September 1999 • NREL/SR-520-26566
`
`Atmospheric Pressure
`Chemical Vapor Deposition of
`CdTe for High Efficiency Thin
`Film PV Devices
`Annual Report
`26 January 1998—25 January 1999
`
`P.V. Meyers
`ITN Energy Systems (cid:125) Wheat Ridge, Colorado
`R. Kee, C. Wolden, L. Raja, V. Kaydanov, T. Ohno,
`R. Collins, M. Aire, and J. Kestner
`Colorado School of Mines (cid:125) Golden, Colorado
`A. Fahrenbruch
`ALF, Inc. (cid:125) Stanford, California
`
`National Renewable Energy Laboratory
`1617 Cole Boulevard
`Golden, Colorado 80401-3393
`NREL is a U.S. Department of Energy Laboratory
`Operated by Midwest Research Institute (cid:118)(cid:118)(cid:118)(cid:118) Battelle (cid:118)(cid:118)(cid:118)(cid:118) Bechtel
`Contract No. DE-AC36-98-GO10337
`
`HANWHA 1057
`
`

`

`September 1999 • NREL/SR-520-26566
`
`Atmospheric Pressure
`Chemical Vapor Deposition of
`CdTe for High Efficiency Thin
`Film PV Devices
`Annual Report
`26 January 1998—25 January 1999
`
`P.V. Meyers
`ITN Energy Systems (cid:125) Wheat Ridge, Colorado
`R. Kee, C. Wolden, L. Raja, V. Kaydanov, T. Ohno,
`R. Collins, M. Aire, and J. Kestner
`Colorado School of Mines (cid:125) Golden, Colorado
`A. Fahrenbruch
`ALF, Inc. (cid:125) Stanford, California
`
`NREL Technical Monitor: H.S. Ullal
`Prepared under Subcontract No. ZAK-8-17619-03
`
`National Renewable Energy Laboratory
`1617 Cole Boulevard
`Golden, Colorado 80401-3393
`NREL is a U.S. Department of Energy Laboratory
`Operated by Midwest Research Institute (cid:118)(cid:118)(cid:118)(cid:118) Battelle (cid:118)(cid:118)(cid:118)(cid:118) Bechtel
`Contract No. DE-AC36-98-GO10337
`
`

`

`NOTICE
`
`This report was prepared as an account of work sponsored by an agency of the United States
`government. Neither the United States government nor any agency thereof, nor any of their employees,
`makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy,
`completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents
`that its use would not infringe privately owned rights. Reference herein to any specific commercial
`product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily
`constitute or imply its endorsement, recommendation, or favoring by the United States government or any
`agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect
`those of the United States government or any agency thereof.
`
`Available to DOE and DOE contractors from:
`Office of Scientific and Technical Information (OSTI)
`P.O. Box 62
`Oak Ridge, TN 37831
`Prices available by calling 423-576-8401
`
`Available to the public from:
`National Technical Information Service (NTIS)
`U.S. Department of Commerce
`5285 Port Royal Road
`Springfield, VA 22161
`703-605-6000 or 800-553-6847
`or
`DOE Information Bridge
`http://www.doe.gov/bridge/home.html
`
`Printed on paper containing at least 50% wastepaper, including 20% postconsumer waste
`
`

