26À37 However, when compared with graphite, the electrochemical lithiation in layered TMDCs is distinct as majority of the lithium is stored by means of a conversion reaction in which Li ion reacts with the TMDC forming Li2S and transition metal phases as the reaction products. More important, this type of a conversion reaction can allow transfer of 2À6 electrons per transition metal compared to a single electron in the case of an intercalation reaction (lithium/carbon system).1,34 Although layered graphite has been ruled out for sodium-based systems (as Na ions do not tend to form staged intercalation compounds with graphite),38À40 a graphene-based freestanding paper-based electrode can provide a porous and exible support structure for a TMDC to undergo a reversible conversion-type reaction with Na ions. It can also act as an e cient electronic current collector, thereby eliminating the need for a metallic substrate (generally a 10 μm thick foil at 10 mg cmÀ2),41,42 electrically conducting additives, and polymeric binders that amount to a total of approximately 10% of the cell weight42 in traditional negative electrodes.43À48 Herein, we provide the rst report of (a) synthesis of compos

ite papers from acid-functionalized MoS2 and reduced graphene oxide akes, (b) improved capacity and high e ciency reversible Na storage in the selfstanding exible MoS2/graphene electrodes at room temperature, and (c) mechanical characterization that highlights the high strain to failure in these composite papers. RESULTS AND DISCUSSION Layered “as-obtained” MoS2 was exfoliated by mechanical sonication in chlorosulfonic acid followed by quenching in DI water (see Methods, Materials, and Instrumentation section). A digital image of the acid-treated MoS2 dispersion immediately after quenching in DI water is shown in Figure 1a. From SEM observations, the particle size for MoS2-raw was observed to be approximately 20À40 μm (Figure 1b), while that of MoS2-SA was less than 20 μm (Figure 1c). Shown in Figure 1dÀf are high-magni cation TEM images of acid-treated MoS2 sheets. All of the sheets were observed to be only a few layers thick with ake size ranging from 100 nm to 1 μm. From literature, we can correlate the reason for exfoliation of MoS2 to electrostatic repulsion forces caused by protonation of MoS2 surfaces.50 Using DLVO theory, ζ-potential measurements can quantify this surface charge onDAVID ET AL.MoS2 sheet and hence help in establishing the dispersion stability. For ζ-potential measurements, the pH was varied by adding 0.01 M NaOH solution, and since the contribution from dissociated OHÀ ions in the measured potential is minimal, it was neglected in the analysis. The lower pH range was limited to protect the instrument electrode. The surface potential showed a range varying from À35 mV at pH 4 to À60 mV at pH 10, shown as an inset in Figure 1g. As higher surface potential (negative) implies more stable suspensions, based on the obtained results, higher pH suggests a larger exposed MoS2 sheet surface. This dependence of surface potential on pH is similar to that observed for exfoliated (surfacefunctionalized) graphene sheets by Coleman's group.51 Further, we utilize their model for graphene stabilization mechanism to explain the superacidÀMoS2 interaction mechanism (see Supporting Information).51,52 Figure 1g is the plot for total interaction energy per unit area of the sheet (VT/A). Further analysis involved Raman spectroscopy and X-ray di raction before and after acid treatment. Raman spectrum (Figure 1h) obtained by use of 633 nm wavelength laser showed typical E2g1, A1g, 2LA(M), and A1gþLA(M) peaks at 380, 407, 460, and 641 cmÀ1, respectively. The in-plane E2g1 peak results from opposite vibration of two S atoms with respect to the Mo atom, while the A1g peak is associated with the out-of-plane vibration of only S atoms in opposite directions.53,54 The intensity of A1g peak arises from the resonance Raman (RR) scattering because the incident laser is in resonance with the direct band gap (～1.96 eV) at the K point. The asymmetric 2LA(M) peak is associated with second-order zone-edge phonon (LA(M)) and a

