Evaluation of Antimony Tri-Iodide Crystals for Radiation Detectors
This study was carried out to examine the potential of antimony tri-iodide (SbI3) as a material for radiation detectors that operate at room temperature. SbI3 is a compound semiconductor with an AsI3-type crystal structure, high atomic number (Sb: 51, I: 53), high density (4.92 g/cm3), and a wide band-gap energy (2.2 eV). In addition, crystalline SbI3 is easy to grow by conventional crystal growth techniques from melting phase because the material exhibits a low melting point (171°C) and undergoes no phase transition in the range of its solid phase. In this study, SbI3 crystals were grown by the Bridgman method after synthesis of SbI3 from 99.9999% pure Sb and 99.999% pure I2. The grown crystals consisted of several large grains with red color and were confirmed to be single-phase crystals by X-ray diffraction analysis. SbI3 detectors with a simple planar structure were fabricated using the cleavage plates of the grown crystals, and the pulse-height spectra were recorded at room temperature using an 241Am alpha-particle (5.48 MeV) source. The detector showed response to the alpha-particle radiation.
Numerous compound semiconductors have been actively investigated for use in radiation detectors [1–9]. High detection efficiency for gamma-rays as well as low-noise operation at room temperature and high charge collection efficiency are important characteristics for radiation detectors with high spectroscopic performance. Although CdTe, CdZnTe , and HgI2 , have recently been considered as ideal detector materials in this application, these materials are more expensive than traditional detector materials such as Si and Ge because of the difficulty associated with their crystal growth.
Antimony tri-iodide (SbI3) is a compound semiconductor with an AsI3-type crystal structure; it has been reported to be a potential semiconductor material for radiation detectors . The characteristics of SbI3 include high atomic number (Sb: 51, I: 53), high density (4.92 g/cm3), and wide band-gap energy (2.2 eV). The physical properties of SbI3 suggest that it can be used in radiation detectors with high detection efficiency and low-noise operation at room temperature without any cooling. Figure 1 shows the attenuation coefficients for Ge, CdTe, and SbI3 . As shown in this figure, SbI3 exhibits high gamma-ray stopping power equivalent to that of CdTe. In addition, SbI3 melts at a low temperature (171°C) and exhibits no phase transition between room temperature and its melting point . Thus, a single SbI3 crystal can be grown from the molten material via the conventional crystal growth technique such as the Bridgman method or the traveling molten-zone method.
The aforementioned attractive properties of SbI3 suggest that SbI3 detectors can be fabricated at low cost compared with CdTe and CdZnTe detectors. BiI3  and PbI2  crystals were also layered structure crystals similar to SbI3 and have been studied as a new radiation detector material. However, gamma-ray energy resolution obtained from these layered materials was poor than CdTe and CdZnTe due to low charge transport properties.
Although SbI3 is a promising material, no previous studies have reported the performance of SbI3 detectors. The objective of this work was to evaluate the potential of SbI3 as a radiation detector.
2.1. Crystal Growth
Commercially available Sb (nominal purity of 99.9999%, Kojundo chemical laboratory Co., Ltd) and I2 (nominal purity of 99.999%, Kojundo chemical laboratory Co., Ltd) were used as starting materials for growing SbI3 crystals. Stoichiometric amounts of the starting materials were loaded into a quartz ampoule (ϕ10 mm) that was subsequently vacuum sealed at approximately 1 Pa. For the synthesis of SbI3, the ampoule was kept in a furnace above the melting point for 24 h. After the synthesis, multipass zone refining was used to reduce the impurities in the synthesized material. The length of the molten zone was approximately 10 mm. The furnace was moved at a speed of 5 cm/h, and the purification was repeated 100 times. After the material had been purified, SbI3 crystals were grown by the Bridgman method  at the middle section of the purified material. Figure 2 shows a schematic of the Bridgman furnace and the temperature distribution used in this study. Crystal growth was carried out in a temperature gradient of approximately 3.4°C/cm. The ampoule was located in the furnace as shown in the figure, and the material in the ampoule was completely melted before the ampoule was lowered. After the material had melted, the ampoule was lowered at a speed of 1 mm/h using a stepping motor. Figure 3 shows cleaved SbI3 crystals grown in this study. The main cleavage plane was parallel to the growth direction and contained several grain boundaries. The crystal color was metallic red by reflected light and red by transmitted light, consistent with the band-gap of SbI3 (2.2 eV).
