polymers

Review Dormant Polymers and Their Role in Living and Controlled Polymerizations; Influence on Polymer Chemistry, Particularly on the Ring Opening Polymerization

Stanislaw Penczek *, Julia Pretula and Piotr Lewi ´nski

Centre of Molecular and Macromolecular Studies of Polish Academy of Sciences, Sienkiewicza 112, 90-363 Lodz, Poland; [email protected] (J.P.); [email protected] (P.L.) * Correspondence: [email protected]; Tel.: +48-42-681-9815

Received: 2 November 2017; Accepted: 23 November 2017; Published: 25 November 2017

Abstract: Living polymerization discovered by Professor Szwarc is well known to all chemists. Some of the living polymerizations involve dormancy, a process in which there is an equilibrium (or at least exchange) between two types of living polymers, namely active at the given moment and dormant at this moment and becoming active in the process of activation. These processes are at least equally important although less known. This mini review is devoted to these particular living polymerizations, mostly polymerizations by the Ring-Opening Polymerization mechanisms (ROP) compared with some selected close to living vinyl polymerizations (the most spectacular is Atom Transfer Radical Polymerization (ATRP)) involving dormancy. Cationic polymerization of tetrahydrofuran was the first one, based on equilibrium between oxonium ions (active) and covalent (esters) dormant species, i.e., temporarily inactive, and is described in detail. The other systems discussed are polymerization of oxazolines and cyclic esters as well as controlled radical and cationic polymerizations of vinyl monomers.

Keywords: dormant polymers; living polymerization; controlled polymerization; ring-opening polymerization

1. Introduction Professor Michael (Michał in Polish) Szwarc’s contributions to the chemical science are not only related to polymers and to his discovery of living and (less known) dormant polymers [1]. Dormant polymers, reversibly formed from the active ones, are at the basis of several living/controlled polymerizations described in the present paper. Moreover, Professor Szwarc is also well known for his work on the bond energy (continued for some time in Syracuse) and on the electron transfer phenomena [2,3]. In 1956, during discussion with Samuel Weissman on the electron transfer in aromatic compounds, Szwarc asked Weissman about styrene (cit.) “Did you transfer electrons to styrene?” The answer he received was: (cit.) “No use, it polymerizes” [4]. Szwarc then realized that electron transfer in the case of styrene not only leads to polymerization, but also to creation of living active species which was indicated by the persistent red color. Since then, working with his students/coworkers, Moshe Levy, Ralph Milkovich, Joseph Jagur, Johannes Smid and several postdocs, he has answered all major questions concerning the anionic polymerization, and is acknowledged by all chemists to be the “father of the living polymerization”. Thus, despite his other important contributions to chemistry, the discovery of the living and dormant polymers is rightly considered to be his greatest achievement. In the recently published “Perspectives” by Grubbs and Grubbs [5], it is amply shown that discovery

Polymers 2017, 9, 646; doi:10.3390/polym9120646 www.mdpi.com/journal/polymers Polymers 2017, 9, 646 2 of 18 of the living polymerization revolutionized practically all areas of polymer chemistry and is still a driving force for novelty in general chemistry and material sciences [6]. The issue strongly connected with living-active polymers, yet not that well known by many chemists, are living-dormant polymers; both have been extensively described in the monumental Szwarc’s monograph on anionic polymerization [1]. The concept is based on the Szwarc’s observation that, in the anionic polymerization of styrene in the presence of anthracene, the temporarily active and inactive (dormant) species are formed reversibly [7]. It will be shown in the present mini review that dormant polymers play a decisive role in understanding living-controlled processes such as ROP [8] and are a fundamental aspect of the controlled radical polymerizations [9–12].

2. Basic Definitions and Their Importance for Better Understanding of the Processes under Study: “Living”, “Controlled”, and “Dormant” The expressions listed in the Section2 title are often used in contexts that are not completely proper. The definition given by International Union of Pure and Applied Chemistry (IUPAC), based on the Szwarc’s description of living and dormant polymers, is as follows: “Ideally, living polymers propagate while their termination or chain transfer are rigorously prevented” [4]. The expression “living polymers” was coined soon after the first experiments had been performed and was used in the two first papers that were published (not without difficulties, due to the expression “living” being restricted for biosciences). These famous papers appeared in 1956 in Nature and Journal of American Chemical Society [13,14]. It is not sufficiently appreciated in general chemistry, as this discovery is important not merely for polymer chemistry. Indeed, until Szwarc’s discovery, chain reactions were described (since Bodenstein’s discovery of chain reactions [15,16]) as composed of at least three elementary reactions: initiation, propagation and termination. This has been described in detail, for instance, by Nicolai Semenov and Sir Cyril Hinshelwood, Nobel Prize Laureates for their work on the chain reactions [17,18]. In contrast to these processes, in living polymerizations, the process is composed exclusively of initiation and propagation, eliminating termination. Therefore, it has led to the novel kind of steady state processes. Until the discovery of living polymers, steady state reactions (as described by Bodenstein [15]) resulted from an interplay of formation of chain carriers in the initiation reaction and their disappearance in the termination reaction by interaction of two chain carriers. This principle is at the bases of the kinetics of the large number of chain reactions, including free radical polymerizations. The novel concept of the steady state is based on the invariant concentration of active centers (growing chains) equal from the very beginning (at certain conditions) to the concentration of the initiator of the polymerization. This approach is true when initiation is complete and the rate of initiation is faster than the rate of propagation. Such processes are considered as fully controlled. However, living polymerizations are not restricted to this particular instance. Slow initiation accompanying fast propagation may also provide living polymers although polymerization is not controlled. These general phenomena are (or should be) taught in every course of polymer science and technology and are used by hundreds of authors in thousands of papers, yet, the actual limitations, discussed further in the text, are seldom taken into the account. IUPAC has defined the terms “living”, “controlled” and “dormant”, simultaneously putting to rest several other expressions that have proven to be no long necessary and were often very confusing (e.g., “immortal”, “quasi living”, etc.) [19]. Therefore, the living polymerization is (cit.) “a chain polymerization from which irreversible chain transfer and chain termination are absent. In many cases, the rate of chain initiation is fast compared with the rate of chain propagation, so that the number of kinetic-chain carriers is essentially constant throughout the polymerization” [20]. Polymers 2017, 9, 646 3 of 18

The term “Controlled polymerization” indicates (cit.) “control of a certain kinetic feature of a polymerization or structural aspect of the polymer molecules formed, or both” [19]. Thus, for living polymerization, termination and irreversible chain transfer should be absent, although reversible deactivation (e.g., reversible transfer), leading to dormant macromolecules, is permitted, as is slow initiation. It may therefore happen that a living polymerization is not a controlled one (due to slow initiation) and controlled polymerizations may not be living, when molar mass and end-groups are under control only within a certain range of conversions and polymerization degrees (usually limited). As the criterion of the absence of termination and irreversible transfer, Szwarc quoted [21] (pp. 14–15) an equation derived in our group [22]: ln(1 ´ Pn¨[I]0/[M]0) = ´kp¨[I]0¨t, where I, M and t are the concentration of the initiator, the concentration of the monomer and time, respectively. Pn stands for the degree of the polymerization. The linearity of the plot of this equation indicates the simultaneous absence of these two mentioned reactions and is characteristic for both living and controlled polymerizations. Dormant species are formed in reversible chain deactivation and this process is described by IUPAC as follows (cit.) “deactivation of a chain carrier in a chain polymerization, reversibly converting an active center into an inactive one and then, within the average lifetime of a growing macromolecule, regenerating an active center on the same original carrier, hence, the temporarily deactivated species created in this process are often described as dormant” [19]. The dormant polymers, i.e., macromolecules, are thus living and only temporarily become unable to grow. Szwarc not only observed temporarily inactive species, coining an expression “dormant”, but also made the following statement on the living-dormant polystyrene complexed with anthracene [7] (cit.) “Complex is in equilibrium with its components, on addition of styrene the polymerization ensues. The exchange between the free living and those complexed with anthracene (dormant) is sufficiently rapid to distribute the polymerized monomer among all polymer molecules”. Thus, the importance of the fast exchange has clearly been stated as a necessary condition for low dispersity (Ð), as computed, for example, for the controlled radical polymerization (e.g., ATRP) by Matyjaszewski [23]. The term “immortal”, which is not supported by IUPAC, was introduced by Inoue for systems with reversible chain transfer (and formation of dormant species) [24]. It is still used by some authors, although IUPAC definition for “living” encompasses reversible transfers. Indeed, as already stated, the resulting from reversible transfer dormant macromolecules retain an ability to grow. Definition of these terms has been important not only from the semantic point of view but also to allow researchers to look at their systems in a more careful manner and to elaborate such conditions for the studied processes that would conform them to these definitions, and, therefore, to better control of the studied systems.

3. Dormancy As mentioned above, it had been in the school of Syracuse that important phenomenon of “dormant polymers” was first observed and defined. Later, understanding of this principle by several other researchers allowed conversion of the otherwise nonliving polymerizations into the living ones. This concept was introduced to account for the fact that “living ends can be protected from termination, but still be reversibly active for propagation”, as recently remained by Levy [6]. There is a plethora of reviews on living and controlled polymerization, yet dormancy has never been comprehensively reviewed. Probably the only exceptions are: a subsection titled “Living and Dormant Polymers” in the last book by Szwarc (published in 1996) [25] (p. 17) and the paper published by our group in 1995 under the title “Polymerizations with contribution of covalent and ionic species” [26]. The present mini review is merely giving a few examples of chain polymerizations, being living-dormant and controlled-dormant, particularly in the ROP and compared with vinyl polymerizations. Polymers 2017, 9, 646 4 of 18

PolymersPolymers 2017 2017, 9,, 6449, 644 4 of4 18 of 18

4. Dormant4. Dormant4. Dormant Species Species Species in in thein the the Ring-Opening Ring-Opening Ring-Opening Polymerization PolymerizationPolymerization (ROP)(ROP) (ROP) Probably,Probably,Probably, the the the first first first processprocess process that thatthat would would would bebe be called called today today “living-dormant”“living-dormant” “living-dormant” was was described described was described by by Flory Flory by Floryalreadyalready already in in1944 in1944 1944for for an foran anionic anionic an anionic polymerization polymerization polymerization ofof ethylene of ethylene oxideoxide in in the oxidethe presence presence in the of of presencean an alcohol alcohol of(Scheme an(Scheme alcohol (Scheme1) 1)[27]. [27].1)[ This 27This]. process This process process leads leads leadsto to thethe to Poisson thePoisson Poisson distdistribution distribution ofof molarmolar of molar massesmasses masses derived derived derived based based based on on this onthis this polymerizationpolymerizationpolymerization [27 [27,28]., 28[27,28].].