`

`Table of Contents
`
`1
`
`Project objective....................................................................................................................................2
`
`1.1 Approach ...........................................................................................................................................2
`
`1.1.1 Deposition technology...............................................................................................................2
`
`1.1.2 Device analysis..........................................................................................................................3
`
`2
`
`APCVD .................................................................................................................................................3
`
`2.1 Background .......................................................................................................................................3
`
`2.1.1 Reaction chemistry ....................................................................................................................3
`
`2.1.2 Mass transport ...........................................................................................................................4
`
`2.2 First year (1998) reactor fabrication progress ...................................................................................5
`
`2.2.1 Adopted stagnant flow APCVD reactor concept.......................................................................5
`
`2.2.2 Developed APCVD “stagnant flow” reactor design..................................................................6
`
`2.2.3 Performed numerical simulations of reactor performance ........................................................7
`
`2.2.4 Fabricated APCVD reactor .......................................................................................................9
`
`2.2.5 Performed “dry runs” of reactor to evaluate and document reactor performance.....................9
`
`2.3 APCVD reactor plans......................................................................................................................10
`
`3 Modeling of CdS/CdTe thin film solar cells .......................................................................................11
`
`3.1 Modeling approach..........................................................................................................................11
`
`3.2 Purposes of modeling ......................................................................................................................11
`
`3.3 Modeling resources .........................................................................................................................12
`
`3.4 Definition of the problem................................................................................................................12
`
`3.5 Input parameters..............................................................................................................................13
`
`3.6 Modeling results..............................................................................................................................14
`
`3.6.1 Modeling results for the front contact and CdS layer..............................................................14
`
`3.6.2 Modeling results for the CdTe layer........................................................................................14
`
`3.7 Prognosis and future work...............................................................................................................15
`
`3.7.1 1D Modeling ...........................................................................................................................15
`
`3.7.2 2D Modeling ...........................................................................................................................16
`
`4
`
`Summary .............................................................................................................................................19
`
`4.1 First year accomplishments .............................................................................................................19
`
`4.2 Planned second year milestones ......................................................................................................20
`
`5 Major articles published during Phase I of the subcontract ................................................................20
`
`6
`
`References ...........................................................................................................................................20
`
`Atmospheric Pressure Chemical Vapor Deposition of
`CdTe for High Efficiency Thin Film PV Devices
`
`1999 Annual Report
`
`1
`
`

`

`1 Project objective
`
`ITN’s three year project Atmospheric Pressure Chemical Vapor Deposition (APCVD) of CdTe for High
`Efficiency Thin Film PV Devices has the overall objectives of improving thin film CdTe photovoltaic
`(PV) manufacturing technology and increasing CdTe PV device power conversion efficiency. Tasks
`required to accomplish the overall goals are grouped into 1) development of APCVD apparatus and
`procedures which enable controlled deposition of device-quality film over large area and 2) development
`of advanced measurement and analytical procedures which provide useful and effective device
`characterization.
`
`1.1 Approach
`
`CdTe deposition by APCVD employs the same reaction chemistry as has been used to deposit 16%
`efficient CdTe PV films, i.e., close spaced sublimation, but employs forced convection rather than
`diffusion as a mechanism of mass transport. Tasks of the APCVD program center on demonstration of
`APCVD of CdTe films, discovery of fundamental mass transport parameters, application of established
`engineering principles to the deposition of CdTe films and, verification of the reactor design principles
`which could be used to design high throughput, high yield manufacturing equipment. Additional tasks
`relate to improved device measurement and characterization procedures which can lead to a more
`fundamental understanding of CdTe PV device operation. Specifically, under the APCVD program,
`device analysis goes beyond conventional one-dimensional device characterization and analysis toward
`two dimension measurements and modeling.
`
`1.1.1 Deposition technology
`
`Although there are many demonstrated methods for producing high-efficiency CdTe solar cells, large-
`scale commercial production of thin-film CdTe PV modules has not yet been realized.1 An important
`contributor to the commercial production of thin-film CdTe will be development of advanced deposition
`reactors. APCVD represents a generation beyond close spaced sublimation (CSS) – the technology which
`has produced the highest efficiency CdTe PV cells to date.2,3 APCVD combines proven CSS reaction
`chemistry with state-of-the-art engineering principles to enable design of thin film deposition reactors for
`the manufacturing environment. APCVD's anticipated advantages include:
`(cid:118) Low equipment cost compared to vacuum processing because equipment will need neither the
`structural strength nor the pumping systems of a vacuum chamber.
`(cid:118) Large area uniformity is achieved through control of temperature and gas flow - both of which are
`subject to rigorous engineering design.
`(cid:118) Simplified process control and source replenishment because the source gas generation is physically
`separated from the deposition chamber.
`(cid:118) CdTe PV device fabrication process compatibility in that APCVD is presently used commercially to
`deposit transparent conducting oxide (TCO) films commonly used in CdTe solar cells. In fact, the
`processing sequence: deposit TCO, deposit CdS, deposit CdTe, dry CdCl2 heat treatment and
`metalorganic CVD of electrodes could be performed in a single continuous process.
`(cid:118) Low raw material costs as CdTe is used in its least expensive form - chunks.
`(cid:118) Simplified continuous processing because gas curtains replace load locks.
`
`Atmospheric Pressure Chemical Vapor Deposition of
`CdTe for High Efficiency Thin Film PV Devices
`
`1999 Annual Report
`
`2
`
`

