`____________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`___________________
`
`
`COALITION FOR AFFORDABLE DRUGS VI LLC
`
`PETITIONER
`
`V.
`
`GRÜNENTHAL GMBH
`
`PATENT OWNER
`
`___________________
`
`CASE NO.: UNASSIGNED
`PATENT NO. 7,994,364
`FILED: DECEMBER 10, 2009
`ISSUED: AUGUST 9, 2011
`INVENTORS: ANDREAS FISCHER, ET AL.
`
`TITLE: CRYSTALLINE FORMS OF (-)-(1R,2R)-3-(3-DIMETHYLAMINO-1-
`ETHYL-2-METHYLPROPYL)-PHENOL HYDROCHLORIDE
`___________________
`
`DECLARATION OF WILLIAM E. MAYO, Ph.D.
`
`
`
`
`
`
`RS 1012 - 000001
`
`
`
`Abbreviations and Nomenclature
`
`2θ
`
`Δ2θ
`
`Bragg angle, defined as the angle between the detector position
`and the incident X-ray beam
`
`Error in peak position between the same peak in the reference
`sample and the test sample
`
`│Δ2θ│
`
`Absolute value of the peak position error
`
`<│Δ2θ│>
`
`Average of all │Δ2θ│ values
`
`d(hkl)
`
`Spacing d between sets of hkl parallel planes
`
`λ
`
`hkl
`
`X-ray wavelength
`
`Set of Miller Indices defining a specific crystal plane
`
`Bragg’s Law Law/equation governing the geometry of diffraction
`
`COD
`
`CSD
`
`Crystallographic Open Database
`
`Cambridge Structure Database
`
`Excipient
`
`Inactive ingredients in the final pharmaceutical product
`
`FOM
`
`Smith Snyder Figure of Merit (range of 0 to 999)
`
`ICDD
`
`International Centre for Diffraction Data (successor organization
`to JCPDS and the ASTM's sponsored consortium Joint Committee
`for Chemical Analysis by Powder Diffraction Methods)
`
`ICSD
`
`Inorganic Crystal Structure Database
`
`Jade
`
`X-ray data analysis program produced by Materials Data Inc.
`Comparable programs are Eva (Bruker) and HighScore
`(Panalytical)
`
`Miller Index /
`Miller Indices
`
`Numerical nomenclature used to represent planes and directions in
`crystals
`
`
`Powder Diffraction File published by the ICDD
`
`
`
`2
`
`RS 1012 - 000002
`
`
`
`Polycrystal
`
`Solid, bulk material made from numerous small grains
`(“crystallites”)
`
`WPF
`
`XRPD
`
`Whole Pattern Fitting
`
`X-ray powder diffraction (a coherent scattering process from a
`periodic array of atoms)
`
`3
`
`
`
`
`
`
`
`
`
`
`
`RS 1012 - 000003
`
`
`
`I, William E. Mayo, declare as follows:
`
`I.
`
`PROFESSIONAL BACKGROUND
`
`1.
`
`I, William E. Mayo, submit this expert declaration on behalf of the
`
`Coalition for Affordable Drugs, in support of its petition for inter partes review of
`
`U.S. Patent No. 7,994,364.
`
`2.
`
`
`
`I make this declaration based on my personal knowledge,
`
`consideration of the materials I discuss herein, and my expert opinions.
`
`
`3.
`
`I am a Professor Emeritus at Rutgers – The State University of New
`
`Jersey and Co-Founder and Chief Scientist at H&M Analytical Services.
`
`
`4.
`
`I have a Ph.D. in Materials Science, and throughout my academic and
`
`professional career, I have focused on the study and practice of X-Ray Powder
`
`Diffraction (“XRPD”).
`
`
`5.
`
`I received a Bachelor of Science Degree in Physics from Carnegie
`
`Mellon University in 1971, after which I worked at Harry Diamond Laboratories as
`
`a Physicist. I received a Master’s Degree in Metallurgy and Materials Science from
`
`Carnegie Mellon in 1974 and subsequently worked as a metallurgist at Olin Metals
`
`Research from 1974 to 1977, and at TRW from 1977 to 1978.
`
`
`6.
`
`In 1978, I began graduate studies at Rutgers University, where I
`
`received a Ph.D. in Mechanics and Materials Science in 1982. My doctoral studies
`
`focused on structural and failure analysis of materials deformed by fatigue and
`
`
`
`4
`
`RS 1012 - 000004
`
`
`
`stress corrosion cracking; development of computerized x-ray testing methods; and
`
`modeling of hardening mechanisms and phase transformations. I also served as a
`
`Post-Doctoral Fellow at Bell Labs in 1982.