rst-order optical phonon peak (A2u).55À58 These results along with electron microscopy results suggest that the structure is relatively undistorted MoS2.59,60 Further, X-ray di raction analysis (XRD) of MoS2-raw and MoS2-SA in Figure 1i showed distinct peak at 14.3 and 13.97° 2θ, respectively. These peaks are associated with 002 re ection from the basal plane of MoS2 with measured “d” spacing of 3.1 and 3.2 Å lattice plane of hexagonal MoS2 (JCPDS #37-1492). This suggests restacking of the MoS2 layers upon drying.61,62 Later, the MoS2/rGO papers were prepared by vacuum ltration of graphene oxide (GO) and molybdenum disul de (MoS2) sheets dispersed in water/ isopropyl alcohol (1:1) solution, which is shown with the help of a schematic in Figure 2a, while Figure 2b is the digital image of one such paper synthesized using this technique. SEM images in Figure 2c,d and Supporting Information Figure S1aÀh show the top-view and corresponding cross-section view of 60MoS2, rGO, 20MoS2, 40MoS2, and 60MoS2-raw free-standing papers, respectively. The papers were approximately 10À20 μm in thickness (that varied with the weight percentage of MoS2 in GO) with a relatively homogeneousVOL. 8’NO. 2’1759 – 1770’2014 1760ARTICLEFigure 1. (a) Photographic image showing acid-treated MoS2 immediately after it was quenched in 1 L of DI water at a concentration of approximately 2 mg mLÀ1. SEM images showing the structure and size distribution of MoS2 (b) before and (c) after superacid treatment. The particle size varied between 20 and 40 μm and approximately 1 and 20 μm for raw and acidtreated MoS2, respectively. The scale bar in the inset is 2 μm. (d,e) High-resolution TEM image of superacid-treated MoS2 (MoS2-SA) sheets. (f) SAED pattern corresponding to TEM image in (e). (g) Graph showing calculated total interaction potential energy (VT), repulsion (VDLVO), and attraction energy (VvdW) (per unit area) with increasing MoS2 sheet separation distance (log scale). Inset: Experimentally measured ζ-potential, showing better dispersion stability at higher pH values. (h) Raman spectra of MoS2 before and after acid treatment. The similarity in the relative intensity and position of the E2g1 and A1g peaks suggests that the structure was largely undistorted MoS2. (i) Change in intensity and fwhm of MoS2 peak at 14° 2θ in the XRD spectra suggests increase in MoS2 interlayer distance after the acid treatment (JCPDS #37-1492).composition (see Supporting Information Figure S2 for high-resolution image). The interleaved structure observed in the cross-sectional images is preferred for easy storage and release of larger Na ions, particularly at higher current densities or C rates. The digital photograph in the inset of Supporting Information Figure S1aÀd con rms the outstanding structural exibility of rGO and MoS2/graphene specimens. Further analysis involved SEM-X-ray energy-dispersive spectroscopy (EDX)

, shown in the inset in Figure 2c. The EDX spectra from spot 1 (square) showed peaks at 0.27 and 0.52 keV, which correspond to carbon (85.43 atom %) and oxygen (10.41 atom %) KR energy, respectively. The low oxygen content indicates that rGO was highly reduced and pristine. In addition toDAVID ET AL.carbon and oxygen, two small peaks at 2.29 and 2.3 keV corresponding to molybdenum LR (1.39 atom %) and sulfur KR (2.78 atom %) energy, respectively, were also observed. At spot 2, peaks corresponding to Mo (26.13 atom %) and S (46.48 atom %) were prominent when compared to that of carbon (24.39 atom %), which unambiguously con rms the presence of MoS2 sheets in the composite. TEM images are shown in Figure 2e and Supporting Information Figure S1iÀl. It is clear that the rGO sheets were layered with few layers of MoS2 forming a very good electron conductive layer and also a support structure for free-standing paper. The inset in Figure 2e is the selected area electron di raction (SAED) pattern that indicates a multiple spot pattern, one of which is due to the polycrystallinity of restackedVOL. 