SchemeScheme 1. An 1. An anionic anionic polymerization polymerization ofof ethyleneethylene oxide oxide in in the the presence presence of an of alcohol. an alcohol. Scheme 1. An anionic polymerization of ethylene oxide in the presence of an alcohol. As followsAs follows from from Scheme Scheme 11,, R’OH R’OH is isthe thechain chain transfer transfer agent that agent is present that isfrom present the beginning from the of Asthe follows polymerization from Scheme (at first,1, R’OH as isa thelow chain molar tr ansfermass agentalcohol that and is presentlater as frommacro the alcohol).beginning beginning of the polymerization (at first, as a low molar mass alcohol and later as macro alcohol). of RO(CHthe polymerization2CH2O)CH2CH 2OH(at isfirst, at theas samea low time mo a larchain mass transfer alcohol agent and and latera dormant, as macro temporarily alcohol). RO(CH CH O)CH CH OH is at the same time a chain transfer agent and a dormant, temporarily RO(CHinactive2 2CH2 polymer2O)CH22 CHchain,22OH reversibly is at the convertisame timeng in a thechain reaction transfer with agent macroanion and a dormant, R’O− into temporarilyan active ´ inactiveinactiveform—a polymer polymer living chain, macromolecule. chain, reversibly reversibly converting converting inin thethe reactionreaction with macroanion R’O R’O− intointo an an active active form—aform—a livingIn living several, macromolecule. macromolecule. later studied ring-opening polymerizations, living-dormant interconversions take Inplace.In several, several, However, later later species studied studied that ring-opening ring-openinglook dormant polymerizations, polymerizations,may not necessarily living-dormant living-dormant be “completely interconversions interconversionsinactive” and may take take place.place.retain However, However, partial speciesactivity species thatas that it lookhas look been dormantdormant documented maymay not in necessarilythe polymerization be be “completely “completely of tetrahydrofuran inactive” inactive” and(THF) and may may retainretaininitiated partial partial with activity activity triflic as asacid it it has (trifluoromethanehas been been documenteddocumented sulfonic in acid, the polymerizationTfOH) [29], or inofof thetetrahydrofuran tetrahydrofuran polymerization (THF) of (THF) initiatedinitiatedoxazolines with with triflicinitiated triflic acid acid with (trifluoromethane (trifluoromethane alkylhalides [30]. sulfonicsulfonicIn these systems,acid, TfOH) TfOH) dormant [29], [29 ], macromoleculesor or in in the the polymerization polymerization may have of of oxazolinesoxazolinessome activity initiated initiated yet with many with alkylhalides times alkylhalides less than [30 [30].].the In active theseIn these ones systems, systems,and rightly dormant dormant could macromolecules be macromolecules considered as may dormant. may have have some The equilibrium involving active and dormant macromolecules that takes place in the ROP of activitysome yetactivity many yet times many less times than less the than active the onesactive and ones rightly and rightly could could be considered be considered as dormant. as dormant. THF initiated by TfOH is given in Schemes 2–5 [31]. TheThe equilibrium equilibrium involving involving active active andand dormantdormant ma macromoleculescromolecules that that takes takes place place in inthe the ROP ROP of of THFTHF initiated initiated by by TfOH TfOH is is given given in in Schemes Schemes2 –2–55[31 [31].].

Scheme 2. The active-inactive (dormant) interconversion in the polymerization of THF initiated with TfOH. SchemeScheme 2. 2.The The active-inactiveactive-inactive (dormant) (dormant) interconversion interconversion in the in polymerization the polymerization of THF of initiated THF initiated with withTfOH. TfOH.Due to the relatively slow ion-ester exchange rate (on the NMR time scale), both the covalent species and the ion pairs can be traced by 1H-NMR (Figure 1). Their relative concentrations strongly dependDue to on the the relativelypolarity of slow the solvent ion-ester [31]. exchange rate (on the NMR time scale), both the covalent Due to the relatively slow ion-ester exchange rate (on the NMR time scale), both the covalent species Theand conversion the ion pairs of the can covalent be traced species by 1H-NMR into the io (Figurenic ones 1). proceeds Their relative with an concentrations anchimeric assistance strongly species and the ion pairs can be traced by 1H-NMR (Figure1). Their relative concentrations strongly depend(a neighboring on the polarity group ofparticipation), the solvent [31].and therefore is much faster than the reaction of the monomer depend on the polarity of the solvent [31]. withThe the conversion covalent species of the covalent(Scheme 3).species into the ionic ones proceeds with an anchimeric assistance This is a unimolecular opposed reaction (an ion pair is kinetically one species), and since its rate (a Theneighboring conversion group ofthe participation), covalent species and therefore into the ionicis much ones faster proceeds than the with reaction an anchimeric of the monomer assistance is sufficiently low, its rate constant was measured by 1H-NMR [32]. It has been noticed that the direct (a neighboringwith the covalent group species participation), (Scheme 3). and therefore is much faster than the reaction of the monomer covalent growth cannot take place, as it is forbidden by the symmetry rules. Covalent species may with theThis covalent is a unimolecular species (Scheme opposed3). reaction (an ion pair is kinetically one species), and since its rate grow only by intermolecular conversion into ionic species, larger by one unit, when the macroester is sufficientlyThis is a unimolecular low, its rate constant opposed was reaction measured (an ion by pair1H-NMR is kinetically [32]. It has one been species), noticed and that since the direct its rate 1 is sufficientlycovalent growth low, itscannot rate constanttake place, was as measuredit is forbidden by H-NMRby the symmetry [32]. It has rules. been Covalent noticed species that the may direct covalentgrow only growth by intermolecular cannot take place, conversion as it is into forbidden ionic species, by the larger symmetry by one rules. unit, when Covalent the macroester species may grow only by intermolecular conversion into ionic species, larger by one unit, when the macroester

Polymers 2017, 9, 646 5 of 18 Polymers 2017, 9, 644 5 of 18

Polymers 2017, 9, 644 5 of 18 reacting withwith incoming monomermonomer moleculemolecule isis convertedconverted intointo an oxonium ion (as schematically shown in Scheme 4): in Schemereacting4 with): incoming monomer molecule is converted into an oxonium ion (as schematically shown in PolymersScheme 2017 4):, 9, 644 5 of 18

reacting with incoming monomer molecule is converted into an oxonium ion (as schematically shown in Scheme 4):

1 ´ Figure 1. The 300-MHz 1H-NMR spectra (the 5.0–5.5 τ region only) of living polyTHF with CF SO O − FigureFigure 1. 1.The The 300-MHz 300-MHz H-NMR 1H-NMR spectra spectra (the(the 5.0–5.5 τ regionregion only) only) of of living living polyTHF polyTHF with with CF CF33SO3SO2O2 −2 O

anion:anion: ( a )( ain ) in CD CD3NONO3NO22; 2;((; b(bb)) )in inin CH CHCH22ClCl22;2 ; and;and and ((cc ())c )inin in CCl CCl44 solvent4solventsolvent (first(first (first published published published in in [31], in [31], [31 the ],the thescale scale scale is inis is inτ in τ Figure 1. The 300-MHz 1H-NMR spectra (the 5.0–5.5 τ region only) of living polyTHF with CF3SO2O− τunits).units).units). anion: (a) in CD3NO2; (b) in CH2Cl2; and (c) in CCl4 solvent (first published in [31], the scale is in τ units).

Scheme 3. The intramolecular conversion of the covalent species (macroesters) into the ionic Scheme 3.3. TheThe intramolecular intramolecular conversion conversion of of the the cova covalentlent species (macroesters) into the ionicionic (macroionScheme pairs)3. The in intramolecular the polymerization conversion of THF of initiated the cova withlent TfOH.species (macroesters) into the ionic (macroion pairs) in the polymerization of THF initiated with TfOH. (macroion(macroion pairs) pairs) in the in the polymerization polymerization of of THF THF initiatedinitiated with with TfOH. TfOH.

SchemeSchemeScheme 4. 4.The 4.The The intermolecular intermolecular intermolecular conversion conversion conversion of ofof aa macroestermacroester inin in reaction reaction reaction with with with a amonomer amonomer monomer molecule molecule molecule into into into anSchemean oxonium anoxonium oxonium 4. The ion ion intermolecular ion in in the in the the polymerization polymerization polymerization conversion of ofof THF THF THFof a initiated initiatedmacroester with with in TfOH. TfOH. reaction with a monomer molecule into an oxonium ion in the polymerization of THF initiated with TfOH.

Polymers 2017, 9, 646 6 of 18 Polymers 2017, 9, 644 6 of 18

The complete set of equilibrating species in polymerization of THF initiated with triflic acid or The complete set of equilibrating species in polymerization of THF initiated with triflic acid or its its esters is schematically presented in Scheme 5 [33]. esters is schematically presented in Scheme5[33].

Polymers 2017, 9, 644 6 of 18

The complete set of equilibrating species in polymerization of THF initiated with triflic acid or its esters is schematically presented in Scheme 5 [33].

SchemeScheme 5. 5. TheThe set set of of ion–ester ion–ester equilibria equilibria in in the the polymerization polymerization of of THF THF initiated initiated with with TfOH. TfOH.

In the polymerization of THF initiated by triflic acid the rate constants for the various conditions In the polymerization of THF initiated by triflic acid the rate constants for the various conditions havehave been been measured. measured. Since Since the the detailed detailed discussion discussion of of the the mechanism mechanism of of an an ionic ionic and and a a covalent covalent Scheme 5. The set of ion–ester equilibria in the polymerization of THF initiated with TfOH. propagationpropagation and and determination determination of of the the rate rate constants constants is is out out of of the the scope scope of of this this paper, paper, the the interested interested readerreader may may find findIn the these polymerization information of THF in initiated [[31–33].31–33 by]. triflicNevertheless, Nevertheless, acid the rate constantsestablishing establishing for the those thosevarious kinetic kineticconditions parameters parameters allowedallowed determination determinationhave been measured. the the times timesSince theof of andetailed an average average discussion macr macromolecule ofomolecule the mechanism to to exist existof an in inionic ionic ionic and and anda covalent covalent covalent state. state. propagation and determination of the rate constants is out of the scope of this paper, the interested Furthermore,Furthermore, it couldcould bebe calculatedcalculated how how many many monomer monomer molecules molecules are are added added at these at these periods periods of time. of time. Thesereader findings may findare theseillustrated information below. in [31–33]. Nevertheless, establishing those kinetic parameters These findingsallowed are determination illustrated the below. times of an average macromolecule to exist in ionic and covalent state. TheThe recordFurthermore, ofof life life it of couldof a macromolecule a bemacromolecule calculated how that many undergoesthat monomer undergoes interconversions molecules interconversions are added between at these between activeperiods and ofactive dormant and dormantstates can statestime. be visualized These can befindings visualized as are sections illustrated as markedsections below. onmarked the line, on wherethe line, the where lengths the of lengths these sections of these aresections equal The record of life of a macromolecule that undergoes interconversions between active and areto theequal time to spentthe time in aspent given in state a given (Scheme state 6(Sch). Thus,eme an6). averageThus, an macromoleculeaverage macromolecule propagates propagates on ionic on ionic speciesdormant for states a certain can be visualizedtime, adding as sections X monomer marked on molecules, the line, where then the thelengths ion-pair of these collapses sections into the species forare a certainequal to the time, time adding spent in Xa given monomer state (Sch molecules,eme 6). Thus, then an average the ion-pair macromolecule collapses propagates into the covalent covalent species. If these are sufficiently less reactive than the ionic species (kp >> kci), or unreactive at species. Ifon these ionic species are sufficiently for a certain time, less reactiveadding X monomer than the molecules, ionic species then the (ion-pairkp >> kcollapsesci), or unreactiveinto the at all, all,then then the the covalentcovalent covalent species. species species If arethese dormant areare sufficientlydormant ones. lessones. Then, reactive Then, the than macromolecule the the macromoleculeionic species spends (kp >> k cispends), some or unreactive time some as attime an inactive as an all, then the covalent species are dormant ones. Then, the macromolecule spends some time as an inactive(dormant) (dormant) and recovers and recovers to the ionic to the state ionic due state to internal due to internal or external or external ionization ionization (k and/or (kci and/ork ). kci’). inactive (dormant) and recovers to the ionic state due to internal or external ionization (kcici and/or kci’). ci’

SchemeScheme 6. Periods 6. Periods of of activity activity an andd inactivity inactivity (dormancy). (dormancy). Scheme 6. Periods of activity and inactivity (dormancy). In Scheme 7, it is shown how these active and inactive periods oscillate in the polymerization of InIn Scheme SchemeTHF in 77, two, it issolvents shown of different how these polarity active (CCl and4 and inactive CH3NO2). periods oscillate in the polymerization of THF in two solventsThus, even of in different CCl4, where polarity a macromolecule (CCl and is CHin a dormantNO ). state much larger than in an ionic THF in twostate solvents (124 s ofvs. different8.1 s), there polarity is approximately (CCl4 and 10 CH times33NO larger22). number of monomer molecules Thus,Thus, introducedeven even in in CCl CClinto 4the4, ,where where chain by a a ionicmacromolecule macromolecule chain ends. This is is ratio in in a ais dormant dormanteven more statethan state 500 much much in CH larger larger3NO2 solvent. than than in in an an ionic ionic statestate (124(124 ss vs. vs.We 8.1 8.1have s), theres),described there is approximately theis ionicapproximately polymerization 10 times 10of THF largertimes in some numberlarger details, number of since monomer it isof the monomer moleculesmost general introducedmolecules one (for ROP) and, moreover, the most thoroughly studied. Other polymerizations, e.g., introducedinto the chain into by the ionic chain chain by ionic ends. chain This ratioends. is This even ratio more is thaneven 500more in than CH3 NO5002 insolvent. CH3NO2 solvent. We havepolymerization described of the oxazolines ionic polymerization or even the controll of THFed radical in some polymerization, details, since either it isproceed the most in the general one We havesame described way or are almostthe ionic identical polymerization kinetically [34]. of THF in some details, since it is the most general one(for ROP)(for and,ROP) moreover, and, moreover, the most thoroughlythe most studied.thoroughly Other studied. polymerizations, Other polymerizations, e.g., polymerization e.g., of polymerizationoxazolines or even of oxazolines the controlled or radicaleven the polymerization, controlled radical either polymerization, proceed in the same either way proceed or are almostin the sameidentical way kinetically or are almost [34]. identical kinetically [34].

