Article pubs.acs.org/IC
Structural and Optical Properties of Sb-Substituted BiSI Grown from Sulfur/Iodine Flux Ryan A. Groom, Allison Jacobs, Marisa Cepeda, Rachel Drummey, and Susan E. Latturner* Department of Chemistry and Biochemistry, Florida State University, 95 Chieftan Way, Tallahassee, Florida 32306, United States S Supporting Information *
ABSTRACT: Bismuth and antimony were reacted in sulfur/iodine flux mixtures at various temperatures and iodine concentrations to explore the effects of these variables on the synthesis and properties of Bi1−xSbxSI products. The products grow as crystals; microprobe elemental analysis and UV/vis/NIR spectroscopy of the Bi1−xSbxSI solid solutions indicate that substitution is homogeneous within individual crystals but varies up to 15% between crystals within each synthesis batch. Raman spectra show a twomode behavior upon substitution, indicating covalent bonding within the structure, and TEM/SEM data confirm no presence of nanoclustering or segregation within the crystals.
■
INTRODUCTION Many heavy-metal sulfides and sulfide iodides such as Bi2S3, SbSI, CuInS2, and mixed transition metal sulfides are semiconductors with potential applications as photovoltaics, thermoelectrics, catalysts, and scintillation materials.1−8 Several members of the AVBVICVII chalcohalide family are known to have ferroelectric, photoconductivity, and piezoelectric properties.9,10 BiSI and SbSI undergo ferroelectric transitions at Tc = 113 and 298 K, respectively; the bismuth analogue also has photoconductivity properties and shows potential as an ionizing radiation detection material.9,11 In recent years BiSI and SbSI have been synthesized using a variety of methods, including solvothermal reactions, spray pyrolysis deposition, vapor transport, and chemical vapor deposition.12−14 Vapor-phase transport methods require careful control of temperature gradients (often mandating the use of multizone furnaces), and differences in equilibria of gas-phase intermediates can hinder the ability to grow solid solution phases. Solvothermal synthesis allows for lower reaction temperatures, but it often requires costly and toxic precursors and solvents such as thiourea, thioacetamide (TAA), tert-butyl sulfide, hexamethyldisilathiane, chloroform, and polyols.15−17 Many of these difficulties can be avoided by utilizing flux chemistry. This is done by using at least one low-melting reactant in a large excess. It melts and dissolves other reagents when heated, allowing a solution-phase reaction with no need for solvent. The solution state helps to lower diffusion barriers, and slow cooling facilitates the growth of bulk crystals. In this work, a mixture of sulfur and iodine is used as a flux. The addition of iodine acts to lower the melting point and the viscosity of sulfur.18,19 It also acts as a reactant, aiding in the formation of sulfide-iodide products. BiSI, SbSI, and (Bi1−xSbx)SI were grown by reacting bismuth and antimony at various ratios in S/I2 mixtures in order to investigate the © XXXX American Chemical Society
effects of this substitution on the properties of the material. Substitution of the halide anions has been previously explored in the BiSX family (X = iodide, bromide, chloride).20−22 Those reports focused on Vegard’s law deviations in BiSX1−xX′x, the potential for BiSBr1−xIx to be used as a tunable band gap material, and the vibrational spectroscopy of the ferroelectric semiconductor SbSBr1−xIx. For this study, the structural and electronic effects of metal substitution in flux-grown Bi1−xSbxSI were observed by crystallography, DSC-TGA, SEM/TEM, Raman spectroscopy, and diffuse reflectance UV/vis measurements. Raman data of (Bi1−xSbx)SI solid solutions shows twomode behavior indicative of covalent bonding.