8’NO. 2’1759 – 1770’2014 1761ARTICLEFigure 2. (a) Schematic representation showing synthesis of rGO/MoS2 composite paper. (b) Digital picture showing largearea composite paper prepared through vacuum ltration. (c) SEM top-view image of 60MoS2 paper; inset shows the EDX spectra of spots in the SEM image indicating the material to be rGO (square) and MoS2 (circle). The scale is 10 μm. (d) Corresponding SEM cross-sectional images show the morphology of the paper. Average thickness of this paper was observed to be ～20 μm. (e) TEM image and SAED pattern of 60MoS2. The MoS2 sheets are observed to be wrapped by much larger graphene sheets. In SAED pattern (inset), multiple spot patterns are observed, one of which is due to the polycrystallinity of restacked rGO sheets while a second set of spot patterns is assigned to MoS2 sheets. The scale is 100 nm. (f) Thermogravimetric analysis data for MoS2-SA, MoS2-rGO composite paper, and rGO paper. (g) Graph of electrical conductivity vs MoS2 loading in the composite paper.rGO sheets, while a second set of spot patterns was due to MoS2 sheets. The hexagonal spot pattern (inset of Supporting Information Figure S1i) indicates that the graphitic AB stacking was preserved in the lattice after thermal reduction. To further observe the distribution of MoS2 in the composite, EDX elemental mapping was performed on the cross section of the 60MoS2DAVID ET AL.specimen (Supporting Information Figure S3). A slightly higher percentage of elemental carbon (graphene) was observed on one end of the paper, which is somewhat typical in a high inclusion content (in this case, MoS2 in graphene) matrix.63 Supporting Information Table S1 summarizes the percentage of each element detected in the corresponding EDX map.VOL. 8’NO. 2’1759 – 1770’2014 1762A

RTICLEFurther evidence showing the presence of MoS2 in rGO was achieved through X-ray photoelectron spectroscopy (XPS) (Supporting Information Figure S4) that compares powered MoS2-SA with 60MoS2 paper before and after reduction. MoS2 peaks that were present in the starting material were also present in the freestanding paper. Notable change was observed in the intensity of oxygen (O1s) and carbon (C1s) peaks for the 60MoS2 specimen (before and after reduction) due to addition of GO. Raman spectroscopy of 60MoS2 paper (Supporting Information Figure S5) showed typical MoS2 peaks at 373 (E2g1), 400 (A1g), and 445 (2LA(M)) cmÀ1 along with characteristic rGO D and G peaks at 1330 (D) and 1560 (G) cmÀ1, respectively. X-ray analysis (Supporting Information Figure S6) of 60MoS2 paper also con rmed the presence of rGO (JCPDS #01-0646) and MoS2 (JCPDS #37-1492) in the composite. Further, the exact amount of MoS2 in the nal composite paper (after reduction) was inferred by carrying out thermogravimetric analysis (TGA) in owing air. As can be seen in Figure 2f, rGO and MoS2 had oxidation events at approximately 450 and 700 °C, respectively. From the TGA data, it was observed that thermally reduced 60MoS2, 40MoS2, and 20MoS2 papers had approximately 73, 53, and 35 wt % of MoS2 in rGO. Later, the e ect of change in ller concentration on electrical conductivity of the composite was studied by use of a four-point measurement technique, which is presented in Figure 2g. The increase in conductivity with increasing rGO concentration in the composite was somewhat nonlinear. This type of behavior is typical when electrically insulating ller (MoS2) is added to a relatively conducting matrix (rGO) because signi cant increase in conductivity can only occur after the rst conducting path through the sample is formed.64,65 Electrochemical Performance. Supporting Information Figure S7 depicts the voltage charge/discharge and differential capacity curves for various paper electrodes with varying MoS2 content. Supporting Information Figure S7a shows the voltage profiles of rGO for the first and second cycle. The first cycle discharge and charge capacities were 784 and 86 mAh gÀ1. The differential capacity profiles in Supporting Information Figure S7b show a primary reduction peak at 200 mV, a secondary reduction peak at 610 mV, and a weak oxidation peak at 0.9 V. The peak at 200 mV, which is present in all subsequent cycles, is associated with intercalation of rGO, while the peak at 610 mV suggests formation of a solid electrolyte interphase (SEI) layer, which exists only in the first cycle. Supporting Information Figure S7c,d shows the initial charge/discharge voltage profile and differential capacity curves for the 20MoS2 electrode. In the first cycle, there are three reduction peaks. A peak at 150 mV is attributed to MoS2/rGO intercalation, while those at 580 mV and 0.8 V are attributed to SEI formation in rGO and MoS2, respectively, as these we

