`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Exhibit 2005
`Exhibit 2005
`
`

`

`
`
`
`
`
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`_____________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`_____________________
`
`
` TAIWAN SEMICONDUCTOR MANUFACTURING COMPANY,
`LTD.GLOBALFOUNDRIES U.S., INC., GLOBALFOUNDRIES
`DRESDEN
`MODULE ONE LLC & CO. KG, GLOBALFOUNDRIES DRESDEN
`MODULE
`TWO LLC & CO. KG
`Petitioners
`
`v.
`
`ZOND, LLC
`Patent Owner
`
`U.S. Patent No. 7,147,759
`
`_____________________
`
`Inter Partes Review Case Nos. IPR2014-00781,
`00782, 01083, 01086, 01087
`
`_____________________
`
`DECLARATION OF LARRY D. HARTSOUGH, Ph.D.
`
`

`

`TABLE OF CONTENTS
`
`
`I. Education and Professional Background ......................................................................1
`
`II. Summary of Opinions ...................................................................................................5
`
`III. Legal Standards ............................................................................................................5
`
`A.
`
`B.
`
`C.
`
`Level of Ordinary Skill in the Art ........................................................................ 5
`
`Legal Standards for Anticipation ......................................................................... 5
`
`Legal Standards for Obviousness ......................................................................... 6
`
`IV. Background Topics .......................................................................................................8
`
`A.
`
`B.
`
`C.
`
`D.
`
`Voltage, current, impedance and power ............................................................... 8
`
`Control systems .................................................................................................. 10
`
`Set point (Controlled Parameter)........................................................................ 13
`
`Power Control vs Voltage Control ..................................................................... 14
`
`E. Magnetron Sputtering History and Operation ........................................................ 16
`
`V. Patent 7,147,759 .........................................................................................................23
`
`VI. Claim Construction .....................................................................................................26
`
`VII. Prior Art ......................................................................................................................26
`
`A. Wang .................................................................................................................. 26
`
`a.
`
`b.
`
`c.
`
`Wang’s Power Pulses ...................................................................................27
`
`Arcing in Wang ............................................................................................30
`
`Variances between Wang’s Target Power Levels and Actual Power ..........33
`
`B.
`
`Kudryavtsev ....................................................................................................... 37
`
`d.
`
`e.
`
`Arcing in Kudryastev ...................................................................................38
`
`Lack of Disclosure of Configured Rise Time or Amplitude........................45
`
`It Would Not Have Been Obvious To Combine The Cylindrical Tube System
`VIII.
`Without A Magnet Of Kudryavtsev With Either The Mozgrin Or Wang Magnetron
`System 46
`
`IX. The Cited References Do Not Teach All Of The Claim Limitations Of Any Claim Of
`The ‘759 Patent ..................................................................................................................56
`
`The cited references do not teach generating “an amplitude and a rise time of the
`A.
`voltage pulse being chosen to increase an excitation rate of ground state atoms that are
`
`

