Scalability study of solid xenon
J. Yoo, H. Cease, W.F. Jaskierny, D. Markley, R.B. Pahlka, D. Balakishiyeva, T. Saab, M. Filipenko
PPreprint typeset in JINST style - HYPER VERSION
Scalability study of solid xenon
J. Yoo ∗ , H. Cease, W. F. Jaskierny, D. Markley, and R. B. Pahlka Fermi National Accelerator Laboratory, Kirk and Pine St., Batavia, IL 60510, USA
D. Balakishiyeva and T. Saab
Department of Physics, University of Florida, Gainesville, FL 32611, USA
M. Filipenko
Erlangen Center for Astroparticle Physics (ECAP), Friedrich Alexander University ofErlangen-Nuremberg, Erwin-Rommel-Straße 1, 91058 Erlangen, Germany A BSTRACT : We report a demonstration of the scalability of optically transparent xenon in the solidphase for use as a particle detector above a kilogram scale. We employed a cryostat cooled byliquid nitrogen combined with a xenon purification and chiller system. A modified
Bridgeman’stechnique reproduces a large scale optically transparent solid xenon.K
EYWORDS : Cryogenic detectors, Photon detectors for UV, visible and IR photons (solid-state),Time projection chambers. ∗ Corresponding Author: [email protected] a r X i v : . [ phy s i c s . i n s - d e t ] A ug ontents
1. Introduction 12. Solid xenon test stand 23. Scalability of solid xenon 44. Discussion and Summary 7
1. Introduction
The Standard Model (SM) of particle physics has been explored with amazing accuracies from thescale of the Hubble radius to the size of nucleons. Despite the remarkable success of the SM, anumber of observations have recently emerged suggesting the incompleteness of our understandingof fundamental interactions. Three main pieces of evidence contribute to this conclusion: the CP-asymmetry of the Universe, the existence of dark matter and the discovery of massive neutrinos. Inall these cases, low background experiments in deep underground sites provide excellent venues todiscover new physics Beyond the SM.Noble elements in both the gas and liquid phases have proven use as excellent low backgroundradiation detectors [1, 2, 3, 4, 5, 6]. Xenon has drawn special attention among the noble elementsdue to several distinct advantages over its lighter counterparts. The liquid phase of xenon pos-sesses a very high scintillation light yield (40 ∼
60 photons/keV) and the vacuum ultraviolet (VUV)wavelength (178 nm) [7] of the scintillation is optically transparent in xenon [8]. Xenon also hasexcellent ionization and electron transport properties and the absence of long-lived radioisotopes inxenon results in no intrinsic background radiation sources. The large atomic mass of xenon (131.3and Z=54) keeps external radioactive decay sources at the outer surfaces of the condensed xenondetector, resulting in improved self-shielding effects. As xenon is a noble element, the chemicalpurification is straightforward by using hot getter and/or gas distillation systems that can removemost of the non-noble contaminants. A wide range of applications has been studied on particletracking and spectroscopy including γ -ray astronomy, neutrinoless double beta decay (NLDBD),dark matter searches, neutrino coherent scattering experiments and medical imaging devices.The solid (crystalline) phase of xenon not only inherits most of these assets from liquid xenon,but also has additional advantages as a particle detector material. Electron drift speeds in thinlayers of solid xenon have been measured to be faster compared to those in the liquid phase [9].Optically transparent solid xenon ("crystal xenon") is also transparent to the xenon scintillationlights which are in the VUV range [7, 10, 11, 12]. Furthermore, an increased amount of lightcollection in the solid phase compared to the liquid phase using the same detector system in alphairradiation has been reported [13]. Therefore, compared to a liquid xenon detector, a solid xenon– 1 –etector can become an ionization detector with faster response and/or a scintillation detector