Quantum Dots of 1T Phase Transitional Metal Dichalcogenides

Nov 29, 2017 - Characterization of the other three 2D MX2 QDs under a 0.001 A/g discharge current density. Panels (a–c), (d–f), and (g–i) are AF...
0 downloads 4 Views 3MB Size
Subscriber access provided by READING UNIV

Article

Quantum Dots of 1T Phase Transitional Metal Dichalcogenides Generated via Electrochemical Li Intercalation Wenshu Chen, Jiajun Gu, Qinglei Liu, Ruichun Luo, Lulu Yao, Boya Sun, Wang Zhang, Huilan Su, Bin Chen, Pan Liu, and Di Zhang ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.7b06364 • Publication Date (Web): 29 Nov 2017 Downloaded from http://pubs.acs.org on November 30, 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 Nano 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 23 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 Nano

Quantum Dots of 1T Phase Transitional Metal Dichalcogenides Generated via Electrochemical Li Intercalation Wenshu Chen,† Jiajun Gu,*,† Qinglei Liu,† Ruichun Luo,‡ Lulu Yao,† Boya Sun,† Wang Zhang,† Huilan Su,† Bin Chen,‡ Pan Liu,‡ and Di Zhang*,† †

State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai,

200240, China ‡

School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai,

200240, China *E-mail (J.-J. Gu): [email protected] *E-mail (D. Zhang): [email protected]

ABSTRACT: We prepare Group VI transitional metal dichalcogenides (TMDs, or MX2) from the 1T phase with quantum-sized and monolayer features via a quasi-full electrochemical process. The resulting two-dimensional (2D) MX2 (M=W, Mo; X=S, Se) quantum dots (QDs) are ca. 3.0-5.4 nm in size with a high 1T phase fraction of ca. 92%-97%. We attribute this to the high Li content intercalated in the 1T-MX2 lattice (mole ratio of Li:M is over 2:1), which is achieved by an increased lithiation driving force and a reduced electrochemical lithiation rate (0.001 A/g).

1

ACS Paragon Plus Environment

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

Page 2 of 23

The high Li content not only promotes the 2H→1T phase transition, but also generates significant inner stress that facilitates lattice breaking for MX2 crystals. Because of their high proportion of metallic 1T phase and sufficient active sites induced by the small lateral size, the 2D 1T-MoS2 QDs show excellent hydrogen evolution reactivity (with a typical ƞ10 of 92 mV, Tafel slope of 44 mV/dec, and J0 of 4.16×10-4 A/cm2). This electrochemical route towards 2D QDs might help boost the development of 2D materials in energy related areas.

KEYWORDS: 2D materials; transitional metal dichalcogenides; quantum dots; 1T phase; electrochemical Li intercalation

As a typical group of two dimensional (2D) materials, monolayer group VI transitional metal dichalcogenides (TMDs) have a broad range of applications in catalysis, batteries, supercapacitors, field effect transistors, etc.1-3 These materials (MX2, including MoS2, WS2, MoSe2, and WSe2, etc.) usually have two phases, i.e., stable 2H and meta-stable 1T phases, which are drastically different in properties.1, 4, 5 For example, compared with its stable and semiconducting 2H counterpart, meta-stable metallic 1T-MoS2 has better performance in electrochemical energy storage6/conversion7-11, surface-enhanced Raman spectroscopy,12 and photothermal applications,13 etc. This can be attributed to 1T-MoS2’s high electrical conductivity, high Fermi energy level, and dense active sites.6,

10-12

Since the quantum dots (QDs) of

monolayer (i.e., 2D) 2H-TMDs have special optical, electronic, and chemical properties14 due to

2

ACS Paragon Plus Environment

Page 3 of 23 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 Nano

some quantum restrictions in their in-plane dimensions and their large surface areas, it is reasonable to suppose that the 2D 1T-TMD QDs should possess different properties as well. However, fabrication methods for 2D TMD QDs with a high fraction of 1T phase is extremely lacking at present. To our knowledge, only a surfactant-assisted ultrasound exfoliation method15 has been reported for the fabrication of MoS2 and WS2 QDs with a high 1T phase fraction. Nevertheless, the Raman spectra of these QDs showed significant E2g1 and A1g signals, which are typical phonon modes for 2H-MoS2 and WS2.16 This indicates that the QDs reported therein contained a certain amount of 2H phase. The way to obtain 2D TMD QDs with a high 1T phase fraction actually lies in the syntheses of 2D TMD flakes. Li intercalation (chemically6,

