High thermal-stable paramagnetism by Rolling

Electronic Supplementary Material (ESI) for Nanoscale. This journal is The Royal Society of Chemistry 2016 Electronic Supplementary information for High thermal-stable paramagnetism by Rolling up MoS 2 nanosheets Da Young Hwang, Kyoung Hwan Choi, Jeong Eon Park and Dong Hack Suh * Division of Chemical Engineering, College of Engineering, Hanyang University, Seoul, 04763, Republic of Korea * Corresponding author. E-mail: dhsuh@hanyang.ac.kr

Experimental Section 1. Materials MoS 2 powder (<2 μm, 99%, Product No. 234842), lithocholic acid (>97 %), and dichlorobenzene (anhydrous) were purchased from Sigma-Aldrich. All chemical reagents were of analytical grades and were obtained commercially. 2. Preparation of the exfoliated Molybdenum disulfide solution The dispersed MoS 2 was prepared according to the previous work 1 ; MoS 2 (1.5g) was added to 300mL dichlorobenzene. The mixture was homogenized at 2400 rpm for 1 hr (IKA-RW20, Stage II), followed by cuphorn sonication for 30 mins (Power%= 50 %, Pulse = 20 sec. (On), 10 sec (Off)). Finally, the slurry was centrifuged at 4400 rpm for 30 minutes. The supernatant was obtained in ODCB. 3. Synthesis of starting material (N-(2-aminoethyl)-3α-hydroxy-5β-cholan-24-amide (LCA)) LCA was synthesized by the reaction of methyl lithocholate with excess of diaminoenthane according to the previous work 2 ; The methyl lithocholate was dissolved in methanol and excess of diaminoethane (20-30 times) was added to the methyl lithocholate solution. The solution was heated with an oil bath (353-363 K) for 2 days. The resulting solution was poured into the water and the precipitate filtered. The product was recrystallized from acetonitrile and dried in a vacuum. 4. Synthesis of 1T@2H Molybdenum disulfide nanoscrolls (1T@2H MNSs)

1T@2H MoS 2 nanoscrolls were synthesized by solution process. 3,4 LCA of 0.02mmol was dissolved in ODCB (1 ml) and heated to 333 K. Then, the dissolved LCA solution was immediately poured into the exfoliated MoS 2 solution (5 ml). The solution with the precipitates was stored at room temperature for 24 hrs. The resulting precipitate was obtained by filtering with a PTFE membrane washed with methanol by centrifugation at 4000 rpm for 20 minutes and followed by decantation. This procedure was at least repeated five times, affording LCA induced MoS 2 nitride nanoscrolls including 1T and 2H phases (1T@2H MNSs). 5. Synthesis of 1T@2H Molybdenum disulfide nanoscrolls with maximum 1T phase (1Tmax.@2H MNSs) 1T@2H MoS 2 nanoscrolls with maximum 1T phase were synthesized by thermal treatment. Thermal treatment was conducted on a hot plate under an argon flow of 100 ml/min. Samples were annealed at a desired temperature for 1 h. 1Tmax.@2H MNSs were obtained at 473K for 1h, followed by XPS measurements. 6. Characterization A drop of the purified sample by a microsyringe was placed on a holey carbon grid and then was allowed to evaporate methanol. HR-TEM measurements were carried out on JEM-2100F with an accelerating voltage of 200 kv. Micro-Raman spectrometer (JASCRO NRS-3100) analysis was carried out with an excitation laser of 532nm wavelength using the 100 X objective. Samples were prepared by drop casting of the dispersion on SiO 2 /Si substrates and the solvent was evaporated at RT. X-ray Photoelectron Spectroscopy measurements were performed in type Theta probe base system (Thermo Fisher Scientific Co.). XPS Peak 4.1

