I, Rachel J. Watters, am a librarian, and the Director of Wisconsin
`
`TechSearch (“WTS”), located at 215 North Randall Avenue, Madison Wisconsin,
`
`53706. WTS is an interlibrary loan department at the University of Wisconsin-
`
`Madison.
`
`I have worked as a librarian at the University of Wisconsin library
`
`system since 1998.
`
`I have been employed at WTS since 2002, first as a librarian
`
`and, beginning in 2011, as the Director. Through the course of my employment, I
`
`have become well informed about the operations of the University of Wisconsin
`
`library system, which follows standard library practices.
`
`This Declaration relates to the dates of receipt and availability of the
`
`following:
`
`Wright, DR et al. "A closed-loop temperature control
`system for a low-temperature etch chuck." Advanced
`Techniques for Integrated Circuit Processing 11, SPIE
`Proceedings, V. 1803, 1992, p. 321-329
`
`Standard operating procedures (or periodicals at the Universigg of
`
`Wisconsin—Madison. When an individual issue of a periodical was received by the
`
`Library, it would be checked in, stamped with the date of receipt, added to library
`
`holdings records, and made available to readers as soon after its arrival as possible.
`
`The procedure normally took a few days or at most a few weeks.
`
`Exhibit A to this Declaration is a true and accurate copy of Advanced
`
`Techniques for Integrated Circuit Processing 11, from the University of Wisconsin-
`
`Madison Library collection. A date stamp on the cover of this copy indicates that
`
`1
`
`Page 1 of 14
`
`Samsung Exhibit 1019
`
`Page 1 of 14
`
`Samsung Exhibit 1019
`
`

`

`this volume, published in 1993, was received by the Kurt F. Wendt Library,
`
`College of Engineering, University of Wisconsin, on June 23, 1993.
`
`Based on the information in Exhibit A, it is clear that the volume was
`
`received by the library on or before June 23, 1993, catalogued and available to
`
`library patrons within a few days or at most a few weeks after June 23, 1993.
`
`I declare that all statements made herein of my own knowledge are true and
`
`that all statements made on information and belief are believed to be true; and
`
`further that these statements were made with the knowledge that willful false
`
`statements and the like so made are punishable by fine or imprisonment, or both,
`
`under Section 1001 of Title 18 of the United States Code.
`
`Date: June 8, 2016
`
`Wisconsin TechSearch
`
`Wendt Commons Library
`215 North Randall Avenue
`
`Madison, Wisconsin 53706
`
`9
`Rachel J.
`
`Director
`
`atters
`
`Page 2 of 14
`
`Page 2 of 14
`
`

`

`
`
`Page 3 of 14
`
`

`

`PROCEEDINGS
`SPIE—The International Society for Optical Engineering
`
`Advanced Techniques for
`lntegra ted Circuit Processing ll
`
`James Bondur
`Gary Castleman
`Lloyd R. Harriott
`Terry R. Turner
`Chairs/Editors
`
`21—23 September 1992
`San Jose, California
`
`Sponsored and Published by
`SPIE—The International Society for Optical Engineering
`
`
`
`
`
`PROCEEDINGS
`S E Fl
`l E 5
`
`Volume 1803
`
`SPIE (The Society of Photo-Optical Instrumentation Engineers) is a nonprofit society dedicated to the
`advancement of opticaland optoelectronic applied science and technology.
`my a warm? trams“;
`cottage or: ENGII‘EE
`
`umverzsaw or WISCONSEN
`Amwast W3 537%
`
`Page 4 of 14
`____——_—__—_._.———-——————u—"
`
`Page 4 of 14
`
`

`

`
`
`
`
`The papers appearing in this book comprise the proceedings of the meeting mentioned on
`the cover and title page. They reflect the authors’ opinions and are published as presented
`and without change,
`in the interests of timely dissemination. Their inclusion in this
`publication does not necessarily constitute endorsement by the editors or by SPIE.
`
`Please use the following format to cite material from this book:
`Author(s), "Title of paper," in Advanced Techniques for Integrated Circuit Processing II,
`James Bondur, Gary Castleman, Lloyd R. Harriott, Terry R. Turner, Editors, Proc. SPIE 1803,
`page numbers (1993).
`
`Library of Congress Catalog Card No. 92-62653
`ISBN 0-8194-1001-2
`
`Published by
`SPIE—The International Society for Optical Engineering
`P.O. Box 10, Bellingham, Washington 98227-0010 USA
`Telephone 206/676-3290 (Pacific Time) 0 Fax 206/647-1445
`
`Copyright©1993, The Society of Photo-Optical Instrumentation Engineers.
`
`Copying of material in this book for internal or personal use, or for the internal or personal
`use of specific clients, beyond the fair use provisions granted by the U.S. Copyright Law is
`authorized by SPIE subject to paymentofcopying fees. The Transactional Reporting Service
`base fee for this volume is $4.00 per article (or portion thereof), which should be paid
`directly to the Copyright Clearance Center (CCC), 27 Congress Street, Salem, MA 01970.
`Other copying for republication, resale, advertising or promotion, or any form of systematic
`or multiple reproduction of any material in this book is prohibited exceptwith permission
`in writing from the publisher. The CCC fee code is 0—8194-1001-2/93/$4.00.
`
`Printed in the United States of America.
`
`Page 5 of 14
`
`
`Page 5 of 14
`
`