■
EXPERIMENTAL SECTION
Synthesis. Stock S/I2 flux preparation and synthesis methods were carried out as described in our recent work on the related bismuth sulfide iodide Bi13S18I2, which is the predominant product for reactions with low iodine content.23 After exploration of a variety of sulfur to iodine ratios, it was determined that Bi1−xSbxSI compounds were optimally obtained using a flux mixture of 10 mmol/0.5 mmol of S/I2. This S/I2 flux mixture was prepared by placing 6.414 g (0.2 mol) of sulfur and 2.538 g (0.01 mol) of iodine in a sealed Pyrex tube, which was heated to 180 °C and then shaken to ensure mixing. The flux was then cooled and ground to use in subsequent reactions. The Bi/Sb/S/I2 flux reactions were prepared by loading 0.447 g of the stock S/I2 mixture (which provides 10 mmol of sulfur and 0.5 mmol of I2) along with 1 mmol of either Bi or Sb or a mixture of both metals into quartz tubes. A piece of quartz wool was placed above the reaction mixture to act as a filter during centrifugation. The tubes were then sealed under vacuum while being cooled in an ice bath and then heated over 4 h to 300, 400, or 500 °C, held for 48 h at that temperature, and cooled to 130 °C over 72 h. The ampules were removed while hot, inverted, and centrifuged to remove excess molten Received: July 19, 2017
A
DOI: 10.1021/acs.inorgchem.7b01839 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry flux. The solid products were removed and washed with carbon disulfide to dissolve any excess flux remaining. Crystallographic Studies. All solid products were analyzed by powder X-ray diffraction (PXRD) on a PANalytical X’Pert PRO powder diffractometer with a Cu Kα radiation source, and the resulting powder patterns were compared to those in the accompanying ICDD database. Rietveld refinement was also performed to determine unit cell parameters to investigate the Vegard’s law behavior of the Bi1−xSbxSI solid solution. The refinements were done using X’Pert HighScore Plus (3.0d). Single-crystal X-ray diffraction data were collected on several Bi1−xSbxSI crystals to ensure that the crystal structure did not change across the solid solution. Crystals were mounted on glass fibers using epoxy. Data were collected with a Bruker APEX2 single-crystal diffractometer with an Mo Kα radiation source. The data were analyzed by direct methods and refined using the ShelXTL software package.24 Structural refinements of (Sb/Bi)SI crystals in space group Pnma were straightforward due to its ordered structure and the single crystallographic site for each element. Thermal Stability Studies. Differential scanning calorimetry thermogravimetric analysis (DSC-TGA) studies were carried out on a TA Instruments SDT Q600 apparatus. Samples were placed in an alumina cup and loaded onto a balance arm, with an empty alumina cup on the reference arm. The heating chamber was kept under an argon flow rate of 100 mL/min. The samples were heated to 100 °C at a rate of 10 °C/min and held there for 5 min, to drive off any solvents or volatile substances. The samples were then heated to 450 °C at a rate of 5 °C/min and then cooled to 100 °C at a rate of 5 °C/min. PXRD was run on samples after heating for phase analysis. Elemental Analysis. Products were analyzed by SEM-EDS using a JEOL JSM-5900 scanning electron microscope at 30 kV acceleration voltage for energy dispersive spectroscopy measurements. Crystals and powdered products were adhered to double-sided carbon tape on an aluminum SEM puck. Larger scale products (crystals or agglomerates) were positioned to have a flat surface facing perpendicular to the electron beam. Measurements were carried out on three to five regions for each sample, measured to at least 500 counts, to get an average representation of elemental atomic percent composition. Element mapping was also used on several crystals to further characterize homogeneity within each crystal. TEM Imaging. Samples were prepared for TEM by chopping crystals to submillimeter dimensions and sprinkling on TED PELLA 200 mesh copper film disks coated with Formvar and carbon. Transmission electron microscopy was carried out on a probe aberration corrected JEOL JEM-ARM200cF with a cold-field emission electron gun at 200 kV. The scanning transmission electron microscopy high angle annular dark field (STEM-HAADF) images were taken with a probe convergence semiangle of 11 mrad and collection angles of 76−174.6 mrad. The STEM resolution of the microscope is 0.78 Å. Optical Characterization. UV−vis−NIR diffuse reflectance measurements were taken using a PerkinElmer Lambda 900 spectrometer equipped with a Labsphere PELA-1000 integration sphere accessory. Each sample was ground with potassium bromide (Certified ACS Reagent, Fisher Chemical) to a dilution of 1% and pressed into a black plastic puck, topped with a quartz disk. Reflectance data were collected at room temperature from 2500 to 320 nm. Two detectors were used, changing from a PbS detector at long wavelengths (NIR−vis) to the PMT detector (UV−vis) at 860 nm; this causes a small artifact at this wavelength. The data were converted from percent reflectance (%R) measurements into Kubelka−Munk (K-M) units using PerkinElmer UV WinLab software. Raman spectra were collected using a JY Horiba LabRam HR800 system excited by a 17 mW Melles-Griot HeNe laser emitting at 633 nm. After the laser passed through a band-pass filter, an edge filter was used to couple the laser beam into the microscope (Olympus BX30) by total reflection. The beam was focused on crystalline or powder products placed on a glass slide using a microscope objective 50× IR (Leica N.A. 0.80). Scattered radiation was collected by the objective, and the laser radiation was filtered out by the edge filter with Raman
scattering coupled into the 800 mm focal length spectrograph through a confocal hole. A grating dispersed the light onto a 1024 × 256 element open CCD detector. The laser power at the sample was 3 mW. Spectra were collected under ambient conditions over a spectral range of 1235−35 cm−1.