17

or electrochemically18,

19

) into TMDs

followed by liquid exfoliation is one of the most common methods to prepare 2D TMD flakes.4, 20

For typical Li intercalation in 2H-MoS2, electrons transfer from a reducing agent (e.g.,

n-butyllithium (n-BuLi)) to the 2H-MoS2 host. This destabilizes the pristine 2H phase (trigonal prismatic coordination) and induces a phase transition from 2H to 1T (octahedral coordination).20-22 Most importantly, such a 2H→1T phase change can simultaneously cause certain defects and fragmentation in TMD grains.4,

20, 23-25

Using transmission electron

microscopy (TEM), Zeng et al. observed the fragmentation of MoS2 flakes in situ during electrochemical lithiation.23 In addition, Azhagurajan et al. reported that extensive intercalation of Li ions disrupted the atomically flat surface of MoS2 and some lithiated small domains were peeled away from the bulk surface.24 On the contrary, Fan et al. reported that by using a dilute

3

ACS Paragon Plus Environment

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

Page 4 of 23

n-BuLi solution for Li intercalation (mole ratio between n-BuLi and MoS2 was 0.1), they could just get large tri-layer 2H-MoS2 flakes with lateral dimensions on the micrometer range.25 These experimental results encourage us to use Li intercalation to simultaneously achieve a complete fragmentation of TMD flakes and a complete 2H→1T phase transition. However, previously reported experimental results suggest that such crystal fragmentation and phase transformation are affected by the intercalated Li amount in the MX2 (x value in LixMX2).24, 25 For example, the LixMoS2 crystal lattice is actually not an ideal 1T phase but offers some distorted 1T structure with the Li intercalation.26, 27 The degree of lattice distortion increases with the x value and the LixMoS2 correspondingly changes from a zig-zag to a diamond like structure.22, 28 Therefore, higher x values can produce more significant distortion in the MoS2 crystals. Because the increased Li content can accelerate the following liquid exfoliation treatment as well,29 there is a high chance of obtaining well fragmented MoS2 via more Li intercalation. Nevertheless, the x value in LixMoS2 is strongly limited by the intercalation methods. For a chemical Li intercalation reaction,4,

20

x is usually smaller than 1.21,

30-32

Even by using

solvothermal methods under enhanced pressures, one could only get x value of 1.12 for MoS2 (at 80 ºC)33 and 1.3 for WS2 (at 100 ºC)34. To further increase the x value (1.5-3.0), thermal evaporation of elementary Li on few-layer MoS2 flakes has to be conducted under 300 ºC in a high vacuum chamber.35 The difficulties in enhancing the x values make determining how the intercalated Li amount affects the fragmentation of the exfoliated MoS2 sheets difficult.

4

ACS Paragon Plus Environment

Page 5 of 23 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 Nano

Here, we demonstrate that the intrinsic reduction potentials for Li intercalation dominate the x values. We enhance the driving force for the Li intercalation via the electrochemical processes and decrease the intercalation rate to achieve quasi-full lithiation. This route yields much higher Li intercalation amounts. Under a discharging current density of 0.001 A/g, the 1.0 mole of the intercalated MX2 (M: Mo, W; X: S, Se) can approximately contain 2.0 moles of Li. The high Li content not only promotes the 2H→1T phase transition, but also generates significant inner stress that facilitates lattice breaking for MX2 crystals. After exfoliation treatment for these materials in water, we successfully prepared 2D MX2 QDs (ca. 3.0-5.4 nm in diameter) with a high 1T phase fraction (ca. 92%-97%). The prepared 2D MoS2 QDs show excellent hydrogen evolution reaction (HER) performance and HER stability. The method and the 2D 1T-MX2 QDs reported here are thus promising for exploring advanced applications in electronics, batteries, supercapacitors, solar cells, sensing, biomedicine, etc. RESULTS AND DISCUSSION The entire electrochemical lithiation for MoS2 can be divided into two steps.23, 24, 36 First, the 2H→1T phase transformation occurs above -2.4 V vs. reversible hydrogen electrode (RHE), which can be addressed as a reduction reaction of MoS2→[MoS2]n- (Figure 1a). This step generates LixMoS2, where x directly relates to the n value. Under deep lithiation, the LixMoS2 transforms into Li2S and Mo clusters (below -2.4 V vs. RHE, the light brown area in Figure 1 a).20, 37 Hence, an electrochemical lithiation process must be cut off above this transformation potential (-2.4 V vs. RHE) to avoid the decomposition of MoS2. 5