`

`A Closed-Loop Temperature Control System
`for a Low-Temperature Etch Chuck
`
`D.R. Wright, W.D. Clark, D.C. Hartman, U.C. SridharanT, SEMATECH, Austin, TX;
`M. Kent, R. Kerns, GS/DRYTEK, Wilmington, MA.
`
`TNow at HEWLETT-PACKARD, Palo Alto, CA.
`
`ABSTRACT
`
`A closed—loop temperature control system has been developed for use in a low-
`temperature (-135°C) plasma etch system.
`The system employs an optical
`fluorescence probe on the chuck (a second probe monitors the wafer temperature as
`well) to provide feedback to the heating element on the input line of the chuck closed-
`Ioop coolant fluid. A simple proportional-integral—derivative (PID) controller with a
`learn mode controls the rate of current pulses applied to the heater.
`Innovations
`include the direct measurement of chuck temperature for the control signal, and the
`coupling of large cooling and heating capacities in close proximity to the chuck along
`with a fast fluid flow to guarantee quick response.
`
`It provides
`The system has been tested in prolonged etch runs of many wafers.
`reliable, tight temperature control (30 as low as O.6°C). This level of control is
`significantly tighter than could be achieved by merely monitoring chiller bath
`temperatures.
`
`1.
`
`INTRODUCTION
`
`As semiconductor features become smaller and process requirements become
`stricter, advanced plasma etch systems require tighter control over their process
`factors. A recent project at SEMATECH worked to develop an etch system that
`would provide tight control of wafer temperature for low-temperature etch
`applications.
`
`This paper discusses the hardware for both the sensors and control systems, as
`well as results of the temperature stability, over various extended runs of hundreds
`of wafers.
`
`2. SUPPORTING HARDWARE
`
`2.1
`
`Fiber optic temperature probe
`
`The fluoroptic thermometric technique developed by Luxtron uses the
`photoluminescent response of a magnesium fluorogermanate phosphor to blue light
`pulses transmitted down a fiber optic cable probe. The rate of decay of the
`fluorescence is measured and correlated to temperature. Two types of fluoroptic
`
`0—8794-7001-2/93/$4.00
`Page 6 of 14
`
`5PIE Vol. 1803 (I992) / 327
`
`Page 6 of 14
`
`

`

`probes are available. The first type ("remote" or non-contacting) uses a phosphor
`dye painted on the sample and receives the data into the fiber from a distance.
`The second type ("contact") uses the phosphor in an encapsulation at the end of
`the fiber. Surrounded by a protective coating, this probe is used to make physical
`and thermal contact with the sample. For the application investigated during the
`project, either type could have been used for the electrode probe. The etch
`system used in the project had two chambers, each with separate "contact"
`probes for monitoring electrode temperature and wafer temperature.
`
`The electrode temperature was monitored to determine the stability of the cooling
`system. The wafer temperature was used to determine the interaction of the etch
`process and the cooling system.
`
`2.2
`
`Chiller systems
`
`Two different cooling systems were used in the tool. For the -70°C chamber, a
`standard bath-type chiller was chosen. This style of chiller controls the
`temperature of a bath through which a heat transfer fluid circulates on its way to
`the electrode (some 5 to 10 meters away).
`
`For the 435°C chamber, a system from Polycold was chosen. This cryo—cooler
`was unique in two ways. First, the refrigerant (a mixture of Freons and argon) was
`circulated through the electrode. This afforded a higher heat removal efficiency
`than circulating a heat transfer fluid between a refrigerated bath and the electrode.
`Second, there was no built—in feedback system for temperature control. These
`systems are normally run at "maximum output" for high—vacuum cold-trap
`applications, with the temperature controlled by the composition of the refrigerant
`mixture. This made it necessary to add a separate system for controlling the
`
`electrode temperature.
`
`3. HEATER/CONTROLLER DESIGN
`
`The throttling approach to temperature control consisted of a valved coolant path
`that bypassed the chuck, allowing some of the cooling fluid to be diverted away
`from the chuck in order to lower the effective cooling power of the Polycold
`system. This approach failed for three main reasons:
`
`0
`
`0
`
`0
`
`Large valves that work reliably at -150°C are prohibitively expensive.
`
`The chuck cooling capacity was lowered (since most of the coolant would
`be bypassed).
`
`The two-phase nature of the refrigerant meant different mixtures of
`constituents flowed in the main and bypass lines, also lowering cooling
`
`efficiency.
`
`322 mm; Vol. 1803 (1992)
`Page 7 of 14
`
`Page 7 of 14
`
`