Polymers 2017, 9, 644 7 of 18

Polymers 2017, 9, 646 7 of 18

Polymers 2017, 9, 644 7 of 18

−1 Scheme 7. Conditions:Scheme 7. Conditions: [THF] [THF]=0 = 8 8 mol·L mol¨L; 25´ °C1; [26,34], 25 ˝ C[where26 τ, is34 time,], where X is the numberτ is time,of monomerX is the number molecules, and indexes i0, c, p and d stand for ionic and covalent propagation and depropagation, of monomer molecules,respectively. and indexes i, c, p and d stand for ionic and covalent propagation and depropagation,Scheme 7. Conditions: respectively. [THF]0 = 8 mol·L−1; 25 °C [26,34], where τ is time, X is the number of monomer Polymerization of oxazolines (OX), with the interconversion between living and dormant molecules, and indexes i, c, p and d stand for ionic and covalent propagation and depropagation, species, has been elaborated in detail by Saegusa and Kobayashi [35,36]. More recently, kinetic respectively. Polymerizationanalysisof of these oxazolines systems has (OX), been given with by the Dworak interconversion [37]. between living and dormant species, The alkyl halides, in order to grow as such (covalent directly to covalent), must form the four- has been elaboratedPolymerizationmembered in detail transitionof oxazolines by state, Saegusa which (OX), andis forbidden with Kobayashi the according interconversion [35 to,36 the]. Moresymmetry between recently, rules. Thus,living kinetic for andthe analysis dormant of these systemsspecies, has beenhaspropagation been given elaborated byto proceed, Dworak in an detail [interconversion37]. by Saeg ofusa covalent-dormant and Kobayashi species [35,36]. into the More ionic onesrecently, is kinetic required. Ionization of covalent dormant species proceeds as for the polymerization of THF, almost Theanalysis alkyl of theseexclusively halides, systems intramolecularly in orderhas been to (Scheme given grow 8).by asDworak Dworak such has (covalent [37].calculated, that directly intermolecular to covalent),ionization with must form the four-memberedThe alkylparticipation transition halides, of in state,monomer, order which to in grow order is as forbiddento suchbe comp (covalentetitive, according would directly require to to the covalent),the symmetryimpossible must to rules.reach form the Thus, four- for the concentration of the monomer equal to ~103 mol/L. propagationmembered to proceed,transition an state, interconversion which is forbidden of covalent-dormant according to the species symmetry into therules. ionic Thus, ones for is required.the Ionizationpropagation of covalent to proceed, dormant an speciesinterconversion proceeds of as covalent-dormant for the polymerization species ofinto THF, the almostionic ones exclusively is required. Ionization of covalent dormant species proceeds as for the polymerization of THF, almost intramolecularly (Scheme8). Dworak has calculated, that intermolecular ionization with participation exclusively intramolecularly (Scheme 8). Dworak has calculated, that intermolecular ionization with of monomer,participation in order of monomer, to be competitive, in order to would be comp requireetitive, the would impossible requireto the reach impossible concentration to reach of the 3 monomerconcentration equal to of ~10 the monomermol/L. equal to ~103 mol/L.

Scheme 8. Polymerization of oxazoline with iodomethane [35,36].

In all these systems, the interconversions are fast comparing with the rates of growth. Thus, there are several interconversions for one monomer molecule that reacted in an act of propagation [37].

5. Active–Dormant Interconversion in Polymerization of Cyclic Esters The first catalytic system, successfully used in the polymerization of cyclic esters, was based on metal alkoxides (e.g., aluminum isopropoxide) and the polymerization could be considered as living

SchemeScheme 8. 8.Polymerization Polymerization ofof oxazoline with with iodomethane iodomethane [35,36]. [35,36 ].

In all these systems, the interconversions are fast comparing with the rates of growth. Thus, there Inare several all these interconversions systems, the for interconversionsone monomer molecu arele that fast reacted comparing in an act with of propagation the rates [37]. of growth. Thus, there are several interconversions for one monomer molecule that reacted in an act of propagation5. Active–Dormant [37]. Interconversion in Polymerization of Cyclic Esters The first catalytic system, successfully used in the polymerization of cyclic esters, was based on 5. Active–Dormant Interconversion in Polymerization of Cyclic Esters metal alkoxides (e.g., aluminum isopropoxide) and the polymerization could be considered as living The first catalytic system, successfully used in the polymerization of cyclic esters, was based on metal alkoxides (e.g., aluminum isopropoxide) and the polymerization could be considered as living and controlled. One of the first comprehensive papers discussing this issue has been published by the Liège group of Teyssié, Jerome and Dubois [38]. Aluminum isopropoxide exists in solutions as a mixture of a trimer and tetramer (Scheme9). Polymers 2017, 9, 644 8 of 18

and controlled. One of the first comprehensive papers discussing this issue has been published by Polymers 2017, 9, 646 8 of 18 the Liège group of Teyssié, Jerome and Dubois [38]. Aluminum isopropoxide exists in solutions as a mixture of a trimer and tetramer (Scheme 9).

Polymers 2017, 9, 644 8 of 18

and controlled. One of the first comprehensive papers discussing this issue has been published by the Liège group of Teyssié, Jerome and Dubois [38]. Aluminum isopropoxide exists in solutions as a mixture of a trimer and tetramer (Scheme 9).

Scheme 9. Trimer (A3) and tetramer (A4). Scheme 9. Trimer (A3) and tetramer (A4). There was a long-lasting controversy, whether all alkoxy groups initiate or only some of them. This controversy has been resolved in our institute, as we have shown in the polymerization of ε- There wascaprolactone a long-lasting (CL) that controversy, in non-protonic solvents whether and at all 25 alkoxy°C in the mixture groups of a initiate trimer and or tetramer only some of them. only trimer (cf. Scheme 9) initiates. When monomer conversion is complete, the tetramer remains This controversyintact has [39].been On the resolvedother hand, Kricheldorf in our institute,used higher temperatures as we have (close shown to 100 °C) in which the caused polymerization of ˝ ε-caprolactoneall (CL) the alkoxy that groups in non-protonic to be active (although solvents the proportions and of at trimer 25 andC tetramer in the were mixture not known) of a trimer and tetramer only trimer[40]. This (cf. means Scheme that at these9) initiates.conditions “dormant When tetramer” monomer has enough conversion time to be is converted complete, into the tetramer reactive trimer before polymerization is over. The difference in reactivities of trimer and tetramer Scheme 9. Trimer (A3) and tetramer (A4). ˝ remains intact [stems39]. Onfrom thethe otherfact, that hand, in trimer Kricheldorf the central Al used atom higher is pentacoordinated temperatures and in (close tetramer to 100 C) which hexacoordinated. Hexacoordinated atom does not have any coordination sites left to coordinate the caused all theThere alkoxy was groupsa long-lasting to be controversy, active (although whether all thealkoxy proportions groups initiate of or trimer only some and of tetramerthem. were monomer (cf. Scheme 9). Apparently, coordination with the central atom is a necessary condition for not known)This [40 controversyinitiation.]. This This means has may been account, that resolved as at Szwarc these in our pointed conditions institute, out while as we discussing “dormant have shown this system tetramer” in the [25] polymerization (p. 123), has for enough the of ε- time to be convertedcaprolactone into reactivethousand-fold (CL) trimer that faster in beforereaction non-protonic of polymerization trimer solvents [39]. and isat over.25 °C in The the differencemixture of ain trimer reactivities and tetramer of trimer and only trimer (cf. Scheme 9) initiates. When monomer conversion is complete, the tetramer remains tetramer stems fromIn many the fact,other systems, that in initiato trimerrs and/or the centralgrowing ionic Al atommacromolecules is pentacoordinated may form aggregates and in tetramer intact and[39]. only On unimers the other are hand, reactive Kricheldorf or one of the used forms higher of aggregates. temperatures Not knowing (close these to subtle100 °C) differences which caused hexacoordinated.all the andalkoxy Hexacoordinatednot realizinggroups tothe be presence active atom (althoughof inactive–dormant does the not proportions have isomers, any ofone trimer coordination may easilyand tetramer incorrectly sites were determine left not to known) coordinate the monomer[40]. (cf. SchemeThisconcentration means9). that Apparently,of active at these centers, conditions and, coordination as follows, “dormant the withinvolved tetramer” the rate centralhas constants enough atom of elementarytime is to a be necessary reactions. converted conditioninto for In the polymerization of cyclic esters formation of dormant aggregated macromolecules has reactive trimer before polymerization is over. The difference in reactivities of trimer and tetramer initiation. Thisbeen may first account, shown for asCL Szwarcand L-lactide pointed (LA), initiated out by while dialkylalkoxy discussing aluminums, this for system which a kinetic [25] (p. 123), for the stems from the fact, that in trimer the central Al atom is pentacoordinated and in tetramer thousand-fold fasterequation reaction has been developed. of trimer This [39 allows]. simultaneous determination of the degree of aggregation hexacoordinated.“m”, equilibrium Hexacoordinated constant and rate atom constants does of not propag haveation any [8]. coordination In this case, the sites extent left of toaggregation coordinate the In manymonomer othercan also (cf. systems, beScheme determined 9). initiators Apparently, first from the and/or coordination dependence growing of witheth log the ionic of centralrate of macromolecules polymerization, atom is a necessary rp on log may condition [P*]total form for aggregates and only unimersinitiation.([P*] aretotalThis =reactive may[I]0), account,where or [ oneP as*]total Szwarc ofis thethe pointed total forms concentration out of while aggregates. discussing of the active Not this knowingand system dormant [25] these (p.species— 123), subtle for the differences and not realizingthousand-foldinstantaneously the faster presence activereaction and of of inactive–dormantdormant trimer [41].[39]. The corresponding isomers, kinetic one system may is shown easily in Scheme incorrectly 10. determine InThe many specific other solution systems, of equations initiato describingrs and/or thisgrowing system ionicfor m =macromolecules 2 has been given bymay Szwarc form [25] aggregates (p. concentration of108) active and the centers, general solution and, for as any follows, “m” has been the proposed involved later rate [8]. constants of elementary reactions. and only unimers are reactive or one of the forms of aggregates. Not knowing these subtle differences In theand polymerization not realizing the presence of cyclic of estersinactive–dormant formation isomers, of dormant one may aggregatedeasily incorrectly macromolecules determine has been firstconcentration shown for of CL active and centers,L-lactide and, as (LA), follows, initiated the involved by rate dialkylalkoxy constants of elementary aluminums, reactions. for which a kinetic equationIn the has polymerization been developed. of cyclic esters This format allowsion simultaneousof dormant aggregated determination macromolecules of the has degree of been first shown for CL and L-lactide (LA), initiated by dialkylalkoxy aluminums, for which a kinetic aggregation “m”, equilibrium constant and rate constants of propagation [8]. In this case, the extent equation has been developed. This allows simultaneous determination of the degree of aggregation of aggregation“m”, equilibrium can also beconstant determined and rate constants first from of propag the dependenceation [8]. In this of case, the the log extent of rate of aggregation of polymerization, rp on log [canP*] alsototal be([ Pdetermined*]total = [ firstI]0), from where the depend [P*]totalenceis of the the totallog of concentrationrate of polymerization, of the rp on active log [P*] andtotal dormant species—instantaneously([P*]total = Scheme[I]0), where 10. Kinetic active [P system*]total and describingis dormantthe total polymerization concentration [41]. with The dormant correspondingof theaggregated active macromolecules and kinetic dormant and system species— is shown in instantaneouslynon-aggregated active andactive dormant macromolecules. [41]. The corresponding kinetic system is shown in Scheme 10. Scheme 10. The specific solution of equations describing this system for m = 2 has been given by The specific solution of equations describing this system for m = 2 has been given by Szwarc [25] (p. Szwarc [25108)] (p. and 108) the andgeneral the solution general for solutionany “m” has for been any proposed “m” has later been [8]. proposed later [8].