withimproved photo collection. In addition, as the density of solid xenon (3.41 g/cc) is higher than thatof liquid xenon (2.95 g/cc), an even further compact detector can be built in the solid phase. Onemay also imagine a low energy bolometric phonon readout from solid xenon with superb energyresolution. Therefore, solid xenon can be a strong contender of low background and low energythreshold dark matter detector. Solid xenon can also be a great NLDBD detector material, where theBa ++ ions that are produced from the xenon NLDBD would be frozen at their decay locations andsubsequently identified via ion-tagging method. The double coincidence (energy and ion-tagging)will substantially reduce the NLDBD background in the energy region of interest.Xenon in the solid phase is chemically stable and forms simple face-centered cubic crystalstructures with interatomic binding energies coming from weak Van der Waals forces . It is knownto be relatively easy to produce small-scale clear and transparent solid noble element specimensthat are virtually perfect crystals [15]. The crystal structure [16, 17, 18, 19] and microscopic de-fects [20] are important properties for experimental applications such as solar axion searches [21].However, for most particle detector applications, it is important to first understand the macroscopicproperties such as density uniformity, optical transparency, charge drift velocity, scintillation, andionization. Particle detectors based on the solid phase of noble element active media have beenthoroughly investigated [9, 13, 22, 23, 12, 24, 25, 11, 26, 27, 28]. Even though most of thesestudies have successfully shown that the solid noble elements are excellent candidates for particledetector material, large scale detectors have yet to be realized. Our first R&D effort, therefore, isfocused on the proof of scaleability of optically transparent solid xenon that is distinguishable fromopaque frozen xenon bulk. In this paper, we describe the instrument details and demonstrate thescalability of solid xenon above a kilogram scale.
2. Solid xenon test stand
Figure 1 shows the schematics of the solid xenon cryostat. The cryostat consists of a stainless steelvacuum jacketed chamber with an outermost diameter of 30 cm with three 15 cm diameter glasswindow ports. Two concentrically placed glass chambers reside inside allowing optical access tothe xenon bulk volume. The larger of the two glass chambers is used as a liquid nitrogen bath forcooling and has a diameter of 23 cm, referred to as the LN (liquid nitrogen) chamber. The smaller10 cm diameter inner chamber houses the xenon volume and is made out of Pyrex with a 5 mmthick side wall and a 10 mm thick flat bottom, referred to as the xenon chamber. The stainless steelchamber also functions as a safety protection chamber in case of unexpected pressure changes in The conditions imposed by mechanical stability limit the possible structures of ideal xenon crystals. These aregenerally face-centered-cubic (fcc) or hexagonal-close-packing (hcp). The crystal structure of xenon changes fcc to hcpunder pressure above ∼
20 GPa [14]. The lattice cell parameter is 6.2 Å in fcc. These crystals nevertheless are polycrystalline in many cases and contains a large number of microscopic defects.The density of structural defects is highly depends on the growing conditions (temperature and pressure) of the crystal.For example, noble element crystals grown under vapor deposition method shows density of dislocations of order 10 cm − , while commercial silicon wafers shows dislocation density of order 10 ∼ − depends on the methods offabrication. For more details about the microscopic properties of noble gas solids, see reference [15]. The directional dark matter searches with crystal detectors would be a challenge due to the short ( ∼ µ m of) tracklength of the recoil nucleus. – 2 – urbo to TurboXenon ! in/out LN ! feed lineLNlevel ! probe Figure 1.