ACS Paragon Plus Environment

ACS Nano 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

Figure 1. Structural fragmentation of MoS2 induced by electrochemical lithiation. (a) Discharge curve and reaction paths during the electrochemical Li intercalation into MoS2. (b) Potential range of three redox couples involved in the chemical and electrochemical Li intercalation of MoS2.20 (c) x values in electrochemically intercalated LixMoS2 compounds obtained via ICP (in solid blue) and discharge curves (in dash blue), and the mean lateral size (in solid red) of exfoliated 2D MoS2 sheets determined via TEM observation with the different the discharge current densities. (d) The large discharge current density induces partial intercalation and results in large lateral size and non-uniform phase distribution of 2D MoS2 sheets. (e) The small discharge current density induces full intercalation and results in small lateral size and uniform phase distribution of the 2D MoS2 sheets. The main difference between chemical and electrochemical Li intercalation for MoS2 is the reduction potentials of the applied redox couples (Figure 1b).20, 38 For electrochemical lithiation (use elementary Li foils as anodes), Li+/Li gives a reduction potential of -3.0 V vs. RHE, which 6

ACS Paragon Plus Environment

Page 6 of 23

Page 7 of 23 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 Nano

is much lower than other chemicals (e.g., n-BuLi (LiC4H9) and lithium borohydrate (LiBH4)) usually used for chemical lithiation. The different reduction potentials for lithiation (E0(BH4-/B2H6), E0(C4H9-/C8H18), and E0(Li/Li+)) cause different initial Li intercalation dynamics (Figure 1b) and determine the final reaction products. When the lithiation reaction achieves equilibrium, the corresponding oxidation potential of MoS2/[MoS2]n- approaches the reduction potentials of the applied intercalation chemicals. Therefore, the reduction potential of the applied intercalation chemicals determines the reduction state of [MoS2]n- corresponding to the x values in the resulted LixMoS2 compounds. A higher driving force for electrochemical lithiation can thus theoretically intercalate higher amounts of Li. However, merely increasing the driving force for Li intercalation is insufficient. To achieve the theoretical Li amount for intercalation, it is necessary to reduce the internal polarization for lithiation. Therefore, similar to the way used to fully charge a Li battery, low electrochemical lithiation rates are preferred. Using the as-prepared 2H-MoS2 grown on carbon fiber paper (CFP) as a cathode (Figure S1-S4), we applied various galvanostatic discharge current densities for these Li-MoS2 batteries to study how the x values change with lithiation rates (Figure 1c). The x values acquired with inductively coupled plasma (ICP) analyses for the intercalated LixMoS2 increase from 1.41 to 1.96 with decreases in the discharge current density (from 1.0 A/g to 0.001 A/g). Moreover, the x values calculated from the discharge curves (Figure 1c and Figure S5) show similar trends. These confirm that a lower discharge current density helps complete the lithiation process. After being exfoliated in water, the MoS2 nanosheets have a decreased mean

7

ACS Paragon Plus Environment

ACS Nano 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

lateral size (from ca. 32.5 nm to 5.4 nm, Figure 1c) with the decrease in discharge current density (from 1.0 A/g to 0.001 A/g, TEM results, Figure S6). The mechanism by which higher x values help fragmentize MoS2 sheets is generalized in Figure 1d and e. Previous studies on the reaction pathway for electrochemical lithiation revealed that the intercalated MoS2 regions consist of Li-ion channels into which the newly inserted Li ions are pushed one-by-one.24 Given this, Li is inserted from the edge to the center part of pristine MoS2 particles. Under a larger discharge current density, the MoS2 is partially intercalated and only the peripheral part is well lithiated. This leads to a smaller x value in the resulting LixMoS2 (Figure 1d). The well lithiated part has more structure distortions and contains more Li content. Thus, it is easily broken in water due to the internal structural distortion and some exfoliation reactions (e.g., 2LixMoS2 + 2xH2O → xH2↑ + 2MoS2 + 2xLiOH29), while the poorly lithiated central part remains intact. This makes the resulting MoS2 sheets (lithiated under 1.0 A/g) show comparably large mean lateral size (ca. 32.5 nm) and a wide lateral size distribution (relative standard deviation (RSD): ca. ±11.9 nm, Figure 1c). As the discharge current density decreases, the Li atoms will be gradually inserted into the central parts of the MoS2. This causes increased x values for the resulted LixMoS2 (Figure 1e). The structure distortion region thus gets into the central part of MoS2. After being exfoliated, much more MoS2 sheets will split into small monolayer sheets, giving rise to the smaller mean size (ca. 5.4 nm) and more uniform size distribution (RSD: ca. ±1.3 nm) for MoS2 QDs (lithiated under 0.001 A/g).

