Optimal bandgap variants of Cd1−xZnxTe for high-resolution X-ray and gamma-ray spectroscopy

https://doi.org/10.1016/S0168-9002(98)01575-7Get rights and content

Abstract

We show that the trade-off between noise and charge generation statistics in Cd1−xZnxTe leads to an optimal band gap of approximately 2.0 eV at room temperature. This implies a ZnTe fraction of approximately 0.7–0.8. We show that for X-rays and relatively low energy gamma-rays Cd0.2Zn0.8Te theoretically offers a significant potential improvement in energy resolution over Cd0.9Zn0.1Te even if compensation of shallow levels is less complete and carrier lifetimes are an order of magnitude lower for the higher x variant. We also show that these calculations are consistent with observed detector performance reported by many workers over a large period of time.

Introduction

Cadmium Zinc Telluride (Cd1−xZnxTe or CZT) has emerged as one of the dominant materials for room-temperature spectroscopy of gamma-rays and hard X-rays. Until recently, the dominant issue in CZT detector technology has been the transport properties of charge carriers, particularly, holes. With the enormous progress that has been made in dealing with hole-tailing problems via electron only devices [1], rise-time discrimination and compensation techniques [2], and improvements in hole transport [3] it would seem that it is appropriate to ask what composition of CZT would optimize energy resolution.

The key material parameter controlling the ultimate energy resolution is the band gap Eg. Since both electron–hole pair production energy and resistivity increase with Eg, statistics argue for a narrow band gap while noise considerations require a wide band gap. In this work, we apply well-known results for the broadening due to noise and charge-generation statistics to obtain an expression for the ultimate energy resolution as a function of material and operating parameters. We use this expression to investigate the question of the optimal band gap as a function of photon energy and temperature.

As one considers the optimal band gap a number of issues must also be taken into account in terms of their effect on detector performance. These include the effect on transport properties of varying composition, the degree of nonuniformity in the material, and the lower atomic number of Zn relative to Cd. In addition, we show that the results obtained by many workers over many years are consistent with the predictions of our model.

Section snippets

Intrinsic energy resolution as a function of material parameters

The statistical variation in the number of electron–hole pairs generated by absorption of a photon of a given energy is [4]σ2N̄2=FN=where N̄ is the mean number of electron-hole pairs created, σ2 is the variance of the (assumed Gaussian) distribution of that number, F is the Fano factor, ε is the average electron–hole pair production energy and is the energy of the incident photon. The broadening due to parallel (leakage current) noise in the detector is [5]σ2p=ILτoqN2pwhere σ2p is the

Composition dependence of transport properties of Cd1−xZnxTe

A key question affecting the viability of high x variants of Cd1−xZnxTe is whether adequate charge collection can be achieved. This in turn depends on the intrinsic mobility and the impact on trap levels and density.

Sensitivity

An important issue in addition to energy resolution is sensitivity, which suffers as x is increased since Zn has a lower atomic number (30) than Cd (48). However, it is important to remember that the key parameter is not the average atomic number, but the effective atomic number, which for photoelectric absorption isZeff,pe=x2Z5Zn+(1−x)2Z5Cd+12Z5Te1/5.Zeff,pe of Cd0.2Zn0.8Te is, therefore, about 95% that of Cd0.8Zn0.2Te so that the absorption coefficient for photoelectric absorption is about

Simulation methodology

We use a Monte Carlo simulation which we have discussed in earlier publications [20], [21] to predict detector performance. The model is a simple, phenomenological one which assumes the validity of the Hecht relation or its generalization to the case of non-uniform drift lengths. Hence the results presented here are most valid for large-area parallel-plate detectors at relatively low energies. The minimum obtainable Gaussian broadening is incorporated based on the analysis of Section 2. The

Experimental verification of dependence of resolution on band gap

With all the factors affecting detector performance ranging from the materials issues discussed here to contact and processing issues that affect the fabricated detector, one may yet remain skeptical as to whether this analysis truly reflects experimental reality. Conducting the experiment that is suggested by this work, namely, the fabrication of detectors in a standard geometry of various composition and testing them at various temperatures and for various energies is clearly not practical.

Conclusions

We conclude from this analysis that for planar detectors with “ohmic” contacts Cd0.2Zn0.8Te may provide a significant improvement in performance over Cd0.9Zn0.1Te at room temperature for X-rays and low-energy gamma rays if sufficiently resistive material with good transport properties can be produced. We have shown that there is substantial leeway in the degree of compensation and carrier lifetimes that must be achieved to obtain some improvement in performance. Our simulations also show that

Acknowledgements

This work was supported in part by the US Department of Energy, Office of Nonproliferation and National Security, Office of Research and Development NN20. The literature search of the final section was greatly facilitated by the review article of D. McGregor and H. Hermon [24].

References (24)

  • J.C Lund et al.

    Nucl. Instr. and Meth. A

    (1996)
  • M.R Squillante et al.

    Semiconductors for room temperature nuclear detector applications

  • D Chattopadhyay

    Solid State Commun.

    (1994)
  • K Suzuki et al.

    J. Crystal Growth

    (1994)
  • C Szeles et al.

    Nucl. Instr. and Meth. A

    (1996)
  • D McGregor et al.

    Nucl. Instr. and Meth. A

    (1997)
  • Z. He, G.F. Knoll, D.K. Wehe, J. Miyamoto, 1996 IEEE Nuclear Science Symposium Conference Record, vol. 1, IEEE,...
  • J.C Lund et al.

    IEEE Trans. Nucl. Sci.

    (1996)
  • K.G Lynn et al.

    MRS Symp. Proc.

    (1998)
  • G.F Knoll

    Radiation Detection and Measurement

  • F.S Goulding et al.

    IEEE Trans. Nucl. Sci.

    (1982)
  • C.A Klein

    IEEE Trans. Nucl. Sci.

    (1968)
  • Cited by (60)

    • Bandgap engineering of Cd<inf>1−x</inf>Zn<inf>x</inf>Te<inf>1−y</inf>Se<inf>y</inf>(0&lt;x&lt;0.27,0&lt;y&lt;0.026)

      2022, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
    • Growth of Cd<inf>1-x</inf>Zn<inf>x</inf>Te thin films with high Zn content by close-spaced sublimation

      2016, Vacuum
      Citation Excerpt :

      In both applications, high Zn composition in CdZnTe is preferred. For the application in X-ray and gamma ray detectors, it is predicted in theory that Cd1-xZnxTe with x value of about 0.7–0.8 has the best performance, whereas a lower x value leads to a lower energy resolution for a planar detector [5]. Besides, Cd1-xZnxTe films with the Zn content of about 0.6–0.8 have the band gap about 1.72 eV, which is a promising candidate used as the absorption layer of top cell in tandem solar cells.

    • Photoluminescence of CdZnTe thick films obtained by close-spaced vacuum sublimation

      2016, Journal of Luminescence
      Citation Excerpt :

      At present, the PL spectra of bulk semiconductor crystals are studied well enough [21–24], but thin and thick films of these materials have not been thoroughly investigated. It was found theoretically that the optimal band gap about of 2.0 eV (x=0.7−0.8) for Cd1−xZnxTe crystals is more suitable for operation at room temperature of high-resolution X-ray and gamma-ray spectroscopy than Cd0.9Zn0.1Te [25]. Therefore it is very important task to study Cd1−xZnxTe thick films with large Zn concentration.

    View all citing articles on Scopus
    View full text