`

`1.1.2 Device analysis
`
`Operation of thin film PV devices is normally analyzed in one dimension – distance perpendicular to the
`device surface. One dimensional (1D) modeling is justified in that thin film PV devices are basically
`comprised of a stack of thin films of various compositions and properties and through which light and
`electricity flow in a direction essentially perpendicular to the plane of the films. There is no question that
`1D modeling successfully describes the fundamentals of thin film PV device operation. Nonetheless,
`quantitative analysis of PV device operation, its dependence on device fabrication procedures, and factors
`affecting stability in the field have not been achieved. Furthermore, we know that individual films are not
`homogeneous, but rather are comprised of grains. Each grain is surrounded by grain boundaries that are
`oriented in all directions and which have different physical, electrical and optical properties than does the
`interior of the grain. Thus tasks of this program are directed toward techniques for quantifying the
`properties of grain boundaries and for quantifying effects grain boundaries may have on thin film PV
`device operation.
`
`An important distinction between the commonly used one dimensional (1D) models and a two
`dimensional (2D) model is that the 2D model allows for electric fields and carrier transport both parallel
`and perpendicular to the direction from which light is incident. These perpendicular components are
`brought about by differences in carrier type, carrier concentration, and carrier lifetime associated with
`grain boundaries. In this project efforts are being directed toward the experimental characterization of
`grain boundaries and the investigation of the effects of these characteristics on working devices. An
`important aspect of this approach is maintenance of a close connection between measurement, modeling
`and analysis.
`
`2 APCVD
`
`2.1 Background
`
`2.1.1 Reaction chemistry
`
`CdTe film deposition by CSS or APCVD is a three-step process that involves (i) generation of elemental
`vapors, (ii) vapor transport, and (iii) condensation and reaction to form CdTe. Congruent sublimation of
`CdTe occurs through the reaction
`
`(1)
`
`Te
`
`)(
`2 g
`
`.
`
`21
`
`CdTe
`
`(cid:11)
`
`heat
`
`(cid:106)
`
`)(
`gCd
`
`(cid:11)
`
`Vapor pressure over solid CdTe in a chemically inert environment depends only on temperature and is
`described by the Antoine equation
`
`.
`
`(2)
`
`log [Psat] = 6.823 – 10,000/T
`where T is temperature Kelvin and saturation pressure, Psat, is expressed in atmospheres. In both APCVD
`and CSS, Cd and Te2 vapors are transported to a substrate that is maintained at a somewhat lower
`temperature than the source. At the substrate temperature, the source gas is supersaturated causing Cd
`and Te2 vapors to react and form CdTe. The degree of supersaturation and the rate of material delivery to
`the surface determine deposition rate. Cd and Te2 can also condense to form their elemental condensed
`phases, but at all temperatures the elemental vapor pressures of Cd and Te2 over their respective
`condensed phases is much higher than either gas over CdTe. Figure 1 displays possible reaction
`mechanisms which may occur on the growing CdTe film surface. In practice, for substrates held above
`~500°C and deposition rates ~ 5 μm/min the deposited films are single phase CdTe.
`
`Atmospheric Pressure Chemical Vapor Deposition of
`CdTe for High Efficiency Thin Film PV Devices
`
`1999 Annual Report
`
`3
`
`

