I I
`
`ENGII'lt:E.RING &
`\YIATI-IE.M :nc.M •.
`scn:.NCES WBR~IW
`
`Thermals
`Page 364 praying
`
`BASF Exhibit 2015, Page 1 of 18
`
`SNF Holding Company et al v BASF Corporation, IPR2015-00600
`
`

`
`CHECKLIST
`
`I:HEmTEI:H
`
`JUNE 1982
`
`... And celebrated it is; but the pyrotechnics are all chemistry,
`says Girone.
`
`Dowsing '82
`Two more real life stories round out Sheridan's view of "the next limiting
`resource."
`
`"We have been retained by a Fortune 500 firm to locate a top-flight
`technologist who ... " Whitaker explains where such letters originate
`and how some end.
`
`How to address the economic long wave is addressed by Assistant
`Secretary of Commerce Designate, Merrifield.
`
`In his Bristol Meyers Award address, front-line fighter Skipper goes to
`strategy ...
`
`. .. while his coworker Montgomery explains tactics.
`
`In view of its complexity, the best bet for coal, Frank shows
`quantitatively, is to tear it down to CO/H 2 and go from there .
`methanol.
`
`: via
`
`It's called thermal spraying and Thorpe shows how to use it to
`fix surfaces.
`
`Silicon in organic chemistry? Dunogues provides another string to the
`synthetic chemist's tool box.
`
`There is life beyond the three-neck flask and resin kettle, Gerrens
`explores it.
`
`356 The Science of the Possible
`in which Donley addresses the Council D in which Weisz wonders at the shape of
`
`VIew from the Top
`
`for Chemical Research.
`
`things.
`
`BOOOOMI
`
`Making the desert bloom ...
`at a price
`
`Executive recruiters:
`a directory
`
`The business environment ahead
`... and how to handle it
`
`Cancer chemotherapy:
`success and failure
`
`targets and projectiles
`
`Methanol: emerging uses,
`new syntheses
`
`How to spray with molten metal
`
`R3SI-a dandy leaving group
`
`How to select
`polymerization reactors
`
`321 The Industrial Chymlst
`explains why the planning process panics
`people.
`
`0
`
`322 Write On
`D in which the Russians catch it. Then
`
`there's a potpourri of thought
`provocation.
`
`326
`D
`
`332
`D
`338
`D
`342
`D
`
`347
`D
`349
`D
`358
`D
`364
`D
`373
`D
`380
`D
`
`324
`
`0
`
`329
`
`0
`
`Heart Cut
`in which there's something different. Can
`you spot it? Do you like it?
`
`OBC The Last Word
`
`0 in which Ramadorai reports on his last
`
`safety meeting.
`
`Volume 12, No. 6, pages 321-384 ISSN 0009-2703
`CHTEDD
`CHEMTECH (ISSN 0009-2703) is published monthly by the
`American Chemical Society at 1155 16th St .. N.W.
`Washington. D.C. 20036. Second-class postage paid at
`Washington. D.C .. and additional mailing offices. POST(cid:173)
`MASTER: Send address changes to Membership & Sub(cid:173)
`scription Services. P.O. Box 3337. Columbus. Ohio
`43210.
`Subscripllons-1-year prices: Members, $20: Nonmem·
`bers (individual personal use), $30; Companies, institutions,
`libraries, $140: Students, $1 0: Foreign subooribers add $6
`lor postage. Send payment to ACS Controller, 1155 16th
`St., N.W .. Washington, D.C. 20036. Single copies, $13.
`Phone orders can be placed for printed, microfiche, and
`microfilm editions by calling, loll free, the sales office at
`(1)(800) 424-6747 from anywhere in the continental U.S.
`
`Can use Visa or Master Card Rates above do not apply to
`nonmember subscribers in Japan, who must enter sub(cid:173)
`scription orders with Maruzen Company Ltd., 3-10 Nihon
`bashi 2-chome, Chuo-ku, Tokyo 103, Japan. Tel: (03)
`272-7211
`Editorial matters: Refer all editorial matters to B. J, Luberoff.
`Editor, CHEMTECH, 48 Maple St., Summit, N.J. 07901:
`(201) 273-4923.
`Advertising: Centcom, Ltd .. T, Koerwer, Pres .. 25 Sylvan
`Road South, Westport. CT 06881: (203) 226-7131 .
