CRYSTAL GROWTH FURNACE SYSTEM CONFIGURATION AND PLANNED EXPERIMENTS ON THE SECOND UNITED STATES MICROGRAVITY LABORATORY MISSION

The Crystal Growth Furnace (CGF) is currently undergoing modifications and refurbishment and is manifested to refly on the Second United States Microgravity Laboratory (USML-2) mission scheduled for launch in September 1995. The CGF was developed for the National Aeronautics and Space Administration (NASA) under the Microgravity Science and Applications Division (MSAD) programs at NASA Headquarters. The refurbishment and refiight program is being managed by the Marshall Space Flight Center (MSFC) in Huntsvil/e, Alabama. Funding and program support for the CGF project is provided to MSFC by the office of Life and Microgravity Sciences and Applications at NASA Headquarters. This paper presents an overview of the CGF system configuration for the USML-2 mission, and provides a brief description of the planned on-orbit experiment operation.


Introduction
The CGF successfully completed its maiden flight in June/July 1992 on the First United States Microgravity Laboratory (USML-1) mission.
The system performed successfully in all aspects, and all the mission objectives were met.Seven samples were successfully processed yielding valuable results.
During the mission, a number of system capabilities were exercised, including demonstration of the crew interaction with the experiment hardware using the Flexible Glovebox (FGBX) for sample insertion and retrieval.
The Principal Investigator (PI) was allowed interaction with the experiment operation by means of realtime ground commanding to control the initiation of crystal growth.The flight hardware was returned to the contractor facility in September 1992 for postflight checkout which was completed in December 1992.At the same time, postflight ground truth science testing in the Ground Control Experiment Laboratory (GCEL) unit for all the four USML-I PIs was also performed and concluded in February 1993.Since then, the CGF system has been undergoing refurbishment and modification to upgrade the system capabilities to accommodate additional science requirements and enhance the system reliability for flight on the USML-2 mission.
The key upgrades to the system are as follows: (1) The addition of the Current Pulse Interface Demarcation (CPID) capability, (2) the development of a new Sample Ampoule/Cartridge Assembly (SACA) to provide the necessary interface for current pulsing through the sample via the CPID system, and (3) modification to the Control and Data Acquisition System (CDAS) to incorporate an 80486 Control Processing Unit (CPU) and a new Remote Acquisition Unit (R.ALI) Interface.The refurbishment and modification of the GCEL unit has been completed and is currently supporting the ground-based science development testing for the reflight of the four USML-2 peer-selected experiments along with an additional Interface Demarcation Flight Test (IDFT).
The flight unit refurbishment and modification activities are progressing well to support the launch schedule of September 1995.The hardware delivery to the Kennedy Space Center (KSC) for mission integration is scheduled for September 1994.
In this paper, a brief description of the various modifications to the CGF system is presented, and an overview of the system configuration for the USML-2 mission is given.A brief description of the planned on-orbit experiments is also presented.

•CGF System Configuration
The CGF USML-1 baseline configuration is described in reference 1.Only modifications/additions to this baseline system configuration are described here briefly.

CGF System Modifieatior_
Modifications to the CGF system for the USML-2 mission have been based upon the following:

•
Lessons learned from USMI.-1 • .PI and mission-specific requirements • Upgrades to system The following changes have been incorporated into the system design as a part of lessons learned from USML-I: • RAU interface board in the CDAS has been redesigned, and the flight software has been modified to prevent "skip" condition and Dedicated Experiment Processor (DEP) load problems.
• Onboard crew displays on the Spacelab Data Display System (DDS) have been streamlined and are being updated to provide additional system data and to include command protection for certain critical, crew-initiated commands.
• Onboard-generated error messages have been updated.
Changes resulting from PI and mission-specific science requirements include the following: The software controls the SIDS functions in accordance with a user-defined timeline, where system • parameters such as the amplitude, width, timing, and polarity of each pulse may be specLt'ied.
The software monitors SIDS system performance to prevent damage to the hardware and to reconfigure the system in an effort to continue operations in the event of a failure.

