Scientific Wet Process Technology for Innovative LSI/FPD Manufacturing

383 Pages·2005·41.35 MB·English

by  Tadahiro Ohmi

Scientific
Wet Process
Technology
for Innovative
LSI/FPD
Manufacturing
© 2006 by Taylor & Francis Group, LLC
Scientific
Wet Process
Technology
for Innovative
LSI/FPD
Manufacturing
edited by
Tadahiro Ohmi
Boca Raton London New York
A CRC title, part of the Taylor & Francis imprint, a member of the
Taylor & Francis Group, the academic division of T&F Informa plc.
© 2006 by Taylor & Francis Group, LLC
Published in 2006 by
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2006 by Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group
No claim to original U.S. Government works
Printed in the United States of America on acid-free paper
10 9 8 7 6 5 4 3 2 1
International Standard Book Number-10: 0-8493-3543-4 (Hardcover)
International Standard Book Number-13: 978-0-8493-3543-3 (Hardcover)
Library of Congress Card Number 2005024414
This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with
permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish
reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials
or for the consequences of their use.
No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or
other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information
storage or retrieval system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.com
(http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA
01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For
organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
Scientific wet process technology for innovative LSI/FPD manufacturing / editor, Tadahiro Ohmi.
p. cm.
Includes bibliographical references and index.
ISBN 0-8493-3543-4 (978-0-8493-3543-3 : alk. paper)
1. Semiconductors--Design and construction. 2. Semiconductors--Cleaning. 3. Integrated circuits--
Design and construction. I. Ohmi, Tadahiro, 1939-
TK7871.85.S3965 2006
621.3815'2--dc22 2005024414
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
Taylor & Francis Group and the CRC Press Web site at
is the Academic Division of T&F Informa plc. http://www.crcpress.com
© 2006 by Taylor & Francis Group, LLC
Preface
In 1965, Gordon E. Moore foretold that the number of transistors in integrated circuits would
increase by a factor of 4 every 3 years. Over the past few decades, semiconductor technologies
have steadily developed and have advanced the integration density of LSI devices in accordance
with Moore’s law. This revolutionary achievement was made possible by improving device
miniaturization technologies, such as photolithography technologies, various plasma processes,
and so on. However, the current molecule reaction-based semiconductor manufacturing technol-
ogiesarenowfacinggreatdifficultyinachievingfurtherdeviceminiaturization.Thepresentstand-
stillinsemiconductortechnologiesresultingfromthishascausedseverestagnationinindustriesall
over the world.
Microprocessors brought to the market in 1971 triggered a drastic change in industrial and
social structure by bringing the so-called microelectronics revolution to all industries and social
systems. Semiconductor technologies have supported the continuous progress of all industries in
subsequent decades. The continuous development of semiconductor technologies must not be
suspendedif the healthyadvancement of the world economyis tobemaintained.
Miniaturization inthe critical dimensions of integrated circuitsisaccompanied by a decrease
in thickness of the gate insulator films of MOS transistors. The biggest reason for the standstill
in current semiconductor technologies is very large leakage current, and a significant increase in
standbyelectricpowerconsumptionsofLSIdevicesuptoafewtenthsofwatts.Largeleakagecur-
rentshavethefollowingtwoconstituents:(1)leakagecurrentthroughtheverythingateinsulator
film (approaching 1nm thickness) and (2) drain leakage current. More than 1(cid:1)103A/cm2 of
leakage current is generated when a voltage of 1V is applied to a current thermal-oxidation-
based gate insulator film having a thickness of about 1nm. Therefore, the current thermal-
oxidation-based gate insulator films can no longer serve as insulator films when their thickness
decreasesto around 1nm.