`

`present in the weakly-ionized plasma to create a multi-step ionization process that
`generates a strongly-ionized plasma,” as recited in independent claim 20 and as
`similarly required by independent claims 1 and 40....................................................... 56
`
`The combination of Wang and Kudryavtsev does not teach a “multi-step
`B.
`ionization process comprising exciting the ground state atoms to generate excited
`atoms, and then ionizing the excited atoms within the weakly-ionized plasma without
`forming an arc discharge,” as recited in claims 1 and 20, and as similarly recited in
`claim 40. ........................................................................................................................ 65
`
`The Combination of Wang, Kudryavtsev and Yamaguchi Does Not Teach
`C.
`“ionizing the feed gas comprises exposing the feed gas to an electrode that is adapted
`to emit electrons,” As Recited In Claim 38. .................................................................. 69
`
`The Combination of Muller-Horche's UV source and Wang Would Not Have
`D.
`Taught or Suggested “the ionizing the feed gas comprises exposing the feed gas to at
`least one of a UV source, an X-ray source, an electron beam source, and an ion beam
`source,” As Recited In Claim 39. .................................................................................. 73
`
`E. The Combination of Wang and the Mozgrin’s Thesis Does Not Teach that “the
`rise time of the voltage pulse is approximately between 0.01 and 100 Vμsec,” As
`Recited In Claim 49 And As Similarly Recited in Claim 44. ....................................... 77
`
`The Combination of Wang and Kudryavtsev Would Not Have Taught or
`C.
`Suggested That “applying the electric field comprises applying a substantially uniform
`electric field,” As Recited In Claim 22. ........................................................................ 80
`
`The Combination of Wang and Kudryavtsev Would Not Have Taught or
`D.
`Suggested That “selecting at least one of a pulse amplitude and a pulse width of the
`electrical pulse that causes the strongly-ionized plasma to be substantially uniform in
`an area adjacent to a surface of the sputtering target,” As Recited In Dependent Claim
`26 And As Similarly Recited in Claim 31..................................................................... 82
`
`E. The Combination of Wang and Kudryavtsev Would Not Have Taught or
`Suggested That “the ions in the strongly-ionized plasma impact the surface of the
`sputtering target in a substantially uniform manner,” As Recited In Claim 30. ........... 84
`
`F. The Combination of Wang and Kudryavtsev Does Not Teach “a temperature
`controller that controls the temperature of the substrate support,” As Recited In Claim
`11. 86
`
`The Combination of Wang and Muller-Horsche Does Not Teach That “the
`G.
`ionization source is chosen from the group comprising a UV source, an X-ray source,
`an electron beam source, and an ion beam source,” As Recited In Claim 17. .............. 87
`
`The Combination of Wang and Kudryavtsev Does Not Teach “a power supply
`H.
`that generates constant power,” as recited in dependent claim 2. ................................. 90
`
`

`

`I. The Combination of Wang and Kudryavtsev Does Not Teach that “the power
`supply generates a constant voltage,” as recited in claim 3. ......................................... 92
`
`J. The Combination of Wang and Kudryavtsev Does Not Teach That “the rise time
`of the voltage pulse is chosen to increase the ionization rate of the excited atoms in the
`weakly-ionized plasma,” As Recited In Dependent Claim 6. ....................................... 95
`
`The Combination of Wang and Kudryavtsev Does Not Teach That “the
`K.
`strongly-ionized plasma is substantially uniform proximate to the sputtering target,” As
`Recited In Dependent Claim 9. ..................................................................................... 96
`
`L. The Combination of Wang and Kudryavtsev Does Not Teach That “volume
`between the anode and the cathode assembly is chosen to increase the ionization rate of
`the excited atoms in the weakly-ionized plasma,” As Recited In Dependent Claim 13.
`
`97
`
`
`
`
`
`
`
`

`

`I, Larry Hartsough, Ph.D., hereby declare:
`
`
`
`
`1.
`
`I am making this declaration at the request of patent owner Zond,
`
`LLC, in the matter of the Inter Partes Reviews (IPRs) of U.S. Patent No.
`
`7,147,759 (the “’759 Patent”), as set forth in the above caption.
`
`2.
`
`I am being compensated for my work in this matter at the rate of
`
`$300 per hour. My compensation in no way depends on the outcome of this
`
`proceeding.
`
`3.
`
`The list of materials I considered in forming the opinions set
`
`forth in this declaration includes the ’759 patent, the file history of the ’759
`
`patent, the Petitions for Inter Partes Review and the exhibits, the PTAB’s
`
`Institution Decisions, the transcript of the deposition of the Petitioners’ expert
`
`on the ‘759 patent, and the prior art references discussed below.
`
`I. Education and Professional Background
`
`4. My formal education is as follows. I received a Bachelors of
`
`Science degree in 1965, Master of Science degree in 1967, and Ph.D. in 1971,
`
`all in Materials Science/Engineering from the University of California,
`
`Berkeley.
`
`5.
`
`I have worked in the semiconductor industry for approximately
`
`30 years. My experience includes thin film deposition, vacuum system
`
`
`
`1
`
`