"T'."T'i"|"
.'" "."l":"i'"_ T':'T'|" "'£' ;"YT"_"f'T':" %';'T'_"I ""_"f"f"_...... |" "i"i" :"'."f"Fi" "',"_ " ""i'":' "t'"f'i'T'l"  Depending upon the configuration of the three relays in the PCS, this current is routed either to the mechanical -pulsing solenoid or to the CPID mate/demate mechanism (both located within the IFEA), where (for CPID operations) it is transmitted across the SEM and into the SACA.This mechanism performs a multiplexing function, such that only the SACA which is located in the processing position receives the electrical or mechanical pulse.A relay in the IFEA, operated under SIDS software control, enables the system to reverse the polarity of the CPID current flowing into the SACA.A switch located in the IFEA allows the crewmember to inhibit SIDS operations during the manual sample exchange process to eliminate the possibility of electrical shock.
Several features have been designed into the CGF SIDS to improve the system operations and to provide additional support of the science requirements.These are as follows:

•
Real-time commands will be available to enable the user to modify the existing or any future timeline segments, to hold and restart the timeline at any point, and to transmit a "pulse now" command.

•
The PCDs are currem sources, where the output current is the controlled parameter.This results in superior current amplitude control performance over a wide range of SACA resistances when compared to the use of traditional voltage sources.

•
The PCDs enable the deIivery of high peak currents into a wide rangeofSACA resistances.Each PCD iscapable ofproviding a peakoutput power ofup to500W, witha peakcompliance voltage exceeding 26 V atfull load.

•
The SIDS protects itself from damage due toelcctrical faults.The PCDs willrecover from operation into either an open-ora short-circuit without damage.Inaddition, thesoftwarecontroller monitors thepower provided by thePCDs.Ifthispower exceedstheratedvalue, the current pulse is terminated.

•
In the event of a CGF hardware failure, the CGF software may reconfigure the PCS relays to enable the SIDS PCDs to be used in support of furnace temperature control SIDS capabilities will be reduced accordingly in this event.