was employed to deconvolve the S 2p and Mo 3d peaks using the Tougaard-type baseline and an iterative least-squared optimization algorithm. Photoluminescence spectroscopy were carried out on HORIBA (ARAMIS; 532 nm laser excitation wavelength). Magnetization studies were carried out using a superconducting quantum interference device (SQUID) magnetometer (T < 300 K). CW X-band EPR (Electronic paramagnetic resonance) spectra were collected on a Bruker EMX plus 6/1 spectrometer equipped with an Oxford Instrument ESR900 liquid He cryostat using an Oxford ITC 503 temperature controller. All spectra were collected with the following experimental parameters: microwave frequency, 9.6 GHz; microwave power, 3 mw; modulation amplitude, 3 G; temperature, 298 K. Electrochemical measurements were performed using a standard three-electrode electrochemical configuration in 0.5 M H 2 SO 4 electrolyte measured by Reference 600TM Potentiostat/Galvanostat/ZRA (GARMY). All measurements were performed in 20 ml of 0.5 M H 2 SO 4 (aq) using MoS 2 derivative samples on glassy carbon electrode (GC) as central disk (3.0-mm diameter, 0.07069 cm 2 area) as the working electrode, a platinum electrode as a counter electrode, and a saturated calomel (SCE) reference electrode (CH Instruments). The electrodes for HER measurements for the MoS 2 derivative samples were prepared by simply drop casting the as-exfoliated solution onto the glass carbon electrodes. For electrochemical measurement, MoS 2 derivatives were prepared by mixing 8 mg of catalyst with 100 μl of Nafion (5 wt% in isopropanol, Sigma-Aldrich) in a solution of 800 μl DI water, 200 μl of EtOH (99.9%), and the mixture was sonicated for 30 min to produce homogeneous slurry. Afterwards, the solution of MoS 2 derivatives (5 μl) was dropped onto the glassy carbon electrode, and dried at RT for 30 min. The resulting MoS 2 derivatives loading on GC was 300 μg cm -2. The reversible hydrogen electrode was calibrated using platinum as both working and counter electrodes to + 0.241 V vs. the SCE reference. The performance of the hydrogen

evolution catalyst is measured using linear sweep voltammetry beginning at + 0.0 V and ending at 0.8 V vs. RHE with a scan rate of 20 mv/s.

Figure S1. TEM images of the dispersed MoS 2 sheets. The concentration of the final exfoliated MoS 2 was measured by using the vacuum filter through a weighed membrane. The multi-layer MoS 2 with size of typically several hundred nanometers is observed.

Figure S2. TEM images of 1T@2H MNSs at room temperature with a low magnification. The tube-like morphology scrolled by interaction of MoS 2 with LCAs. The TEM images at a low magnification represent the produced 1T@2H MNSs.

Figure S3. Collected TEM images 1T@2H MNSs from several samples with a diameter from 40 to 100 nm.

Figure S4. Calculation of the strain energy for 1T@2H MNSs. We now discuss that the observed behavior can be explained by considering the 1T@2H MNSs induced uniaxial strain on Raman modes responsible for E 1 2g and A 1g, correspondingly. The uniaxial strain-induced peak splitting for the E 1 2g mode enables us to calculate parameters: the Grüneisen parameter, γ, and the shear deformation potential, β, and the Δω solution of the secular equation for the E 2g mode E2g, 5 where Δω is the change of frequency in the Raman mode, ω 0 E 2g is the E 2g peak position at zero strain, ν is Poisson s ratio and ε is the induced uniaxial tensile strain. The Grüneisen parameter of 1.1 and the shear deformation potential of 0.78 for both monolayer and bilayer MoS 2 attaching to a substrate (ν = 0.33) are reported. 6 In case of formation of 1T@2H MNSs, these calculated parameter is used to estimate the strain applied to free-standing multilayer MoS 2. To insert ω 0 E 2g = 380.9 cm -1 from Raman spectra, γ E2g = 1.1, β E2g = 0.78 and Poisson s ω ratio ν = 0.125 for free-standing MoS 2 in Equation 3, 7 we can obtain the E 2g for E2g + / ε ~ 2.0 cm 1 /% strain. These results are almost equal to the linear relationship with a slope 1.7 cm 1 / % strain obtained by applying uniform tensile strain to few-layer MoS 2. 8 Combining the above results, 1T@2H MNSs formed through the rolling 2H MoS 2 sheets can sustain the high local strain greater than ~ 1.7 % compared with that of MoS 2 sheets, which represents

the reliable results by virtue of the layer-by-layer Vander Waals interaction energy of overlapping (stability). Figure S5. Summary the induced strain in scrolls obtained from the frequency shift of E 1 2g mode for 1T@2H MNSs. To estimate the applied strains for the temperature dependent 1T@2H MNSs, we first consider the available total frequency shift data of individual components of the system as a ω 0 E function of temperature based on standard peak position at 2g = 380.9 cm -1 and ω 0 A1g = 406.44, respectively. As seen from the Figure 3b, both E 1 2g and A 1g modes follow a systematic red-and blue shifts with increase in temperature. The frequency red-shifts of both peaks increase in the range of -3.44, -5.84 and -6.30 cm -1 with increasing temperature at 473K, which converted to the induced uniaxial tensile strain resulting in that of ~ 1.7, 2.9 and

