Elsevier

Journal of Crystal Growth

Volume 470, 15 July 2017, Pages 20-26
Journal of Crystal Growth

Multi-ampoule Bridgman growth of halide scintillator crystals using the self-seeding method

https://doi.org/10.1016/j.jcrysgro.2017.03.038Get rights and content

Highlights

  • KCaI3:Eu crystals were grown using the multi-ampoule Bridgman method.

  • Scintillator performance was compared from different sections of the boule.

  • 25 mm × 25 mm cylinders were produced with energy resolutions of 3.5–4.7% at 662 keV.

Abstract

We investigate the multi-ampoule growth at 25 mm diameter of ternary iodide single crystal scintillator KCaI3:Eu using the randomly oriented self-seeded Bridgman method. We compare scintillation performance between cubic inch scale crystals containing small variations of low nominal europium concentrations previously shown to balance light yield with self-absorption in the host crystal. Growth conditions were optimized in the developmental furnace and four 2 in3 KCaI3:Eu crystals were grown simultaneously producing a total of six 25 mm × 25 mm cylinders. Small variations in activator concentration did not result in significant performance differences among the six measured crystals. A range of energy resolutions of 3.5–4.7% at 662 keV was achieved, surpassing that of NaI:Tl crystals commonly used in spectroscopic detection applications. The function and basic design of the multi-ampoule furnace as well as the process of growing single crystals of KCaI3 is included here.

Introduction

Halide based single crystal scintillators are widely used in spectroscopic detectors of ionizing radiation such as X-rays and gamma-rays in order to identify and intercept radiological materials. Materials used within the detectors in these applications must have excellent energy resolution in order to distinguish the signatures of special radiological material and naturally occurring radiation (background). Halide scintillators such as thallium doped NaI (NaI:Tl) and CsI (CsI:Tl) can attain energy resolutions of 6–7% at 662 keV. These materials have been in use for decades and their growth processes using variations of the Bridgman or Czochralski methods have been well developed, making the costs low enough for widespread use. Recently an emphasis has been placed upon enhancing isotope identification technology as a response to the increasing threat of nuclear attacks on a global scale and in support of enforcing nuclear non-proliferation treaties. This effort is reliant upon development of cost effective alternatives to the superior spectroscopic capabilities of semiconductor technologies such as high purity germanium (HPGe) based detectors [1]. Room-temperature operating direct-collection semiconductor detectors such as cadmium zinc telluride (CZT) have also been in development for some time [2]. However, despite energy resolution of ≈1% or better at 662 keV, the combination of special operating conditions, the extremely high purity required of these materials, and the difficulties encountered in crystal growth results in high costs and limited availability for widespread deployment.

Development of high performance scintillators such as LaBr3:Ce [3] and SrI2:Eu [4] has resulted in energy resolutions in the 3% range thereby bridging the performance gap between existing semiconductor based detectors and scintillator based detectors using NaI:Tl and CsI:Tl. Nonetheless, these new materials suffer from high costs that can be attributed in part to lower yields of single crystal growth processes and the high cost of raw materials. Consequently, present efforts on behalf of a great many research entities spanning national labs, private industry, and research universities are engaged in the search for yet another generation of scintillator and semiconductor materials with a greater emphasis on cost reductions to improve the commercial availability.

Challenges to scintillator production can be generally divided into those encountered in the growth of large (1 in3 scale) single crystal monoliths, and those pertaining to achieving uniform composition/performance characteristics. In the case of scintillation materials used in a spectroscopic detector, the crystals must be of sufficient optical quality and free of significant inclusions and cracking to ensure efficient and uniform light collection. The achievable energy resolution is the most important metric to isotope identification which is primarily a function of the scintillation light yield, non-proportionality of response to a range of excitation energies, and uniformity of scintillation light production and collection within the crystal and photosensor respectively [5]. For activated or extrinsic scintillators, the uniformity of scintillation light production relies upon the homogeneity of defects and activator species throughout the crystal bulk.

Until recently, Potassium containing scintillators have been overlooked for their use in radioisotope identification devices due to the drawbacks associated with the natural abundance of 40K. Investigations into the ternary iodide materials such as KCaI3:Eu, K(Ca,Sr)I3:Eu and KSr2I5:Eu have shown these compounds to share desirable properties with high scintillation light yields and capable of attaining energy resolution in small crystals of ≈3% or better at 662 keV [6], [7], [8]. Scale-up efforts have shown cubic inch scale crystals of these materials maintain the excellent light yield and energy resolution and warrant their continued development [9], [10], [11]. The presence of 40K in these materials produces a pronounced beta background along with the characteristic 1460 keV gamma photopeak in pulse-height spectra and negatively affects time-to-detection figure of merit for weak source activity. In LaBr3:Ce and CeBr3 scintillators, a similar elevated background due to natural abundance of 138La and actinium impurities hinder their use in low count rate applications but are desirable in applications requiring excellent energy resolution in high source activity environments [12]. Sourcing material with reduced concentrations of the 138La isotope and improved processing have been effective in reducing the elevated background in these materials [13].