CPID SACA Dc_ign
The development of a SACA to support the processing of gallium arsenide (GaAs:Se) and gallium-dopcd germanium (Ge:Ga) using C-'PID has posed a significant design challenge.Chemical compatibility among a cartridge material, the sample material (in the event of an ampoule failure), other internal SACA components, and the fumace environment is a serious issue that requires a detailed understanding of the high temperature reactions involved.
The first problem is that a single cartridge material may not be suitable for all applications, unless a properly designed coating can be developed.In addition, thermocouples and CPID conductors must survive in the internal SACA environment.
Choices with respect to sheathed versus unsheathed thermocouples, ampoule failure detection sensors, ampoule feedthrough design for CPID samples, and insulation pre-treatment must all be addressed to ensure a proper thermochemical balance within the confines of a SACA at elevated temperatures.This balance is driven by potential platinum/alumina gas rings that are driven by low oxygen partial pressures in the gas environment between platinum and alumina components, outgassing of various components within the SACA, and diffusion of thermocouple elemental material, causing thermocouple decalibration.These requirements become even more significant when coupled with the geometric constraints imposed by the CGF system and PIs, including a maximum cartridge diameter of 1.009 inches on a 24-inch-long tube closed on one end and with a 0.030-inch wall thickness.
Over the past 3 years, the CGF team has expended significant effort to develop SACAs for the USML-I and for the upcoming USML-2 flight of the CGF.Various cartridge materials have been evaluated including alumina, pyrolytic boron nitride, graphite, molybdenum, TZM, various carbides, and rhenium, in addition to the two materials that flew on USML-I: WC-103 (Nb/Hffri alloy) with silicide coating and Inconel.For the USML-2 flight of the CGF, the new design efforts are mainly directed towards the production of flight SACAs for the GaAs:Se sample, and the IDFT (Ge:Ga) both of which utilize the CPID capability.This effort has led to the requirement for the development of new cartridge material/fabrication methods and/or a coating on the interior of the cartridge.The coating must not only contain liquid sample materials and possible decomposition products, but it must be able to be applied uniformly down the length of the tube with a large aspect ratio.In addition, it must be compatible with the cartridge material and all internal SACA components.
Materials compatibility testing on various cartridge materials/coatings and possible chemical reactions with Ga, As, or GaAs have been performed.
The CPID SACA design is essentially similar to the SACA design flown on USML-1, with the exception of provisions made to incorporate components required to support electrical pulsing through the sample material.These components include both input and output electrical conductors, means to attach these conductors to the ampoule, and an additional connector devoted to CPID power delivery.A typical CPID SACA design configuration is depicted in Figure 3.As indicated above,theproposed transport experiment is verybasic and isexpected toprovide valuable information for the time dependence of mass transportand growth phenomena duringtheearly stages ofgrowth.A quantitative comparison of several of theabove properties forground control and spacecrystalswilldelineate theeffects ofresidual convecfivc contributions on these properties.Combined with theUSML-I experiments forgrowthtimesof6 and 8 hours, theresults of theplannedUSMI¢-2 experimentfor2 to 4 hours'growth periodwillyieldfunctional relationships (trends) foralltheaboveproperties.Thiswillconsiderably enhancethereliability ofthese observations.SACA Configuration: The ampoule ismade of fusedsilica (quartz).The sample consists of HgCdTe sourcematerial, and the substrates is<100> oriented CdTe singlecrystal wafer.Mercury iodide (HgI2)isusedasthetransport agent.There aresixK-type thermocouples placedtomonitorthe temperature ofthesource, thesubstrate, and thetemperature atotherpointsalongtheampoule.The ampoule isplaced inside an Inconcl cartridge, md thecartridge issealed.
• To rcaffn'm thatpmprocessed HgZnTe alloy crystals can be successfully quenched,back-melted, and regrown maintaining nearly steady-state compositions.

•
To freeze in the diffusionboundary layer under controlled solidification conditions and firom analysis of theboundarylayercomposition verify thevalueforthe HgTe-ZnTe interdiffusion coefficient forthex = 0.16 alloy composition.

•
To performdetailed characterization oftherapidly frozen portions of the ingot to assess the Potential benefits of the reduced gravity environment for obtaining homogeneous alloy castings.
Seeded Growth of(Cd,Zn)Te by Directional Solidification Objectives: The plannedexperiment isnearly identical tothe experiments that were performed on USML-1.Two different ampoule configurations will be used for the USML-2 experiment development.
The specific objectives for the USML-2 phase of the investigation are as follows:

•
To quantify the effects of the gg environment on the mechanical suain and defect distribution within the (Cd,Zn)Te crystal.
The approach involves the development of a model of the quasi-steady-state thermomechanical stress field and the comparison of the predicted stress fields with quantitative measurements from well-characterized one-g and gg samples.

•
To empirically and analytically investigate the dislocation and defect content of one-g and _tg processed crystals and to relate the defects to growth conditions.

•
To quantitatively examine the transport conditions using numerical models.These models focus on the prediction of the Ixg transport conditions in order to assess the sensitivity of the experiment to the acceleration environment and the effect of thermal and gravitational asymmetry on one-g and _tg tzansport.

