Accessing γ-Ga2S3 by solventless thermolysis of gallium xanthates: A low-temperature limit for crystalline products

Article abstract of DOI:10.1039/c9dt02061f

Alkyl-xanthato gallium(iii) complexes of the form [Ga(S₂COR)₃], where R = Me (1), Et (2), ⁱPr (3), ⁿPr (4), ⁿBu (5), ^sBu (6) and ⁱBu (7), have been synthesized and fully characterised. The crystal structures for 1 and 3-7 have been solved and examined to elucidate if these structures are related to their decomposition. Thermogravimetric analysis was used to gain insight into the decomposition temperatures for each complex. Unlike previously explored metal xanthate complexes which break down at low temperatures (<250 °C), to form crystalline metal chalcogenides, powder X-ray diffraction measurements suggest that when R ≥ Et these complexes did not produce crystalline gallium sulfides until heated to 500 °C, where γ-Ga₂S₃ was the sole product formed. In the case of R = Me, Chugaev elimination did not occur and amorphous Ga_xS_y products were formed. We conclude therefore that the low-temperature synthesis route offered by the thermal decomposition of metal xanthate precursors, which has been reported for many metal sulfide systems prior to this, may not be appropriate in the case of gallium sulfides.

Full text of DOI:10.1039/c9dt02061f

Dalton
Transactions
View Article Online
View Journal
PAPER
Accessing γ-Ga2S3 by solventless thermolysis of
gallium xanthates: a low-temperature limit for
Cite this: DOI: 10.1039/c9dt02061f
crystalline products†
c
Suliman A. Alderhami,a,b David Collison,a David J. Lewis,
*
Paul D. McNaughter, a Paul O’Brien, a,c Ben F. Spencer,
c
Inigo Vitorica-Yrezabala and George Whitehead
a
i
Alkyl–xanthato gallium(III) complexes of the form [Ga(S2COR)3], where R = Me (1), Et (2), Pr (3), nPr (4),
nBu (5), Bu (6) and Bu (7), have been synthesized and fully characterised. The crystal structures for 1 and
3–7 have been solved and examined to elucidate if these structures are related to their decomposition.
Thermogravimetric analysis was used to gain insight into the decomposition temperatures for each
complex. Unlike previously explored metal xanthate complexes which break down at low temperatures
(<250 °C), to form crystalline metal chalcogenides, powder X-ray diffraction measurements suggest that
when R ≥ Et these complexes did not produce crystalline gallium sulfides until heated to 500 °C, where
γ-Ga2S3 was the sole product formed. In the case of R = Me, Chugaev elimination did not occur and
amorphous GaxSy products were formed. We conclude therefore that the low-temperature synthesis
route offered by the thermal decomposition of metal xanthate precursors, which has been reported for
many metal sulfide systems prior to this, may not be appropriate in the case of gallium sulfides.
s
i
Received 16th May 2019,
Accepted 26th July 2019
DOI: 10.1039/c9dt02061f
rsc.li/dalton
Molecular precursor routes towards gallium sulfides in par-
ticular present an excellent opportunity to control the phase
Introduction
Gallium sulfides are refractory materials which are produced and stoichiometry of the materials produced. As there are
by traditional synthetic routes due to their existence in four seven potential phases that can be formed, the control of the
stoichiometric forms: gallium monosulfide (GaS), gallium ses- phase achieved is of greater concern as compared to other
quisulfide (Ga2S3), Ga2S and Ga4S5. Additionally, both GaS and metal chalcogenides with fewer potential phases, e.g. PbS, CdS
Ga2S3 exhibit polymorphism.1–4 In contrast, aluminium sul- and ZnS, and this therefore remains a challenge, which is
fides possess only one stoichiometric form, Al2S3, with four reflected in the range of materials produced in the literature.
polymorphs, whilst gallium selenides and tellurides are known For example, Barron et al. used gallium–chalcogen cubanes to
to form only three and four phases, respectively.3,5 However, deposit cubic GaS using MOCVD. The cubic [tBuGaS]4 clusters
gallium sulfides possess wide band gaps in the range of act as seeds for oriented nucleation followed by the growth of
3.05–2.85 eV and have potential for use in photovoltaics and cubic GaS.10 By using [Cp*Ga(µ-S)]4, where Cp* is (C5Me5) or
optoelectronics and as passivating layers in III–V group semi- (C5Me4Et), Barron et al. prepared γ-Ga2S3.10,11 In contrast,
conductor devices.6–9 Thus, low temperature ‘soft’ processing [Ga(SiPr)2(μ-SiPr)]2 has been used as a molecular precursor in
routes that produce well-defined phases and stoichiometries of LP-MOCVD experiments to produce different phases of
these materials would potentially be of tremendous utility.