■
RESULTS AND DISCUSSION There are two known Bi/S/I phasesBiSI and the metastable Bi13S18I2 (previously reported as Bi19I3S27 and Bi(Bi2S3)9I3, among other formulas).25,26 Our exploration of reactions of bismuth in S/I2 fluxes at various temperatures indicates that BiSI can be obtained in very high yield with no Bi13S18I2 if the iodine content in the flux is equal to the metal; lower iodine content promotes formation of the Bi13S18I2 phase. Antimony forms only SbSI regardless of S/I2 ratio; the analogous Sb13S18I2 is not known and could not be synthesized. DSC-TGA measurements indicate that SbSI decomposes into Sb2S3 with loss of SbI3 from 300 to 360 °C, and this was confirmed by PXRD (see Figures S1 and S2 in the Supporting Information). This is in contrast to the behavior of BiSI, which decomposes to Bi13S18I2 at 400 °C with loss of BiI3 and I2.24 Synthesis. BiSI crystals are about 0.5−1 mm in length and 20−50 μm in diameter; these needlelike black crystals splay into smaller fiber-like crystals when pressed (Figure 1). EDS
Figure 1. Crystals of Bi1−xSbxSI on millimeter grid paper, grown from reactions of Bi/Sb in S/I2 flux (0.5/0.5/10/0.1) at 400 °C. Left inset: single crystal of BiSI. Right inset: single crystals and ground powder of SbSI.
measurements on BiSI typically indicate near-identical atomic percentages of all three elements (Bi, S, and I in the 30−35% range). The yield of optimized reactions (at a Bi:S:I ratio of 1:10:0.5) is roughly 85%, with 95−97% purity of the BiSI phase. In addition to iodine content, the maximum reaction temperature is also a crucial parameter. The best yield is found for reactions run at 400 °C; reactions run at higher temperature produce larger crystals, but this is accompanied by partial thermal decomposition to form Bi13S18I2 byproduct.23 Reactions of antimony in S/I2 flux yield only SbSI, forming similarly needle shaped crystals which are dark red. Reactions run at 400 °C produced nearly pure SbSI product (>90% yield); lower temperature (300 °C) reaction products contained traces of unreacted antimony, while reactions at 500 °C contained traces of Sb 2 S 3 (presumably from decomposition of SbSI). While SbSI decomposes above 350 °C under atmospheric conditions (vide infra), the reactions in this work take place within a sulfur/iodine flux in a sealed ampule. The presence of excess sulfur and iodine in this closed B
DOI: 10.1021/acs.inorgchem.7b01839 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry system environment modifies the decomposition equilibrium, preventing the breakdown of SbSI into the binary products Sb2S3 and SbI3. Reactions were explored using mixtures of antimony and bismuth in the S/I2 flux, to determine whether these elements would segregate into different products or incorporate into one product as a solid solution. These elements appear to mix well in the flux; reactions in S/I2 flux at 400 °C produce well-formed (Bi1−xSbx)SI crystals for the entire substitution range. For all reactions, crystalline products were only observed to form within the flux media and not within the headspace. Crystals grew from the conical-shaped bottom of the quartz ampule, upward toward the top of the flux, in a starburst fashion (Figure S3 in the Supporting Information). This would indicate that, despite the elevated temperature, the crystals grow within the flux solution as opposed to using a vapor transport mechanism. Structure. BiSI, SbSI, and the solid solution (Bi1−xSbx)SI all have the same structure in orthorhombic space group Pnma (Figure 2) at room temperature and above.27,28 A ferroelectric
Figure 3. SEM images of crystals from a reaction of Bi/Sb/S/I2 (0.5/ 0.5/10/0.5 mmol) at 400 °C. EDS measurements were taken on spots 1−4; spots 1 and 2 had a stoichiometry of Bi0.5Sb0.5SI, while spot 3 was Bi0.55Sb0.45SI and spot 4 was Bi0.65Sb0.35SI.
solidification of metal alloys such as Cu1−δNiδ, where the elemental makeup of the core region of precipitated grains differs from that of the outer region. Rietveld fitting of PXRD data indicates that, while the overall cell volume of BiSI shrinks as Bi is replaced by Sb, the effect is not isotropic. The b and c axes shrink, but the a axis elongates slightly. The dependence of each axis length on substitution level is shown in Figure 4. The b axis shows a linear dependence (Vegard’s law behavior), but slight deviations are seen for the a and c axes. Both axes stay somewhat constant for Figure 2. Structure of BiSI looking down the b axis at the ac plane. The black circle highlights the Bi−S 1-D chain.