`
`
`7.
`
`I joined the faculty at Rutgers University in the Department of
`
`Mechanics and Material Science in 1982 as an Assistant Professor and then
`
`progressed to Associate Professor in 1988, Full Professor in 1995, and Emeritus
`
`Professor in 2008. Through a series of mergers and name changes, I have been
`
`associated with the Department of Mechanics and Material Science, Department of
`
`Ceramic Engineering, Department of Ceramic and Materials Engineering, or
`
`Department of Materials Science and Engineering during my 26-year academic
`
`career.
`
`
`8.
`
`Prior to the start of my academic career, I was a Physicist in the
`
`Microminiature Branch at Harry Diamond Research Laboratory (Army Material
`
`Command). From 1974 until 1977, I was a Metallurgist in the Physical Metallurgy
`
`Section at Olin Metals Research Laboratory. From 1977 until 1978, I was a
`
`Metallurgist in the Materials Engineering Section at TRW (Reda Pump Division).
`
`My primary responsibilities focused on materials R&D.
`
`
`9.
`
`I have also co-founded four companies (H&M Analytical Services,
`
`NanoPac, XStream Systems, and Veracity Networks) in the private sector that
`
`were closely linked to my Rutgers research and experiences. X-Stream and
`
`
`
`5
`
`RS 1012 - 000005
`
`
`
`Veracity focused on the detection of counterfeit pharmaceuticals utilizing a novel
`
`XRPD method that I developed with FAA funding for rapid detection of
`
`explosives hidden inside checked baggage. NanoPac commercialized work that I
`
`had done with NSF and DOD funding to develop novel nanoscale materials via a
`
`controlled transformation route of metastable starting materials. Finally, H&M
`
`Analytical Services was founded as a consulting and testing company to take
`
`advantage of my nearly 50 years of experience in XRPD methods. I have been
`
`involved with these companies on a part-time (1997 – 2008) or full time (2008 –
`
`present) basis for more than 18 years.
`
`10.
`
` Through H&M Analytical Services, I have consulted for nearly 500
`
`different customers ranging from government labs (e.g., Brookhaven, CDC,
`
`DARPA, EPA, FAA, FDA, JPL, NASA, TVA, and Sandia), to Universities (e.g.,
`
`BU, Colo. State, Dartmouth, MIT, Northwestern, Ohio State, Purdue, RPI,
`
`Rutgers, Stevens, Tufts, U. Mass, U. Virginia, Washington Univ., and Yale), large
`
`companies (e.g., Caterpillar, Colgate Palmolive, Du Pont, EXXON, Fujitsu, GE,
`
`Georgia Pacific, Gillette, Honeywell, IBM, J&J, L-3, Libbey, Lockheed Martin, M
`
`& M Mars, Marathon Oil, Monsanto, Northrup Grumman, Perkin Elmer, Pfizer,
`
`Raytheon, Sandoz, and United Technologies), small companies, testing labs, and
`
`individuals.
`
`11.
`
` As part of my academic duties at Rutgers, I taught nineteen different
`
`
`
`6
`
`RS 1012 - 000006
`
`
`
`courses in various aspects of material characterization and general aspects of
`
`Material Science, all of which are itemized in Exhibit 1013. Of particular relevance
`
`to this action are three undergraduate courses (150:254, 150:309 & 150:408) and
`
`five graduate courses (150:520, 150:521, 150:524, 150; 563, and 655:xx (an
`
`experimental course)) dealing with various aspects of basic and advanced aspects
`
`of XRPD. I have also provided training in XRPD at the post-graduate level.
`
`Finally, I supervised three different XRPD labs where I was responsible for
`
`purchasing, maintaining, instructing, and supervising all aspects of the labs. In
`
`toto, I have utilized more than 40 different X-ray diffractometers and personally
`
`analyzed well in excess of 100,000 diffraction patterns during my career.
`
`
`12.
`
`In addition to my teaching duties related to XRPD, I have been very
`
`active in providing services to the XRPD community. In recognition of this fact, in
`
`2006 I was awarded the title of Fellow by the International Centre for Diffraction
`
`Data (“ICDD”), which is the world’s foremost source of X-ray powder diffraction
`
`data. This honor was bestowed on me in recognition of my 18 years as an Editor
`
`for the Powder Diffraction File (“PDF”), 8 years as an Editor for New Data for the
`
`Journal Powder Diffraction and 35+ years of effort in advancing the development
`
`of XRPD techniques.
`
`13.
