HIGHLIGHT

Anionic Polymerization: High Vacuum Techniques

NIKOS HADJICHRISTIDIS, HERMIS IATROU, STERGIOS PISPAS, MARINOS PITSIKALIS Department of Chemistry, University of Athens, Panepistimiopolis Zografou, 15771 Athens, Greece

Received 5 June 2000; accepted 8 June 2000

ABSTRACT: Anionic polymerization is a powerful tool for the synthesis of a variety of model materials with well-dened molecular characteristics. However specially designed apparatuses and appropriate high vacuum techniques are needed in order to exclude from the reaction environment all reac-

tive contaminants with the anionic centers. This review describes the basic principles of anionic polymerization as well as detailed experimental methods for the purication of the reagents usually used for the synthesis of model polymeric materials. In addition a few examples of the synthesis of poly-

mers with complex architecture are given. 2000 John Wiley & Sons, Inc. J
Polym Sci A: Polym Chem 38: 32113234, 2000

Keywords: anionic polymerization; molecular architecture; living polymers; glass blowing

FROM LEFT TO RIGHT: H. IATROU, S. PISPAS, M. PITSIKALIS, N. HADJICHRISTIDIS

Hermis Iatrou: B. Sc. University of Athens, Greece (1989); Ph.D. University of Athens, Greece (1993); Postdoctoral Research: Institute of Material Science in the Center of Nuclear Science in Juelich, Germany with Prof. Dr. Dieter Richter (1994 1995) and University of Alabama at Birmingham, USA with Prof. Jimmy Mays (August 1997February 1998); Research Associate, University of Athens (since 1998); 27 papers in refereed scientic journals; 20 announcements in international scientic conferences. Stergios Pispas: B. Sc. University of Athens, Greece (1989); Ph.D. University of Athens, Greece (1994); Postdoctoral Research: University of Alabama at Birmingham, USA with Prof. J. W. Mays (1994 1995); Research Associate, University of Athens (since 1997); Visiting Scientist: Foundation for Research and Technology, Greece (November 1990), University of Alabama at Birmingham, (AugustSeptember 1992 and SeptemberOctober 1993), Max Plank

Correspondence to: N. Hadjichristidis (E-mail: hadjichristidis@ chem.uoa.gr)

Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 38, 32113234 (2000) 2000 John Wiley & Sons, Inc.

3211

3212

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000)

Institute for Polymer Science, Germany (February 1994), University of Massachusetts at Amherst, (July September 1995), IBM Almaden Research Center, USA (October 1995), Foundation for Research and Technology, Greece (July 1998); American Institute of Chemists Foundation Award for Distinguished Postdoctoral Research (1995); 35 papers in refereed scientic journals; 24 announcements in international scientic conferences. Marinos Pitsikalis: B. Sc. University of Athens, Greece (1989); Ph.D. University of Athens, Greece (1994); Postdoctoral Research: University of Birmingham, USA with Prof. J. W. Mays (19951996); Lecturer, Industrial Chemistry Laboratory, Department of Chemistry, University of Athens (since 1998); Visiting Scientist: University of Milano (March 1991), University of Alabama at Birmingham, USA (SeptemberOctober 1993), Max Plank Institute for Polymer Science, Germany (August 1994), National Institute for Standards and Technology, USA (December 1995), IBM Almaden Research Center, USA (February 1995), 30 papers in refereed scientic journals, 24 announcements in international scientic conferences. Nikos Hadjichristidis: B. Sc. University of Athens, Greece (1966); Ph.D. University of Liege, Belgium (1971); ` D. Sc. University of Athens Greece (1978); Postdoctoral Research: University of Liege with Professor V. Desreux ` (19711972) and National Research Council of Canada with Dr. J. Roovers (19721973); Lecturer (1973), Assistant Professor (1982), Associate Professor (1985), Full Professor (1988), Director of Industrial Chemistry Laboratory (since 1994) and Chairman of the Chemistry Department (19911995 and since 1999) of the University of Athens; Visiting Scientist, University of Liege (Summers 1974, 1975); Visiting Research ofcer, ` NRC of Canada (Summer 1976); Visiting Scientist, University of Akron, at Dr. L. Fetters Laboratory (Summers 1977 through 1982); Distinguished Visiting Scientist, NRC of Canada (1983); Visiting Professor, Exxon Research Engineering Co., NJ, since 1984, every year 12 months. President of the European Polymer Federation (19951996); Member of the National Advisory Research Council since 1994; President of the State Highest Chemical Board since 1995; Director of the Institute of Organic and Pharmaceutical Chemistry of the National Hellenic Research Foundation since 2000; Academy of Athens Award for Chemistry (1989); Empirikion Award for Sciences (1994) and Greek Chemists Association Award (2000); more than 200 papers in refereed scientic journals.

INTRODUCTION
Anionic polymerization offers powerful methods for the controlled manipulation of macromolecular architecture15 and is still today unique for the synthesis of a wide variety of materials with a high degree of molecular and compositional homogeneity.6 12 This advantage makes living anionic polymerization very attractive for the synthesis of the model polymers that are needed for establishing the relationships between structure and properties. However, the high reactivity of anionic centers towards atmospheric (moisture, oxygen, carbon dioxide) and other contaminants dictates specially designed apparatuses and appropriate techniques, in order to avoid premature living chain termination. Both inert atmosphere and high vacuum techniques have been successfully employed for anionic polymerization. Inert atmosphere techniques are simpler and lead more easily to the preparation of large quantities of materials. In contrast high vacuum techniques are very demanding in glass blowing practice, more time consuming and usually lead to the synthesis of only a few grams of model polymeric materials. Nevertheless, these limitations

for high-vacuum techniques are a small price to pay given their potential for generating complex macromolecular model compounds. In 1960, Worsfold and Bywater13 described an apparatus for carrying out anionic polymerization under high vacuum environment. In 1974, Fetters and Morton,12 in a detailed review, presented apparatuses and reactors used for high vacuum techniques. Nevertheless, the high vacuum methodology given in these two articles is applied mainly to the synthesis of simple structures, for example, linear and symmetric star polymers. Since 1977 the high vacuum technique has been further developed and modied for the synthesis of polymeric materials with complex macromolecular architecture. In this review besides the detailed description of the new apparatuses and reactors, a few examples for the synthesis of complex architectures are given.

FUNDAMENTALS OF ANIONIC POLYMERIZATION

Anionic polymerization is a living polymerization method due to the absence, in most cases, of any spon-

HIGHLIGHT

3213

taneous termination reaction. In order to obtain polymers with predictable molecular weights and narrow molecular weight distributions it is required to have a homogeneous reaction throughout initiation and propagation, to have an initiation rate higher than the rate of propagation and to exclude all possible terminating impurities from the system. Under these conditions true model polymers can be prepared. Monomers The monomers that are susceptible to anionic polymerization are those that can form stable carbanionic species under polymerization conditions. The most studied monomers are styrenic, dienic, and also cyclic that can react with nucleophiles leading to ring opening.1,2,14,15 The double bond must have substituents that can stabilize the negative charge, by charge withdrawing making the anions stable to possible nucleophilic attack from other species. Consequently strong electrophilic groups or relatively acidic proton donating groups, such as amino-, carboxyl-, hydroxyl, or acetylene functional groups will interfere during anionic polymerization and thus must be excluded from the monomer unit or be protected to a suitable derivative. In contrast aromatic rings, double bonds, carbonyl, ester, cyano, sulfone groups, and so forth, stabilize the negative charge and promote the anionic polymerization. The existence of polar substituents, such as carbonyl, cyano, or nitro groups leads to a complex situation, because they may react with the initiator or with the propagating anionic species. The living polymerization of monomers bearing such polar substituents is possible only under very specic conditions (e.g., low polymerization temperature, use of bulky initiators, selected counterion, etc.). In many cases termination reactions occur leading to poor control over the molecular weight and the molecular weight distribution. Therefore the anionic polymerization of acrylonitrile, vinyl silanes, halogenated monomers, isocyanates, nitroalkenes, and so forth, does not proceed in a living manner. Monomers that have been successfully polymerized under anionic polymerization conditions are styrene and styrene derivatives ( -methyl styrene, p-methyl styrene, t-butyl styrene, etc.),16 20 dienes (e.g., isoprene, butadiene, etc.)10,2125, (meth)acrylates (at low temperatures, using bulky initiators to avoid side reactions with the carbonyl group),26 30 vinyl pyridines3133 (at low temperatures, due to the reactivity of the pyridine ring), cyclic siloxanes,34 epoxides,35 and lactones.36,37 Initiators The most versatile and useful anionic initiators are the alkyllithium compounds, since most of them are com-