Scheme 10. Kinetic system describing polymerization with dormant aggregated macromolecules and Scheme 10. Kinetic system describing polymerization with dormant aggregated macromolecules and non-aggregated active macromolecules. non-aggregated active macromolecules.

If, based on Scheme 10, we assume that:

+ + + m[(Pn )m] >> [Pn ], then m[(Pn )m] « [I]0 Polymers 2017, 9, 644 9 of 18

If, based on Scheme 10, we assume that:

m[(Pn+)−m] >> [Pn+], then m[(Pn+)m] ≈ [I]0 As the rate of the polymerization equals:

rp = −dln[M]/dt (1) The combination of these two equations leads to (full derivation is given in [8]):

rp1−m = −m·Ka/kpm−1 + kp·[I]0·rp−m (2)

Thus, “m” is known, and kp and Ka can be determined from Equation (2) by plotting rp1−m (the left hand side) as a function of [I]0·rp−m (the second term of the right hand side of this equation). PolymersPlots2017 ,showing9, 646 the dependence of polymerization rate on the initial concentration of an initiator9 of 18 in the polymerization of ε-caprolactone (CL) with various R2AlOR’alcoholates are given in Figure 2. AsDepending the rate of on the the polymerization structure of active equals: species and conditions of polymerization, propagation proceeds on the species that either retain their integrity independently of concentration (green squires) or partially aggregate into the temporarilyrp = ´dln[ inactiveM]/dt species as described by the Equation (2). (1) The ability to aggregate and the size of the aggregates depend on the entourage of the active species.The Thus, combination for R2AlOR’ of these as an two initiator, equations there leads is aggregation to (full derivation into trimer is given when in R [ 8=]): C2H5, into a dimer for i-C4H9 and no aggregation when active species are solvated by amino ligands. The absence of 1´m m´1 ´m aggregation for propagation onr pAl(i-OPr)= ´m3 (not¨Ka/ shown)kp + iskp particularly¨[I]0¨rp spectacular, since the initiator(2) itself is highly aggregated (cf. above) and disaggregates upon propagation, when i-OPr groups on m k K r 1´m the AlThus, atoms “ ”are is known,converted and intop and macromolecules.a can be determined from Equation (2) by plotting p (the left I ¨r ´m handThe side) presented as a function method of [ ](Equation0 p (the (2)) second of analysis term of of the polymerizations right hand side with of this aggregation equation). of active centersPlots that showing leads to the dormancy dependence allowed of polymerization analyzing the rate already on the published initial concentration data not only of an for initiator ROP but in ε thealso polymerization for some vinyl ofpolymerizations-caprolactone (Figure (CL) with 3). various R2AlOR’alcoholates are given in Figure2.

FigureFigure 2.2. Bilogarithmic dependencies ofof thethe relativerelative ratesrates ofof polymerizationpolymerization ((rrpp)) ofof εε-caprolactone-caprolactone initiatedinitiated withwith dialkyaluminumdialkyaluminum alkoxidesalkoxides onon thethe startingstarting concentrationsconcentrations ofof initiatorinitiator ([([I])I])0,0, showingshowing influenceinfluence of of aggregation aggregation on onthe kinetics the kinetics of the polymerization. of the polymerization. Conditions: Conditions: THF solvent, THF 25 °C, solvent, [ε-CL] ˝ −1 ´1 25= 2 C,mol·L[ε-CL]. Points = 2 mol experimental¨L . Points for experimental initiators: blue for initiators: triangles (blue▲), (C triangles2H5)2AlOC (N),2H (C5;2 Hred5)2 circlesAlOC2 H(●5),; red[(CH circles3)2CHCH (‚), [(CH2]2AlOCH3)2CHCH3; black2]2AlOCH circles3; black (● circles), (C (‚2H),5 (C)2AlOCH2H5)2AlOCH2CH=CH2CH=CH2; green2; green squares squares ((■),), (C(C22HH55))22AlOCH2CH=CHCH=CH22/(CH/(CH3)NCH3)NCH2CH2CH2NH(C2NH(C2H25H) 5(1/2) (1/2 mole mole ratio). ratio). Experimental Experimental data data have beenbeen takentaken from from [ 41[41]]. [34].

Depending on the structure of active species and conditions of polymerization, propagation proceeds on the species that either retain their integrity independently of concentration (green squires) or partially aggregate into the temporarily inactive species as described by the Equation (2). The ability to aggregate and the size of the aggregates depend on the entourage of the active species. Thus, for R2AlOR’ as an initiator, there is aggregation into trimer when R = C2H5, into a dimer for i-C4H9 and no aggregation when active species are solvated by amino ligands. The absence of aggregation for propagation on Al(i-OPr)3 (not shown) is particularly spectacular, since the initiator itself is highly aggregated (cf. above) and disaggregates upon propagation, when i-OPr groups on the Al atoms are converted into macromolecules. Polymers 2017, 9, 646 10 of 18

The presented method (Equation (2)) of analysis of polymerizations with aggregation of active centers that leads to dormancy allowed analyzing the already published data not only for ROP but alsoPolymers for some 2017, vinyl9, 644 polymerizations (Figure3). 10 of 18 Polymers 2017, 9, 644 10 of 18

+ FigureFigure 3. 3.Green Green plot: plot: The Thepolymerization polymerization of of 2-methoxystyrene 2-methoxystyrene with with Li Li ion,+ ion, in toluene in toluene as a assolvent, a solvent, at Figure 3. Green plot: The polymerization of 2-methoxystyrene with Li+ ion, in toluene as a solvent, at at20 ˝°CC (the experimental data data taken from [42 [42]).]). Pink: Pink: An An anionic anionic polymerization polymerization of ofmethyl methyl 20 °C (the experimental+ data taken from [42]). Pink: An anionic polymerization of methyl methacrylate with Li+ cation, in THF as a solvent, at ´−65 °C˝ (the experimental data taken from [43]). methacrylate with Li cation,+ in THF as a solvent, at 65 C (the experimental data taken from [43]). Blue:methacrylate The polymerization with Li cation, of oxirane in THF on as − aCH solvent,2CH2O at−, Cs´−65+ ion°C+ (thepairs, experimental in THF as a datasolvent, taken at 70from °C [43]).(the˝ Blue: The polymerization of oxirane on ´CH2CH2O− , Cs+ ion pairs, in THF as a solvent, at 70 C experimentalBlue: The polymerization data taken from of oxirane [44]). Red: on − TheCH2 CHpolymerization2O , Cs ion pairs,of hexamethylcyclotrisiloxane in THF as a solvent, at 70 (D °C3) (theon (the experimental data taken from [44]). Red: The polymerization of hexamethylcyclotrisiloxane (D3) experimental− data+ taken from [44]). Red: The polymerization of hexamethylcyclotrisiloxane (D3) on −Si(CH3)2O , Li´ ion+ pairs, in THF as a solvent, at 22 °C (the˝ experimental data taken from [45]). Graph on ´Si(CH3)2O− ,+ Li ion pairs, in THF as a solvent, at 22 C (the experimental data taken from [45]). has−Si(CH been3) taken2O , Li from ion pairs,our paper in THF [8]. as a solvent, at 22 °C (the experimental data taken from [45]). Graph Graphhas hasbeen been taken taken from from our paper our paper [8]. [8]. In Figure 4, two chosen dependences from Figure 3 are analyzed according to Equation (2). The ratesIn FigureInof polymerizationFigure4, two4, two chosen chosen (with dependences dependen 1-m exponent)ces from are Figureplotted Figure 33 againstare are analyzed analyzed the total according accordingconcentration to Equation to Equationof the (2). active The (2). Thecentersrates rates of of (equalpolymerization polymerization to the starting (with (with 1-mconcentration 1-m exponent) exponent) ofare an plotted are initiator plotted against multiplied against the total the by concentration totalrp−m). concentrationThe straight of the activelines of the centers (equal to the starting concentration of an initiator multiplied by rp−m´).m The straight lines activeprovide centers kp from (equal intercepts to the starting and Ka concentration from the slopes. of an It initiatoris shown multiplied in the original by rp paper). The that straight improper lines provide kp from intercepts and Ka from the slopes. It is shown in the original paper that improper providevaluesk ofp fromm do interceptsnot give straight and K lines.a from the slopes. It is shown in the original paper that improper valuesvalues of mof dom do not not give give straight straight lines. lines.

Figure 4. Polymerization of: oxirane (a); and hexamethylcyclotrisiloxane-D3 (b). Conditions are in the Figure 4. Polymerization of: oxirane (a); and hexamethylcyclotrisiloxane-D (b). Conditions are in the descriptionFigure 4. Polymerization of Figure 3. Graphs of: oxirane have (beena); and taken, hexamethylcyclotrisiloxane-D as above, from [8]. 3 (b). Conditions are in the descriptiondescription of Figureof Figure3. Graphs3. Graphs have have been been taken, taken, as as above, above, from from [ 8[8].]. The determined in this way values of constants have been found practically equal to the values measuredThe determined by the authors in thisof the way original values papers of constants who were have using been numerical found practically method ofequal fitting. to the It proves values themeasured utility ofby the the linearauthors analytical of the original approach papers and who the werecorresponding using numerical equation method that givesof fitting. an access It proves to thethe analytical utility of descriptionthe linear analytical of the polymerization approach and with the acorresponding reversible formation equation of thatdormant gives aggregates. an access to the analytical description of the polymerization with a reversible formation of dormant aggregates.

Polymers 2017, 9, 646 11 of 18

The determined in this way values of constants have been found practically equal to the values measured by the authors of the original papers who were using numerical method of fitting. It proves thePolymers utility 2017 of, 9 the, 644 linear analytical approach and the corresponding equation that gives an access11 to of the 18 analytical description of the polymerization with a reversible formation of dormant aggregates. 6. Dormancy in Sn(Oct)2/ROH Catalyzed/Initiated Polymerization of Cyclic Esters: L-Lactide (LA) 6.and Dormancy ε-Caprolactone in Sn(Oct) (CL)2 /ROH Catalyzed/Initiated Polymerization of Cyclic Esters: L-Lactide (LA) and ε-CaprolactonePolymers 2017 (CL), 9, 644 11 of 18 Sn(Oct)2/ROH catalyzed/initiated polymerization of cyclic esters will be discussed separately for its importanceSn(Oct)2/ROH 6.in Dormancy a commercial catalyzed/initiated in Sn(Oct) polymerization2/ROH Catalyzed/Initiated polymerization of LA Polymerization in the of cyclic melt of Cyclic and esters Esters:a wide will L-Lactide be application discussed (LA) in separately research and ε-Caprolactone (CL) forworks its importanceon a cyclic inesters a commercial polymerization polymerization (altogethe ofr LAover in 1000 the meltpapers, and according a wide application to the Web in researchof Sci.). Sn(Oct)2/ROH catalyzed/initiated polymerization of cyclic esters will be discussed separately for works on a cyclic esters polymerization (altogether2 over 1000 papers, according to the Web of Sci.). In polymerizationits importance of cyclic in a esterscommercial Sn(Oct) polymerization as such of LA is in notthe melt active. and a wideIt gains application the inactivity research only after the Inreaction polymerization with, workse.g., of onalcohols, cyclica cyclic esters esters when polymerization Sn(Oct) it is converte2 (altogetheas suchrd over isfirstnot 1000 into active.papers, mono- according It gains and to the thefinally Web activity of Sci.).into onlydialkoxy after thetin In polymerization of cyclic esters Sn(Oct)2 as such is not active. It gains the activity only after the reaction(Sn(OR)2 with,) (cf. Scheme e.g., alcohols, 11). when it is converted first into mono- and finally into dialkoxy tin (Sn(OR)2) reaction with, e.g., alcohols, when it is converted first into mono- and finally into dialkoxy tin (cf. Scheme 11).(Sn(OR)2) (cf. Scheme 11).