transition from Pnma to Pna21 is seen for SbSI below a Tc value of 295 K, but not for BiSI.29,30 All measurements were carried out above this temperature. Rietveld refinements were performed on PXRD patterns of the bulk product matrix from reactions at 400 °C in a 10/0.5 mmol S/I2 flux. Under these synthetic conditions, the Bi:Sb ratio in the reactant mixture is largely maintained in the averaged Bi1−xSbxSI product matrix, as determined by PXRD (Figure S4 in the Supporting Information). However, closer examination by SEM-EDS shows that, while there is homogeneity in the substitution within a crystal, deviations in the Bi/Sb ratio from crystal to crystal can be observed. Element mapping confirmed that there was uniform substitution within crystals, but EDS sampling of neighboring crystals showed as much as a 10% deviation in substitution from the average (Figure 3). For example, products of a 0.5/0.5/10/0.5 mmol ratio Bi/Sb/S/I2 reaction at 400 °C yielded crystals with EDS ratios ranging from Bi0.45Sb0.55SI to Bi0.65Sb0.35SI, with the average substitution at Bi0.55Sb0.55SI. This may lead to difficulties in interpreting data from characterization techniques that involve samples comprised of large numbers of crystals (such as powder XRD and UV/vis diffuse reflectance). This type of segregation of elements into different crystals may be caused by differing solubilities or reactivities of bismuth vs antimony in the S/I2 flux as the reaction mixture is heated or cooled. Crystals richer in the less soluble element will precipitate at higher temperatures; as the temperature drops, the proportion of Bi/Sb will change. Similar “coring” effects result during
Figure 4. Dependence of unit cell parameters on substitution in Bi1−xSbxSI from products of reactions at 400 °C (M/S/I2: 1/10/0.5 mmol). C
DOI: 10.1021/acs.inorgchem.7b01839 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
(CZTS).35,36 This behavior is often ascribed to local structural distortions which occur on substitution. The extent of such local atom displacements can have a nonlinear dependence on optical properties; in some cases it may result in clustering or create localized states in the band gap.37 In addition, the substitution of a heavier atom for a lighter one results in a greater spin−orbit coupling contribution to the band energies, leading to nonlinear behavior.38 The inert pair of the pnictide cations might also have electronic contributions near the band gap. Previous computational studies of BiSI and SbSI indicate that they are indirect gap semiconductors. The highest occupied states in the valence band are derived from filled iodine 5p orbitals, and the lowest unoccupied states in the conduction band are derived from empty bismuth 6p orbitals (or antimony 5p orbitals for SbSI). However, advanced computational studies on BiSI support the possibility of significant Bi 6s character near the valence band maximum, caused by hybridization between the Bi 6s band with the iodine 5p band, despite the positioning of the former well below the Fermi level.39−42 While similar calculations have not been reported for SbSI, it is possible that a similar contribution to the valence band maximum from the Sb 5s electrons could occur. Given the inequivalent stereoactivities of the inert pair for Sb3+ versus Bi3+, it is likely that the positioning of this s electron state near the VBM will not have a linear dependence with substitution; this would contribute to the nonlinear dependence of Eg on substitution. Another factor in the nonlinear dependence of Eg on composition may be the inhomogeneity of the bulk samples used for UV−vis−NIR reflectance measurements. As previously mentioned, SEM-EDS showed that there is homogeneity of substitution within individual crystals but deviation of substitution from crystal to crystal within the product matrix. Some crystals in a given batch of Bi1−xSbxSI will have bismuth content higher than that indicated by the averaged stoichiometry. It is these crystals (and their smaller band gap) that will dominate the absorption properties of the overall sample. For example, reactions of Bi/Sb/S/I2 (0.5/0.5/10/0.5 mmol) at 400 °C yielded crystals with an average substitution of 55% bismuth and 45% antimony, but bismuth substitution reached as high as 65%. The crystals with a substitution of 65% bismuth, which have a band gap smaller than those with the average substitution, will dominate the reflectance spectra. As stated before, the observed band gap begins to shift toward that of SbSI when the products reach an Sb-rich average composition (>50% Sb). To determine if this sample inhomogeneity is the deciding factor, it would be necessary to carry out UV/vis absorption measurements on individual crystals, but the instrumentation was not available. Raman spectra were collected on single-crystal Bi1−xSbxSI samples at varying substitution amounts. The spectra for BiSI and SbSI are similar to those previously reported for both phases in the paraelectric Pnma phase.29,43 Work by Teng et al. indicated that MSI compounds are made up of covalent metal sulfide units (M3+S2−)+ which are charge balanced by iodide (MS)+I−, creating an overall covalent−ionic model. The Raman-active modes which involve covalent metal−sulfur bonds show large changes in frequency when one metal is exchanged for another: for example, the Γ1 mode at 286.5 cm−1 for BiSI is found at 320 cm−1 for SbSI. However, modes involving iodine motion are predominantly unchanged by the exchange of bismuth for antimony, with the Γ2 mode at 220 cm−1 for BiSI found at 235 cm−1 for SbSI.43 Our Raman studies