Meanwhile, in an attempt to improve the mobility of electrons and holes to enhance the
speedperformanceofLSIdevices,thesemiconductorindustryhasintroducedSiGetoMOStran-
sistors,wherethegermaniumconcentrationinSiGeisgraduallyincreasedtoimprovethemobility
of the electrons and holes. This attempt is theoretically incorrect. It is very well known that
the bandgaps of Si and Ge are 1.12 and 0.69eV, respectively, resulting in a huge difference in
the leakage currents in the reverse direction of the pn diodes of Si and Ge by a factor of
2.8(cid:1)104 at room temperature. Thus, the increase in germanium concentration in the SiGe of
MOS transistorsis inevitably accompanied by a drastic increase in the drain leakage current of2
to3 orders ofmagnitude.
The author has developed radical-reaction-based semiconductor manufacturing using
microwave-excited very low-electron-temperature high-density plasma equipment instead of
the current molecule reaction-based semiconductor manufacturing in order to overcome the
difficulties thatlimit the progress of semiconductor technologies.
Current plasma equipment cannot be used for transistor fabrication, but it can be applied to
interconnect fabrications. This is because of the major disadvantages of using the current plasma
equipment,suchasseveremetalliccontaminationonthesubstratesurfacebyhigh-energyionbom-
bardment at the inner surface of the process chamber, severe damage to the substrate surface by
high-energy ion bombardment, severe charge-up damage due to the residual electric charges on
the substrate surface just after turning off the plasma, the charges induced by secondary electron
emissioncaused by high-energyion bombardment, andso on.
© 2006 by Taylor & Francis Group, LLC
It has been established that the newly developed microwave-excited high-density plasma
having very low electron temperatures overcomes all these disadvantages of the current plasma-
processing equipment, and can be applied to transistor fabrication, including gate insulator
formation. High-density plasma is excited by circularly polarized microwaves with a frequency
of 2.45GHz in the region of around 10 to 20mm under the ceramic shower plate. This is
carried out in order to introduce the flow of plasma excitation gas and radical generation gas
onto the substrate surface in a uniform manner. In the newly developed high-density plasma,
thereexistsaplasmadiffusionregionjustunderthisplasmaexcitationregion.Thisregionischarac-
terizedbyverylowelectrontemperatures,forexample,1.0eVforArgas,0.7eVforKrgas,and
0.5eV for Xe gas. Thus the bombarding ion energies are less than the critical values that would
cause metal sputtering and substrate surface damage.
Sisubstratesaresetinthisdiffusionplasmaregionwhereanelectroncurrentandanioncurrent
flowingontothesubstratesurfaceareequalwitheachotheratanyinstant,sothatasurfaceelectric
chargedoesnotremainevenaftertheplasmaisturnedoff.Thismeansthatthemostseveredisad-
vantages, that is, charge-up damage and high-energy ion bombardment-induced damage, are
essentially eliminated.
The introduction of this new plasma processing, free of charge-up damage, brings about a
drastic change in circuit layout pattern regulation, that is, a very limited antenna ratio, which is
defined as the ratio of the area of an interconnect of the gate electrode of a MOS transistor with
the area of the gate electrode. At present, the antenna ratio of circuit layout patterns of LSI
devices is strictly limited to less than 100 to 200, in order to obtain reasonable manufacturing
yields. When circuit layout patterns having very high antenna ratios such as 106 are designed in
LSI chips, all the chips on the Si substrates suffer fatal charge-up damage when using the
current plasma processing, resulting in zero percent yield. The number of plasma processing
stages throughout the entire LSI manufacturing process is up to several tens, so the antenna ratio
inthecircuitlayoutpatternisstrictlylimitedtolessthan100to200atpresent.Thenewlydevel-
opedplasmaprocessing,freeofcharge-updamage,isthemicrowave-excitedhigh-densityplasma,
which has very low electron temperatures and enables realization of LSI of any circuit layout
patternhaving arbitraryantennaratios.
