Dissolution of Sn, SnO, and SnS in a Thiol-Amine Solvent Mixture: Insights into the Identity of the Molecular Solutes for Solution-Processed SnS

Article abstract of DOI:10.1021/acs.inorgchem.6b00243

Binary solvent mixtures of alkanethiols and 1,2-ethylenediamine have the ability to readily dissolve metals, metal chalcogenides, and metal oxides under ambient conditions to enable the facile solution processing of semiconductor inks; however, there is little information regarding the chemical identity of the resulting solutes. Herein, we examine the molecular solute formed after dissolution of Sn, SnO, and SnS in a binary solvent mixture comprised of 1,2-ethanedithiol (EDT) and 1,2-ethylenediamine (en). Using a combination of solution ¹¹⁹Sn NMR and Raman spectroscopies, bis(1,2-ethanedithiolate)tin(II) was identified as the likely molecular solute present after the dissolution of Sn, SnO, and SnS in EDT-en, despite the different bulk material compositions and oxidation states (Sn⁰ and Sn²⁺). All three semiconductor inks can be converted to phase-pure, orthorhombic SnS after a mild annealing step (～350 °C). This highlights the ability of the EDT-en solvent mixture to dissolve and convert a variety of low-cost precursors to SnS semiconductor material.

Full text of DOI:10.1021/acs.inorgchem.6b00243

Article
pubs.acs.org/IC
Dissolution of Sn, SnO, and SnS in a Thiol−Amine Solvent Mixture:
Insights into the Identity of the Molecular Solutes for Solution-
Processed SnS
Jannise J. Buckley,† Carrie L. McCarthy,† Joselyn Del Pilar-Albaladejo,‡ Golam Rasul,†, §
and Richard L. Brutchey*,†
†Department of Chemistry and §Loker Hydrocarbon Institute, University of Southern California, Los Angeles, California 90089,
United States
‡Department of Chemistry, Iowa State University, Ames, Iowa 50011, United States
S
* Supporting Information
ABSTRACT: Binary solvent mixtures of alkanethiols and 1,2-
ethylenediamine have the ability to readily dissolve metals,
metal chalcogenides, and metal oxides under ambient
conditions to enable the facile solution processing of
semiconductor inks; however, there is little information
regarding the chemical identity of the resulting solutes. Herein,
we examine the molecular solute formed after dissolution of
Sn, SnO, and SnS in a binary solvent mixture comprised of 1,2-
ethanedithiol (EDT) and 1,2-ethylenediamine (en). Using a
combination of solution 119Sn NMR and Raman spectros-
copies, bis(1,2-ethanedithiolate)tin(II) was identified as the
likely molecular solute present after the dissolution of Sn, SnO,
and SnS in EDT−en, despite the different bulk material compositions and oxidation states (Sn0 and Sn2+). All three
semiconductor inks can be converted to phase-pure, orthorhombic SnS after a mild annealing step (∼350 °C). This highlights
the ability of the EDT−en solvent mixture to dissolve and convert a variety of low-cost precursors to SnS semiconductor
material.
the stoichiometry of the final product is often dependent on the
nominal stoichiometry of the dissolved bulk material or
combinations of materials. The challenge of this approach,
however, lies in the fact that most bulk metals, metal oxides,
and metal chalcogenides are insoluble in common solvents, as
dissolution of these bulk materials involves breaking the strong
bonds that make up these frameworks.
Hydrazine is one solvent capable of dissolving metal
chalcogenide frameworks to yield soluble semiconductor inks.
Mitzi and co-workers at IBM first employed hydrazine in the
presence of elemental chalcogen as an effective solvent for the
dissolution of bulk metal chalcogenides.9−11 This dissolution
process is facilitated by nucleophilic attack by chalcogenides
(E2− = S2−, Se2−, and Te2−) formed via the in situ reduction of
elemental chalcogen in hydrazine. Reaction between the
resulting E2− and bulk metal chalcogenides yields soluble
hydrazinium chalcogenidometallate species, which are soluble
in hydrazine and other polar solvents, and can be easily solution
deposited to form high-quality metal chalcogenide thin films
upon annealing. A number of studies have investigated the
nature of the molecular solutes, the solute/solvent interactions,
INTRODUCTION
■
Solution processing of inorganic semiconductors for the
fabrication of affordable and scalable electronic thin films is
emerging as a commercially viable alternative to traditional
high-vacuum deposition methods.1 Using this approach,
semiconductor thin films can be printed onto various substrates
(e.g., plastics, glasses) at relatively low temperatures using
existing high-throughput technologies, such as roll-to-roll
manufacturing and spray coating.2 Innovations in the area of
molecular semiconductor inks have led to high-quality
electronic thin films, which have shown promise in photo-
voltaics,1,3 thermoelectrics,4 thin film transistors,5 and phase
change memory materials.6
Molecular semiconductor inks are comprised of molecular
solutes and are typically prepared using one of two methods:
(i) a discrete inorganic coordination complex can be dissolved
as a premade soluble precursor,7,8 or (ii) bulk materials having
the desired elemental composition of the recovered semi-
conductor may be directly dissolved.9−13 Direct dissolution
offers a number of advantages over other methods, that is, in a
single step, cheap bulk materials can be used as elemental
sources for high-quality semiconductors using a simple
“dissolve and recover” process. Additionally, this approach
offers the ability to tune semiconductor composition because
Received: February 2, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.6b00243
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Thermal Recovery of SnS. Solvent removal entailed heating the
drop-cast inks on a temperature-controlled hot plate to 100 °C under
flowing nitrogen followed by heating to 107 °C under vacuum (24 h).
