Tungsten(VI) selenide tetrachloride, WSeCl₄ – synthesis, properties, coordination complexes and application of [WSeCl₄(SeⁿBu₂)] for CVD growth of WSe₂ thin films

Abstract

WSeCl₄ was obtained in good yield by heating WCl₆ and Sb₂Se₃in vacuo. Green crystals grown by sublimation were shown by single crystal X-ray structure analysis to contain square pyramidal monomers with apical W=Se, and powder X-ray diffraction (PXRD) analysis confirmed this to be the only form present in the bulk sample. Density functional theory (DFT) calculations using the B3LYP-D3 functional replicated the structure, identified the key bonding orbitals, and were used to aid assignment of the IR spectrum of WSeCl₄. Reaction of WSeCl₄ with ligands L gave [WSeCl₄(L)] (L = MeCN, DMF, thf, py, OPPh₃, 2,2′-bipy, SeMe₂, SeⁿBu₂), whilst the dimers [(WSeCl₄)₂(μ-L–L)] were formed with L–L = Ph₂P(O)CH₂P(O)Ph₂, 1,4-dioxane and 4,4′-bipyridyl. The complexes were characterised by microanalysis, IR and ¹H NMR spectroscopy, and single crystal X-ray structures determined for [WSeCl₄(L)] (L = OPPh₃, MeCN, DMF) and [(WSeCl₄)₂(μ-L–L)] (L–L = 1,4-dioxane, 4,4′-bipyridyl). All except the 2,2′-bipy complex, which is probably seven-coordinate, contain six-coordinate tungsten with the neutral donor trans to W=Se. Alkylphosphines, including PMe₃ and PEt₃, decompose WSeCl₄ upon contact, forming phosphine selenides (SePR₃). In contrast, the selenoether complexes [WSeCl₄(SeMe₂)] and [WSeCl₄(SeⁿBu₂)] were isolated and characterised. The crystal structure of the minor W(VI) by-product, [(WSeCl₄)₂(μ-SeMe₂)], was determined and using SMe₂, a few crystals of the W(V) species, [{WCl₃(SMe₂)}₂(μ-Se)(μ-Se₂)], were obtained and structurally characterised. The isolated W(VI) complexes are compared with those of WOCl₄ and WSCl₄ and the combination of experimental and computational data are consistent with WSeCl₄ being a weaker Lewis acid and its complexes significantly less stable than those of the lighter analogues, especially in solution. Low pressure chemical vapour deposition (LPCVD) using [WSeCl₄(SeⁿBu₂)] in the range 660–700 °C (0.1 mmHg) produced highly reflective thin films, which were identified to be WSe₂ by grazing incidence X-ray diffraction (XRD), scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS) and Raman spectroscopy. XRD analysis of the thinner films revealed them to be highly oriented in the <00l> direction.

---

Introduction

Transition metal oxide halides have played a major role in the development of coordination and organometallic chemistry and in the applications of early d-block elements.^1–3 In contrast, the heavier analogues, the sulfide and selenide halides, are little known and have been much less studied.^4,5 There are no examples of molecular chalcogenide iodides, and whilst chalcogenide fluorides exist for Mo and W, their coordination chemistry is unexplored.⁶ The synthesis and chemistry of MEX₃ (M = Nb, Ta; E = S, Se; X = Cl, Br) and WEX₄ were developed in the period 1965–1985, much of it by the groups of Fowles, Rice and Drew.^4,5 We have recently undertaken re-investigations and comparisons of the coordination chemistry of WSCl₄ and WSCl₃ with those of WOCl₄ and WOCl₃ with a range of N- and O-donor ligands,⁷ as well as with the softer phosphines and arsines,⁸ and thioethers.⁹ Comparisons of the chemistries of TaOCl₃ and TaSCl₃ have also been reported.^10,11 Thioether complexes of NbSCl₃, of type [{NbSCl₃(SR₂)}₂] (R = Me, ⁿBu) are obtained from reaction of NbCl₅, SR₂ and S(SiMe₃)₂ in CH₂Cl₂, whilst [NbSCl₃(L–L)] (L–L = MeS(CH₂)₂SMe, MeS(CH₂)₃SMe, ⁱPrS(CH₂)₂SⁱPr and ⁿBuS(CH₂)₃SⁿBu) are made from reaction of L–L with pre-formed [NbSCl₃(MeCN)₂].¹² Selected examples can function as single source precursors for low pressure chemical vapour deposition (LPCVD) of thin films of 3R-NbS₂.¹² [WSCl₄(SMe₂)] and [(WSCl₄)₂{μ-RS(CH₂)₂SR}] (R = Me, Ph, ⁱPr) and [(WSCl₄)₂{μ-MeS(CH₂)₃SMe}] have been made by direct reaction of WSCl₄ with the thioether in anhydrous CH₂Cl₂ solution.⁹ We also showed that [(WSCl₄)₂{μ-ⁱPrS(CH₂)₂SⁱPr}] functions as a single source precursor to deposit 2H-WS₂ thin films by low pressure chemical vapour deposition (LPCVD),⁹ providing further motivation to develop the chemistry of WSeCl₄ and its complexes.

Attempts to prepare dithio- or diseleno-ether complexes of NbSeCl₃ were unsuccessful, but reaction of NbCl₅, SeⁿBu₂ and (Me₃Si)₂Se in CH₂Cl₂ gave black [NbSe₂Cl₃(SeⁿBu₂)].¹² The structure of the latter is unknown, but may contain a diselenide unit ([Se₂]²⁻). This compound deposits 2H-NbSe₂ films by LPCVD at 600–700 °C, 0.05 mmHg.¹²

The early transition metal dichalcogenides ME₂ (M = Nb, Ta, Mo, V, W, etc., E = S, Se or Te) are layered semiconductors with good stabilities, and their properties can be tuned through the choice of metal and chalcogen.¹³ This versatility gives rise to a plethora of applications in (opto)electronics, magnetic materials, spintronics, sensors and electrocatalysis, which are very active areas of research.^13–16 Production of the materials as thin films maximises the anisotropy of their magnetic or electronic properties, hence techniques to deposit 2D layers are of particular interest. Chemical vapour deposition (CVD) is widely used in industry as a low-cost, scalable and very versatile method to deposit such films using either dual or single source precursors.^13–19 The latter can offer better control of film stoichiometry, cost-effective use of reagents, as well as the ability to selectively deposit the metal chalcogenide film onto lithographically patterned substrates.¹⁸ The application of molecular transition metal complexes as single source CVD precursors for the growth of metal chalcogenide thin films has been reviewed recently.²⁰

Although WSeCl₄ was first described in 1974,²¹ the few coordination complexes known are described in papers focussed rather on WSCl₄ adducts,^5,22–24 and a more detailed investigation of the selenide chloride is required to develop its chemistry and to determine if it could be the basis of a single source low pressure CVD route to WSe₂. A small number of dual source CVD routes to WSe₂ films have been described,^19,20 including using WCl₆ and SeEt₂ in atmospheric pressure CVD,²⁵ however, to the best of our knowledge there are no single source CVD reagents for tungsten diselenide films.

Here we report a detailed study of the synthesis, single crystal and powder X-ray structure and properties of WSeCl₄, its coordination chemistry with a range of neutral N-, O- and Se-donor ligands, and demonstrate that the new complex, [WSeCl₄(SeⁿBu₂)], can function as a single source precursor for the CVD growth of continuously interconnected WSe₂ thin films.

