throbber
Samsung Electronics Co., Ltd. v. Demaray LLC
`Samsung Electronic's Exhibit 1041
`Exhibit 1041, Page 1
`
`

`

`DISCLAIMER
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`This publication is based on sources and information believed to be reliable, but the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`authors and Lattice Press disclaim any warrantyorliability based on orrelating to the
`
`
`
`
`contents ofthis publication.
`
`
`
`
`
`Published by:
`
`LATTICE PRESS
`
`
`Post Office Box 340
`
`
`
`
`
`
`
`
`
`Sunset Beach, California 90742, U.S.A.
`
`
`
`
`
`
`
`
`
`
`
`
`Cover design by Roy Montibon, New Archetype Publishing, Los Angeles, CA.
`
`
`
`
`
`
`
`Copyright © 2000 by Lattice Press.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`All rights reserved. Nopart of this book may be reproduced ortransmitted in any form
`
`
`
`
`
`
`
`
`
`
`
`
`
`or by any means,electronic or mechanical, including photocopying, recording or by any
`
`
`
`
`
`
`
`
`
`
`
`information storage and retrieval system without written permission from the publisher,
`
`
`
`
`
`
`
`
`
`except for the inclusion of brief quotations in a review.
`
`
`
`
`
`
`
`Library of Congress Cataloging in Publication Data
`
`
`
`
`
`
`Wolf, Stanley and Tauber, Richard N.
`
`
`
`
`
`
`
`
`Silicon Processing for the VLSI Era
`
`
`
`
`Volume 1: Process Technology
`
`
`
`Includes Index
`
`
`
`
`
`
`
`1. Integrated circuits-Very large scale
`
`
`
`
`
`integration. 2. Silicon. I. Title
`
`ISBN 0-9616721-6-1
`
`
`
`
`98765432
`
`
`
`
`
`
`
`
`
`
`PRINTEDIN THE UNITED STATES OF AMERICA
`
`
`
`
`
`
`
`
`
`Ex. 1041, Page 2
`
`Ex. 1041, Page 2
`
`

`

`dominant technique for depositing a variety of metallic films in VLSI and ULSI fabrication,
`including aluminum alloys, titanium,titanium:tungsten,titanium nitride, tantalum, and cobalt.
`Sputtering displaced the original PVD technique for depositing metal films (evaporation) for
`the following reasons:
`1. Sputtering can be accomplished from large-area targets, which simplifies the problem of de-
`positing films with uniform thickness over large wafers.
`2. Film thickness control is relatively easily achieved by selecting a constant set of operating
`conditions, and then adjusting the deposition timeto reachit.
`3. The alloy composition of sputter-deposited films can be more tightly (and easily) controlled
`than that of evaporatedfilms.
`4. Many important film properties, such as step coverage and grain structure can be controlled
`by varying the negative bias and heat applied to the wafers. Other film properties (including
`stress and adhesion), can be controlled by altering such process parameters as power & pressure.
`5. The surface of the substrates can be sputter-cleaned in vacuum prior toinitiating film depos-
`ition (and the surface is not exposed again to ambient after such cleaning).
`6. There is sufficient material in most sputter targets to allow many deposition runs before target
`replacementis necessary.
`7. Device damage from x-rays generated during electron-beam evaporation is eliminated (al-
`though someother radiation damage maystill occur).
`Asis true with other processes, however,sputtering also has its drawbacks. They include:
`1. Sputtering processes involve high capital equipmentcosts;
`2. Since the processis carried out in low-medium vacuum ranges (comparedto the high vacuum
`conditions under which evaporation is conducted), there is greater possibility of incorporating
`impurities into the deposited film.
`3. Better step coverage can generally be achieved using CVD.
`In general, the sputtering process consists of four steps:
`1. Ions are generated anddirected ata target.
`2. The ions sputter target atoms;
`3. The ejected (sputtered) atoms are transported to the substrate.
`4. Upon reaching the substrate they condense and formathinfilm.
`Althoughit is of interest to note that sputtering can be conducted by generating the energetic
`incident ions by other means(e.g., ion beams),
`in virtually all VLSI and ULSI sputtering
`processes their source is a glow-discharge. The discussion of sputtering in this section will be
`limited to glow-discharge sputtering.4>.
`11.2.1 Introduction to Glow Discharge Physics
`The energetic particles used to strike target materials to be sputtered in ULSI sputter deposition
`systemsare generated by glow-discharges.*> A glow-discharge is a self-sustaining type of plas-
`ma(a plasmais defined as a partially ionized gas containing an equal numberofpositive and
`negative charges as well as some numberof neutral gas particles). In Fig. 11-3 a simple de-
`diode type system that can be employed to study properties of glow discharges used in
`
`
`
`Ex. 1041, Page 3
`
`