SACA C0nfigurati0n:
The (Cd, Zn)Te sample consists of a high quality (111) _ crystal which is in contact with a (Cd, Zn)Te bulk.The sample (seed and bulk material) is enclosed in an evacuated, fused silica ampoule that has internal carbon nonwetting coating.The ampoule is instrumented with sixKtypethcrmocouples and integrated intoa WC-I03 cartridge coatedinsideand outsidewith chrome-ironsilicide and sealed.
Seed_ Crystal Growth of Selenium-doped G_llium Ar_cnid_ (GaA_:Se) by Direetiorlal Sglidificatign Objectives: The experiment is designed to specifically characterize and controUably modify the melt-solid interface shape during the growth to achieve uniform radial segregation of the dopant in the solid.This experiment will be a direct follow-up of the USML-1 experiments which focused on obtaining axial dopant uniformity.
In the planned experiments, the booster heater and the gradient zone configuration of the RFM will be used to achieve a planar or near-planar interface shape in order to minimize radial dopant variation.
Since control of the interface shape during the growth is the principal goal of the USML-2 experiments, it is necessary to measure the interface shape during growth.The technique of interface demarcation by current pulsing (Peltier pulsing) will be extensively used, This technique has been used successfitIly to determine the segregation behavior of semiconductors on a rnicroscale.Thus, theseexperiments are expected toprovide new and important experimental datafor thecontinuation andupgrading ofheattransfer andfluid-flow models on a commercially important semiconductor material system.The current requirement for the pulses will be based on obtaining the necessary current density in a 1.50-cm diameter sample to establish the interface demarcation.

SACA Confimtratign;
The ampoule is made of fused silica (quartz).The growth boule (sample material), seleniumdoped gallium arsenide, is enclosed inside a pyrolytic boron nitride (PBN) sleeve.The PBN sleeve is closed at one end by means of a graphite pedestal, and the other end has a graphite chamber in which a graphite plunger is provided to support a PBN leaf spring to allow expansion of the boule volume.
The graphite contacts have molybdenum wire contacts attached to them.This assembly is hermetically sealed by seaIing the ampoule around the molybdenum foil of the feedthrough.These feedthroughs terminate in a loop where the interface demarcation current leads are attached.The sample ampoule is integrated into the specially configured CPID SACA.
The SACA is backfilled with argon to a desired pressure and sealed.Six S-type thermocouples are located inside the SACA to monitor experiment processing.
IDFT: Growth of Ge:Ga by Directional Solidification Objectives: This flight test is designed to assess the feasibility of interface demarcation in a microgravity environment and to identify the effects on interface shape which may be caused by low-level accelerations.
Since control of the interface shape during growth is a significant goal of three of the USML-2 experiments, it is necessary to measure the interface shape during growth.This test is designed to characterize the melt-solid interface shape during the growth process.For the USML-2 mission, current pulsing (Peltier pulsing) will be the technique of interface demarcation.The current requirement for the pulses will be based on obtaining the required current density in a 1.40-cm diameter sample.
SACA Configuration: The ampoule is made of Ge#214 fused silica (quartz).The growth boule (sample material), Ge:Ga, is enclosed inside this ampoule between two graphite cups.These cups serve as current contacts for interface demarcation.The graphite cups have platinum wire contacts which are spot-welded to molybdenum foil.Each molybdenum foil forms a hermetic seal where it passes through the end of the quartz ampoule.Spot-welded to the molybdenum foil outside the ampoule is another platinum wire which terminates in a loop where the interface demarcation current leads are attached.The sample ampoule is integrated into the specially configured CPID SAC#.and sealed.
Six S-type thermocouples are located inside the SACA to monitor experiment processing.

E.xt)eriment Processing Scenario for USML-1
The experiment processing scenario for the USML-2 mission is defined below: • Twelve (SACAs) samples will be carried on board as stowed items.

•
Six selected samples will be manually loaded into the SEM by a crcwmember following CGF activation and preparation for manual sample exchange.
• Processing of the samples in the predef'med sequence will then be performed automatically, and five samples will be nominally processed.
• Upon completion of processing, the processed SACAs will be retrieved by a crewmembcr and restowed for the remm flight.