Results and discussion

WSeCl₄

WSeCl₄ was prepared by heating together a finely ground mixture of WCl₆ and Sb₂Se₃ in a 3 : 1 molar ratio in vacuo, according to the general method of Britnell et al.²¹ Obtaining good yields of the pure product required very careful control of the conditions, and hence the preparation and apparatus used in this work are described in detail in the Experimental section. The green powdered WSeCl₄ is moisture sensitive, but can be stored in a sealed container in the glove box indefinitely. Attempts to obtain WSeCl₄ by reaction of WCl₆ and Se(SiMe₃)₂ in a variety of organic solvents, including CH₂Cl₂, toluene or MeCN, and at various temperatures from ambient to −78 °C, failed. This contrasts with the ready synthesis of WSCl₄ from WCl₆ and S(SiMe₃)₂, and WOCl₄ from WCl₆ and O(SiMe₃)₂.^7–9,26 Refluxing WSeCl₄ with Cl₂C=CCl₂ also failed to produce WSeCl₃, whereas the similar reaction of WSCl₄ and Cl₂C=CCl₂ forms WSCl₃.²⁷

Green single crystals of WSeCl₄, which were obtained during the synthesis by sublimation, contained two independent molecules in the asymmetric unit, both monomeric square pyramids, with similar bond lengths and angles (Fig. 1). In contrast, WSCl₄ adopts two structurally distinct polymorphs, a dimer, [SCl₃W(μ-Cl)₂WCl₃S],²⁸ and a tetramer, [Cl₄SW(μ-Cl)SCl₂W(μ-Cl)₂WCl₂S(μ-Cl)WSCl₄].²⁹ WSCl₄ monomers are present in WSCl₄·S₈, which seems to be best described as co-crystallised WSCl₄ and cyclo-octasulfur, rather than an adduct.³⁰

Fig. 1 The crystal structure of the two independent molecules of WSeCl₄ in the asymmetric unit showing the atom numbering scheme and with the ellipsoids drawn at the 50% probability level. Selected bond lengths (Å) and angles (°) are: W1–Se1 = 2.2099(4), W1–Cl3 = 2.2944(9), W1–Cl2 = 2.3048(8), W1–Cl4 = 2.3045(8), W1–Cl1 = 2.3038(9), Se1–W1–Cl3 = 101.46(2), Se1–W1–Cl2 = 101.15(2), Se1–W1–Cl4 = 101.98(2), Se1–W1–Cl1 = 101.66(3), Cl3–W1–Cl2 = 87.74(3), Cl3–W1–Cl4 = 87.68(3), Cl1–W1–Cl2 = 88.06(3), Cl1–W1–Cl4 = 87.30(3), W2–Se2 = 2.2128(4), W2–Cl5 = 2.3419(8), W2–Cl7 = 2.2762(9), W2–Cl6 = 2.3949(8), W2–Cl8 = 2.2689(8), Se2–W2–Cl5 = 98.24(2), Se2–W2–Cl7 = 101.18(2), Se2–W2–Cl6 = 97.65(2), Se2–W2–Cl8 = 101.49(2), Cl5–W2–Cl6 = 85.15(3), Cl7–W2–Cl6 = 88.32(3), Cl8–W2–Cl5 = 88.58(3), Cl8–W2–Cl7 = 91.45(3).

Powder X-ray diffraction (PXRD) data from the powdered bulk WSeCl₄ (ESI Fig. S1†) shows the same polymorph as identified in the single crystal structure, with no evidence for a second form. This solid state structure is very similar to that reported for the compound from vapour phase electron diffraction.³¹

The IR spectrum of WSeCl₄ (Nujol mull) shows (ESI. Fig. S2†) a medium feature at 387 cm⁻¹ assigned as ν(W=Se) and intense features at 353 and 367 cm⁻¹ assigned as ν(W–Cl). The ν(W=Se) is in good agreement with literature data, viz. ν(W=Se) 388,³¹ or 396 cm⁻¹,²¹ and is supported by DFT calculations (below). The IR spectrum of WSeCl₄ vapour, isolated in a N₂ matrix at 10 K showed vibrations at 399 and 369 cm⁻¹, both tentatively assigned as ν(W–Cl), while ν(W=Se) was not identified.³² The reassignment of the band at 399 cm⁻¹ as ν(W=Se) seems merited in the light of these new results.

Density functional theory (DFT) calculations

To further explore the structure of WSeCl₄, to investigate the frontier orbitals and to complement the experimental data with IR spectroscopic assignments, the ground state geometry of WSeCl₄ was optimised using DFT methods, starting from the solid-state structure. These calculations were carried out with the B3LYP-D3 functional,³³ (lanl2dz on W; 6-311G(d) on Se and Cl atoms) and the geometrical parameters were found to be in good agreement with the solid-state values (Fig. 2 and ESI Table S2†). The predicted IR harmonic wavenumbers in Table 1 confirm the assignment of the ν(W=Se) peak in the experimental IR spectrum at a slightly higher frequency than ν(W–Cl).

Fig. 2 Schematic showing the DFT calculated structures for (a) WECl₄ and (b) [ECl₃W(μ-Cl)₂WCl₃E] (E = S, Se).

Table 1 Selected experimental and calculated IR stretching and bending vibrations

Bond	Experimental/cm^−1	DFT (B3LYP-D3)	Intensity/km mol^−1
		Adjusted^a wavenumber/cm^−1
a Scaling factor used 0.966 (ref. 34).
W=Se	388	370	25
W–Cl	367
W–Cl	353	354	6
W–Cl	336	333	119

---

As discussed above, the known polymorphs of WSCl₄ in the solid state are a dimer or tetramer, while the crystals obtained of WSeCl₄ show this to be a monomer. To further explore this difference, DFT calculations were undertaken to compare the energy of formation of the optimised selenide and sulfide dimer (with bridging Cl), i.e. the energy difference upon going from 2 WECl₄ to [ECl₃(W(μ-Cl)₂WCl₃E]. This showed the energy of formation of the sulfide-bridged dimer (ΔE) to be more negative (−78.4 kJ mol⁻¹) than the selenide bridged dimer (−76.3 kJ mol⁻¹), supporting the conclusion that for WSCl₄ the dimer is more stable. This is further evidenced by examining the molecular orbitals of WSCl₄ and WSeCl₄ (ESI Fig. S30†), particularly the HOMO and HOMO−1 orbitals, which show more extensive contribution from the chlorides in WSCl₄ compared to the WSeCl₄.

Finally, calculations were undertaken to explore the relative Lewis acidities of the WECl₄ species (E = S, Se) by considering the following reaction:

WECl₄ + OPPh₃ → WECl₄(OPPh₃) (E = S, Se);

where ΔE = E([WSCl₄(OPPh₃)]) − E(OPPh₃) − E(WSCl₄).

These showed that the ΔE for E = S (−161.2 kJ mol⁻¹) was slightly more negative than for E = Se (−156.5 kJ mol⁻¹), consistent with WSCl₄ being the more Lewis acidic of the pair.

Complexes with N- and O-donor ligands

Despite the lower Lewis acidity of WSeCl₄, the syntheses of a range of its complexes with neutral Group 15 and 16 donor ligands was undertaken (Scheme 1), and their properties compared with those of the WOCl₄ and WSCl₄ analogues. The reactions with ligands occurred readily, although the products were often unstable in solution and sometimes also in the solid state. Successful syntheses used carefully chosen (low polarity, weakly-coordinating, volatile) solvents and required rapid isolation of the products. Attempted recrystallisation of the complexes often resulted in significant decomposition rather than purification. The solution spectra also needed to be recorded from freshly prepared solutions to minimise significant decomposition.

Scheme 1 Synthesis routes to WSeCl₄ and its complexes.

The reaction of WSeCl₄ with one molar equivalent of DMF, thf, pyridine, OPPh₃ and 2,2′-bipy, as described in the Experimental section, resulted in green powders [WSeCl₄(L)]; often the choice of solvent was key to obtaining the target complex in a pure form. In contrast, [WSeCl₄(MeCN)] could be obtained from a range of solvents, as well as from neat MeCN, and could be used as a synthon for other complexes, although it proved to be little different from using WSeCl₄ itself in most cases. The bidentates, Ph₂P(O)CH₂P(O)Ph₂ (dppmO₂), 1,4-dioxane and 4,4′-bipyridyl (L–L), reacted with two equivalents of WSeCl₄ to form the ligand-bridged dimers, [(WSeCl₄)₂(μ-L–L)]. With one exception, ([WSeCl₄(2,2′-bipy)] below), the spectroscopic data show the complexes contain six-coordinate distorted octahedral tungsten environments, which was confirmed by X-ray crystal structures of five examples (Fig. 3–7), showing the neutral ligand occupying the vacant site trans to W=Se in WSeCl₄. As is found in other complexes of this type, the tungsten lies out of the Cl₄ plane towards the selenium. The effect of coordination on the W=Se and W–Cl bond lengths is small, with only slightly longer bonds in the six-coordinate complexes. The ¹H and (where appropriate) ³¹P{¹H} NMR spectra confirm the presence of the coordinated neutral ligand, with chemical shifts generally very similar to those in the WOCl₄ and WSCl₄ analogues.⁷ The poor solubility and solution instability generally precluded using ⁷⁷Se NMR spectroscopy (⁷⁷Se: 7.58%, I = ½, Ξ = 19.09 MHz, D_c = 2.98) for their characterisation.