`

`design, and plasma processing of materials. I made significant contributions
`
`to the development of magnetron sputtering hardware and processes for the
`
`metallization of silicon integrated circuits. Since the late 1980’s, I have also
`
`been instrumental in the development of standards for semiconductor
`
`fabrication equipment published by the SEMI trade organization.
`
`6.
`
`From 1971-1974, I was a research metallurgist in the thin film
`
`development lab of Optical Coating Laboratory, Inc. In 1975 and 1976, I
`
`developed and demonstrated thin film applications and hardware for an in-
`
`line system at Airco Temescal. During my tenure (1977-1981) at Perkin
`
`Elmer, Plasma Products Division, I served in a number of capacities from
`
`Senior Staff Scientist, to Manager of the Advanced Development activity, to
`
`Manager of the Applications Laboratory. In 1981, I co-founded a
`
`semiconductor equipment company, Gryphon Products, and was VP of
`
`Engineering during development of the product. From 1984-1988, I was the
`
`Advanced Development Manager for Gryphon, developing new hardware and
`
`process capabilities. During 1988-1990, I was Project Manager at General
`
`Signal Thinfilm on a project to develop and prototype an advanced cluster tool
`
`for making thin films. From 1991-2002, I was Manager of PVD (physical
`
`vapor deposition) Source Engineering for Varian Associates, Thin Film
`
`Systems, and then for Novellus Systems, after they purchased TFS. Since
`
`
`
`2
`
`

`

`then, I have been consulting full time doing business as UA Associates, where
`
`my consulting work includes product development projects, film failure
`
`analysis, project management, technical presentations and litigation support.
`
`7.
`
`Throughout my career, I have developed and/or demonstrated
`
`processes and equipment for making thin films, including Al, Ti-W, Ta, and
`
`Cu metallization of silicon wafers, RF sputtering and etching, and both RF
`
`and dc magnetron reactive sputtering, for example SiO2, Al2O3, ITO (Indium-
`
`Tin Oxide), TiN, and TaN. I have been in charge of the development of two
`
`sputter deposition systems from conception to prototype and release to
`
`manufacturing. I have also specialized in the development and improvement
`
`of magnetically enhanced sputter cathodes. I have experience with related
`
`technology areas, such as wafer heating, power supply evaluation, wafer
`
`cooling, ion beam sources, wafer handling by electrostatics, process pressure
`
`control,
`
`in-situ wafer/process monitoring, cryogenic pumping, getter
`
`pumping, sputter target development, and physical, electrical and optical
`
`properties of thin films.
`
`8.
`
`I am a member of a number of professional organizations
`
`including the American Vacuum Society, Sigma Xi (the Scientific Research
`
`Society), and as a referee for the Journal of Vacuum Science & Technology.
`
`I have been a leader in the development of SEMI Standards for cluster tools
`
`
`
`3
`
`

`

`and 300mm equipment, including holding various co-chair positions on
`
`various standards task forces. I have previously served as a member of the
`
`US Department of Commerce’s Semiconductor Technical Advisory
`
`Committee.
`
`9.
`
`I have co-authored many papers, reports, and presentations
`
`relating to semiconductor processing, equipment, and materials, including the
`
`following:
`
`a. P. S. McLeod and L. D. Hartsough, "High-Rate Sputtering of
`Aluminum for Metalization of Integrated Circuits", J. Vac. Sci.
`Technol., 14 263 (1977).
`b. D. R. Denison and L. D. Hartsough, "Copper Distribution in
`Sputtered Al/Cu Films", J. Vac. Sci. Technol., 17 1326 (1980).
`c. D. R. Denison and L. D. Hartsough, "Step Coverage in Multiple Pass
`Sputter Deposition" J. Vac. Sci. Technol., A3 686 (1985).
`d. G. C. D’Couto, G. Tkach, K. A. Ashtiani, L. Hartsough, E. Kim, R.
`Mulpuri, D. B. Lee, K. Levy, and M. Fissel; S. Choi, S.-M. Choi,
`H.-D. Lee, and H. –K. Kang, “In situ physical vapor deposition of
`ionized Ti and TiN thin films using hollow cathode magnetron
`plasma source” J. Vac. Sci. Technol. B 19(1) 244 (2001).
`
`10. My areas of expertise include sputter deposition hardware and
`
`processes, thin film deposition system design and thin film properties. I am a
`
`named inventor on twelve United States patents covering apparatus, methods
`
`or processes in the fields of thin film deposition and etching. A copy of my
`
`CV is attached as Attachment A.
`
`
`
`
`
`4
`
`