`
` My experiences with the ICDD include an important role in the
`
`editorial review of the PDF patterns submitted by scientists from around the world,
`
`
`
`7
`
`RS 1012 - 000007
`
`
`
`as well as contribution of new patterns. Also, through my participation on
`
`Technical Committees as a member and Chair, I am experienced in the interactions
`
`that the ICDD and its predecessor organizations have had with governmental
`
`agencies such as the National Institute of Standards (NIST) with regard to the
`
`development and publication of powder diffraction data.
`
`
`14.
`
`I have conducted numerous research projects in Materials Science,
`
`primarily dealing with a) polymorphic phase transformation in metastable
`
`materials, b) development of new X-ray analytical techniques, and c) development
`
`of new applications for nanoscale materials. This research was funded by various
`
`private and government sources and totaled more than $22,000,000. Even though I
`
`have retired from academic life, I still continue to do research as a research affiliate
`
`at Rutgers through a DARPA project to develop the next generation of ceramic
`
`armor via phase transformation of metastable materials.
`
`
`15.
`
`I have numerous publications including 8 books written, 25
`
`monographs edited, 4 book chapters, 3 instructional texts, 133 refereed articles, 83
`
`archival abstracts, 100 conference presentations, and 6 patents/applications. Most
`
`of these publications focus on the use or development of XRPD methods. In
`
`addition, a large body of work conducted by my group to develop a new XRPD
`
`method to detect explosives on airplanes could not be published due to national
`
`security concerns.
`
`
`
`8
`
`RS 1012 - 000008
`
`
`
`
`16.
`
`In addition to teaching, research, publications, and editorial work
`
`related to XRPD, I have designed and constructed numerous XRPD instruments
`
`for the FAA, DHS, X-Stream, Veracity, L-3 Communications, and Rutgers. Much
`
`of this work revolved around the design of advanced algorithms and neural
`
`networks for phase identification of various crystalline materials, including their
`
`polymorphs.
`
` During my 26-year career at Rutgers, I taught computer programming
`17.
`
`to approximately 14,000 engineering and science undergraduate students and
`
`various XRPD and general Materials Science courses to approximately 1,000
`
`undergraduate, graduate and post-graduate students. I also directly supervised the
`
`Ph.D. research of 10 students and trained/advised many other Ph.D. candidates
`
`(approximately 25) who were using XRPD in their thesis or post-doctoral research.
`
`18.
`
` My research has resulted in the filing of six patent applications, three
`
`of which resulted in issued patents. Of particular note are: Combinatorial
`
`Contraband Detection Using Energy Dispersive X-Ray Diffraction (U.S. Patent
`
`Application No. 2006/0104414) and Analysis Methods for Energy Dispersive X-
`
`Ray Diffraction Patterns (U.S. Patent No. 6,118,850).
`
` Based on my education, practical training, teaching, research,
`19.
`
`equipment design, editorial work, patents, consulting, and industrial experience, I
`
`consider myself an expert in the area of X-ray diffraction, material
`
`
`
`9
`
`RS 1012 - 000009
`
`
`
`characterization, use of XRPD databases for general phase identification and
`
`quantification, with a special emphasis on detection of polymorph phases.
`
` My qualifications are further detailed in my curriculum vitae, a copy
`20.
`
`of which is attached hereto as Exhibit 1013.
`
`
`21.
`
`I have been asked to provide my opinions and views on the
`
`patentability of U.S. Patent No. 7,994,364 based upon my review and analysis of
`
`this literature, as well as my education, training, and experience in XRPD, and
`
`testing of tapentadol HCl polymorphs.
`
`II.
`
`FEES
`
`22.
`
`I have no financial interest in the outcome of this litigation. I invoice
`
`at a rate of $375 per hour.
`
`III. MATERIALS REVIEWED
` The opinions and the statements I make in this declaration are based
`23.
`
`on my personal knowledge, testing of the tapentadol HCl samples, and professional
`
`experience. In addition, I rely on and incorporate by reference the documents and
`
`information cited in the declaration itself and listed below:
`
`
`24.
`
`I have reviewed certain literature and patents pertaining to crystalline
`
`forms of tapentadol hydrochloride: U.S. Patent No. 7,994,364 (“the ’364 patent”)
`
`(Ex. 1001); 4-18-08 Letter from Grünenthal to EPO (Ex. 1004); EP 0 693 475 (Ex.