Fig. 3 The structure of [WSeCl₄(OPPh₃)] showing the atom numbering scheme and ellipsoids are shown at the 50% probability level. H atoms are omitted for clarity. Selected bond lengths (Å) and angles (°) are: W1–Se1 = 2.2321(2), W1–Cl1 = 2.3514(5), W1–Cl2 = 2.3577(5), W1–Cl3 = 2.3217(5), W1–Cl4 = 2.3218(5), W1–O1 = 2.1669(14), P1–O1 = 1.5066(15), Se1–W1–Cl1 = 97.836(15), Se1–W1–Cl2 = 94.628(14), Se1–W1–Cl3 = 96.375(15), Se1–W1–Cl4 = 97.047(15), Cl1–W1–Cl2 = 87.552(17), Cl3–W1–Cl2 = 88.861(18), Cl3–W1–Cl4 = 91.204(19), Cl4–W1–Cl1 = 89.494(18), O1–W1–Cl1 = 82.53(4), O1–W1–Cl2 = 84.52(4), O1–W1–Cl3 = 83.23(4), O1–W1–Cl4 = 83.81(4), P1–O1–W1 = 167.82(10).

Fig. 4 The structure of [WSeCl₄(DMF)] showing the atom numbering scheme and ellipsoids are shown at the 50% probability level. H atoms and a C₆H₆ solvent molecule are omitted for clarity. Note that the disorder in the structure was modelled using split atom occupancies and only the major form is shown here. Selected bond lengths (Å) and angles (°) are: W1–Se1 = 2.2478(7), W1–Cl1 = 2.3283(15), W1–Cl2 = 2.3252(15), W1–Cl3 = 2.3422(14), W1–Cl4 = 2.3466(14), W1–O1 = 2.173(4), Se1–W1–Cl1 = 98.79(4), Se1–W1–Cl2 = 98.22(4), Se1–W1–Cl3 = 95.60(4), Se1–W1–Cl4 = 96.22(4), Cl1–W1–Cl4 = 88.69(5), Cl2–W1–Cl1 = 89.50(6), Cl2–W1–Cl3 = 89.44(6), Cl3–W1–Cl4 = 88.75(5), O1–W1–Cl1 = 83.06(12), O1–W1–Cl2 = 81.74(13), O1–W1–Cl3 = 82.55(12), O1–W1–Cl4 = 83.82(13).

Fig. 5 The structure of [WSeCl₄(MeCN)] showing the atom numbering scheme. Ellipsoids are shown at the 50% probability level and H atoms are omitted for clarity. Selected bond lengths (Å) and angles (°) are W1–Se1 = 2.2312(9), W1–Cl1 = 2.3248(17), W1–Cl2 = 2.3208(17), W1–N1 = 2.330(7), Se1–W1–Cl1 = 97.83(4), Se1–W1–Cl2 = 99.32(4), Cl1–W1–Cl1 = 90.30(9), Cl1–W1–N1 = 81.05(13), Cl2–W1–Cl1 = 88.83(6), Cl2–W1–Cl2 = 86.95(9), Cl2–W1–N1 = 81.83(14).

Fig. 6 The structure of [(WSeCl₄)₂(μ-4,4′-bipy)] showing the atom numbering scheme. Ellipsoids are shown at the 50% probability level and H atoms and benzene solvent molecules are omitted for clarity. The asymmetric unit consists of two of these molecules, only one of which is shown. Selected bond lengths (Å) and angles (°) are: W1–Se1 = 2.2424(7), W1–Cl4 = 2.3292(17), W1–Cl1 = 2.3150(16), W1–Cl3 = 2.3113(17), W1–Cl2 = 2.3214(18), W1–N1 = 2.391(5), Se1–W1–Cl4 = 97.47(4), Se1–W1–Cl1 = 98.68(5), Se1–W1–Cl3 = 98.66(5), Se1–W1–Cl2 = 98.08(5), Cl4–W1–N1 = 81.40(14), Cl1–W1–Cl4 = 88.40(6), Cl1–W1–Cl2 = 88.92(7), Cl1–W1–N1 = 81.42(13), Cl3–W1–Cl4 = 88.66(7), Cl3–W1–Cl2 = 89.34(7), Cl3–W1–N1 = 81.22(13), Cl2–W1–N1 = 83.05(14).

Fig. 7 The structure of [{WSeCl₄}₂(1,4-dioxane)] showing the atom numbering scheme. Ellipsoids are shown at the 50% probability level and H atoms are omitted for clarity. Selected bond lengths (Å) and angles (°) are: W1–Se1 = 2.2285(4), W1–Cl1 = 2.3138(8), W1–Cl2 = 2.3273(8), W1–Cl3 = 2.3069(8), W1–Cl4 = 2.3081(8), W1–O1 = 2.398(2), Se1–W1–Cl1 = 98.44(2), Se1–W1–Cl2 = 98.97(2), Se1–W1–Cl3 = 99.60(2), Se1–W1–Cl4 = 98.73(2), Cl1–W1–Cl2 = 88.20(3), Cl1–W1–O1 = 80.83(6), Cl2–W1–O1 = 81.40(6), Cl3–W1–Cl2 = 88.80(3), Cl3–W1–Cl4 = 89.51(3), Cl3–W1–O1 = 81.12(6), Cl4–W1–Cl1 = 87.95(3), Cl4–W1–O1 = 80.88(6).

In the IR spectra, the ν(W–Cl) occurred as two strong overlapping bands in the region 350–310 cm⁻¹ and ν(W=Se) as a weak or medium intensity band in the range 380–350 cm⁻¹, somewhat lower frequency than in WSeCl₄ itself, in accord with the increase in the coordination number.

Once isolated, the [WSeCl₄(2,2′-bipy)] proved to very poorly soluble in suitable solvents, and X-ray quality crystals could not be obtained. However, the ¹H NMR spectrum showed equivalent pyridyl rings, whilst the ν(W–Cl) = 318, 304 cm⁻¹ and ν(W=Se) = 345 cm⁻¹ are lower than in the six-coordinate complexes, providing supporting evidence that [WSeCl₄(2,2′-bipy)] is probably seven-coordinate and pentagonal bipyramidal. The corresponding [WOF₄(2,2′-bipy)], [WOCl₄(2,2′-bipy)] and [WSCl₄(2,2′-bipy)] are also thought to be seven-coordinate based upon spectroscopic data, and [WOF₄{o-C₆H₄(PMe₂)₂}], [WOCl₄{o-C₆H₄(AsMe₂)₂}] and [WSCl₄{o-C₆H₄(PMe₂)₂}] have been established as pentagonal bipyramids by single crystal X-ray analysis.^8,35 Rather surprisingly, attempts to isolate the (unknown) anion, [WSeCl₅]⁻ with a range of cations, including [PPh₄]⁺ and [NEt₄]⁺, have been unsuccessful thus far, whereas the [WSCl₅]⁻ and [WS₂Cl₄]²⁻ ions are well known.^36,37