Bi1−xSbxSI up to x = 0.5 and then begin to shift, similar to what is observed for the band gap energies of these compounds (vide infra). This nonlinear behavior of these two axes may be related to the stereoactivity of the inert pair on the Bi or Sb atoms, which would be positioned in the ac plane of the structure (see Figure 2). This can be compared to the substitution behavior in the bismuthinite−stibnite system (Bi2−xSbxS3). The structure of these sulfides features ribbons of (Bi/Sb)S units separated by spaces that accommodate the Bi/Sb inert pair. As is seen for (Bi/Sb)SI, the (Bi/Sb)2S3 structure exhibits similarly anisotropic changing of the unit cell axes as substitution occurs, along with deviation from Vegard’s law.31 Instead of just one factor affecting the structural geometry upon substitution, there are twothe size of the atom being substituted and the stereoactivity of the inert pair. The average radius of the pnictide ion decreases as more antimony is incorporated. However, the stereoactivity (and corresponding effective size) of the inert pair is larger for antimony.32 The counterbalancing of these two factors produces a deviation from expected Vegard’s law behavior, particularly for the a and c axes which define the plane in which the inert pair is positioned. Electronic and Optical Properties. Formation of a Bi1−xSbxSI solid solution is a very useful way of controlling the band gap and transport properties of this charge-balanced semiconducting compound. Since the use of these materials as photovoltaics is of interest, diffuse reflectance measurements (UV−vis−NIR) were done on Bi1−xSbxSI at several substitution levels to optically determine the band gaps of the fluxgrown materials (Figure 5). Band gaps of 1.55 and 1.86 eV were
Figure 5. UV−vis diffuse reflectance spectra of various MSI (M = Bi/ Sb) phases. The spectra that lie between BiSI and Bi/Sb 50/50 spectra are of various Bi1−xSbxSI products from reactions containing a Bi/Sb ratio between 100/0 and 50/50.
observed for BiSI and SbSI, respectively, close to the reported values of 1.57−1.63 and 1.84−1.88 eV.12,1,33,34 Diffuse reflectance spectra of flux-grown Bi1−xSbxSI compounds are shown in Figure 5. As the antimony content in the product increases from x = 0 to x = 0.5, the size of the band gap remains close to that of BiSI. It is not until the compound reaches an antimony-rich composition (>50% Sb) that the band gap begins to shift toward that of SbSI. This disagrees with the expected linear dependence of band gap energy on substitution for simple compound semiconductors. However, band gap bowing (deviation from the linear trend) is reported in several systems, such as CuAl 1−x In x S 2 and Cu 2 Mn x Zn 1−x SnS 4 D
DOI: 10.1021/acs.inorgchem.7b01839 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry of Bi1−xSbxSI focused on modes in the region of 250−350 cm−1. As Sb is substituted for Bi, the Raman spectrum of each solid solution compound features two Γ1 vibrational modes, at energies characteristic of both parent compounds (315 cm−1 for SbSI and 280 cm−1 for BiSI) (see Figure 6). The presence of
Figure 7. TEM images of Bi0.55Sb0.45SI material (Bi/Sb/S/I2: 0.5/0.5/ 10/0.5 at 400 °C).
Figure 6. Raman spectra of BiSI, SbSI, and Bi1−xSbxSI (x = 0.37, 0.45).
both parent modes instead of one averaged mode indicates a two-mode Raman behavior (where two vibrational modes are seen in the spectrum corresponding to “pure” BiSI and SbSI). This is similar to what is seen for Zn1−xMgxSe and is often an indication of strong covalent contributions to bonding in a solid, creating separate localized modes.44 The relative peak intensity from the spectra is related to the level of substitution in the phase. This is in contrast to one-mode behavior common in more ionic systems, where a solid solution exhibits vibrational modes between those of its parent end members. TEM Imaging. The Raman and UV−vis spectra also indicated the possibility that, instead of forming a homogeneous solid solution, the Bi and Sb were segregating in the product, resulting in nanoclustering of BiSI and SbSI domains. This was observed in the (PbS)1−x(PbTe)x system, where nanoclustering of PbS and PbTe occurred on the 5−10 nm scale throught the material.45 This would yield a Raman spectrum that is a combination of SbSI and BiSI (as is seen), and the absorption spectrum would be dominated by the BiSI regions. To test for this, Z-contrast TEM measurements were carried out on a Bi0.55Sb0.45SI sample to determine if nanoclustering was evident. The intensity of atomic columns in STEM-HAADF images is proportional to the atomic number Zn, where n is close to 2: i.e., they are Z-number-sensitive images (Z-contrast).46,47 Bi atoms show brighter contrast because of the higher Z number in comparison to Sb atoms. In Figure 7 there was no considerable contrast difference observed, demonstrating a lack of nanoclustering in this substituted structure.
dependence on x that is observed for the unit cell parameters and the band gap. The variation in stereoactivity of the lone pair as Bi is substituted for Sb may also be a factor. The twomode behavior in the Raman spectra of single crystals indicates a high degree of covalency in the bonding of these compounds. Investigation of further substitution chemistry (for instance, BiS1−xSexI) would be of interest to explore if similar effects are seen.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b01839. Microscope images of crystals, thermal decomposition data (DSC-TGA), and powder X-ray diffraction data (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail for S.E.L.:
[email protected]. ORCID
Susan E. Latturner: 0000-0002-6146-5333 Notes
The authors declare no competing financial interest.