`

`ALUMINUMTHIN FILMS AND PHYSICAL VAPOR DEPOSITION IN ULSI
`
`439
`
`GLASS ENVELOPE
`ANODE
`
`(+)
`
`
`
`ANODE DARK
`SPACE
`
`SPACE
`AK
`
`COLUMN
`
`CATHODE
`
`ZAAA
`ra
`
`
`NEGATIVE
`
`| DARK SPACE GLOW
`
`
`
`alaaee
`
`
`VOLTAGE pe|
`
`(c)
`
`Fig. 11-3 (a) Structure of a glow discharge in a de diode system. (b) Charged particle concentration in a
`glow discharge. (c) Voltage variation in a de diode glow discharge.
`sputtering is shown.It consists of a glass tube that is evacuated and then re-filled with a gas at
`low pressure. Within the tube there are two electrodes (a positively charged anode and a
`negatively charged cathode) and a dc potential difference is applied between them.
`11.2.2 The Creation of Glow Discharges
`Consider the system shownin Fig. 11-3 to examine the case whena tubeis filled with Ar at an
`initial pressure of 1 torr, the distance between the electrodes is 15 cm, and a 1.5 kV potential
`difference is applied between them. Atthe outset no current flowsin the circuit, as all the Ar gas
`atoms are neutral and there are no charged particles in the gas. The full 1.5 kV is thus dropped
`between the two electrodes. If a free electron enters the tube (most likely created from the
`ionization of an Ar atom by a passing cosmic ray), it will be accelerated by the electric field
`existing between the electrodes (whose magnitude is: E= V/d = 1.5 kV/15 cm = 100 V/cm).
`The average distance that a free electron will travel at P = 1 torr before colliding with an Ar
`atom (i.e., the mean free path 4) is 0.0122 cm (Chap. 3). Most electron-atom collisions are
`elastic, in which virtually no energy is transferred between the electron and gas atom. Such
`elastic collisions occur because the mass of the electron is much smaller than that of the atom.
`Thus, the minumum distance an electron musttravel before it can undergo an inelastic collision
`(in which significant energy is transferred to the atom, either by the excitation of an atomic
`electron to a higher energy level, or to cause its escape from the atom) is about ten times A, or
`0.122 cm.If this is the minimum distance that must be traveled by electrons between inelastic
`collisions, there must be a significant number of electron path lengths in the range of 0.5-1.0
`cm.If a free electron travels 1 cm in the 100 V/cm electric field, it will have picked up 100 eV
`of kinetic energy. With this amount ofenergy, the free electron can transfer enough energy to an
`Ar electron to cause it to be excited or ionized. If this transferred energy Eis less than the
`ionization potential (e.g., 11.5 eV< E < 15.7 eV forAr), the orbital electron will be excited to a
`
`ee _
`Ex. 1041, Page 4
`
`J
`
`