The deposited inks were then annealed on a temperature-controlled
hot plate to 350 °C under flowing nitrogen followed by a slow cool
down to room temperature.
and the molecular condensation reactions that occur upon
annealing to recover the semiconductor.11,14−19 This knowl-
edge has provided key insights into the phase formation
mechanism of as-deposited films and has ultimately allowed for
more control over the resulting crystalline phase, composition,
band gap, and film morphology.
Material Characterization. Solution 119Sn NMR spectra were
recorded on a Varian 500 MHz spectrometer at 186.5 MHz. Spectra
were obtained at room temperature without deuterium locking of the
main magnetic field. Each experiment used a 10 μs pulse width and a
0.5 s relaxation delay and collected 1024 scans. A spectral window
from −1000 to +1500 ppm was used to search for the signals in 500
ppm increments. δ119Sn were referenced to the internal standard of the
probe. The NMR samples were prepared in 5 mm J-Young tubes
under nitrogen. Electrospray ionization mass spectrometry (ESI-MS)
was performed using a Waters LCT Premier operated using
electrospray ionization with negative ion detection. The carbon,
hydrogen, nitrogen, and sulfur chemical composition of the dried
solutions was analyzed using a Flash 2000 CHNS Elemental Analyzer.
Samples (3 mg) were prepared for elemental analysis by drying under
vacuum at 107 °C for 24 h. Raman spectra of the solutions were
recorded under ambient conditions using a Horiba Jobin Yvon
spectrometer, equipped with a liquid sample holder. An excitation
source of 785 nm from a diode laser was employed at a power level of
50 mW. UV−vis absorption spectra were taken in a nitrogen
atmosphere using a PerkinElmer Lambda 950 spectrophotometer.
SnO and SnS solutions were prepared as described above. A 1:10 vol/
vol mixture of EDT−en was used for the blank, and each solution was
diluted with this same solvent mixture prior to data collection to reach
the reported concentrations. The spectrum of the EDT−en solvent
mixture was measured against an empty cuvette. FT-IR spectra were
recorded on a Bruker Vertex 80 spectrometer. All samples were drop
cast onto ZnSe windows and either dried at 107 °C under vacuum for
24 h or annealed to 350 °C and then immediately cooled. The
resulting spectra were then baseline corrected using a rubber band
mode with 5 iterations and normalized to the largest peak. Thermal
analysis measurements were made on a Netzsch DSC/TGA-MS/IR
STA449 F1 using sample sizes of ∼20 mg. Argon was used as a
protective and flow gas. Mass flow controllers were set to 40 mL min−1
for the protective gas and 10 mL min−1 for the purge gas. The thermal
range examined was 50−500 °C with a heating rate of 10 °C min−1.
Powder X-ray diffraction (XRD) data was collected with a Rigaku
Ultima IV diffractometer in parallel beam geometry (2 mm beam
width) using Cu Kα radiation (λ = 1.54 Å). Scanning electron
microscope-energy-dispersive X-ray spectroscopy (SEM-EDX) was
used for elemental analysis on a JEOL JSM-6610 scanning-electron
microscope. Diffuse reflectance UV−vis−NIR spectroscopy was
performed using the reflectivity mode in a PerkinElmer Lambda 950
that was equipped with a 150 mm integrating sphere. The samples
were prepared by thoroughly grinding 9 mg of material with 290 mg of
MgO using a powder sample holder at the end of the integrating
sphere.
In addition to hydrazine, an alternative binary solvent system
comprised of alkanethiols and 1,2-ethylenediamine (en) has
been developed by our group for the facile dissolution of a wide
array of metals, metal oxides, and metal chalcogenides. The
dissolved materials can be directly recovered as phase-pure
binary metal chalcogenides at relatively low temper-
atures.12,13,20,21 The solvent system can also be used to dissolve
multiple bulk components in order to controllably deposit
more compositionally complex semiconductors (e.g., CuIn-
(SxSe1−x)2, Cu2ZnSnS4−xSex or CZTSSe, and Cu2SnSe3).22−26
For example, alkanethiol−en-processed CZTSSe photovoltaic
devices have been reported with power conversion efficiencies
> 7%.27
Recently, the EDT−en solvent system has been applied to
the solution processing of SnS semiconductor thin films.20,21,24
SnS is an indirect narrow band gap semiconductor (Eg = 1.08
eV) comprised of earth-abundant elements that has been
receiving increasing attention as an absorber material for single-
junction solar cells,28 with devices currently achieving power
conversion efficiencies > 4% where the absorber layer was
deposited by atomic layer deposition.29 The solution deposition
of SnS may ultimately allow for the low-cost fabrication of large
area devices. Bulk SnO and SnS have been previously dissolved
using the alkanethiol−en solvent system for the recovery of
phase-pure SnS,20,21,24 and herein, we report the dissolution of
elemental Sn toward the solution processing of SnS for the first
time; however, nothing has heretofore been reported on the
nature of the dissolved species. Toward this end, here we
explore the solute identity of three bulk tin-containing materials
(i.e., Sn, SnO, and SnS) dissolved in a 1,2-ethanedithiol
(EDT)−en solvent system to give semiconductor inks to SnS.