Various reactive radicals (such as oxygen radicals O† and NH† radicals) generated in the
plasma excitation region diffuse to the Si substrate surface and form very high-integrity
SiO and Si N films on any crystal orientation Si substrate surface at low temperatures (such as
2 3 4
400 to 6008C). The leakage current through this radical-reaction-based gate insulator film has
been confirmed to be smaller by a factor of 3 orders of magnitude compared with that of the
current molecule reaction-based gate insulator film. The stagnation of current Si technologies
resulting from the existence of very large leakage currents through the gate insulator has thus
beencompletelyovercomebyintroducingtheseradical-reaction-basedgateinsulators.Moreover,
theintroductionofaradical-reaction-basedSi N filmtothegateinsulatorimprovesthemobility
3 4
of electrons and holes by a factor of at least 2 compared with that of the current thermal oxide
MOS transistor. It is therefore not necessary to introduce a SiGe region to MOS transistors in
order to improve the speed performance of LSI. Any difficulties arising from very large drain
leakage currents are completely overcome by introducing this radical-reaction-based Si N gate
3 4
insulator.
The flicker noise component, that is, 1/f noise, is also decreased by the introduction of
radical-reaction-based gate insulators, by a factor of 2 orders of magnitude. This is very
important to the future progress of LSI device miniaturization, while accompanied by a
gradual decrease in supply and signal voltages and without introduction of operation errors of
the devices.
Whereas the current molecule reaction-based thermal oxidation can produce relatively
high-integrity SiO films only on (100) Si surface orientations, the radical-reaction-based direct
2
oxidation and direct nitridation have been proven to produce very high-integrity insulator films
© 2006 by Taylor & Francis Group, LLC
on any crystal orientation of Si substrate surface. Thus, LSI device fabrication by the current
molecule reaction-based semiconductor manufacturing process is limited only to the (100) Si
surface and the structure of MOS transistors is limited to the two-dimensional planar structure.
As a result, only a small part of the capabilities of the Si material has been used for practical
applicationsso far.
On the other hand, the radical-reaction-based semiconductor manufacturing process has been
shown to utilize almost all of the Si material capabilities, for example, by LSI device fabrication
anycrystalorientationoftheSisubstratesurfaceandin3-dimensionalMOStransistors.Theelec-
triccurrentdrivabilityofpMOStransistorsfabricatedona(110)Sisubstratesurfacewithk110l
direction is 3 times larger than that of a pMOS transistor on a (100) Si surface, resulting in an
improvement of CMOS speed performance by afactor of2.
Dual-shower plate microwave-excited high-density plasma processing has been developed
simultaneously for application to various plasma CVD film formations and various material
pattern etchings. In this method, the lower shower plate, supplying source gases for film for-
mation and pattern etchings, is introduced to the diffusion plasma region described previously.
The plasma potential in the plasma diffusion region may be limited to less than 10V, at most,
due to the very low electron temperatures. This completely eliminates the sputtering of the
surface of the lower shower plate due to ion bombardment. Source material gases are supplied
to the plasma diffusion region, which has very low electron temperatures, and not to the plasma
excitation region, which has relatively high electron temperatures. This arrangement ensures that
the source material gas molecules are not decomposed so much. In other words, the source
material gas molecules can be supplied even to the bottom of the contact and the through
holes, however narrow the diameter and however deep, resulting in the realization of processing
of any narrow contact, through holes etchings, and film depositions at a constant process speed.
This process, which is free of the microloading effect, indicates that the speed of processes such
as etching rate and film-deposition rate are independent of the pattern size (for example, from
10mm to 25nm). Current semiconductor equipment commonly exhibits very severe microload-
ing effects, that is, the equipment can be applied to 100-nm generation LSI fabrication, but not
to 65-nm generation LSI fabrication. Thus, a huge amount of investment is continuously
required for new generation LSI fabrication. The new radical-reaction-based semiconductor
equipment, such as the microwave-excited high-density plasma with single and dual-shower
plate structures, can be used continuously for all LSI generations, to 25-nm. This is a revolu-
tionary change in semiconductor manufacturing.