`

`ALUMINUMTHIN FILMS AND PHYSICAL VAPOR DEPOSITION IN ULSI
`
`443
`
`SHIELD
`
`SHIELD TARGET
`RR
`
`
`
`Fig. 11-7 (a) Potential distribution in vicinity of cathode shield (b) Reducing rim effect by extending
`cathode shield. (c) Reducing rim effect by wrapping shield around the cathode.8 From L. Maissel and R.
`Glang, Eds., Handbook of Thin Film Technology 1970. Reprinted by permission of McGraw-Hill Book Co.
`dark space however, the ion production rate becomes reduced, and the voltage across the
`electrodes must rise to increase the secondary electron emission. Such a glow is known as an
`obstructed glow. In most practical sputter deposition systems the glow is obstructed. Thatis, in
`order to most effectively collect the sputtered material onto the substrate, the anode (on which
`the wafers are sometimes mounted)is placed as close to the cathode as possible (typically just
`far enough awayto avoid extinguishing the negative glow).
`It is typically necessary to insure that sputtering is allowed to occur only at the front side of
`the target, as the backside contains cooling coils and attachmentfixtures which are definitely not
`to be sputtered. To guarantee that no sputtering takes place except from desired surfaces, a
`shield of metal (at a potential equal to that of the anode) is placed at a distance less than the
`Crookes dark space at all other cathode surfaces (Fig. 11-7). Since no discharge will occur
`between two electrode surfaces separated by less than this distance, such shielding (termed
`dark-space shielding)is effective in preventing sputtering from unwanted cathode surfaces.
`
`11.3 THE PHYSICS OF SPUTTERING
`
`Whena solid surface is bombarded by atoms, ions, or molecules, many phenomena occur. The
`kinetic energy of the impinging particles largely dictates which are the most probable events.
`For low energy particles (<10 eV), most interactions occur only at the surface of the target
`material. At very low energies (<5 eV) such events are limited to reflection or physisorption of
`the bombarding species. For low energies which exceed the binding energy of the target
`material (S—10 eV), surface migration and surface damage effects can take place. At much
`higher energies (>10 keV), the impinging particles travel well into the bulk of the sample before
`slowing downand depositing their energy. Thus, such particles are most likely to be embedded
`in the target, and this mechanismis the basis of ion-implantation. At energies between the two
`extremes, two other effects also arise: 1) some fraction of the energy of the impinging ionsis
`transferred to the solid in the form of heat, and lattice damage; and 2) another fraction of such
`energy causes atoms from the surface to be dislodged and ejected into the gas phase
`(sputtering).
`
`Ex. 1041, Page 5
`
`

`

`444 SILICON PROCESSING FOR THE VLSI ERA
`
`3
`
`aay.
`oOo
`1
`/ NEUTRALIZED
`~~ —*
`GAS MOLECULE,
`REBOUNDED FROM
`THE TARGET SURFACE
`
`~
`
`BOMBARDING IONIZED
`
`GAS MOLECULE TARGET ANDIAS
`
`se INDICATE DINECIION OF
`MOMEN LUM TRANSFER.
`
`x
`
`\
`“
`}Nf
`
`‘
`
`SPUTTERED ATOM
`
`Fig. 11-8 (a) Binary collision between atom A and B, followed by a binary collision between atom B and
`C.(b) Collision process responsible for sputtering and fast neutral generation.
`11.3.1 The Billiard Ball Model of Sputtering
`The exact mechanisms which lead to the ejection of atoms under ion bombardmentare not
`known, and a comprehensive theory of sputtering is not likely to be developed in the near future
`since many parameters are involved. These include the kinetic energy of the ions, lattice
`structure, and binding energy oflattice atoms. Someofthe details, however, are reasonably well
`understood and can be aptly described with a relatively simple momentum-transfer model. G.K.
`Wehner, whose theoretical work first established a solid scientific basis for sputtering, often
`described sputtering as a game of three-dimensional billiards, played with atoms.? Using this
`analogy,it is possible to visualize how atoms maybe ejected from a surface as the result of two
`binary collisions (Figs, 11-8b and 11-8c) when a surface is struck by a particle with a velocity
`normal to the surface (e.g., atom A in Fig. 11-8b). Note that a binary collision is one in which
`the primary incoming particle (e.g., atom A)strikes a single object (e.g., atom B in Fig. 11-8b),
`and gives up a significant fraction of its energy to the struck atom, while retaining the remaining
`fraction. As a consequenceofthe collision, atom B may leave the point of impact at an angle
`greater than 45°.
`If atom B then undergoes a secondary collision with atom C, the angle at
`which atom C leaves the secondary impact point may again be greater than 45°. Thus, it is
`possible that atom C can have a velocity component greater than 90° (and thus be directed
`toward the surface). As a result, there is a finite probability that atom C will be ejected from the
`surface as a result of the surface being struck by atom A.
`Whenthe directions of sputtered atoms from the surface of polycrystalline materials (and
`most cathode materials in sputter applications are polycrystalline) are measured for the case of
`normalincidence, it is found that the ejected atoms leave the surface in essentially a cosine
`distribution. A cosine distribution, however, does not describe the sort of small-angle ejections
`that would be expected from the simple collision processes described above. Evidently in actual
`sputtering events, more than twocollisions are involved, and the energy delivered by impinging
`ions during normal incidence is so randomly distributed that
`the effect of the incident
`momentum vector is lost. Note that the energy range of sputtered atoms leaving the targetis
`
`i1
`
`
`
`Ex. 1041, Page 6
`
`