Reactions with soft P- and Se-donor ligands

The addition of PMe₃ or PEt₃ to a solution of WSeCl₄ in CH₂Cl₂ immediately produced orange-brown solutions in which the only species evident in the ³¹P{¹H} NMR spectra, in addition to small amounts of the free phosphine, were the corresponding phosphine selenides SePMe₃ (δ_P = +9.4)³⁸ and SePEt₃ (δ_P = +45.2, ¹J_PSe = 680 Hz).³⁹ The reaction of WECl₄ (E = O, S, Se) with a large excess of PPh₃ over several days in toluene was reported to form [WCl₄(PPh₃)₂] and EPPh₃.⁴⁰ Similar, but much faster abstraction of the selenium occurs on contact with alkyl phosphines. Similarly, reaction of o-C₆H₄(PMe₂)₂ with WSeCl₄ in CH₂Cl₂ instantly produced a green solution and a brown precipitate, and the ³¹P{¹H} NMR spectrum showed several species present in the solution, with the major resonances at δ = +20 and +33, which may be due to phosphine selenides. This contrasts with the reactions of o-C₆H₄(PMe₂)₂ with WOCl₄ and WSCl₄, which gave high yields of the seven-coordinate complexes [WECl₄{o-C₆H₄(PMe₂)₂}] (E = O, S), both of which were characterised, including the X-ray structure of the latter (as noted earlier).⁸ Decomposition of [WECl₄{o-C₆H₄(PMe₂)₂}] in CH₂Cl₂ solution over a few days gave crystals of a number of tungsten(V) species including [WOCl₃{o-C₆H₄(PMe₂)₂}] and [WCl₄{o-C₆H₄(PMe₂)₂}₂]⁺.⁸ This contrasts with the immediate reduction of the metal in the WSeCl₄ systems in the present study. A few crystals obtained from the mother liquor of the o-C₆H₄(PMe₂)₂/WSeCl₄ reaction were identified by X-ray analysis as [WCl₄{o-C₆H₄(PMe₂)₂}₂]Cl; the same eight-coordinate cation was found among the decomposition products of [WSCl₄{o-C₆H₄(PMe₂)₂}] in CH₂Cl₂.⁸

The reaction of WSeCl₄ with SeMe₂ in benzene resulted in green [WSeCl₄(SeMe₂)], identified by microanalysis and its IR spectrum, which showed ν(W=Se) at 387 cm⁻¹ and ν(W–Cl) at 343 cm⁻¹. Use of SeⁿBu₂ in this reaction gave a dark red oil identified as [WSeCl₄(SeⁿBu₂)] from its IR, ¹H and ⁷⁷Se{¹H} NMR spectra, but attempts to isolate complexes with diselenoethers such as [(WSeCl₄)₂{μ-RSe(CH₂)₂SeR}] were unsuccessful. Similarly, attempts to prepare [(WSCl₄)₂{μ-MeSe(CH₂)₂SeMe}] were unsuccessful, although the [WSCl₄(SeMe₂)] analogue is well characterised.⁹ A few dark green crystals isolated from a solution of the [WSeCl₄(SeMe₂)] monomer in C₆D₆ unexpectedly proved to be the dinuclear [(WSeCl₄)₂(μ-SeMe₂)] (Fig. 8), which shows two square pyramidal WSeCl₄ groups linked by a bridging SeMe₂ ligand. The d(W=Se) and d(W–Cl) are similar to those in the complexes with hard donor ligands (above), while the d(W–Se_Me2) is long at ∼3.0 Å. Selenoethers bridging between two metal centres and formally using both lone pairs are known for many d-block metals,⁴¹ although it was unexpected here given the modest acceptor ability of WSeCl₄. This dimer presumably crystallised as the least soluble species in a partially dissociated solution of [WSeCl₄(SeMe₂)].

Fig. 8 The structure of [(WSeCl₄)₂(μ-SeMe₂)] showing the atom numbering scheme. Ellipsoids are shown at the 50% probability level and H atoms are omitted for clarity. Selected bond lengths (Å) and angles (°) are W1–Se1 = 2.2374(8), W2–Se3 = 2.2255(9), W1–Se2 = 3.0107(8), W2–Se2 = 3.0814(8), W1–Cl = 2.2871(19)–2.3311(19), W2–Cl = 2.3006(18)–2.3130(18), Se1–W1–Se2 = 178.42(3), Se3–W2–Se2 = 176.28(3), W1–Se2–W2 = 127.82(3), Se1–W1–Cl = 98.42(5) – 100.95(5), Se2–W1–Cl = 80.03(5) – 81.91(5), Se3–W2–Cl = 98.86(6) – 101.89(6), Se2–W2–Cl = 76.67(5) – 81.78(5).

Similar reaction of WSeCl₄ with SMe₂ did not yield a pure product, however, a few crystals obtained by slow evaporation of CH₂Cl₂ from the reaction mixture were shown by X-ray crystallographic analysis to contain the W(V) dimer, [{WCl₃(SMe₂)}₂(μ-Se)(μ-Se₂)] (Fig. 9). Previous work showed that decomposition of several WSCl₄ thioether complexes (or reduction with excess thioether) affords the corresponding W(V) complexes.^5,9 However, decomposition of the WSeCl₄ complexes prepared in the present study typically produced dark intractable solids, that were insoluble in organic solvents and could not be identified. Structurally characterised niobium(IV) complexes similar to [Cl₃(SMe₂)W(μ-Se)(μ-Se₂)WCl₃(SMe₂)] include [(tht)Br₂Nb(μ-Se)(μ-Se₂)NbBr₂(tht)] (tht = tetrahydrothiophene) and [(SMe₂)Cl₂Nb(μ-Se₂)₂NbCl₂(SMe₂)].^5,21,42,43 In fact, very little is known about WSeCl₃ and no complexes with neutral ligands have been described.⁵

Fig. 9 The structure of [{WCl₃(SMe₂)}₂(μ-Se)(μ-Se₂)] showing the atom numbering scheme. Ellipsoids are shown at the 50% probability level and H atoms are omitted for clarity. Selected bond lengths (Å) and angles (°): W1–W1 = 2.8021(6), W1–Se1 = 2.4310(9), W1–Se2 = 2.5687(9), W1–Se3 = 2.5589(9), Se2–Se3 = 2.2633(14), W1–Cl1 = 2.4259(16), W1–Cl2 = 2.3944(16), W1–Cl3 = 2.3791(16), W1–S1 = 2.6017(18), W1–Se1–W1 = 70.39(3), W1–Se2–W1 = 66.11(3), W1–Se3–W1 = 66.40(3), Cl1–W1–S1 = 71.16(6), Cl2–W1–S1 = 76.35(6), Cl3–W1–S1 = 82.26(6), Cl3–W1–Cl1 = 153.31(6), Cl3–W1–Cl2 = 88.76(6), Cl2–W1–Cl1 = 87.33(6).

Growth of WSe₂ thin films via low pressure CVD

The isolation of [WSeCl₄(SeⁿBu₂)] as an oil, containing a 1 : 2 ratio of tungsten to selenium, and with the n-butyl groups on the selenoether providing a potential low energy thermal decomposition pathway (e.g. via β-hydride elimination), raised the prospect that this might serve as a single source CVD precursor for the growth of tungsten diselenide thin films onto fused silica substrates, in a similar manner to [(WSCl₄)₂{μ-ⁱPrS(CH₂)₂SⁱPr}], which deposits WS₂ films.⁹ LPCVD using [WSeCl₄(SeⁿBu₂)] produced highly reflective, silvery films of WSe₂ at 0.1 mmHg and 660–700 °C.

Grazing incidence XRD analysis of the films obtained from [WSeCl₄(SeⁿBu₂)] (Fig. 10 and S29†) are consistent with polycrystalline 4H WSe₂ in space group P6₃/mmc. Lattice parameters were refined as: a = 3.2829(11), c = 13.130(6) (R_wp = 7.82%, literature: a = 3.285(1); c = 12.961(1) Å from the XRD pattern in Fig. S29a†).⁴⁴ The grazing incidence diffraction patterns of the thinner films, for example, Fig. 10, are almost completely dominated by the 002 reflection, consistent with significant preferred orientation in the <00l> direction. This orientation is not unusual, as these layered materials typically grow initially with the c axis normal to the substrate, and suggests that the majority of the platelet crystals have grown in that orientation, with the platelets flat to the substrate. The average crystallite size in the WSe₂ film was determined to be 9.4(14) nm based on the grazing incidence XRD data in Fig. S29a,† using the Williamson-Hall method.