■
■
ACKNOWLEDGMENTS This project was supported by the Council on Research and Creativity of Florida State University. The assistance of Dr. Yan Xin in the TEM facility at the National High Magnetic Field Laboratory is greatly appreciated. The TEM facility is funded and supported by the Florida State University Research Foundation, and the National High Magnetic Field Laboratory, which was supported in part by the National Science Foundation Cooperative Agreement DMR-1157490, the State of Florida.
CONCLUSION Metal sulfide iodides BiSI and SbSI as well as substitutional variants Bi1−xSbxSI (0 ≤ x ≤ 1) were successfully synthesized from reactions in sulfur/iodine flux. Synthesis of Bi1−xSbxSI in a flux produces crystals with varying substitution between crystals, averaging to that of the reactant ratio. This lack of substitution homogeneity is an aspect of flux chemistry which needs further exploration; it is likely that the addition of stirring or modification of heating profile may correct this problem. This sample inhomogeneity may be the cause of the nonlinear E
DOI: 10.1021/acs.inorgchem.7b01839 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
■
(22) Audzijonis, A.; Ž igas, L.; Kvedaravičius, A.; Ž altauskas, R. The Experimental and Theoretical Investigation of Vibration Spectra in Ferroelectric Semiconductor SbSBrxI1−x Crystals. Phys. B 2009, 404, 3941−3946. (23) Groom, R.; Jacobs, A.; Cepeda, M.; Drummey, R.; Latturner, S. E. Bi13S18I2: (Re)discovery of a Subvalent Bismuth Compound Featuring [Bi2]4+ Dimers Grown in Sulfur/Iodine Flux Mixtures. Chem. Mater. 2017, 29, 3314−3323. (24) Sheldrick, G. M. SHELXTL NT/2000; Bruker AXS, Inc., Madison, WI, 2000. (25) Otto, H. H.; Strunz, H. The Crystal Chemistry and Synthesis of Lead-Bismuth-Antimony. Neues Jahrb. Mineral., Abh. 1968, 108, 1−19. (26) Miehe, G.; Kupčik, V. The Crystal Structure of Bi(Bi2S3)9I3. Naturwissenschaften 1971, 58, 219−220. (27) Haase-Wessel, W. Die Kristallstruktur Des Wismutsulfidjodids (BiSJ). Naturwissenschaften 1973, 60, 474−474. (28) Kikuchi, A.; Oka, Y.; Sawaguchi, E. Crystal Structure Determination of SbSI. J. Phys. Soc. Jpn. 1967, 23, 337−354. (29) Balkanski, M.; Teng, M. K.; Shapiro, S. M.; Ziolkiewicz, M. K. Lattice Modes and Phase Transition in SbSI. Phys. Status Solidi B 1971, 44, 355−368. (30) Nako, K.; Balkanski, M. Electronic Band Structures of SbSI in the Para- and Ferroelectric Phases. Phys. Rev. B 1973, 8, 5759. (31) Poleti, D.; Karanovic, L.; Balic-Zunic, T.; Grzetic, I. Crystal Structure of (Bi0.94 Sb1.06)S3 and Reconsideration of Cation Distribution over Mixed Sites in the Bismuthinitestibnite SolidSolution Series. Neues Jahrb. Mineral., Abh. 2012, 189, 177−187. (32) Kyono, A.; Kimata, M. Structural Variations Induced by Difference of the Inert Pair Effect in the Stibnite-Bismuthinite Solid Solution Series (Sb,Bi) 2 S 3. Am. Mineral. 2004, 89, 932−940. (33) Park, S.-A.; Kim, M.-Y.; Lim, J.-Y.; Park, B.-S.; Koh, J.-D.; Kim, W.-T. Optical Properties of Undoped and V-Doped VA−VIA−VIIA Single Crystals. Phys. Status Solidi B 1995, 187, 253−260. (34) Nowak, M.; Szperlich, P. Temperature Dependence of Energy Band Gap and Spontaneous Polarization of SbSI Nanowires. Opt. Mater. (Amsterdam, Neth.) 2013, 35, 1200−1206. (35) Kumar, S.; Joshi, S.; Gupta, S. K.; Auluck, S. Band Gap Engineering of CuAl1− xInxS2 Alloys for Photovoltaic Applications: A First Principles Study. J. Phys. D: Appl. Phys. 2016, 49, 205103. (36) Chen, L.; Deng, H.; Cui, J.; Tao, J.; Zhou, W.; Cao, H.; Sun, L.; Yang, P.; Chu, J. Composition Dependence of the Structure and Optical Properties of Cu2MnxZn1−xSnS4 Thin Films. J. Alloys Compd. 