`

`ALUMINUM THIN FILMS AND PHYSICAL VAPOR DEPOSITION IN ULSI
`
`445
`
`typically 3-10 eV, and the bombarding species also recoil from the cathode face with some en-
`ergy. Thus, the target surface is a source of sputtered atoms and energetic backscattered species.
`For the case when the surface is bombarded by ionsat an oblique angle (i.e., 45°-90°), there
`is a higher probability that the primary collision between incident ions and surface atoms will
`lead to sputtering events. Furthermore, oblique incidence confines the action closer to the sur-
`face, and thus sputtering is enhanced.In cases of oblique bombardment, the incident momentum
`vector becomes important, and sputtered atomsare ejected strongly in the forward direction. In
`addition, the sputter yield (defined as the number of atoms ejected per incident ion), may be as
`muchas an order of magnitude larger than that resulting from normal incidence by bombarding
`ions (Fig. 11-9). This effect also leads to faceting, whichis covered in Sect. 11.7.4.
`11.3 2 Sputter Yield
`The sputter yield is important because it largely (but not completely) determines the rate of
`sputter deposition. Sputter yield depends on a number of factors besides the direction of
`incidence of the ions, and these include: a) the target material; b) the mass of bombardingions;
`and c) the energy of the bombarding ions. There is a minimum energy threshold for sputtering
`that is approximately equal to the heat of sublimation (e.g., 13.5 eV for Si). In the energy range
`of sputtering (10-5000 eV), the yield increases with ion energy and mass. Figure 11-10 shows
`the sputter yield of copper as a function of energy for various noble gas ions. The sputter yields
`of various materials in argon atdifferent energies is given in Table 11-2.9
`Several matters related to sputter yield should be noted. First, although the sputtering yields
`of various materials are different, as a group they are muchcloserin value to one anotherthan,
`for example,
`the vapor pressure of comparable materials. This makes the deposition of
`multilayer or multi-component films much more controllable by sputtering. The details of sput-
`tering from multi-componenttargets are discussed in the section on Process Considerations in
`Sputter Deposition. Second, since the bombarding ions are by no means monoenergetic in glow-
`discharge sputtering, it is not necessarily valid to use the sputter yield values for pure metals if
`alloys, compounds, or mixtures are being sputtered. Tabulations of sputtering yields, however,
`are useful for obtaining rough indications of deposition or etch rates of various materials.
`
` INCIDENT 10N
`
`SPUTTERING
`
`YIELD,Ste)
`
`max
`ANGLE OF INCIDENCE 141
`
`m2
`
`Fig. 11-9 Schematic diagram showing variation of sputtering yield with ion angle of incidence.
`
`Ex. 1041, Page 7
`
`