Fig. 10 Grazing incidence XRD pattern (top) from a WSe₂ thin film deposited by low pressure CVD using [WSeCl₄(SeⁿBu₂)]. The broad feature at 2θ = 20–25° is from the SiO₂ substrate. XRD pattern for bulk WSe₂ (bottom).⁴⁴

Fig. 11 presents top-view SEM images of the as-deposited WSe₂ thin film. The film displays a smooth underlying interconnected polycrystalline layer with some more randomly oriented crystallites on top. The underlying layer comprises of hexagonal plate-like crystallites oriented with the c-axis normal to the substrate. This is supported by the preferential growth observed in the XRD patterns in Fig. 10 and Fig. S29b.†

Fig. 11 Morphological characterisation via SEM imaging for an as-deposited WSe₂ film produced via low pressure CVD using [WSeCl₄(SeⁿBu₂)], taken at magnifications (a) ×10 000 and (b) ×40 000.

The Raman spectrum for the WSe₂ film deposited via LPCVD is presented in Fig. 12. The two significant peaks located at 248.7 cm⁻¹ and 258.5 cm⁻¹ are commonly known as the E_2g¹ (green) and A_1g (orange) vibrational modes, respectively, and are in good agreement with current literature.⁴⁵ The inset is a schematic of the E_2g¹ and the A_1g modes, which represent the in-plane and out-of-plane vibrational modes, respectively.⁴⁶ The relative intensities of the E_2g¹ and the A_1g modes suggest the suppression of the out-of-plane A_1g mode being due to the thin nature of the film.^46,47 The peaks labelled 2M are likely to be related to the combinational second order Raman modes.^46–48 The origin of peak x located at 307.3 cm⁻¹ is likely to be a combinational mode of low frequency modes due to its relatively large frequency. It has been suggested that this peak is linked to coupling of the E_2g¹ mode and an intra-layer shear force, due to reports that suggest suppression of this peak in monolayered samples.^46,49,50

Fig. 12 Raman spectrum of the WSe₂ film obtained using [WSeCl₄(SeⁿBu₂)].

The film composition was investigated by X-ray photoelectron spectroscopy (XPS). Fig. 13a presents the XPS survey spectrum of the film. Both the W and Se peaks are clearly presented, with no impurities (e.g. C, Cl) observed, suggesting the WSe₂ film is of high quality. Elemental scans were conducted of both the W 4f and Se 3d orbitals (shown in Fig. 13b–c). From the elemental scan of the W 4f orbital, two significant peaks positioned at 32.2 eV and 34.3 eV were observed. These peaks are in good agreement with literature and represent electron emission from the 4f_7/2 and 4f_5/2 atomic orbitals.⁵¹ An additional peak located at 37.5 eV was noted in the W scan, which has been suggested to be related to W 5p_3/2.⁵² The Se scan in Fig. 13 displays two significant peaks at 54.6 eV and 55.7 eV related to the 3d_5/2 and 3d_3/2 orbitals respectively.⁵² From the spectra shown in Fig. 13, a WSe_2.1 composition for the deposited films has been derived (i.e. W : Se = 32.1% : 67.9%).

Fig. 13 XPS characterisation of an as-deposited WSe₂ film produced via low pressure CVD using [WSeCl₄(SeⁿBu₂)]. (a) XPS survey scan of binding energies in the range of 0 eV to 650 eV, with peaks related to the atomic orbitals of W (blue) and Se (orange) labelled. Elemental scans for the as-deposited WSe₂ film showing the peaks associated with W (b) and Se (c).

Experimental

Syntheses were performed using standard Schlenk and glovebox techniques under a dry N₂ atmosphere. Solvents and ligands were dried by distillation from CaH₂ (CH₂Cl₂, MeCN, DMF,) or Na/benzophenone ketyl (toluene, n-hexane, diethyl ether, thf, 1,4-dioxane, py). WCl₆, PMe₃, PEt₃, SMe₂, SeMe₂, 2,2′-bipy and 4,4′-bipy were obtained from Sigma-Aldrich. Ph₂P(O)CH₂P(O)Ph₂ was obtained by air oxidation of the corresponding diphosphine in CH₂Cl₂, catalysed by SnI₄.⁵² Solid ligands were dried in vacuo for several hours immediately before use.

Infrared spectra were recorded on a PerkinElmer Spectrum 100 spectrometer in the range 4000–200 cm⁻¹, with samples prepared as Nujol mulls between CsI plates. ¹H and ³¹P{¹H} NMR spectra were recorded using a Bruker AV 400 spectrometer and referenced to the residual protio-resonance of the solvent and 85% H₃PO₄. Microanalyses on new compounds were undertaken by London Metropolitan University or Medac Ltd. Note that the complexes were generally very unstable, readily decomposing within a few hours in some cases. This hindered attempts to obtain satisfactory microanalyses on some samples from outsourced measurements – as noted below.

X-ray experimental

Crystals were grown from slow evaporation of saturated solutions in CH₂Cl₂ or by liquid–liquid diffusion using CH₂Cl₂ and hexane. Data collections used either a Rigaku OD UG2 goniometer equipped with a HyPix 6000HE hybrid pixel array detector mounted at the window of an FR-E + SuperBright molybdenum (λ = 0.71073 Å) rotating anode generator with VHF Varimax optics (70 micron focus) with the crystal held at 100 K (N₂ cryostream), or a Rigaku OD UG2 goniometer equipped with a HyPix 6000HE hybrid pixel array detector mounted at the window of a MicroMax 007HF copper (λ = 1.54184 Å) rotating anode generator with VHF Varimax optics (70 micron focus) with the crystal held at 100 K (N₂ cryostream). Crystallographic parameters are presented in Table S1.† Structure solution and refinement were performed using SHELX(T)-2018/2, SHELX-2018/3 through Olex2⁵³ and were mostly straightforward. H atoms were added and refined with a riding model. Where additional restraints were required, details are provided in the cif file for each structure found on CCDC.†

Computational experimental

Geometry optimisation and frequency calculations were performed on WSeCl₄ with the B3LYP hybrid functional, with the Grimme correction for dispersion (D3(BJ)),³³ using Gaussian 16.⁵⁴ A 6-311G(d) basis set was used on Se and Cl atoms while the RECP basis set LANL2DZ was used on the W atom. Calculations on WSCl₄ were based on the crystal structure for WSeCl₄, while the starting geometries for the dimers ([ECl₃W(μ-Cl)₂WCl₃E]) (E = S, Se) were based on the structure for E = S reported in the literature.²⁸

For the complexes, the starting geometries for [WSeCl₄(OPPh₃)] and [WSCl₄(OPPh₃)]⁹ came from their respective crystal structures with the same functional and basis set as used for WSeCl₄. Energy minima were confirmed by the absence of any imaginary frequencies using a frequency calculation and calculated frequencies were adjusted by standard scaling factors.³⁴

Following the method of Britnell et al.,²¹ freshly ground WCl₆ (6.19 g, 15.6 mmol) and Sb₂Se₃ (2.5 g, 5.2 mmol) were mixed in a glove box and added to a glass reaction tube (Fig. 14) under a dinitrogen atmosphere. The glass tube was then placed under vacuum and sealed with a gas torch at point (1). Section A was placed in a tube furnace at 120 °C for 4 days, with section C at room temperature, to allow the side products to deposit in the cooler section. This was then removed by sealing the glass at point (2). Section A was placed back in the furnace, with section B at room temperature for 4 days until more side product had collected in section B. At this point, joint (3) was broken in a glovebox and the solid remaining in section A removed and ground up, and the resulting green powder placed in a glass tube and heated at 170 °C for 6 h in vacuo. Any remaining SbCl₃ pumped away and a grey solid had sublimed in the glass tube, leaving a brown residue which was discarded. The grey solid was removed in a glovebox and ground up to afford a bright green solid. Yield: 4.3 g, 68%. IR spectrum (Nujol, cm⁻¹): 377 W=Se, 367, 353, 336 W–Cl.

Fig. 14 (a) A borosilicate glass tube with three 1 cm constrictions, and a B14 joint, used for the solid-state reaction of WCl₆ and Sb₂Se₃ to form WSeCl₄; (b) grey-green solid remaining in the reaction tube after heating.