3.2 % for the temperature at 298, 373 and 473K, respectively (which follows a linear relationship with a slope 2.0 cm -1 / % strain). We suggest that the different Raman frequency red-shift on temperature dependent peak shifts of E 1 2g and A 1g modes can be discussed in the thermal expansion coefficient. When temperature rises further, 1T@2H MNSs may slip on the layer-by-layer S-plane because the tensile strain significantly increases. We interpret that the maximum rolling effect occurred in 1T@2H MNSs at temperatures in range of 373 to 473 K. Furthermore, we found the decrease of frequency red-shift of the E 2g and A 1g modes in 1T@2H MNSs at temperatures over 473 K. The metastable 1T phase can be maintained due to van der Waals interactions between layers of the nanoscrolls, which are relatively stronger than the bending strain that contributes to the instability of the phase structure. In a temperature range between 473 to 573 K, the 1T to 2H phase relaxation occurs spontaneously due to the thermally activated atom displacement with a 1T phase fraction about ~26 % at 773K. Therefore, we suggest that the induced strain at 473K for 1T@2H MNSs gradually release leading to phase transition from 1T to 2H relaxation, which derives the decrease of the strain from 3.2 to 2.5 % with the increasing temperature. Although the bending strain assisted by van der Waals force is remained, the 1T 2H phase relaxation is occurred by supplying the thermal activation energy that was required for atomic displacement around 353 K for 1T MoS 2 sheets.

Figure S6. Raman spectra of the MoS 2 sheets located at 380.9 and 406.4 cm 1 with FWHMs of 4.6 and 5.0 cm 1, respectively. The corresponding bands for MoS 2 sheets are located at 380.9 and 406.4 cm 1 with FWHMs of 4.6 and 5.0 cm 1, respectively. The E 1 2g mode involves in-plane displacement and shear force of Mo and S atoms, whereas the A 1g mode involves out-of-plane symmetric displacement and compressive force of S atoms along the c axis.

Table S1. Elemental Analyses of the 2H MoS 2 sheets and 1T@2H MNSs using XPS. Samples Atomic Ratio of Mo/S 2H MoS 2 sheets at 298K 1 : 2.16 1T@2H MoS 2 scrolls at 298K 1 : 2.37 1T@2H MoS 2 scrolls at 373K 1 : 2.42 1T@2H MoS 2 scrolls at 473K 1 : 2.45 1T@2H MoS 2 scrolls at 573K 1 : 2.32 1T@2H MoS 2 scrolls at 673K 1 : 2.19

Figure S7. Magnetization vs. magnetic field curve of 2H MoS 2 nanosheets.

Figure S8. Temperature dependent Magnetization vs. magnetic field curve of 1T@2H MNSs at 4K.

Figure S9. EPR spectra with 1T MoS 2 sheets.

Figure S10. Electrochemical effects of the 2H MoS 2 nanosheets and their nanoscrolls on the HER activity. Current density at -0.3V vs RHE for 2H MoS 2 (blue curve) sheets and 1T@2H MNSs (red curve) and comparison of Tafel plots from the polarization curves.

References 1 Christopher E. Hamilton, Jay R. Lomeda, Zhengzong Sun & James M. Tour. High- Yield Organic Dispersions of Unfunctionalized Graphene. Nano letters 9, 3460-3462 (2009). 2 Löfman, M., Koivukorpi, J., Noponen, V., Salo, H. & Sievänen, E. Bile acid alkylamide derivatives as low molecular weight organogelators: Systematic gelation studies and qualitative structural analysis of the systems. Journal of Colloid and Interface Science 360, 633-644, doi:http://dx.doi.org/10.1016/j.jcis.2011.04.112 (2011). 3 Hwang, D. Y. & Suh, D. H. Formation of hexagonal boron nitride nanoscrolls induced by inclusion and exclusion of self-assembling molecules in solution process. Nanoscale 6, 5686-5690, doi:10.1039/c4nr00897a (2014). 4 Hwang, D. Y., Yook, J. Y. & Suh, D. H. Inclusion and exclusion of self-assembled molecules inside graphene scrolls and control of their inner-tube diameter. RSC Advances 4, 35943-35949, doi:10.1039/c4ra04556d (2014). 5 Mohiuddin, T. M. G. et al. Uniaxial strain in graphene by Raman spectroscopy: G peak splitting, Gruneisen parameters, and sample orientation. Physical Review B 79, 205433 (2009). 6 Conley, H. J. et al. Bandgap Engineering of Strained Monolayer and Bilayer MoS2. Nano Letters 13, 3626-3630, doi:10.1021/nl4014748 (2013). 7 Castellanos-Gomez, A. et al. Local Strain Engineering in Atomically Thin MoS2. Nano Letters 13, 5361-5366, doi:10.1021/nl402875m (2013). 8 Rice, C. et al. Raman-scattering measurements and first-principles calculations of strain-induced phonon shifts in monolayer MoS2. Physical Review B 87, 081307

(2013).