`

`The method chosen to provide the Polycold-chilled electrode with temperature
`control in the Drytek pre-productlon prototype low-temperature etch system was
`by heating the cryogenic fluid in the supply line and controlling the heater power to
`maintain the desired electrode temperature. The Polycold cooling power that must
`be overcome by the heater depends on the fluid delivery temperature in a nonlinear
`manner.
`It was verified that the amount of heating power required to be
`transferred to the Polycold fluid would be within the capability of the unit and
`would not pose any difficulty for its normal operation.
`
`The heater must be insulated to prevent water-ice condensation on cryogenic
`surfaces. Enclosing the entire device in a vacuum envelope would be the preferred
`method because it also provides mechanical protection for the electrical leads to
`the heaters. The use of thermal insulation would be acceptable provided it is
`capable of cryogenic operation and is compatible with the equipment installation.
`
`In the first heater design, a brass block containing cartridge heaters was brazed to
`a straight stainless steel tube which carried the refrigerant inside a cylindrical
`vacuum enclosure.
`
`The following problems were encountered with the first design:
`
`0
`
`0
`
`0
`
`The stainless steel tube had a very low thermal conductivity. This low
`conductivity was compounded by the presence of the brazed connection to
`the heater block.
`
`The Polycold fluid moves at a velocity of several meters per second, so that
`the contact time inside the 1—inch heater was only 2 to 10 msec.
`
`"Channelling flow" occurred inside the stainless steel tube because the
`Polycold uses two-phase flow, consisting of liquid droplets entrained in a
`carrier gas. As illustrated in Figure 1, expansion of the gas as the liquid
`droplets vaporized caused a laminar gas barrier to form that impeded heat
`transfer, prevented succeeding droplets from contacting the wall,
`'
`"channelled" them into the center of the fluid stream, and prevented
`
`turbulent mixing of the two-phase fluid stream.
`
`These effects combined to give the first heater design very poor performance. No
`more than 6 to 8 degrees Centigrade (C°) of temperature rise could be obtained at
`the electrode, even with the brass block at 300°C.
`
`The heater system was redesigned to address the limitations discovered in the first
`heater design. The control principle of the second design is illustrated in Figure 2.
`In this case, the process temperature to be controlled was the same as the
`feedback signal to the heater controller. The electrode temperature was measured
`by the Luxtron probe. The analog output from the Luxtron controller, which was
`proportional to the electrode temperature, was used as feedback to a three-term
`
`Page 8 of 14
`’
`fi¥_______—_—__——_—————
`
`SPIE Vol. 1803 (1992)/323
`
`Page 8 of 14
`
`