[WSeCl₄(MeCN)]. MeCN (0.3 mL) in CS₂ (2 mL) was added to a green suspension of WSeCl₄ (0.150 g, 0.37 mmol) in CS₂ (5 mL). The resulting bright green suspension was stirred for 20 min and concentrated to 50% volume before the mother liquor was removed by filtration, to obtain a green powder, which was dried in vacuo. Yield: 0.125 g, 76%. Required for C₂H₃Cl₄NSeW (445.66): C, 5.39; H, 0.68; N, 3.14. Found: C, 5.51; H, 1.03; N, 2.90%. IR spectrum (Nujol, cm⁻¹): 2308 C=N, 2281, ν(C–C) + δ(CH₃), 373 W=Se, 347, 332 W–Cl. ¹H NMR (CD₂Cl₂): δ = 2.30 (s). The same complex can be obtained using CH₂Cl₂ or C₆H₆ as solvent or neat MeCN.

[WSeCl₄(THF)]. THF (0.01 mL, 0.14 mmol) was added to a stirring solution of WSeCl₄ (0.060 g, 0.14 mmol) in benzene (2 mL). The resulting green solution was frozen in liquid nitrogen, then allowed to warm to ca. −5 °C and the benzene was sublimed under vacuum to leave a green solid. Yield: 0.045 g, 67%. Required for C₄H₈Cl₄OSeW (476.72): C, 10.08; H 1.69. Found: C, 10.75; H, 1.80%. IR spectrum (Nujol, cm⁻¹): 1089, 846 thf, 373 W=Se, 344, 330 W–Cl. ¹H NMR (CD₂Cl₂): δ = 4.08 (s, [4H]), 1.96 (s, [4H]).

[WSeCl₄(OPPh₃)]. A solution of OPPh₃ (0.069 g, 0.2 mmol) in dichloromethane (5 mL) was added slowly to a suspension of WSeCl₄ (0.100 g, 0.2 mmol) in dichloromethane (5 mL). The resulting green solution was then stirred for 30 min, concentrated to ∼1 mL in vacuo, filtered, and the resulting green solid washed with hexane (2 × 1 mL), and dried in vacuo. Yield: 0.105 g, 76%. Required for C₁₈H₁₅Cl₄OPSeW (682.9): C, 31.66; H, 2.21. Found: C, 31.86; H, 2.15%. IR spectrum (Nujol, cm⁻¹): 1131 P=O, 360 W=Se, 337, 309 W–Cl. ¹H NMR (CD₂Cl₂): δ = 7.77 (m, [6H]), 7.67 (m, [3H]), 7.63 (m, [6H]). ³¹P{¹H} NMR (CD₂Cl₂): δ = +45.7 (s). The same complex was obtained using C₆H₆, toluene or CS₂ as solvent or by reaction of OPPh₃ with [WSeCl₄(MeCN)] in a 1 : 1 molar ratio.

[WSeCl₄(DMF)]. DMF (0.019 mL, 0.25 mmol) was added to a stirring solution of WSeCl₄ (0.100 g, 0.25 mmol) in benzene (2 mL). The resulting green solution was frozen in liquid nitrogen, before being allowed to warm to ca. −5 °C under dynamic vacuum to remove the benzene via sublimation to afford a green solid. The complex decomposes in a few hours even in the glove box, therefore a satisfactory microanalysis could not be obtained, and it was identified by a single crystal X-ray structure determination. Yield: 0.080 g, 66%. IR spectrum (Nujol, cm⁻¹): 1644 C=O, 388 W=Se, 367, 331 W–Cl. ¹H NMR (C₆D₆): δ = 7.57 (s, [H]), 1.89 (s, [3H]), 1.35 (s, [3H]).

[WSeCl₄(2,2′-bipy)]. 2,2′-Bipy (0.039 g, 0.25 mmol) in toluene (3 mL) was added slowly to a purple suspension of WSeCl₄ (0.100 g, 0.25 mmol) in toluene (4 mL). The red suspension was stirred for 10 min, before the colourless solution was filtered off the red/brown solid, which was washed in CH₂Cl₂ (1 mL) and dried in vacuo. Yield: 0.090 g, 56%. Required for C₁₀H₈Cl₄N₂SeW·CH₂Cl₂ (645.73): C, 20.46; H, 1.56; N, 4.34. Found: C, 20.36; H, 1.61; N, 5.50%. IR spectrum (Nujol, cm⁻¹): 341 W=Se, 313 W–Cl. ¹H NMR (CD₂Cl₂): δ = 8.92 (d, [2H], J_HH = 4.7 Hz), 8.89 (d, [2H], J_HH = 8.2 Hz), 8.28 (dt, [2H], J_HH = 1.7, 7.8 Hz), 7.75 (dt, [2H], J_HH = 1.0, 6.8 Hz).

[WSeCl₄(NC₅H₅)]. Pyridine (0.020 g, 0.25 mmol) was added slowly to a purple suspension of WSeCl₄ (0.100 g, 0.25 mmol) in CH₂Cl₂ (5 mL). The resulting green suspension was stirred for 10 min before the volatiles were removed in vacuo, to afford a green-brown solid. Yield: 0.076 g, 63%. Required for C₅H₅Cl₄NSeW (483.71): C, 12.42; H, 1.04; N, 2.90. Found C, 13.40; H, 1.54; N, 2.96%; see comment above regarding instability of the complexes. IR spectrum (Nujol, cm⁻¹): 364 W=Se, 327 W–Cl. ¹H NMR (CD₂Cl₂): δ = 8.65 (br [2H]), 8.25 (br, [1H]), 7.82 (br, [2H]).

[(WSeCl₄)₂(μ-4,4′bipy)]. A solution of 4,4′-bipy (0.013 g, 0.08 mmol) in CH₂Cl₂ (2 mL) was added to a stirring solution of [WSeCl₄(MeCN)] (0.075 g, 0.16 mmol) in CH₂Cl₂ (3 mL). The green suspension was stirred for 5 min before concentrating to approx. half the volume. The brown/green solid was isolated by filtration and dried in vacuo. Yield: 0.035 g, 45%. Required for C₁₀H₈Cl₈N₂Se₂W₂ (965.41): C, 12.44; H, 0.81, N, 2.90. Found: C, 12.23; H, 2.00, N, 2.74%. IR spectrum (Nujol, cm⁻¹): 365 W=Se, 333 W–Cl. ¹H NMR (CD₂Cl₂): δ = 9.20 (m, [4H]), 7.74 (m, [4H]).

[(WSeCl₄)₂(1,4-dioxane)]. 1,4 dioxane (0.01 mL, 0.13 mmol) was added to a stirring solution of WSeCl₄ (0.100 g, 0.25 mmol) in CH₂Cl₂ (2 mL). The green solution was stirred for 5 min before the volatiles were removed under vacuum to afford a green solid. Yield: 0.082 g, 70%. Required for C₄H₈Cl₈O₂Se₂W₂ (897.33): C, 5.35; H, 0.90. Found: C, 4.50; H, 1.15%. IR spectrum (Nujol, cm⁻¹): 377 W=Se, 361, 333 W–Cl. ¹H NMR (C₆D₆): δ = 3.47 (s).

[(WSeCl₄)₂(μ-dppmO₂)]. A solution of dppmO₂ (0.053g, 0.125 mmol) in CH₂Cl₂ (3 mL) was added to a stirring suspension of WSeCl₄ (0.100 g, 0.25 mmol) in CH₂Cl₂ (3 mL) to give a green-brown suspension. After 5 min the brown solution was filtered away and the remaining green solid was dried in vacuo. Yield: 0.083 g, 54%. Required for C₂₅H₂₂Cl₈O₂P₂Se₂W₂ (1225.61): C, 24.5; H, 1.81. Found: C, 25.5; H, 1.94%. IR spectrum (Nujol, cm⁻¹): 1074 P=O, 366 W=Se, 338, 337 W–Cl. ¹H NMR (C₆D₆): δ = 7.80 (m, [8H]), 7.61 (m, [4H]), 6.90 (m, [8H]), 4.64 (t, [2H], ²J_H–P = 15.6 Hz). ³¹P{¹H} NMR (C₆D₆): δ = +45.9 (s). The product was very poorly soluble and the dilute solution showed significant decomposition, therefore solution data were collected in situ, but also show some decomposition.