EXPERIMENTAL SECTION
■
Materials and Methods. 1,2-Ethylenediamine (en, ≥99%), tin(II)
sulfide (SnS, 99.99%), and tin(IV) chloride pentahydrate (SnCl4·
5H2O, 98%) were purchased from Sigma-Aldrich. 1,2-Ethanedithiol
(EDT, 98+%), elemental tin (Sn, 99.85%), and tin(II) oxide (SnO,
99%) were purchased from Alfa Aesar. Tin(II) chloride dihydrate
(SnCl2·2H2O, 98%) was purchased from Strem Chemicals. 1,2-
Ethylenediamine and 1,2-ethanedithiol were purged using nitrogen gas
for a minimum of 10 min before use in the dissolution experiments; all
other reagents and solvents were used as received without further
purification.
RESULTS AND DISCUSSION
■
Preparation of Inks. The dissolved inks were prepared by mixing
the appropriate bulk material (50 mg) with en (1.00 mL) and EDT
(0.1 mL) at room temperature under a nitrogen atmosphere. The
solutions were stirred for ca. 5 days to yield transparent light-yellow
solutions. Complete dissolution was observed for bulk SnS and SnO.
While elemental Sn did not fully dissolve, excess Sn was easily
removed by filtration through a 450 nm PTFE membrane.
Dissolution of Sn, SnO, and SnS. Inks were prepared by
the dissolution of bulk Sn, SnO, and SnS in a 1:10 vol/vol
mixture of EDT−en according to previous reports.12 The
solubilities, expressed as wt % solute in the saturated solution of
EDT−en at room temperature and ambient pressure, of both
SnO and SnS have been previously determined.20,21 SnO when
dissolved at room temperature for a period of 15 h was soluble
up to 15 wt % in a 1:4 vol/vol ratio of EDT−en, while SnS was
soluble up to 12 wt % in an a 1:11 vol/vol ratio of EDT−en
after mild heating (50 °C, 12 h). Here, a lower concentration of
4.5 wt % in a 1:10 vol/vol mixture of EDT−en was used for all
three materials due to the lower solubility limit of elemental Sn
in the solvent system. At this concentration, both SnO and SnS
were fully soluble after stirring for 24 h. Longer dissolution
Synthesis of (EDT)2Sn(IV) (1). 1 was synthesized by following a
literature procedure30 where an aqueous mixture of 1,2-ethanedithiol
and potassium hydroxide was added to a solution of SnCl4·5H2O; the
resulting white precipitate was recrystallized from DCM to yield
colorless crystals. Anal. Calcd for C4H8S4Sn: C, 15.85; H, 2.66; S,
1
42.32. Found: C, 15.96; H, 2.73; S, 43.38. H NMR (CDCl3, 25 °C,
500 MHz): δ 3.22 ppm. 13C{1H} NMR (CDCl3, 25 °C, 125 MHz): δ
35.9 ppm. 119Sn NMR (CDCl3, 25 °C, 186.5 MHz): δ 281 ppm. Mp:
186 °C.
B
DOI: 10.1021/acs.inorgchem.6b00243
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
times were required for elemental Sn; even after 5 days a small
amount of Sn remained undissolved. Despite the relatively
reduced solubility of elemental Sn, all three materials gave light-
yellow, optically transparent solutions that were free of visible
scattering and stable for a period of months at a concentration
of 4.5 wt % (Figure 1a).
resulting white precipitate was recrystallized from dichloro-
methane to yield colorless crystals that were confirmed to be 1
by NMR spectroscopy, melting point determination, and
combustion analysis (vide supra). Compound 1 has been
previously characterized using NMR with a reported δ119Sn
=
277 ppm in thioformaldehyde.31 While this value does not
match the δ119Sn = 217 ppm for the dissolved Sn, SnO, and SnS
species, it was plausible that the EDT−en solvent system could
cause differences in chemical shift. To test this hypothesis, 1
was characterized in several solvents using 119Sn NMR (Figure
2). The δ119Sn of 1 varied quite drastically between +280 and
Figure 1. (a) Photograph of bulk elemental Sn, SnO, and SnS
dissolved in a 1:10 vol/vol EDT−en solvent mixture at 4.5 wt %. (b)
Solution 119Sn NMR spectra of Sn, SnO, and SnS dissolved in EDT−
en giving a single resonance at δ119Sn = 217 ppm.