gallium sulfides depending on the substrate used; γ-Ga2S3 is
produced on glass, α-Ga2S3 on silicon (100), and highly
oriented γ-Ga2S3 on YSZ (111), suggesting that an epitaxial
aSchool of Chemistry, The University of Manchester, Oxford Road, Manchester,
M13 9PL, UK
growth
mechanism
occurs
in
these
cases.12
bChemistry Department, Faculty of Science and Arts, Al-Baha University,
Al Makhwah, Saudi Arabia
[MeGa(SCH2CH2S)]3 was used to synthesize nanometric
β-Ga2S3 when pyrolysed between 300 and 500 °C.13
Ramalingam et al. prepared α-Ga2S3 nanoparticles from tris
(cyclo-hexylmethyldithiocarbamato) gallium(III) and tris(cyclo-
hexyl-ethyldithiocarbamato) gallium(III), using a solvothermal
method.14 Malik et al. synthesised monodisperse cubic GaS
nanoparticles using [Ga(S2CNEt2)3] in 4-ethylpyridine at
cSchool of Materials, The University of Manchester, Oxford Road, Manchester,
M13 9PL, UK. E-mail: david.lewis-4@manchester.ac.uk
†Electronic supplementary information (ESI) available: IR spectra, NMR data,
single-crystal X-ray crystallography, and TGA, SEM and EDX analyses. CCDC
1900046–1900052. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c9dt02061f
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
167 °C.15 [Ga(S2CNMenHex)3], prepared from an asymmetrical min−1 under N2. Powder X-ray diffraction (PXRD) patterns were
amine, yielded α-Ga2S3 when using LP-MOCVD at 500 °C.16 obtained using an X’Pert diffractometer with a Cu-Kα source,
The tris-monothiocarbamate complex, [Ga(SOCNEt)3], pro- and scanning electron microscopy (SEM) was performed using
duced cubic GaS at 350 °C by AACVD.17 Barron et al. studied a Tescan Mira 3 FEG scanning electron microscope (SEM)
the preparation of di-tert-butyl gallium dithiocarbamate pre- equipped with an energy-dispersive X-ray (EDX) spectrometer.
cursors [(tBu)2Ga(S2CNR2)], where R = Me or Et. Both precur- Single-crystal X-ray diffraction data for all precursors were col-
sors formed hexagonal GaS by AP-MOCVD at relatively low lected on a dual source Rigaku FR-X rotating anode diffracto-
temperatures.18 Guihua Shang et al. used thiocarboxy com- meter with a Mo-Kα wavelength source at 150 K for precursors
plexes of the form [Ga(SCOMe)2Me(dmpy)] (where dmpy = 3,5- 1, 2, 3, 4, 5, and 7 and with a Cu-Kα wavelength source at 200 K
dimethylpyridine) to deposit hexagonal α-Ga2S3 by AACVD.19 for precursor 6, and reduced using a CrysAlisPro 171.39.30c.
However, despite these reports, the use of molecular precursor The structure was solved and refined using SHELXTL 2016,
routes to produce gallium sulfides is relatively unexplored implemented through Olex2 v1.2.9.
compared with other routes towards these materials.
XPS was performed using a Kratos Axis Ultra spectrometer
Metal xanthate precursors offer an attractive route to metal (Kratos Analytical, UK) with a monochromatic Al K_alpha X-ray
sulfides due to their low decomposition temperatures and for- source (1486 eV, 10 mA emission, and 150 W). Survey spectra
mation of volatile bi-products upon decomposition, i.e. SCO, were measured with a pass energy of 80 eV and high-resolution
H2S and an alkyl-ene.20–22 These properties have been spectra with a pass energy of 20 eV. A charge neutraliser was
exploited by us and others for solventless thermolysis of metal used to remove differential charging often exhibited by
xanthate precursors to form the corresponding metal chalco- powders. Binding energy calibration was performed using the
genides, e.g. CdS, FeS,23 CuS,21,24 PbS,25 ZnS,26 and SnS,27
C 1s peak at 284.8 eV associated with adventitious
chalcogenide spinels (AIIB2IIIX4)28 and Cu2ZnSnS4.29 Metal hydrocarbons.
xanthate complexes have also been proven to be successful pre-
Synthesis of potassium alkyl xanthates
cursors for a wide range of nanometric binary, ternary and
quaternary target metal sulphides in other media, e.g. in solu- Potassium alkyl xanthate (KS2COR) ligands with various alkyl
tion or within polymers.30–35
groups (methyl, ethyl, isopropyl, n-propyl, n-butyl, s-butyl and
In this work, we report the synthesis of a series of linear i-butyl) were prepared according to the method of McNaughter
and branched alkyl xanthato gallium(III) complexes and et al.25 A typical ligand synthesis is as follows for K(S2COMe):
examine the influence of the alkyl chain length on their break- potassium hydroxide was dissolved (3.5 g, 62.3 mmol) in
down profiles. We determine the suitability of gallium(III) excess methanol by stirring for one hour and was subsequently
xanthates as precursors for gallium sulfides by low-tempera- cooled to 0 °C. Carbon disulfide (4.7 g, 3.7 ml, 62.3 mmol) was
ture thermolysis. It should be noted that whilst the complexes added dropwise with stirring for one hour. The yellow precipi-
tris(O-ethyl xanthato)gallium(III) and dimethylpentan-3-yl tate was filtered, recrystallized from acetone and dried in air to
xanthato gallium(III) have been previously reported, their obtain fine yellow crystals of potassium methyl xanthate.