Development of Bridgman growth processes at the cubic inch scale for KCaI3:Eu has made rapid progress resulting from extensive growth experiments in transparent and conventional furnaces. As a result, this composition has been the focus of current efforts to investigate its potential for further development. One challenge encountered for scale-up of europium doped scintillators is maintaining performance as the crystal volume is increased due to complications arising from self-absorption [14]. A previous examination into the effects of increasing crystal size and europium concentrations in this compound show self-absorption can be minimized with use of low Eu doping in the 0.5–1.0 at% range. Use of lower Eu dopant can maintain shorter scintillation lifetimes on the order of 2 µs and maintain excellent energy resolution in the 4% range for cubic inch scale crystals.

Optimized growth conditions for KCaI3 single crystals have been incorporated within a multi-ampoule growth station (MAGS) developed at our research facility at the University of Tennessee. The developmental Bridgman-type furnace adopts the concept of growing multiple boules of smaller diameter simultaneously when it is economically unfeasible, or exceedingly difficult, to produce the same volume and quality of crystal in one boule with a larger diameter. Such multi-ampoule furnace designs incorporating an inductively or resistively heated crucible for growth of GaAs as well as fluoride based materials have been proposed [15], [16], [17]. Similarly, multi-ampoule furnaces have been designed at the Shanghai Institute for Ceramics, Chinese Academy of Sciences (SICCAS) for large-scale production of 34 × 34 × 360 mm PbWO4 crystals for use as detectors in electromagnetic calorimeters [18], as well as piezo electric crystals Li2B4O7 (at ∅105 mm) [19] and Sr3Ga3Ge4O14 (at ∅50 mm) [20]. In each case of the SICCAS developed furnaces, the conventionally cylindrical bore is instead elongated to accommodate multiple ampoules in a side by side configuration, all using a single crystal seed fitted in the bottom of the ampoule to promote growth of a single grain of a desired orientation. This approach was successful in each case in producing multiple single crystal boules of high quality.

To our knowledge, this modification of the vertical Bridgman process has not yet been employed in the growth of iodide based crystals. One major difficulty in adapting the process to iodide crystal growth is the fabrication and handling of seed crystals with a desired orientation from the typically hygroscopic and brittle single crystals. During the course of our research surrounding KCaI3 utilizing the MAGS we have exclusively employed growth using a randomly oriented self-seeded approach using grain isolating capillary geometries [21], [22].

In this work we compare scintillation performance between cubic inch scale crystals of KCaI3:Eu grown in the MAGS. We examine the effects of varying low nominal europium concentrations on the basis of spectroscopic performance, namely energy resolution, as well as scintillation decay time. We will illustrate the basic design and functionality of the MAGS and procedures used for growth of multiple single crystal boules of KCaI3:Eu at ∅25 mm, simultaneously, using a randomly oriented self-seeded approach. Obstacles encountered in optimizing this process for KCaI3:Eu will also be discussed.

Section snippets

The multi-ampoule growth station

The multi-ampoule growth station (MAGS) was designed and constructed using widely available ceramic materials and industrial grade power handling components. The furnace has a maximum operating temperature of 1000 °C and is divided into two independently controlled heating zones separated by a 2.5″ insulated region housing a diaphragm. The basic construction is comprised primarily of firebrick supported by a steel frame. Use of the diaphragm between the heating zones is critical in maintaining a

Results

The resulting as-grown boules within their ampoules are shown in Fig. 4. Prior to opening, each boule was first inspected by inverting the ampoule and carefully sliding the crystal into the non-coated section. Boule #1 (1st on the left) appeared to have severe cracking throughout the boule while the remaining boules had only minor cracking isolated to the last-to-freeze region and not extending beyond a few mm into the boule. The outer surface of each boule contains trace carbon from the

Conclusion

The multi-ampoule Bridgman method has been successfully applied for growth of ∅1″ ternary iodide single crystals using a randomly oriented self-seeded approach. The same approach can be applied to growth of ∅1.5″ or larger crystals in the developmental furnace and is an ongoing effort. The omission of a seed crystal can be prone to failure in the self-seeding process but can be obviated by optimizing the capillary geometry, namely widening the capillary diameter to 3–4 mm to reduce bubble

Acknowledgments

This work has been supported by the U.S. Department of Homeland Security, Domestic Nuclear Detection Office, under competitively awarded grant # 2012-DN-077-ARI067-05. This support does not constitute an express or implied endorsement on the part of the Government. The authors would also wish to thank Dr. Kan Yang, Dr. Harold Rothfuss, and William McAlexander for their contribution to the early stages of this investigation. We are also grateful for the technical services of Doug Fielden, Larry

References (32)

Cited by (11)

  • Effect of OH content in the quartz crucible on the growth and quality of CsI single crystals and remedies

    2020, Journal of Crystal Growth
    Citation Excerpt :

    Crucible shape and size plays a central role in Bridgman technique as it contains melt and facilitate directional solidification and grain selection during single crystal growth. In general, when it comes to halide and rare earth halides, fused silica or quartz is material of choice for crucible fabrication owing to its excellent thermo-mechanical and chemical properties featuring very low thermal expansion coefficient, high temperature stability, inertness and low cost [13–17]. Different grade of fused silica with varying impurity contents and concentration are commercially available for crucible fabrication.

  • Effects of zirconium codoping on the optical and scintillation properties of SrI<inf>2</inf>:Eu<sup>2+</sup> single crystals

    2020, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
  • Growth of large size ([Formula presented] 38 mm diameter) KCaI<inf>3</inf>:Eu scintillator crystals

    2019, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
View all citing articles on Scopus
View full text