re present only during the firstDAVID ET AL.cycle. Only one subtle anodic peak at 1.35 V was observed. As the percentage of MoS2 increased from 40% (Supporting Information Figure S7e,f) and 60% (Figure 3a,b), the domination of Na intercalation in MoS2 over rGO increased, which was evidently seen with an increase in the intensity of the reduction peak at around 0.8 and 0.9 V in the first cycle (peak at 580 mV observed in rGO electrode was relatively absent). In the case of the 60MoS2-raw (Supporting Information Figure S7g,h) electrode, similar peaks to that of 60MoS2 were observed. Later, reaction kinetics of rGO and 60MoS2 electrodes were compared by performing galvanostatic intermittent titration (GITT) cycling and calculating the reaction resistance to Na insertion and extraction for the two electrodes. For rGO electrodes (Supporting Information Figure S8a,c,e), the reaction resistance for Na insertion was observed to be fairly constant at ～10 Ohm g (or less) while it increased exponentially to ～50 Ohm g in the case of Na extraction at an extraction voltage of ～1.25 V. For the MoS2 electrode (Supporting Information Figure S8b,d,f), reaction resistance increased gradually to ～20 Ohm g during Na insertion. However, for the extraction half, the resistance remained stable at ～20 Ohm g until approximately 2.5 V and then saw a sudden increase reaching ～50 Ohm g. The sudden increase in reaction resistance could be attributed to successive stage transformation processes during sodiation/desodiation or lithiation/ deliathiation observed in layered intercalation compounds.66 Interestingly, both the electrodes showed a reaction resistance value of ～10 Ohm g during initial insertion (discharge) in the upper voltage range of 2 to 1.0 V (Supporting Information Figure S8c,d). This suggests that the initial Na insertion in 60MoS2 composite electrode in the upper voltage range could be an intercalation reaction, which is later followed by a conversion-type reaction in the lower voltage range (indicated by rise in resistance). Figure 3c shows the charge capacities and Coulombic e ciency of rGO, 20MoS2, 40MoS2, 60MoS2, and 60MoS2-raw anodes cycled at a constant current density of 25 mA gÀ1. For rGO, the charge capacity was stable at ～70.5 mAh gÀ1 in the 20th cycle, while the high irreversible rst cycle capacity is attributed to electrochemical reaction contributing to SEI layer formation. In the case of rGO/MoS2 composite electrodes, the rst cycle charge capacity increased with increasing percentage of MoS2 in the composite, that is, 20MoS2, 40MoS2, and 60MoS2 showed 139 , 263, and 338 mAh gÀ1, respectively. After initial drop in the capacity, rGO/MoS2 composite electrode remained constant at 123, 172, and 218 mAh gÀ1 for 20MoS2, 40MoS2, and 60MoS2, respectively. 60MoS2 anode was the best performing with 83% capacity retention and approximately 98% average e ciency. The MoS2-raw electrode showed a rst cycle charge capacity ofV

OL. 8’NO. 2’1759 – 1770’2014 1763ARTICLEFigure 3. (a) Voltage pro le of 60MoS2 free-standing electrode along with its corresponding (b) di erential capacity curves for the rst two cycles. (c) Sodium charge capacity of various electrodes at a constant current density of 25 mA gÀ1. (d) Sodium charge capacity and corresponding Coulombic e ciency of 60MoS2 electrode cycled at varying current densities.233 mAh gÀ1 that reduced to below ～100 mAh gÀ1 after 20 cycles. For the MoS2-raw electrode, it is possible that the formation of SEI with successive cycles (and unexfoliated nature of the akes) may have hindered the di usion of Na into the bulk of the specimen, resulting in capacity fading on consecutive cycling. For the acid-treated MoS2-SA electrode, the more open and interleaved structure enabled it to utilize the entire bulk of the material in the electrode, resulting in exceptional cyclic stability. Presence of conducting graphene sheets further provided the necessary platform on which volume or morphology changes due to conversion reaction could occur without any breakdown in the electrical contact. Later, the composite paper electrode with maximum possible MoS2 loading (i.e., 90% MoS2 in rGO) was cycled under similar conditions (Supporting InformationDAVID ET AL.Figure S9). However, these papers were very brittle and required special handing during cell assembly. The rst cycle discharge and charge capacities were observed at 943 and 347 mAh gÀ1, respectively. Even though the rst charge capacity was higher than that of other composite electrodes, the electrode started to show random spikes in the voltage pro le with a capacity drop after the second cycle. Therefore, rate capability tests were only performed on the best performing electrode specimen (i.e., 60MoS2). As shown in Figure 3d, the electrode stabilized to a charge capacity of 240 mAh gÀ1 at a current density of 25 mA gÀ1 (with respect to total weight of the electrode) after the initial ve cycles. The charge capacity remained stable (214 mAh gÀ1, 90% retention) even at current densities as high 100 mA gÀ1. The electrode regained most of its charge capacity (230 mAh gÀ1, 96% retention) whenVOL. 