Solution 119Sn NMR Spectroscopy. Solution 119Sn NMR
spectroscopy was used to probe the identity of the dissolved Sn,
SnO, and SnS in EDT−en. The 119Sn isotope is useful in this
regard because it yields narrow signals over a very wide
chemical shift (δSn119) range of ca. 6500 ppm for organotin and
inorganic tin compounds, making the chemical shift sensitive to
even small structural changes.31−33 For example, a correlation
exists between the chemical shift range and the tin coordination
number, whereby irrespective of the formal oxidation state, the
coordination number can be inferred from 119Sn chemical
shift.32,33 The 119Sn NMR spectra, obtained directly after
dissolution of the bulk Sn, SnO, and SnS materials in the
EDT−en solvent system, reveal an identical, single sharp
resonance at δSn119 = 217 ppm for all three materials (Figure
1b). Thus, even though the three systems have different bulk
material compositions (Sn, SnO, and SnS) and oxidation states
(Sn0 and Sn2+), they all yield an equivalent solute on the NMR
time scale upon dissolution in EDT−en. The lack of JSn−Sn
coupling in the 119Sn NMR spectra suggests that the solutes are
mononuclear. The observed resonance at δSn119 = 217 ppm is
consistent with those chemical shifts reported for mononuclear
tin tetrathiolate compounds in the relatively narrow chemical
shift window of ca. 20−300 pm; the chemical shifts of tin
tetrathiolates are in the range from 138 to 26 ppm for
(EtS)4Sn(IV) and (tBuS)4Sn(IV), respectively.34,35 This
chemical shift range is extended further downfield when tin is
coordinated by a chelating dithiolate, such as EDT. For
example, (EDT)2Sn(IV) gives a chemical shift of δSn119 = 277
ppm.31 From these initial 119Sn NMR spectra it can be inferred
that regardless of the bulk material that was dissolved, a single
four-coordinate tin thiolate compound was formed in the
EDT−en solvent system.
Figure 2. 119Sn NMR spectra of 1 in various solvents showing the
dependence of δ119Sn on solvent and tin coordination number. Red and
blue shaded areas indicate the general δ119Sn ranges for four- and six-
coordinate tin compounds, respectively.
−350 ppm depending on the solvent used; recall that the
chemical shifts are very characteristic of tin coordination
environments where δ119Sn of 20−300 ppm are indicative of
four-coordinate tin thiolates, while upfield chemical shifts
ranging from δ119Sn of ca. −125 to −515 ppm are indicative of
six-coordinate tin compounds.36 In CDCl3, compound 1 gives a
single peak in the 119Sn NMR with a δ119Sn of 281 ppm,
consistent with a four-coordinate tin thiolate and in good
agreement with literature values.31 In pure EDT, the δ119Sn
shifted slightly upfield to 269 ppm and was significantly
broadened, implying dynamic exchange between the ethanedi-
thiolate ligands and EDT on the NMR time scale. Addition of
deprotonated EDT2− to 1 in D2O resulted in the binding of an
additional ethanedithiolate ligand to 1 rather than exchange
with the existing ligands, as evidenced by the large upfield shift
to −352 ppm, a region characteristic of six-coordinate tin
compounds.37,38 Similarly, when en was used as a solvent it
appeared to coordinate to 1, yielding an apparent six-coordinate
compound having a δ119Sn at −263 ppm. The negative ion ESI-
MS spectrum of 1 in the presence of en reveals that 1 is readily
chelated by en, as evidenced by the main ion cluster observed at
m/z = 362.9 that corresponds to a formula of [Sn-
(EDT)2(en)]− (see Supporting Information, Figure S2).
Finally, when 1 was dissolved in the binary EDT−en solvent
system, the resulting δ119Sn of −346 ppm was significantly
broadened and located in a chemical shift range also consistent
with a six-coordinate tin compound.