`Service: For change of address, send old label with new
`address to ACS Membership and Subscription Services,
`P.O. Box 3337, Columbus, Ohio 43210 Allow 6 weeks to
`effect change. Refer claims for missing issues, information
`on status of accounts, etc., to this office Claims for missing
`issues will not be allowed if (a) notice of address change was
`
`not received in specific time: (b) North American claim was
`made 90 days beyond issue date: (c) other claim was made
`one year beyond issue dale.
`Disclaimer-The American Chemical Society assumes no
`responsibility for opinions advanced by contributors to its
`publications Views expressed in the editorials are those of
`the author and do not necessarily represent the official
`position of the Society.
`Reproduction: Permission is granted by the American
`Chemical Society to make reprographic copies for personal,
`noncommercial, or internal use. This permission does nol
`extend to copying for promotional purposes, creating new
`collective works, resale, or to any material not copyright to
`ACS. For details, contact B. Friedman, ACS. 1155 16th St.,
`N.W .. Washington, D.C. 20036: (202) 872-4367.
`
`Copyright © 1982 by the American Chemical Society
`
`IFC CHEMTECH JUNE 1982
`
`BASF Exhibit 2015, Page 2 of 18
`
`SNF Holding Company et al v BASF Corporation, IPR2015-00600
`
`

`
`How to select
`polymerization
`reactors
`part I
`
`There is life beyond
`the three-neck flask
`and resin kettle
`
`Heinz Gerrens
`
`A !though commercial polymerization reactors come in
`
`many configurations, only a few accepted guidelines exist
`for the complex problem of selecting a reactor for a specific
`polymerization reaction. Previous reviews by Daumiller (1)
`and Platzer (2) excluded polyaddition and polycondensation,
`and were restricted to either polymerization in the narrower
`sense (1) or to radical polymerization of ethylene, styrene,
`and vinyl chloride (2). The present two-part contribution
`is an attempt to review the entire range of polymerization
`reactions
`including polymer linkage. Problems of
`macromolecular chemistry are discussed as well as those of
`industrial processes.
`Industrial production of polymers differs from the
`common syntheses of compounds of low molar mass in some
`vital ways.
`Most polymers of industrial interest have molar masses
`between 104 and 107 and consist of mixtures with a definite
`molar mass distribution. High molar masses result in very
`high viscosities in solution or melt. For example, in solution
`polymerization of styrene, the viscosity can increase by over
`six powers of 10 as conversion progresses from 0 to 60% (3).
`Since linkage of monomer molecules into a single
`macromolecule always produces a decrease in entropy, a
`negative enthalpy gradient must provide the driving force
`of the reaction; hence, virtually all the polymerization
`reactions are exothermic. In the adiabatic polymerization
`of ethylene, for example, calculation shows the temperature
`increase on achieving a complete conversion to be about
`1800 K. Thus, considerable quantities of heat must be
`removed. This is made more difficult by the fact that high
`viscosities often result in low heat transfer coefficients in
`stirred reactors.
`Polymerization kinetics constitute another special feature.
`Radical and ionic polymerizations are chain reactions, and
`the steady-state concentration of chain carriers, i.e., (growing
`polymer radicals or ions) in industrial formulations, is only
`about w-s mol/L. These polymerization reactions are
`therefore highly sensitive to all impurities that could react
`with the chain carriers. Similar conditions prevail in
`polyaddition and polycondensation of bifunctional
`monomers, where high conversions are required to achieve
`sufficiently high molar masses, e.g., a conversion of 99.5%
`for a (number average) degree of polymerization P n = 200.
`
`380 CHEMTECH JUNE 1982
`
`Monofunctional compounds interfere with this type of
`polymerization reaction.
`In contrast to most low molar mass products, no polymer
`can be adequately characterized by a chemical formula and
`some purity specification. The following parameters are used
`to describe polymers:
`• chemistry (monomer building blocks)
`• molar mass distribution
`• grafting, crosslinking, end groups
`•
`tacticity, crystallinity, morphology of crystals
`• for copolymers: copolymer composition, sequence of
`monomers, statistics of grafting
`• for heterogeneous polymerization: particle shape and size
`distribution, porosity, mutual distribution of polymer
`phases
`• additives and impurities.
`Yet these parameters are frequently insufficient for the
`characterization of a plastic material that is bought because
`of its performance in a particular application. The relation
`between this performance and precisely defined chemical
`and physicaJ data is often not sufficiently clear, so that
`several empirical tests are also required.