A schematic diagram of the solid xenon test stand the glass chambers. The LN chamber is equipped with an effective phase separator at the bottom,made with a combination of aluminum and polyethylene blocks. The phase separator, when itis pre-cooled to liquid nitrogen temperature, transfers nitrogen liquid into the LN chamber withminimal evaporation in the internal transfer line. The turbo line on the vacuum side is used foractive vacuum shielding of the cryogenic chamber, and the turbo line on the xenon chamber is usedto evacuate the chamber before the xenon transfer. The system has safety backup system whichincludes a 440 liter xenon recovery cylinder, a 250 liter buffer tank, two 4 liter xenon storagecylinders and a 4 liter cylinder for calibration gas. A commercial hot getter (PF4-C3-R-1 andMonotorr PS4-MT3-R-1 by SAES) and a circulation loop allows continuous purification of thexenon.For the temperature control of the inner chamber, we employ ten Nichrome 32-gauge heaterwires and platinum Resistance Temperature Detectors (RTD: PT-100) that are affixed to the bottomand barrel surface of the xenon chamber using cryogenic epoxy. A total of nine RTDs measure thethermal gradients of the outer surface of the xenon chamber and one thermometer-heater loop isinstalled near the bottom of the inner wall of the xenon chamber. While the cryogenic features ofthe system reported here largely overlap with those of liquid-based systems, we require additionaladvanced controls for temperature and pressure in order to keep the optical transparency withoutobvious defects in the solid xenon. The glass chambers place constraints on the pressure in theinner glass chamber, set to about 1 bar.A main control panel is designed to handle the xenon transfer and monitoring at a central-– 3 –zed location. It allows convenient monitoring and controlling of parameters, such as the xenonflow rate, cooling bath level, temperatures, pressures, and weights of the xenon storage cylinders.The cooling LN is fed from a continuously-serviced 20 ton external LN tank that facilitates un-interrupted cryogenic system operation. Using the cold nitrogen gas above the liquid to cool thexenon chamber, the liquid nitrogen level in the LN chamber is controlled by a feed-back system ofpneumatic control valves and differential pressure level meters with an accuracy of about 0.5 cm.The liquid nitrogen level is set to about 2 cm below the bottom of the inner glass chamber duringnormal operation and is therefore not in direct contact with the xenon chamber. The temperatureof the glass chamber is controlled using three sets of thermometer–heater feedback loops (bottom,barrel, and top). The insulation vacuum between the stainless steel chamber and the LN chamberis maintained at about 10 − Torr using a dedicated turbo pump.The xenon chamber is cleaned in a ultrasonic bath before the installation and is then bakedat 40 ◦ C for a few days using the attached heaters. The vacuum level of the xenon vessel reaches10 − Torr at room temperature with no detector instruments in the chamber and stabilizes at thisvalue at a temperature of 164 K. With detector instruments installed in the chamber, the vacuumlevel reaches below 10 − Torr after two days of evacuation. We used research grade xenon with99.999% purity level. The other components reported by the gas provider are: krypton ( < < < < < < < < <
100 ppm). No gascontamination above the RGA background level is observed in the xenon gas.
3. Scalability of solid xenon
The main focus of the scalability study is to understand the conditions required to produce opticallytransparent solid-phase xenon which is distinguishable from frozen opaque volumes or other typesof solid phases. Due to the density difference between the liquid phase (2.95 g/cc) and the solidphase (3.41 g/cc), the growing process of solid xenon requires special care in maintaining andcontrolling the growing speed.The configuration of the heater wires on the inner glass chamber is designed to produce 2 kg ofsolid xenon at maximum. We found that a transparent layer of solid xenon can be reliably grown viavapor deposition methods in a pre-cooled chamber. However, the growth rate of solid xenon wasquite slow ( ∼ Bridgeman’s technique [15] was adopted to grow solid xenon on a kg-scale. InBridgeman’s method, a temperature gradient of 1 ∼ ∼ L = (cid:112) ( K ∆ T ) / ( H ρ ) √ t ,– 4 – ime [hour]0 10 20 30 40 50 60 Thickness [cm] D Total solid xenon mass [kg]
Figure 2.