Given the indication by 119Sn NMR data that the molecular
solute was a four-coordinate tin thiolate compound, it was
hypothesized that the dissolved species could be similar in
nature to the known compound (EDT)2Sn(IV) (1). To test
this hypothesis, 1 was synthesized following a literature
procedure whereby an aqueous mixture of EDT and KOH
was added to a solution of SnCl4·5H2O in water.30 The
C
DOI: 10.1021/acs.inorgchem.6b00243
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
These results collectively suggest that the molecular solute of
Sn, SnO, and SnS dissolved in EDT−en was not 1, since the
Sn(IV) compound appears to be easily chelated by EDT2− or
en in the solvent system to form six-coordinate compounds. In
contrast, the dissolved Sn, SnO, and SnS all give a molecular
solute that remains four coordinate in the EDT−en solvent
system. Four-coordinate Sn(IV) compounds have the propen-
sity to go six coordinate in the presence of donor solvents,
whereas the analogous four-coordinate Sn(II) compounds do
not.39 On the basis of the apparent coordination number of
four in EDT−en by 119Sn NMR for the molecular solutes of the
dissolved Sn, SnO, and SnS, it appears that the dissolved
species are in the Sn2+ oxidation state. The influence of tin
oxidation state (Sn2+ vs Sn4+) on δ119Sn is rarely reported in the
literature; thus, further control experiments were performed to
help identify the dissolved tin species. As a simple test, Sn2+ and
Sn4+ in the form of SnCl2·2H2O and SnCl4·5H2O, respectively,
were dissolved in EDT−en and characterized using 119Sn NMR
(see Supporting Information, Figure S3). Dissolution of SnCl4·
5H2O yielded a single resonance at δ119Sn = −347 ppm,
consistent with our previous experiment showing 1 forms a six-
coordinate compound when dissolved in EDT−en. Dissolution
of SnCl2·2H2O in EDT−en, on the other hand, gives a single
resonance by 119Sn NMR that is nearly identical to that of the
dissolved Sn, SnO, or SnS at δ119Sn = 210 ppm. It is likely that
this species represents the same compound as that formed from
the dissolution of Sn, SnO, and SnS, as a 1:1 wt/wt mixture of
SnCl2·2H2O and SnO dissolved in EDT−en yields a single
resonance at δ119Sn = 216 ppm. This further supports the
hypothesis that the dissolved Sn, SnO, and SnS species are in
the 2+ oxidation state and remain four coordinate in the EDT−
en solvent system, with en acting as a noncoordinating solvent
and possible countercation (in the form of enH+/enH22+).
To further validate this hypothesis, a DFT calculation was
carried out to model the gas-phase 119Sn NMR chemical shift of
bis(1,2-ethanedithiolate)tin(II). The calculation was performed
using Gaussian 09 at the B3PW91/TZP level of theory.
Vibrational frequencies were used to characterize stationary
points as minima (number of imaginary frequency (NIMAG) =
0). The δ119Sn was calculated using direct implementation of the
gauge including atomic orbitals (GIAO) method on the gas-
phase-optimized geometry of B3PW91/TZP (see Supporting
Information, Figure S4), with the δ119Sn being referenced to
Me4Sn (calculated absolute shift σSn = 2751). The B3PW91
functional has been previously found to give reliable trends for
tin compounds.40 The calculated δ119Sn of bis(1,2-
ethanedithiolate)tin(II) (with σSn = 2443) was found to be
308 ppm, which is within the experimentally determined 119Sn
NMR window of four-coordinate tin thiolate compounds and
lies within the range previously reported between experimental
and gas-phase calculated δ119Sn for a variety of Sn(II)
compounds.41
Figure 3. Solution Raman spectra of Sn, SnO, and SnS dissolved in
EDT−en at 4.5 wt %.
and 322 ( 3) cm−1. It should be further noted that there are no
significant bands in these spectra associated with the v(Sn−N)
symmetric stretch at ∼178 cm−1, which is readily observed with
coordination of amines to 1 (see Supporting Information,
Figure S6).42,43 On this basis, it is unlikely that en in the solvent
mixture is coordinating to tin, leaving the molecular solute
solely ligated by two dithiolate ligands (vide supra).
UV−vis spectroscopy was used to probe the possibility of
polysulfide formation from the sulfur as a result of SnS
dissolution. The EDT−en solvent system has a strong
absorption onset in the UV at ca. 320 nm with a smaller
absorption band at λmax = 365 nm. To test for polysulfide anion
formation, solutions of dissolved SnO (having no sulfur) were
compared against dissolved SnS. Solutions of dissolved SnO
and SnS both show strong, red-shifted absorption onsets with
no other distinguishable absorption features at lower energies
(see Supporting Information, Figure S7). The dissolved SnO
and SnS also both show a similar slight red shift in absorption
onset with an increase in concentration and no other
absorption bands at lower energies. While UV−vis spectrosco-
py has been show to have limitations for detecting individual
polysulfide species in nonaqueous solvents due to concen-
tration-dependent disproportionation equilibria, it has been
established that higher order polysulfides anions have
absorption maxima at wavelengths > 330 nm, while sulfides
and disulfides (such as S2 and S2−) absorb at shorter
2−
wavelengths < 280 nm.44,45 As a result of the lack of absorption
features at lower energies, higher order polysulfide anions may
be ruled out as the fate of the sulfur when dissolving SnS at
these concentrations; however, smaller sulfides or disulfides
(supported by enH+ or enH2 cations) may be present with
2+
absorption features at wavelengths that are masked by the
strong absorption onset of the solutes at 375−400 nm.