`

`
`
`446 SILICON PROCESSING FOR THE VLSI ERA
`
`Table 11-2 SPUTTER YIELDS FOR METALSIN ARGON (ATOMS/ION)
`
`Target
`
`At.Wt/Dens.
`
`100 eV
`
`300 eV
`
`600eV
`
`1000 eV
`
`2000 eV
`
`Al
`Au
`Cu
`Ni
`Pt
`Si
`Ta
`Ti
`Ww
`
`10.0
`10.2
`7.09
`6.6
`9.12
`12.05
`10.9
`10.62
`14.06
`
`0.11
`0.32
`0.5
`0.28
`0.2
`0,07
`0.1
`0.08
`0.12
`
`0.65
`1.65
`1.6
`0.95
`0.75
`0.31
`0.4
`0.33
`0.41
`
`1.2
`2.8
`2.3
`1.5
`1.6
`0.5
`0.6
`0.41
`0.75
`
`1.9
`3.6
`3.2
`24
`
`0.6
`0.9
`0.7
`
`2.0
`5.6
`4.3
`
`0.9
`
`11.3.3 Selection Criteria for Process Conditions and Sputter Gas
`The information gleaned from sputter yields and the physics of sputtering can also be applied
`toward an understanding of how process conditions and materials are selected for sputtering
`including: a) type of sputtering gas; b) pressure range of operation; and c) electrical conditions
`for the glow discharge.4® That is,
`in purely physical sputtering (as opposed to reactive
`sputtering) it is important that the ions or atomsof the sputtering gas not react with the growing
`film. This limits the selection of sputtering species to the noble gases. Furthermore, argon is
`generally the gas of choice, since it is easily available (hence low in cost), and its mass is a good
`match to those of the elements most frequently sputtered (Al, Cu, Si, and Ti), giving adequate
`sputtering yields for these elements. The pressure range of operationis set by the requirements
`of the glow discharge (lower limit ~2-3 mtorr for magnetron sputtering) and the scattering of
`sputtered atoms by the sputter gas (upper limit 100 mtorr). In addition, a desired goal of sputter
`deposition is to obtain maximum deposition rates. As a result, electrical conditions are selected
`
`Almen & Bruce
`(1961)
`
`xe
`
`Copper
`
` 20+
`
`=°
`
`3E°g <~
`
`0
`
`10
`
`20
`
`30
`
`40
`
`—_*
`50
`60
`70
`kev
`
`Fig. 11-10 Sputtering yields of the noble gases on copper, as a function of energy.!9
`
`lon Energy
`
`Ex. 1041, Page 8
`
`

`

`ALUMINUM THIN FILMS AND PHYSICAL VAPOR DEPOSITION IN ULSI
`
`
`
`
`
`
`
`
`
`
`447
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`to give a maximum sputter yield per unit energy. That is, as energy is increased, each energy
`
`
`
`
`
`
`
`
`
`
`
`
`
`increment gives a progressively smaller increase in sputter yield. This occurs because higher
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`energy ions implant themselves, and thus end up dissipating a greater proportion of their energy
`
`
`
`
`
`
`
`
`
`
`
`
`via non-sputtering processes. The most efficient ion energies for sputtering are typically
`
`
`
`
`
`
`
`
`obtained for electrode voltages of several hundred volts.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`In general, the higher the current at the cathode, the higher is the film deposition rate, since
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`more ions are striking the cathode (and thus cause more sputtering). The product of the cathode
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`current and the electrode voltage gives the input power of the sputtering process. In magnetron
`sputtering, cathode current densities of 10-100 mA/em?at a few hundred volts are typical.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`From the foregoing discussion on the mechanism of sputtering one can see that sputtering is
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`a highly inefficient process. In fact, ~70% of the energy consumed during the sputtering process
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`is dissipated as heat in the target, and ~25% by emission of secondary electrons and photons
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`from the target. This heating can raise target temperatures to levels capable of damaging the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`target, associated vacuum components, or the material that bonds the target to the backing
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`electrode. The target must therefore be cooled to avoid these problems. As we will discusslater,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`this is normally accomplished by bonding the target to a water-cooled Cu backingplate.
`
`
`
`
`
`
`
`
`11.3.4 Secondary Electron Production for Sustaining the Discharge
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Nearthe outset of the discussion on glow-discharges, it was observed that glow-discharges must
`
`
`
`
`
`
`
`
`
`
`
`
`
`be continuously provided with free electrons to keep them sustained. In most dc-sputtering
`
`
`
`
`
`
`
`
`
`
`
`
`
`systems, the source of such electrons is secondary-electron emission from the target. An
`
`
`
`
`
`
`
`
`
`
`
`
`important mechanism that generates such secondary electrons is Auger emission, although other
`
`
`
`
`
`
`
`
`
`
`
`
`emission mechanismsare also responsible for their production. The Auger emission process
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`occurs according to the sequence of events shown in Fig. 11-11. Target electrons require a
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`minimum amount of energy g® to escape into free space (where Mis the work function of the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`target material, and g® is typically 3~5 eV). However, whena positive ion (such as Art) comes
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`close to the target surface it appears as a potential well of 15.67 eV to target electrons occupying
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`energy states near the Fermi level. Thus, some of the target electrons can escape by tunneling
`
`
`
`
`
`
`
`
`
`
`
`
`
`into these potential wells. The energy difference between 15.67 eV and the mini-mum free-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`space escape energy is released in the form of a photon. If this photon (e.g., of energy 10.67 eV
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`= [15.67 eV - 5 eV]) is absorbed by another electron near the target surface, this electron(i.e.,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`the Auger electron) may posses enough energy to be emitted from the target into free space.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Since this sequence of events is rather improbable, several ions must strike the cathode in order
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`for a single Auger electron to be emitted. The secondary electron yield per bombarding ion ¥;
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`has been measured to be 0.05-0.1 for metal targets. Thus, each secondary electron needs to
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`generate as much as 20 ion-electron pairs before it reaches the anode to sustain the discharge.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Besides providing a source of electrons to sustain the discharge, electron emission is also
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`important to the sputtering process in other ways. First, ions bombarding the cathode are
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`neutralized. Thus, it is highly probable that each ion that closely approaches the target will
`extract an electron from the target, and return to the discharge as a neutral atom. Second, the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`total target current, IJ, is the sum of the ion flux striking the target, I,, and the secondary electron
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`current leaving the electrode, I. Since y,
`is larger for dielectric materials, their I, is also larger.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`This implies that for the same cathode current, I, dielectrics will sputter more slowly than
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`metals. That is, in dielectric sputtering a larger fraction of I is due to electron emission, and thus
`
`
`
`
`
`
`
`
`
`
`
`
`
`for equal valuesof I, a smaller ion flux is striking the dielectric target.
`
`Ex. 1041, Page 9
`
`Ex. 1041, Page 9
`
`