A simple thermal model of the expected solid xenon growth (red curve), and an example of actualgrowth of solid xenon (blue dots). The temperatures at the bottom and top of the liquid xenon volume areset to 145 and 161 K, respectively. The disagreement between the ideal 1-dimensional thermal model (redcurve) and actual growing speed (blue dots) might be due to different thermal configuration of the realisticxenon glass cylinder and the local convection effect of the liquid xenon above the solid xenon. where L is the thickness, t is time, the latent heat of xenon fusion is H =17.5 J/g, the thermal conduc-tivity of solid xenon is K =0.001 W/(cm · K) [29], and the solid xenon density is ρ =3.41/cc. Figure 2shows a simple thermal model of the expected solid xenon growth rate (red curve) and an exampleof actual growth data (blue dots). In this example, the bottom temperature is set to 145 K and toptemperature is set to 161 K. The actual growth rate of solid xenon is a little slower than expectedby the simple model. In order to grow 2 kg of solid xenon, which is equivalent about 9 cm high inthe xenon chamber, it requires more than 48 hours of stable temperature control. The details of thethermal configuration in the solid xenon chamber system requires fine tuning of the temperaturegradient. We found the following setup reliably reproduces the optically transparent solid xenon inour system.First, the xenon is liquified in the glass chamber at a set temperature of 163 K and pressure of14.5 ± ± ± igure 3. Photographs of various types of solid xenon. The top left photo shows a concave structure ofthe solid surface due to the over-cooling at the barrel. There is also an opaque substructure at the bottomcenter due to the significant temperature gradient at the bottom surface. These are produced when there wasa rapid variation of cooling liquid level. The top right photo shows frozen layers of solid xenon. Thesestructures can be easily created when the pressure drops below the vapor pressure. The bottom photo showssolid xenon in a polycrystalline state on the order of a few mm, provided when the temperature and pressurecontrol become unstable near the triple point (161.4 K and 11.9 PSIA). The dark brown color at the bottomof the chamber is kapton tape. The spacing of the heater between each wire is about 1 cm, which is a goodmeasure of the approximate height of xenon. xenon; including opaque spots, filamentary structures and voids. We do not observe any significantpressure dependence of the optical quality of the solid xenon so long as the pressure is kept wellabove the vapor pressure. Rapid pressure variations near the vapor pressure can easily produceopaque layers at the top surface of the solid volume. Failure to maintain temperature and pressurestability near the triple point creates a few millimeters of polycrystalline xenon which is clearlyidentified by the birefringence of each polycrystalline cell. Figure 3 shows a few examples of thesolid phases of xenon which were produced during the initial test of the solid xenon cryogenics.Figure 4 shows an example of the clear visual difference between frozen opaque xenon (left)and transparent solid xenon (right). Opaque xenon is produced when the temperature at the bottomof the glass chamber was reduced quickly ( ∆ T / ∆ t < ∼ igure 4. Photographs of opaque solid xenon (left) and transparent solid xenon (right). The left photographshows about 3 cm of opaque solid xenon and 1 cm of liquid xenon above the solid phase. The right photo-graph shows about 3 cm high (about 650 g) of transparent solid xenon and 1 cm (about 189.4 g) high liquidxenon on top of the solid xenon. ∆ T (cid:39)
4. Discussion and Summary
We demonstrated the scalability of transparent solid xenon above a kg-scale. We found a modified
Bridgeman’s technique is suitable to reproduce optically transparent solid xenon in a consistentmanner. The solid xenon production is automatically controlled using heater and thermometerloops in a gas-cooled glass chamber system. Our system is designed to maximize visual access– 7 – igure 5.
The time projection chamber (TPC) installed in the xenon glass chamber. In this particular sample,the amount of solid xenon in the chamber is about ∼ to the xenon volume while keeping the glass chamber in a safe condition. We produced about2 kg of optically transparent solid xenon at a temperature of 157 K. To our knowledge, this is thelargest single monolithic volume of optically transparent solid xenon that have been produced andreported. Developing solid xenon larger than a few kg would require substantial modifications toour existing system as the current set up may not be the best design to achieve a uniform thermalconfiguration for a larger scale solid xenon. A competitive dark matter experiment would require aton-scale background free detector. However, it would take more than 200 days to grow solid xenonto a height of 1 m using the same method that we used with a similar thermal configuration. There-fore, a substantial improvement in solidifying speed is required to develop a ton-scale detector. Forexample, a cold-bath method, where the xenon chamber slowly sinks into a temperature-calibratedcryogenic liquid bath, would be an attractive alternative when building a larger scale detector.The next interesting R&D topics are understanding the scintillation and charge transport prop-erties in a kg scale of solid xenon. Bolometric readout of energy deposit in heat channel wouldbe quite an interesting research topic in the future and will open up the possibility of measuringall three energy deposition channels of particle interactions – scintillation, ionization and heat.However developing solidification process of large scale crystal xenon at sub-Kelvin temperatureappears technically challenging. The vapor deposition method would be a good test option for alow temperature and a thin-layer (less than a cm scale) of solid xenon detector development. Wecan consistently reproduce transparent solid xenon bulk that was 5 mm thick at the temperature of77 K via the vapor deposition method.In summary, the research and development efforts towards employing solid xenon as a particledetector were presented. Using a three-chamber vessel with automated liquid nitrogen cooling– 8 –onditions, we demonstrated the scalability of optically transparent solid xenon above a kilogramscale. Acknowledgments
We are very grateful to M. Miyajima, J. White, and A. Bolozdnya for the initial discussions ofthe solid xenon particle detector and sharing their ideas. We thank R. Barger, D. Butler, R. Davis,A. Lathrop, L. Harbacek, K. Hardin, C. Kendziora, W. Miner, K. Taheri, M. Rushmann, E. Skup,M. Sarychev and J. Vorin at Fermilab for their tireless hard work to provide us the experimentalsetup with highest standard. We also thank V. Anjur, A. Anton and B. Loer for their participationof the system setup. This work supported by the Department Of Energy Advanced Detector R&Dfunding.