Conversion to SnS upon Annealing. Coupled thermog-
ravimetric analysis/differential scanning calorimetry-mass spec-
trometry (TGA/DSC-MS) was used to study the decom-
position of the dried inks. TGA traces for the dried inks of
dissolved Sn, SnO, and SnS all display the same thermal
decomposition characteristics from 100 to 500 °C, with a
single-step mass loss event ranging from 42% to 46% occurring
at ca. 300 °C (Figure 4a). The similarities in decomposition
behavior are again consistent with all three inks containing
identical molecular solutes upon the dissolution of Sn, SnO,
and SnS in EDT−en. By DSC, the dried inks display exotherms
at ∼275 °C and endotherms at ∼315 °C, associated with ∼15%
and ∼30% mass losses during decomposition, respectively. The
major volatile decomposition products observed in the mass
spectra at ∼275 and ∼315 °C are consistent with EDT
Raman and UV−vis Spectroscopy. Raman spectra
collected for Sn, SnO, and SnS dissolved in EDT−en support
the conclusion that identical solutes exist in solution. Figure 3
gives the Raman spectra of Sn, SnO, and SnS dissolved in the
EDT−en system. Vibrational modes observed after dissolution
of Sn, SnO, and SnS are attributed to the molecular solute,
assuming that the Raman bands from the EDT−en solvent
mixture remain constant (see Supporting Information, Figure
S5). The Raman spectra obtained after dissolution of Sn, SnS,
and SnO in EDT−en are nearly identical to one another in
peak position, with each possessing major bands at ∼197, 264,
D
DOI: 10.1021/acs.inorgchem.6b00243
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 4. (a) TGA/DSC traces comparing the decomposition of dried inks of Sn, SnO, and SnS dissolved in EDT−en. (b) FT-IR spectra of the
dried ink corresponding to SnO dissolved in EDT−en before and after annealing to 350 °C. (c) Powder XRD patterns of orthorhombic Pnma SnS
obtained after annealing the dried inks to 350 °C under flowing nitrogen. (d) Tauc plots derived from UV−vis−NIR spectra showing the indirect
optical band gaps of the recovered SnS.
pyrolysis from 1, with the molecular ion peak at m/z = 120
corresponding to [C4H8S2]+ from EDT decomposition (see
Supporting Information, Figure S8), with no further signals
from en being observed. Therefore, it is reasonable to infer that
the dry inks contain coordinated EDT similar to 1, with en
being completely expelled during the drying stage of the sample
preparation (∼105 °C under vacuum, 24 h).
after annealing the dried inks to 350 °C were investigated by
diffuse reflectance UV−vis−NIR spectroscopy using an
integrating sphere (see Supporting Information, Figure S9).
Figure 4d shows the indirect optical band gap approximations
obtained from Tauc plots of the UV−vis−NIR diffuse
reflectance spectra. The SnS material derived from all three
inks possess the same approximate indirect band gap transitions
(Eg,ind = 1.1 eV), which agree well with literature values.20,29
Conversion of the light-yellow inks to the corresponding
black SnS material was achieved using a mild annealing step
under flowing nitrogen. The minimum annealing temperature
required for conversion to SnS was determined by TGA, which
showed that thermal decomposition events were completed by
350 °C. FT-IR spectra of the annealed SnS lacked any features
attributable to organic species, confirming that 350 °C is an
appropriate annealing temperature for the recovery of organic-
free material (Figure 4b). Powder X-ray diffraction (XRD)
established that the material obtained after annealing the dried
inks to 350 °C was crystalline and phase-pure orthorhombic
(Pnma) SnS. Figure 4c gives the diffraction patterns of the
materials resulting from the inks for dissolved Sn, SnO, and
SnS. The diffraction patterns are all identical, possessing a 100%
intensity peak (2θ = 31.9°) indexed to the (111) reflection of
orthorhombic Pnma SnS structure (JCPDS no. 01-073-1859).
No crystalline Sn, SnOx, or other sulfide (SnS2, Sn2S3, Sn3S4,
etc.) impurities were observed by XRD for any of the samples,
further indicating their phase purity. Elemental analysis of five
randomly selected large areas of powdered samples by scanning
electron microscope energy-dispersive X-ray spectroscopy
(SEM-EDX) gave an average composition of 53 atom % Sn
to 47 atom % S for SnS recovered from the ink with dissolved
Sn, 57 atom % Sn to 43 atom % S for SnS recovered from the
ink with dissolved SnO, and 56 atom % Sn to 44 atom % S for
SnS recovered from the ink with dissolved SnS. These atomic
percentages closely match those obtained for SnS recovered
from EDT−en in previous studies20 and also for a commercial
SnS standard. Moreover, the optical properties of SnS obtained
CONCLUSION
■
We investigated the nature of the solutes produced after
dissolution of three different bulk materials (i.e., Sn, SnO, and
SnS) in the binary EDT−en solvent system. Our findings reveal
that upon dissolution of these three bulk materials a single and
identical molecular solute is produced, as evidenced by 119Sn
NMR, Raman spectroscopy, and TGA/DSC. These data
suggest that the likely identity of the molecular solute is
bis(1,2-ethanedithiolate)tin(II). While this compound has not
been previously reported as an isolated compound in the
literature, it is likely forming from the dissolution of these tin-
containing species in EDT−en. In all cases, mild annealing of
the resulting semiconductor inks yields crystalline and phase-
pure orthorhombic SnS with an indirect optical band gap of 1.1
eV. Therefore, since all of the bulk materials converge to an
identical molecular solute in the alkahest solution, which upon
annealing the semiconductor inks give phase-pure material, the
most inexpensive sources of tin (such as elemental Sn) can be
readily used for the facile deposition of SnS.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.6b00243.