mercially available, they can be rather easily prepared by the reaction of the corresponding alkyl chloride with Li metal, they are soluble in a variety of solvents, including hydrocarbons and most importantly they are efcient initiators for anionic polymerization. The unique characteristic of the organolithium compounds, among the other organic compounds with alkali metals, is that the COLi bond exhibits properties of both covalent and ionic bonds.38,39 This is a consequence of the fact that lithium, compared to the other alkali metals has the smallest radius, the highest electronegativity and the highest ionization potential.40 In addition rather low energy unoccupied p orbitals are available for bonding. The covalent character of the COLi bond along with the strong aggregation of the ionic pairs41 43 is responsible for the higher solubility of the organolithium compounds in hydrocarbon solvents, compared to the solubility of the anions with other alkali metals as a counter ion. However, it was found that in the solid state and in solution these compounds form aggregates. The structure of the organic moiety greatly inuences the degree of aggregation. Unhindered alkyllithium compounds form hexameric aggregates in hydrocarbon solvents, but when the alkyl group has a branching point at the - or -carbon then the aggregates are tetrameric. The degree of association is also inuenced by the nature of the solvent, the solution concentration and temperature. In general the degree of association is decreased by decreasing concentration, by using a strongly solvating solvent (the aromatic solvents are stronger solvating agents than aliphatic solvents), by increasing the temperature and by using an organic group capable to delocalize electrons.44 49 It was shown that the reactivity of the alkyllithium initiators is directly connected with their degree of association: the lower the degree of association the higher the reactivity of the initiator.50,51 This is evident by the data, concerning the relative reactivities of various alkyllithium initiators versus the degree of association (given in parenthesis) for styrene and diene polymerizations, given below:52 menthyllithium 2 sec-BuLi 4 i-PrLi (4-6) n-BuLi (6)

t-BuLi 4

Alkyllithium initiators are usually preferred for the polymerization of styrene and dienes, but are too reactive for the polymerization of methacrylates, because they attack the carbonyl group. For this reason less reactive initiators26 30 have been used successfully, for example, diphenylhexyllithium (DPHLi), prepared by the reaction of n-BuLi with 1,1-diphenylethylene, DPE:

3214

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000)

Kinetics of Anionic Polymerization Another class of initiators are the radical anions formed by the reaction of alkali metals with aromatic hydrocarbons in polar aprotic solvents.53 The most common case is the reaction of naphthalene with Na metal in THF:

Initiation Reaction
The initiation kinetics of styrene and dienes using alkyllithium initiators in hydrocarbon solvents are the best studied examples in anionic polymerization. The classic work of Worsfold and Bywater60 concerning the initiation reaction of styrene polymerization with n-BuLi in benzene led to the following relationship: Ri n-BuLi]1/6 [St]

These radical anions react with monomers by reversible electron transfer to form the corresponding monomer radical anions. The example with styrene is given below:

The fractional kinetic order of the initiation rate on the total initiator concentration implies that the active species for the polymerization are a small fraction of dissociated initiator molecules, produced by the following equilibrium: (n-BuLi)6 6n-BuLi Similar kinetic order dependencies have been observed in other systems in aromatic solvents. However this conclusion was questioned on the basis of theoretical results and experimental measurements of the energy of activation of the initiation reaction, which appears to be too low to include also the energy of the complete dissociation of the aggregates.61 63 Consequently it seems more likely that an incomplete or stepwise dissociation exists leading to aggregates with lower degrees of association.64 These intermediate aggregates are not totally inactive towards the polymerization of the monomer. The situation is much more complex in aliphatic solvents, since the inverse correspondence between the reaction order and the degree of alkyllithium aggregation is not always observed.65 Consequently it seems that the polymerization starts exclusively from the initiator aggregates. A further complication arises from the appearance of cross-associated structures between the organolithium compounds and growing living polymer chains having benzylic or allyliclithium end groups.66 68 Generally, initiation is faster in aromatic than in aliphatic solvents and is accelerated by the addition of Lewis bases, because they facilitate the dissociation of the alkyllithium aggregates. Propagation Reaction The propagation kinetics for styrene polymerization with lithium as the counterion has been studied in both aro-

It is obvious from this reaction scheme that the radical anions lead to difunctional growing chains, useful for the synthesis of triblock copolymers, cyclic polymers, telechelics, and more complicated structures as the H-, super H- and -shaped polymers. Similar difunctional initiators can be prepared by the reaction of alkali metals with monomers having rather low ceiling temperatures, such as -methylstyrene. Tetrameric dianions are the main product from the reaction of -methylstyrene with Na metal54 and dimeric dianions from the corresponding reaction with Na/K alloy.55 The fact that these initiators are formed only in polar solvents limits their utility for the polymerization of dienes, since the polar solvents do not favor the formation of the often desired 1,4 microstructure of the polydienes. To solve this problem difunctional initiators soluble in hydrocarbon solvents are needed. The major problem is that these initiators are usually highly associated in hydrocarbon solvents, leading to the formation of insoluble products, which are not reactive initiators. One of the most effective difunctional initiators, soluble in hydrocarbon media, is the one formed by the addition reaction of sec-BuLi with 1,3-bis(1-phenyl-ethenyl)benzene.56 59

HIGHLIGHT

3215

Figure 1. Vacuum line.

matic and aliphatic solvents yielding the same equation,69,70 shown below: Rp [PStLi]1/2 [St]

where [PStLi] and [St] are the living polymer and the monomer concentrations, respectively. In analogy with the initiation kinetics the one-half kinetic order dependence on the living polymer concentration is due to the association of the living polymer chains into dimers in the hydrocarbon solvents. These dimers are not reactive for monomer addition and consequently a dissociation step to unassociated living chains is required. A more complex situation was observed for the propagation kinetics of dienes. In the polymerization of both isoprene and butadiene a rst order dependence on the monomer concentration and fractional order dependencies on the concentration of the living chain ends were observed. The fractional order dependence ranges from one-half to one-fourth or even one-sixth.7175 Considering the dissociation model, as previously mentioned, the degree of association of the active chain ends has to be dened. However, contradictory results have been reported in the literature. It can be concluded that the degree of association of the polydienyllithium chains is at least two depending on several parameters, such as the molecular weight of the chain, the temperature, and the nature of the solvent. Tetrameric, hexameric, and even larger structures76 have been also reported. These results are not always compatible with the observed propagation rates and thus the exact mechanism cannot be unambiguously described.

