- Email: [email protected]
High yield production of photoluminescent tungsten disulphide nanoparticles
Journal of Colloid and Interface Science 396 (2013) 160–164
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
High yield production of photoluminescent tungsten disulphide nanoparticles Shannon M. Notley Faculty of Life and Social Sciences, Swinburne University of Technology, Hawthorn 3122, VIC, Australia
a r t i c l e
i n f o
Article history: Received 27 November 2012 Accepted 15 January 2013 Available online 7 February 2013 Keywords: WS2 Raman Nanoparticle Photoluminescence Steric stabilisation
a b s t r a c t Single and few layered tungsten disulphide (WS2) nanoparticles were prepared using a surfactant assisted ultrasonication exfoliation technique with concentrations of up to 0.4 mg/mL. The lateral dimension of the particles was in the range of 100–250 nm. The exfoliated WS2 was stabilised against re-aggregation through adsorption of a tri-block non-ionic polymeric surfactant (PEO–PPO–PEO). These nanoparticles were characterised by absorption, Raman and photoluminescence spectroscopy (PL). Broadening of the E2g peak in the Raman spectrum was observed due to phonon confinement within a single layer of WS2. The exfoliated particles have significantly different properties than the bulk WS2 material, in particular, the emergence of strong photoluminescence at 1.97 eV in energy coincidental with the excitonic peak in the UV–Vis spectrum. The emergent PL emission suggests that the monolayer WS2 is a direct gap material analogous to other dichalcogenides such as MoS2. Ó 2013 Elsevier Inc. All rights reserved.
1. Introduction Atomically thin materials have received great attention, the most commonly cited example being graphene [1–3]. Other van der Waals bonded solids though, such as the dichalcogenides, also have interesting optical and electronic properties particularly in the limit of 2D [4]. Chiefly among these, the transition metal dichalcogenide semiconductors molybdenite (MoS2) and tungstenite (WS2) have attracted interest for potential use in advanced applications such as photovoltaics and photocatalysis due to tunable band gaps depending on size. There is a strong push from the renewable energy interests to create materials with appropriate valence and conduction bands which allow the splitting of water to form hydrogen and oxygen through the utilisation of the maximum amount of incident solar radiation. Typically, TiO2 has been used in such applications however the large band gap of 3.2 eV means less than 5% of the incident solar photons can be absorbed. Both MoS2 and WS2 in the bulk form are indirect band gap semi-conducting materials with band gaps of 1.22 eV and 1.20 eV respectively. The magnitude of this gap is insufficient for many photocatalytic applications, specifically the photoelectrochemical splitting of water [5,6]. Thermodynamically, the minimum energy required is 1.23 eV but practical considerations means a band gap greater than 1.4 eV is needed. Furthermore, the positioning of the valence and conduction bands is such that the generation of hydroxyl radicals is also not possible reducing the potential for these materials to be used in many photocatalytic applications. Hence,
E-mail address: [email protected] 0021-9797/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcis.2013.01.035
much effort has been taken in tuning the band gap properties of the dichalcogenides to more suitable values through quantum confinement effects [7]. 3D confinement has led to increased band gaps of up to 2.5 eV for crystallite sizes of 4–5 nm for MoS2. Precipitation inside micelles has typically been used in order to produce these nanoparticles [8–10] and analogous WS2 particles [11]. Recent studies have shown that confinement effects are also demonstrated by the 2D analogues of MoS2 with emerging photoluminescence suggesting a transition to a direct band gap material [12–14]. Thus, it is timely that the methods used for generating large scale quantities of graphene be adapted for producing single and few layered materials of the transition metal dichalcogenides. Many methods have been described for the production of 2D materials such as graphene [15]. Whilst large area graphene was first produced using exfoliation with adhesive tape [1,16,17], subsequent methods such as intercalation, reduction of graphene oxide [18], solvothermal techniques [19] and sonication in appropriate solvent systems [20–28] have all led to increased quantities of material. Aqueous solution processing has many advantages, particularly in terms of cost and the environment. One of the more promising techniques is the use of ultrasonic exfoliation of graphene and related materials in the presence of aqueous surfactant solutions [21,23,29]. The surfactant is necessary to promote exfoliation for two reasons. The first, it reduces the surface tension of the aqueous phase to a magnitude of the order of the cohesive energy required to separate the solid sheets. Secondly, it adsorbs onto the surface of the exfoliated particles creating an extra repulsive term that inhibits re-aggregation of the produced particles [30,31]. This surfactant assisted ultrasonication technique was recently adapted from the graphene system to other van der Waals bonded solids [32], including the semi-conducting dichalcogenides MoS2 and
S.M. Notley / Journal of Colloid and Interface Science 396 (2013) 160–164
WS2. In this paper, a method for the high yield production of photoluminescent WS2 nanoparticles under aqueous conditions is described. Surfactant is added continuously throughout the sonication thereby giving rise to a much greater concentration of exfoliated WS2 particles. These particles have been characterised in terms of their size and their opto-electronic properties, with interesting photoluminescence observed.
2. Experimental section 2.1. Materials Powdered Tungsten(IV) Sulphide with a bulk particle size of 2 lm was purchased from Sigma–Aldrich Australia and used as received. A tri-block co-polymeric surfactant, Pluronic F108 (BASF, Ludwigshafen) with a molecular weight of 14.6 kDa was used to stabilise the exfoliated particles as well as control the liquid–vapour interfacial tension. Milli-Q water was used in the preparation of all suspensions.
161
2.2.3. UV–Vis spectrophotometry UV–Vis spectra of the diluted WS2 suspensions were measured using a Shimadzu UV-3101PC spectrophotometer with matched quartz cuvettes over the wavelength range of 250–800 nm. 2.2.4. Raman spectroscopy and photoluminescence spectra The synthesised WS2 particles were characterised using Raman spectroscopy with a Horiba John Yvon T64000 Raman system. Laser excitation at 532 nm was used. Typically, WS2 particles were deposited onto an oxidised silicon wafer. The photoluminescence spectrum of the particles was also measured using the laser excitation at 532 nm. 2.2.5. Zeta potential measurements The zeta potential of the WS2 particles was determined using a Malvern Zetasizer Nano. The measurements were performed on the suspension in Milli-Q water (pH 6.5) after extensive dialysis to remove any unbound non-ionic surfactant. 3. Results and discussion
2.2. Methods 2.2.1. Production of aqueous suspensions of WS2 particles WS2 nanoparticle suspensions were prepared using the surfactant assisted ultrasonic exfoliation technique as previously described [21,23,27,29,31–34]. A 2% w/w suspension of bulk WS2 was added to Milli-Q water. Under continuous sonication at 60 W using a ‘‘Cell Disruptor’’ Model W-220F (Heat Systems-Ultrasonics Inc) sonicator, non-ionic surfactant was added from a highly concentrated solution to maintain the concentration of surfactant in water at approximately 0.1% w/w. The exfoliation procedure results in a vast increase in solid–liquid surface area which leads to rapid depletion of the surfactant through adsorption to the particles surface. Thus, continuously adding surfactant during sonication leads to a significantly greater yield of nanoparticulate WS2 [29]. This concentration of surfactant is suitable for maintaining the optimum surface tension as described previously at 40– 42 mJ/m2 [32]. The resultant exfoliated suspension was then dialysed in MilliQ water for 48 h to remove any non-adsorbed surfactant from the solution. The suspension was then typically centrifuged at 1500 rpm for 5 min to sediment further any large (non-exfoliated) particles. The supernatant was collected and was found to be stable for up to 3 months. Some sedimentation was observed due to the relatively high density of WS2 (7.5 g/mL) but the particles could be readily re-dispersed through simple agitation. The particles were subsequently characterised in terms of size, charge and spectroscopic properties as described below. The yield of single and few layer WS2 nanoparticles produced using this method was determined gravimetrically to be as high as 0.4 mg/mL depending on the rate of centrifugation. Furthermore, density measurements of the suspensions were also performed to confirm the concentration of WS2 particles. The thickness of the WS2 particles was typically characterised based upon the photoluminescence measurements. That is, the proportion of particles which demonstrated strong luminescence at 1.97 eV, with at least 100 particles measured. 2.2.2. Transmission Electron Microscopy (TEM) imaging The dried suspension of WS2 particles was imaged using TEM to confirm that single and few layered material was prepared as well as to determine the lateral sizes of the particles. A Hitachi H7100FA transmission electron microscope with an accelerating voltage of 125 kV was used. Furthermore, electron diffraction patterns of the particles were measured.