8’NO. 2’1759 – 1770’2014 1764ARTICLEFigure 4. (a) TEM images and corresponding SAED patterns of 60MoS2 electrode before and after rst discharge cycle (0.01 V). The ring-like SAED pattern suggests formation of nanocrystallites or amorphization of MoS2 in the sodiated (discharged) electrode.67,68 (b) XRD pattern of 60MoS2 electrode before and after rst discharge cycle (0.01 V). Broadened MoS2 (JCPDS #37-1492), rGO (JCPDS #01-0646), Mo (JCPDS #01-1207), and Na2S (JCPDS #65-0525) peaks were observed in the sodiated electrode.68À70the current density was brought back to 25 mA gÀ1 after 15 cycles. On further increasing the current density to 200 mA gÀ1, 72% (173 mAh gÀ1) of

the initial stable capacity was retained. Again, when the current density was brought back to the initial 25 mA gÀ1, the electrode recovered 87% of its capacity and remained stable for another ve cycles. It is remarkable that, even at 200 mA gÀ1, the electrode had a stable charge/ discharge cycle with no abnormalities, which suggests the improved mechanical stability of this interleaved architecture. Summary of the electrochemical data is presented in Table S3. Further, to check the integrity of the electrode specimen, the cells were disassembled and the electrodeDAVID ET AL.was recovered for further characterization. Supporting Information Figure S10 shows the digital photographs (aÀe), low-resolution (fÀj), and high-resolution (kÀo) SEM images of the dissembled cells after 20 cycles. No evidence of surface cracks, volume change, or physical imperfections could be observed in the SEM image, suggesting high mechanical/structural strength of the MoS2/rGO composite paper. In all cases, the evidence of formation of a thin layer covering the electrode surface, possibly the SEI layer, could be observed. The contamination in the specimen, indicated by the arrows, is from the residue of glass separator bers. Also, these anodes may have been exposed to air during the transfer process, resulting in oxidation of Na species,VOL. 8’NO. 2’1759 – 1770’2014 1765ARTICLEFigure 5. (a) Tensile test setup (1, load cell, xed; 2, clamps, top clamp not shown; 3, computer-controlled movable translation stage) with sample after fracture from loading (inset shows zoomed-in view of two such specimens). (b) Engineering stressÀstrain plot for rGO, 40MoS2, and 60MoS2 free-standing papers. Photographic image of (c) 40MoS2 and (d) 60MoS2 paper tested in this study. (e,f) Corresponding SEM cross-sectional images showing the fractured surface. The scale bar is 20 μm.which appeared as bright spots in the images (due to its nonconducting nature). Supporting Information Figure S11 shows the images obtained by EDX mapping of the electrode surface. Table S2 shows the atomic weight percent of various elements detected during the EDX mapping. A high percentage of sodium (19.46 atom %) was observed on the electrode surface, which is generally attributed to formation of a SEI layer during the electrochemical cycling process. Even higher percentage of surface oxygen (41.46 atom %) was observed, which may have come from oxidation of intercalated sodium metal. Further analysis involved disassembling 60MoS2 cell after rst discharge cycle. TEM images in Figure 4a show degradation of MoS2 sheets and possible amorphization (ring-like SAED pattern).67 Complementing results were observed in XRD and XPS analysis of the fully sodiated electrode, as shown in Figure 4b and Supporting Information Figure S12, respectively. Broad Mo peak and NaÀS peaks could be identi ed in both XRD (Mo at 29°, NaÀS at 72° 2θ) and XPS (Mo at 231 eV,