`

`II. Summary of Opinions
`
`11.
`
`It is my opinion that none of the claims of the ‘759 patent are
`
`obvious.
`
`III. Legal Standards
`
`12.
`
`In this section I describe my understanding of certain legal
`
`standards. I have been informed of these legal standards by Zond’s attorneys.
`
`I am not an attorney and I am relying only on instructions from Zond’s
`
`attorneys for these legal standards.
`
`A. Level of Ordinary Skill in the Art
`
`13.
`
`In my opinion, given the disclosure of the ’759 Patent and the
`
`disclosure of the prior art references considered here, I consider a person of
`
`ordinary skill in the art at the time of filing of the ‘759 Patent to be someone
`
`who holds at least a bachelor of science degree in physics, material science,
`
`or electrical/computer engineering with at least two years of work experience
`
`or equivalent in the field of development of` plasma-based processing
`
`equipment. I met or exceeded the requirements for one of ordinary skill in the
`
`art at the time of the invention and continue to meet and/or exceed those
`
`requirements.
`
`B.
`
`Legal Standards for Anticipation
`
`14.
`
`I understand that a claim is anticipated under 35 U.S.C. § 102 if
`
`
`
`5
`
`

`

`(i) each and every element and limitation of the claim at issue is found either
`
`expressly or inherently in a single prior art reference, and (ii) the elements and
`
`limitations are arranged in the prior art reference in the same way as recited
`
`in the claims at issue.
`
`C. Legal Standards for Obviousness
`
`15.
`
`I understand that obviousness must be analyzed from the
`
`perspective of a person of ordinary skill in the relevant art at the time the
`
`invention was made. In analyzing obviousness, I understand that it is
`
`important to understand the scope of the claims, the level of skill in the
`
`relevant art, and the scope and content of the prior art, the differences between
`
`the prior art and the claims.
`
`16.
`
`I understand that even if a patent is not anticipated, it is still
`
`invalid if the differences between the claimed subject matter and the prior art
`
`are such that the subject matter as a whole would have been obvious at the
`
`time the invention was made to a person of ordinary skill in the pertinent art.
`
`17.
`
`I understand that a person of ordinary skill in the art provides a
`
`reference point from which the prior art and claimed invention should be
`
`viewed. This reference point prevents one from using his or her own insight
`
`or hindsight in deciding whether a claim is obvious.
`
`18.
`
`I also understand that an obviousness determination includes the
`
`
`
`6
`
`

`

`consideration of various factors such as (1) the scope and content of the prior
`
`art; (2) the differences between the prior art and the asserted claims; and (3)
`
`the level of ordinary skill in the pertinent art.
`
`19.
`
`I also understand that a party seeking to invalidate a patent as
`
`obvious must demonstrate that a person of ordinary skill in the art would have
`
`been motivated to combine the teachings of the prior art references to achieve
`
`the claimed invention, and that the person of ordinary skill in the art would
`
`have had a reasonable expectation of success in doing so. This is determined
`
`at the time the invention was made. I understand that this temporal
`
`requirement prevents the forbidden use of hindsight. I also understand that
`
`rejections for obviousness cannot be sustained by mere conclusory statements
`
`and that the Petitioners must show some reason why a person of ordinary skill
`
`in the art would have thought to combine particular available elements of
`
`knowledge, as evidenced by the prior art, to reach the claimed invention.” I
`
`also understand that the motivation to combine inquiry focuses heavily on
`
`“scope and content of the prior art” and the “level of ordinary skill in the
`
`pertinent art.”
`
`20.
`
`I have been informed and understand that the obviousness
`
`analysis requires a comparison of the properly construed claim language to
`
`the prior art on a limitation-by-limitation basis.
`
`
`
`7
`
`