Scheme 11. Sn(Oct)2, (Oct)SnOR, Sn(OR)2 equilibria. Scheme 11. Sn(Oct)2, (Oct)SnOR, Sn(OR)2 equilibria. Dialkoxy tinScheme formally behaves11. Sn(Oct) as other2, (Oct)SnOR, metal alkoxides Sn(OR) [38]. The2 equilibria. alcohol added to the system acts both as an initiator and as a chain transfer agent, which reversibly forms dormant species until Dialkoxy tin it becomes formally an end behavesgroup of a macromoleculeas other metal (Scheme alkoxidesal 12).koxides [[38].38]. The alcohol added to the system acts both asas anan initiatorinitiator andand asas aa chainchain transfer transfer agent, agent, which which reversibly reversibly forms forms dormant dormant species species until until it becomesit becomes an an end end group group of of a macromoleculea macromolecule (Scheme (Scheme 12 ).12).

Scheme 12. Chain transfer and creation of dormant species in the polymerization of ε-caprolactone

with the Sn(Oct)2/ROH system.

The [ROH]0/[Sn(Oct)2]0 ratio first strongly influences the rate of the polymerization until it reaches plateau. The established relationship is illustrated in Figure 5.

Scheme 12. Chain transfer and creation of dormant species in the polymerization of ε-caprolactone Scheme 12. Chain transfer and creation of dormant species in the polymerization of ε-caprolactone with the Sn(Oct)2/ROH system. with the Sn(Oct)2/ROH system.

The [ROH]0/[Sn(Oct)2]20] 0ratioratio first first strongly strongly influences influences the the rate rate of of the the polymerization until it reaches plateau. The established relationship is illustrated inin FigureFigure5 5.. Approximately tenfold excess of BuOH over Sn(Oct)2 is needed to convert all Sn(Oct)2 in the 5 system into active species. When no BuOH is added, Mn of the obtained polymer approaches 10 and apparently traces amounts of H O or other proton donors present in the system serve as a coinitiator. 2 According to Figure5, at first, the rate of the polymerization increases with increasing [ n-BuOH] Figure 5. Dependence of the rate of polymerization of L-Lacide (LA) ((rp = ln{([LA]0 − [LA]eq)/([LA] − (the active species[LA] areeq)/t) formed on [n-BOH]0 according) at constant concentration to Scheme of Sn(Oct) 112 =) 0.05 and mol·L then−1. Polymerization it reaches at 50 °C, a plateau when the concentration of activein THF solvent. sites The becomes numbers on invariant. the plot relate to Since Mn values the [46]. number of active species is constant and Pn is equal to the [monomer]0/[BuOH]0 ratio at complete conversion, the system is both living and controlled.

Figure 5. Dependence of the rate of polymerization of L-Lacide (LA) ((rp = ln{([LA]0 − [LA]eq)/([LA] −

[LA]eq)/t) on [n-BOH]0) at constant concentration of Sn(Oct)2 = 0.05 mol·L−1. Polymerization at 50 °C,

in THF solvent. The numbers on the plot relate to Mn values [46].

Polymers 2017, 9, 644 11 of 18

6. Dormancy in Sn(Oct)2/ROH Catalyzed/Initiated Polymerization of Cyclic Esters: L-Lactide (LA) and ε-Caprolactone (CL)

Sn(Oct)2/ROH catalyzed/initiated polymerization of cyclic esters will be discussed separately for its importance in a commercial polymerization of LA in the melt and a wide application in research works on a cyclic esters polymerization (altogether over 1000 papers, according to the Web of Sci.). In polymerization of cyclic esters Sn(Oct)2 as such is not active. It gains the activity only after the reaction with, e.g., alcohols, when it is converted first into mono- and finally into dialkoxy tin (Sn(OR)2) (cf. Scheme 11).

Scheme 11. Sn(Oct)2, (Oct)SnOR, Sn(OR)2 equilibria.

Dialkoxy tin formally behaves as other metal alkoxides [38]. The alcohol added to the system acts both as an initiator and as a chain transfer agent, which reversibly forms dormant species until it becomes an end group of a macromolecule (Scheme 12).

Scheme 12. Chain transfer and creation of dormant species in the polymerization of ε-caprolactone

with the Sn(Oct)2/ROH system.

PolymersThe2017 [ROH], 9, 6460/[Sn(Oct)2]0 ratio first strongly influences the rate of the polymerization until12 of 18it reaches plateau. The established relationship is illustrated in Figure 5.

Polymers 2017, 9, 644 12 of 18

Approximately tenfold excess of BuOH over Sn(Oct)2 is needed to convert all Sn(Oct)2 in the system into active species. When no BuOH is added, Mn of the obtained polymer approaches 105 and apparently traces amounts of H2O or other proton donors present in the system serve as a coinitiator. According to Figure 5, at first, the rate of the polymerization increases with increasing [n-BuOH] Figure 5. Dependence of the rate of polymerization of L-Lacide (LA) ((rp = ln{([LA]0 − [LA]eq)/([LA] − (the activeFigure 5.speciesDependence are formed of the rate according of polymerization to Scheme of L -Lacide11) and (LA) then ((r pit= reaches ln{([LA] 0a´ plateau[LA]eq)/([LA] when the [LA]eq)/t) on [n-BOH]0) at constant concentration of Sn(Oct)2 = 0.05 mol·L−1. Polymerization´1 at 50 °C, concentration´ [LA]eq)}/ oft) active on [n-BuOH] sites becomes0) at constant invariant. concentration Since the of Sn(Oct)number2 = of 0.05 active mol ¨speciesL . Polymerization is constant and at Pn in50 THF˝C, in solvent. THF solvent. The numbers The numbers on the onplot the relate plot to relate Mn values to M values[46]. [46]. is equal to the [monomer]0/[BuOH]0 ratio at complete conversion,n the system is both living and controlled.

Apparently, a similar (formall (formally)y) mechanism takes place for the polymerization of LA initiated/catalyzed by by a a strong strong base base (e.g., (e.g., 1,8-diaz 1,8-diazabicyclo[5.4.0]undec-7-eneabicyclo[5.4.0]undec-7-ene (DBU)) (DBU)) and and an an alcohol. alcohol. When alcohol is absent, then as it it has has been been shown shown by by Waymouth, Waymouth, macrocyclic macrocyclic PLA PLA are are formed formed [47,48]. [47,48]. Recently, a a comprehensive comprehensive study study of of the the lactide lactide LA LA polymerization polymerization with with DBU/alcohol DBU/alcohol has been has publishedbeen published [49] [42]. [49 This]. This study study encompasses encompasses the earl theier earlier works worksand shows, and in shows, agreement in agreement with previous with observations,previous observations, a simultaneous a simultaneous coexistence coexistence of an unreacted of an unreactedDBU, DBU DBU, H-bonded DBU H-bonded with growing with macromoleculesgrowing macromolecules and dormant and HO-ended dormant macromolecul HO-ended macromolecules.es. Authors have Authorsshown that have when shown sufficient that whenexcess sufficientof an alcohol excess over of DBU an alcohol is used, over the DBUprocess is involves used, the activated process alcohol involves pathway, activated in alcohol which growingpathway, species in which are growing presented species as activated are presented –OH asgrowing activated species –OH (Scheme growing 13). species However, (Scheme since 13). terminationHowever, since has terminationbeen found, has it beenis considered found, it isliv considereding and controlled living and only controlled in a certain only in region a certain of concentrationsregion of concentrations of monomer, of monomer, DBU and DBU an and alcohol, an alcohol, as well as well as only as only at atlimited limited conversion conversion and polymerization degree.

Scheme 13. PolymerizationPolymerization of LL-lactide-lactide with with DBU/alcohol DBU/alcohol system. system. Charges Charges (in (in place place of of partial partial charges) charges) are schematically shownshown forfor clarity clarity although although the the complete complete ionization ionization and and presence presence of ionsof ions have have not not yet yetbeen been experimentally experimentally established. established.

According to the postulated mechanism, the first order deactivation of the active According to the postulated mechanism, the first order deactivation of the active macromolecules macromolecules takes place, forming temporarily inactive, dormant, –OH ended macromolecules, takes place, forming temporarily inactive, dormant, –OH ended macromolecules, accompanied with accompanied with deactivation of DBU. deactivation of DBU. Macromolecules with –OH end-groups are dormant, since can be reactivated by DBU (Scheme 14). This is, however, for as long as there is still DBU present, since it is deactivated as shown above. At that stage, when there is no more DBU, formation of the –OH ended macromolecules (Scheme 14) would be a termination reaction.

Scheme 14. Active–dormant conversion in the polymerization of L-lactide with DBU/alcohol system.

Polymers 2017, 9, 644 12 of 18

Approximately tenfold excess of BuOH over Sn(Oct)2 is needed to convert all Sn(Oct)2 in the system into active species. When no BuOH is added, Mn of the obtained polymer approaches 105 and apparently traces amounts of H2O or other proton donors present in the system serve as a coinitiator. According to Figure 5, at first, the rate of the polymerization increases with increasing [n-BuOH] (the active species are formed according to Scheme 11) and then it reaches a plateau when the concentration of active sites becomes invariant. Since the number of active species is constant and Pn is equal to the [monomer]0/[BuOH]0 ratio at complete conversion, the system is both living and controlled. Apparently, a similar (formally) mechanism takes place for the polymerization of LA initiated/catalyzed by a strong base (e.g., 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU)) and an alcohol. When alcohol is absent, then as it has been shown by Waymouth, macrocyclic PLA are formed [47,48]. Recently, a comprehensive study of the lactide LA polymerization with DBU/alcohol has been published [49] [42]. This study encompasses the earlier works and shows, in agreement with previous observations, a simultaneous coexistence of an unreacted DBU, DBU H-bonded with growing macromolecules and dormant HO-ended macromolecules. Authors have shown that when sufficient excess of an alcohol over DBU is used, the process involves activated alcohol pathway, in which growing species are presented as activated –OH growing species (Scheme 13). However, since termination has been found, it is considered living and controlled only in a certain region of concentrations of monomer, DBU and an alcohol, as well as only at limited conversion and polymerization degree.

Scheme 13. Polymerization of L-lactide with DBU/alcohol system. Charges (in place of partial charges) are schematically shown for clarity although the complete ionization and presence of ions have not yet been experimentally established. Polymers 2017, 9, 646 13 of 18 According to the postulated mechanism, the first order deactivation of the active macromolecules takes place, forming temporarily inactive, dormant, –OH ended macromolecules, Macromoleculesaccompanied with deactivation –OH end-groups of DBU. are dormant, since can be reactivated by DBU (Scheme 14). This is, however,Macromolecules for as long aswith there –OH isend-groups still DBU are present, dormant, since since itcan is be deactivated reactivated by as DBU shown (Scheme above. At that 14). This is, however, for as long as there is still DBU present, since it is deactivated as shown above. stage, whenAt therethat stage, is no when more there DBU, is no more formation DBU, form ofation the –OHof the ended–OH ended macromolecules macromolecules (Scheme (Scheme 14) 14 ) would be a terminationwould be reaction. a termination reaction.

Polymers 2017, 9, 644 13 of 18

Related way of formation of dormant macromolecules has also been postulated for other strong Scheme 14. Active–dormant conversion in the polymerization of L-lactide with DBU/alcohol system. Schemebase + 14.alcoholActive–dormant systems [48] conversion as well inas thefor polymerization catalyst acting of simultaneouslyL-lactide with DBU/alcohol as initiator system. (INICAT) [50]. In Polymers 2017, 9, 644 13 of 18 the latter system, we postulated the following equilibrium with formation of dormant species as Related way of formation of dormant macromolecules has also been postulated for other strong shown inRelated Scheme way of15. formation of dormant macromolecules has also been postulated for other strong base + alcoholbase + systemsalcohol systems [48] as[48] well as well as as for for catalyst acting acting simultaneously simultaneously as initiator as (INICAT) initiator [50]. (INICAT) In [50]. In the latterthe system, latter system, we postulatedwe postulated the the following following eq equilibriumuilibrium with with formation formation of dormant of dormant species as species as shown in Schemeshown in Scheme15. 15.