NaÀS at 160.6 eV) spectra. Also, the primary MoS2 peak at 14° 2θ (002) appeared broadened, further indicating degradation of MoS2 structure, most likely due to a conversion-type reaction with Na ions.68À70 On the basis of the voltage pro les, di erential capacity plots, GITT cycling data, and post-cycling analysis, we predict the mechanism of Na ion's reaction with the MoS2/graphene composite to be a combination of intercalation and conversion-type reaction thatDAVID ET AL.is generally observed in Li/TMDC34 and cathodes for Na-ion batteries.71,72 Supporting Information Figure S13 shows an idealized rGO/MoS2 structure (in reality, however, the acid-treated MoS2 sheets are wrapped by much larger rGO sheets) to illustrate the predicted reaction mechanism in the 2.0 to 0.1 V range. Step 1 is seen as a combination of Na intercalation reaction into the ordered MoS2 (～0.9 V) and later into the disordered NaxMoS2 layers (～0.8 V). Step 2 represents the conversion reaction resulting in breakdown of MoS2 into Mo and Na2S, as can be seen in the TEM (Figure 4a), XRD (Figure 4b), and XPS data from the fully sodiated cell in Supporting Information Figure S12 (further con rmed by the lower voltage plateau at ～0.12 V that was not observed in the discharge of rGO electrode, Supporting Information Figure S7a). A more detailed analysis may be obtained from in situ synchrotron powder di raction73 and spectroscopy studies similar to those demonstrated by Grey's group on LIBs.74,75 The tensile strength and strain to failure are important parameters for any exible battery electrode. Therefore, the rGO, 40MoS2, and 60MoS2 papers were subjected to static uniaxial tensile testing in a custombuilt setup (see Methods, Materials, and Instrumentation and Figure 5a). As can be seen, the specimen strip is secured on one end by a computer-controlled movable stage, while the other end is xed to a load cell, which in turn is xed to an immovable stage. Engineering stressÀstrain plots derived from loadÀdisplacementVOL. 8’NO. 2’1759 – 1770’2014 1766ARTICLETABLE 1. Summary of Tensile Test Data for rGO, 40MoS2,and 60MoS2 Free-Standing Composite Papersspecimen modulus (MPa) tensile strength (MPa) failure strain (%)rGO 40MoS260MoS2897.86 ( 18.86 874.62 ( 18.37 427.03 ( 8.97 434.5 ( 9.12 386.01 ( 8.11 450 ( 9.45 424.8 ( 8.92 130.8 ( 2.75 115.92 ( 2.43 102.76 ( 2.16 131.22 ( 2.76 125.85 ( 2.6412.57 ( 0.13 11.37 ( 0.12 9.48 ( 0.1 8.69 ( 0.09 7.45 ( 0.08 7.56 ( 0.08 6.06 ( 0.06 3.44 ( 0.04 2.84 ( 0.03 2.23 ( 0.02 2.69 ( 0.03 2.13 ( 0.021.4 ( 0.014 1.3 ( 0.014 2.22 ( 0.023 2 ( 0.021 1.93 ( 0.02 1.68 ( 0.017 1.43 ( 0.015 2.63 ( 0.028 2.45 ( 0.026 2.17 ( 0.023 2.05 ( 0.021 1.69 ( 0.018each that are shown in the photographic image in Figure 5c,d, respectively. Data are summarized in Table 1. Figure 5e,f is the corresponding SEM images of the fractured edge for 40MoS2 and 60MoS2, respectively. The edge was obs

erved to be more regular and smooth for 60MoS2 than for 40MoS2. The variation in strain to failure for the composite specimen is attributed partially to the likely inhomogeneity in the specimens (large size of the paper and higher loading 4 mg cmÀ2) and crumpled nature of the rGO layers. Subsequently, combined with observations in SEM images, 60MoS2 had even larger variation in failure strain as the more slippery MoS2 sheets can slide better than crumpled rGO sheets. CONCLUSION We have demonstrated synthesis of a compositelayered paper consisting of acid-exfoliated MoS2 nano akes in an rGO matrix. Mechanical tests involving static uniaxial tension reveal mechanical strength that was approximately 2À3 MPa and high failure strain (approximately 2%) in these materials. Further, the composite paper was directly utilized as counter electrode in Na-ion battery half-cell, and its performance was evaluated as a potential anode for use in a Na-ion battery full cell. These tests revealed high rst cycle electrochemical capacity of 338 mAh gÀ1 with respect to total weight of the electrode with excellent cyclability of Na ions. This study provides the rst experimental evidence of reversible electrochemical storage of Na in a layered self-standing MoS2 composite electrode at room temperatures and is expected to open new avenues for use of large-area free-standing binder-free exible electrodes for rechargeable battery applications.data are shown in Figure 5b. 40MoS2 showed higher fracture strength and modulus (approximately 7.8 and 424 MPa) than 60MoS2 (approximately 2.1 and 120 MPa) composite paper. On comparison, rGO had fracture strength and modulus of approximately 12 and 885 MPa, respectively. The tensile strength of rGO paper is comparable to those reported by Nyugen's group involving in situ reduced GO papers.76 These values are, however, much lower than GO papers, but this is hardly surprising considering that our papers were annealed at high temperatures (500 °C for 2 h and 900 °C for 5 min) and the mechanical strength of GO generally decreases with increasing annealing temperatures caused by release of oxygen groups that disturb the structure of the paper, resulting in a highly crumpled con guration.77 The strain to failure was higher in the case of the 60MoS2 specimen, reaching values in excess of approximately 2%. A total of ve specimens were tested for 40MoS2 and 60MoS2 paper,METHODS, MATERIALS, AND INSTRUMENTATIONThe ζ-potential surface measurements were carried out on a ZetaPlus ζ-potential analyzer (Brookhaven's Inst. Corp.). The e ect of ionic concentration on the potential measured is minimized by using a low concentration of basic (0.01 M NaOH) solution for controlling the pH. Scanning electron microscopy (SEM) of the synthesized material was carried out on a Carl Zeiss EVO MA10 system with incident voltage of 5 to 30 kV. TEM images were digitally acquired by use of a Phillips CM100 operated at 100 KV. Material c

haracterization was made using an X-ray di ractometer (XRD) operating at room temperature with nickel- ltered Cu KR radiation (λ = 1.5418 Å). Thermogravimetric analysis was performed using Shimadzu 50 TGA (limited to 800 °C). Samples weighing ～2.5 mg were heated in a platinum pan at a rate of 10 °C minÀ1 in air owing at 20 mL minÀ1. Raman spectra were measured using a LabRAM ARMIS Raman spectrometer using 633 nm laser excitation (laser power of 17 mW) as the light source. The surface chemical composition was studied by X-ray photoelectron spectroscopy (XPS, PHI Quantera SXM) using monochromatic Al KR radiation. Static uniaxial in-plane tensile tests were conducted in a simple test setup. The sample strip is secured on one end by a computer-controlled movable stage (M-111.2DG from PI), while the other end is xed to a 1N load cell (ULC-1N Interface), which in turn is xed to an immovable stage. All tensile tests wereconducted in controlled strain rate mode with a strain rate of 0.2% minÀ1. The samples were cut with a razor into rectangular strips of approximately 5 Â 15 mm2 for testing without further modi cation. Electrical conductivity measurements were carried out by use of a four-point probe setup and Keithley 2636A (Cleveland, OH) dual channel sourcemeter in the ohmic region. Electrochemical cycling of the assembled cells was carried out using multichannel battery test equipment (Arbin-BT2000, Austin, TX) at atmospheric conditions. Preparation of Graphene Oxide. Sodium nitrate (99.2%), potassium permanganate (99.4%), sulfuric acid (96.4%), hydrogen peroxide (31.3% solution in water), hydrochloric acid (30% solution in water), and methanol (99.9%) were purchased from Fisher Scientific. All materials were used as received without further purification. Modified Hummer's method was used to make graphene oxide.49 Concentrated H2SO4 (130 mL) was added to a mixture of graphite flakes (3 g) and NaNO3 (1.5 g). The mixture was cooled using an ice bath. KMnO4 was added slowly to this mixture. The mixture was stirred for 12 h at 50 °C. Then it was quenched with water (400 mL) with 30% H2O2 (3 mL) while in an ice bath such that the temperature does not go beyond 20 °C. The remaining material was then washed in succession with 200 mL of water twice, 200 mL of 30% HCl, and 200 mL of ethanol. The material remaining after these extended washes is coagulated with 200 mL of ether and filtered throughDAVID ET AL.VOL. 8’NO. 2’1759 – 1770’2014 1767ARTICLEa paper filter. The filtrate is dried overnight to obtain dry graphene oxide. Preparation of Exfoliated or Acid-Treated MoS2 Flakes. MoS2 powder (2 mg mLÀ1, 99%, Sigma Aldrich) was sonicated for 30 min in concentrated chlorosulfonic acid (superacid, 99%, Sigma Aldrich), and the non-exfoliated sheets were allowed to settle. Please note that the superacid was very slowly added to the MoS2 powder in an argon-filled glovebox (dew point À50 °C)

本文来源：https://www.bwwdw.com/article/be2j.html