`

`IV. Background Topics
`
`21. The prior art references cited in the Petition and the Board’s
`
`Decision (Wang and Kudryavtsev) describe pulses for generating a plasma,
`
`but do not disclose the type of method and apparatus described in the ‘759
`
`patent and its claims.
`
`A. Voltage, current, impedance and power
`
`22. As is commonly known, when a voltage “V” is applied across an
`
`impedance “I,” an electric field is generated that forces a current I to flow
`
`through the impedance. For purely resistive impedance, the relation between
`
`the voltage and the resultant current is given by: V = I * R
`
`23. A common analogy is that voltage is like a pressure that causes
`
`charge particles like electrons and ions to flow (i.e., current), and the amount
`
`of current depends on the magnitude of the pressure (voltage) and the amount
`
`of resistance or impedance that inhibits the flow. The ‘759 patent and the prior
`
`art considered here involve the flow of current through an assembly having a
`
`pair of electrodes with a plasma in the region between them. The effective
`
`impedance of such an assembly varies greatly with the density of charged
`
`particles in the region between the electrodes. Although such an impedance is
`
`more complex than the simple resistive impedance of the above equation, the
`
`general relation is similar: a voltage between the electrode assembly forces a
`
`
`
`8
`
`

`

`current to flow through the plasma, such that the amount of current is
`
`determined by the amplitude of the voltage and the impedance of the plasma.
`
`Thus, the current through the electrode assembly increases with the electrode
`
`voltage and, for a given electrode voltage, the current will increase with a drop
`
`in the impedance of the plasma.
`
`24. The impedance varies with the charge density of the plasma:
`
`With a high density of charged particle the impedance is relatively small, and
`
`with a low density of charge particles the impedance is relatively large.
`
`Simply, the more ions and electrons to carry the charge, the less resistance.
`
`However, the charges and fields react with each other in a very complicated
`
`manner.
`
`25.
`
`In response to the electric field in the region between the
`
`electrodes (i.e., the voltage across the electrodes), all charged particles in the
`
`region (the electrons and positive ions) feel a force that propels them to flow.
`
`This flow is an electric current “I.” Obviously, the amount of current depends
`
`upon the number of charged particles. When there are no charged particles
`
`(i.e., no plasma), there is no current flow in response to the electric field. In
`
`this condition, the impedance of the electrode assembly is extremely high, like
`
`that of an open circuit. But when there is a dense plasma between the
`
`electrodes (with many charged particles), a substantial current will flow in
`
`
`
`9
`
`

`

`response to the electric field. In this condition, the impedance of the electrode
`
`assembly is very low. Thus, in general, the impedance of an electrode
`
`assembly varies greatly with the charge density of the plasma: The impedance
`
`is effectively infinite (an open circuit) when there is no plasma, and is very
`
`low when the charge density is very high.
`
`26.
`
` It is also well known that electric power (P) is the product of
`
`voltage (V) and current (I): P = V * I. This too is a complex relationship.
`
`When the voltage and current are perfectly in phase, then one may simply
`
`multiply them to yield the power. Thus, for a given voltage across an electrode
`
`assembly, the amount of power will depend on the amount of corresponding
`
`current flowing through the electrode assembly. If there is no current flow
`
`(such as when there is no plasma between the electrodes), the power is zero,
`
`even if the voltage across the electrodes is very large. Similarly, at very low
`
`electrode voltages, the power can still be quite high if the current is large.
`
`27. To provide context for understanding this aspect of the ‘759
`
`patent, I consider below some known basic principles of control systems (such
`
`as used in all power supplies and all such control systems) for controlling a
`
`parameter such as voltage amplitude.
`
`B. Control systems
`
`28. The power supply mentioned in the ‘759 patent is an example of
`
`
`
`10
`
`

`

`a control system. This system controls the voltage amplitude of a voltage
`
`pulse. A simplified block diagram of a common feedback control system is
`
`shown the figure below from a text by Eronin.1
`
`
`
`
`
`Figure 1 Control system simplified block diagram
`
`29. The “reference input signal” represents a “desired value” or “set-
`
`point” of the controller. The “forward elements” directly control the
`
`“controlled variable.” In response to the difference between the set-point and
`
`a feedback signal (which represents the condition of the controlled variable),
`
`the forward elements direct the controlled variable in an attempt to reduce the
`
`difference to zero, thereby causing the controlled variable to equal the set
`
`
`1 Ex. 2010, Eronini Umez-Eronini, System Dynamics and Control, Brooks
`
`Cole Publishing Co., CA, 1999, pp. 10-13.
`
`
`
`11
`
`