reagents used. All impurities capable of deactivating the initiator and the propagating chain ends (moisture, oxygen, etc) must be excluded from the polymerization mixture as much as possible if full use of the living character of the anionic systems is desired. For this purpose the use of a high vacuum line is the usual choice (Fig. 1). The main parts of a high vacuum line are the oil pump (A), the mercury diffusion pump (B), the liquid nitrogen trap (C), the upper (D) and lower (E) glass tube rigs, and the stopcocks. The oil pump is used in order to reduce the pressure in the system in a level where mercury or polysiloxane vaporization by heating through a heating mantle is possible. After mercury distillation is achieved the mercury diffusion pump can bring the pressure to a value around 10 5 mmHg. The use of a turbopump alone is also possible. A liquid nitrogen trap is used to condense any volatile substances incorporated in the system and protect in this way the oil and the diffusion pumps from contamination. The upper rig is separated from the lower one by high vacuum teon stopcocks and it is used to connect the parts of the lower rig to the vacuum separately. This way it is possible to use the smaller parts of the lower rig separately and perform different procedures (e.g., distillations, degassing) simultaneously. Reactors and other glass apparatuses are joined to the vacuum line through the lower ring via high vacuum o ring stopcocks and ground joints. The taps are constructed from a teon piston and a glass outer part, which are connected by viton or teon o rings. A Tesla coil can be used in order to check the quality of the vacuum. When the vacuum line is properly evacuated, the sound of the Tesla coil is the least noisy when it is close to the glass. Glassware Polymerization reactors, dilution apparatuses, and ampoules are constructed from Pyrex glass asks and tub-

HIGH VACUUM TECHNIQUES

High Vacuum Line For the synthesis of polymers by anionic polymerization special care should be exercised in the purication of the

3216

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000)

tained, as it is evidenced from the sound of the Tesla coil, the stopcock is closed and the liquid is thawed. The freeze-thaw cycle is repeated for two or three times. In this way during every cycle a new equilibrium is attained between the gasses dissolved in the liquid and the part of the ask above the liquid. The gases occupying the free space are driven out of the liquid to the vacuum line while the liquid is maintained in the solid state.

Distillation
Since high vacuum levels are attained in the vacuum line, distillation of liquids (solvents, monomers, etc.) can be accomplished simply by heating the round bottom ask containing the liquid and freezing the receiving ask. The two asks are usually connected to the same part of the lower rig in neighboring stopcocks in order to minimize the distillation pathway. Before the beginning of the distillation the liquid is degassed. While the liquid is thawed the receiving ask is pumped down thoroughly and brought to liquid nitrogen temperature. The part of the lower rig is isolated from the high vacuum by closing the appropriate stopcocks and distillation begins. The temperature of the ask containing the liquid is limited by the ability of the grease on the ground joint to remain in a gel form. After the whole volume of the liquid has been collected in the receiving ask, this ask is connected to the vacuum by opening the appropriate stopcock and the frozen liquid is degassed. At the end of degassing the stopcock connecting the ask to the lower rig is closed and the liquid is thawed and allowed to be stirred.

Figure 2. Breaker with break-seal and constriction.

ing. The clean glassware is rinsed with deionized water several times and dried in an oven at 150 C. Then the equipment is constructed and it is let to be annealed in a furnace at 520 C in order to remove tensions in the glass. At the beginning of the experimental procedure the apparatus is evacuated on the vacuum line, checked for possible pinholes with a Tesla coil and after it is certain that no holes are present it is ame dried with a hand torch, using a soft ame, under continuous pumping. In this way volatile contaminants present on the inner surface of the glass can be removed. For the introduction of reagents under vacuum conditions into the main reactor, break-seals are used. They are constructed with a thin glass tip, which is easily ruptured with a cylindrical magnet encapsulated in an evacuated Pyrex glass tube (Fig. 2). These breakers can be moved from outside the apparatus by moving the magnets. Removal of the reagents is performed by heat sealing of constrictions (Fig. 2). Simple High Vacuum Procedures

Short Path Distillation

In the case where very high boiling liquids (usually monomers) have to be distilled, a special glass rig is constructed (Fig. 3). The apparatus is connected to the vacuum line and checked for pin holes. Then the liquid to be distilled is transferred into ask (A), by disconnecting the apparatus from the line or via an ampoule already attached to the rig (B), degassed on the vacuum line and then distilled into calibrated ampoules (C). Purification of Reagents

Degassing
In order to remove oxygen and other gasses from a liquid system the simple procedure of degassing is employed. The ask containing the liquid is connected to the vacuum line through a stopcock. The stopcock is opened two or three times for a short period of time (seconds) until bubbles are produced in its bulk. The liquid is then frozen, with the stopcock closed, using a liquid nitrogen or dry ice/isopropanol bath and then the stopcock is opened and sufcient time is allowed for the whole system to be pumped down. After high vacuum is at-

Solvents
Benzene. Benzene is usually puried by stirring over concentrated sulfuric acid for a week in a conical ask inside a hood. This procedure removes the thiophenes, substituted phenyl-compounds like toluene and vinyl compounds, that may alter the microstructure of polydienes, or participate in the polymerization. Then it is washed with an aqueous solution of NaOH, and after

HIGHLIGHT

3217

Figure 3. Apparatus for short-path distillation of high-boiling point substances.

with water many times until it becomes neutral and nally is dried with CaCl2. Then it is transferred carefully in a round bottom ask containing fresh nely grounded CaH2 and a magnetic stir bar, attached to the vacuum line and degassed. It is left for reaction of CaH2 with moisture overnight. Then it is again degassed and distilled in a calibrated cylinder containing n-BuLi and styrene. The persistence of the bright orange color of PSLi produced after the thawing of benzene indicates the purity of the solvent. Depending on the moisture levels of the solvent, an intermediate drying step over sodium or sodium dispersion, before the distillation onto PSLi is sometimes necessary. Toluene. This solvent is puried in a way similar to benzene except from the exposure to sulfuric acid. Alkanes. Alkanes (hexane, cyclohexane, etc.) are puried in a manner similar to benzene. Tetrahydrofuran. THF is usually reuxed over Na, under N2 atmosphere in the presence of benzophenone, until a bright deep purple color is attained, before being collected in a round-bottom ask containing fresh nely grounded CaH2. This ask is then connected to the vacuum line, the solvent is degassed and distilled in a ask containing a Na/K (1/3 by weight) alloy. The bright blue color, which develops after stirring for some time indicates that the solvent is free from impurities. Alternatively, THF can be puried by stirring over CaH2 on the vacuum line overnight, subsequent distillation over sodium mirrors until no degradation of the mirror is observed and nally distillation onto Na/K alloy or polystyrylsodium (produced by Na and styrene) or diphenylhexyllithium (produced by the reaction of 1,1-diphenylethylene and n-BuLi). In the latter two cases

the purity of THF is indicated by the permanent redorange color of PSNa or the deep red color of DPHLi. Ethers. Other ethers are puried by ways similar to THF. Usually stirring over CaH2 followed by exposure to Na mirrors is a procedure that ensures pure compounds. Amines. Purication of amines (usually tertiary amines like triethylamine, tetramethylethylene diamine, or dipiperidine ethane) presents some problems. Fortunately these compounds are used in small amounts as modiers of the kinetics of polymerization or the microstructure of polydienes. The procedure usually employed is composed of thorough drying over CaH2, exposure to several sodium mirrors and nal distillation into ampoules. The use of BuLi should be avoided because it reacts with amines.