The surfactant assisted sonication method was used to exfoliate single and few layer WS2 from the bulk particles in the presence of the non-ionic tri-block surfactant. This technique involves a relatively high powered sonicator probe however little oxidation was observed as determined through the measurement of the nanoparticle zeta potential. Typically, the f potential of the WS2 at pH 6.5 was less than 5 mV. This particle charge is significantly lower than the graphene analogues however the dangling edges of the WS2 are stabilised by sulphur atoms [35] leading to reduced overall charge in comparison to the graphene which is predominately stabilised through oxygen containing moieties. This overall surface charge is typically too low to stabilise the particles in suspension. However, the non-ionic surfactant adsorbs to the surface allowing stability of the WS2 for in excess of 3 months. Fig. 1 shows an image of the nanoparticlulate WS2 suspension, which has a slight green/brown1 colour. The UV–Vis spectrum of a 100 diluted suspension of WS2 nanoparticles was measured in the range of 250–800 nm. As can be seen in Fig. 1, the particles absorb strongly in the UV region and furthermore, a peak is observed at 627 nm (1.97 eV), commonly assigned as the ‘‘A exciton’’ and related to the direct excitonic transition at the K point of the Brillouin zone [13]. This peak at 627 nm is also present in the bulk material which demonstrates that there is no a significant shift in the spectrum as may be expected from simple two dimensional confinement. Wilcoxon et al. have demonstrated that changes in the adsorption edge only occur for confinement in three dimensions [8]. Furthermore, the spectrum shows strong absorption at wavelengths below 627 nm, particularly in the UV region in agreement with previous spectrophotometric studies [7]. Confinement in 3D leads to a shift in the absorption edge and decreases in the position of the exciton. It is clear however that although the size of the WS2 particles are confined in terms of thickness, the relatively large lateral dimensions leads to maintenance of the position of the excitonic peak at 627 nm. Previous studies suggest that there is only a very small shift (0.15 eV) due to confinement in 2D [8]. The peak associated with the indirect B excitonic transition is not as prominent as for the isomorphous MoS2 analogues in either the bulk or exfoliated samples shown in Fig. 1. The difference in energies of the A and B excitons peaks upon exfoliation is not expected to change significantly under these 2D confinement conditions unlike in 3D con1 For interpretation of color in Fig. 1, the reader is referred to the web version of this article.
162
S.M. Notley / Journal of Colloid and Interface Science 396 (2013) 160–164
Fig. 1. UV–Vis spectrum of the 100 diluted WS2 nanoparticulate suspension (inset) prepared using the surfactant assisted sonication exfoliation technique compared with the bulk WS2 starting material. Position of the ‘‘A’’ exciton highlighted.
finement where the spin–orbit splitting can be heavily influenced by the small lateral sizes [7,8]. The exfoliated WS2 nanoparticles were imaged using TEM to determine lateral sizes as well as to demonstrate that the thicknesses of the particles were of the order required for transmission of the electrons. It was not possible to directly determine the number of layers from TEM measurements. Fig. 2 shows a typical image of the particles. Most particles had lateral sizes in the 100–250 nm range. The proportion of single layer WS2 (determined from PL measurements) was dependent upon the sedimentation or centrifugation step in the preparation. Typically, faster centrifugation resulted in much smaller particles, particularly in terms of thickness, and a resulting reduction in concentration in suspension as shown in Fig. 2. This is due to the relative high density of the particles in comparison to water. As centrifugation separates on mass, it is likely that particles of small lateral dimensions but with multiple layers are sedimented but larger single layer particles would also be separated. However, a high proportion of single layer material could be achieved in this way and although sedimentation was observed over time, the particles could be easily redispersed through
simple agitation. The electron diffraction pattern of single layer WS2 nanoparticles was also determined and is shown in Fig. 2. Clearly, the typical hexagonal pattern as expected from the trigonal bipyramidal spatial arrangement is demonstrated. Raman spectroscopy has been demonstrated to be a valuable tool in research into 2D materials as shifts in peak positions and shapes give much information on the structure of the material, particularly the number of layers and also the presence of any defects [16]. Fig. 3 shows Raman spectra for the bulk WS2 powder as well as the exfoliated material deposited onto a silicon wafer. The silicon peak at 520 cm 1 was used to calibrate the absolute peak positions for both samples (Si peak not shown). Two peaks, at 347 cm 1 and 410 cm 1 in the bulk sample are shown in Fig. 3 and these are assigned to E2g and A1g respectively. The spectra demonstrate that there is a significant blue shift of the two peaks in region of 340–420 cm 1 upon exfoliation to single layer material. This shift in the peaks could potentially be due to stress induced in the lateral plane through drying onto the substrate and has been observed previously in studies involving WS2 nanotubes [36]. However the reduction of the interlayer interaction for the exfoliated particles may also lead to the observed shift in these peaks. Furthermore, the intensity of the peak at 350 cm 1 for the exfoliated sample, is much higher and broader. The broadening of this Raman band is due to phonon confinement within the single layer and demonstrates that the lateral dimensions of the particle are in the nanometre range [4,37]. It is interesting to note though that both E2g and A1g peaks are blue shifted which is in direct contrast to previous experimental and theoretical studies on the analogous MoS2 material which show a decrease in A1g frequency and increase in E2g [12,38]. The A1g is the out of plane phonon mode and is therefore almost exclusively related to the sulphur atoms. Exfoliation to a single layer hence removes the interactions with neighbouring interlayer sulphur atoms which leads to a decrease in the frequency. However, the shift in the E2g peak is not as straight forward to explain. The E2g peak is the in-plane bending phonon mode and it can be argued as above that the weak interlayer attraction dominates. This is contrasted though by the ab initio calculations by Molina-Sanchez and Wirtz where they found that longer range Coulombic interactions is most influential, a situation supported by experimental determinations of peaks in MoS2. It is possible though that the heavier W
Fig. 2. (A) Yield of WS2 particles (closed circles) and proportion of particles showing photoluminescence (open squares) as a function of centrifugation rate for a set time period of 5 min. (B) Schematic of spatial arrangement of atoms in crystalline WS2. (C) TEM image of single and few layer WS2 nanoparticles. (D) Electron diffraction pattern of WS2 nanoparticles.