`If the synthesized polymex does not imm diately attain
`the desired properties, it is virtually impossible to improve
`the quality by subsequent processing, in contrast to the
`synthesis of low molar mass compounds. Distillation and
`crystallization fail. Fractionation methods are far too
`expensive for industrial application. Even extraction of low
`molar mass components such as monomer residues or
`initiators can be difficult and expensive. One must choose
`the most suitable reactor and operating conditions to attain
`the required properties in the polymerization reaction.
`
`Polymerization reactions and reactors
`We can classify polymerization reactions based on their
`kinetics (Table 1). In monomer linkage or addition
`polymerization, single monomer molecules are attached
`successively to the growing polymer molecule. In radical
`polymerization, monomer linkage occurs with termination
`either bY" disproportionation or by combination. If the
`growing species is an ion, the ions of the same charge repel
`each other and the monomer linkage occurs without
`termination (e.g., anionic "living" polymerization). Cationic
`
`BASF Exhibit 2015, Page 3 of 18
`
`SNF Holding Company et al v BASF Corporation, IPR2015-00600
`
`

`
`Table 1. Kinetics of simple polymerization
`reactions
`
`Monomer linkage with termination
`(e.g., radical polymerization)
`
`Mn + Mm
`kt.~/
`Mr~ + Mm
`kt,c\
`
`Termination by
`
`Initiation
`
`Propagation
`
`Disproportionation
`
`Combination
`
`Monomer linkage without termination
`(e.g., anionic "living" polymerization)
`k;
`I+ M-+Mi
`
`Initiation
`
`Propagation
`
`Polymer linkage (e.g., polycondensation)
`
`Propagation
`
`BR
`CPFR
`I
`k
`kd
`k;
`kp
`kt,c
`kt,d
`
`M
`Mi,Mr~
`R·
`SCSTR
`
`t
`v
`v
`T
`
`Symbols used
`batch reactor
`continuous plug flow reactor
`initiator
`rate constant
`initiator decomposition constant
`initiation constant
`propagation constant
`termination constant (combination)
`termination constant
`(disproportionation)
`monomer
`growing polymer radical
`initiator radical
`segregated continuous stirred tank
`reactor
`time
`volume
`volumetric flow rate
`residence time
`
`Proceu
`Solution polymerization
`
`Solution polymerization with
`phase separation
`Precipitation polymerization
`
`Bead polymerization
`
`Emulsion polymerization
`
`Table 2. Polymerization processes
`
`Characterization
`a. Monomer acts as solvent for
`polymer
`b. Solvent miscible with monomer,
`dissolves polymer
`Polymer A dissolved in monomer B
`
`a. Polymer insoluble in monomer
`b. Monomer miscible with precipitant
`for polymer
`Monomer and polymer insoluble in
`water; initiator soluble in monomer
`a. Protective colloid
`b. Pickering emulsifier
`Monomer and polymer insoluble in
`water; initiator soluble in water
`a.
`Ionic emulsifier
`b. Protective colloid
`
`Example
`Bulk polymerization of methyl
`methacrylate
`Styrene in ethylbenzene
`
`High impact polystyrene with
`polybutadiene rubber
`Bulk polymerization of vinyl chloride
`Acrylonitrile in water
`
`Vinyl chloride with cellulose derivative
`Styrene with hydroxyapatite
`
`Butadiene-styrene with Na-Dresinate
`·vinyl acetate with polyvinyl alcohol
`
`CHEMTECH JUNE 1982 381
`
`BASF Exhibit 2015, Page 4 of 18
`
`SNF Holding Company et al v BASF Corporation, IPR2015-00600
`
`

`
`Reactor
`
`~
`
`v
`
`100
`%
`
`0
`
`T
`
`100
`%
`
`.... L
`1UL
`
`Residence time distributions Concentt'lltlon profile
`
`moi/L
`
`0.5
`
`I
`
`moi,IL
`
`v
`
`0
`
`t
`
`T
`
`t
`
`0.&
`
`I
`
`100
`o;o
`
`10.0L
`
`moi/L
`
`0 '----~---
`
`0.5
`
`t
`
`Figure 1. Model reactors
`
`(Ziegler-type)
`coordination
`and
`polymerization
`polymerization with transition metal catalysts are also
`classified as this type of reaction. The kinetics of both these
`polymerizations are far less well known than that of radical
`or anionic polymerization. Apparently, the original initiator
`is reconstituted after each macromolecule is formed, and
`hence no termination reaction occurs in the kinetic sense ( 4,
`5). The lifetime of a growing polyethylene molecule can be
`about 30 min (5). Polymer-linkage between
`two
`in polycondensation and
`macromolecules occurs
`polyaddition reactions also known as step reactions.