[WSeCl₄(SeMe₂)]. A solution of SeMe₂ (0.016 g, 0.15 mmol) in benzene (1 mL) was added to a stirring solution of WSeCl₄ (0.060 g, 0.15 mmol) in benzene (2 mL) forming a green/black suspension. After 5 min the solvent was removed in vacuo to give a red/brown sticky solid. Yield: 0.035 g, 46%. Required for C₂H₆Cl₄Se₂W (513.64): C, 4.68; H, 1.18. Found: C, 4.67; H, 1.42%. IR spectrum (Nujol, cm⁻¹): 371 W=Se, 344 W–Cl. ¹H NMR (CD₂Cl₂): 1.41 (br s, [6H]).

[WSeCl₄(SeⁿBu₂)]. A solution of SeⁿBu₂ (0.047 g 0.24 mmol) in benzene (1 mL) was added to a stirring solution of WSeCl₄ (0.1 g, 0.24 mmol) in benzene (2 mL) forming a green/black suspension. After 5 min the solvent was removed in vacuo to give a viscous red oil, which was not sufficiently stable to obtain (out-sourced) microanalytical measurements. IR spectrum (Nujol, cm⁻¹): 380 W=Se, 352, 321 W–Cl. ¹H NMR (in situ in C₆D₆): 2.48 (t, J = 7.6 Hz, [2H] CH₂), 1.46 (m, [2H], CH₂), 1.17 (m, [2H] CH₂), 0.73 (t, J = 7.3 Hz, [3], CH₃). ⁷⁷Se{¹H} NMR (CD₂Cl₂, 295 K): δ = 292; (183 K): δ = 295.

LPCVD growth of WSe₂ films using [WSeCl₄(SeⁿBu₂)]

The precursor (50–60 mg) was loaded into the precursor bulb at the closed end of a silica tube in a N₂ purged glove box, silica substrates (ca. 1 × 8 × 20 mm³) were then positioned end-to-end lengthways along the tube away from the precursor. The tube was then set horizontally in a furnace so that the substrates were in the hot zone and the precursor protruded ca. 4 cm beyond the end of the tube furnace. The tube was evacuated to 0.1 mm Hg and the furnace heated to 700 °C and left for 15 min to allow the temperature to stabilise. The tube was gradually moved towards the hot zone until evaporation of the precursor began and the position was maintained until no further evaporation occurred, leaving a small amount of dark sticky residue in the precursor bulb. After ca. 10 min the tube was cooled to room temperature and the substrates were removed and stored for characterisation. Uniformly shiny silver films were observed on the substrates from the hot zone at temperatures between 650–700 °C (determined by temperature profiling), i.e. 8–18 cm away from the precursor.

Film characterisation

X-ray diffraction (XRD) patterns were collected in grazing incidence mode (θ₁ = 1°) using a Rigaku SmartLab diffractometer (Cu-K_α, λ = 1.5418 Å) with parallel X-ray beam and a Hypix detector operated in 1D mode. Phase matching and lattice parameter calculations (WS₂) used the PDXL2 software package⁵⁵ and diffraction patterns from ICSD.⁵⁶ High resolution scanning electron microscopy (SEM) measurements were carried out with a field emission SEM (Jeol JSM 7500F) at an accelerating voltage of 2 kV. X-ray photoelectron spectroscopy (XPS) data were obtained using a ThermoScientific Theta Probe system with Al-K_α radiation (photon energy = 1486.6 eV). All peaks are calibrated against the adventitious C 1s peak at 284.6 eV. Raman spectra of the deposited films were measured at room temperature on a Renishaw InVia Micro Raman Spectrometer using 532 nm excitation. The incident laser power was adjusted to 0.1 mW for all samples.

Conclusions

Single crystal and powder XRD data for WSeCl₄ reveal discrete square pyramidal monomers, which contrasts with the weakly chloride bridged dimer and tetramer units in WSCl₄, while WOCl₄ is a strongly oxido-bridged polymer.^3,57 The weaker Lewis acidity of WSeCl₄ that this implies, is supported by DFT calculations and is also manifest in its coordination chemistry with neutral ligands. The synthesis and spectroscopic characterisation of the first extended series of complexes with neutral N-, O- and Se- donor ligands are described, which, with the exception of the seven-coordinate [WSeCl₄(2,2′-bipy)], contain six-coordinate tungsten with the neutral donor trans to the W=Se. The structures were confirmed by six single crystal X-ray studies, the first on WSeCl₄ complexes. In contrast to the behaviour with WSCl₄ and WOCl₄,⁸ alkyl phosphines abstract Se from WSeCl₄ upon contact. The complexes of WSeCl₄ are less robust both in solution and in the solid state compared to those of WSCl₄ and WOCl₄, with some decomposing rapidly in solution.

Notably, despite its limited solution stability, [WSeCl₄(SeⁿBu₂)] has been shown to function as a very effective LPCVD reagent for the deposition of thin films of the key WSe₂ semiconductor; the first example of a single source precursor for this technologically important material. The resulting films were characterised by grazing incidence X-ray diffraction and scanning electron microscopy, both of which indicate very strong preferred growth in the <00l> direction, together with XPS and Raman spectroscopy. Further work focusing on optimising the film deposition and to explore the electrical properties of these CVD grown films is underway.

In contrast to the behaviour of [WSCl₄(L)] analogues, the [WSeCl₄(L)] generally do not lead to complexes of W(V), [WSeCl₃(L)], on standing either at ambient temperature, by heating in solution, or by reaction with excess L. A few crystals of one W(V) example, [{WCl₃(SMe₂)}₂(μ-Se)(μ-Se₂)], were isolated from the attempted preparation of [WSeCl₄(SMe₂)].

Conflicts of interest

The authors have no conflicts to declare.

Acknowledgements

We thank EPSRC for support through EP/P025137/1 and the ADEPT Programme Grant (EP/N035437/1). We would also like to thank Prof. J. M. Dyke for helpful discussions regarding the DFT calculations reported herein. We wish to acknowledge the use of the EPSRC funded Physical Sciences Data-science Service hosted by the University of Southampton and STFC under grant number EP/S020357/1.