SchemeScheme 15.SchemeThe 15. 15. TheThe proposed proposedproposed intramolecular intramolecular intramolecular mechanism mechanism mechanism of active–dormant of of active–dormant active–dormant interconversion interconversioninterconversion for 3-[(4,5- for for 3-[(4,5- dihydro-1H-imidazol-2-yl)amino]-propanol INICAT system (in [50] the intermolecular activation has 3-[(4,5-dihydro-1dihydro-1H-imidazol-2-yl)amino]-propanol INICAT INICAT system system (in [50] (in the [50 ]intermolecular the intermolecular activation has also been discussed). activationalso has been also discussed). been discussed). 7. Formation of Dormant Species in Polymerizations of Vinyl Monomers 7. Formation7. Formation ofIn this Dormant subsection,of Dormant Species formation Species in Polymerizationsof dormant in Polymerizations species in of the Vinyl active of MonomersVinyl Monomersinactive interconversion in vinyl polymerization will be briefly presented for comparison with ROP. In this subsection, formation of dormant species in the active inactive interconversion In this subsection,The best known formation example of of dormantsuch transformation species is in an the anionic active polymerization inactive of dienesinterconversion with Li+ in vinyl polymerizationin vinylcounterion polymerization discussed will be in briefly papers will beby presented brieflySzwarc, Bywaterpresented for comparison and forFetters comparison [25] with (p. ROP. 34). Thewith aggregation ROP. in this The bestsystemThe known bestgives known a example very good example ofexample such of transformationof such dormant transformation species isformation. an anionic is anHowever, anionic polymerization discussion polymerization used of to dienes be so of dienes with Li with+ Li+ counterioncounterionextensive discussed anddiscussed so in well papers known in papers by that Szwarc, we by will Szwarc, Bywaterrestrain Bywaterourselves and Fetters from and getting Fetters [25] into (p. [25] 34).any (p.details. The 34). aggregation The aggregation in this in this No less spectacular is the “Controlled Radical Polymerization: CRP” of vinyl monomers. In CRP, system gives a very good example of dormant species formation. However, discussion used to be so system givesstable a very(persistent) good radicals example are ofused dormant that are species able to react formation. with growing However, macroradicals, discussion forming used to be so extensiveextensive anddormant so and wellspecies, so known wellalthough known that are we unable that will weto restrain initiate will restpolymerization ourselvesrain ourselves from or to gettingreact from between getting into themselves. any into details. any The details. No lessinactivatingNo spectacular less spectacular agents isarethe in is thethe “Controlled so“Controlled called “nitroxide Radical Radical moderated Polymerization: Polymerization: polymerization” CRP”CRP” (mostly of of vinyl vinyl 2,2,6,6- monomers. monomers. In CRP, In CRP,stable stabletetramethylpiperidinyl-1-oxide (persistent) (persistent) radicals radicals are (TEMPO)) are used used itselfthat that anared areits ab numerousle able to toreact derivatives react with with (Schemegrowing growing 16) macroradicals,[51]. macroradicals, The forming creation of dormant macromolecules in CRP of styrene with TEMPO is presented in Scheme 17. formingdormant dormant species, species, although although are unable are unable to initiate to initiatepolymerization polymerization or to react or between to react themselves. between The themselves.inactivating The inactivatingagents are agentsin the areso called in the “nitroxide so called “nitroxidemoderated moderated polymerization” polymerization” (mostly 2,2,6,6- (mostlytetramethylpiperidinyl-1-oxide 2,2,6,6-tetramethylpiperidinyl-1-oxide (TEMPO)) (TEMPO))itself and its itself numerous and itsderivatives numerous (Scheme derivatives 16) [51]. The (Schemecreation 16)[51 ].of Thedormant creation macromolecul of dormant macromoleculeses in CRP of styrene in CRP with of TEMPO styrene withis presented TEMPO isin presentedScheme 17. in Scheme 17. Even more widely used is the process known as atom transfer radical Polymerization (ATRP), discovered in 1995 by MatyjaszewskiScheme [11 16.]. TEMPO—a This process stable radical. involves a reversible reaction of the macroradicals with CuII salts, that provides reversible radical complex which is neither able to initiate polymerization, nor to react with itself (like TEMPO), (Scheme 18).

Scheme 16. TEMPO—a stable radical.

Scheme 17. CRP of styrene with TEMPO [52].

Scheme 17. CRP of styrene with TEMPO [52].

Polymers 2017, 9, 644 13 of 18

Related way of formation of dormant macromolecules has also been postulated for other strong base + alcohol systems [48] as well as for catalyst acting simultaneously as initiator (INICAT) [50]. In the latter system, we postulated the following equilibrium with formation of dormant species as shown in Scheme 15.

Polymers 2017, 9, 644 13 of 18

Related way of formation of dormant macromolecules has also been postulated for other strong base + alcohol systems [48] as well as for catalyst acting simultaneously as initiator (INICAT) [50]. In the latter system, we postulated the following equilibrium with formation of dormant species as shown in Scheme 15.

Scheme 15. The proposed intramolecular mechanism of active–dormant interconversion for 3-[(4,5- dihydro-1H-imidazol-2-yl)amino]-propanol INICAT system (in [50] the intermolecular activation has also been discussed).

7. Formation of Dormant Species in Polymerizations of Vinyl Monomers Scheme 15. The proposed intramolecular mechanism of active–dormant interconversion for 3-[(4,5- In this subsection,dihydro-1H-imidazol-2-yl)amino]-propanol formation of dormant INICATspecies system in the (in [50] active the intermolecular inactive activation hasinterconversion in vinyl polymerizationalso been discussed). will be briefly presented for comparison with ROP. + The best7. Formation known ofexample Dormant of Species such intransformation Polymerizations isof anVinyl anionic Monomers polymerization of dienes with Li counterion discussed in papers by Szwarc, Bywater and Fetters [25] (p. 34). The aggregation in this In this subsection, formation of dormant species in the active inactive interconversion system givesin vinyl a very polymerization good example will be ofbrieflydormant presented species for comparison formation. with However, ROP. discussion used to be so extensive and Theso wellbest known known example that we of such will transformation restrain ourselves is an anionic from polymerization getting into of any dienes details. with Li + No lesscounterion spectacular discussed is the in papers“Controlled by Szwarc, Radical Bywater Polymerization: and Fetters [25] (p. CRP” 34). The of aggregationvinyl monomers. in this In CRP, stable (persistent)system gives radicals a very good are example used thatof dormant are abspeciesle to formation. react with However, growing discussion macroradicals, used to be so forming extensive and so well known that we will restrain ourselves from getting into any details. dormant species,No less although spectacular are is unable the “Controlled to initiate Radical polymerization Polymerization: CRP” or to of react vinyl monomers.between themselves.In CRP, The inactivatingstable agents (persistent) are inradicals the soare usedcalled that “nitroxide are able to reactmoderated with growing polymerization” macroradicals, forming(mostly 2,2,6,6- tetramethylpiperidinyl-1-oxidedormant species, although (TEMPO))are unable to itselfinitiate an polymerizationd its numerous or to react derivatives between themselves. (Scheme The16) [51]. The inactivating agents are in the so called “nitroxide moderated polymerization” (mostly 2,2,6,6- creation of dormant macromolecules in CRP of styrene with TEMPO is presented in Scheme 17. Polymers 2017tetramethylpiperidinyl-1-oxide, 9, 646 (TEMPO)) itself and its numerous derivatives (Scheme 16) [51]. The 14 of 18 creation of dormant macromolecules in CRP of styrene with TEMPO is presented in Scheme 17.

Scheme 16. TEMPO—a stable radical. SchemeScheme 16. 16. TEMPO—aTEMPO—a stable stable radical. radical.

Polymers 2017, 9, 644 14 of 18

Even more widely used is Schemethe process 17. CRP known of styrene as withatom TEMPO transfer [52]. radical Polymerization (ATRP), discovered in 1995 by Matyjaszewski [11] This process involves a reversible reaction of the macroradicals with CuII salts, that provides reversible radical complex which is neither able to initiate Scheme 17. CRP of styrene with TEMPO [52]. polymerization, nor to reactScheme with itself 17. (likeCRP TEMPO), of styrene (Scheme with TEMPO 18). [52].

SchemeScheme 18. Schematic 18. Schematic illustration illustration of the of key thereactions key reacti inons ATRP. in ATRP. (L stands (L forstands ligand—mostly for ligand—mostly derivatives of amines,derivatives and Mof amines, is Cu on and the M oxidation is Cu on the state oxidationn or n +state 1). n or n + 1).

The fast buildup of Mn+1·LX (e.g., CuIICl2/ligand) establishes the persistent radical effect and n+1 II Thecontrols fast termination buildup of [9,11]. M However,¨LX (e.g., in Cu radicalCl2/ligand) polymerizations establishes at least the some persistent termination radical by effect and controlscombination termination or disproportionation [9,11]. However, of the macroradicals in radical is polymerizations inevitable. Nevertheless, at least conditions some termination have by combinationbeen found allowing or disproportionation carrying the polymerization of the macroradicals to the full monomer is inevitable. conversion Nevertheless, with only conditionsa few havepercent been found of macromolecules allowing carrying terminated the polymerizationat the end of the to process. the full To monomer the same conversioncategory belongs with only a fewReversible percent addition-fragmentation of macromolecules terminatedchain-transfer at polymerization the end of the (RAFT), process. as described To the in same Reference category [52]. belongs Reversible addition-fragmentation chain-transfer polymerization (RAFT), as described in The cationic polymerization can also be a controlled process if the nucleophiles are introduced Reference [52]. into the system converting reversibly the instantaneously active into the dormant ones (Scheme 19). The cationic polymerization can also be a controlled process if the nucleophiles are introduced into the system converting reversibly the instantaneously active into the dormant ones (Scheme 19). This process, at certain conditions, is well controlled. It is not living, since carbocations easily undergo proton transfer (except at very low temperatures). The cationic vinyl polymerization with added nucleophiles has been comprehensively described in terms of reversible formation of dormant species in the Sigwalt’s laboratory [53] (Scheme 20). Indeed, Sigwalt has measured the ratio kp/ktr in systems with and without nucleophiles and have found the same values [54]. Thus, it has been clearly demonstrated that the only propagating species are usual carbocations and the other species, formed in reaction with nucleophiles, are just non-propagating, dormant macroions (oxonium, ammonium ions, etc.). These onium cations reversibly form back carbocations. Therefore,Scheme controlled19. The cationic polymerization, polymerization of closestyrene to with the added living, nucleophile. results from formation of dormant macromolecules in the reaction of macrocations with nucleophiles. Polymerization becomes much slowerThis andprocess, better at certain conditions conditions, for the is control well controlled. of the process It is not are living, provided. since carbocations easily undergo proton transfer (except at very low temperatures). The cationic vinyl polymerization with added nucleophiles has been comprehensively described in terms of reversible formation of dormant species in the Sigwalt’s laboratory [53] (Scheme 20). Indeed, Sigwalt has measured the ratio kp/ktr in systems with and without nucleophiles and have found the same values [54]. Thus, it has been clearly demonstrated that the only propagating species are usual carbocations and the other species, formed in reaction with nucleophiles, are just non- propagating, dormant macroions (oxonium, ammonium ions, etc.). These onium cations reversibly form back carbocations. Therefore, controlled polymerization, close to the living, results from formation of dormant macromolecules in the reaction of macrocations with nucleophiles. Polymerization becomes much slower and better conditions for the control of the process are provided. In this way, the original idea, that addition of nucleophiles (ethers, esters, sulfides) forms a new kind of active (i.e., growing) species has finally been put to rest. Thus, let us quote Szwarc, who had

Polymers 2017, 9, 644 14 of 18

Even more widely used is the process known as atom transfer radical Polymerization (ATRP), discovered in 1995 by Matyjaszewski [11] This process involves a reversible reaction of the macroradicals with CuII salts, that provides reversible radical complex which is neither able to initiate polymerization, nor to react with itself (like TEMPO), (Scheme 18).

Scheme 18. Schematic illustration of the key reactions in ATRP. (L stands for ligand—mostly derivatives of amines, and M is Cu on the oxidation state n or n + 1).

The fast buildup of Mn+1·LX (e.g., CuIICl2/ligand) establishes the persistent radical effect and controls termination [9,11]. However, in radical polymerizations at least some termination by combination or disproportionation of the macroradicals is inevitable. Nevertheless, conditions have been found allowing carrying the polymerization to the full monomer conversion with only a few percent of macromolecules terminated at the end of the process. To the same category belongs Reversible addition-fragmentation chain-transfer polymerization (RAFT), as described in Reference [52].