S.M. Notley / Journal of Colloid and Interface Science 396 (2013) 160–164
Fig. 3. Raman spectra of bulk WS2 (bottom) and exfoliated WS2 nanoparticles (top) deposited onto silicon wafer. Absolute peak positions were calibrated against the silicon peak at 540 cm 1, the exfoliated material results in a blue shift of the peaks assigned as E2g and A1g. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
atoms provide the impetus for a greater interlayer attraction giving rise to the observed increase in frequency as shown in Fig. 3. Recent studies have shown that single and few layer MoS2 particles demonstrate photoluminescence, the intensity of which increases with decreasing thickness of the particles [13,14]. Indeed, even a double layer can see a significant drop in intensity as well as a shift to lower energy of the primary peak. The PL spectrum of the stabilised WS2 particles produced using the surfactant assisted exfoliation technique outlined above was measured here and is shown in Fig. 4. The proportion of particles which displayed this strong luminescence was used in order to determine the relative efficiency of the exfoliation process. In comparison to the bulk WS2 which shows no observable PL emission, a strong peak of about 40 meV width is observed for the exfoliated WS2 particles. The position of this PL peak at the photon energy of 1.97 eV is coincident with the A1 exciton peak in the UV–Vis spectrum of WS2 as also shown in Fig. 4 (data from Fig 1 converted from wavelength to energy). Of the order of 63% of particles surveyed using the PL measurements after excitation at 532 nm demonstrated this strong and sharp peak at 1.97 eV. It is unusual for indirect band gap semiconductors to show appreciable photoluminescence although similar properties have been observed for the analogous transition metal dichalcogenide, MoS2 [13]. Here, two peaks corresponding to the positions of the
163
A1 and B1 excitons were observed yet in another study, only one strong emission peak was resolved at the A1 energy in the limit of a single MoS2 layer [14]. These studies suggested that the photoluminescence of MoS2 resulted from the transition to a direct band gap semiconducting material with decreasing thickness. Evidence from ab initio calculations supported this result [39]. That the position and width of the PL peak, in terms of energy, corresponds with the exciton peak measured using UV–Vis spectroscopy provides further compelling evidence beyond the fact that photoluminescence was indeed observed. Similarly, the results in Fig. 4 demonstrate that the analogous material, WS2, shows the same behaviour. That is, a transition from an indirect to direct band gap semiconductor in the limit of 2D and that only a single PL peak is observed. Typically the surfactant assisted exfoliation technique produces a distribution of single and few layered van der Waals bonded materials. WS2 nanoparticles should photoluminesce, albeit with reduced intensity, for exfoliated particles with more than a single layer. This means that this production method for the WS2 is capable of giving a large quantity of particles suitable for use in photocatalytic applications as the band gap of the dichalcogenide has been increased from 1.20 eV up to 1.97 eV. Importantly, the magnitude of the band gap and the positioning of the valence and conduction bands give rise to the potential use of these particles in the production of hydrogen in a photoelectrochemical cells through the splitting of water. Whilst the magnitude of the bandgap is not of the same order as for nanocrystals of the transition metal dichalcogenides, the ability to tune the band gap through 2D rather than 3D confinement is nevertheless advantageous. This is particularly important when the availability of solar electromagnetic radiation is considered. The bandgap of WS2 observed here means approximately 40% of incident solar photons possess the energy for absorption and electron–hole separation, in comparison to TiO2 where less than 4% of photons can be absorbed. Stable aqueous suspensions of semiconducting nanoparticles with appropriate band gaps have been promoted as an alternate means for generating hydrogen rather than the use of electrodes [6]. Many examples of semi-conductor catalysts use visible light to drive photocatalytic reactions. Quantum dots such as CdSe and CdS have bandgaps of 1.7 eV and 2.4 eV respectively which makes them highly suited for this application, particularly as they have high quantum efficiencies. However, many of the materials currently described are far from stable and pose potential environmental problems upon degradation which limits their use in colloidal bed type reactors for hydrogen productions. WS2 and the analogue MoS2 are far more stable against degradation and
Fig. 4. Left: PL spectrum of bulk WS2 (black) and exfoliated WS2 (red) under laser excitation at 532 nm. Right: PL spectrum and UV–Vis spectrum of the exfoliated WS2 particles in the region of 565–685 nm showing coincident PL emission at the A1 exciton peak at 1.97 eV. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
164
S.M. Notley / Journal of Colloid and Interface Science 396 (2013) 160–164
are potentially less risky in terms of environmental hazards. The production method for producing photoluminescent WS2 particles is furthermore easily scalable and is of relatively low cost compared to other visible light catalysts. It is important also to recognise that whilst the magnitude of the bandgap is important in terms of the energy of photons that may be absorbed, the positioning of the valence and conduction bands ultimately determines the ability for these materials to participate in redox reactions. Previous studies have shown the conduction band for confined WS2 to be shifted to more negative potential relative to the hydrogen electrode allowing these materials to be used to produce H2 [6,7]. Furthermore, the bands of MoS2 and WS2 are less influenced by solution pH in comparison to TiO2 where hydrogen production typically occurs at pH 1 (or 1 M salt) which has a profound influence on stability of dispersions. In order to maintain the potential benefits of the increased band gap, it is hence important to be able to control the stability of the particles, as re-aggregation will almost certainly result in the transition back to the indirect band gap material and inability to participate in many important photocatalytic reactions. In order to inhibit aggregation in aqueous suspension, surfactant is added during the exfoliation procedure. The surfactant used here was a nonionic tri-block copolymer with the adsorbing group, polypropylene oxide. The long polyethylene oxide blocks act to sterically stabilise the particles [40]. As these polymer chains are non-ionic, the particles are hence stable against flocculation up to ionic strengths in excess of 0.15 M however beyond this, the ethylene oxide groups lose hydration and hence ability to extend beyond the range of the strong dispersion forces of the WS2 sheets. However, the polymer may act as an ‘‘intercalating’’ layer and hence the interesting observed electronic and optical behaviour of the WS2 nanoparticles may be maintained. For incorporation into colloidal bed type reactors for the solar production of hydrogen, typical salt concentrations are of the order of 0.1 M or less. Hence, the high yield production method detailed here for producing a semiconducting material such as WS2 represents a potential route for realising the goal of harvesting hydrogen from water on a large scale. 4. Conclusions A method for the high yield production of photoluminescent WS2 particles in aqueous suspension has been described. The particles are highly stable in suspension promoted through the use of a non-ionic block co-polymer surfactant which creates a steric barrier to aggregation. These particles were stable for in excess of 3 months however some sedimentation was observed over time. The particles could though be easily redispersed through agitation. The properties of the exfoliated WS2 material were substantially different to the bulk WS2 powder. In particular, the electronic properties showed significant changes on confinement of the particles to two dimensions, i.e., a single layer of WS2. The raman spectra showed a significant blue shift of the E2g and A1g peaks as well as changes in intensity. The most interesting observed change on exfoliation was the emergence of a strong emission in the photoluminescence spectrum of the single layer WS2 sheets which was absent in the bulk material. This peak suggests that single layer WS2 is a direct band-gap semiconducting material with energy of 1.97 eV. The magnitude of this band gap is highly suited to a range of photocatalytic applications. Hence, this high yielding preparation technique provides a potential route to realising many goals in terms of water purification and potential in photoelectrochemical splitting of water to produce hydrogen.