2015, 627, 388−392. (37) Schnohr, C. S. Compound Semiconductor Alloys: From Atomic-Scale Structure to Bandgap Bowing. Appl. Phys. Rev. 2015, 2, 031304. (38) Im, J.; Stoumpos, C. C.; Jin, H.; Freeman, A. J.; Kanatzidis, M. G. Antagonism between Spin−Orbit Coupling and Steric Effects Causes Anomalous Band Gap Evolution in the Perovskite Photovoltaic Materials CH3NH3Sn1− xPbxI3. J. Phys. Chem. Lett. 2015, 6, 3503− 3509. (39) Shi, H.; Ming, W.; Du, M. H. Bismuth Chalcohalides and Oxyhalides as Optoelectronic Materials. Phys. Rev. B: Condens. Matter Mater. Phys. 2016, 93, 104108. (40) Ganose, A. M.; Butler, K. T.; Walsh, A.; Scanlon, D. O. Relativistic Electronic Structure and Band Alignment of BiSI and BiSeI: Candidate Photovoltaic Materials. J. Mater. Chem. A 2016, 4, 2060−2068. (41) Audzijonis, A.; Zaltauskas, R.; Sereika, R.; Zigas, L.; Reza, A. Electronic Structure and Optical Properties of BiSI Crystal. J. Phys. Chem. Solids 2010, 71, 884−891. (42) Audzijonis, A.; Gaigalas, G.; Ž igas, L.; Pauliukas, A.; Ž altauskas, R.; Č erškus, A.; Narusis, J. Theoretical Investigation of the Electronic Structure of a Ferroelectric SbSI Cluster at a Phase Transition. Cent. Eur. J. Phys. 2005, 3, 382−394. (43) Teng, M. K.; Balkanski, M.; Massot, M.; Ziolkiewicz, M. K. Optical Phonon Analysis in the AVBVICVII Compounds. Phys. Status Solidi B 1974, 62, 173−182.
REFERENCES
(1) Cho, I.; Min, B.-K.; Joo, S. W.; Sohn, Y. One-Dimensional Single Crystalline Antimony Sulfur Iodide, SbSI. Mater. Lett. 2012, 86, 132− 135. (2) Kuku, T. A.; Azi, S. O. Optical Properties of Evaporated PbSnS3 Thin Films. J. Mater. Sci. 1998, 33, 3193−3196. (3) Fiechter, S.; Tomm, Y.; Kanis, M.; Scheer, R.; Kautek, W. On the Homogeneity Region, Growth Modes and Optoelectronic Properties of Chalcopyrite-Type CuInS2. Phys. Status Solidi B 2008, 245, 1761− 1771. (4) Yang, H.; Jauregui, L. A.; Zhang, G.; Chen, Y. P.; Wu, Y. Nontoxic and Abundant Copper Zinc Tin Sulfide Nanocrystals for Potential High-Temperature Thermoelectric Energy Harvesting. Nano Lett. 2012, 12, 540−545. (5) Ivanovskaya, A.; Singh, N.; Liu, R.-F.; Kreutzer, H.; Baltrusaitis, J.; Van Nguyen, T.; Metiu, H.; McFarland, E. Transition Metal Sulfide Hydrogen Evolution Catalysts for Hydrobromic Acid Electrolysis. Langmuir 2013, 29, 480−492. (6) Zhao, L.-D.; Lo, S.-H.; He, J.; Li, H.; Biswas, K.; Androulakis, J.; Wu, C.-I.; Hogan, T. P.; Chung, D.-Y.; Dravid, V. P.; Kanatzidis, M. G. High Performance Thermoelectrics from Earth-Abundant Materials: Enhanced Figure of Merit in PbS by Second Phase Nanostructures. J. Am. Chem. Soc. 2011, 133, 20476−20487. (7) Johnsen, S.; Liu, Z.; Peters, J. A.; Song, J.-H.; Peter, S. C.; Malliakas, C. D.; Cho, N. K.; Jin, H.; Freeman, A. J.; Wessels, B. W.; Kanatzidis, M. G. Thallium Chalcogenide-Based Wide-Band-Gap Semiconductors: TlGaSe 2 for Radiation Detectors. Chem. Mater. 