`

`448 SILICON PROCESSING FOR THE VLSI ERA
`
`
`
`
`
`
`
`
`
`
`
`4
`SMOLIN,
`C/A SEA OF CONDUCTION ©
`“ ELECTRONS ~~" .-
`
`LL Le ONS
`LLL KE 2
`
`
`
`eV;
`
`yy = IONIZATION POTENTIAL
`15.76 VOLTS FOR ARGON
`
`[elOFCAPTUREDELECTION
`
`TRANSITION OF EXCITED ELECTRON (AUGER PROCESS)
`
`
`
`
`
`
`
`
`
`
`
`
`Fig. 11-11 Potential energy diagram for ion approaching a metal target.
`
`
`
`
`
`
`
`
`
`
`
`
`11.3.5 Sputter Deposited Film Growth
`
`Upon being ejected from the target surface, sputtered atoms have velocities of 3-6x10° cm/sec
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`and energies of 10-40 eV.It is desirable that as many of these sputtered atoms as possible be
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`deposited upon the substrates to form the specified thin film. To accomplish this, the target and
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`wafers are closely spaced, with spacings of 5-10 cm being typical. The mean free path, A, of
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`sputtered atoms at typical sputter pressures is less than 5-10 cm (e.g., at 5 mtorr, A =1 cm).
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Thus, it is likely that sputtered atoms will suffer one or more collisions with the sputter-gas
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`atoms before reaching the substrate (Fig. 11-12). The sputtered atoms may therefore: a) arrive at
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`the substrate with reduced energy (~1—2 eV); b) be backscattered to the target or the chamber
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`walls; or c) lose enough energy so that thereafter, they move by diffusion in the same manner as
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`neutral sputter gas atoms. These events imply that the sputtering gas pressure can impact various
`
`
`
`
`
`
`
`
`
`
`
`
`
`film deposition parameters, such as the deposition rate and compositionof the film.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The formation and growth of the thin film on the substrate proceeds according to the general
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`discussion on thin film formation given in Chap. 4. Therefore, this discussion is restricted to the
`
`
`
`
`
`
`
`
`
`
`
`
`events that uniquely impact the formation and growth of glow-discharge sputteredfilms.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The substrate surface (onto which the desired film of the sputtered target material is
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`deposited) is also subjected to impingement by many species. The sputtered atoms arrive and
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`condense onto the substrate. For a typical deposition rate of 20 nm/min, a monolayer of
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`deposited film will form approximately every second (assuming the size of a typical atom is
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`~0.3 nm). For this case, and even for much higher deposition rates of Al
`(e.g., 620 nm/min), the
`
`
`
`
`
`
`
`
`
`
`
`
`
`heat of condensation is not necessarily the most important source of substrate heating.
`
`
`
`
`
`
`
`
`
`11.3.6 Species that Strike the Wafer During Film Deposition
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`In addition to the sputtered atoms, the substrate is also struck by many other species (Fig. 11-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`13), the most important being: 1) fast neutral sputter gas atoms, which retain significant energy
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`after striking and recoiling from the cathode. As they impinge upon the substrate, some may
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`embed themselvesin the growing film; 2) negative ions,
`formed near the cathode surface by
`
`
`
`Ex. 1041, Page 10
`
`Ex. 1041, Page 10
`
`