8

ACS Paragon Plus Environment

Page 8 of 23

Page 9 of 23 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 Nano

Figure 2. Characterization of exfoliated MoS2 nanosheets. (a)-(d) AFM analyses of the exfoliated MoS2 nanosheets lithiated under the discharge current density of (a) 1.0 A/g, (b) 0.1 A/g, (c) 0.01 A/g, and (d) 0.001 A/g, respectively. (e) TEM and (f) HAADF-STEM image of the exfoliated 2D MoS2 QDs lithiated under 0.001 A/g. The enlarged lattice image in (e) is provided in Figure S7 and the right image in (f) is a deconvoluted result. (g) Intensity profile along the red dashed line indicated in (f). (h) Arrangement of atoms on the (001) plane of monolayer 1T-MoS2. Scale bars, (a)-(d) 200 nm; (e) 20 nm (top), 5 nm (bottom-left), and 1 nm (bottom-right); (f) 1 nm (left), 0.5 nm (right). Atomic force microscopy (AFM) analyses (Figure 2a-d) show that the resulting MoS2 sheets tend to show monolayer features (ca. 0.8 nm in height) and reduced mean lateral sizes (from 32.0 to 14.6 nm, Figure S8) with a decrease in the discharge current density (from 1.0 to 0.001 A/g). It should be noted that the mean lateral size obtained from AFM scans is larger than that from TEM observation (Figure 1c and Figure S6). These deviations are caused by the tip-size effect of AFM, which was also observed for 2D graphene ribbons39 and InAs QDs40. 9

ACS Paragon Plus Environment

ACS Nano 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

Since the lateral sizes of the resulting MoS2 sheets are ca. 4 to 6 times of magnitude larger than the thicknesses, we take these materials as “2D TMD QDs”. In addition, high-resolution transmission electron microscopy (HRTEM) (Figure 2e and Figure S7) images show that the MoS2 QDs fabricated via the full lithiation (under 0.001 A/g) plus exfoliation process are highly crystallized. To further identify the atom positions for these 2D MoS2 QDs, we conducted high resolution scanning TEM (STEM) in a high angle annular dark-field (HAADF) imaging mode (Figure 2f and Figure S9). The bright spots denote the positions of Mo atoms because of the larger atomic number of Mo than S. Both deconvoluted result (the right image in Figure 2f) and line intensity profile (Figure 2g) confirm that these 2D QDs are of monolayer 1T phase (Figure 2h). For most of the obtained 2D MoS2 QDs, we just observed the 1T structure (see Figure S9a-f). Nevertheless, we found the coexistence of H and T structure in a few areas (e.g., Figure S9g and h). It should be noted that we did not observe any 1T’ structures (distorted 1T structure with zig-zag Mo-Mo chains) in the resulting 2D MoS2 QDs, which were sometimes found in TMD nanosheets prepared via chemical Li-intercalation methods.26 These TEM and HAADF-STEM results strongly indicate that most of the obtained 2D MoS2 QDs are of highly crystalline 1T phase. Several macroscopic analyses also confirm that full lithiation under a smaller discharge current density generates higher 1T phase fraction. Figure 3a and Figure S10 show the X-ray Diffraction (XRD) results of the exfoliated MoS2 sheets. As the discharge current density decreases, the {001} plane of 1T-MoS241 gradually appears with an increase in its full width at half maximum (FWHM). Also, a broad peak of 1T-MoS2 (from 17º to 25º, shown between two 10

ACS Paragon Plus Environment

Page 10 of 23

Page 11 of 23 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 Nano

Figure 3. (a) XRD, (b) Raman, and (c) XPS results of the raw 2H-MoS2 grown on CFP and the exfoliated 2D MoS2 (0.001 A/g). dashed lines in Figure S10) occurs for samples treated under low current densities. These results suggest that the proportion of 1T-MoS2 in the products increases and the mean lateral size of MoS2 decreases with decreasing lithiation current density. The structural evolution also results in significant differences in the characteristic Raman features of the as prepared MoS2. In Figure 3b, the Raman shifts at 146, 196, 223, 281, 336, and 350 cm-1 of the exfoliated MoS2 QDs are especially associated with the phonon modes of 1T-MoS2.16,