`In the early days of industrial polymerization, the
`heterogeneous polymerization processes of precipitation,
`bead (suspension) and emulsion polymerization (Table 2)
`were already developed with the view of avoiding the
`difficulties arising from the high viscosity of the reaction
`mixture. The resulting polymer suspension, frequently in
`an aqueous medium, exhibits low viscosity and an ease of
`handling together with a favorable heat transfer. The
`polymer dispersions and powders produced in these
`processes have long since become commercial products in
`
`their own right. Their colloidochemical properties are
`among th important parameters describing a polymer.
`Ch mica! reaction ngineering normally defines simple
`i lealized model reactor , a presented in Figure 1, together
`with th ir respective re idence tim distributions and the
`concentration profiles for a second order reaction (6). They
`are shown in the order of increasing width of the residence
`time distribution. The batch reactor (BR) and continuous(cid:173)
`plug flow reactor (CPFR) are best suited for attaining high
`conversion, while in a homogeneous continuous stirred tank
`reactor (HCSTR), complete conversion cannot be achieved
`on principle. A cascade of continuous stirred tank reactors
`represents an intermediate state. With an increasing number
`of reactors, it approaches the CPFR.
`important for
`Not shown, but probably quite
`polymerization reactions, is the segregated continuous stirred
`tank reactor (SCSTR). In it, the reaction mixture is regarded
`as consisting of many small separate elements. Each element
`includes a large number of molecules and reacts as a small
`batch reactor. The elements are ideally mixed and have the
`same residence time distribution as an HCSTR (7). Ideal
`
`382 CHEMTECH JUNE 1982
`
`BASF Exhibit 2015, Page 5 of 18
`
`SNF Holding Company et al v BASF Corporation, IPR2015-00600
`
`

`
`Polymerization
`reactiOM and
`proc••••
`
`Solution polymet1Dtlon
`
`PNclpltdon polymerlutlon
`
`Emul.ton polymerization
`
`Solution or melt polycondenMtlon
`
`=-u If
`
`Figure 2. Polymerization processes
`and industrially employed reactors
`
`homogeneous miXmg down to molecular dimensions is
`difficult to achieve at high viscosities of the reaction mixture.
`As shown by Nauman (8), the majority of all bulk and
`solution polymerizations probably proceed in segregated
`systems.
`For the three simple homogeneous polymerization
`reactions in Table 1 and three different model reactors, there
`are nine possible combinations, yielding quite different
`theoretical molar mass distributions, as shown in Table 3.
`Further details have been published in a review (9).
`In a similar grid, the polymerization reactions and
`processes enumerated in Tables 1 and 2 can be related to the
`ideal reactors shown in Figure 1. Figure 2 lists the types of
`reactors used for the different polymerization reactions.
`
`Adapted with permission from Ger. Chem . Eng. , 1981,4, 1-13.
`
`References
`(1) Daumiller, G. Chem . lng. Tech. 1968, 40 , 673.
`(2) Platzer, N. Ind. Eng. Chern. 1970, 62, 6.
`(3) Nishimura, N. ]. Macromol. Chern. 1966,1 , 257.
`
`(4) Russell, K. E.; Wilson, G. J. ln "Polymerization Processes";
`Schildknecht, C. E.; Skeist, l. , Eds.; Wiley: New York, 1977; p. 312.
`(5) Keii, T. "Kjnetics o( Ziegler- attn Polymerization"; Kodansha, Tokio,
`and Chapman & Hall: London, 1972; p. 142 f£.
`(6) Fill.er, E.; Fritz, W. "Technische Chemic"; pringer-Verlag: Berlin,
`1975; pp. 72- 75.
`(7) Fitzer, E.; F ritz, W. "Technische Chemie"; Springer-Verlag: Berlin,
`1975; p. 352 ff.
`(8) Nauman, E. B. }. Macromol. Sci.-Rev. Macromol. Chern. 1974, ClO,
`75.