`

`Proportional-lntegraI-Derivative (PID) process controller with auto-tune capability
`that provided time-proportioned power to the heater via a solid state relay. Direct
`control of the electrode temperature could be maintained with optimal transient
`response over the widest possible range of operating conditions.
`
`The heater design in Figure 3 was chosen to provide direct contact of the Polycold
`fluid with a heater material (brass) of high thermal conductivity, and to provide a
`flow path that produced turbulence in the Polycold fluid (avoiding "channelling
`flow"). Calculations showed that 600 W of heater power would be required to
`raise the Polycold fluid temperature from -130°C to -80°C. Space was provided
`for up to 2400 W of heater capacity to ensure that the time-proportioning
`controller could be operated within its control band. The number of heaters
`installed or connected could be varied, providing a simple way to adjust the
`controlled range of heater power. The heater was mounted at atmospheric
`pressure with self—adhesive thermal insulating tape applied to prevent
`condensation. A thermocouple was provided to monitor heater temperature.
`
`The following results were obtained with this design:
`
`0
`
`0
`
`0
`
`0
`
`0
`
`A maximum electrode temperature rise of over 60°C was observed.
`
`The system provided rapid transient response and stable operation over the
`range from -130°C to -80°C.
`
`The heater thermocouple indicated that the heater was maintained at
`cryogenic temperature very near the electrode temperature. This fact
`confirms the excellent heat transfer characteristics of this design.
`
`The system is capable of controlling to :l:1°C.
`
`The system worked very reliably throughout its installation at SEMATECH.
`
`4. OPERATIONAL TESTS
`
`4.1
`
`Hardware acceptance test
`
`As part of the hardware acceptance test prior to the shipment of the tool to
`SEMATECH, a test using a total of 550 RF (plasma) cycles was performed to
`monitor the thermal stability of the cold electrodes. Each wafer was passed from
`the cassette, via the flatfinder, to chamber 2 (oxide etch), which was held at -
`30°C by the bath—type chiller circulating methanol.
`In chamber 2, the wafer
`received five RF cycles of 20-second duration, separated by 20 seconds. The
`recipe used was a standard oxide etch CZF6 (hexafluoroethane or Freon-116)
`chemistry with 600 W of power all to the bottom electrode. After processing in
`chamber 2, each wafer then passed into chamber 3 (poly etch), which was
`maintained at -105°C by the Polycold with the heater and controller discussed
`
`3 zEageEQchfi «MS (1992)
`
`Page 9 of 14
`
`

`

`in chamber 3 each wafer received six RF cycles of 20-second duration,
`previously.
`separated by 20 seconds. The recipe was the SFe/argon poly etch with 400 W of
`power, almost all to the top electrode. Fifty wafers (25 in each of two cassettes)
`were processed for a grand total of 550 plasma cycles.
`
`Using the parallel output (R5232) port of the Luxtron controller, a personal
`computer recorded the temperatures of both electrodes and both wafers as well as
`flags indicating whether the plasma was on in each chamber. These data were
`captured to disk and later analyzed using a spreadsheet program.
`
`Figure 4 shows a plot of the continuously sampled temperature of both wafer and
`electrode for six plasma cycles (one wafer) in chamber 3 (poly). Figure 5 shows a
`plot of the continuously sampled temperature of both wafer and electrode for five
`plasma cycles (one wafer) in chamber 2 (oxide). At this scale the exponential
`behaviors of both the heating and cooling of the wafer are apparent. Figure 6 then
`shows a more long-term timeline, showing stability over several hours.
`
`Table A summarizes all of the electrode temperature data taken during the plasma
`hardware acceptance test.
`in the poly chamber, there was a standard deviation
`(1a) of 0.21 °C for the electrode while in the oxide chamber there was a standard
`
`deviation (1a) of O.78°C for the electrode. The electrode data were taken from
`
`the entire test (with or without a wafer present).
`
`Table A: Summary of Hardware Acceptance Test Results
`— Pew Chamber
`Oxide Chamber
`Number of Data Points “ 1648
`-1os-83°c
`-ao-7e°c
`
`
`
`
`
`
`
`
`
`
`
`
`
`Standard Deviation
`
`0.21°C
`
`O.78°C
`
`'Both
`The results show a highly repeatable temperature control in both chambers.
`the bath-type chiller, using its internal feedback control, and the Polycold chiller,
`with the external heater controller, provided adequate cooling and repeatable
`temperatures over the entire test.
`
`4.2 Marathon
`
`Following the hardware acceptance, the low-temperature etch tool was installed at
`SEMATECH. After the optimization of the etch process was completed, an
`extended test was performed on the Polycold/electrode system. Over several days
`of 24 hour-per—day operation, more than 1200 wafers were processed through the
`tool. As before, computer files were maintained of the wafer and electrode
`
`temperatures at roughly five-second intervals. At the completion of the marathon,
`30,165 measurements of the electrode temperature had been recorded. The
`
`Page 10 of 14
`
`SPIE Vol. 7803 (7992) / 325
`
`Page 10 of 14
`
`