`

`Cd
`
`Te2
`
`Te(s)
`
`Cd(s)
`
`CdTe(d)
`
`Figure 1 Possible CdTe deposition mechanism - Eley-Rideal mechanism. On the left side Cd(g)
`atoms react with condensed Te to form CdTe before the Te can re-evaporate. The analogous
`process – with Te2(g) reacting with momentarily condensed Cd, is displayed on the right side.
`
`It is emphasized that these reactions depend only on temperature and on the concentrations of the source
`gases immediately above the superstrate. In particular, the reactions do not depend upon the pressure of
`inert gases such as N2, Ar or He. Thus the heterogeneous reaction chemistry, i.e., the set of reactions that
`take place on the substrate, is the same for CSS, elemental vapor deposition, or APCVD. The primary
`difference between CSS and APCVD is the mechanism of mass transport.
`
`2.1.2 Mass transport
`
`2.1.2.1 Diffusion
`Mass transport of the gas source species occurs through a combination of diffusion and convection. Most
`previous work, including CSS and elemental vapor deposition, has been performed in closed systems
`where diffusion is the lone transport mechanism. Mass flux due to diffusion is proportional to the product
`of concentration gradient and diffusion coefficient. In CSS, the concentration gradient is determined by
`the difference between the equilibrium partial pressures at the source and superstrate and by the physical
`separation between them.4 For typical CSS conditions (P = 10-50 torr, source-superstrate spacing of 3-10
`mm, and temperature differences of 10-80°C) deposition rates are on the order of 4 μm/min. Use of CSS
`at atmospheric pressure has been demonstrated, but deposition rates were in the 0.2 μm/min range.5
`
`2.1.2.2 Forced convection
`Mass transport rate can be increased and control improved over that achieved by diffusion alone by using
`forced convection. Convection processes are critical in APCVD for both generating elemental vapors and
`for transporting material to the substrate. Vaporization of source material occurs in a packed bed
`containing CdTe chunks as illustrated in Figure 2. Mass transport from a packed bed is an efficient and
`well-characterized method of producing source material-laden gases.6 Flow in packed beds is turbulent,
`creating high mass transport coefficients. Partial pressure of CdTe vapors, PCdTe, coming out of the source
`gas generator is given by
`
`PCdTe/Psat = (1-exp{- hm *Ap,t * (cid:69) /[vo * Ac,b ]})
`where Psat is the equilibrium partial pressure at the bed temperature (Eq. 2), Ap,t is the total surface area of
`the particles in the packed bed, Ac,b is the cross sectional area of the bed, hm is the mass transfer
`coefficient, vo is the “open velocity” of the gas (i.e., the velocity the gas would have if there were no
`particles in the bed), and (cid:69) is the void fraction of the bed. Calculations for the case of a 10 cm diameter
`bed packed with ~4 kg of 1 cm diameter particles, maintained at 1000 K (727 °C), and a gas flow rate of
`7.6 l/s indicate that the outlet gas will be >99% saturated and the pressure drop will be ~4.5 psi. Source
`
`Atmospheric Pressure Chemical Vapor Deposition of
`CdTe for High Efficiency Thin Film PV Devices
`
`1999 Annual Report
`
`4
`
`