2.2.1. Optical Transmittance Spectra
A UV–Vis spectrophotometer (Shimadzu UV-1800) was used to evaluate the band-gap energy of the grown SbI3 crystal. After the crystal growth, the SbI3 crystal was cleaved into thin plates (thickness of 0.1–0.3 mm) using a cutter blade. Flat transparent SbI3 plates with a thickness of 0.1–0.2 mm were selected as samples for this evaluation.
2.2.2. X-Ray Diffraction
X-ray diffraction (XRD) analysis (Rigaku 2500HF) was used to confirm the crystallinity of the grown crystal. The voltage and current of the X-ray tube (Cu-Kα) were 40 kV and 200 mA, respectively. Cleaved plates and powdered SbI3 were prepared from the grown crystal. Commercially available SbI3 powder was used to obtain reference data for the SbI3 phase.
2.2.3. Current–Voltage Characteristics
The current–voltage characteristics of the SbI3 detectors were measured to evaluate the resistivity of the SbI3 detectors. The bias voltage was changed from 0 V to 200 V, and the leakage current of the detectors was measured at room temperature.
2.2.4. Fabrication of an SbI Detector
The grown crystal was cleaved into thin plates (thickness of 0.1–0.3 mm), and gold electrodes (3 mm in diameter) were deposited onto both sides of the SbI3 crystals by the vacuum evaporation method. After electrode deposition, the SbI3 crystals were placed on a stage used for detector fabrication and palladium wires were connected to each electrode from a terminal on the stage. The detector was enclosed in an aluminum case, and the signal from the detector was taken from a BNC terminal on the case.
2.2.5. Radiation Response
For initial operation before evaluating gamma-ray detector performance of the SbI3 detectors, 241Am alpha-particle energy spectra were recorded using the SbI3 detectors in a conventional measurement system comprising a preamplifier, bias supply, shaping amplifier, and a multichannel pulse-height analyzer. The 241Am alpha-particle source (185 kBq) and the detector were enclosed in the aluminum case used for the current–voltage measurements. Distance between the alpha-particle source and the SbI3 crystal was approximately 2 mm and the measurement was carried out in the atmosphere. After the initial evaluation, gamma-ray response was also evaluated using the same measurement system and 137Cs gamma-ray source (3.7 MBq).
3. Results and Discussion
Figure 4 shows the transmittance spectrum as a function of wavelength. The cutoff wavelength was approximately 576 nm, suggesting that the band-gap energy of the grown crystal was approximately 2.15 eV, consistent with a previously reported value .
Figure 5 shows the XRD patterns obtained from an SbI3 plate, the powder, and the commercial SbI3 powder. All of the diffraction peaks in the pattern of the powdered crystal match the JCPDS data , indicating that SbI3 was successfully synthesized using the starting materials and that single-crystalline SbI3 was grown by the Bridgman method.
Figure 6 shows the extended XRD patterns in the region of the (006) plane. The peaks of the (006) plane in the pattern of the grown crystal and the powdered crystal were slightly shifted toward higher angles compared with the corresponding peak in the pattern of the commercial powder. This result implies that the lattice constant of the grown and powdered crystal samples was somewhat smaller than that of the commercial SbI3 powder. The origin of this small difference requires further investigation beyond the scope of the present work.
3.2. Detector Performance
Figure 7 shows the leakage current as a function of the bias voltage obtained from the SbI3 detectors. The thickness of the SbI3 crystal was 0.339 mm. The resistivity of the SbI3 detectors was estimated to be approximately 1.0 × 1010 Ω·cm.