The cationicPolymers polymerization 2017, 9, 644 can also be a controlled process if the nucleophiles15 of 18 are introduced intoPolymers the2017 system, 9, 646 converting reversibly the instantaneously active into the dormant ones (Scheme15 19). of 18 written on this subject: “No new mechanism operates ‘living’ cationic polymerization. Neither does a new kind of species participate in these polymerizations” [25] p. 142). It has to be understood that onium ions are not propagating by themselves. Formation of dormant species has also been shown in styrene cationic polymerization initiated by protonic acids. Even though the cationic polymerization of styrene initiated by perchloric acid (HClO4) does not have the importance of ATRP, it has to be at least briefly mentioned, as it has triggered several further works involving dormant species. It shows additionally that the same system, depending on conditions (e.g., temperature), is living, living/dormant or just with termination. During the cationic polymerization of styrene initiated with HClO4, carried out at −97 °C, conversion of styrene stops at a certain conversion, that depends on the [styrene]0/[HClO4]0 ratio [55] [48]. One molecule of acid produces at this temperature one macromolecule. This result is explained by the collapse of the living ion pairs and formation of the unreactive polystyrylperchloric ester (it cannot e.g., ionize back, transfer a proton and form any species able to restart polymerization at this temperature). Choosing proper [styrene]0/[HClO4]0 ratio polymerization may go to completion, showing then typical features of living/controlled process. Thus, in terms of our pictograms (as above forSchemeScheme THF), there 19. is TheThe one cationicperiod cationic of growthpolymerization polymerization and, after collapse of styrene of counterions, with added the system nucleophile. is not active anymore. This process is a termination reaction (Scheme 20).

This process, at certain conditions, is well controlled. It is not living, since carbocations easily undergo proton transfer (except at very low temperatures). The cationic vinyl polymerization with added nucleophiles has been comprehensively described in terms of reversible formation of dormant species in the Sigwalt’s laboratory [53] (Scheme 20). Indeed, Sigwalt has measured the ratio kp/ktr in systems with and without nucleophiles and have found the same values [54]. Thus, it has been clearly demonstrated that the only propagating species are usual carbocations and the other species, formed in reaction with nucleophiles, are just non- propagating, dormant macroionsScheme 20. The (oxonium, polymerization ofammonium styrene with HClO ions,4 to partial etc.). conversion These only. onium cations reversibly Scheme 20. The polymerization of styrene with HClO4 to partial conversion only. form back carbocations.The perchloric Therefore, ester at this controlledtemperature is ponot lymerization,dormant, but just theclose termination to the product. living, If, results from formation of however,dormant the macromoleculesratio [styrene]0/[HClO4 ]0 inis properlythe re chosen,action then, of polymerizationmacrocations may withgo to nucleophiles. PolymerizationIn this way,completion. becomes the original At highermuch idea, temperatures slower that addition and(e.g., −better70 of°C), nucleophiles the conditions conversion of (ethers, formacroion the esters,intocontrol macroester sulfides) of isthe formsprocess a neware kind of active (i.e.,reversible growing) and polymerization species hascan be finally carried beenout to putcompletion—the to rest. Thus, formed letester us is quotea dormant Szwarc, who had provided. species. Nevertheless, at still higher temperatures (e.g., at r.t.) the proton transfer takes place and up writtenIn this on this way,to subject: 100the unsaturated original “Nonew macromolecules idea, mechanism that addition are formed operates offrom nucleophiles one “living” acid molecule. cationic (ethers, polymerization. esters, sulfides) Neither forms doesa new a new kind of speciesA similar participate system, namely in these with HOSO polymerizations”2CF3 as an initiator, has [25 been] (p. discussed 142). inIt details has by to Vairon, be understood that kind of active (i.e.,who hasgrowing) determined species the involved has lifetimes finally in beenthe stop put flow to experiments rest. Thus, [56]. leAtt some us quote conditions, Szwarc, who had onium ions arethere not is propagating a living polymerization by themselves. with dormant species, at higher temperatures termination takes place. Formation of dormant species has also been shown in styrene cationic polymerization initiated by 8. Conclusions protonic acids. Even though the cationic polymerization of styrene initiated by perchloric acid (HClO4) The major common feature of the presented processes, both in ROP and in some vinyl does not have thepolymerizations, importance is the of fast ATRP, interconversion it has to between be at active least and briefly dormant mentioned, species. Then, only as ita fraction has triggered several further works involvingof all macromolecules dormant can propagat species.e instantaneously. It shows additionally The total lifetime thatof an average the same macromolecule system, depending on is much higher with an introduction of dormant species, as it is a sum of the time of activity and conditions (e.g., temperature), is living, living/dormant or just with termination. dormancy. The periods of activity and dormancy are characterized by the ka/kd (rate constants of ˝ During theactivation/deactivation) cationic polymerization ratio that can be of altered styrene by experimentalist initiated by with choosing HClO proper4, components carried out at ´97 C, conversion of styreneof the system. stops An atimportant a certain result conversion, of the slowing down that is depends the ability onof all the macromolecules [styrene] to/[HClO start ] ratio [55]. polymerization at the same time leading to polymers of low dispersity if the rate of exchange growing0 4 0 One molecule of acid dormant produces is fast inat comparison this temperature with the rate of one propagation. macromolecule. In this way, polymerizations This result is explained by the collapse of the living ion pairs and formation of the unreactive polystyrylperchloric ester (it cannot e.g., ionize back, transfer a proton and form any species able to restart polymerization at this temperature). Choosing proper [styrene]0/[HClO4]0 ratio polymerization may go to completion, showing then typical features of living/controlled process. Thus, in terms of our pictograms (as above for THF), there is one period of growth and, after collapse of counterions, the system is not active anymore. This process is a termination reaction (Scheme 20). The perchloric ester at this temperature is not dormant, but just the termination product. If, however, the ratio [styrene]0/[HClO4]0 is properly chosen, then, polymerization may go to completion. At higher temperatures (e.g., ´70 ˝C), the conversion of macroion into macroester is reversible and polymerization can be carried out to completion—the formed ester is a dormant species. Nevertheless, at still higher temperatures (e.g., at r.t.) the proton transfer takes place and up to 100 unsaturated macromolecules are formed from one acid molecule. A similar system, namely with HOSO2CF3 as an initiator, has been discussed in details by Vairon, who has determined the involved lifetimes in the stop flow experiments [56]. At some conditions, there is a living polymerization with dormant species, at higher temperatures termination takes place. Polymers 2017, 9, 644 13 of 18

Related way of formation of dormant macromolecules has also been postulated for other strong base + alcohol systems [48] as well as for catalyst acting simultaneously as initiator (INICAT) [50]. In the latter system, we postulated the following equilibrium with formation of dormant species as shown in Scheme 15.

Polymers 2017, 9, 646 16 of 18

8. Conclusions The major common feature of the presented processes, both in ROP and in some vinyl polymerizations, is the fast interconversion between active and dormant species. Then, only a fraction of all macromolecules can propagate instantaneously. The total lifetime of an average Scheme 15. The proposed intramolecularmacromolecule mechanism of active–dormant is much higher interconversion with an introduction for 3-[(4,5- of dormant species, as it is a sum of the time dihydro-1H-imidazol-2-yl)amino]-propanol INICAT system (in [50] the intermolecular activation has of activity and dormancy. The periods of activity and dormancy are characterized by the ka/kd also been discussed). (rate constants of activation/deactivation) ratio that can be altered by experimentalist by choosing proper components of the system. An important result of the slowing down is the ability of all 7. Formation of Dormant Species in Polymerizations of Vinyl Monomers macromolecules to start polymerization at the same time leading to polymers of low dispersity if the In this subsection, formation of dormantrate ofspecies exchange in the growing active dormantinactive isinterconversion fast in comparison with the rate of propagation. In this in vinyl polymerization will be briefly presentedway, polymerizations for comparison knownwith ROP. for the long time as notoriously not living and without control are The best known example of such transformationconverted into is an systems anionic with polymerization these desired of features. dienes with Li+ counterion discussed in papers by Szwarc, BywaterMore and examples Fetters of[25] the (p. formation34). The aggregation of the dormant in this macromolecules in ROP and in vinyl system gives a very good example of dormantpolymerizations species formation. (e.g., inHowever, so-called discussion “group transfer used to polymerization”) be so could be given, however the extensive and so well known that we will discussedrestrain ourselves examples from are getting the most into representative any details. ones and the described phenomena also apply to No less spectacular is the “Controlledother Radical cases. Polymerization: CRP” of vinyl monomers. In CRP, stable (persistent) radicals are used that are able to react with growing macroradicals, forming dormant species, although are unable to initiateAcknowledgments: polymerizationThis or work to react was realizedbetween in themselves. a frame of the The project of National Science Centre Poland Grant No. DEC-2016/23/B/ST5/02448. inactivating agents are in the so called “nitroxide moderated polymerization” (mostly 2,2,6,6- Author Contributions: Stanislaw Penczek proposed the original subject of the paper and discussed the choice tetramethylpiperidinyl-1-oxide (TEMPO)) itself and its numerous derivatives (Scheme 16) [51]. The of examples. Julia Pretula has chosen a few examples and prepared the manuscript. Piotr Lewi´nskidiscussed creation of dormant macromolecules in CRPcritically of styrene all the with chosen TEMPO examples is presented and provided in Scheme some parts 17. of the manuscript. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Szwarc, M. Carbanions, Living Polymers and Electron Transfer Processes; Interscience Publishers: New York, NY, USA, 1968. 2. Szwarc, M.; Jagur-Grodzinski, J. Ions and Ion Pairs in Electron Transfer Reactions of Radical Anions, Carbanions and Solvated Electrons in Ions and Ion Pairs in Organic Reactions; Wiley: New York, NY, USA, 1974. Scheme 16.3. TEMPO—aSzwarc, M.stable Review radical. Lecture, Electron Transfer Reactions and Living Polymers. Proc. R. Soc. 1964, 279, 260–290. [CrossRef] 4. Szwarc, M. Living Polymers. Their Discovery, Characterization, and Properties. J. Polym. Sci. Part A Polym. Chem. 1998, 36, IX–XV. [CrossRef] 5. Grubbs, R.R.; Grubbs, R.H. 50th Anniversary Perspective: Living Polymerization-Emphasizing the Molecule in Macromolecules. Macromolecules 2017, 50, 6979–6997. [CrossRef] 6. Levy, M. Impact of the Concept of “Living Polymers” on Material Science. Polym. Adv. Technol. 2007, 18, 681–684. [CrossRef] 7. Khanna, S.N.; Levy, M.; Szwarc, M. Complexes Formed by Antracene with “Living” Polystyrene.