`

`point value.
`
`30. For example, the set-point for filling a water tank may be 1,000
`
`gallons, or full. The desired value, set-point or desired level is the value “full”
`
`or “1000 gallons.” An open loop control system might just fill the tank for a
`
`pre-calibrated time that result in the tank being full. The control system might
`
`be set to fill the tank once per day based on historical water usage. However,
`
`if water usage is not consistent, the tank may run empty before it is filled, or
`
`may overflow because there was less water usage than normal. On the other
`
`hand, a closed loop system such as shown above uses feedback control. For
`
`example, it measures the water level, and only adds the needed amount. It
`
`might have a switch or sensor that detects when the tank is full, and turns off
`
`the flow of water. The set-point is the desired value. “Here the comparison
`
`of the tank level signal with the desired value of the tank level (entered into
`
`the system as a set-point setting) and the turning of the pump on or off are all
`
`performed by appropriate hardware in the controller.” 2 Further, a closed loop
`
`system could be left on to fill the tank if the level dropped to low. “In feedback
`
`control, a measurement of the output of a system is used to modify its input
`
`
`2 Ex. 2007, Eronini Umez-Eronini, System Dynamics and Control, Brooks
`
`Cole Publishing Co., CA, 1999, pp. 10-13.
`
`
`
`12
`
`

`

`in such a way that the output stays near the desired value3.”
`
`C.
`
`Set point (Controlled Parameter)
`
`31. The parameter that is directed to a desired value is called the
`
`“controlled variable,” as shown in the figure from Eronini. Eronini’s diagram
`
`also shows that while controlling the “controlled variable,” the system may
`
`“manipulate” another control parameter that Eronini calls the “manipulated
`
`variable.”
`
`32. For example, Eronin’s text on control systems shows a control
`
`system that directs the “controlled variable” to its desired value (or “set
`
`point”):
`
`33. Eronin’s diagram also shows that while controlling the
`
`“controlled variable,” the system may “manipulate” another control parameter
`
`that Eronini calls the “manipulated variable.” Another reference by Weyrick
`
`
`
`
`3 Id. at p. 12.
`
`
`
`13
`
`

`

`uses the same terminology as Eronin:
`
` “The controlled output is the process quantity being controlled.”
`
` “The manipulated variable is the control signal which the control
`
`elements process.”4
`
`34. Similarly, Kua and Sinka also show that the “controlled
`
`parameter” is widely understood to mean the parameter being controlled by
`
`the control system.5
`
`
`
`35. With this understanding, I now consider the difference between
`
`controlling the amplitude of a voltage and controlling the power.
`
`D.
`
`Power Control vs Voltage Control
`
`36. To demonstrate the difference between the control of voltage and
`
`the control of power, I will refer to the generic diagram of a feedback control
`
`system from Dorf. In a system for controlling voltage, the set point is a
`
`specified voltage and the “controlled variable” obviously is voltage. Thus, in
`
`a feedback control system as shown in Dorf, a feedback signal representative
`
`of the measured voltage is fed back and compared to the desired voltage level
`
`or “set point.” Based on the difference between the measured voltage and the
`
`
`4 Ex. 2011, Weyrick at 13
`
`5 Ex. 2009, Kua, Automatic Control, Prentice Hall Inc., 1987; Ex. 2006,
`
`Sinha, Naresh, K., Control Systems, Holt, Rinehart and Winston, 1986.
`
`
`
`14
`
`

`

`desired voltage or set point, the forward elements are instructed to drive or
`
`restrain the voltage in an attempt to move the actual voltage to match the
`
`desired voltage.
`
`37.
`
`In a system for controlling power, the set point is a specified
`
`power value and the controlled variable is power. In such a system, the
`
`voltage and/or current can be driven by the feed forward elements to whatever
`
`levels are needed to achieve the target power level. Thus, in the example of a
`
`system for controlling the power of a plasma electrode assembly, if there is
`
`no plasma between the electrodes (and therefore little or no current) a
`
`controller attempting to achieve a target power level will drive the voltage
`
`extremely high in an attempt to achieve the target power P, i.e., P = V * I,
`
`(because I is very low or zero in this situation).
`
`38.
`
` Thus, in a control system for controlling power to a desired set
`
`point, voltage will vary as the controller attempts to achieve the desired power
`
`level (i.e., a desired product of voltage and current). However, the amplitude
`
`of the voltage is not controlled and instead the voltage and/or the current vary
`
`as needed to achieve the desired power.
`
`39. The rise time of a voltage therefore, is a different parameter than
`
`the rise time of power. For example, consider a scenario in which a voltage
`
`source outputs a constant voltage. If that source is connected across an
`
`
`
`15
`
`