Acknowledgments Financial support from the Australian Research Council’s Future Fellowship scheme is gratefully acknowledged. Animesh Basak, Centre for Advanced Microscopy at the ANU is thanked for performing the TEM imaging. References [1] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V. Grigorieva, A.A. Firsov, Science 36 (2004) 666. [2] A.K. Geim, K.S. Novoselov, Nat. Mater. 6 (2007) 183. [3] A.K. Geim, Science 324 (2009) 1530–1534. [4] H.S.S. Ramakrishna Matte, A. Gomathi, A.K. Manna, D.J. Late, R. Datta, S.K. Pati, C.N.R. Rao, Angew. Chem. Int. Ed. 49 (2010) 4059–4062. [5] R.M. Navarro Yerga, M. Consuelo Alvaraz Galvan, F. del Valle, J.A. Villoria de la Mano, J.L.G. Fierro, ChemSusChem 2 (2009) 471–485. [6] K. Maeda, J. Photochem. Photobiol., C 12 (2011) 237–268. [7] B.L. Abrams, J.P. Wilcoxon, Crit. Rev. Solid State Mater. Sci. 30 (2005) 153–182. [8] J.P. Wilcoxon, P.P. Newcomer, G.A. Samara, J. Appl. Phys. 81 (1997) 7934–7944. [9] J.P. Wilcoxon, J. Phys. Chem. B 104 (2001) 7334. [10] T.R. Thurston, J.P. Wilcoxon, J. Phys. Chem. B 103 (1999) 11. [11] J.M. Huang, R.A. Laitinen, D.F. Kelley, Phys. Rev. B 62 (2000) 10995–11005. [12] C. Lee, H. Yan, L.E. Brus, T.F. Heinz, J. Hone, S. Ryu, ACS Nano 4 (2010) 2695– 2700. [13] A. Splendiani, L. Sun, Y. Zhang, T. Li, J. Kim, C.Y. Chim, G. Galli, F. Wang, Nano Lett. 10 (2010) 1271–1275. [14] K.F. Mak, C. Lee, J. Hone, J. Shan, T.F. Heinz, Phys. Rev. Lett. 105 (2010) 136805. [15] S. Park, R.S. Ruoff, Nat. Nanotechnol. 4 (2009) 217–224. [16] A.C. Ferrari, J.C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K.S. Novoselov, S. Roth, A.K. Geim, Phys. Rev. Lett. 97 (2006) 187401. [17] K.S. Novoselov, D. Jiang, F. Schedin, T.J. Booth, V.V. Khotkevitch, S.V. Morozov, A.K. Geim, Proc. Natl. Acad. Sci. USA 102 (2005) 10451. [18] D. Li, M.B. Muller, S. Gilje, R.B. Kaner, G.G. Wallace, Nat. Nanotechnol. 3 (2008) 101–105. [19] M. Choucair, P. Thordarson, J.A. Stride, Nat. Nanotechnol. 4 (2009) 30–33. [20] Y. Hernandez, M. Lotya, D. Rickard, S.D. Bergin, J.N. Coleman, Langmuir 26 (2010) 3208–3213. [21] Y. Hernandez, V. Nicolosi, M. Lotya, F.M. Blighe, Z. Sun, S. De, I.T. McGovern, B. Holland, M. Byrne, Y.K. Gun’ko, J.J. Boland, P. Niraj, G. Duesberg, S. Krishnamurthy, R. Goodhue, J. Hutchison, V. Scardaci, A.C. Ferrari, J.N. Coleman, Nat. Nanotechnol. 3 (2008) 563. [22] M. Lotya, Y. Hernandez, P.J. King, R.J. Smith, V. Nicolosi, L.S. Karlsson, F.M. Blighe, S. De, W. Zhiming, I.T. McGovern, G.S. Duesberg, J.N. Coleman, J. Am. Chem. Soc. 131 (2009) 3611–3620. [23] M. Lotya, P.J. King, U. Khan, S. De, J.N. Coleman, ACS Nano 4 (2010) 3155–3162. [24] X. Cui, C. Zhang, R. Hao, Y. Huo, Nanoscale 3 (2011) 2118–2126. [25] D. Nuvoli, L. Valentini, V. Alzari, S. Scognamillo, S.B. Bon, M. Piccinini, J. Illescas, A. Mariani, J. Mater. Chem. 21 (2011) 3428–3431. [26] Y.T. Liang, M.C. Hersam, J. Am. Chem. Soc. 132 (2010) 17661–17663. [27] J.-W.T. Seo, A.A. Green, A.L. Antaris, M.C. Hersam, J. Phys. Chem. Lett. 2 (2011) 1004–1008. [28] X. Wang, P.F. Fulvio, G.A. Baker, G.M. Veith, R.R. Unocic, S.M. Mahurin, M. Chi, S. Dai, Chem. Commun. 46 (2010) 4487–4489. [29] S.M. Notley, Langmuir 28 (2012) 14110–14113. [30] R.J. Smith, M. Lotya, J.N. Coleman, New J. Phys. 12 (2010) [Art No. 125008]. [31] A. Griffith, S.M. Notley, J. Colloid Interface Sci. 369 (2012) 210–215. [32] J.N. Coleman, M. Lotya, A. O’Neill, S.D. Bergin, P.J. King, U. Khan, K. Young, A. Gaucher, S. De, R.J. Smith, I.V. Shvets, S.K. Arora, G. Stanton, H.-Y. Kim, K. Lee, G.T. Kim, G.S. Duesberg, T. Hallam, J.J. Boland, J.J. Wang, J.F. Donegan, J.C. Grunlan, G. Moriarty, A. Shmeliov, R.J. Nicholls, J.M. Perkins, E.M. Grieveson, K. Theuwissen, D.W. McComb, P.D. Nellist, V. Nicolosi, Science 331 (2011) 568– 571. [33] R.J. Smith, P.J. King, M. Lotya, C. Wirtz, U. Khan, S. De, A. O’Neill, G.S. Duesberg, J.C. Grunlan, G. Moriarty, J. Chen, J. Wang, A.I. Minett, V. Nicolosi, J.N. Coleman, Adv. Mater. 23 (2011) 3944–3948. [34] S.M. Notley, J. Colloid Interface Sci. 375 (2012) 35–40. [35] N. Bertram, J. Cordes, Y.D. Kim, G. Gantefor, S. Gemming, G. Seifert, Chem. Phys. Lett. 418 (2006) 36–39. [36] M. Virsek, A. Jesih, I. Milosevic, M. Damnjanovic, M. Remskar, Surf. Sci. 601 (2007) 2868–2872. [37] G.L. Frey, R. Tenne, M.J. Matthews, M.S. Dresselhaus, G. Dresselhaus, Phys. Rev. B 60 (1999) 2883–2892. [38] A. Molina-Sanchez, L. Wirtz, Phys. Rev. B 84 (2011) 155413. [39] S. Lebegue, O. Eriksson, Phys. Rev. B 79 (2009) 115409. [40] D.H. Napper, Polymeric Stabilization of Colloidal Dispersions, Academic Press, New York, 1983.