“Dormant Polymers”. Trans. Faraday Soc. 1962, 58, 747–761. [CrossRef] Scheme 17. CRP8. of Duda,styrene A.; with Penczek, TEMPO S. [52]. Determination of the Absolute Propagation Rate Constant in Polymerization with Reversible Aggregation of Active Centres. Macromolecules 1994, 27, 4867–4870. [CrossRef] 9. Greszta, D.; Mardare, D.; Matyjaszewski, K. “Living” Radical Polymerization. 1. Possibilities and Limitations. Macromolecules 1994, 27, 638–644. [CrossRef] 10. Veregin, R.P.N.; Georges, M.K.; Kazmaier, P.M.; Hamer, G.K. Free Radical Polymerizations for Narrow Polydispersity Resins: Electron Spin Resonance Studies of the Kinetics and Mechanism. Macromolecules 1993, 26, 5316–5320. [CrossRef] 11. Wang, J.-S.; Matyjaszewski, K. Controlled/“Living” Radical Polymerization, Atom Transfer Radical Polymerization in the Presence of Transition Metal Complexes. J. Am. Chem. Soc. 1995, 117, 5614–5615. [CrossRef] 12. Fisher, H. The Persistent Radical Effect in Controlled Radical Polymerization. J. Polym. Sci. Part A Polym. Chem. 1999, 37, 1885–1901. [CrossRef] Polymers 2017, 9, 646 17 of 18

13. Szwarc, M. “Living” Polymers. Nature 1956, 178, 1168–1169. [CrossRef] 14. Szwarc, M.; Milkovich, R.; Levy, M. Polymerization Initiated by Electron Transfer to Monomer. A new Method of Formation of Block Polymers. J. Am. Chem. Soc. 1956, 78, 2656–2657. [CrossRef] 15. Bodenstein, M. Eine Theorie der photochemischen Reaktionsgeschwindigkeiten. Z. Phys. Chem. 1913, 85, 390–421. [CrossRef] 16. Penczek, S.; Pretula, J.B. Fundamental Aspects of Chain Polymerization. In Polymer Science: A Comprehensive Reference; Matyjaszewski, K., Moeller, M., Eds.; Elsevier: Amsterdam, The Netherlands, 2012; Volume 3, pp. 4–37. 17. Semenov, N.N.; translated by Frankel, J.J. Chemical Kinetics and Chain Reactions; Clarendon Press: Oxford, UK, 1935. 18. Hinshelwood, C. Chemical Kinetics in the Past few Decades; Nobel Lectures, Chemistry; Elsevier: Amsterdam, The Netherlands, 1964. 19. Penczek, S.; Moad, G. Glossary of Terms Related to Kinetics Thermodynamics, and Mechanisms of Polymerization (IUPAC Recommendation 2008). Pure Appl. Chem. 2008, 80, 2163–2193. [CrossRef] 20. Jenkins, A.D.; Kratochvíl, P.; Stepto, R.F.T.; Suter, U.W. Glossary of Basic Terms in Polymer Science (IUPAC Recommendations 1996). Pure Appl. Chem. 1996, 68, 2287–2311. [CrossRef] 21. Szwarc, M.; van Beylen, M. Ionic Polymerization and Living Polymers; Chapman & Hall: London, UK, 1993. 22. Penczek, S.; Kubisa, P.; Szymanski, R. On the Diagnostic Criteria of the Livingness of Polymerization. Macromol. Chem. Rapid Commun. 1991, 12, 77–80. [CrossRef] 23. Matyjaszewski, K. Radical Polymerization. In Controlled and Living Polymerizations; Mueller, A.H.E., Matyjaszewski, K., Eds.; Wiley: Weinheim, Germany, 2009; Chapter 3. 24. Inoue, S. Immortal Polymerization: The Outset, Development and Application. J. Polym. Sci. Part A

Polym. Chem.Polymers 20172000, 9, 644 , 38, 2861–2871. [CrossRef] 13 of 18

25. Szwarc, M.RelatedIonic way Polymerization. of formation of dormant macromolecul Fundamentalses has also; been Introduction; postulated for other strong Hanser Publishers: Munich, Germany, 1996; base + alcohol systems [48] as well as for catalyst acting simultaneously as initiator (INICAT) [50]. In p. 17. the latter system, we postulated the following equilibrium with formation of dormant species as 26. Penczek,shown S.; in Szymanski, Scheme 15. R.; Duda, A. Polymerization with Contribution of Covalent and Ionic Species. Makromol. Symp. 1995, 98, 193–216. [CrossRef] 27. Flory, P. Molecular Size Distribution in Ethylene Oxide Polymers. J. Am. Chem. Soc. 1940, 62, 1561–1565. [CrossRef] 28. Flory, P. Fundamental Principles of Condensation Polymerization. Chem. Rev. 1948, 39, 137–197. [CrossRef] 29. Penczek, S. Cationic Ring-Opening Polymerization (CROP) Major Mechanistic Phenomena. J. Polym. Sci. Part A Polym. Chem. 2000, 38, 1919–1933. [CrossRef]

30. Makina, A.;Scheme Kobayashi, 15. The proposed S. intramolecular Chemistry mechanism of of2-oxazolines: active–dormant interconversion A Crossing for 3-[(4,5- of Cationic Ring-Opening Polymerization dihydro-1H-imidazol-2-yl)amino]-propanol INICAT system (in [50] the intermolecular activation has and Enzymaticalso been Ring-Openingdiscussed). Polymerization. J. Polym. Sci. Part A Polym. Chem. 2010, 48, 1251–1270.

[CrossRef7. ]Formation of Dormant Species in Polymerizations of Vinyl Monomers 31. Matyjaszewski,In this subsection, K.; Penczek, formation of S. dormant The species Macroester in the active inactiveMacroion interconversion Equilibrium in the Cationic Polymerization of in vinyl polymerization will be briefly presented1 for comparison with ROP. THF ObservedThe best Directly known example by of 300 such transformation MHz H is NMR. an anionic J.polymerization Polym. Sci.of dienes Polym. with Li+ Chem. Ed. 1974, 12, 1905–1912. [CrossRef] counterion discussed in papers by Szwarc, Bywater and Fetters [25] (p. 34). The aggregation in this 32. Buyle, A.M.;system Matyjaszewski,gives a very good example of K.; dormant Penczek, species formation. S. Kinetics However, discussion and Thermodynamics used to be so of Interconversion of Macroesters extensive and so well known that we will restrain ourselves from getting into any details. and MacroionNo less Pairsspectacular in is the Cationic “Controlled Radical Polymerization Polymerization: CRP” ofof vinyl Tetrahydrofuran. monomers. In CRP, Macromolecules 1977, 10, 269–274. stable (persistent) radicals are used that are able to react with growing macroradicals, forming [CrossRefdormant] species, although are unable to initiate polymerization or to react between themselves. The inactivating agents are in the so called “nitroxide moderated polymerization” (mostly 2,2,6,6- 33. Matyjaszewski,tetramethylpiperidinyl-1-oxide K.; Kubisa, (TEMPO)) P.; Penczek, itself and its S.numerous Ion derivativesEster (Scheme Equilibrium 16) [51]. The in the Living Cationic Polymerization of Tetrahydrofuran.creation of dormant macromoleculJ. Polym.es Sci.in CRP Polym.of styrene with Chem. TEMPO Ed.is presented1974 in ,Scheme12, 1333–1336.17. [CrossRef] 34. Penczek, S.; Kubisa, P.; Matyjaszewski, K.; Szymanski, R. Structures and Reactivities in the Ring Opening and Vinyl Cationic Polymerization. In Cationic Polymerizations and Related Processes; Goethals, E.J., Ed.;

Academic Press: New York, NY, USA, 1984; p. 184.

35. Saegusa, T.; Chujo, Y. MacromolecularScheme 16. TEMPO—a stable Engineering radical. on the Basis of the Polymerization of 2-Oxazolines. Makromol. Chem. Macromol. Symp. 1991, 51, 1–10. [CrossRef] 36. Kobayashi, S. Polymerization of Oxazolines. In Polymer Science: A Comprehensive Reference; Penczek, S., Grubbs, R.H., Eds.; Elsevier: Amsterdam, The Netherlands, 2012; Volume IV, pp. 397–423. 37. Dworak, A.; Trzebicka, B.; Kowalczuk, A.; Tsvetanov, C.H.; Rangelov, S. Polyoxazolines–Mechanism of

Synthesis and Solution Properties.Scheme 17. CRP of Polimerystyrene with TEMPO2014 [52]. , 59, 88–94. [CrossRef]

Polymers 2017, 9, 646 18 of 18

38. Dubois, P.; Jerome, R.; Teyssie, P. Macromolecular Engineering of Polylactones and Polylactides. 7. Structural i Analysis of Copolymers of ε-Caprolactone and L- or D,L-Lactide initiated by Al(O Pr)3. Macromolecules 1992, 25, 37–44. [CrossRef] 39. Kowalski, A.; Duda, A.; Penczek, S. Polymerization of L,L-Lactide Initiated by Aluminum Isopropoxide Trimer and Tetramer. Macromolecules 1998, 31, 2114–2122. [CrossRef] 40. Kricheldorf, H.R.; Besl, M.; Schvarnagl, N. Poly(lactones). 9. Polymerization Mechanism of Metal Alkoxide Initiated Polymerizations of Lactide and Various Lactones. Macromolecules 1988, 21, 286–293. [CrossRef] 41. Duda, A.; Penczek, S. Kinetics of Polymerization Involving Reversible Deactivation due to Aggregation of Active Centres—Analytical vs. Numerical—Solution for the Epsilon-Caprolactone Dialkylalkoxide Aluminum Systems. Macromol. Rapid Commun. 1994, 15, 559–566. [CrossRef] 42. Geerts, J.; van Beylen, M.; Smets, G. Anionic polymerization of o- and p-methoxystyrene. J. Polym. Sci. Part A Polym. Chem. 1969, 7, 2859–2873. [CrossRef] 43. Kunkel, D.; Mueller, A.H.E.; Janata, M.; Lochman, L. The role of association/complexation equilibria in the anionic polymerization of (meth)acrylates. Makromol. Chem. Macromol. Symp. 1992, 60, 315–326. [CrossRef] 44. Kazanskii, K.S.; Solovyanov, A.A.; Entelis, S.G. Polymerization of ethylene oxide by alkali metal-naphthalene complexes in tetrahydrofuran. Eur. Polym. J. 1971, 7, 1421–1433. [CrossRef] 45. Wilczek, L.; Kennedy, J.P. Aggregation in the Anionic Polymerization of Hexamethyl cyclotrisiloxane with Lithium Counterion. Polym. J. 1987, 19, 531–538. [CrossRef] 46. Kowalski, A.; Duda, A.; Penczek, S. Kinetics and Mechanism of Cyclic Esters Polymerization Initiated with Tin(II)Octoate. 3. Polymerization of L,L-Dilactide. Macromolecules 2000, 33, 7359–7370. [CrossRef] 47. Brown, H.A.; De Crisci, A.G.; Hedrick, J.L.; Waymouth, R.M. Amidine-Mediated Zwitterionic Polymerization of Lactide. ACS Macro Lett. 2012, 1, 1113–1115. [CrossRef] 48. Lohmeijer, B.G.; Pralt, R.C.; Leibforth, F.; Logan, J.W.; Long, D.A.; Dove, A.P.; Nedesberg, F.; Choi, J.; Wade, C.; Waymouth, R.M.; et al. Guanidine and Amidine Organocatalysts for Ring-Opening Polymerization of Cyclic Esters. Macromolecules 2006, 39, 8574–8583. [CrossRef] 49. Scherck, N.J.; Kim, H.Ch.; Won, Y-Y. Elucidating a Unified Mechanistic Scheme for the DBU—Catalyzed Ring-Opening Polymerization of Lactide to Poly(lactic acid). Macromolecules 2016, 48, 4699–4713. [CrossRef] [PubMed] 50. Lewinski, P.; Sosnowski, S.; Penczek, S. L-Lactide Polymerization—Living and Controlled—Catalyzed by Initiators: Hydroxy alkylated Organic Bases. Polymer 2017, 108, 265–271. [CrossRef] 51. Jenkins, A.D.; Jones, R.G.; Moad, G. Terminology for Reversible—Deactivation Radical Polymerization Previously Called “Controlled” Radical or “Living” Radical Polymerization. Pure Appl. Chem. 2010, 82, 483–491. [CrossRef] 52. Moad, G.; Solomon, D.H. The Chemistry of Radical Polymerization, 2nd ed.; Elsevier: Oxford, UK, 2005. 53. Matyjaszewski, K.; Sigwalt, P. Unified Approach to Living and Nonliving Cationic Polymerization of Alkenes. Polym. Int. 1994, 35, 1–26. [CrossRef] 54. Sigwalt, P.; Poulton, A.; Tardi, M. Transfer and Termination Reactions in Living Carbocationic Polymerization. J. Macromol. Sci. 1994, 31, 953–968. [CrossRef] 55. Darcy, L.E.; Millerine, W.P.; Pepper, D.C. Molecular Weights in the Cationic Polymerization of Styrene by Perchloric Acid. Chem. Commun. 1968, 22, 1441–1442. [CrossRef] 56. Charleux, B.; Rives, A.; Vairon, J.-P.; Matyjaszewski, K. Stopped Flow Investigations of Trifluoromethane Sulfonic Acid Initiated Cationic Oligomerization of Trans-1,3-diphenyl-1-butene. 2. A Model Kinetic Study of Styrene Cationic Polymerization. Macromolecules 1998, 31, 2403–2408. [CrossRef]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).