`

`(cid:74)(cid:68)(cid:86)(cid:3)(cid:82)(cid:73)(cid:3)(cid:87)(cid:75)(cid:76)(cid:86)(cid:3)(cid:87)(cid:75)(cid:85)(cid:82)(cid:88)(cid:74)(cid:75)(cid:83)(cid:88)(cid:87)(cid:3)(cid:68)(cid:81)(cid:71)(cid:3)(cid:86)(cid:68)(cid:87)(cid:88)(cid:85)(cid:68)(cid:87)(cid:76)(cid:82)(cid:81)(cid:3)(cid:79)(cid:72)(cid:89)(cid:72)(cid:79)(cid:3)(cid:76)(cid:86)(cid:3)(cid:86)(cid:88)(cid:73)(cid:73)(cid:76)(cid:70)(cid:76)(cid:72)(cid:81)(cid:87)(cid:3)(cid:87)(cid:82)(cid:3)(cid:70)(cid:82)(cid:68)(cid:87)(cid:3)(cid:68)(cid:3)(cid:20)(cid:19)(cid:19)(cid:3)(cid:70)(cid:80)(cid:21)(cid:3)(cid:86)(cid:88)(cid:83)(cid:72)(cid:85)(cid:86)(cid:87)(cid:85)(cid:68)(cid:87)(cid:72)(cid:3)(cid:90)(cid:76)(cid:87)(cid:75)(cid:3)(cid:20)(cid:19)(cid:3)(cid:151)(cid:80)(cid:3)(cid:82)(cid:73)(cid:3)(cid:38)(cid:71)(cid:55)(cid:72)
`(cid:76)(cid:81)(cid:3)(cid:82)(cid:81)(cid:72)(cid:3)(cid:80)(cid:76)(cid:81)(cid:88)(cid:87)(cid:72)
`
`Source gas is convected to the deposition
`chamber via transport piping. Deposition onto
`piping walls is retarded by heating them to
`temperatures above the packed bed temperature.
`
`(1998)
`year
`2.2 First
`fabrication progress
`
`reactor
`
`During Phase I an APCVD reactor was
`designed, all components were constructed and
`assembly was nearly completed. In order to
`verify reactor performance a “dry run”
`(operation of reactor without CdTe) was
`performed. Although the reactor performance
`was close to that predicted, some deficiencies
`were discovered. Modifications to the original
`design are expected to be minor and the APCVD
`reactor program is expected to be back on
`schedule in the second year of the program.
`
`(cid:41)(cid:76)(cid:74)(cid:88)(cid:85)(cid:72)(cid:3) (cid:21)(cid:3) (cid:3) (cid:54)(cid:70)(cid:75)(cid:72)(cid:80)(cid:68)(cid:87)(cid:76)(cid:70)(cid:3) (cid:82)(cid:73)(cid:3) (cid:87)(cid:75)(cid:72)(cid:3) (cid:179)(cid:83)(cid:68)(cid:70)(cid:78)(cid:72)(cid:71)(cid:3) (cid:69)(cid:72)(cid:71)(cid:180)
`(cid:86)(cid:82)(cid:88)(cid:85)(cid:70)(cid:72)(cid:3)(cid:74)(cid:68)(cid:86)(cid:3)(cid:74)(cid:72)(cid:81)(cid:72)(cid:85)(cid:68)(cid:87)(cid:82)(cid:85)
`
`2.2.1 Adopted stagnant flow APCVD reactor concept
`
`A design team consisting of engineers and scientists from ITN and CSM considered various design
`options for the APCVD reactor. Although all design options involved atmospheric pressure reactors and
`mass transport by forced convection, the design team considered many possible configurations.
`Fundamental design issues to be resolved included issues such as: whether source gas flow should be top-
`down or bottom-up, whether source gas flow would be generally parallel or perpendicular to the substrate
`surface, whether flow should be turbulent or laminar, and whether gas injectors should be large or small
`or round or square. After consideration of these various issues, the design team selected the Stagnant
`Flow Reactor (SFR) design concept, an example of which is displayed in Figure 3.
`
`Source Gas Inlet
`
`Curtain Gas
`Plenum
`
`Nozzle
`Heater
`
`Nozzle
`
`Substrate
`
`Exhaust
`
`Figure 3 Conceptual stagnation flow reactor which displays the essential features of the APCVD
`reactor for deposition of thin film CdTe.
`
`Atmospheric Pressure Chemical Vapor Deposition of
`CdTe for High Efficiency Thin Film PV Devices
`
`1999 Annual Report
`
`5
`
`