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
High yield production of photoluminescent tungsten disulphide nanoparticles Shannon M. Notley Faculty of Life and Social Sciences, Swinburne University of Technology, Hawthorn 3122, VIC, Australia
a r t i c l e
i n f o
Article history: Received 27 November 2012 Accepted 15 January 2013 Available online 7 February 2013 Keywords: WS2 Raman Nanoparticle Photoluminescence Steric stabilisation
a b s t r a c t Single and few layered tungsten disulphide (WS2) nanoparticles were prepared using a surfactant assisted ultrasonication exfoliation technique with concentrations of up to 0.4 mg/mL. The lateral dimension of the particles was in the range of 100–250 nm. The exfoliated WS2 was stabilised against re-aggregation through adsorption of a tri-block non-ionic polymeric surfactant (PEO–PPO–PEO). These nanoparticles were characterised by absorption, Raman and photoluminescence spectroscopy (PL). Broadening of the E2g peak in the Raman spectrum was observed due to phonon confinement within a single layer of WS2. The exfoliated particles have significantly different properties than the bulk WS2 material, in particular, the emergence of strong photoluminescence at 1.97 eV in energy coincidental with the excitonic peak in the UV–Vis spectrum. The emergent PL emission suggests that the monolayer WS2 is a direct gap material analogous to other dichalcogenides such as MoS2. Ó 2013 Elsevier Inc. All rights reserved.
1. Introduction Atomically thin materials have received great attention, the most commonly cited example being graphene [1–3]. Other van der Waals bonded solids though, such as the dichalcogenides, also have interesting optical and electronic properties particularly in the limit of 2D [4]. Chiefly among these, the transition metal dichalcogenide semiconductors molybdenite (MoS2) and tungstenite (WS2) have attracted interest for potential use in advanced applications such as photovoltaics and photocatalysis due to tunable band gaps depending on size. There is a strong push from the renewable energy interests to create materials with appropriate valence and conduction bands which allow the splitting of water to form hydrogen and oxygen through the utilisation of the maximum amount of incident solar radiation. Typically, TiO2 has been used in such applications however the large band gap of 3.2 eV means less than 5% of the incident solar photons can be absorbed. Both MoS2 and WS2 in the bulk form are indirect band gap semi-conducting materials with band gaps of 1.22 eV and 1.20 eV respectively. The magnitude of this gap is insufficient for many photocatalytic applications, specifically the photoelectrochemical splitting of water [5,6]. Thermodynamically, the minimum energy required is 1.23 eV but practical considerations means a band gap greater than 1.4 eV is needed. Furthermore, the positioning of the valence and conduction bands is such that the generation of hydroxyl radicals is also not possible reducing the potential for these materials to be used in many photocatalytic applications. Hence,
E-mail address: [email protected] 0021-9797/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcis.2013.01.035
much effort has been taken in tuning the band gap properties of the dichalcogenides to more suitable values through quantum confinement effects [7]. 3D confinement has led to increased band gaps of up to 2.5 eV for crystallite sizes of 4–5 nm for MoS2. Precipitation inside micelles has typically been used in order to produce these nanoparticles [8–10] and analogous WS2 particles [11]. Recent studies have shown that confinement effects are also demonstrated by the 2D analogues of MoS2 with emerging photoluminescence suggesting a transition to a direct band gap material [12–14]. Thus, it is timely that the methods used for generating large scale quantities of graphene be adapted for producing single and few layered materials of the transition metal dichalcogenides. Many methods have been described for the production of 2D materials such as graphene [15]. Whilst large area graphene was first produced using exfoliation with adhesive tape [1,16,17], subsequent methods such as intercalation, reduction of graphene oxide [18], solvothermal techniques [19] and sonication in appropriate solvent systems [20–28] have all led to increased quantities of material. Aqueous solution processing has many advantages, particularly in terms of cost and the environment. One of the more promising techniques is the use of ultrasonic exfoliation of graphene and related materials in the presence of aqueous surfactant solutions [21,23,29]. The surfactant is necessary to promote exfoliation for two reasons. The first, it reduces the surface tension of the aqueous phase to a magnitude of the order of the cohesive energy required to separate the solid sheets. Secondly, it adsorbs onto the surface of the exfoliated particles creating an extra repulsive term that inhibits re-aggregation of the produced particles [30,31]. This surfactant assisted ultrasonication technique was recently adapted from the graphene system to other van der Waals bonded solids [32], including the semi-conducting dichalcogenides MoS2 and
S.M. Notley / Journal of Colloid and Interface Science 396 (2013) 160–164
WS2. In this paper, a method for the high yield production of photoluminescent WS2 nanoparticles under aqueous conditions is described. Surfactant is added continuously throughout the sonication thereby giving rise to a much greater concentration of exfoliated WS2 particles. These particles have been characterised in terms of their size and their opto-electronic properties, with interesting photoluminescence observed.
2. Experimental section 2.1. Materials Powdered Tungsten(IV) Sulphide with a bulk particle size of 2 lm was purchased from Sigma–Aldrich Australia and used as received. A tri-block co-polymeric surfactant, Pluronic F108 (BASF, Ludwigshafen) with a molecular weight of 14.6 kDa was used to stabilise the exfoliated particles as well as control the liquid–vapour interfacial tension. Milli-Q water was used in the preparation of all suspensions.
161
2.2.3. UV–Vis spectrophotometry UV–Vis spectra of the diluted WS2 suspensions were measured using a Shimadzu UV-3101PC spectrophotometer with matched quartz cuvettes over the wavelength range of 250–800 nm. 2.2.4. Raman spectroscopy and photoluminescence spectra The synthesised WS2 particles were characterised using Raman spectroscopy with a Horiba John Yvon T64000 Raman system. Laser excitation at 532 nm was used. Typically, WS2 particles were deposited onto an oxidised silicon wafer. The photoluminescence spectrum of the particles was also measured using the laser excitation at 532 nm. 2.2.5. Zeta potential measurements The zeta potential of the WS2 particles was determined using a Malvern Zetasizer Nano. The measurements were performed on the suspension in Milli-Q water (pH 6.5) after extensive dialysis to remove any unbound non-ionic surfactant. 3. Results and discussion
2.2. Methods 2.2.1. Production of aqueous suspensions of WS2 particles WS2 nanoparticle suspensions were prepared using the surfactant assisted ultrasonic exfoliation technique as previously described [21,23,27,29,31–34]. A 2% w/w suspension of bulk WS2 was added to Milli-Q water. Under continuous sonication at 60 W using a ‘‘Cell Disruptor’’ Model W-220F (Heat Systems-Ultrasonics Inc) sonicator, non-ionic surfactant was added from a highly concentrated solution to maintain the concentration of surfactant in water at approximately 0.1% w/w. The exfoliation procedure results in a vast increase in solid–liquid surface area which leads to rapid depletion of the surfactant through adsorption to the particles surface. Thus, continuously adding surfactant during sonication leads to a significantly greater yield of nanoparticulate WS2 [29]. This concentration of surfactant is suitable for maintaining the optimum surface tension as described previously at 40– 42 mJ/m2 [32]. The resultant exfoliated suspension was then dialysed in MilliQ water for 48 h to remove any non-adsorbed surfactant from the solution. The suspension was then typically centrifuged at 1500 rpm for 5 min to sediment further any large (non-exfoliated) particles. The supernatant was collected and was found to be stable for up to 3 months. Some sedimentation was observed due to the relatively high density of WS2 (7.5 g/mL) but the particles could be readily re-dispersed through simple agitation. The particles were subsequently characterised in terms of size, charge and spectroscopic properties as described below. The yield of single and few layer WS2 nanoparticles produced using this method was determined gravimetrically to be as high as 0.4 mg/mL depending on the rate of centrifugation. Furthermore, density measurements of the suspensions were also performed to confirm the concentration of WS2 particles. The thickness of the WS2 particles was typically characterised based upon the photoluminescence measurements. That is, the proportion of particles which demonstrated strong luminescence at 1.97 eV, with at least 100 particles measured. 2.2.2. Transmission Electron Microscopy (TEM) imaging The dried suspension of WS2 particles was imaged using TEM to confirm that single and few layered material was prepared as well as to determine the lateral sizes of the particles. A Hitachi H7100FA transmission electron microscope with an accelerating voltage of 125 kV was used. Furthermore, electron diffraction patterns of the particles were measured.