2011, 23, 3120−3128. (8) Wang, P. L.; Liu, Z.; Chen, P.; Peters, J. A.; Tan, G.; Im, J.; Lin, W.; Freeman, A. J.; Wessels, B. W.; Kanatzidis, M. G. Hard Radiation Detection from the Selenophosphate Pb2P2Se6. Adv. Funct. Mater. 2015, 25, 4874−4881. (9) Sasaki, Y. Photoconductivity of a Ferroelectric Photoconductor BiSI. Jpn. J. Appl. Phys. 1965, 4, 614−615. (10) Nitsche, R.; Merz, W. J. Photoconduction in Ternary V-VI-VII Compounds. J. Phys. Chem. Solids 1960, 13, 154−155. (11) Audzijonis, A.; Ž igas, L.; Kvedaravičius, A.; Ž altauskas, R. Origin of Ferroelectric Phase Transitions of Bi X Sb 1-X SI Mixed Crystals. Ferroelectrics 2009, 392, 45−54. (12) Hahn, N. T.; Self, J. L.; Mullins, C. B. BiSI Micro-Rod Thin Films: Efficient Solar Absorber Electrodes? J. Phys. Chem. Lett. 2012, 3, 1571−1576. (13) Grigas, J.; Talik, E.; Adamiec, M.; Lazauskas, V.; Nelkinas, V. XPS and Electronic Structure of Quasi-One-Dimensional BiSI Crystals. J. Electron Spectrosc. Relat. Phenom. 2006, 153, 22−29. (14) Aguiar, I.; Mombrú, M.; Barthaburu, M. P.; Pereira, H. B.; Fornaro, L. Influence of Solvothermal Synthesis Conditions in BiSI Nanostructures for Application in Ionizing Radiation Detectors. Mater. Res. Express 2016, 3, 025012. (15) Deshpande, M. P.; Sakariya, P. N.; Bhatt, S. V.; Garg, N.; Patel, K.; Chaki, S. H. Characterization of Bi2S3 Nanorods Prepared at Room Temperature. Mater. Sci. Semicond. Process. 2014, 21, 180−185. (16) Zhou, X.; Soldat, A. C.; Lind, C. Phase Selective Synthesis of Copper Sulfides by Non-Hydrolytic Sol−gel Methods. RSC Adv. 2014, 4, 717. (17) Shen, G.; Chen, D.; Tang, K.; Huang, L.; Qian, Y. Large-Scale Synthesis of (Bi(Bi2S3)9I3)0.667 Submicrometer Needle-like Crystals via a Novel Polyol Route. J. Cryst. Growth 2003, 249, 331−334. (18) Meyer, B. Elemental Sulfur. Chem. Rev. 1976, 76, 367−387. (19) Fanelli, R. Modifying the Viscosity of Sulfur. Ind. Eng. Chem. 1946, 38, 39−43. (20) Kunioku, H.; Higashi, M.; Abe, R. Low-Temperature Synthesis of Bismuth Chalcohalides: Candidate Photovoltaic Materials with Easily, Continuously Controllable Band Gap. Sci. Rep. 2016, 6, 32664. (21) Schultz, P.; Keller, E. Strong Positive and Negative Deviations from Vegard’s Rule: X-Ray Powder Investigations of the Three QuasiBinary Phase Systems BiS X 1 − X Y X (X, Y = Cl, Br, I). Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater. 2014, 70, 372−378. F
DOI: 10.1021/acs.inorgchem.7b01839 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry (44) Valakh, M. Y.; Litvinchuk, A. P.; Dzhagan, V. M.; Yukhymchuk, V. O.; Havryliuk, Y. O.; Guc, M.; Bodnar, I. V.; Izquierdo-Roca, V.; Pérez-Rodríguez, A.; Zahn, D. R. T. Optical Properties of Quaternary Kesterite-Type Cu2Zn(Sn1−xGex)S4 Crystalline Alloys: Raman Scattering, Photoluminescence and First-Principle Calculations. RSC Adv. 2016, 6, 67756−67763. (45) Johnsen, S.; He, J.; Androulakis, J.; Dravid, V. P.; Todorov, I.; Chung, D. Y.; Kanatzidis, M. G. Nanostructures Boost the Thermoelectric Performance of PbS. J. Am. Chem. Soc. 2011, 133, 3460−3470. (46) Pennycook, S. J.; Boatner, L. A. Chemically Sensitive StructureImaging with a Scanning Transmission Electron Microscope. Nature 1988, 336, 565−567. (47) Loane, R. F.; Xu, P.; Silcox, J. Thermal Vibrations in Convergent-Beam Electron Diffraction. Acta Crystallogr., Sect. A: Found. Crystallogr. 1991, 47, 267−278.
G
DOI: 10.1021/acs.inorgchem.7b01839 Inorg. Chem. XXXX, XXX, XXX−XXX