`
`1006); WO 03/935953 (“Bartholomaeus”) (Ex. 1009); U.S. Patent No. 6,248,737
`
`
`
`10
`
`RS 1012 - 000010
`
`
`
`B1 (Ex. 1017); U.S. Patent No. 6,344,558 B1 (Ex. 1018); EP 1612203 (Ex. 1022);
`
`H.P. Klug & L.E. Alexander, “X-Ray Diffraction Procedures”, 2nd Ed., Wiley,
`
`Chapter 7, pp 505-531 (Ex. 1023); D. Krawitz, “Introduction to Diffraction in
`
`Materials Science and Engineering”, Wiley (2001), Ch. 8, pp 215-234 (Ex. 1024);
`
`F.H. Chung & D.K. Smith, “Industrial Applications of X-Ray Diffraction”,
`
`Dekker, (2000), p21 (Ex. 1025); V. K. Pecharsky & P.Y. Zavalij, “Fundamentals
`
`of Powder Diffraction and Structural Characterization of Materials”, 2nd Ed.,
`
`Springer (2009) pp. 380-382 (Ex. 1026); J.D. Hanawalt, “Phase Identification by
`
`X-Ray Powder Diffraction Evaluation of Various Techniques”, Adv. X-ray
`
`Analysis., v.20 (1976) pp.63-73 (Ex. 1027); R.L. Snyder, “A Hanawalt Type Phase
`
`Identification Procedure for a Minicomputer”, Adv. In X-ray Analysis, v.24 (1980)
`
`pp. 83-90 (Ex. 1028); D.K. Smith, S.Q. Hoyle & G.G. Johnson, “Phase
`
`Identification Using Whole-Pattern Matching”, Adv. X-Ray Analysis, v. 36 (1993)
`
`pp. 287-299 (Ex. 1029); B.D. Cullity, “Elements of X-ray Diffraction”, 2nd Ed,
`
`Addison Wesley, (1978), p.402 (Ex. 1030); C. Suryanarayana & M.G. Norton, “X-
`
`Ray Diffraction – A Practical Approach”, Plenum, (1998), p. 240 (Ex. 1031); J.
`
`Faber, C.A. Weth & J. Bridge, “A Plug-in Program to Perform Hanawalt or Fink
`
`Search-Indexing Using Organics Entries in the ICDD PDF-4/Organics 2003
`
`Database”, Adv. XRay Analysis, v. 47 (2004) pp166-173 (Ex. 1032); H.G.
`
`Brittain, Ed. “Polymorphism in Pharmaceutical Solids”, Marcel Dekker, 1999, p.
`
`
`
`11
`
`RS 1012 - 000011
`
`
`
`236 (Ex. 1034); V. K. Pecharsky & P.Y. Zavalij, “Fundamentals of Powder
`
`Diffraction and Structural Characterization of Materials”, 2nd Ed., Springer (2009)
`
`p 385 (Ex. 1033); Powder Diffraction File, published annually by the International
`
`Centre for Diffraction Data; D. Krawitz, “Introduction to Diffraction in Materials
`
`Science and Engineering”, Wiley (2001), Ch. 8, pp 269-277 (Ex. 1036);V. K.
`
`Pecharsky & P.Y. Zavalij, “Fundamentals of Powder Diffraction and Structural
`
`Characterization of Materials”, 2nd Ed., Springer (2009) pp. 524-545 (Ex. 1037);
`
`R.E. Dinnebier & S.J.L. Billinge, “Powder Diffraction – Theory and Practice”,
`
`RSC Publishing, 2008, pp. 266-281 (Ex. 1038); Website www.wikipedia.com;
`
`Website www.intechopen.com; Website http://mutuslab.cs.uwindsor.ca; Website
`
`http://prism.mit.edu/x-ray; G.S. Smith & R.L. Snyder, “A Criterion for Rating
`
`Powder Diffraction Patterns and Evaluating the Reliability of Powder Pattern
`
`Indexing”, J. Appl. Cryst., v. 12 (1979) pp. 60-65 (Ex. 1035); G.S. Pawley, Unit-
`
`Cell Refinement from Powder Diffraction Scans, J. Appl. Cryst., 14 (1981) 357-
`
`361 (Ex.1039); R.E. Dinnebier & S.J.L. Billinge, “Powder Diffraction – Theory
`
`and Practice”, RSC Publishing, 2008, pp. 153-159 (Ex. 1040);“X-ray Diffraction”,
`
`the United States Pharmacopeia <941>, USP 38/NF33, (2015) p. 10/10 (Ex. 1021).
`
`
`25.