`

`
`
`ALUMINUM THIN FILMS AND PHYSICAL VAPOR DEPOSITION IN ULSI
`
`
`
`
`
`
`
`
`
`
`
`
`461
`
`
`
`Fig. 11-23 An spent aluminum circular-planar-magnetrontarget(right) is being changed.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`reported that deposition rates of 1200-nm/min can be achieved when depositing Al:(0.5%)Cu
`
`films on 200-mm wafers.>9 In fact, the maximum sputter deposition rates with such sources are
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`usually not limited by any aspect of the plasma (as in dc and rf diode cases), but rather by the
`
`
`
`
`
`
`
`
`
`
`
`
`ability to cool the cathode target to keep it from melting.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The fact that the annular shape of the plasmain a circular planar magnetron results in a non-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`uniform erosion of the target also causes the deposition onto nearby wafers to be non-uniform.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`In fact the deposition profile mirrors the race track when the waferis close to the target, but this
`deposition profile becomes more smeared as the sample is moved further from the target (see
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Fig. 11-24a). Thus, a tradeoff is often made between deposition rate and thickness uniformity
`
`
`
`
`
`
`
`
`
`
`
`
`
`(i.e., uniformity improves as the target-to-wafer spacing is increased, but the deposition rate
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`decreases). Note also from Fig. 11-24 that the target diameter is made larger than the wafer
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`diameter to improve thickness uniformity. As an example, for 200-mm (8 in) wafers, the target
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`diameter is typically 330 mm (14 in). Figure 11-24b shows such a target sputtering onto a
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`smaller wafer. In practice, thickness uniformities of 5% (30) within a wafer can be achieved.
`
`
`
`
`
`
`
`11.6 VLSI AND ULSI SPUTTER DEPOSITION EQUIPMENT
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`A number of different sputtering systems have been designed for commercial use in IC
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`fabrication. Only the dominant sputtering systems designs that were used in VLSI and ULSI
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`generations will be described (i.e., those used for IC fabrication on 125-mm, 150-mm, and 200-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`mm wafers). Sputtering systems used for earlier IC technologies are described in thefirst edition
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`of Vol. 1. In this section the generic components of a sputtering system are discussed first. Then
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`the sputtering systems used in 125-mm and 150-mm wafer processing are briefly examined. The
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`focusof this discussion is on the sputtering systems used for 200-mm wafers (and larger).
`
`
`
`
`
`
`
`
`11.6.1 The Components of a Generic Sputtering System
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The schematic of a basic sputtering system is shown in Fig. 11-25. It consists of the following
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`subsystems: a) the sputter chamber, in which the substrate holder and sputtering source reside
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`(the source also includes the target); b) vacuum pumps; c) power supplies (dc and/or rf); d)
`
`
`
`
`
`
`
`
`
`
`
`sputtering gas supply and flow controllers; e) monitoring equipment (pressure gauges, volt-
`
`Ex. 1041, Page 11
`
`Ex. 1041, Page 11
`
`