thermal decomposition has not been studied in detail – we
Synthesis of tris(O-methyl xanthato)gallium(III), [Ga(S2COMe)3]
(1)
also report these aspects here.36,37
Gallium(III) chloride (1.51 g, 8.6 mmol) was dissolved in dry
toluene (40 ml) resulting in a colourless solution. Potassium
Experimental section
methyl xanthate (3.76 g, 25.8 mmol) was added to the solution
Anhydrous toluene (≥99.8%, Sigma-Aldrich), anhydrous of GaCl3 and it was stirred for two hours. The precipitate of
gallium(III) chloride (99.9%, Sigma-Aldrich), methanol KCl formed was separated by filtration under air and the
(≥99.8%, Sigma-Aldrich), ethanol (≥99.8%, Sigma-Aldrich), solvent from the mother liquor was removed in vacuo to
2-propanol (≥99.5%, Sigma-Aldrich), 1-propanol (99%, Sigma- produce a slightly opaque viscous liquid. Hexane (8 ml) was
Aldrich), 1-butanol (99%, Sigma-Aldrich), 2-butanol (99%, added to the liquid and the mixture was stored at 4 °C for
Sigma-Aldrich), isobutanol (99%, Sigma-Aldrich), carbon di- 12 hours, which resulted in the formation of off-white shiny
sulfide (99%, Sigma-Aldrich), hexane (≥97%, Sigma-Aldrich) crystals. The crystals were collected by filtration and washed
and potassium ethyl xanthogenate (96%, Sigma-Aldrich) were with cold hexane. M.p. 59–60 °C. Yield: 51%. FT-IR (solid)
used as received. All syntheses were performed under an N2 νmax/cm−1: 2938 (w), 1436 (w), 1236 (s), 1170 (s), 1027 (s) and
1
environment using the Schlenk technique unless otherwise 945 (m). H NMR (CDCl3, 400 MHz) δ/ppm: 4.18 (s, 3H; OMe).
stated. H and 13C NMR spectra were obtained using a Bruker 13C {1H} NMR (CDCl3, 400 MHz) δ/ppm: 45.0 (OMe); 202.7
1
AC400 FT-NMR spectrometer. Melting points were determined (S2C–O). Elemental analysis: found (calc.) C, 18.91 (18.43); H,
using melting point apparatus (Stuart; Cole-Palmer, UK); infra- 2.31 (2.32); S, 48.38 (49.11); Ga, 14.12 (17.85).
red spectra were recorded via ATR (4000–400 cm−1, resolution
Synthesis of tris(O-ethyl xanthato)gallium(III), [Ga(S2COEt)3] (2)
4 cm−1); and elemental analysis was performed with a Carlo
Erba EA 1108. Thermogravimetric analyses (TGA) were per- The procedure was the same as for (1) with gallium(III) chloride
formed using a Seiko SSC/S200 at a heating rate of 10 °C (1.51 g, 8.6 mmol) and potassium ethyl xanthate (4.12 g,
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
25.8 mmol) being used. Off-white shiny crystals were collected M.p. 70–71 °C. Yield: 35%. FT-IR (solid) νmax/cm−1: 2971 (w),
and washed with cold hexane. M.p. 82–83 °C. Yield: 65% 1453 (w), 1233 (s), 1098 (s), 1030 (s) and 869 (m). 1H NMR
(1.21 g). FT-IR (solid) νmax/cm−1: 2987 (w), 1433 (w), 1244 (s), (CDCl3, 400 MHz) δ/ppm, 1.00 (t, J = 8 Hz, 3H; OCH(Me)
1117 (s), 1023 (s) and 860 (m). 1H NMR spectra (CDCl3, CH2M
400 MHz): 1.45–1.47 ppm (t, Hz, 3H; OCH2M ); CH2); 5.02–5.10 (m, 1H; OCH
5.18–5.27 ppm (q, 2H; OCH
2–). 13C {1H} NMR (CDCl3, 400 MHz), 9.57 (OCH(Me)CH2M
400 MHz) δ/ppm: 14.0 (OCH2CH3); 74.7 ppm (OCH2–); 226 (OCH(Me)–); 88.8 (OCH(Me)–); 226 (s, S2C
(S2C–O). Elemental analysis: found (calc.) C, 24.96 (25.19); H, lysis: found (calc.) C, 34.83 (34.80); H, 5.27 (5.17); S, 37.12
̲e̲); 1.42 (d, 3H; OCH(M̲e̲
8
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
3.49 (3.75); S, 44.34 (43.55); Ga, 16.12 (16.09).
(36.90); Ga, 13.49 (13.43).
Synthesis of tris(O-i-butyl xanthato)gallium(III), [Ga(S2COiBu)3]
(7)
Synthesis of tris(O-isopropyl xanthato)gallium(III),
[Ga(S2COiPr)3] (3)
The procedure was the same as for (1) using gallium(III) chlor-
ide (1.51 g, 8.6 mmol) and K(S2COiPr) (4.48 g, 25.8 mmol),
resulting in off-white shiny crystals of gallium(III) isopropyl
xanthate. M.p. 87–88 °C. Yield: 41% (1.75 g). FT-IR (solid)
νmax/cm−1: 2978 (w), 1447 (w), 1240 (s), 1084 (s), 1020 (s) and
The procedure was the same as for precursor (1), using
K(S2COiBu) (4.84 g, 25.8 mmol) and GaCl3 (1.51 g, 8.6 mmol).