References [1] E. Aprile and T. Doke, Rev. Mod. Phys. , 2053 (2010).[2] EXO Collaboration, N. Ackerman et al. , Phys. Rev. Lett. , 212501 (2011).[3] XENON100 Collaboration, E. Aprile et al. , Phys. Rev. Lett. , 181301 (2012).[4] LUX Collaboration, D. S. Akerib et al. , Phys. Rev. Lett. , 091303 (2014).[5] KamLAND-Zen Collaboration, A. Gando et al. , Phys. Rev. Lett. , 062502 (2013).[6] XMASS Collaboration, K. Abe et al. , Phys. Rev. Lett. , 121301 (2014).[7] J. Jortner, L. Meyer, S. A. Rice, and E. G. Wilson, J. Chem. Phys. (1965).[8] M. Szydagis et al. , JINST , P10002 (2011), 1106.1613.[9] L. Miller, S. Howe, and W. Spear, Phys.Rev. , 871 (1968).[10] W. Baum, S. Gotz, P. Heeg, M. Mutterer, and J. Theobald, IEEE Transactions on Nuclear Science (1988).[11] D. Varding, I. Reimand, and G. Zimmerer, Phys. Stat. Sol (b) , 301 (1994).[12] S. Kubota, M. Hishida, M. Suzuki, and J.-z. Ruan, Nucl Instr Meth Phys. Res. , 101 (1982).[13] E. Aprile et al. , Nucl. Instrum. Meth. A343 , 129 (1994).[14] W. A. Caldwell et al. , Science , 930 (1997).[15] M. L. Klein and J. A. Venables, Rare Gas Solid, Academic Press , 610 (1977).[16] J. A. Venables and D. J. Ball, Proc. Roy. Soc. London , 331 (1971).[17] H. M. Kramer and J. A. Venables, J. Crystal Growth , 329 (1972).[18] H. M. Kramer, Journal of Crystal Growth , 65 (1976).[19] K. F. Niebel and J. A. Venables, Proc. Roy. Soc. A , 365 (1974).[20] J. A. Venables and D. J. Ball, 6th International Conference on Electron Microscopy , 333 (1966).[21] CDMS Collaboration, Z. Ahmed et al. , Phys.Rev.Lett. , 141802 (2009), 0902.4693.[22] A. I. Bolozdynya et al. , JETP Letters , 401 (1977). – 9 –
23] S. Himi, T. Takahashi, J.-z. Ruan, and S. Kubota, Nucl. Instr. Meth.Phys. Res. , 153 (1982).[24] R. Kink, K. Kalder, A. Lohmus, and H. Niedrais, Phys. Status Solidi B , 321 (1987).[25] G. Baldini, Phys. Rev. , 1562 (1962).[26] R. Michniak, R. Alleaume, D. McKinsey, and J. Doyle, Nucl. Instrum. Meth.
A482 , 387 (2002).[27] E. Gushchin, A. Kruglov, and I. Obobovskii, Sov. Phys. JETP (1982).[28] E. Aprile, K. Giboni, and C. Rubbia, Nucl.Instrum.Meth.
A241 , 62 (1985).[29] O. I. Purskii and N. N. Zholonko, Physics of the Solid State , 2015 (2004)., 2015 (2004).