E
DOI: 10.1021/acs.inorgchem.6b00243
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
NMR spectra of 1; ESI-MS of 1 in en; 119Sn NMR
Article
(20) Antunez, P. D.; Torelli, D. A.; Yang, F.; Rabuffetti, F. A.; Lewis,
N. S.; Brutchey, R. L. Chem. Mater. 2014, 26, 5444−5446.
(21) McCarthy, C. L.; Webber, D. H.; Schueller, E. C.; Brutchey, R.
L. Angew. Chem., Int. Ed. 2015, 54, 8378−8381.
(22) Tian, Q.; Cui, Y.; Wang, G.; Pan, D. RSC Adv. 2015, 5, 4184−
4190.
spectra of SnCl2 and SnCl4 dissolved in EDT−en; gas-
phase-optimized geometry of bis(1,2-ethanedithiolate)-
tin(II) by DFT; Raman spectra of the EDT−en solvent
mixture and of 1 dissolved in EDT−en; UV−vis spectra
of dissolved SnO and SnS; mass spectra of the volatiles
produced by TGA/DSC upon decomposition of the
semiconductor inks; diffuse reflectance UV−vis-NIR
spectra of the recovered SnS (PDF)
(23) Arnou, P.; Cooper, C. S.; Malkov, A. V.; Bowers, J. W.; Walls, J.
M. Thin Solid Films 2015, 582, 31−34.
(24) Liu, F.; Zhu, J.; Xu, Y.; Zhou, L.; Li, Y.; Hu, L.; Yao, J.; Dai, S.
Chem. Commun. 2015, 51, 8108−8111.
(25) Liu, F.; Zhu, J.; Li, Y.; Wei, J.; Lv, M.; Xu, Y.; Zhou, L.; Hu, L.;
Dai, S. J. Power Sources 2015, 292, 7−14.
AUTHOR INFORMATION
Corresponding Author
*E-mail: brutchey@usc.edu.
■
(26) Liu, F.; Zhu, J.; Hu, L.; Zhang, B.; Yao, J.; Nazeeruddin, Md. K.;
Gratzel, M.; Dai, S. J. Mater. Chem. A 2015, 3, 6315−6323.
̈
(27) Tian, Q.; Cui, Y.; Wang, G.; Pan, D. RSC Adv. 2015, 5, 4184−
4190.
Notes
(28) Albers, W.; Haas, C.; Vink, H. J.; Wasscher, J. D. J. Appl. Phys.
1961, 32, 2220−2225.
The authors declare no competing financial interest.
(29) Sinsermsuksakul, P.; Sun, L.; Lee, S. W.; Park, H. H.; Kim, S. B.;
Yang, C.; Gordon, R. G. Adv. Energy Mater. 2014, 4, 1400496.
(30) Epstein, L. M.; Straub, D. K. Inorg. Chem. 1965, 4, 1551−1554.
(31) Davies, A. G.; Slater, S. D.; Povey, D. C.; Smith, G. W. J.
Organomet. Chem. 1988, 352, 283−294.
ACKNOWLEDGMENTS
■
This work was supported by the National Science Foundation
under DMR-1506189. J.J.B. acknowledges the National Science
Foundation for a Graduate Research Fellowship. J.D.P.-A.
thanks ISU AGEP for a Postdoctoral Fellowship. Acknowl-
edgement is also made to Dr. J. Greaves for performing ESI-MS
measurements and providing assistance and access to the Mass
Spectrometry facilities at UC Irvine.
(32) Gielen, M. In Tin Chemistry Fundamentals Frontiers, and
Applications; Davies, A. G., Pannell, K. H., Tiekink, E. R. T., Eds.;
Wiley: U.K., 2008.
(33) Casella, G.; Ferrante, F.; Saielli, G. Eur. J. Org. Chem. 2009,
2009, 3526−3534.
(34) Kennedy, J. D.; McFarlane, W.; Pyne, G. S.; Clarke, P. L.;
Wardell, J. L. J. Chem. Soc., Perkin Trans. 2 1975, 2, 1234−1239.
(35) Holmes, R. R.; Shafieezad, S.; Holmes, J. M.; Day, R. O. Inorg.
Chem. 1988, 27, 1232−1237.
REFERENCES
■
(1) Todorov, T. K.; Gunawan, O.; Gokmen, T.; Mitzi, D. B. Prog.
Photovoltaics 2013, 21, 82−87.
(2) Mitzi, D. B. J. Mater. Chem. 2004, 14, 2355−2365.
(3) Wang, W.; Winkler, M. T.; Gunawan, O.; Gokmen, T.; Todorov,
T. K.; Zhu, Y.; Mitzi, D. B. Adv. Energy Mater. 2014, 4, 1301465.
(4) Milliron, D. J.; Raoux, S.; Shelby, R. M.; Jordan-Sweet, J. Nat.