Monomers
Styrene. This high boiling monomer (bp: 145 C) is dried overnight, on the vacuum line, over CaH2 in a short path distillation apparatus similar to the one shown in Figure 3. Then it is distilled in large ( 50 mL) ampoules equipped with break-seals. One of these ampoules is attached to another short path distillation rig (Fig. 3), an appropriate amount of dibutylmagnesium (DBMg) 1 M in heptane ([monomer]/[DBMg solution] 20/1) is introduced via the side arm in ask (A), and the arm is sealed with a torch ame at (a). The Mg compound having one n-butyl and one iso-butyl group is preferred due to its higher solubility in hydrocarbon solvents. After removal of heptane in the line break-seal (b) is ruptured and the ampoule is removed by simultaneously pumping down the frozen styrene in ask (A). The monomer is

3218

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000)

Figure 4. Apparatus for the purication of the 1,1-diphenylethylene and 1,3-bis(1phenylethenyl)benzene.

thawed and allowed to stand overnight with continuous stirring. A bright yellow-greenish color develops as the pure monomer starts to polymerize. Polymerization of styrene with DBMg compounds proceeds very slowly. The next day the pure monomer can be distilled in the precalibrated ampoules (C). Other Styrenic Monomers. Other styrenic monomers (tbutylstyrene (bp: 91 C/9mmHg), p-methylstyrene (bp: 170 C), -methylstyrene (bp: 167 C), DPE (bp: 270 C), etc. can be puried in the same way as styrene. A short path distillation apparatus is used in all cases. In the case of -methylstyrene n-BuLi can be used, alternatively to DBMg, since this monomer has a low ceiling polymerization temperature and no polymerization occurs at room temperature. The purity of the monomer is determined from the red color of the anions formed. Diphenylethylene can also be puried in the nal step by exposure to n-BuLi, as described below, since no polymerization addition is possible due to steric hindrance. 1,1 Diphenylethylene (DPE). The apparatus used for the purication of DPE (Fig. 4) is checked for pinholes, a small amount of DPE is loaded in ask (A) and a slight ow of pure nitrogen is inserted through the ground joint (B). The outlet of the gas is the needle placed in septum (C). Then, DPE is titrated by introducing a solution of n-BuLi in hexane, by a syringe through the septum under stirring, until a deep red color is permanently formed. The ow of the gas is stopped, the needle and the syringe are removed from the septum and the apparatus is immediately evacuated and the septum is removed by sealing constriction (D). After all the hexane has been dis-

tilled off, ask (A) is immersed in an oil bath and the temperature is set at 85 C. DPE starts to distill and when the appropriate amount is collected in the cylinder (F), the ask is removed by heat sealing constriction (E). The appropriate amount of puried solvent is distilled in the cylinder in order to dilute DPE, and the apparatus is removed from the vacuum line by sealing at (H). Depending on the amount needed, the solution of DPE is placed in ampoules (G), removed from the main apparatus by heat sealing the constriction and stored at 20 C. Isoprene (bp:34 C). This diene is dried over nely grounded CaH2 on the vacuum line overnight. It is then distilled over n-BuLi where it remains with continuous stirring for 30 min at 0 C. This step is repeated as many times needed until a light yellow color remains and the viscosity is slightly increased, before the pure monomer is distilled in precalibrated ampoules equipped with break-seals. It can be stored at 20 C for six months. Butadiene (bp: 4 C). This monomer is a gas at room temperature so rst it must be condensed in a graduated cylinder that contains n-BuLi at 78 C (dry ice/isopropanol bath [Fig. 5, (A)]. The monomer is stirred for 30 min at 10 C (ice/salt bath). Then it is distilled into another ask (B) containing new n-BuLi and it is allowed to stand for another 30 min at 10 C until the viscosity is slightly increased, due to polymerization. It is nally distilled at 78 C in a calibrated ampoule already attached to the polymerization apparatus. After the distillation is complete the polymerization apparatus is sealed off the vacuum line, the break-seal of the ampoule is ruptured while the butadiene is kept at 78

HIGHLIGHT

3219

Figure 5. Purication of butadiene and distillation to the ampoule prior to the polymerization.

C, then distilled in the main apparatus containing the solvent and subsequently polymerized. Alternatively, it is possible to make an ampoule of butadiene by diluting the monomer with puried solvent in a ratio of [monomer]/[solvent] lower than 0.2. 2-Vinylpyridine: (bp: 158 C). This high boiling monomer is dried over CaH2 in a short path apparatus overnight and then it is distilled into ampoules. The monomer free from stabilizer can be stored at 20 C for about a week. The nal purication steps involve exposure to Na mirror for a few minutes in an apparatus similar to the one shown in Figure 3 and subsequent distillation in another ampoule. The ampoule is then connected to a new short path apparatus with a side arm for the introduction of trioctylaluminum. After stirring for a few minutes over trioctylaluminum the puried 2-vinylpyridine is distilled in a precalibrated ampoule, which is then attached to the polymerization reactor. Polymerization of this monomer is effected the same day, following the nal two purication steps, since the monomer obtained is so pure that it polymerizes by standing overnight at 20 C.77 Methacrylates and Acrylates. These monomers are usually puried as given in ref. 78. Commercial monomer (usually 98 99%) is dried over CaH2 overnight in a short path apparatus attached to the vacuum line (Fig. 3). The next day the monomer is distilled into calibrated ampoules. The inhibitor free monomer can be stored at 20 C for one week. It is nally puried just prior to polymerization by a short exposure to triethyl (TEA) or

trioctylaluminum (TOA) using a short path distillation apparatus. As soon as a yellow-greenish color persists in the bulk of the monomer this is considered pure enough for anionic polymerization and it is distilled in a precalibrated ampoule, which is then attached to the polymerization reactor. Traces of alcohols present from the synthesis of the monomer are removed during this last exposure since they react readily with the aluminum compounds. TOA is usually preferred with the higher boiling monomers of the family (i.e., tert-butylmethacrylate) since TEA can be sublimed under high vacuum into the monomer ampoules. TOA is also more efcient in removing secondary or tertiary alcohols. The yellowgreenish color is due to the complex formed between the aluminum compound and the carbonyl group of the monomer and sometimes may be misleading since complexation may occur before complete reaction with the remaining alcohol. Ethylene Oxide (bp:10 C). This low boiling monomer is initially condensed like butadiene in a graduated cylinder containing CaH2. It is stirred over CaH2 for about 30 min at 0 C and distilled over n-BuLi. During its exposure to n-BuLi EO is kept at 0 C with stirring (for 30 min). This step is repeated twice using new n-BuLi each time and an exposure period of 30 min. Finally the pure monomer is distilled into graduated ampoules and stored at 20 C till needed.79 Hexamethylcyclotrisiloxane. This monomer (D3) is a solid (melting point: 70 C) and this poses some problems in its handling under high vacuum condi-

3220

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000)

tions. Therefore it is diluted with puried benzene and stirred over CaH2 overnight in a specially designed short path distillation apparatus (Fig. 4 without the tubes for split-down the distilled species). Then the solvent along with the monomer are sublimed into an ampoule by heating at around 40 C. The ampoule is attached to a new short path apparatus connected to the vacuum line and PSLi is introduced (Fig. 4). The monomer is allowed to stand in contact with PSLi for about 2 h at room temperature and then it is distilled along with the solvent into the cylinder and split-down in precalibrated ampoules.80 Other siloxane monomers can be puried in a similar manner. Terminating Agents. The compounds used for deactivation of the living anions are usually alcohols. Methanol is widely used as the terminating agent. This compound is degassed on the vacuum line several times and it is distilled into ampoules. Other terminating agents used in special cases, that is, (CH3)3SiCl can be puried by a simple degassing since they are used in large excess compared to the concentration of the living ends. Coupling Agents. The coupling agents widely used in anionic polymerization are chlorosilanes. The low molecular weight compounds of this family (Me2SiCl2, SiCl4, MeSiCl3) are liquids and they can be easily puried by distillation from CaH2, where they are usually allowed to stand overnight with continuous stirring. The next day the volatile silane compound is distilled into calibrated ampoules. A middle fraction is collected. Solid silanes are dissolved usually in hexane and sublimed in the vacuum line from CaH2 by using a short path distillation apparatus. The solid coupling agents like bromo- or chloromethyl derivatives of benzene can be sublimed under vacuum. Solid commercial products of high purity can be dried on the vacuum line. They are placed in a dilution apparatus, which is left overnight on the vacuum line by continuous pumping. Subsequently they are diluted by distilling the appropriate solvent in the apparatus. Care should be taken when diluting silanes with functionality greater than two. SiOCl groups can react with impurities on the glass walls of the dilution apparatus (i.e., SiOOH groups). In this case the dilution apparatus is washed with a solution of (CH3)3SiCl in benzene or hexane in a similar way to the s-BuLi case (see Fig. 7). Then the solution is collected in the purge section cooled with liquid nitrogen and the purge section is separated from the rest of the rig by sealing off the appropriate constriction. The apparatus is connected again to the vacuum line via a break-seal. The break-seal is ruptured and the last traces of volatile monochlorosilane are re-

moved in the vacuum line overnight. The next day the required amount of solvent is distilled in and the apparatus is separated from the line by heat sealing the appropriate constriction.