The surfactant assisted sonication method was used to exfoliate single and few layer WS2 from the bulk particles in the presence of the non-ionic tri-block surfactant. This technique involves a relatively high powered sonicator probe however little oxidation was observed as determined through the measurement of the nanoparticle zeta potential. Typically, the f potential of the WS2 at pH 6.5 was less than 5 mV. This particle charge is significantly lower than the graphene analogues however the dangling edges of the WS2 are stabilised by sulphur atoms [35] leading to reduced overall charge in comparison to the graphene which is predominately stabilised through oxygen containing moieties. This overall surface charge is typically too low to stabilise the particles in suspension. However, the non-ionic surfactant adsorbs to the surface allowing stability of the WS2 for in excess of 3 months. Fig. 1 shows an image of the nanoparticlulate WS2 suspension, which has a slight green/brown1 colour. The UV–Vis spectrum of a 100 diluted suspension of WS2 nanoparticles was measured in the range of 250–800 nm. As can be seen in Fig. 1, the particles absorb strongly in the UV region and furthermore, a peak is observed at 627 nm (1.97 eV), commonly assigned as the ‘‘A exciton’’ and related to the direct excitonic transition at the K point of the Brillouin zone [13]. This peak at 627 nm is also present in the bulk material which demonstrates that there is no a significant shift in the spectrum as may be expected from simple two dimensional confinement. Wilcoxon et al. have demonstrated that changes in the adsorption edge only occur for confinement in three dimensions [8]. Furthermore, the spectrum shows strong absorption at wavelengths below 627 nm, particularly in the UV region in agreement with previous spectrophotometric studies [7]. Confinement in 3D leads to a shift in the absorption edge and decreases in the position of the exciton. It is clear however that although the size of the WS2 particles are confined in terms of thickness, the relatively large lateral dimensions leads to maintenance of the position of the excitonic peak at 627 nm. Previous studies suggest that there is only a very small shift (0.15 eV) due to confinement in 2D [8]. The peak associated with the indirect B excitonic transition is not as prominent as for the isomorphous MoS2 analogues in either the bulk or exfoliated samples shown in Fig. 1. The difference in energies of the A and B excitons peaks upon exfoliation is not expected to change significantly under these 2D confinement conditions unlike in 3D con1 For interpretation of color in Fig. 1, the reader is referred to the web version of this article.
162
S.M. Notley / Journal of Colloid and Interface Science 396 (2013) 160–164
Fig. 1. UV–Vis spectrum of the 100 diluted WS2 nanoparticulate suspension (inset) prepared using the surfactant assisted sonication exfoliation technique compared with the bulk WS2 starting material. Position of the ‘‘A’’ exciton highlighted.
finement where the spin–orbit splitting can be heavily influenced by the small lateral sizes [7,8]. The exfoliated WS2 nanoparticles were imaged using TEM to determine lateral sizes as well as to demonstrate that the thicknesses of the particles were of the order required for transmission of the electrons. It was not possible to directly determine the number of layers from TEM measurements. Fig. 2 shows a typical image of the particles. Most particles had lateral sizes in the 100–250 nm range. The proportion of single layer WS2 (determined from PL measurements) was dependent upon the sedimentation or centrifugation step in the preparation. Typically, faster centrifugation resulted in much smaller particles, particularly in terms of thickness, and a resulting reduction in concentration in suspension as shown in Fig. 2. This is due to the relative high density of the particles in comparison to water. As centrifugation separates on mass, it is likely that particles of small lateral dimensions but with multiple layers are sedimented but larger single layer particles would also be separated. However, a high proportion of single layer material could be achieved in this way and although sedimentation was observed over time, the particles could be easily redispersed through
simple agitation. The electron diffraction pattern of single layer WS2 nanoparticles was also determined and is shown in Fig. 2. Clearly, the typical hexagonal pattern as expected from the trigonal bipyramidal spatial arrangement is demonstrated. Raman spectroscopy has been demonstrated to be a valuable tool in research into 2D materials as shifts in peak positions and shapes give much information on the structure of the material, particularly the number of layers and also the presence of any defects [16]. Fig. 3 shows Raman spectra for the bulk WS2 powder as well as the exfoliated material deposited onto a silicon wafer. The silicon peak at 520 cm 1 was used to calibrate the absolute peak positions for both samples (Si peak not shown). Two peaks, at 347 cm 1 and 410 cm 1 in the bulk sample are shown in Fig. 3 and these are assigned to E2g and A1g respectively. The spectra demonstrate that there is a significant blue shift of the two peaks in region of 340–420 cm 1 upon exfoliation to single layer material. This shift in the peaks could potentially be due to stress induced in the lateral plane through drying onto the substrate and has been observed previously in studies involving WS2 nanotubes [36]. However the reduction of the interlayer interaction for the exfoliated particles may also lead to the observed shift in these peaks. Furthermore, the intensity of the peak at 350 cm 1 for the exfoliated sample, is much higher and broader. The broadening of this Raman band is due to phonon confinement within the single layer and demonstrates that the lateral dimensions of the particle are in the nanometre range [4,37]. It is interesting to note though that both E2g and A1g peaks are blue shifted which is in direct contrast to previous experimental and theoretical studies on the analogous MoS2 material which show a decrease in A1g frequency and increase in E2g [12,38]. The A1g is the out of plane phonon mode and is therefore almost exclusively related to the sulphur atoms. Exfoliation to a single layer hence removes the interactions with neighbouring interlayer sulphur atoms which leads to a decrease in the frequency. However, the shift in the E2g peak is not as straight forward to explain. The E2g peak is the in-plane bending phonon mode and it can be argued as above that the weak interlayer attraction dominates. This is contrasted though by the ab initio calculations by Molina-Sanchez and Wirtz where they found that longer range Coulombic interactions is most influential, a situation supported by experimental determinations of peaks in MoS2. It is possible though that the heavier W
Fig. 2. (A) Yield of WS2 particles (closed circles) and proportion of particles showing photoluminescence (open squares) as a function of centrifugation rate for a set time period of 5 min. (B) Schematic of spatial arrangement of atoms in crystalline WS2. (C) TEM image of single and few layer WS2 nanoparticles. (D) Electron diffraction pattern of WS2 nanoparticles.