`
`I also rely on my nearly 50 years of experience in teaching, research,
`
`and consulting, my considerable hands-on experience in XRPD testing and
`
`equipment design, my education and training in the area of X-ray diffraction that
`
`
`
`12
`
`RS 1012 - 000012
`
`
`
`relate to phase and polymorph identification of crystalline and amorphous
`
`materials.
`
`IV. LEVEL OF ORDINARY SKILL IN THE ART
` A person of ordinary skill in the art (“POSA”) in connection with the
`26.
`
`’364 patent would typically have a Ph.D. in fields relevant to small molecule drug
`
`development, such as biochemistry, medicinal chemistry, organic chemistry, or the
`
`equivalent, or a bachelor’s degree in the same field(s) with four to six years of
`
`practical experience.
`
`V.
`
`THE MEANING OF SELECTED TERMS IN THE CLAIMS OF THE
`’364 PATENT
`
`27.
`
`It is my understanding that the claim terms in a patent subject to IPR
`
`must be understood in their broadest reasonable interpretation in light of the
`
`specification of the patent.
`
` The terms in the claims of the ’364 patent are used in accordance with
`28.
`
`their plain and ordinary meaning, as exemplified by the terms presented below.
`
`
`29.
`
`It is my opinion that a POSA would have understood that the terms in
`
`Claims 1–4 and 24–27 are plain on their face. I have given the terms their plain and
`
`ordinary meaning under a broadest reasonable interpretation in light of the
`
`specification.
`
`
`30.
`
`I note that because the ’364 patent does not mention of the purity of
`
`the crystalline compound recited in claims 1–4 or 24–27, a POSA would not
`
`
`
`13
`
`RS 1012 - 000013
`
`
`
`attribute any particular level of purity to the crystalline compound recited in claims
`
`1–4 or 24–27.
`
`VI. LEGAL STANDARDS GOVERNING ANTICIPATION
`
`I understand that an anticipation analysis involves comparing a claim
`31.
`
`to the prior art to determine whether a POSA would anticipate the claimed
`
`invention in view of the prior art, and in light of the general knowledge in the art. I
`
`also understand that to anticipate a claim, a prior art reference must disclose each
`
`and every claim limitation, either expressly or inherently. I further understand that
`
`to explain the meaning of a prior art reference, a POSA can refer to a secondary
`
`reference or to the knowledge of one of ordinary skill in the art.
`
`VII. BACKGROUND SCIENTIFIC INFORMATION
` Background material describing the process of phase identification by
`32.
`
`XRPD is given in the following sections.
`
`A. Crystals, Crystal forms, and Polymorphs
` Many natural and man-made materials are considered “crystalline,” as
`33.
`
`opposed to “amorphous” or non-crystalline. Although there are not always precise
`
`boundaries between “crystalline” and “amorphous” materials, a “crystalline”
`
`material is one that has a crystal structure, meaning a structure in which all of the
`
`atoms arrange themselves in a periodic and predictable way. By way of illustration,
`
`the simple structure of MgB2 is shown in Figure A:
`
`
`
`14
`
`RS 1012 - 000014
`
`
`
`
`
`Figure A: (www.intechopen.com – J. Nagamatsu, Nature 410(2001)63).
`
`34.
`
` Crystalline materials have a high degree of order, i.e. the atoms form a
`
`regular, periodic structure in three dimensions and are located in predictable
`
`positions. But, when the atoms become disordered, an amorphous material forms,
`
`as demonstrated in Figure B:
`
`Crystalline SiO2
`
`Noncrystalline SiO2
`
`
`
`Figure B: (W. D. Callister, “Material Science and Engineering”, 5th Ed.,
`
`Wiley, 2000, p.58).
`
` Many crystalline materials will exhibit multiple crystalline forms,
`35.
`
`which are termed polymorphs (for multi-element compounds) or allotropes (for
`
`elements). One common example is shown below in Figure C, where carbon atoms
`
`form a variety of crystal structures.
`
`
`
`
`
`15
`
`RS 1012 - 000015
`
`
`
`Figure C: Various
`carbon allotropes a)
`diamond, b) graphite, c)
`lonsdalite, d – f)
`fullerenes, g)
`amorphous, h) single
`wall nanotube (Source:
`Wikipedia.org).
`
`
`
`
`
` Bulk, solid materials typically consist of individual grains in which
`36.
`
`the same crystal structure exists in each grain but are rotated or twisted with
`
`respect to their neighbors and is referred to as a “polycrystalline” material (e.g.
`
`individual grains that make up rock candy as shown in Figure D).
`
`Figure D: Individual
`grains of sugar in rock
`candy
`
`
`
`B. X-Ray Powder Diffraction Testing
` X-ray powder diffraction (“XRPD”) has been in use for almost a
`37.