`

`462 SILICON PROCESSING FOR THE VLSI ERA
`
`
`Scm THROW
`
`Cathode
`
`
`
`SI
`
`Xu
`
`|
`Sem
`
`ETCH TRACK DIAMETER
`
`Spring
`
`Wafer
`=< ¢—_
`clip
`
`Heater
`
`block
`
`
`(a)
`
`o
`2
`a 10000
`5
`
`$000
`
`&B
`
`w
`
`
`
`
`
`insulation
`Heater
`
`element
`Thermocouple
`Argon
`Fig. 11-24 a) The deposition profile and deposition rate on a sample in front of a magnetron cathode as a
`function of sample distance.” Reprinted with permission of Academic Press. b) Sputter deposition from a
`planar magnetrontarget whose diameteris larger than thatof the wafer.
`(pressure gauges, voltmeters, and residual-gas analyzers); f) wafer holders and handling
`mechanisms; and g) microcomputer controller.
`In modern sputter tools the sputter chambers are isolated from the ambient by a load lock.
`That is, a cassette of wafers to be sputter deposited is placed into the loadiock. Then the valve
`connecting the loadlock to the outside ambient is closed. The loadlock is pumped down, and
`upon reaching the base pressure of the transfer chamber the vacuum valve to the transfer
`chamber is opened. A robotic arm in the transfer chamber removes a wafer from the cassette
`and bringsit into the transfer chamber. The loadlock/transfer chambervalve is then closed. The
`sputter chamber awaiting the wafer has also been pumped to the appropriate low base pressure
`(e.g., 10°°-10°torr), The vacuum valveto the sputter chamberis opened,
`andthe waferis then
`DC/RF
`RF VOLTAGE
`
`
`
`
`prose (V)
`
` P
`
`DC VOLTAGE
`
`
`OSCILLOSCO
`
`
`
`
`
`GAS SUPPLY
`
`DC/RF/GROUND
`Fig. 11-25 Schematic drawing showing someof the components of a sputtering system.
`
`2
`
`Ex. 1041, Page 12
`
`

`

`ALUMINUM THIN FILMS AND PHYSICAL VAPOR DEPOSITION IN ULSI
`
`
`
`
`
`
`
`
`
`
`463
`
`
`
`
`
`
`
`Fig. 11-26 Examples of sputtering targets for planar circular magnetron sources.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`inserted into the sputter chamber by the robot arm. After the arm withdraws, the vacuum valve
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`between the sputter chamber and the transfer chamber is closed, and sputter gas is flowed into
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`the chamber to the appropriate pressure (e.g., 5 mtorr) and flow rate. The sputter process can
`then bestarted.
`
`
`
`
`
`
`11.6.1.1 Sputtering Targets: Sputtering targets consist of the material that is to be sputter-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`deposited onto the wafer. Targets in circular magnetronsare circular discs 3 to 10 mmthick,as
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`shown in Fig. 11-26. These are bonded to a water-cooled Cu backing plate for good thermal
`
`
`
`
`
`
`
`
`
`
`
`
`contact, usually by soldering with a low-melting-temperature metal (or with a conductive
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`epoxy). Since over 70% of the energy incident onto the ta

This document is available on Docket Alarm but you must sign up to view it.


Or .

Accessing this document will incur an additional charge of $.

After purchase, you can access this document again without charge.

Accept $ Charge
throbber

Still Working On It

This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.

Give it another minute or two to complete, and then try the refresh button.

throbber

A few More Minutes ... Still Working

It can take up to 5 minutes for us to download a document if the court servers are running slowly.

Thank you for your continued patience.

This document could not be displayed.

We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.

Your account does not support viewing this document.

You need a Paid Account to view this document. Click here to change your account type.

Your account does not support viewing this document.

Set your membership status to view this document.

With a Docket Alarm membership, you'll get a whole lot more, including:

  • Up-to-date information for this case.
  • Email alerts whenever there is an update.
  • Full text search for other cases.
  • Get email alerts whenever a new case matches your search.

Become a Member

One Moment Please

The filing “” is large (MB) and is being downloaded.

Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.

Your document is on its way!

If you do not receive the document in five minutes, contact support at support@docketalarm.com.

Sealed Document

We are unable to display this document, it may be under a court ordered seal.

If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.


Access Government Site

We are redirecting you
to a mobile optimized page.





Document Unreadable or Corrupt

Refresh this Document
Go to the Docket

We are unable to display this document.

Refresh this Document
Go to the Docket