Yield: 50%. M.p. 55–57 °C. FT-IR (solid) νmax/cm−1: 2961 (w),
1448 (w), 1205 (s), 1177 (s), 1036 (s) and 975 (m). 1H NMR
(CDCl3, 400 MHz) δ/ppm, 1.02 (t, 6H; OCH2CH(M
2.11–2.25 (m, 1H; OCH2CH(Me)2); 4.23 (t, J = 8 Hz, 2H;
OCH
2CH(Me)2. 13C {1H} NMR (CDCl3, 400 MHz) δ/ppm: 19.0
(s, OCH2CH(Me)M ); 19.0 ppm (s, OCH2CH(M ∼)); 27.9 ppm
(OCH2CH(Me)); 84.7 ppm (s, OCH2–); 227 (s, S2C–O).
̲e̲)2);
1
899 (m). H NMR (CDCl3, 400 MHz) δ/ppm: 1.46 (d, J = 8 Hz,
̲
6H; OCH(M
NMR (CDCl3, 400 MHz), 21.44 (OCH(C
NCH–); 225.06 ppm (s, S2C–O). Elemental analysis: found
̲e̲
)2); 5.18–5.27 ppm (sep, 1H; OCH–). 13C {1H}
̲
̲H3)2); 83.9 ppm (s,
̲
̲
̲e̲
̲
̲
̲
̲
(calc.) C, 30.33 (30.69); H, 4.46 (4.65); S, 40.41 (39.23); Ga,
14.96 (14.52).
Elemental analysis: found (calc.) C, 34.83 (34.19); H, 5.27
(5.12); S, 37.12 (35.51); Ga, 13.49 (13.15).
Synthesis of tris(O-n-propyl xanthato)gallium(III),
Pyrolysis of precursors by the melt reaction method
[Ga(S2COnPr)3] (4)
Pyrolysis was conducted in a ceramic boat by heating 0.2 g of
each precursor to 500 °C for one hour in a tube furnace
(Carbolite Gero 30–3000 °C) under N2 gas. Upon cooling, black
powders were collected.
The procedure was the same as for precursor (1), using
K(S2COnPr) (4.48 g, 25.8 mmol) and GaCl3 (1.51 g, 8.6 mmol).
M.p. 58–60 °C. Yield: 41%. FT-IR (solid) νmax/cm−1: 2969 (w),
1455 (w), 1248 (s), 1138 (s), 1033 (s) and 939 (m). 1H NMR
(CDCl3, 400 MHz): 1.04 ppm (t, 8 Hz, 3H; OCH2CH2Me);
1.83–1.92 ppm (m, 2H; OCH2CH2Me); 4.40–4.44 ppm (t, 2H;
OCH2CH2Me). 13C {1H} NMR (CDCl3, 400 MHz) δ/ppm: 10.3 (s,
Results and discussion
OCH2CH2M
̲e̲); 21.8 (s, NCH2C̲H2); 80.3 (s, OC̲H2–); 227 (s, Seven alkyl xanthato gallium(III) complexes with different alkyl
S2C–O). Elemental analysis: found (calc.) C, 30.33 (30.57); H, groups, [Ga(S2COMe)3] (1), [Ga(S2COEt)3] (2), [Ga(S2COiPr)3]
4.46 (4.35); S, 40.41 (40.38); Ga, 14.96 (14.70).
(3), [Ga(S2COnPr)3] (4), [Ga(S2COnBu)3] (5), [Ga(S2COsBu)3] (6)
and [Ga(S2COiBu)3] (7), were prepared by metathesis of
gallium(III) chloride with the relevant potassium alkyl xanthate
in dry toluene. The products were characterised using their
Synthesis of tris(O-n-butyl xanthato)gallium(III),
[Ga(S2COnBu)3] (5)
The procedure was the same as for precursor (1), using FTIR and NMR spectra (ESI Fig. 1.1, 2.1–2.7†), and elemental
K(S2COnBu) (4.84 g, 25.6 mmol) and GaCl3 (1.51 g, 8.6 mmol). analysis. All complexes were powders with an off-white shiny
M.p. 40–42 °C. Yield: 40%. FT-IR (solid) νmax/cm−1: 2952 (w), appearance and were stored at −20 °C. All seven complexes
1452 (w), 1214 (s), 1140 (m), 1036 (s) and 917 (m). 1H NMR readily dissolved in chloroform, toluene and DCM.
spectra (CDCl3, 400 MHz): 0.96–0.99 ppm (8 Hz, t, 3H;
OCH2CH2CH2M ); 1.43–1.52 ppm (sex, 2H; OCH2CH2CH2Me); novel; while complex 2 has been reported previously, our data
1.79–1.86 ppm (m, 2H; OCH2CH2–); 4.44–4.48 ppm (t, 2H; are included for completeness (Fig. 1). The crystal data and
OCH
2–). 13C {1H} NMR (CDCl3, 400 MHz) δ/ppm: 13.7 (s, refinement parameters associated with these structures are
OCH2CH2CH2M ); 19.0 (OCH2CH2CH2); 30.35
ppm shown in ESI Tables 3.1–3.3.†36
(OCH2CH2–); 78.7 (OCH2–); 227 (S2C–O). Elemental analysis: Examination of the crystal structures revealed inter-
found (calc.) C, 34.83 (35.01); H, 5.27 (5.24); S, 37.12 (37.17); molecular S–S distances below the sum of the van der Waals
The crystal structures of complexes 1, 3, 4, 5, 6 and 7 are
̲
̲
̲
̲
̲
̲e̲
̲
̲
Ga, 13.49 (13.22).