Summary
The Crystal Growth Furnace was designed and developed to support the United States Microgravity Laboratory (USML) and Microgravity Science Laboratory (MSL) flight opportunities for microgravity research and performed flawlessly on its maiden flight on the USML-I mission.
The flight and prototype ground systems have undergone extensive modifications and enhancements in order to provide better operational flexibility in achieving scientific objectives and to increase programmatic confidence.The CGF has demonstrated the capability for supporting needed research and development in a microgravity environment on various important electronic and photonic materials of interest.
This sophisticated second generation high temperature processing facility is providing the means for furthering the understanding of the complex phenomena inherent with both diffusion controlled and vapor transport growth and will lead to significant improvement of pr0cesses and materials for future applications.
The important CPID upgrade for the USML-2 mission will enable the investigator to mark the growth interface during processing in order to determine more precisely growth rates and interface shape within the melt, and will mcrease the science yield from the flight.In addition, the information provided by the CPID experiment and flight test will not only serve to characterize the melt-solid interface shape during the growth process, but will also denote the effects on interface shape caused by low level accelerations imposed by various Orbiter attitudes during processing.
The four reflight experiments are proposed from both private industry and leading research institutions.They wero selected from a peer review process and represent the leading flight research programs in this country.The experiment and investigation teams have been working diligently and tirelessly in analyzing the previous flight results and in conducting ground-based testing in an attempt to carry their scientific investigations to the highest level.
The requirements of materials technology development continues to provide the impetus for the continually improving furnace facilities to be utilized in a mierogravity environment.
The evolution of furnace facilities such as the CGF with its current enhancements and upgrades, while currently an R&D development, offers new and expanded options for the characterization, development, and exploitation of new materials to meet future commercial and industrial needs.Understanding these processes that occur in microgravity will greatly enhance the technology base for the development of important electronic and photonic semiconductor materials.

Significant
research findings have resulted from directional solidification growth experiments conducted in a microgravity environment and compared to terrestrially grown crystals.Crystals grown by chemical vapor transport in a microgravity environment have also demonstrated improved crystal morphology, lower defect densities, and higher growth rates than demonstrated in ground-based processing.
The planned series of USML and MSL flights require the transportation of the processing facility into Earth orbit on each mission.
With the concept of free-flyers or Space Station furnace facilities, longer processing time will be available which will greatly expand the technology base for accommodating materials research.This will afford the opportunity for processing a greater number of different samples and for processing new materials requiring much slower growth rates with more precisely controlled timelines.
The CGF furnace, as it has evolved, is currently manifested as a primary payload on the USML-2 mission scheduled for launch in mid-September 1995.In addition, the facility has also been recently manifested as part of the payload complement on the MSL-1 flight now scheduled for launch in early 1997.A derivative of this design capability is currently being developed to fly on the Space Station and will provide a valuable resource and a long-term and continuing capability for the United States for materials research and development well into the next century.

Table I .
In addition, a new flight RFM has been built to the same cortfiguration as flown on the • Addition of the CPID capability to send current pulses into the sample being processed to mark and locate the crystal growth interface Coymeans of Peltier effect) • Development of the SACA to provide interface for CPID • Upgrades to science data and graphics displays for PI use on the ground Upgrades to the system consists of the following: * Program Scientist, Senior Member AIAA t Program Manager,:_Lead SystemsEngineer, § Lead EngineerSACA Development,Member A/AA Project Manager

Booster Heater Temperature Gradient Zone Length Sample Size Heatin[[ Rite Absoh_te Control Set Point Accuracy]Stability Pumace Translation Rate Induced Aoeeleratio_t in the Sample
At thetopof theampoule,sixplatinumwires arewelded to thelugsand alsowelded toa platinuminput electrode which islocated in the centerof theSACA and terminates atthetop of theSACA in a leadwire.The six wires allow fordifferential thermalexpansionof intcrnal SACA componentsduringthcrmal cycling.At thebottomof ARGONe J,C_-F'IL(.Reconllgurable Furnace ModuleHot Zone Temperat'm_Cold Zone Tempenttu_ the effects on interface shape which may be caused by lowlevel accelerations.The experiment tides and the Investigator teams for the respective experiment are given in Table m.