`(9) Gerrens, H. Proc. 4th Int./ 6th European Symp. Chern. Reaction Eng.
`Heidelberg, 1976, 6.-'8,4,76, Vol. 2. Survey Papers p. 585.
`
`Heinz Gerrens is director of the Department of
`Process Development at BASF Aktiengesellschaft
`(D-6700 Ludwigshafen, West Germany). He
`joined BASF in 1954 after receiving his
`Dr.rer.nat. in physical chemistry from Mainz
`University. He holds an honorary professorship
`at the Technical University of Karlsruhe and is
`author of nu·merous publications on applied
`polymer science.
`
`BIG GAME HUNTER
`
`THE JELLY BEAN REVISITED
`
`" War, I'm sorry to say, is a very exhilarating
`experience. Everything else is a little bit duller. The
`war experience binds together those who have
`experienced it in a way that's not possible in any other
`kind of experience."
`
`Jack Thompson, journalist,
`remembering D-Day, WW II
`(from the New York Times
`June 8, 1981)
`
`It seems that years ago in Princeton, a 12-year-old girl
`used to spend a surprising amount of time visiting her
`neighbor, Albert Einstein. In due course, the mother
`became concerned and when the occasion presented
`itself, asked the great thinker if her daughter's
`presence was an annoyance. Dr. Einstein is reported
`to have said, " No Madam, not at all. You see, she likes
`the way I do her algebra and I enjoy the jelly beans she
`brings me."
`
`CHEMTECH JUNE 1982 383
`
`BASF Exhibit 2015, Page 6 of 18
`
`SNF Holding Company et al v BASF Corporation, IPR2015-00600
`
`

`
`JULY 1982
`
`Page 444
`
`BASF Exhibit 2015, Page 7 of 18
`
`SNF Holding Company et al v BASF Corporation, IPR2015-00600
`
`

`
`CHECKLIST
`
`[HEmTE[H
`
`JULY 1982
`
`Who's creative?
`
`How to judge a job applicant
`(or be one)
`
`A new corporate design
`
`The politics of the
`"Clean Water Act"
`
`A theological view of
`nuclear energy
`
`Nutrition and vitamin C
`
`How to select polymerization
`reactors-Part II
`
`Keeping up with the
`instrument explosion
`
`392
`
`0
`
`396
`
`0
`404
`0
`
`413
`0
`
`420
`0
`
`428
`
`0
`434
`0
`444
`0
`
`To answer, chemist Lyon tested colleagues, and more important, tested
`tests.
`
`Few who interview (on either side of the desk) have been trained for the
`task. Management consultant Wareham provides such training.
`
`Today's expectations, education, and externalities call into question the
`basic nature of corporations conceived in another age. Who better than
`Forrester to find a fix?
`
`Dowsing '82
`In the 1970s Congress addressed environmental law and the proliferation
`of laws. Linton relates how the original law happened, timely now that
`it's up for review.
`
`Some say that you can find "anything you seek" in the Bible (or in
`physics). Some say it (it) contains only "Truth." This
`was priest I physicist Pollard's challenge.
`
`If we knew " ... how it acts at the molecular level" we'd feel a lot more
`comfortable using as much vitamin C as other species do. Brin provides
`rationale for our doing so.
`
`The theory laid out by Gerrens in Part I now proves useful.
`
`Even if you don't use instruments you use their results, pay for them, wait
`for them, and question them. So Settle, Pleva, and Jackson are really
`addressing you.
`
`385
`0
`386
`0
`
`The Industrial Chymist
`suggests where to look for gold.
`
`View from the Top
`in which Osgood examines the purpose of
`the nonprofit research organization.
`
`387
`0
`
`389
`0
`
`Write On
`in which trademark convention,
`inventorship, personal pronouns, filters,
`and formaldehyde are called into
`question.
`
`Heart Cut
`in which Kreps and Thyagarajan found
`how to use membranes. singlet oxygen,
`ceramic paper. and prosthetic plastics.
`
`424 The Science of the Possible
`0
`in which Weisz tells us where the needle
`is in 28 orders of magnitude of a
`haystack.
`
`OBC The Last Word
`
`0 in which Teison tells how to travel smart.
`
`Volume 12, No.7, pages 385-448 ISSN 0009·2703
`CHTEDD
`CHEMTECH (ISSN 0009-27031 is published monthly by lhe
`American Chemical Society al 1155 16th Sl .. N W.