PREPARATION OF ORGANOLITHIUM INITIATORS

Synthesis of Monofunctional Initiators The most commonly used anionic initiator is sec-BuLi. This initiator is commercially available, but it is less thermally stable than n-BuLi. The reported rate of decomposition is 1.4% per month at 20 C.9 Therefore it is better to synthesize this initiator and keep it in solution at low temperature. Under these conditions the concentration does not change appreciably over a period of several months. sec-BuLi is prepared by the reaction of sec-BuCl with Li metal according to the reaction:81,82 secBuCl 2 Li 3 secBuLi LiCl

The sec-BuCl ( 5 mL) is dried over CaH2 overnight on the vacuum line under continuous stirring, degassed several times and nally distilled into a calibrated ampoule, which is then attached to the main apparatus, used for the synthesis of the initiator, as shown in Figure 6. The reactor is attached to the vacuum line, checked for pinholes and ame-dried. After pumping for sufcient time a lithium dispersion in hexane containing 2% sodium is placed in ampoule (A) by breaking the vacuum for a short time. A 710 fold excess of Li over the sec-BuCl is used. Then the hexane is evaporated through the vacuum line. A small quantity (35 mL) of n-BuLi solution 1.6 M in hexane is introduced through septum (B) by a syringe, for purging of the reactor. After removal of the hexane, constriction (C) is heat-sealed. Puried hexane (typically 20 30 mLs) is distilled through the vacuum line into ask E and the reactor is sealed at (D). The ampoule (A) is washed with the n-BuLi solution in order to transfer the Li metal quantitatively to ask (F) and is sealed. The whole reactor is washed with the n-BuLi solution, to react with the impurities on the glass and is rinsed thoroughly, to remove the excess n-BuLi, by condensing hexane on the walls. Finally the hexane is distilled from ask (E) to ask (F) and the purge section is heat sealed at (I). Flask (F) is immersed in an icewater bath. Afterwards the break seal (G) is ruptured and sec-BuCl is left to distill slowly into ask (F) containing the Li dispersion in hexane under continuous stirring. After distillation is completed the reaction is allowed to continue for

HIGHLIGHT

3221

According to this method a sample of the sec-BuLi solution is directly hydrolyzed and titrated and another sample of the same solution is reacted with benzyl chloride in 1,2-dibromomethane, followed by acid titration as well. With the second titration the residual alkalinity is determined due to any materials initially present in the solution which do not contain CLi bonds.

Polymerization of Styrene
A known amount of styrene, puried according to the standards of anionic polymerization is polymerized by a known amount of initiator solution. The active concentration is calculated by means of the equation: moles initiator gr monomer/Mn

where Mn is the number-average molecular weight of the polystyrene formed, determined by size exclusion chromatography or preferably by membrane osmometry for better accuracy. Dilution of the Initiator
Figure 6. Apparatus used for the preparation of the monofunctional initiator sec-BuLi, from sec-BuCl and lithium metal.

another 4 h under stirring and nally ask (F) is left in an icewater bath at 0 C overnight. The reaction mixture is then ltered through the glass lter to remove the unreacted Li and the sec-BuLi solution in hexane is collected into ampoules (H) for subsequent use. Analysis of the Initiator The active concentration of the initiator can be determined by several methods. The simplest and most accurate methods are the following:

Titration Method
A known amount of the sec-BuLi solution is hydrolyzed, by breaking the ampoule and immediately pouring the initiator solution in excess water. The resulting LiOH is titrated with aqueous HCl solution to a phenolphthalein end point. Alternatively the double titration method of Gilman and Haubein83,84 can be used.

The concentration of the sec-BuLi solution obtained from the synthesis is usually between 11.6 M. This means that it usually has to be diluted to the desired concentration and split-down into calibrated ampoules. The apparatus used for the dilution and the split-down of the initiator is shown in Figure 7. Ampoule (A) containing the concentrated sec-BuLi solution is sealed on the apparatus. It is connected to the vacuum line through ground joint (B) and it is pumped down, checked for pinholes and ame-dried. A small amount of the purging agent, that is, 35 mL of n-BuLi solution 1.6 M in hexane is introduced through septum (C) by a syringe. After removal of hexane, constriction (D) is heat-sealed. The required amount of puried hexane is distilled through the vacuum line into ask (E), the reactor is sealed at (F) and removed from the line. The apparatus is washed with the n-BuLi solution and then it is rinsed thoroughly, by condensing hexane on the walls. Finally the hexane is distilled from ask (E) into the calibrated cylinder (G) and the purge section is cut off at (H). The break-seal (A) is ruptured and the concentrated sec-BuLi solution is allowed to drain into cylinder G. Then ampoule (A) is rinsed several times with hexane so that all traces of the initiator are transferred to the cylinder and it is removed from the apparatus by heat sealing at (J). The dilute solution of the initiator is distributed over the ampoules and the cylinder by manual manipulations. When the desired amount of initiator solution is collected constrictions (K) are

3222

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000)

Figure 7. Apparatus for the dilution of reagents.

rinsed with solvent to remove the initiator traces and both the ampoules and the cylinder are removed by heat sealing and are stored at 20 C for subsequent use. The amount of initiator left in the cylinder can be further split-down or diluted in a similar apparatus like in Figure 7.

Preparation of the Difunctional Initiator (1,3phenylene)bis(3-methyl-1-phenylpentylidene) Dilithium Initiator (DLI) The reactions used for the preparation of DLI57 are shown below:

The apparatus used is shown in Figure 8. 1,3-bis(1phenylethenyl)benzene (I), (PEB) is puried by shortpath distillation from n-BuLi, and diluted with hexane, in a similar way to the purication of 1,1,diphenyl ethylene (Fig. 4). The DLI is prepared by reaction of sec-BuLi with PEB. Initially, 30 mLs of hexane is distilled. The apparatus is washed, rinsed, the solvent is distilled into ask (A) and the purge section is removed from the main

reactor by heat sealing constriction (C). PEB and secBuLi are added in ask (A) in a ratio [sec-BuLi]/[PEB] 2.1/1. The DLI produced is insoluble in this solvent and remains at the bottom of the ask. All liquids are transferred in ask (B) and the reactor is rinsed with hexane, by immersing ask (B) in a 40 C water bath. Then ask (B) is removed by heat sealing constriction (D), the main reactor is attached to the vacuum line

HIGHLIGHT

3223

Figure 8. Apparatus for the preparation of the (1,3-phenylene)bis(3-methyl-1-phenylpentylidene) dilithium initiator.

through the ground joint, the break-seal (E) is ruptured and benzene is distilled in ask (A). The solubility of DLI in benzene is limited and the concentration should not be higher than 1% (w/v). The reactor is removed from the vacuum line by heat-sealing constriction (F), DLI is dissolved in benzene and the initiators solution is poured in ampoules (G) and sealed off by heat sealing the constrictions.85

POLYMERIZATION PROCEDURES
Polymerizations in Nonpolar Solvents

Linear Homopolymers, Block Copolymers, Star Homopolymers, and Star-Block Copolymers