radii, i.e. below 3.78 Å, suggesting possible S–S interactions in
the solid state for 6, 5 and 2 structures with distances of
3.6852(19) Å, 3.475(2) Å and 3.535(1) Å, respectively. This was
also observed by Hoskins et al. for complex 2 with a slightly
Synthesis of tris(O-s-butyl xanthato)gallium(III), [Ga(S2COsBu)3]
(6)
The procedure was the same as for precursor (1), using different S–S distance of 3.598(4) Å and the difference is attrib-
K(S2COsBu) (4.84 g, 25.8 mmol) and GaCl3 (1.51 g, 8.6 mmol). uted to greater errors with the earlier measurement.36
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 1 X-ray crystal structures of gallium xanthates. (a) [Ga(S2COMe)3] (1), (b) [Ga(S2COEt)3] (2), (c) [Ga(S2COiPr)3] (3), (d) [Ga(S2COnPr)3] (4), (e) [Ga
(S2COnBu)3] (5), (f) [Ga(S2COsBu)3] (6) and (g) [Ga(S2COiBu)3] (7). Green = Ga, red = O, black = C, and yellow = S. Thermal ellipsoids at 50% prob-
ability. CCDC range is 1900046–1900052.†
It has been suggested by Harcourt and Winter that a feature interactions has little asymmetric bonding between the S donor
exhibited by many metal xanthate complexes is the presence of sites caused by intramolecular interactions. Careful examination
an asymmetric binding mode by the xanthate to the metal of S–S contacts is required to distinguish between M–S differ-
centre.38 To determine if this is an intrinsic property of the ences internal to the molecule and M–S differences caused in
molecule or as a consequence of the observation of the structure the crystal packing. The intrinsic bond length differences may
as a crystalline solid the possible intermolecular interactions, influence the breakdown of the precursor and potentially the
e.g. S–S van der Waals interactions, were examined and com- phase created when the breakdown occurs within the solid or
pared to the decomposition temperatures and products.
liquid states. In the case of gallium xanthates, the large Ga–S
Complex 5 possesses the shortest S–S intermolecular dis- bond differences can be attributed to intermolecular S–S inter-
tance of S(1)–S(5) = 3.4749(24) Å. S(5) and S(1) reside in adja- actions, and consequentially a relationship between Ga–S bond
cent xanthate ligands within the complex, and show large differences of [Ga(S2COR)3] in the solid state and breakdown
Ga–S bond differences of 0.1013(7) and 0.1194(8) Å (Fig. 2(b)). temperatures cannot be drawn.
The remaining ligand with no S–S interactions has a Ga–S
Examination of the TGA profiles revealed a variety of multi-
bond difference of 0.0021(7) Å, suggesting that in Ga xanthate step breakdowns for both the linear alkyl chains (Fig. 3) and
complexes the primary contribution to the asymmetric Ga–S the branched alkyl chains (Fig. 4). The residual mass of 26.4%
bonding is from intermolecular S–S interactions in the solid for [Ga(S2COMe)3] (1) corresponds closely to the expected mass
state. This is in contrast to the observations of the binding of of GaS of 26.0% (Table 1). The TGA profile step sizes revealed
xanthates to group 14 and 15 elements where large M–S bond the loss of two potential species for each step, i.e. SCOR–H
differences of approximately 0.5 Å are seen.25,38
(SCO and the alkene) or multiples of H2S (ESI Fig. 4.1–4.3 and
The clearest example of a complex with no intermolecular ESI Table 4.2†). In the cases of [Ga(S2COEt)3] and longer alkyl
S–S interactions is complex 7 with the closest S–S distance chains regardless of branching, all lost three equivalents of
above the sum of the van der Waals radii at 4.286(26) Å. This SCO and the corresponding alkene followed by the loss of half
i
s
complex possesses a distorted octahedral shape presumably an H2S. The Et, Pr and Bu show a clean two-step process,
caused by the coordination of the xanthate ligand (Fig. 2(a)). where the first step corresponds to the loss of three equiva-
The distorted octahedron consists of two parallel near-equilat- lents of SCO and the alkene, and the second step corresponds
eral triangles and six isosceles triangles created from the S–Ga–S to the loss of half an equivalent of H2S as expected for
bite angles of the xanthate ligands, each having vdW inter- Chugaev elimination-based decomposition. For nPr, nBu and
actions with two other xanthate ligands. The differences in the iBu, the initial step involving the loss of three equivalents of
Ga–S bonds from within the same xanthate are small, all below the alkene and SCO shows two features, as highlighted by the
0.031 Å, suggesting that the complex with no intermolecular S–S first derivative (ESI Fig. 4.1–4.3†), although the total weight
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
Fig. 3 TGA profiles for the n-alkyl series of [Ga(S2COR)3] precursors,
where R = Me, Et, nPr and nBu.