`Washington, D.C. 20036. Second-class pos1age paid al
`Washington. D.C .. and additional mailing ollices POST(cid:173)
`MASTER: Send address changes lo Membership & Sub(cid:173)
`scription Services. P 0 Box 3337. Columbus, Ohio
`43210.
`Subscrlpllons-1-year prices: Members. $20: Nonmem(cid:173)
`bers (individual personal usel. $30: Companies. inslilulions.
`libraries, $140: Students. $10: Foreign subscribers add $6
`lor postage Send paymenllo ACS Con1roller. 1155 16th
`Sl .. N W .. Washington, D.C 20036 Single copies. $13
`Phone orders can be placed for printed, microfiche, and
`m1crolilm editions by calling, loll tree. the sales ollice al
`( 11(800) 424-67 4 7 from anywhere in I he conlinenlal U.S.
`
`Can use V1sa or Master Card Rates above do nol apply lo
`nonmember subscribers in Japan, who must enter sub(cid:173)
`scription orders with Maruzen Company Lid. 3-10 Nihon
`bashi 2-chome. Chuo-ku, Tokyo 103, Japan. Tel: (03)
`272-721 1.
`Ed~orial maHers: Refer all editorial matters loB J Luberoll.
`Editor. CHEMTECH. 48 Maple Sl . Summit. N.J 07901 .
`(20 11 273-4923.
`Advertising: Cent com. Lid .. 'r Koerwer, Pres . 25 Sylvan
`Road South. Westport. CT 06881: (203) 226-7131 .
`
`Service: For change of address. send old label wilh new
`address lo ACS Membership and Subscription Services.
`P 0 Box 3337. Columbus. Ohio 43210 Allow 6 weeks Ia
`effect change Refer claims lor missing issues, information
`on status of accounts. etc. to this oHice Claims for missing
`issues will not be allowed if (a} notice of address change was
`
`not rece1ved in spE:cific t1me: (b) North American claim was
`made 90 days beyond 1ssue date: (c) other claim was made
`one year beyond issue date
`Disclaimer-The American Chemical Society assumes no
`responsibility for opinions advanced by contributors to its
`publications Views expressed in the editorials are those of
`\he author a fad do not necessarily represent I he ofticial
`position of the Society
`Reproduction: Permission is granted by the American
`Chemical Society lo make reprographic copies lor personal.
`noncommerc1al, or internal use This permission does not
`extend to copying for promotional purposes, creating new
`collective works. resale, or to any mal erial nol copynghl lo
`ACS For details, contact B Friedman. ACS. 1155 161h Sl.
`N W. Washington. DC 20036: (202) 872-4367
`
`Copyright © 1982 by lhe American Chem1cal Society
`
`IFC CHEMTECH JULY 1982
`
`BASF Exhibit 2015, Page 8 of 18
`
`SNF Holding Company et al v BASF Corporation, IPR2015-00600
`
`

`
`How to select
`polymerization reactors
`part II
`
`Heinz Gerrens
`
`In our first installment (CHEMTECH, 1982, June, p. 380)
`we looked broadly at the problems facing those who want
`to make polymers on a commercial scale. Now we look at
`actual reactor selection. The kinetic behavior of the
`particular polymerization is the primary determinant of
`which reactor configuration we use. It is convenient to
`classify reactions into three kinetic types:
`• monomer linkage with termination
`• monomer linkage without termination
`• polymer linkage.
`Let's look at each in turn.
`Monomer linkage with termination. Figure 3 presents
`the most important kinetic correlations, the chain transfer
`reaction being neglected. The overall reaction rate Rp and
`number average degree of polymerization P n are functions
`of the initiation rate R; (or the initiator concentration [I])
`and the monomer concentration [ M] and also of the
`temperature (via the temperature dependence of the
`individual rate constants . k). At constant monomer
`concentration (small conversion increment or CSTR) the
`so-called Schulz- F lory molar mass distribution is produced.
`The dispersion index. D = P w! P n is 1.5 for termination by
`combination and 2.0 for that by disproportionation. A high
`degree of polymerization P n is reached already at very small
`fractional conversions x. If the monomer concenh·ation [Ml
`decreases during the polymerization (BR or CPFR), a
`number of Schulz-Flory distsributions overlap, with
`decreasing average degree of polymerization P n, and the
`molar mass distribution becomes wider.