S.M. Notley / Journal of Colloid and Interface Science 396 (2013) 160–164
Fig. 3. Raman spectra of bulk WS2 (bottom) and exfoliated WS2 nanoparticles (top) deposited onto silicon wafer. Absolute peak positions were calibrated against the silicon peak at 540 cm 1, the exfoliated material results in a blue shift of the peaks assigned as E2g and A1g. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
atoms provide the impetus for a greater interlayer attraction giving rise to the observed increase in frequency as shown in Fig. 3. Recent studies have shown that single and few layer MoS2 particles demonstrate photoluminescence, the intensity of which increases with decreasing thickness of the particles [13,14]. Indeed, even a double layer can see a significant drop in intensity as well as a shift to lower energy of the primary peak. The PL spectrum of the stabilised WS2 particles produced using the surfactant assisted exfoliation technique outlined above was measured here and is shown in Fig. 4. The proportion of particles which displayed this strong luminescence was used in order to determine the relative efficiency of the exfoliation process. In comparison to the bulk WS2 which shows no observable PL emission, a strong peak of about 40 meV width is observed for the exfoliated WS2 particles. The position of this PL peak at the photon energy of 1.97 eV is coincident with the A1 exciton peak in the UV–Vis spectrum of WS2 as also shown in Fig. 4 (data from Fig 1 converted from wavelength to energy). Of the order of 63% of particles surveyed using the PL measurements after excitation at 532 nm demonstrated this strong and sharp peak at 1.97 eV. It is unusual for indirect band gap semiconductors to show appreciable photoluminescence although similar properties have been observed for the analogous transition metal dichalcogenide, MoS2 [13]. Here, two peaks corresponding to the positions of the
163
A1 and B1 excitons were observed yet in another study, only one strong emission peak was resolved at the A1 energy in the limit of a single MoS2 layer [14]. These studies suggested that the photoluminescence of MoS2 resulted from the transition to a direct band gap semiconducting material with decreasing thickness. Evidence from ab initio calculations supported this result [39]. That the position and width of the PL peak, in terms of energy, corresponds with the exciton peak measured using UV–Vis spectroscopy provides further compelling evidence beyond the fact that photoluminescence was indeed observed. Similarly, the results in Fig. 4 demonstrate that the analogous material, WS2, shows the same behaviour. That is, a transition from an indirect to direct band gap semiconductor in the limit of 2D and that only a single PL peak is observed. Typically the surfactant assisted exfoliation technique produces a distribution of single and few layered van der Waals bonded materials. WS2 nanoparticles should photoluminesce, albeit with reduced intensity, for exfoliated particles with more than a single layer. This means that this production method for the WS2 is capable of giving a large quantity of particles suitable for use in photocatalytic applications as the band gap of the dichalcogenide has been increased from 1.20 eV up to 1.97 eV. Importantly, the magnitude of the band gap and the positioning of the valence and conduction bands give rise to the potential use of these particles in the production of hydrogen in a photoelectrochemical cells through the splitting of water. Whilst the magnitude of the bandgap is not of the same order as for nanocrystals of the transition metal dichalcogenides, the ability to tune the band gap through 2D rather than 3D confinement is nevertheless advantageous. This is particularly important when the availability of solar electromagnetic radiation is considered. The bandgap of WS2 observed here means approximately 40% of incident solar photons possess the energy for absorption and electron–hole separation, in comparison to TiO2 where less than 4% of photons can be absorbed. Stable aqueous suspensions of semiconducting nanoparticles with appropriate band gaps have been promoted as an alternate means for generating hydrogen rather than the use of electrodes [6]. Many examples of semi-conductor catalysts use visible light to drive photocatalytic reactions. Quantum dots such as CdSe and CdS have bandgaps of 1.7 eV and 2.4 eV respectively which makes them highly suited for this application, particularly as they have high quantum efficiencies. However, many of the materials currently described are far from stable and pose potential environmental problems upon degradation which limits their use in colloidal bed type reactors for hydrogen productions. WS2 and the analogue MoS2 are far more stable against degradation and
Fig. 4. Left: PL spectrum of bulk WS2 (black) and exfoliated WS2 (red) under laser excitation at 532 nm. Right: PL spectrum and UV–Vis spectrum of the exfoliated WS2 particles in the region of 565–685 nm showing coincident PL emission at the A1 exciton peak at 1.97 eV. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
164
S.M. Notley / Journal of Colloid and Interface Science 396 (2013) 160–164
are potentially less risky in terms of environmental hazards. The production method for producing photoluminescent WS2 particles is furthermore easily scalable and is of relatively low cost compared to other visible light catalysts. It is important also to recognise that whilst the magnitude of the bandgap is important in terms of the energy of photons that may be absorbed, the positioning of the valence and conduction bands ultimately determines the ability for these materials to participate in redox reactions. Previous studies have shown the conduction band for confined WS2 to be shifted to more negative potential relative to the hydrogen electrode allowing these materials to be used to produce H2 [6,7]. Furthermore, the bands of MoS2 and WS2 are less influenced by solution pH in comparison to TiO2 where hydrogen production typically occurs at pH 1 (or 1 M salt) which has a profound influence on stability of dispersions. In order to maintain the potential benefits of the increased band gap, it is hence important to be able to control the stability of the particles, as re-aggregation will almost certainly result in the transition back to the indirect band gap material and inability to participate in many important photocatalytic reactions. In order to inhibit aggregation in aqueous suspension, surfactant is added during the exfoliation procedure. The surfactant used here was a nonionic tri-block copolymer with the adsorbing group, polypropylene oxide. The long polyethylene oxide blocks act to sterically stabilise the particles [40]. As these polymer chains are non-ionic, the particles are hence stable against flocculation up to ionic strengths in excess of 0.15 M however beyond this, the ethylene oxide groups lose hydration and hence ability to extend beyond the range of the strong dispersion forces of the WS2 sheets. However, the polymer may act as an ‘‘intercalating’’ layer and hence the interesting observed electronic and optical behaviour of the WS2 nanoparticles may be maintained. For incorporation into colloidal bed type reactors for the solar production of hydrogen, typical salt concentrations are of the order of 0.1 M or less. Hence, the high yield production method detailed here for producing a semiconducting material such as WS2 represents a potential route for realising the goal of harvesting hydrogen from water on a large scale. 4. Conclusions A method for the high yield production of photoluminescent WS2 particles in aqueous suspension has been described. The particles are highly stable in suspension promoted through the use of a non-ionic block co-polymer surfactant which creates a steric barrier to aggregation. These particles were stable for in excess of 3 months however some sedimentation was observed over time. The particles could though be easily redispersed through agitation. The properties of the exfoliated WS2 material were substantially different to the bulk WS2 powder. In particular, the electronic properties showed significant changes on confinement of the particles to two dimensions, i.e., a single layer of WS2. The raman spectra showed a significant blue shift of the E2g and A1g peaks as well as changes in intensity. The most interesting observed change on exfoliation was the emergence of a strong emission in the photoluminescence spectrum of the single layer WS2 sheets which was absent in the bulk material. This peak suggests that single layer WS2 is a direct band-gap semiconducting material with energy of 1.97 eV. The magnitude of this band gap is highly suited to a range of photocatalytic applications. Hence, this high yielding preparation technique provides a potential route to realising many goals in terms of water purification and potential in photoelectrochemical splitting of water to produce hydrogen.