Fig. 2 Coordination spheres of complexes. (a) [Ga(S2COiBu)3], (7),
showing no intermolecular S–S interactions and (b) [Ga(S2COnBu)3], (5),
showing the intermolecular S–S interactions (dashed lines) occurring
between metal xanthate complexes. Gallium (green), sulfur (yellow) and
oxygen (red) with carbons and hydrogens from the alkyl groups are
omitted for clarity.
loss of the step corresponds to the correct amount for the
Chugaev elimination of all three ligands (ESI Table 4.2).† The
deconvolution of these features and the elucidation of the pro-
ducts formed require further study to fully understand the
mechanistic differences between chains.
Fig. 4 TGA profiles for the branched alkyl series of [Ga(S2COR)3] pre-
cursors, where R = iPr, sBu and iBu.
The 13C NMR shifts of the xanthate ipso carbon were com-
pared to the temperature at the first breakdown step of the
complexes from TGA. The Me–Et–iPr–iBu series showed a
i
i
decrease in shift from Me to Pr and an increase for Bu (ESI
Fig. 5.1(a)).† This is in contrast to the expected result if the
position of the shift was governed purely by the inductive
influence of the alkyl group, i.e. the same trend as the pKa
values of the corresponding alcohols. A deviation from the
expected order was also observed for the Me–Et–nPr–nBu series
(ESI Fig. 5.1(b)).† When plotted against the temperature at the
first breakdown step, the 13C shifts show a negative correlation
between the electron density at the ipso carbon as reflected by
the 13C shift increasing with a decrease in the temperature of
the first breakdown step (ESI Fig. 5.2).†
Table 1 TGA final residues and matched phases
Calculated, %
Precursor
Final residue, %
GaS
Ga2S3
[Ga(S2COMe)3]
[Ga(S2COEt)3]
[Ga(S2COiPr)3]
[Ga(S2COnPr)3]
[Ga(S2COnBu)3]
[Ga(S2COsBu)3]
[Ga(S2COiBu)3]
26.4
26.4
25.3
24.2
22.4
22.5
23.2
26.0
—
—
—
—
—
27.2
24.8
24.8
22.8
22.8
22.8
—
—
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
XPS of the powders produced allowed the examination of
The observed difference between the crystallinity of the
the composition of the film and any contamination. powders resulting from the furnace experiments and that of the
Examination of the Ga 2p profile showed a major contribution TGA may be attributable to the difference in the heating profiles
from a sulfur-bound species at 1117.8 eV and a minor contri- used in both experiments. In the case of TGA, the sample was
bution from an oxygen-bound species at 1119 eV resulting in heated from room temperature to 600 °C at a heating rate of
between 0.9 and 4.9% of binding to oxygen (ESI Table 6.1 and 10 °C min−1, whereas the sample in the furnace was heated
ESI Fig. 6.1).† The S 2p 3/2 peak shows only one species at rapidly to the target temperature of 85 °C min−1 and was held
approx. 161.5 eV, with no intensity at the predicted location for at this temperature for one hour. Only at 500 °C the reflections
sulfur bound to oxygen at 168–170 eV (ESI Fig. 6.2).† For the corresponding to the crystalline phase of γ-Ga2S3 were observed
complexes of alkyl group Et or greater the Ga/S value was in when breaking down [Ga(S2COEt)3]. We must conclude that the
the region of 1.3 to 1.8 indicating likely formation of a Ga2S3 TGA experiments did not provide sufficient energy to the
phase whereas for Me the S/Ga was 2.0. A S/Ga of two gives a sample to enable the formation of a crystalline phase of gallium
stoichiometry of GaS2, which is not observed as a crystalline sulfide. Subsequently 1 to 7 were all heated at 500 °C. The pro-
phase and is likely to contain both the +1 and +3 oxidation ducts of thermolysis from precursors 2 to 7 have PXRD patterns
states of Ga, e.g. GaIGaIIIS4.
corresponding to γ-Ga2S3 with 1 being clearly amorphous
Initial thermolysis of [Ga(S2COEt)3] at 250 °C (i.e. the temp- despite the same heating profile (Fig. 5, bottom). These low-
erature at which full decomposition was observed by TGA) for intensity reflections correspond to the (111), (220) and (311)
one hour resulted in an amorphous material which showed no planes of the defect of zinc blende γ-Ga2S3 (ICDD no. 00-043-
ˉ
Bragg reflections in its powder X-ray diffraction (PXRD) pattern 0916; F43m, a = 5.210 Å) and are significantly broadened indi-
(Fig. 5, top). The thermolysis was repeated at 350, 450 and cating small, possibly nanoscale, crystalline domains.
500 °C, with these temperatures held for 1 h to assist in the
The breakdown products from [Ga(S2COMe)3] when exam-
formation of a crystalline phase. It was found that despite ined by XPS and TGA provided contrasting results with respect
being well in excess of the predicted temperature for the mole-
cular breakdown, as observed by TGA, only a temperature of
500 °C produced crystalline γ-Ga2S3 powders.