42

As the discharge current density decreases, the intensity of these 1T peaks

increases while the intensity of two typical phonon modes (E2g1 and A1g) of 2H-MoS2 decreases (Figure S11), indicating increased 1T phase proportion. It should be noted that in most previous literatures,6, 8, 19, 43 the exfoliated nanosheets always contained the characteristic E2g1 and A1g 11

ACS Paragon Plus Environment

ACS Nano 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

peaks of 2H-MoS2, suggesting a multiphase nature of the MoS2 reported therein. In comparison, these two peaks can hardly be found for the MoS2 QDs (0.001A/g) reported here. This reveals that the 1T phase proportion is extremely high for these materials. Thus, we used X-ray photoelectron spectroscopy (XPS) to study the 1T-MoS2 fraction and the bonding states of the elements for the exfoliated nanosheets. The deconvolution results of the Mo 3d peaks are shown in Figure 3c and Figure S12, respectively. The Mo 3d spectrum of 2H-MoS2 consists of two peaks located at approximately 229.8 and 232.9 eV, which correspond to the 3d5/2 and 3d3/2 components, respectively. For 1T-MoS2, these two peaks shift to lower binding energies by approximately 0.9 eV.44 By calculating the area ratio of each state of Mo, the 1T phase content increases (from 4% to 94%) with the decrease in the discharge current density (from 1.0 to 0.001 A/g). The exfoliated MoS2 QDs (0.001 A/g) have a high 1T phase fraction (ca. 94%), which is higher than the MoS2 sheets prepared via chemical lithiation6, 17, 43 and other un-controlled electrochemical lithiation methods8, 19. To demonstrate the application of these materials, we measured the HER performance of the 1T-MoS2 QDs. Figure 4a shows the representative linear sweep voltammetry (LSV) curves of glass carbon (GC) electrodes, commercial 20 wt.% Pt/C catalysts, and exfoliated MoS2 nanosheets. The corresponding Tafel plots processed from the LSV data are provided in Figure 4b. Of these three electrodes, the GC electrodes show negligible HER activity, and the commercial Pt/C catalysts deliver the best HER overpotential (ƞ10) of 39 mV (vs. RHE) to achieve a geometric current density of 10 mA/cm2 and a Tafel slope of 30 mV/dec. For MoS2 electrodes with a decrease in the lithiation current density applied in the fabrication process, the 12

ACS Paragon Plus Environment

Page 12 of 23

Page 13 of 23 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 Nano

Figure 4. HER activity of the exfoliated MoS2 sheets. (a) LSV curves for the glass carbon electrode, commercial 20 wt.% Pt/C catalyst and as-exfoliated MoS2 nanosheets (1.0-0.001 A/g). (b) Corresponding Tafel plots of the LSV curves in (a). (c) Nyquist plots of the exfoliated MoS2 nanosheets (1.0-0.001 A/g). (d) LSV curves of the 0.001 A/g MoS2 QDs before and after 10000 cycles. (e) Chronopotentiometry curves of 0.001 A/g MoS2 QDs sheets and commercial 20 wt.% Pt/C catalyst under a HER current density of 200 mA/cm2 for 80 h. HER activity increases significantly. The ƞ10 of MoS2 electrodes (1.0-0.001A/g) decreases from 276 to 92 mV (vs. RHE), and the corresponding Tafel slopes decrease from 130 to 44 mV/dec.