The basic reactor used for the synthesis of linear homopolymers, block copolymers, star homopolymers, and star-block copolymers is shown in Figure 9. The ampoules containing all necessary reagents, that is, the initiator and puried, monomers are attached to the main

reactor. The apparatus is attached to the vacuum line through ground joint (A), checked for pinholes, ame dried, and pumped for 1 h in order to remove the volatile species such as air and humidity. Then 23 mL of concentrated solution of n-BuLi in hexane is injected in ask (B) through the septum and the tube holding the septum is removed by heat sealing at (a). The apparatus is pumped for another half hour in order to remove hexane and air inserted in the apparatus during the injection, followed by distillation of the appropriate amount of solvent into ask (B). Usually benzene is used as solvent for polymerization of styrene and hexane or benzene for polymerization of isoprene or butadiene. For a typical polymerization the amount of solvent is chosen such that the nal polymer concentration is 510%, depending on the molecular weight of the polymer. The reactor is removed from the vacuum line by heat sealing at (C). The inner walls of the reactor are washed by the solution of n-BuLi, this solution is transferred back to ask (B) and placed in a 50 C water bath. Because of the vacuum inside the apparatus, benzene reuxes and condenses

3224

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000)

Figure 11. The main polymerization apparatus.

Figure 9. Apparatus for the preparation of linear homo- and copolymers, along with symmetric star homopolymers and star-block copolymers, consisting of the purge section and the main polymerization reactor.

over the whole reactor, rinsing traces of n-BuLi from the glass walls. As a consequence, all n-BuLi inserted along with the products of reaction impurities are collected in ask (B) of the purge section. The apparatus is tilted in a way shown in Figure 10. A water bath at 25 C is placed under the purge section and

an icewater bath under ask (D). Due to the vacuum in the apparatus, the solvent is distilled into the main reactor but the nonvolatile species, that is, n-BuLi and its reaction products with impurities, remain in ask (B). The purge section is removed by heat sealing of constriction (E) leaving a clean reactor lled with an appropriate amount of pure solvent. First, the break-seal of the monomer is ruptured, by moving the pair of magnets. The monomer is poured (or distilled in the case of isoprene or butadiene) in ask (D), followed by addition of the initiator and the ampoule is rinsed with solvent (Fig. 11). The polymerization is left to proceed until all monomer is consumed (usually for nine half life-times for a linear homopolymer). The living ends of the polymeric chains can initiate the polymerization of another monomer resulting in a

Figure 10. Intermediate step involving rinsing of the polymerization apparatus by distillation of the solvent from the purge section to the main apparatus.

HIGHLIGHT

3225

Figure 12. Reactor used for the preparation of the 3- and 4-miktoarm star copolymers of the ABn type, where n 2,3.

block copolymer or can react with a linking agent forming a star homopolymer. Furthermore, in the former case the living diblock copolymer can react with a linking agent to form a star-block copolymer. In the case of the synthesis of star polymers, the ratio of molar (COLi)/ (active sites of linking agent) is chosen to be 1.21.3 in order to accelerate the linking reaction. The excess of living chains can be removed by solvent/nonsolvent fractionation. Flask (F) (Fig. 11) can be used as reservoir of a solution of living arm, that will be used for the preparation of molecules with more complex architectures. Termination can be accomplished by addition of several drops of well-degassed methanol enclosed in a separate ampoule (not shown).

3- and 4-Miktoarm Star ( -Star) Copolymers of the ABn Type, where n 23

Schematically, the reactions used for the preparation of the 3- and 4- -star copolymers of the ABn Type, where n 23 are the following:

The basic reactor used for the synthesis of 3- and 4-miktoarm star copolymers is shown in Figure 12. The apparatus is attached to the vacuum line through ground joint (A), checked for pinholes, ame dried and pumped for 1 h. A small amount of concentrated solution of n-BuLi is injected through the septum, the side tube (C) is rinsed and removed by heat sealing. Twenty to thirty milliliters of benzene are distilled into ask (B) and the apparatus is removed from the vacuum line by heat sealing the constriction (D). Once the solvent is thawed the reactor is purged with the solution of n-BuLi. Then a water bath of 50 C is placed under ask (B) and benzene starts to reux rinsing the walls of the apparatus and is nally collected in ask (B). Therefore, traces of n-BuLi along with reaction products of impurities are collected in this ask. The purge section containing the solution of n-BuLi is removed by heat sealing constriction (E). An ampoule of living arm ALi (H) ( 5% solution in benzene) is attached to the main reactor as shown in Figure 13 (the preparation is discussed previously). The space between the two break-seals is evacuated by attaching the reactor to the vacuum line via ground joint (F), followed by removal of the apparatus by heat sealing constriction (G). Following distillation of the linking agent in the main ask, the ask is placed in an icewater bath and a hot towel is placed simultaneously on ampoule (H). At the same time the living polymer is added by rupturing the two break-seals. During the addition, the mixture of the inserted solution and linking agent is

3226

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000)

Figure 13. Intermediate step in the preparation of the 3- and 4-miktoarm star copolymers: adding the living arm ALi to the excess of the linking agent.

vigorously stirred. The hot towel increases the vapor pressure of ampoule (H), and addition of the living arm to the silane solution is ensured (the opposite, which would result in undesired coupling reactions, is impossible under these conditions). The ampoule is removed from the reactor by heat sealing of constriction (I). A ground joint is sealed in one of the two remaining breakseals of the apparatus through which it is attached to the

vacuum line (Fig. 14). The solvent and excess linking agent are removed by long pumping on the vacuum line and by redissolving the polymer in benzene and removing the solvent several times. Once the removal of excess linking agent is complete, 50 100 mLs of solvent are distilled into the main ask, followed by removal of the apparatus from the vacuum line by heat sealing at (J). After the preparation of an ampoule containing excess of

Figure 14. Intermediate step in the preparation of the 3- and 4-miktoarm star copolymers: removing the excess of the volatile linking agent by long pumping of the resulted ASiCln 1 on the vacuum line.

HIGHLIGHT

3227

the solution of living polymer is added to the solution of the macromolecular linking agent ASiCl23. After the completion of the linking reaction the star copolymer is separated from the parent material by fractionation.86,87

3-Miktoarm Star Terpolymer of the ABC Type

A schematic representation of the sequence of reactions used to prepare the ABC -star terpolymer is the following:

Figure 15. Final step of the synthesis of the 3- and 4-miktoarm star copolymers of the ABn type: addition of the living arm BLi to the macromolecular linking agent ASiCln 1.

living arm B the two solutions are connected by using an apparatus shown in Figure 15. The space between the two break-seals is pumped, and after heat-sealing at (K)

The apparatus used for the preparation of the 3-miktoarm star terpolymer is shown in Figure 16. Apart from some modications it is similar to the one used for the preparation of the -stars of the AB2/3 type discussed in the previous section. The addition of the rst arm A is made in exactly the same way described in the previous section. Then an ampoule of living arm B is sealed to the main reactor as shown in Figure 17, and the space between the asks is evacuated. The break-seals between the two asks are ruptured and the macromolecular linking agent ASiCl2 is titrated by the solution of living arm B, by lifting the ampoule in the way that is shown in the latter Figure. The linking reaction is monitored by taking small aliquots of solution in tubes, and measured by SEC

Figure 16. Reactor used for the preparation of the 3-miktoarm star terpolymer of the ABC type.

3228

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000)

Figure 17. Titration of living arm B with ASiCl2. The addition of the living chains is made by lifting ampoule (B).