`

`SFR design features include a nozzle with an area greater than the substrate out of which the source gas
`flows uniformly over the nozzle area (cm3/s*cm2). The nozzle face is parallel to the substrate surface and
`gas flow from the nozzle is laminar (non-turbulent). Source gas flow is perpendicular to the plane of the
`surface creating a layer of essentially stagnant gas above the substrate. Considerations which led the
`design team to adopt the SFR concept at this stage of the project included:
`(cid:118) SFR design enables uniform deposition rate over the substrate surface
`(cid:118) Stagnant Flow Reactors are used commercially in many CVD processes
`(cid:118) Planar geometry allows 1-D simulation of reactor performance
`(cid:118) SFR design enables determination of engineering parameters required for next-generation commercial
`scale APCVD reactor design
`
`2.2.2 Developed APCVD “stagnant flow” reactor design
`
`Having adopted the SFR design concept, the design team proceeded to design a reactor suitable for
`APCVD of CdTe. Components of the APCVD reactor are shown schematically in Figure 4. Deposition
`takes place in a cold wall reactor with a design specification for the source gas of 800ºC. Hot source gas
`flows laminarly downward and impinges perpendicularly onto the heated substrate at a nominal speed of
`50 cm/s. Eddy currents are eliminated and reactor walls are kept cool by a curtain gas which flows
`parallel to the source gas but outside of and on all four sides of the nozzle. The curtain gas also serves to
`prevent deposition onto reactor walls.
`
`Substrate temperature control is achieved through a specially designed substrate support which includes
`both a 500 W heater and water cooling. Inclusion of both heating and cooling capabilities enables active
`temperature control of the substrate within the temperature range expected during deposition.
`Specifically, heat can be supplied in order to keep the substrate at the desired temperature - expected to be
`in the range of 550ºC to 625ºC - but heat from the source gas can be removed under conditions when the
`source gas transfers excess heat to the substrate.
`
`Compressed nitrogen is used both as the carrier gas for the Cd and Te2 vapors and as the curtain gas.
`Before entering the CdTe bubbler (i.e., the packed bed containing chunks of CdTe), carrier gas is
`preheated to approximately the temperature of the CdTe chunks. Carrier gas passes upward through the
`CdTe chunks in the packed bed structure (see Figure 2) where Cd and Te2 gasses are picked up by the
`carrier gas - creating the source gas for CdTe deposition.
`
`Partial pressures of Cd and Te2 gasses leaving the bubbler are expected to be close to the equilibrium
`partial pressure at the bubbler temperature. Source gas can be characterized by its “saturation
`temperature”, Ts, which is determined by solving equation (2) for temperature given the actual Cd and Te2
`gas partial pressures. Thus Ts (cid:98) Tbubbler. After passing through the packed bed the source gas is
`transported to the reactor through hot piping. Transport tubing walls are kept at a temperature greater
`than Ts in order to prevent deposition onto tubing walls.
`As a safety precaution the APCVD system is designed to operate slightly below room pressure by a few
`inches of water (pressure). Thus, after passing through the reactor, exhaust gas is drawn out the
`deposition zone by a blower (pump) with sufficient capacity to maintain negative partial pressure
`everywhere in the APCVD system from the exit of the bubbler to the exhaust blower. Before reaching
`the blower, however, gas exiting the reactor chamber first passes through a heat exchanger and a high
`efficiency particle accumulator (HEPA) filter which serve to cool and clean the exhaust gas, respectively.
`Note that any residual Cd and Te2 gasses remaining in the exhaust gas will deposit onto the cold, i.e.,
`room temperature, walls of the heat exchanger or will be captured as particulates by the HEPA filter. As
`an additional safety precaution, the entire reactor system will be contained within an enclosed plexiglass
`
`Atmospheric Pressure Chemical Vapor Deposition of
`CdTe for High Efficiency Thin Film PV Devices
`
`1999 Annual Report
`
`6
`
`