`

`process that was used in the marathon was different from the one used in the
`hardware acceptance test. The etch process was performed in two parts, with an
`initiation step and a main etch step. Each wafer received approximately one
`minute of plasma exposure time. The electrode temperature was set to -120°C.
`
`The results are shown in Table B. The control of the chuck temperature was not
`quite as tight as that of the hardware acceptance test. Some reasons include the
`lower base temperature, and a different etch process (with slightly higher heat load
`into the wafer while the plasma was on).
`
`Table B: Summary of Marathon Results
`
`_ Polvsilicon chamber
`Numbér of Data Poms “
`-1zo-7°c
`
`0.52°C
`
`Standard Deviation
`
`5. CONCLUSIONS
`
`A closed-loop temperature control system has been developed for use in a low-
`temperature (-135°C) plasma etch system. The system employs an optical
`fluorescence probe on the chuck to provide feedback to the heating element on the
`input line of the chuck closed-loop coolant fluid. A simple proportional-integral-
`derivative (PID) controller with 3 learn mode controls the rate of current pulses
`
`applied to the heater.
`
`Unlike the indirect control signal of the (remote) bath temperature, the system uses
`the direct measurement of chuck temperature for the control signal. The use of
`large cooling and heating capacities in close proximity to the chuck along with a
`fast fluid flow guarantee quick response. Careful design insures turbulence in the
`flow to guarantee good heat transport between the walls and the fluid.
`
`It provides
`The system has been tested in prolonged etch runs of many wafers.
`reliable, tight temperature control (30 as low as O.6°C). This level of control is
`
`significantly tighter than could be achieved by merely monitoring chiller bath
`temperatures.
`
`6. ACKNOWLEDGEMENTS
`
`The authors would like to thank Paul Grosshart for his excellent data transfer
`
`program, and Masoud Mayer for his help on the heater design and fabrication.
`This work was performed as part of a joint development project between Drytek
`and SEMATECH, a consortium of twelve semiconductor manufacturers and the
`
`federal government (DARPA).
`
`3i5’elgé’fit’ot8iiil’99”
`
`Page 11 of 14
`
`

`

`WALL UF STAINLESS STEEL TUBE
`
`~ DRUPLET IN CONTACT
`WITH VALL
`
`
`FDRMATIUN DF
`VAPDR BARRIER
`AT TUBE SURFACE
`
`
`
`
`
`
`@%
`
`
`GAS EXPANDING ARUUND DRDPLET
`
`CHANELLING FLDV
`
`VAPDRIZING
`
`FLUID DRDPLETS
`ENTRAINED IN
`CARRIER GAS
`
`*—
`DIRECTIUN CIF
`
`FLUV
`
`Figure 1
`Two-Phase Refrigerant Flow In the Polycold Line Heater
`
`\
`I
`UPTIC CABLE
`LUXTRDN FIBER “\erl—u
`
`
`LUXTRUN
`
`SEMAIECH
`LUV TEMPERATURE
`ETCH TDDL
`
` CHAMBER 3
`
`(ZHUCK
`
`.’T‘\
`
`‘¥ ANALDG DUTPUT SIGNAL
`PRDPDRTIDNAL TD CHUCK
`TEMPERATURE
`
`
`
`CUNTRDL SIGNA_
`FUR SDLID STATE RELAY
`
`Figure 2
`Schematic of Heater Controller (Second Design)
`
`Page 12 0f14
`
`SPIE Vol. 1803 (1992)/327
`
`\k ,&
`
`
` ._
`
`
`
`HEATER
`‘—
`
`SUPPLY
`
`
`
`RETURN_.
`
`
`PULYCULD
`
`|
`* PVI—F— 110 VAC
`
`\" HEATI‘R PEIVLk
`
`
`
`/‘\
` \
`.w
`PID—
`I
`/
`CONTROLLER wT—r—A
`
`
`
`Slll In-ETATL
`RELAY
`
`Page 12 of 14
`
`

`

`ECCENTRIC "LET/WRIT PUTS TD
`CAUSE TURELANCE AND NIGER til—USU}!
`RATE VITH n: UNDER INTEFJm VALLS
`
`
`
`8 EATER-S CAPACITY
`0-2400 WATTS M
`
`Figure 3
`Heater Diagram (Second Design)
`
`
`20:1 :35
`20:15:22
`20:16:05
`20:16:“
`20:17:31
`
`—+——|
`
`20:16:11
`
`20:15:50
`
`~10!
`
`405
`
`403 ° 40!.
`
`QIro
`
`I '
`
`c 406
`
`-107
`
`406
`
`409
`
`-110
`
`. 0°.o°o..00¢oo o
`O
`o ’00..
`and!”
`
`o
`000 o 00
`°
`0 on... .
`
`’ 0’9
`0
`‘00.
`
`0
`
`11— (Mu-:II)
`
`Figure 4
`Poly Etch Chuck and Wafer Temperature for 6 RF Cycles
`
`wagé'igow “992)
`
`Page 13 of 14
`
`

`

`
`
`Figure 6
`Low-Temperature Electrode and Wafer Temperatures Duri nnnnnnnnnnnun
`
`Page 14 of 14
`
`