Acknowledgments Financial support from the Australian Research Council’s Future Fellowship scheme is gratefully acknowledged. Animesh Basak, Centre for Advanced Microscopy at the ANU is thanked for performing the TEM imaging. References [1] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V. Grigorieva, A.A. Firsov, Science 36 (2004) 666. [2] A.K. Geim, K.S. Novoselov, Nat. Mater. 6 (2007) 183. [3] A.K. Geim, Science 324 (2009) 1530–1534. [4] H.S.S. Ramakrishna Matte, A. Gomathi, A.K. Manna, D.J. Late, R. Datta, S.K. Pati, C.N.R. Rao, Angew. Chem. Int. Ed. 49 (2010) 4059–4062. [5] R.M. Navarro Yerga, M. Consuelo Alvaraz Galvan, F. del Valle, J.A. Villoria de la Mano, J.L.G. Fierro, ChemSusChem 2 (2009) 471–485. [6] K. Maeda, J. Photochem. Photobiol., C 12 (2011) 237–268. [7] B.L. Abrams, J.P. Wilcoxon, Crit. Rev. Solid State Mater. Sci. 30 (2005) 153–182. [8] J.P. Wilcoxon, P.P. Newcomer, G.A. Samara, J. Appl. Phys. 81 (1997) 7934–7944. [9] J.P. Wilcoxon, J. Phys. Chem. B 104 (2001) 7334. [10] T.R. Thurston, J.P. Wilcoxon, J. Phys. Chem. B 103 (1999) 11. [11] J.M. Huang, R.A. Laitinen, D.F. Kelley, Phys. Rev. B 62 (2000) 10995–11005. [12] C. Lee, H. Yan, L.E. Brus, T.F. Heinz, J. Hone, S. Ryu, ACS Nano 4 (2010) 2695– 2700. [13] A. Splendiani, L. Sun, Y. Zhang, T. Li, J. Kim, C.Y. Chim, G. Galli, F. Wang, Nano Lett. 10 (2010) 1271–1275. [14] K.F. Mak, C. Lee, J. Hone, J. Shan, T.F. Heinz, Phys. Rev. Lett. 105 (2010) 136805. [15] S. Park, R.S. Ruoff, Nat. Nanotechnol. 4 (2009) 217–224. [16] A.C. Ferrari, J.C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K.S. Novoselov, S. Roth, A.K. Geim, Phys. Rev. Lett. 97 (2006) 187401. [17] K.S. Novoselov, D. Jiang, F. Schedin, T.J. Booth, V.V. Khotkevitch, S.V. Morozov, A.K. Geim, Proc. Natl. Acad. Sci. USA 102 (2005) 10451. [18] D. Li, M.B. Muller, S. Gilje, R.B. Kaner, G.G. Wallace, Nat. Nanotechnol. 3 (2008) 101–105. [19] M. Choucair, P. Thordarson, J.A. Stride, Nat. Nanotechnol. 4 (2009) 30–33. [20] Y. Hernandez, M. Lotya, D. Rickard, S.D. Bergin, J.N. Coleman, Langmuir 26 (2010) 3208–3213. [21] Y. Hernandez, V. Nicolosi, M. Lotya, F.M. Blighe, Z. Sun, S. De, I.T. McGovern, B. Holland, M. Byrne, Y.K. Gun’ko, J.J. Boland, P. Niraj, G. Duesberg, S. Krishnamurthy, R. Goodhue, J. Hutchison, V. Scardaci, A.C. Ferrari, J.N. Coleman, Nat. Nanotechnol. 3 (2008) 563. [22] M. Lotya, Y. Hernandez, P.J. King, R.J. Smith, V. Nicolosi, L.S. Karlsson, F.M. Blighe, S. De, W. Zhiming, I.T. McGovern, G.S. Duesberg, J.N. Coleman, J. Am. Chem. Soc. 131 (2009) 3611–3620. [23] M. Lotya, P.J. King, U. Khan, S. De, J.N. Coleman, ACS Nano 4 (2010) 3155–3162. [24] X. Cui, C. Zhang, R. Hao, Y. Huo, Nanoscale 3 (2011) 2118–2126. [25] D. Nuvoli, L. Valentini, V. Alzari, S. Scognamillo, S.B. Bon, M. Piccinini, J. Illescas, A. Mariani, J. Mater. Chem. 21 (2011) 3428–3431. [26] Y.T. Liang, M.C. Hersam, J. Am. Chem. Soc. 132 (2010) 17661–17663. [27] J.-W.T. Seo, A.A. Green, A.L. Antaris, M.C. Hersam, J. Phys. Chem. Lett. 2 (2011) 1004–1008. [28] X. Wang, P.F. Fulvio, G.A. Baker, G.M. Veith, R.R. Unocic, S.M. Mahurin, M. Chi, S. Dai, Chem. Commun. 46 (2010) 4487–4489. [29] S.M. Notley, Langmuir 28 (2012) 14110–14113. [30] R.J. Smith, M. Lotya, J.N. Coleman, New J. Phys. 12 (2010) [Art No. 125008]. [31] A. Griffith, S.M. Notley, J. Colloid Interface Sci. 369 (2012) 210–215. [32] J.N. Coleman, M. Lotya, A. O’Neill, S.D. Bergin, P.J. King, U. Khan, K. Young, A. Gaucher, S. De, R.J. Smith, I.V. Shvets, S.K. Arora, G. Stanton, H.-Y. Kim, K. Lee, G.T. Kim, G.S. Duesberg, T. Hallam, J.J. Boland, J.J. Wang, J.F. Donegan, J.C. Grunlan, G. Moriarty, A. Shmeliov, R.J. Nicholls, J.M. Perkins, E.M. Grieveson, K. Theuwissen, D.W. McComb, P.D. Nellist, V. Nicolosi, Science 331 (2011) 568– 571. [33] R.J. Smith, P.J. King, M. Lotya, C. Wirtz, U. Khan, S. De, A. O’Neill, G.S. Duesberg, J.C. Grunlan, G. Moriarty, J. Chen, J. Wang, A.I. Minett, V. Nicolosi, J.N. Coleman, Adv. Mater. 23 (2011) 3944–3948. [34] S.M. Notley, J. Colloid Interface Sci. 375 (2012) 35–40. [35] N. Bertram, J. Cordes, Y.D. Kim, G. Gantefor, S. Gemming, G. Seifert, Chem. Phys. Lett. 418 (2006) 36–39. [36] M. Virsek, A. Jesih, I. Milosevic, M. Damnjanovic, M. Remskar, Surf. Sci. 601 (2007) 2868–2872. [37] G.L. Frey, R. Tenne, M.J. Matthews, M.S. Dresselhaus, G. Dresselhaus, Phys. Rev. B 60 (1999) 2883–2892. [38] A. Molina-Sanchez, L. Wirtz, Phys. Rev. B 84 (2011) 155413. [39] S. Lebegue, O. Eriksson, Phys. Rev. B 79 (2009) 115409. [40] D.H. Napper, Polymeric Stabilization of Colloidal Dispersions, Academic Press, New York, 1983.