`
`century and is one of the most useful tools for studying crystalline materials and
`
`determining polymorphic forms. With the advent of fast and cheap computers, and
`
`
`
`16
`
`RS 1012 - 000016
`
`
`
`advances in electronics and powerful X-ray sources, XRPD has moved to the
`
`forefront for characterizing a wide variety of materials. Since this method is also
`
`the key tool used in the ’364 Patent to identify the various polymorphs of
`
`tapentadol HCl, a review of the basics of XRPD may prove useful.
`
` X-rays are high-energy photons that have a wavelength comparable to
`38.
`
`the atomic spacing in solid materials. As a result, X-rays are able to scatter from
`
`these materials in a special way called diffraction that provides useful information
`
`about the crystalline structure. This process is shown schematically in Figures E(i)
`
`and (ii).
`
` (i) Destructive interference
`
`
`
`
` (ii) Constructive interference
`
`
`
`Figure E: Scattering from a periodic array of atoms leading to diffraction1
`
` The incident X-ray beam (i.e., the beam emanating from the X-ray
`39.
`
`tube) appears on the left side of Figure E(i) and interacts with parallel rows of
`
`atoms separated by a distance dhkl. Each atom scatters the X-ray beam, and these
`
`scattered waves in general will destructively interfere with each other and produce
`
`1 (http://mutuslab.cs.uwindsor.ca).
`
`
`
`17
`
`RS 1012 - 000017
`
`
`
`no useful information. But at a few unique angles, the scattered waves will add to
`
`produce a diffracted beam as shown in Figure E(ii).
`
` To collect these diffracted beams, a diffractometer is a conceptually
`40.
`
`simple device that is widely used and is illustrated in Figure F. In this device, the
`
`X-ray tube and the detector are located on either side of the sample to be tested. In
`
`the most common diffractometer design, the tube and detector rotate around a
`
`common axis to collect the spectrum and the scattered X-ray intensity will be
`
`collected as a function of the angle 2θ, defined as the angle between the incident
`
`and diffracted beams.
`
`
`
`
`
`X-ray tube
`
`Detector
`
`
`
`
`
`
`
`
`
`Figure F: Schematic diagram of a powder X-ray diffractometer2
`
` For a typical sample, the diffraction pattern will consist of a series of
`41.
`
`peaks separated by regions of low background intensity as shown in Figure G.
`
`
`2 (http://prism.mit.edu/x-ray).
`
`
`
`18
`
`RS 1012 - 000018
`
`
`
`2
`
`2
`
`2
`
`
`
`Figure G: Illustration that only a few peaks contribute to the diffraction
`pattern at any one time, depending on the grain orientations3
`
` At the lowest angle (≈20.5° in this example), a small number of the
`42.
`
`crystals making up the sample will be in the proper orientation to produce a strong
`
`diffraction signal. But, once the diffractometer has rotated past that angle, the
`
`intensities are very low until the second peak begins (≈29.2°) and so forth until the
`
`entire pattern has been collected. And, each peak has a set of indices in text above
`
`it (called Miller Indices) to indicate which set of crystal planes give rise to that
`
`peak.
`
` The relationship between the separation distance dhkl of a set of hkl
`43.
`
`planes and the diffraction angle 2θ is given by Bragg’s Law:
`
`nλ = 2dhkl sin(θhkl)
`
`where n = an integer value 1, 2, ...
`
`
`3 Ibid
`
`
`
`19
`
`RS 1012 - 000019
`
`
`
`
`
`
`
`
`
` λ = X-ray wavelength
`
` dhkl = spacing between parallel planes with hkl Miller Indices
`
` θhkl = one half of the diffraction angle.
`
` The experimental procedure is to collect the diffraction pattern over a
`44.
`
`wide range to capture a sufficient number of peaks and to measure the angular
`
`positions and intensities of each peak in the pattern. Then, using Bragg's law, each
`
`peak position is used to convert the 2θ value into the corresponding d values that
`
`describe the various crystalline planes in the test sample.
`
` The intensity axis in XRPD patterns is usually represented with an
`45.
`
`arbitrary scale, and the intensities are typically reported on a relative scale of 0 to
`
`100%, with the strongest peak set at 100%. Any peak below 1% intensity,
`
`however, is likely to be buried in the experimental noise and difficult to measure
`
`accurately. Historically, peaks whose intensity are less than three times the
`
`background noise (known as the three sigma (3σ) rule) usually cannot be reliably
`
`differentiated from the background signal. Therefore, these very weak peaks are
`
`usually ignored.