Mater. 2007, 6, 352−356.
(5) Milliron, D. J.; Mitzi, D. B.; Copel, M.; Murray, C. E. Chem.
Mater. 2006, 18, 587−590.
(6) Wang, R. Y.; Caldwell, M. A.; Jeyasingh, R. G. D.; Aloni, S.;
Shelby, R. M.; Wong, H.-S. P.; Milliron, D. J. J. Appl. Phys. 2011, 109,
113506.
(7) Zhou, H.; Hsu, C.- J.; Hsu, W.-C.; Duan, H.-S.; Chung, C.-H.;
Yang, W.; Yang, Y. Adv. Energy Mater. 2013, 3, 328−336.
(8) Sun, Y.; Zhang, Y.; Wang, H.; Xie, M.; Zong, K.; Zheng, H.; Shu,
Y.; Liu, J.; Yan, H.; Zhu, M.; Lau, W. J. Mater. Chem. A 2013, 1, 6880−
6887.
(36) Otera, J. J. Organomet. Chem. 1981, 221, 57−61.
(37) Melen, R. L.; McPartlin, M.; Wright, D. S. Dalton Trans. 2011,
40, 1649−1651.
(38) de Assis, F.; Chohan, Z. H.; Howie, R. A.; Khan, A.; Low, J. N.;
Spencer, G. M.; Wardell, J. L.; Wardell, S. M. S. V. Polyhedron 1999,
18, 3533−3544.
(39) Zubieta, J. A.; Zuckerman, J. J. Prog. Inorg. Chem. 1978, 24,
251−475.
(40) Avalle, P.; Harris, R. K.; Karadakov, P. B.; Wilson, P. J. Phys.
Chem. Chem. Phys. 2002, 4, 5925−5932.
(41) Wang, L.; Kefalidis, C. E.; Roisnel, T.; Sinbandhit, S.; Maron, L.;
Carpentier, J.-F.; Sarazin, Y. Organometallics 2015, 34, 2139−2150.
(42) Shamir, J.; Starostin, P.; Peruzzo, V. J. Raman Spectrosc. 1994,
25, 251−254.
(43) Buttrus, N. H.; Suliman, M. M.; Al-Allaf, T. A. K. Synth. React.
Inorg. Met.-Org. Chem. 2001, 31, 837−848.
(9) Mitzi, D. B.; Kosbar, L. L.; Murray, C. E.; Copel, M.; Afzali, A.
Nature 2004, 428, 299−303.
(44) Canas, N. A.; Fronczek, D. N.; Wagner, N.; Latz, A.; Friedrich,
̃
K. A. J. Phys. Chem. C 2014, 118, 12106−12114.
(10) Mitzi, D. B. Adv. Mater. 2009, 21, 3141−3158.
(11) Yuan, M.; Mitzi, D. B. Dalton Trans. 2009, 6078−6088.
(12) Webber, D. H.; Brutchey, R. L. J. Am. Chem. Soc. 2013, 135,
15722−15725.
̂
(45) Barchasz, C.; Molton, F.; Duboc, C.; Lepretre, J.-C.; Patoux, S.;
Alloin, F. Anal. Chem. 2012, 84, 3973−3980.
(13) Webber, D. H.; Buckley, J. J.; Antunez, P. D.; Brutchey, R. L.
Chem. Sci. 2014, 5, 2498−2502.
(14) Mitzi, D. B.; Yuan, M.; Liu, W.; Kellock, A. J.; Chey, S. J.;
Deline, V.; Schrott, A. G. Adv. Mater. 2008, 20, 3657−3662.
(15) Bob, B.; Lei, B.; Chung, C.-H.; Yang, W.; Hsu, W.-C.; Duan, H.-
S.; Hou, W. W.; Li, S.-H.; Yang, Y. Adv. Energy Mater. 2012, 2, 504−
522.
(16) Nørby, P.; Overgaard, J.; Christensen, P. S.; Richter, B.; Song,
X.; Dong, M.; Han, A.; Skibsted, J.; Iversen, B. B.; Johnsen, S. Chem.
Mater. 2014, 26, 4494−4504.
(17) Chung, C.-H.; Li, S.- H.; Lei, B.; Yang, W.; Hou, W. W.; Bob, B.;
Yang, Y. Chem. Mater. 2011, 23, 964−969.
(18) Yang, B.; Xue, D.-J.; Leng, M.; Zhong, J.; Wang, L.; Song, H.;
Zhou, Y.; Tang, J. Sci. Rep. 2015, 5, 10978.
(19) Yang, W.; Duan, H.-S.; Cha, K. C.; Hsu, C.-J.; Hsu, W.-C.;
Zhou, H.; Bob, B.; Yang, Y. J. Am. Chem. Soc. 2013, 135, 6915−6920.
F
DOI: 10.1021/acs.inorgchem.6b00243
Inorg. Chem. XXXX, XXX, XXX−XXX