13

ACS Paragon Plus Environment

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

The relatively low Tafel slope of 0.001 A/g MoS2 (44 mV/dec) suggests a two electron transfer process following a Volmer-Tafel mechanism of bimolecular adsorption and hydrogen evolution (with a theoretical Tafel slope of 45 mV/dec45). From the intercept of the linear region of Tafel plots, geometric exchange current densities (J0) for these exfoliated MoS2 sheets can be calculated (Table S1). The increased J0 values (from 6.97×10-5 to 4.16×10-4 A/cm2) indicate increased geometric active sites with a decrease in the lateral size and an increase in the 1T phase fraction. This is consistent with our analyses as well. We further conducted electrochemical impedance spectroscopy (EIS) experiments to study the HER kinetics of the MoS2 electrodes (Figure 4c). The equivalent circuit model is drawn in the inset of Figure 4c, and the calculated values of the series resistance (Rs) and charge transfer resistance (Rct) are given in Table S1. The Rct values dramatically decrease with the lithiation current density applied in the fabrication process. The LSV, Tafel slope, and EIS data show an enhanced HER activity with a decrease in the discharge current density. The HER performance of the 0.001 A/g MoS2 QDs with a typical ƞ10 of 92 mV, Tafel slope of 44 mV/dec, and J0 of 4.16×10-4 A/cm2 surpasses that of MoS2-based catalysts reported previously (Table S2). Such an excellent performance originates from the features of the monolayer, rather high 1T phase content (94%), and rather small lateral size (5.4 ± 1.3 nm) of the MoS2 QDs. These merits offer the catalysts excellent electron transfer abilities and high density of HER active sites. We also examined the HER stability of the 0.001 A/g MoS2 2D QDs. The HER activity slightly decreases after 10000 cycles (Figure 4d). Table S1 shows the comparison of some calculated HER parameters before and after 10000 cycles. The increased values of ƞ10, Tafel 14

ACS Paragon Plus Environment

Page 14 of 23

Page 15 of 23 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 Nano

Figure 5. Characterization of other three 2D MX2 QDs under a 0.001 A/g discharge current density. Panels (a-c), (d-f) and (g-i) are AFM images, TEM images, and XPS data of as-prepared MoSe2, WS2, and WSe2, respectively. Inset in the AFM images are the height distribution collected from AFM data. Scale bars: (a), (d) and (g) 200 nm; (b), (e) and (h) 20 nm, and 1 nm (inset). slope, Rct, and the correspondingly decreased J0 values reveal reduced charge transfer kinetics. We conducted XPS analyses for the 0.001 A/g MoS2 monolayer QDs before and after HER 15

ACS Paragon Plus Environment

ACS Nano 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

cycling (Figure S14). The results indicate that ca. 20% 1T-MoS2 converted to 2H-MoS2 after the cycling. Moreover, the intensity of Mo and S signals became relatively weak likely because some MoS2 monolayer QDs dissolved into the H2SO4 electrolyte. This can explain the decreased Rs values (from 4.8 to 3.2 Ω/cm2) after the cycling. The HER durability of the 0.001 A/g MoS2 monolayer QDs was also investigated via a long time constant current hydrogen generation test. The catalyst continually produces H2 under a high current density of 200 mA/cm2 for 80 h with only a ca. 5.2% increase in overpotential. This shows the excellent hydrogen evolution stability (Figure 4e). In comparison, although the commercial Pt/C catalyst showed a better initial HER activity than our MoS2 QDs, it quickly decreased (in 1 h) in properties and was much less active afterwards. We further successfully prepared 1T-MoSe2, 1T-WS2, and 1T-WSe2 QDs using the same methods described in this work (Figure 5). The pristine layered MX2 materials are hundreds of nanometers to several micrometers in lateral size (Figure S15). They were discharged under 0.001 A/g and the electrochemical processes were cut-off as soon as the Li:M ratio achieved 2.0 as calculated from the discharge time (Figure S16). After exfoliation, these materials show monolayer characteristics and very small lateral sizes (ca. 3.0 nm, see Figure 5b, e, and h). The XPS results (Figure 5c, f, and i) show that the 2D MX2 QDs have a high proportion of 1T phase. Determined with the deconvolution results of XPS data, the 1T ratios were ca. 92%, 97%, and 97% for 2D MoSe2, WS2, and WSe2 QDs, respectively. A comparison of the full XPS spectra between raw materials and exfoliated 2D MX2 QDs show that the resulting materials retain the elemental ingredients after the discharge processes (Figure S17). 16

ACS Paragon Plus Environment

Page 16 of 23

Page 17 of 23 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 Nano

CONCLUSIONS We have demonstrated that 2D MX2 QDs (M = Mo, W; X = S, Se) with a high 1T phase fraction can be successfully prepared via an electrochemical lithiation plus liquid exfoliation process. The electrochemical lithiation can insert sufficient Li atoms into MX2 lattice (the mole ratio of Li:M is as high as 2.0) by controlling the lithiation rate to be sufficiently low (0.001 A/g). The intercalated Li amount surpasses the commonly used chemical lithiation methods (usually x