`

`chamber (not shown) which will also be kept at negative pressure with respect to room ambient and will
`be vented through a second HEPA filter and out of an exhaust duct.
`
`Figure 4 Schematic representation of the APCVD reactor system.
`
`In order to monitor reactor performance, pressure, temperature and flow will be continuously measured at
`various locations on both the inlet and exhaust sides of the reactor. Temperature, pressure an flow data
`will be recorded automatically by an HP Data Acquisition Unit and stored on a computer.
`
`2.2.3 Performed numerical simulations of reactor performance
`
`Researchers modeled reactor temperature and flow using computer programs which numerically solved
`the Navier-Stokes equations using boundary conditions appropriate to the APCVD reactor. An example,
`shown in Figure 5, shows the temperature distribution and flow streamlines for a specified set of process
`parameters. Note in particular the uniformity of temperature across the substrate and the absence of eddy
`currents within the deposition zone.
`
`The hot, flowing, source gas transfers heat, momentum (pressure), and material to the substrate. Each
`form of transport has its characteristic boundary layer - i.e., thermal, momentum and mass - which are
`related to one another, but are not identical. For example in the case of film deposition rate, although the
`majority of mass transport of Cd and Te2 from the bubbler to the substrate is by forced convection, Cd
`and Te2 must make the final 1 - 2 mm of their trip to the substrate by diffusion through the stagnation
`layer. Figure 6 shows calculated mass fraction and temperature profiles above the substrate for certain
`assumed process conditions and surface reaction chemistry. Surface reaction chemistry and growth can
`be characterized by the probability that a gas molecule striking the substrate will react to form
`
`Atmospheric Pressure Chemical Vapor Deposition of
`CdTe for High Efficiency Thin Film PV Devices
`
`1999 Annual Report
`
`7
`
`

`

`Figure 5 Simulated temperature distribution
`and gas flow streamlines in the APCVD SFR
`design. The substrate is directly below the
`inlet
`
`Figure 6 Concentration and temperature
`dependence on distance above substrate for
`specified assumptions.
`
`Figure 7 CdTe film growth rate as a function
`of reaction probability for various nozzle gas
`velocities.
`
`Figure 8 Concentration contours showing
`carrier gas concentrations and material
`utilization efficiency for two values of nozzle
`gas velocity.
`
`CdTe. Figure 7 shows results of a calculation of deposition rate as a function of boundary layer thickness
`and the rate of chemical reaction at the surface. Modeling has been carried further to simulate growth rate
`and material utilization efficiency in the present APCVD reactor design. It can be seen in Figure 8 that
`
`Atmospheric Pressure Chemical Vapor Deposition of
`CdTe for High Efficiency Thin Film PV Devices
`
`1999 Annual Report
`
`8
`
`