It is well known that current plasma equipment for plasma CVD and RIE cannot maintain
plasma uniformity on the entire Si substrate surface for different gas pressures, different gas
combinations and concentrations, different substrate electrode self-biases, and different Si sub-
strate surfaces (different surface materials and different surface patterns). In such equipment,
plasmauniformityontheentireSisubstratesurfacehastoberealizedbytuningvariousequipment
parameters, which is quite tedious and time-consuming. Therefore, the current semiconductor-
manufacturing technologies work very effectively only for very large volume production of a
small variety of LSIdevices.
The key business area in the field of electronics is rapidly becoming digital consumer
electronics rather than of conventional personal-computer-oriented business, which requires only
very large-volume production of microprocessors and DRAMs. Digital electronics, however,
requires a very wide variety of LSIs with very small volume production. The new microwave-
excited high-density plasma with single- and dual-shower plate structures maintains plasma
uniformity on the entire substrate surface even under widely varying operation parameters such
as gas working pressure, gas combinations and concentrations, self-biases of the substrate elec-
trode, and substrate surface patterns and materials. Thus, Si substrates of any kind and structure
can be continuously processed at optimum conditions for each individual Si substrate without
tuning the various equipmentparametersand structures.
© 2006 by Taylor & Francis Group, LLC
The new radical-reaction-based semiconductor manufacturing using microwave-excited
high-density plasma will realize the production of a wide variety of LSI devices in very small
volumes, as is required in the digital consumer electronics era, in keeping with its very
high productivity. The semiconductor industry is now facing a real revolution. The present
Radical-Reaction-BasedSemiconductor-ManufacturingSeriesdescribesallthesenewlydeveloped
technologies ina very academicmanner.
TadahiroOhmi
© 2006 by Taylor & Francis Group, LLC
About the Editor
TadahiroOhmiwasborninTokyo,Japan,in1939.HeearnedhisB.S.,M.S.,andPh.D.degrees
in electrical engineering from Tokyo Institute of Technology, Tokyo, in 1961, 1963, and 1966,
respectively.
Prior to 1972, he served as a research associate in the Department of Electronics of
Tokyo Institute of Technology, where he worked on Gunn diodes such as velocity overshoot
phenomena, multi-valley diffusion and frequency limitation of negative differential mobility
due to an electron transfer in the multi-valleys, high-field transport in semiconductor such as
unified theory of space–charge dynamics in negative differential mobility materials, Bloch-
oscillation-induced negative mobility and Bloch oscillators, and dynamics in injection lasers.
In 1972, he moved to Tohoku University and is now a professor at the New Industry Creation
Hatchery Center (NICHe), Tohoku University. He is currently engaged in research on high-
performance ULSI free from gate and drain leakage currents, threshold voltage fluctuations of
MOS transistors and 1/f noises such as ultra-high-speed ULSI by introducing directly nitrided
Si N gate insulation as a high-K gate dielectric, nonporous fluorocarbon film having dielectric
3 4
constant less than 1.90 as an inter-metallic dielectric and Si (110) surface, and metal substrate
SOI,ULSIandlargesizeflat-paneldisplay,andadvancedsemiconductorandFPDprocesstechnol-
ogies by developing radical-reaction-based semiconductor and FPD manufacturing due to
microwave-excited very low electron temperature high-density plasma free from charge-up
damages andbombarding ion-induced damages.
Dr. Ohmi’s research activities include 2000 original papers and 1800 patent applications. He
received the Ichimura Award in 1979, the Inoue Harushige Award in 1989, the best paper award
of IEEE Transactions on Semiconductor Manufacturing in 1989, the Ichimura Prizes in
Industry-Meritorious Achievement Prize in 1990, the Okouchi Memorial Technology Prize in
1991, the Minister of State for Science and Technology Award for the promotion of invention in
1993, the Invention Prize and 4th International Conference on Soft Computing (IIZKA’96) Best
Paper Award in 1996, the IEICE Achievement Award in 1997, the Werner Kerm Award in
2001, ECS Electronics Division Award in 2003, the Medal with Purple Ribbon from the govern-
ment ofJapan in 2003, andthe BestCollaborationAward (the Prime Minister’s Award) in2003.