`The development of industrial reactors for the bulk and
`solution polymerizations of styrene is shown in Figure 4.
`When the industrial production of polymers was first begun,
`no processes were known for removal of residual monomers
`at the end of polymerization. Complete conversion is most
`easily achieved in a BR. However, if solid polymer is the
`desired product, only the heterogeneous processes of
`precipitation, bead, and emulsion polymerization can be
`applied if no 'miner 's pickax" is to be employed. The
`so-called cast polymerization (2, 10), still used in the U.S. in
`197 4 (11 ), is a bulk polymerization in thin sheets. The process
`is often carried out in a BR up to 30-40% conversion. The
`viscous syrup is then cast in vertical molds clamped together
`in a filter press and polymerized to completion. A similar
`process is used for producing acrylic glass plates from methyl
`methacrylate (12).
`In a continuous process developed by BASF (13, 14) as
`early as 1936, two parallel stirred tank reactors precede a
`tower reactor. Also, in this case, as complete a conversion as
`
`In order to mafntatn in each issue both po/ydiscipltnary balance and
`t-tmelfn~s, we sometimes sertalt-ze arttcles originally written as· single
`contributions. When we do, we Insert appropriate bridging text to alert
`the reader. Unfortunately, that te:tt Intended to accompany the first
`installment of this article "got lest ." For this tire editors apologize to Dr.
`Cemms and the readers.
`
`434 CHEMTECH JULY 1982
`
`possible is desired so that the residence time distribution of
`the CPFR is required. The temperature rises from 80 oc in
`the CSTRs to 220 oc at the outlet of the tower reactor,
`thereby limiting the viscosity of the melt. The increasing
`temperature produces different instantaneous degrees of
`polymerization (see viscosity average P 71 in Figure 4) and
`
`BR
`CPFR
`0= Pwl Pn
`d
`I
`[IJ
`K
`k
`k;
`kp
`kt,d
`
`R;
`Rp
`SCSTR
`
`T
`VK-tube
`W(P)
`
`X
`
`Symbols used
`batch reactor
`continuous plug flow reactor
`dispersion index
`reactor diameter
`initiator
`initiator concentration
`equilibrium constant
`rate constant
`initiation constant
`propagation constant
`termination constant
`( disproportionation)
`length of reactor
`monomer
`monomer concentration
`mole fraction of glycol
`mole fraction of water
`number of monomer units in P-mer
`degree of polymerization
`(number average)
`degree of polymerization
`(mass average)
`degree of polymerization
`(viscosity average)
`rate of initiation
`overall rate of polymerization
`segregated continuous stirred tank
`reactor
`temperature
`tower reactor for polyamide 6
`mass distribution function of the degree
`of polymerization
`fractional conversion
`
`Subscripts
`0
`X
`
`initial
`at conversion x
`
`Editor's note
`If you need to refresh yourself on free radical kinetics.
`as we did. consult Bill Lloyd's series in CHEMTECH's
`inaugural volume ( 1) or Cheves Walling's classic (2). The
`steady-state assumption
`really does make
`life
`bearable.
`References
`(1) Lloyd, W. G. CHEMTECH 1971, March. 176; June, 371; November,
`687.
`(2) Walling. C. "Free Radicals in Solu1ion"; John Wiley & Sons: New
`York. 1957.
`
`BASF Exhibit 2015, Page 9 of 18
`
`SNF Holding Company et al v BASF Corporation, IPR2015-00600
`
`

`
`1,.
`
`a:
`&D
`
`0
`
`s t
`I
`I
`I i I
`' f
`..
`j
`I
`~
`
`='
`
`•
`I
`i!
`~
`c
`8
`
`Q.
`Ill
`
`l
`1: I
`c
`=
`8 !
`
`I I
`I
`
`=::n
`
`PfDIIIII ..
`
`Solution polymwlntion
`
`·~pol~lon
`
`if
`
`Figure 2. Polymerization processes
`and Industrially employed reactors
`
`a rather wide molar mass distribution (15).
`A continuous polymerization in the presence of a solvent
`and with incomplete conversion was made possible by the
`development of equipment for the removal of monomer and
`solvent. The process of the Dow Chemical Co. makes use of
`a cascade of three tower reactors equipped with stirrers and
`internal heat-exchange tubes, in addition to the usual jacket.