Fig. 5 Powder X-ray diffraction of thermolysis products. Top: PXRD
patterns of γ-Ga2S3 powders obtained from pyrolysis of [Ga(S2COEt)3] at
different temperatures (a) 250 °C, (b) 350 °C, (c) 450 °C and (d) 500 °C
for one hour under N2. Bottom: PXRD patterns of γ-Ga2S3 powders
obtained from pyrolysis of precursors 1–7 at 500 °C for one hour under
N2. The stick pattern below corresponds to cubic γ-Ga2S3 (ICDD no. 00-
043-0916; F4¯3m, a = 5.210 Å).
Fig. 6 Secondary electron SEM image of a crystallite of γ-Ga2S3 pro-
duced from the decomposition of [Ga(S2COEt)3] at 500 °C (a) with
corresponding EDX elemental maps for S Kα (b) and Ga Lα (c).
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
to the stoichiometry of the GaxSy powder produced. XRD has
demonstrated that [Ga(S2COMe)3] does not produce a crystal-
Acknowledgements
line phase unlike all longer alkyl chain complexes which The authors would like to acknowledge the Royal Embassy
produce γ-Ga2S3. As the Me complex does not possess a Saudi Arabia Culture Bureau London and Al-Baha University
β-hydrogen required for the six-membered cyclic transition for financial support. PDM and DJL would like to acknowledge
state, the syn-elimination cannot occur. An alternative pathway the EPSRC (grant EP/R022518/1).
for the ligands is to react as a free xanthic acid which is
unstable and reform CS2 and the free alcohol. The TGA profile
for [Ga(S2COMe)3] shows a single step whereby the ligands are
lost rapidly and indicates that all the ligands are lost by the
References
same mechanism, although a complete understanding of this
cannot be attained from our data.
1 G. Shen, D. Chen, P.-C. Chen and C. Zhou, ACS Nano, 2009,
3, 1115–1120.
SEM images of 2–7 showed randomly shaped aggregated
crystals of γ-Ga2S3 (ESI Fig. 7.1†). EDX spectroscopic analysis
showed that the γ-Ga2S3 powders were gallium rich by between
ca. 0.3–3.2% away from the expected value (ESI Table 7.1†). Ga-
rich γ-Ga2S3 powders are an unexpected result from the per-
spective of each precursor containing six sulphur atoms and
one gallium, but reports by O’Brien et al. suggest that on using
metal dithiocarbamate precursors at high temperatures, metal-
rich phases are observed.39 EDX elemental mapping of gallium
sulfide powders showed uniform distribution of gallium and
sulfur (Fig. 6).
2 P. Rao, S. Kumar and N. K. Sahoo, Mater. Chem. Phys.,
2015, 149–150, 164–171.
3 K. K. Banger and A. F. Hepp, in Encyclopedia of Inorganic
and Bioinorganic Chemistry, American Cancer Society, 2011.
4 M. Lazell, P. O’Brien, D. J. Otway and J.-H. Park, J. Chem.
Soc., Dalton Trans., 2000, 4479–4486.
5 A. J. Downs and C. R. Pulham, Chem. Soc. Rev., 1994, 23,
175–184.
6 G. Shang, K. Kunze, M. J. Hampden-Smith and
E. N. Duesler, Chem. Vap. Deposition, 1996, 2, 242–244.
7 C. Sanz, C. Guillén and M. T. Gutiérrez, J. Phys. D: Appl.
Phys., 2008, 41, 235103.
8 J.-H. Park, M. Afzaal, M. Helliwell, M. A. Malik, P. O’Brien
and J. Raftery, Chem. Mater., 2003, 15, 4205–4210.
9 D. P. Dutta, G. Sharma, A. K. Tyagi and S. K. Kulshreshtha,
Mater. Sci. Eng., B, 2007, 138, 60–64.
Conclusions
In this work, we have reported the synthesis and full character-
ization of seven gallium xanthate complexes with low break- 10 A. N. MacInnes, M. B. Power and A. R. Barron, Chem.
down temperatures. Despite their low breakdown temperatures Mater., 1992, 4, 11–14.
in the region of 250 °C, it was found that heating at tempera- 11 S. Schulz, E. G. Gillan, J. L. Ross, L. M. Rogers, R. D. Rogers
tures lower than 500 °C did not result in any crystalline phases and A. R. Barron, Organometallics, 1996, 15, 4880–4883.
of GaxSy. Performing the breakdown at 500 °C for one hour 12 S. Suh and D. M. Hoffman, Chem. Mater., 2000, 12, 2794–2797.
yielded γ-Ga2S3 for all [Ga(S2COR)3] precursors, where R ≥ Et. 13 P. U. Sastry and D. P. Dutta, J. Alloys Compd., 2009, 487,
When R = Me, the complex is unable to undergo Chugaev elim-
ination and the resulting powders showed various GaxSy stoi- 14 K. Ramalingam, G. S. Sivagurunathan and C. Rizzoli, Main
chiometries and produced an amorphous material. Previous Group Met. Chem., 2015, 38, 75–82.
reports using metal xanthate complexes with increasing linear 15 M. A. Malik, P. O’Brien, S. N. Mlondo and N. Revaprasadu,
alkyl chain lengths show TGA profiles changing from a single- Mater. Res. Soc. Symp. Proc., 2005, 879E, Z7.13.1.
step to a two-step breakdown process.25 At comparatively high 16 M. R. Lazell, P. O’Brien, D. J. Otway and J.-H. Park, Chem.