Dr. Ohmi is a member of the Institute of Electronics, Information and Communication
EngineersofJapan(fellow),theJapanSocietyofAppliedPhysics,theECS,andtheIEEE(fellow).
© 2006 by Taylor & Francis Group, LLC
Contributors
Nobukazu Ikeda Masaaki Nagase
Fujikin Incorporated Fujikin Incorporated
Osaka, Japan Osaka,Japan
TakashiImaoka
KojiNishino
ORGANOCorporation
Fujikin Incorporated
Tokyo, Japan
Osaka,Japan
Nobuhiko Inoue
Tadahiro Ohmi
Oki Electric Industry Corporation
New Industry Creation
Tokyo, Japan
HatcheryCenter
TohokuUniversity
HirotoIzumi
Sendai, Japan
Stella ChemifaCorporation
Osaka, Japan
Senri Ojima
Nomura Micro Science Co., Ltd.
HirohisaKikuyama
Kanagawa, Japan
Stella ChemifaCorporation
Osaka, Japan
Hiroshi Sugawara
ORGANO Corporation
Masafumi Kitano
Tokyo,Japan
New Industry CreationHatcheryCenter
Tohoku University
Sendai, Japan JunTakano
Stella Chemifa Corporation
Kenichi Mitsumori Osaka,Japan
Alps Electric Co., Ltd.
Sendai, Japan
Akinobu Teramoto
New Industry Creation
MasayukiMiyashita
HatcheryCenter
Stella ChemifaCorporation
TohokuUniversity
Osaka, Japan
Sendai, Japan
HitoshiMorinaga
Tatsuhiro Yabune
New Industry CreationHatcheryCenter
Stella Chemifa Corporation
Tohoku University
Osaka,Japan
Sendai, Japan
Hiroshi Morita Ikunori Yokoi
Kurita Water Industries Ltd. KuritaWater Industries Ltd.
Tokyo, Japan Tokyo,Japan
© 2006 by Taylor & Francis Group, LLC
Contents
Chapter1 SurfaceChemicalElectronicsat the Semiconductor Surface . . . 1
TadahiroOhmi
Chapter2 Principles ofSemiconductorDevice Wet Cleaning . .. . . .. . . 35
Hitoshi Morinaga
Chapter3 High-Performance Wet CleaningTechnology . .. . . .. . . .. . . 61
HiroshiMorita,Akinobu Teramoto, Hitoshi Morinaga, Senri Ojima,
and Kenichi Mitsumori
Chapter4 Etching of Various SiO . . .. . . .. . . .. . . .. . .. . . .. . . .. . 153
2
Tatsuhiro Yabune, Masayuki Miyashita, HirohisaKikuyama,
Jun Takano, andAkinobu Teramoto
Chapter5 Silicon Etching . .. . . .. . .. . . .. . . .. . . .. . .. . . .. . . .. . 251
KenichiMitsumoriand Nobuhiko Inoue
Chapter6 ChemicalComposition Control Technology . . .. . . .. . . .. . 271
Tatsuhiro Yabune, Masayuki Miyashita, HirohisaKikuyama,
and JunTakano
Chapter7 Wet VaporResistStripping Technology . . .. . .. . . .. . . .. . 285
Senri Ojima and TadahiroOhmi
Chapter8 Antistatic Technology .. . .. . . .. . . .. . . .. . .. . . .. . . .. . 293
KenichiMitsumoriand Takashi Imaoka
Chapter9 ChemicalWasteReclamation Technology .. . .. . . .. . . .. . 315
HiroshiSugawaraand Takashi Imaoka
Chapter10 Advanced Ultrapure Water and Liquid ChemicalSupply
Systemand Materials for Fluctuation-Free Facility .. . . .. . 331
IkunoriYokoi, Masaaki Nagase, Koji Nishino, Nobukazu Ikeda,
Masafumi Kitano, HirotoIzumi,and TadahiroOhmi
© 2006 by Taylor & Francis Group, LLC