`

`the simulated film deposition rate and film uniformity improve with source gas velocity, while material
`utilization efficiency is higher for lower gas flow. Simulations will be compared with experimental
`deposition runs in order to better establish mass and thermal transport coefficients and thereby to improve
`our ability to model reactor performance.
`
`2.2.4 Fabricated APCVD reactor
`
`ITN engineers, with support from the CSM design team, prepared detailed engineering drawings,
`specified required components, and ordered required parts. CSM and ITN personnel assembled the
`
`Gas Preheat
`
`CdTe Bubbler
`
`Hot Transfer
`Piping
`
`Exhaust
`Filter
`
`Reactor
`Chamber
`
`Figure 9 APCVD reactor system as it appeared at the end of the first year of the program.
`
`APVCD components and installed the reactor in the CdTe deposition lab in the Physics Department
`building at CSM. Figure 9 displays the APCVD reactor as it appeared at the end of 1999.
`
`2.2.5 Performed “dry runs” of reactor to evaluate and document reactor performance
`
`By the end of this first year of the program researchers had begun trials of reactor performance to
`determine whether design specifications had been achieved. Evaluations runs were performed “dry”, i.e.,
`without CdTe in the bubbler, in order to avoid contamination of reactor components in the event that the
`reactor would require redesign or re-assembly. During the dry runs gas flow and pressure were monitored
`as well as temperature at all key points. Figure 10 shows representative temperature acquired during a
`dry run.
`
`Gas supply, water cooling, temperature control, exhaust gas blower, process monitors and data acquisition
`were all evaluated and found to work generally as expected.
`
`Although the dry run was generally successful in that the required gas flows and temperatures were
`readily achieved, the dry run also identified certain deficiencies in the reactor design. Specifically, it was
`discovered that there were certain “cold spots” in the hot piping system and that gas exiting the nozzle
`was only about 650ºC. In addition, not all sensors performed as required. Although the remedy for these
`deficiencies is straightforward, researchers decided not to operate the reactor until these deficiencies have
`been successfully addressed.
`
`Atmospheric Pressure Chemical Vapor Deposition of
`CdTe for High Efficiency Thin Film PV Devices
`
`1999 Annual Report
`
`9
`
`

`

`Temperature during warm up
`Dry Run -January 1999
`
`500
`
`1000
`
`1500
`
`2000
`
`2500
`
`3000
`
`3500
`
`4000
`
`4500
`
`Time (sec)
`
`Temperatures during Cool Down
`Dry Run - January 1999
`
`Hot Piping 2
`Hot Piping 3
`Hot Piping 4
`Bubbler 2
`Bubbler 3-Exhaust
`Bubbler 4-Intake
`Nozzle
`Bypass 2
`Substrate-Hot
`Substrate-Cold 1
`Substrate-Cold 2
`
`Hot Piping 2
`Hot Piping 3
`Hot Piping 4
`Bubbler 2
`Bubbler 3-Exhaust
`Bubbler 4-Intake
`Nozzle
`Bypass 2
`Substrate-Hot
`Substrate-Cold 1
`Substrate-Cold 2
`
`1000
`
`900
`
`800
`
`700
`
`600
`
`500
`
`400
`
`300
`
`200
`
`100
`
`0
`
`0
`
`Temp (C)
`
`800
`
`700
`
`600
`
`500
`
`400
`
`300
`
`200
`
`100
`
`Temp (C)
`
`A)
`
`0
`6500
`
`B)
`
`7000
`
`7500
`
`8000
`
`8500
`
`9000
`
`9500
`
`10000
`
`Time (sec)
`
`Figure 10 Representative graphs showing temperature data recorded during A) heat-up and B)
`cool-down portions of the APCVD reactor dry runs.
`
`2.3 APCVD reactor plans
`
`ITN first task of the second year of the APCVD program will be to correct the deficiencies of the existing
`reactor. Additional heating zones will be installed and all sensors will be made to operate correctly.
`Results of these modifications will be evaluated during a second set of dry runs.
`
`As soon as reactor performance has been verified, CdTe film depositions and characterization studies will
`begin. Initial runs will evaluate the reactor performance space through a series of designed experiments
`which will establish the dependencies of film growth and film parameters on deposition procedure and
`process parameters. Actual reactor performance will be compared with simulated performance to
`establish more accurate values for mass and thermal transport parameters. Growth rate, thickness
`uniformity, and CdTe film structure will be monitored to determine the relationships between process
`parameters and film characteristics. Additional film characterization will include optical transmission,
`SEM micrographs, in-plane AC impedance, Seebeck coefficient, and radio frequency photoconductivity
`measurements as required.
`
`Device fabrication will be central to evaluation of film properties. CSM researchers will use their
`facilities to produce CdS layers, and back electrodes and to perform oth