Figures 8 and 9 show the alpha-particle energy spectra obtained from a typical SbI3 detector at room temperature. The thickness of the detector was 0.229 mm. The spectra were obtained by changing the polarity of the bias voltage to evaluate the difference in charge transport properties between electrons and holes. The bias voltage was 200 V, and the amplifier shaping time was 30 μs. Both energy spectra recorded under alpha-particle irradiation show increased counts above the noise counts. Penetration depth of 5.48 MeV alpha-particles is calculated to be less than 40 μm in water and the range for the alpha-particles is much smaller than the crystal thickness. Because the penetration depth of alpha-particles into the crystals was very shallow, the maximum count channel in the energy spectra depends on the transport properties of holes or electrons. As shown in Figures 8 and 9, the maximum count channel was independent of the polarity of the bias voltage. Thus, the charge transport properties of electrons and holes in the SbI3 detector were approximately equivalent. In addition, because the SbI3 detector exhibited no full-energy peaks corresponding to 5.48 MeV of alpha-particles, the μτ products for both carriers in the SbI3 detector were substantially smaller than those for other compound semiconductor detectors such as CdTe, CdZnTe, and TlBr.
Figure 10 was the alpha-particle energy spectra acquired for 30 s and 180 s and total counts were 15,834 counts and 50,969 counts for 30 s and 180 s, respectively. Although the count rate linearity could not be observed from the SbI3 detector, the linearity may be improved by reducing charge trapping and detector noise during operation.
Figure 11 shows the 137Cs gamma-ray energy spectra obtained from the SbI3 detector at room temperature. Counts from the detectors were slightly increased by the irradiation of gamma-rays. These results suggested that almost gamma-rays from the source were penetrating through the detector because the thickness of the SbI3 crystal was 0.229 mm.
3.3. Detector Stability
Stable detector performance is an important characteristic for radiation detectors. During operation of semiconductor detectors, temporal changes in detector performance known as the polarization phenomenon are usually observed as a decrease in pulse height. This phenomenon depends on the semiconductor material used for the detector and has been previously observed for CdTe detectors and TlBr detectors. Methods for improving or suppressing this phenomenon have been reported [17–20]. As previously mentioned, confirmation of the polarization phenomenon and research overcoming the phenomenon are necessary for realizing semiconductor detectors suitable for applications such as medical devices and industrial imaging.
In the present study, the SbI3 detector was continuously operated at room temperature and alpha-particle energy spectra were recorded as a function of time to evaluate the polarization phenomenon in the detector. Figure 12 shows the operation-time dependence of the 241Am alpha-particle energy spectra obtained using the SbI3 detector. The spectral response of the SbI3 detector was stable for almost 24 h under the bias voltage (200 V). Although a reduction in pulse height was observed after 90 h of operation, the pulse height immediately recovered when the bias voltage was cutoff. Recovery from the polarization phenomenon after cutoff of the applied bias voltage has also been reported for CdTe detectors , whose recovery behavior differs from that of TlBr detectors. TlBr detectors required a long time to recover their performance because the accumulation of ionic charge at the electrodes caused the polarization phenomenon. Therefore, the polarization phenomenon observed for the SbI3 detector is similar to that of CdTe detectors and is speculatively attributed to trapped charges.
SbI3 crystals were grown by the Bridgman method, and the performance of SbI3 detectors was evaluated on the basis of alpha-particle energy spectra recorded at room temperature. Although the spectra recorded using the fabricated SbI3 detectors showed no full-energy peak of the alpha-particles, the detector responded to alpha-particle irradiation. The SbI3 detector exhibited a decrease in pulse height during long-term operation, similar to other semiconductor detectors based on CdTe and TlBr; however, stable operation for 24 h was observed and the pulse height immediately recovered when the applied bias was cut off. Thus, SbI3 is a promising material for use in radiation detectors that exhibit stable operation. Further research on purification of the starting material and on improving carrier transport is required to improve the spectroscopic performance of the SbI3 detectors.
The data used to support the findings of this study are available from the corresponding author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
“Powder diffraction file (No. 70273),” JCPDS, 1997.View at: Google Scholar
T. Onodera, K. Hitomi, and T. Shoji, “Spectroscopic performance and long-term stability of thallium bromide radiation detectors,” Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, vol. 568, no. 1, pp. 433–436, 2006.View at: Publisher Site | Google Scholar
K. Hitomi, T. Shoji, and Y. Nizeki, “A method for suppressing polarizarion phenomena in TlBr detectors,” Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, A585, pp. 102–104, 2008.View at: Google Scholar