(Fig. 18). At the end point ampoule (B) is removed from the reactor by heat sealing constriction (C). Finally, the last arm C is added to the macromolecular linking agent ABSiClMe by using a similar apparatus shown in Figure 15. The order of addition of the living chains for the preparation of 3-miktoarm star terpolymers has been discussed for the case where A is polystyrene (PS), B is polyisoprene, and C is polybutadiene (PB).88

scheme used for the preparation of the quaterpolymer is shown below:

4- -Star Copolymer of the A2B2 Type

The reactions used for the preparation of the 4-miktoarm star copolymer of the A2B2 type are the following: The synthesis of the miktoarm stars relies on (a) the inability of the carbanions of the rst (A) arm to undergo chain coupling with the chlorosilane, (b) the inability of the sterically hindered second or third arm to undergo complete reaction with trifunctional and difunctional macromolecular linking agent, (when titration is required) and (c) the ability of the less sterically hindered carbanion of the last arm to react completely with the macromolecular linking agent.

The reaction of the two rst arms A with SiCl4 are made in the same way arms A and B were incorporated in the synthesis of ABC. Finally, by reacting the macromolecular linking agent A2SiCl2 with an excess of BLi, the miktoarm copolymer A2B2 is formed. 4-Miktoarm star copolymers of the A2B2 has been synthesized, where A is PS and B is PB.89

Vergina Star Copolymer(PI8PS8)

The reactions used for the synthesis of the vergina star copolymer are shown below:90

4- -Star Quaterpolymer of the ABCD Type

Along the same lines the 4-miktoarm star quaterpolymer of the ABCD type has been synthesized89 in a similar way, where A is PS, B poly(4-methylstyrene) (P4MeS), C is PI and D is PB. The synthesis is similar to the one presented for the ABC star terpolymer. In this case the arms B and C have to be added by titration. The reaction

where SOCl16 is Si{[CH2CH2Si(CH3)][CH2CH2Si(CH3)Cl2]2}4. The main reactor used for the linking reaction was similar to the one shown in Figure 12. The

HIGHLIGHT

3229

Figure 18. SEC chromatograms of the synthesis of the 3- ABC terpolymer. (a) Titration of the macromolecular linking agent ASiCl2 with the living arm BLi; (b) Arm B; (c) Arm C; (d) Unfractionated; and (e) Fractionated 3- star terpolymer of the ABC type.

appropriate amount of PSLi in benzene was added to a benzene solution of SOCl16, in order to incorporate only eight arms of PS. The formation of the eight-arm PS star required 23 weeks, depending on the molecular weight of the arm. A small sample of the eight-arm star linking agent (PS)8OSOCl8 was removed and properly characterized. A 20% excess of living PILi chains were then added to the macromolecular linking agent (PS)8OSOCl8, and was left to react for 3 4 weeks, depending on the molecular weights of the PI and PS arms. The SEC chromatograms of the precursors along with the (PS)8OSOCl8, unfractionated and fractionated Vergina star copolymer are shown in Figure 19. Polymerizations in THF The polymerization of several monomers, like methacrylates or 2-vinyl pyridine is conducted in THF at low

temperatures. The principles of the polymerization procedure are the same as in the case of the polymerizations that take place in benzene. However a different apparatus is used as shown in Figure 20. A characteristic example for the polymerization of methyl methacrylate, (MMA) in THF at 78 C will be given. The apparatus is connected to the vacuum line, pumped down, checked for pinholes and ame-dried. A small amount of n-BuLi solution 1.6 M in hexane and DPE is added through septum (A) to the purge section ask resulting in the formation of a deep red DPHLi solution. After removal of the hexane, constriction (B) is heat-sealed. The appropriate amount of puried THF, sufcient to give a nal monomer concentration in the range of 5 to 10%, is distilled in the purge section ask through the vacuum line and the whole reactor is sealed off at constriction (C). Diphenylhexyllithium is preferred over n-BuLi since it is known that THF may react with

3230

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000)

Figure 19. SEC chromatograms of the synthesis of the Vergina star copolymers. (a) PS arm; (b) PI arm; (c) (PS)8OSOCl8; (d) Unfractionated product after the linking reaction of the (PS)8OSOCl8 with the living PI arms; (e) Fractionated Vergina star copolymer.

the n-BuLi giving several byproducts. The apparatus is thoroughly washed with the DPHLi solution. The purging solution is then transferred back to the purge section

ask and the apparatus is rinsed several times by distilling THF from the purge section to the walls of the reactor. Finally the solvent is distilled into the polymer-

Figure 20. Apparatus for the synthesis of polymers by using THF as a solvent at 78 C.

HIGHLIGHT

3231

ization ask (E), frozen using liquid nitrogen and the purge section is removed by heat sealing. Afterwards the apparatus is attached to the vacuum line through joint (F), the appropriate amount of puried MMA is distilled in the ampoule and the monomer section is sealed off at (G). The initiator used for the polymerization of MMA is the DPHLi, a bulky initiator to avoid the side reactions with the carbonyl group. The DPE from the ampoule (I) is added to the reaction ask in the solvent, which is then chilled to 78 C using a dry iceisopropanol bath. The sec-BuLi (H) is then added to the cold DPE solution, immediately forming the DPHLi initiator. At this point the break seal (J) is broken and the monomer is allowed to distill into the polymerization ask under rigorous stirring. The red color of the initiator instantaneously disappears, indicating the rapid initiation process. After completion of the monomer distillation the reaction mixture is allowed to react at 78 C, under continuous stirring for about 30 min (or more depending on the desired molecular weight) to ensure completion of the poly-

Figure 22. Examples of graft copolymers with different architectures.

merization. Finally the living polymer solution is deactivated by addition of degassed MeOH (K).

3- -Star Terpolymer of the (PS)(PI)(PMMA) type

A schematic representation of the sequence of reactions used to prepare the (PS)(PI)(PMMA) -star terpolymer is the following:91

Figure 21. Examples of single and double miktoarm stars.

Figure 23. Examples of multiarm miktoarm stars.

3232

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000)

These model materials have facilitated many studies on the relationship between the structure and the properties of the copolymers.

CONCLUSIONS
The high vacuum methodology is the most appropriate for the synthesis of well-dened polymers by anionic polymerization. The main drawbacks of this technique are the building of the complicated reactors by glass blowing and the longtime required for performing the reaction sequences. Nevertheless these limitations are a small prize to pay given the potential of the high vacuum technique for generating true model polymers with complex architectures.
The authors would like to acknowledge Dr. J. Roovers for his helpful discussions and suggestions.

The controlled substitution of two chlorine atoms of trichloromethylsilane in order to synthesize the macromolecular linking agent (PI)(PS)SiCl has been presented in the synthesis of the 3- terpolymer of the ABC type. The resulting monofunctional macromolecular chlorosilane solution was slowly added to a dilute solution containing a stoichiometric amount of , dilithium initiator (I). The procedure was followed with SEC in order to ensure complete reaction of the monofunctional macromolecular chlorosilane with only one active center of the initiator. The macromolecular initiator so created was used to polymerize MMA at 78 C in THF. The last step differentiates the strategy from the one used in synthesizing miktoarm stars having only diene and styrene arms, since PMMA does not give the desired linking product with chlorosilane. The (PS)(PI)(PMMA) was separated from the side products by fractionation. By using a combination of these techniques, a wide variety of complex macromolecular architectures has been synthesized, which are presented in Figures 21 23. Most of these miktoarm stars ( -stars) have welldened structures with high degree of molecular and compositional homogeneity. The time required to perform the linking reactions is usually between 12 months, depending on the number of chlorine atoms per silicon of the linking agent. During this time the synthesis of other polymers can be initiated and nally in two months a series of copolymers can be prepared.