temperatures between 300 and 500 °C required for the for- Mater., 1999, 11, 3430–3432.
mation of Ga2S3, the subtleties of the molecular breakdown 17 G. A. Horley, M. R. Lazell and P. O’Brien, Chem. Vap.
below 250 °C have a reduced influence on the morphology of Deposition, 1999, 5, 203–205.
the material prepared unlike PbS which requires far lower 18 A. Keys, S. G. Bott and A. R. Barron, Chem. Mater., 1999, 11,
temperatures to form. Hence, the gallium sulfide system offers 3578–3587.
an insight into the balance of the precursor breakdown charac- 19 G. Shang, M. J. Hampden-Smith and E. N. Duesler, Chem.
teristics versus the temperature required for the formation of Commun., 1996, 1733–1734.
the target phase. These findings suggest that metal xanthate 20 M. A. Malik, M. Afzaal and P. O’Brien, Chem. Rev., 2010,
complexes with low temperature breakdown requirements are 110, 4417–4446.
best suited for target phases that readily form at low 21 M. Al-Shakban, P. D. Matthews, G. Deogratias,
351–353.
temperatures.
P. D. McNaughter, J. Raftery, I. Vitorica-Yrezabal, E. B. Mubofu
and P. O’Brien, Inorg. Chem., 2017, 56, 9247–9254.
22 N. Alam, M. S. Hill, G. Kociok-Köhn, M. Zeller, M. Mazhar
and K. C. Molloy, Chem. Mater., 2008, 20, 6157–6162.
23 L. Almanqur, I. Vitorica-yrezabal, G. Whitehead, D. J. Lewis
and P. O’Brien, RSC Adv., 2018, 8, 29096–29103.
Conflicts of interest
There are no conflicts of interest to declare.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
24 M. Akhtar, Y. Alghamdi, J. Akhtar, Z. Aslam, 32 F. T. F. O’Mahony, U. B. Cappel, N. Tokmoldin, T. Lutz,
N. Revaprasadu and M. A. Malik, Mater. Chem. Phys., 2016,
180, 404–412.
R. Lindblad, H. Rensmo and S. A. Haque, Angew. Chem.,
Int. Ed., 2013, 52, 12047–12051.
25 P. D. McNaughter, S. A. Saah, M. Akhtar, K. Abdulwahab, 33 M. Al-Shakban, P. D. Matthews, X. L. Zhong, I. Vitorica-
M. Azad Malik, J. Raftery, J. A. M. Awudza and P. O’Brien,
Dalton Trans., 2016, 45, 16345–16353.
Yrezabal, J. Raftery, D. J. Lewis and P. O’Brien, Dalton
Trans., 2018, 47, 5304–5309.
26 T. C. Vagvala, S. S. Pandey, S. Krishnamurthy and 34 T. Rath, M. Edler, W. Haas, A. Fischereder, S. Moscher,
S. Hayase, Z. Anorg. Allg. Chem., 2016, 642, 134–139.
27 M. Al-Shakban, Z. Xie, N. Savjani, M. A. Malik and
P. O’Brien, J. Mater. Sci., 2016, 51, 6166–6172.
28 M. Dilshad Khan, G. Murtaza, N. Revaprasadu and
P. O’Brien, Dalton Trans., 2018, 47, 8870–8873.
29 G. Kociok-Köhn, K. C. Molloy and A. L. Sudlow,
Can. J. Chem., 2014, 92, 514–524.
30 E. A. Lewis, P. D. McNaughter, Z. Yin, Y. Chen, J. R. Brent,
A. Schenk, R. Trattnig, M. Sezen, G. Mauthner, A. Pein,
D. Meischler, K. Bartl, R. Saf, N. Bansal, S. A. Haque,
F. Hofer, E. J. W. List and G. Trimmel, Adv. Energy Mater.,
2011, 1, 1046–1050.
35 P. D. McNaughter, J. C. Bear, A. G. Mayes, I. P. Parkin and
P. O’Brien, R. Soc. Open Sci., 2017, 4, 170383.
36 B. F. Hoskins, E. R. T. Tiekink, R. Vecchiet and G. Winter,
Inorg. Chim. Acta, 1984, 90, 197–200.
S. A. Saah, J. Raftery, J. A. M. Awudza, M. A. Malik, 37 T. Chand Vagvala, S. Sudhir Pandey, Y. Ogomi and
P. O’Brien and S. J. Haigh, Chem. Mater., 2015, 27, 2127–
2136.
31 A. J. MacLachlan, T. Rath, U. B. Cappel, S. A. Dowland,
S. Hayase, RSC Adv., 2014, 4, 24658–24661.
38 R. D. Harcourt and G. Winter, J. Inorg. Nucl. Chem., 1975,
37, 1039–1041.
H. Amenitsch, A.-C. Knall, C. Buchmaier, G. Trimmel, 39 S. Mlowe, D. J. Lewis, M. Azad Malik, J. Raftery,
J. Nelson and S. A. Haque, Adv. Funct. Mater., 2015, 25,
409–420.
E. B. Mubofu, P. O’Brien and N. Revaprasadu, Dalton
Trans., 2016, 45, 2647–2655.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019