`
` The diffraction peaks are usually sharp, permitting accurate
`46.
`
`determination of peak position, with an accuracy of about 0.005˚ or better. A
`
`typical report of diffraction data contains diffraction angle 2θ, the d value and the
`
`
`
`20
`
`RS 1012 - 000020
`
`
`
`relative intensity for each sufficiently intense peak observed in the diffraction
`
`pattern, as illustrated in Table 1 for “Form A” of tapentadol HCl taken from the
`
`’364 patent.
`
`Table 1 Diffraction Data for Tapendatol HCl “Form A”
`
`Peak No.
`
`2θ (°)
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`9.07
`
`10.11
`
`14.51
`
`15.08
`
`15.39
`
`15.69
`
`15.96
`
`16.62
`
`17.00
`
`18.24
`
`18.88
`
`20.00
`
`20.39
`
`21.66
`
`22.54
`
`24.27
`
`d (Å)
`
`9.750
`
`8.749
`
`6.104
`
`5.875
`
`5.757
`
`5.648
`
`5.553
`
`5.334
`
`5.215
`
`4.864
`
`4.700
`
`4.439
`
`4.355
`
`4.103
`
`3.945
`
`3.667
`
`Intensity (%)
`
`10
`
`9
`
`100
`
`24
`
`11
`
`22
`
`24
`
`13
`
`20
`
`63
`
`28
`
`23
`
`47
`
`47
`
`41
`
`28
`
`
`
`21
`
`RS 1012 - 000021
`
`
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`29
`
`30
`
`31
`
`32
`
`25.03
`
`25.47
`
`25.84
`
`26.04
`
`26.94
`
`27.29
`
`27.63
`
`28.33
`
`28.72
`
`29.09
`
`29.29
`
`29.76
`
`30.37
`
`30.74
`
`31.70
`
`34.37
`
`3.557
`
`3.497
`
`3.448
`
`3.422
`
`3.309
`
`3.268
`
`3.228
`
`3.150
`
`3.108
`
`3.070
`
`3.049
`
`3.002
`
`2.943
`
`2.908
`
`2.823
`
`2.609
`
`13
`
`43
`
`20
`
`27
`
`13
`
`29
`
`28
`
`20
`
`12
`
`12
`
`21
`
`11
`
`23
`
`11
`
`14
`
`11
`
`Note: “d” spacings calculated using λ = 1.54178 Å
`
`(Ex. 1001 at 8:20-52.)
`
`C.
`
`Interpreting XRPD Results
` The most common use of XRPD of powders is to identify the
`47.
`
`crystalline phases present in a material. Also, if the phase has multiple polymorphs,
`
`it is important to determine which polymorph is present. This process is generally
`22
`
`
`
`RS 1012 - 000022
`
`
`
`referred to simply as phase identification or “phase ID,” which are used here
`
`interchangeably. For the present analysis, XRPD will be used to identify which
`
`polymorph of tapendatol HCl is present in various products.
`
` Once a diffraction pattern has been collected, certain information must
`48.
`
`be extracted from the pattern to enable a search of the known phases and
`
`polymorphs (See Ex. 1023; Ex. 1024). By far, the most common set of features
`
`used for phase ID are the position, d, and relative intensity, I, of each peak of
`
`sufficient intensity that can be located reliably. This information is termed the d/I
`
`pair (See Ex. 1025), an example of which is given above in Table 1.
`
` After the feature set has been extracted, a subset of those data is then
`49.
`
`compared to the library of known phases and polymorphs that have been archived
`
`in databases such as the Powder Diffraction File (PDF), Inorganic Crystal
`
`Structure Database (ICSD), Cambridge Structural Database (CSD),
`
`Crystallographic Open Database (COD), and Pearson Crystal Database (PCD),
`
`among others (See Ex. 1026). Other sources of phase information may be patents,
`
`the open literature, private corporations, and universities. The most commonly
`
`used database is the PDF published by the ICDD, which has developed two
`
`popular search/match algorithms known as the Hanawalt and Fink methods to
`
`identify a phase (See Ex. 1027; Ex. 1028; Ex. 1029).
`
`50.
`
` The recommended method for identifying a phase or polymorph has
`
`
`
`23
`
`RS 1012 - 000023
`
`
`
`been in place for approximately 60 years and is described in numerous XRPD texts
`
`(e.g. Exs. 1023-1033). The recommended procedure consists of the following
`
`steps:
`
`
`1.
`
`Use the position of the three strongest diffraction lines to isolate
`
`a number of candidate phases.
`
`
`2.