References

1. Comprehensive Coordination Chemistry, ed. G. Wilkinson, R. D. Gillard and J. A. McCleverty, Pergamon, Oxford, 1987, vol. 3.
2. Comprehensive Coordination Chemistry II, ed. J. A. McCleverty and T. J. Meyer, Elsevier, Oxford, 2003, vol. 4.
3. R. A. Walton, Prog. Inorg. Chem., 1972, 16, 1.
4. D. A. Rice, Coord. Chem. Rev., 1978, 25, 199.
5. V. K. Greenacre, W. Levason, G. Reid and D. E. Smith, Coord. Chem. Rev., 2020, 424, 213512.
6. W. Levason, F. M. Monzittu and G. Reid, Coord. Chem. Rev., 2019, 391, 90.
7. V. K. Greenacre, A. L. Hector, W. Levason, G. Reid, D. E. Smith and L. Sutcliffe, Polyhedron, 2019, 162, 14.
8. W. Levason, G. Reid, D. E. Smith and W. Zhang, Polyhedron, 2020, 179, 114372.
9. D. E. Smith, V. K. Greenacre, A. L. Hector, W. Levason, G. Reid, F. Robinson and S. Thomas, Dalton Trans., 2020, 49, 2496.
10. R. D. Bannister, W. Levason, M. E. Light, G. Reid and W. Zhang, Polyhedron, 2019, 167, 1.
11. R. D. Bannister, W. Levason, G. Reid and F. Robinson, Polyhedron, 2019, 169, 129.
12. Y.-P. Chang, A. L. Hector, W. Levason and G. Reid, Dalton Trans., 2017, 46, 9824.
13. Chemical Vapour Deposition: Precursors, Processes and Applications, ed. A. C. Jones and M. L. Hitchman, The Royal Society of Chemistry, 2009.
14. M. Chhowalla, Z. Liu and H. Zhang, Chem. Soc. Rev., 2015, 44, 2584 (ed. – themed issue on metal dichalcogenides).
15. M. Chhowalla, H. S. Shin, G. Eda, L.-J. Li, K. P. Loh and H. Zhang, Nat. Chem., 2013, 5, 263; J. Liu and X.-W. Liu, Adv. Mater., 2012, 24, 4097.
16. J. R. Brent, N. Savjani and P. O'Brien, Prog. Mater. Sci., 2017, 89, 411.
17. Q. Xiang, J. Yu and M. Jaroniec, J. Am. Chem. Soc., 2012, 134, 6575; K. Lee, R. Gatensby, N. McEvoy, T. Hallam and G. S. Duesberg, Adv. Mater., 2013, 25, 6699; K. Xu, P. Chen, X. Li, C. Wu, Y. Guo, J. Zhao, X. Wu and Y. Xie, Angew. Chem., Int. Ed., 2013, 52, 10477; Z. Yan, C. Jiang, T. R. Pope, C. F. Tsang, J. L. Stickney, P. Goli, J. Renteria, T. T. Salguero and A. A. Balandin, J. Appl. Phys., 2013, 114, 20430.
18. S. L. Benjamin, C. H. de Groot, C. Gurnani, A. L. Hector, R. Huang, K. Ignatyev, W. Levason, S. J. Pearce, F. Thomas and G. Reid, Chem. Mater., 2013, 25, 4719.
19. M. D. Khan, M. A. Malik and N. Revaprasadu, Coord. Chem. Rev., 2019, 388, 24.
20. V. Brune, M. Grisch, R. Weißing, F. Hartle, M. Frank, S. Mishra and S. Mathur, Dalton Trans., 2021, 50, 12365.
21. D. Britnell, G. W. A. Fowles and D. A. Rice, J. Chem. Soc., Dalton Trans., 1974, 2191.
22. M. G. B. Drew, E. M. Page and D. A. Rice, J. Chem. Soc., Dalton Trans., 1983, 61.
23. D. Britnell, G. W. A. Fowles and D. A. Rice, J. Chem. Soc., Dalton Trans., 1975, 213.
24. G. W. A. Fowles, D. A. Rice and K. J. Shanton, J. Chem. Soc., Dalton Trans., 1978, 1658.
25. N. D. Boscher, C. J. Carmalt and I. P. Parkin, J. Mater. Chem., 2006, 16, 122.
26. V. C. Gibson, T. P. Kee and A. Shaw, Polyhedron, 1990, 9, 2293.
27. U. Müller and P. Klingelhöfer, Z. Anorg. Allg. Chem., 1984, 510, 109.
28. M. G. B. Drew and R. Mandyczewky, J. Chem. Soc. A, 1970, 2815.
29. F. A. Cotton, P. A. Kibala and R. B. W. Sandor, Inorg. Chem., 1989, 28, 2485.
30. D. L. Hughes, J. D. Lane and R. L. Richards, J. Chem. Soc., Dalton Trans., 1991, 1627.
31. E. M. Page, D. A. Rice, K. Hagen, L. Hedberg and K. Hedberg, Inorg. Chem., 1982, 21, 3280.
32. P. J. Jones, W. Levason, J. S. Ogden, J. W. Turff, E. M. Page and D. A. Rice, J. Chem. Soc., Dalton Trans., 1983, 2625.
33. (a) S. Grimme, J. Comput. Chem., 2006, 27, 1787; (b) S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem., 2011, 32, 1456; (c) C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785; (d) P. J. Stephens, F. J. Devlin, C. F. Chabalowski and M. J. Frisch, J. Phys. Chem., 1994, 98, 11623.
34. From https://cccbdb.nist.gov/vibscalejust.asp, Computational Chemistry Comparison and Benchmark Database.
35. J. W. Emsley, W. Levason, G. Reid, W. Zhang and G. De Luca, J. Fluorine Chem., 2017, 197, 74.
36. M. G. B. Drew, G. W. A. Fowles, E. M. Page and D. A. Rice, J. Chem. Soc., Dalton Trans., 1981, 2409.
37. P. Klingelhöfer and U. Müller, Z. Anorg. Allg. Chem., 1988, 536, 70.
38. W. McFarlane and D. S. Rycroft, J. Chem. Soc., Dalton Trans., 1973, 2162.
39. N. Muller, P. C. Lauterbur and J. Goldenson, J. Am. Chem. Soc., 1956, 78, 3556.
40. M. G. B. Drew, E. M. Page and D. A. Rice, J. Chem. Soc., Dalton Trans., 1983, 61.
41. W. Levason, S. D. Orchard and G. Reid, Coord. Chem. Rev., 2002, 225, 159.
42. M. G. B. Drew, D. A. Rice and D. M. Williams, J. Chem. Soc., Dalton Trans., 1983, 2251.
43. M. G. B. Drew, D. A. Rice and D. M. Williams, J. Chem. Soc., Dalton Trans., 1984, 1087.
44. J. A. Champion, Br. J. Appl. Phys., 1965, 16, 1035.
45. W. Zhao, Z. Ghorannevis, K. K. Amara, J. R. Pang, M. Toh, X. Zhang, C. Kloc, P. H. Tan and G. Eda, Nanoscale, 2013, 5, 9677.
46. H. Zeng, G.-B. Liu, J. Dai, Y. Yan, B. Zhu, R. He, L. Xie, S. Xu, X. Chen, W. Yao and X. Cui, Sci. Rep., 2013, 3, 1608.
47. Y. Q. Ma, B. L. Liu, A. Y. Zhang, L. Chen, M. Fathi, A. Abbas, M. Y. Ge, C. F. Shen and C. W. Zhou, ACS Nano, 2015, 9, 7383.
48. X. Wang, X. Chen, Y. Zhou, C. Park, C. An, Y. Zhou, R. Zhang, C. Gu, W. Yang and Z. Yang, Sci. Rep., 2017, 46694.
49. X. Zhang, W. P. Han, J. B. Wu, S. Milana, Y. Lu, Q. Q. Li, A. C. Ferrari and P. H. Tan, Phys. Rev. B: Condens. Matter Mater. Phys., 2013, 87, 115413.
50. Y. Zhao, X. Luo, H. Li, J. Zhang, P. T. Araujo, C. K. Gan, J. Wu, H. Zhang, S. Y. Quek, M. S. Dresselhaus and Q. Xiong, Nano Lett., 2013, 13, 1007.
51. R. Zhang, D. Drysdale, V. Koutsos and R. Cheung, Adv. Funct. Mater., 2017, 1702455.
52. W. Levason, R. Patel and G. Reid, J. Organomet. Chem., 2003, 688, 280.
53. (a) G. M. Sheldrick, Acta Crystallogr., Sect. C: Struct. Chem., 2015, 71, 3; (b) G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008, 64, 112; (c) O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339.
54. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman and D. J. Fox, Gaussian 16, Revision C.01, Gaussian, Inc., Wallingford, CT, 2016.
55. S. Grazulis, D. Chateigner, R. T. Downs, A. F. Yokochi, M. Quiros, L. Lutterotti, E. Manakova, J. Butkus, P. Moeck and A. Le Bail, J. Appl. Crystallogr., 2009, 42, 726.
56. ICSD: Inorganic Crystal Structure Database (ICSD) accessed via the EPSRC funded Physical Sciences Data-science Service hosted by the University of Southampton and STFC.
57. H. Hess and H. Hartung, Z. Anorg. Allg. Chem., 1966, 344, 157.

---

Footnote

† Electronic supplementary information (ESI) available: Crystallographic parameters (Table S1), IR and NMR spectra for the new complexes (Fig. S2–S28), grazing incidence XRD patterns for further WSe₂ films obtained in this work (Fig. S29), together with computational data from DFT calculations (Tables S2–S5 and Fig. S30). CCDC 2122137 (WSeCl₄), 2122138 ([WSeCl₄(OPPh₃)]), 2122139 ([WSeCl₄(DMF)]), 2122140 ([WSeCl₄(MeCN)]), 2122141 ([(WSeCl₄)₂(4,4′-bipy)]), 2122142 ([WSeCl₄(dioxane)]), 2122143 ([(WSeCl₄)₂(μ-SeMe₂)]), 2122144 ([(WSeCl₄)₂(μ-Se)(μ-Se₂)]). For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/d1dt03980f

---