`Ethylbenzene solvent helps to maintain a low viscosity and
`the temperature rises only froin llO octo 170 oc. A vacuum
`degasser removes the residual styrene and ethylbenzene,
`which are returned to the first reactor. The resulting
`polystyrene has a narrower molar mass distribution than that
`made by 'the old BASF process, because of the more uniform
`
`temperature profile and the exclusion of the range of very
`low monomer concentrations [M] in which the dispersion
`index of the polymer changes rapidly (9). High-impact
`polystyrene and ABS polymers can also be produced by the
`Dow process.
`A further step in the direction of uniform temperature
`the application of the
`and concentration profiles is
`completely mixed continuous stirred tank reactor followed
`by a vacuum degasser. In the ideal steady state, T and [M]
`remain constant and the Schulz-Flory distribution is
`produced as the narrowest attainable molar mass distribution
`in radical polym~rization. In this case, the reactor can
`contain large quantities of monomer; precautions against
`
`S~hulz.Fiory
`
`"I 0
`rrl:-a.e 0.5
`
`Distribution
`
`p _ p
`(1 - x)
`_o.•-
`n,o
`Pn,o = 1000
`
`D= * = 1.S Comblnldlan,
`
`n
`
`2.0 Dlsproportlonatlan
`
`0 0~~~~~~~~~~~.
`0.6
`1.0
`Conversion x
`
`0.8
`p. 1t-8
`
`Figure 3. Kinetics of monomer linkage with termination; radical polymerization
`
`CHEMTECH JULY 1982 435
`
`BASF Exhibit 2015, Page 10 of 18
`
`SNF Holding Company et al v BASF Corporation, IPR2015-00600
`
`

`
`the occurrence of a runaway reaction in case of a breakdown
`must be carefully considered (16). The CSTR is also
`particularly suited to the production of copolymers such
`as styrene/acrylonitrile. At steady state, constant
`concentrations of the comonomers set in ·and a copolymer
`with a uniform chemical composition is formed even from
`nonazeotropic monomer mixtures (17, 18).
`Another important example of the development of
`polymerization reactors is the high-pressure polymerization
`of ethylene. At sufficiently high pressures this reaction is
`homogeneous. Both the continuous stirred tank reactor (ICI)
`(19) and the continuous plug flow reactor (BASF) (20) are
`used for this polymerization (see Figure 5). The maximum
`attainable conversion is limited by heat removal. A critical
`temperature range of >280-330 a C, in which explosive
`decomposition of ethylene into carbon and methane occurs,
`must be avoided. A tube reactor has a favorable ratio of
`surface area to volume and relatively thin walls. This makes
`heat removal easier, but the reaction mixture must be
`preheated to the starting temperature of the polymerization.
`In a stirred tank reactor with thick walls and an unfavorable
`surface area to volume ratio, the polymerization proceeds
`almost adiabatically. However, cooling is possible by
`addition of cold feed.
`In a simple tube reactor, ethylene conversions of 20- 25%
`are being achieved. However, on subsequent addition of cold
`gas and initiator and application of tubes with increasing
`diameters (21 ), conversions of up to 35% are possible. Also,
`the simple, compact stirred autoclave has been developed
`into a slim (l/d ratio = 10 to 20) vessel with several
`compartments. Multiple feed
`into
`the
`individual
`compartments can alter the temperature profile, and
`
`backmixing in the reactor can be varied by changing the
`pitch of the stirrer blades (22). These measures also res~lt
`in higher conversions.
`It follows from the patent literature and a review by Luft
`(23) that a polymer with a relatively narrow molar mass
`distribution is formed in the single tube reactor. Multiple
`feed with correspondingly extended temperature profile
`produces wider distributions. Surprisingly, in the compact
`stirred autoclave, where temperature and concentration
`should be practically uniform, a polymer with a much wider·
`molar mass distribution (D = 12 to 16) is formed. On the
`other hand, the autoclave with compartments yields a more
`uniform polymer with D = 4 to 8 in spite of much wider
`temperature and concentration profiles. Finally, very wide,
`asymmetric molar mass distributions with dispersion indices
`of up to D =50 can be produced with a radial feed into the
`middle part of a slim, stirred autoclave, not divided into
`compartments (24) .
`At present, the reasons for this can only be speculated on.
`The occurrence of segregation appears possible and should
`lead to much wider distributions in the SCSTR than in a
`CPFR (see Table 3 and Reference 9). On the other hand, a