`
`After the closest match has been found, reduce the number of
`
`candidate phases by comparing the experimental intensities
`
`with those for the candidate phase given in the database.
`
` When good agreement has been found for the three strongest
`3.
`
`lines, compare the entire experimental pattern with the database
`
`reference pattern.
`
` When full agreement is obtained, identification is generally
`4.
`
`complete. (Exs. 1030-1031).
`
` With regard to the first step, there may be some variation in the
`51.
`
`methods for the first round of screening. The Hanawalt method uses the three
`
`strongest lines in an experimental diffraction pattern to identify the top candidates
`
`for a phase match. (See Ex. 1031.) Since the three strongest lines may have some
`
`variation due to experimental errors and/or preferred orientation, the order of the
`
`three lines are permuted. For example, if the three strongest lines are d1, d2 and d3,
`
`searches are conducted for matches to (d1 d2 d3), (d2 d1 d3), (d3 d2 d1) and so forth.
`
`
`
`24
`
`RS 1012 - 000024
`
`
`
`By contrast, the Fink method places less reliance on peak intensities and uses the
`
`eight strongest lines for the search (See Ex. 1032). Both the Hanwalt and Fink
`
`methods predate widespread use of computers, but variants of these methods still
`
`serve as the basis for automated, computer-based search methods.
`
`52.
`
` Regardless of which method is used, the search results are only a
`
`preliminary finding, and the careful analyst must still match the full experimental
`
`pattern with all of the lines from the reference database.
`
` Turning to the remaining steps in the phase ID process, when
`53.
`
`comparing the full experimental pattern to the full reference pattern, the match of
`
`the intensity profiles is very important (See Ex. 1023; Ex. 1024; Ex. 1026; Exs.
`
`1029-1033). Although the individual peak intensities are less accurate than the
`
`peak positions, there should be general agreement between the test and reference
`
`patterns. This is best expressed by the observations by Pecharsky & Zavalij (Ex.
`
`1033):
`
`When there are a few strong reflections in the database record, all
`should be present in the analyzed experimental pattern. When even
`one of the strong peaks is missing in the analyzed pattern, or it is
`present but has very low intensity, this match is likely incorrect,
`unless an extremely strong preferred orientation is possible in either
`pattern (but not in both) and there is a legitimate reason for the two to
`be different.
` As a final, critical step in the phase identification process, it is
`54.
`
`
`
`25
`
`RS 1012 - 000025
`
`
`
`customary to compute a Figure of Merit (“FOM”) to quantitatively assess the
`
`proposed match (See Ex. 1028; Ex. 1029; Ex. 1032). This is routinely done for all
`
`of the patterns in the PDF database with the aid of the Smith-Snyder FOM (See Ex.
`
`1035), which is the ratio of the fraction of observable lines that are actually found,
`
` (cid:1832)(cid:1841)(cid:1839)(cid:4666)(cid:1840)(cid:4667)(cid:3404)
`
`divided by the average error in peak positions.
`
`(cid:2869)(cid:2996)|(cid:2940)(cid:2870)(cid:2968)|(cid:2997) (cid:3015)(cid:3015)(cid:3291)(cid:3290)(cid:3294)(cid:3294)
`
`where,
` N
`
`
`
`
`FOM(N) is the figure of merit for N observed lines,
`
`
`poss is the number of observable lines, and
`<│Δ2θ│> is the average peak positional errors.
`
`
`
` By convention, the range for the Smith-Snyder FOM goes from 0 to
`55.
`
`999 and high quality experimental patterns have typical values of 100 (inorganic
`
`phases) or 50 (organic phases) based on my experience of 18 years as an editor for
`
`the PDF, 8 years as an editor for the Journal Powder Diffraction and 37 years of
`
`involvement with the ICDD. When considering a pattern for inclusion in the PDF
`
`database in my editorial function, I used a minimum FOM value of 10 before I will
`
`recommend acceptance of a pattern. And for patterns with an FOM between 10 and
`
`15, I would often recommend that the pattern be assigned a low quality rating.
`
`Thus, the FOM is often used to assess the reliability of a match between the
`
`proposed phase and the experimental data.
`
` Outside of the editorial process within the ICDD organization and its
`56.
`
`PDF database, the FOM is also used by analysts in the lab when attempting to
`
`
`
`26
`
`RS 1012 - 000026
`
`
`
`match their own experimental diffraction pattern against potential matches.
`
`Analysis programs such as Jade, will use a proprietary FOM to create a ranking of
`
`proposed hits in a search/match request and put the most likely hit at the top of the
`
`list and the least likely