`

`impedance that gradually drops, the current will increase as the impedance
`
`drops. Since power is the product of voltage (here a constant) and current, the
`
`power too will rise as the current increases. Thus, in this situation, power rises
`
`at a rate determined by the rate at which the impedance decreases. But there
`
`is no rise in voltage because the source maintains a static, constant voltage at
`
`its output in this example. This demonstrates that a rise time in voltage is a
`
`different parameter than rise time in power.
`
`40. This example can also be used to demonstrate the difference
`
`between a controlled change in the output of a voltage source, and a reaction
`
`to a change in impedance. If the impedance drops so fast that the voltage
`
`source cannot maintain the voltage at its target level, the voltage output by the
`
`source can drop due to limitations of the voltage source. This drop in voltage
`
`is not a controlled drop, caused by the power supply in response to a
`
`programmed change in the voltage set point: It is a transient drop caused by a
`
`change in the impedance load that exceeds the capacity of the voltage source.
`
`E. Magnetron Sputtering History and Operation
`
`41. Since the late 1970s, DC magnetron sputtering has become the
`
`preferred method for the deposition of thin metal films for many applications,
`
`including semiconductor devices and protective layers on cutting tools.
`
`Several significant advantages of this method over alternatives, such as
`
`
`
`16
`
`

`

`thermal evaporation or diode sputter deposition, are higher deposition rate and
`
`improved film structure.
`
`42. The higher deposition rate is possible because the closed loop
`
`magnetic field of the magnetron traps the secondary electrons (produced when
`
`the inert gas ions bombard the metal target that is attached to the cathode
`
`assembly held at a negative voltage of several hundreds of volts). These
`
`electrons gain energy as they are accelerated across the dark space. Since
`
`most of the voltage drop from anode to cathode occurs in this region, the
`
`electrons arrive in the discharge region with more than enough energy to
`
`ionize the neutral gas atoms there. The crossed electric and magnetic fields
`
`create a force on the electrons that causes them to circulate in a path that
`
`follows the shape of the magnetic loop and is only a few mm from the face of
`
`the target. The circulating current in this loop is about 10x the anode-cathode
`
`current of the sputtering discharge. It is these electrons that collide with, and
`
`create large numbers of ions of, the inert neutral sputtering gas atoms (usually
`
`argon) that have diffused into this region. The ions are accelerated toward the
`
`target and bombard it with energies that are nearly the full cathode-anode
`
`voltage. As the secondary electrons create an ion, they lose energy and move
`
`closer to the anode. After several ionizing collisions they no longer have
`
`enough energy to create ions. It is the secondary electrons that sustain a
`
`
`
`17
`
`

`

`normal magnetron discharge.
`
`43. The magnetron discharge is characterized by higher current and
`
`lower voltage (i.e., lower impedance) compared to a diode discharge. This
`
`allows higher powers to be delivered than would be possible with diode
`
`sputtering, because the drop in yield with lower voltage is more than made up
`
`for by the increase in the number of ions. In DC magnetron sputtering,
`
`repeatability of film thickness is usually achieved by operating the power
`
`supply in power control mode and depositing for a specific time.
`
`44. The sputtered metal atoms are ejected from the target with high
`
`velocity, compared to evaporation, which contributes to film adhesion and
`
`microstructure. However, this high velocity means that relatively few of the
`
`metal atoms have a chance to become ionized as they traverse the thin zone
`
`of high energy electrons on their way from the source target to the substrate
`
`workpiece (e.g., silicon wafer or razor blade).
`
`45. As research progressed over the ensuing decades, the advantages
`
`of increasing the ionization of the sputtered atoms became evident. Ions
`
`impacting the growing film improved qualities such as hardness, adhesion and
`
`density even further. Furthermore, the trajectories of the incoming ions could
`
`be made more perpendicular to the substrate surface by application of a
`
`negative bias. This allowed more film to be deposited in the bottom of high
`
`
`
`18
`
`

`

`aspect ratio holes enabling the production of semiconductor devices with
`
`ever-decre