REFERENCES AND NOTES

1. Rempp, P.; Franta, E.; Herz, J.-E. Adv Polym Sci 1988, 86, 145. 2. Morton, M. Anionic Polymerization: Principles and Practice; Academic Press: New York, 1983. 3. Webster, O.W. Science 1991, 251, 887. 4. Grubbs, R. H.; Tumas, W. Science 1989, 243, 907. 5. Kennedy, J. P.; Ivan, B. Designed Polymers by Carbocationic Macromolecular Engineering: Theory and Practice; Hanser Publishers: Munich, 1992. 6. Szwarc, M. Nature 1956, 178, 1168. 7. Szwarc, M.; Levy, M.; Milkovich, R. J Am Chem Soc 1956, 78, 2656. 8. Bywater, S. Encyclopedia of Polymer Science and Engineering, 2nd ed.; Kroschwitz, J. I., Ed.; Wiley Interscience: New York, 1985; Vol. 2, p 1. 9. Hsieh, H. L.; Quirk, R. P. Anionic Polymerization: Principles and Practical Applications; Marcel Dekker: New York, 1996. 10. Young, R. N.; Quirk, R. P.; Fetters, L. J. Adv Polym Sci 1984, 56, 1. 11. Pitsikalis, M.; Pispas, S.; Mays, J. W.; Hadjichristidis, N. Adv Polym Sci 1998, 135, 1. 12. Morton, M.; Fetters, L. J. Rubber Chem Technol 1975, 48, 359. 13. Worsfold, D. J.; Bywater, S. Can J Chem 1960, 38, 1891. 14. Riess, G.; Hurtrez, G. Encyclopedia of Polymer Science and Engineering, 2nd ed.; Kroschwitz, J. I., Ed.; Wiley: New York, 1985; Vol. 2, p 324. 15. Jerome, R.; Fayt, R.; Ouhadi, T. Prog Polym Sci 1984, 10, 87. 16. Mays, J. W.; Hadjichristidis, N. Polym Bull 1989, 22, 471.

HIGHLIGHT 17. Conlon, D. A.; Crivello, J. V.; Lee, J. L.; O Brien, M. J. Macromolecules 1989, 22, 509. 18. Konigsberg, I.; Jagur-Grodzinski, J. J Polym Sci Polym Chem Ed 1983, 21, 2535. 19. Okay, O.; Funke, W. Macromolecules 1990, 23, 2623. 20. Fetters, L.J.; Morton, M. Macromolecules 1969, 2, 453. 21. Elgert, K.-F.; Ritter, W. Makromol Chem 1976, 177, 2021. 22. Zhongde, X.; Mays, J. W.; Xuexin, N.; Hadjichristidis, N.; Schilling, F. C.; Bair, H. E.; Pearson, D. S.; Fetters, L. J. Macromolecules 1985, 18, 2560. 23. Blondin, D.; Regis, J.; Prudhomme, J. Macromolecules 1974, 7, 187. 24. Suzuki, T.; Tsuji, Y.; Tagekami, Y.; Harwood, H. J. Macromolecules 1979, 12, 234. 25. Zhong, X. F.; Francois, B. Makromol Chem 1990, 191, 2743. 26. Hatada, K.; Kitayama, T.; Ute, K. Prog Polym Sci 1988, 13, 189. 27. Yuki, H.; Hatada, K. Adv Polym Sci 1979, 31, 1. 28. Allen, R. D.; Long, T. E.; McGrath, J. E. Polym Bull 1986, 15, 127. 29. Varshney, S. K.; Jacobs, C.; Hautekeer, J. P.; Bayard, P.; Jerome, R.; Fayt, R.; Teyssie, P. Macromolecules 1991, 24, 4997. 30. Ozaki, H.; Hirao, A.; Nakahama, S. Macromolecules 1992, 25, 1391. 31. Luxton, A. R.; Quig, A.; Delxaux, M.-J.; Fetters, L. J. Polymer 1978, 19, 1320. 32. Muller, M.; Lenz, R. W. Makromol Chem 1989, 190, 1153. 33. Soum, A. H.; Tien, C.-F.; Hogen-Esch, T. A.; DAccorso, N. B.; Fontanille, M. Makromol Chem Rapid Commun 1983, 4, 243. 34. Fisch, D.; Khan, I. M.; Smid, J. Makromol Chem Macromol Symp 1990, 32, 241. 35. Inoue, S.; Aida, T. G. Encyclopedia of Polymer Science and Engineering, Suppl Vol.; Kroschwitz, J. I., Ed.; Wiley: New York, 1990; p 412. 36. Kricheldorf, H. R.; Berl, M.; Scharnagl, N. Macromolecules 1988, 21, 286. 37. Dubois, P.; Jerome, R.; Teyssie, P. Makromol Chem Macromol Symp 1991, 42/43, 103. 38. Lambert, C.; von Rague Schleyer, P. Angew Chem Int Ed Engl 1994, 33, 1129. 39. Schade, C.; von Rague Schleyer, P. Adv Organomet Chem 1987, 27, 169. 40. Rundle, R. E. J Phys Chem 1957, 61, 45. 41. Brown, T. L. Adv Organomet Chem 1965, 3, 365. 42. Brown, T. L. Pure Appl Chem 1970, 23, 447. 43. Dietrich, H. J. Organomet Chem 1981, 205, 291. 44. Brown, T. L.; Dickerhoff, D. W.; Baffus, D. A. J Am Chem Soc 1962, 84, 1371. 45. Lewis, H. L.; Brown, T. L. J Am Chem Soc 1970, 92, 4664. 46. Margerison, D.; Pont, J. D. Trans Faraday Soc 1971, 67, 353. 47. Fraenkel, G.; Beckenbaugh, W. E.; Yang, P. P. J Am Chem Soc 1976, 98, 6878. 48. Bywater, S.; Worsfold, D.J. J Organomet Chem 1967, 10, 1. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77.

3233

78. 79. 80. 81. 82.

Weiner, M.; Vogel, C.; West, R. Inorg Chem 1962, 1, 654. Hsieh, H. L. J Polym Sci A-3 1965, 153, 163. Selman, C. M.; Hsieh, H. L. Polym Lett 1971, 9, 219. Hsieh, H. L.; Glaze, W. H. Rubber Chem Technol 1970, 43, 22. Szwarc, M. Adv Polym Sci 1983, 49, 1. Richards, D. H.; Williams, R. L. J Polym Sci 1973, 11, 89. Lee, C. L.; Smid, J.; Szwarc, M. J Phys Chem 1962, 66, 904. Tung, L. H.; Lo, T. Y.-S. Macromolecules 1994, 27, 2219. Bredeweg, C. J.; Gatzke, A. L.; Lo, T. Y.-S.; Tung, L. H. Macromolecules 1994, 27, 2225. Lo, T. Y.-S.; Otterbacher, E. W.; Gatzke, A. L.; Tung, L. H. Macromolecules 1994, 27, 2233. Lo, T. Y.-S.; Otterbacher, E. W.; Pews, R. G.; Tung, L. H. Macromolecules 1994, 27, 2241. Worsfold, D. J.; Bywater, S. Can J Chem 1960, 38, 1891. Clark, T.; Schleyer, P. V.; Pople, J. A. J Chem Soc Chem Commun 1978, 137. Baird, N. C.; Barr, R. F.; Datta, R. K. J Organomet Chem 1973, 59, 65. Van Beylen, M.; Bywater, S.; Smets, G.; Szwarc, M.; Worsfold, D. J. Adv Polym Sci 1988, 86, 87. Graham, G.; Richtsmeier, S.; Dixon, D. A. J Am Chem Soc 1980, 102, 5759. Guyot, A.; Vialle, J. J Polym Sci Part B 1968, 6, 403. Roovers, J. E. L.; Bywater, S. Macromolecules 1968, 1, 328. Schue, F.; Bywater, S. Macromolecules 1969, 2, 458. Morton, M.; Pett, R. A.; Fetters, L. J. Macromolecules 1970, 3, 333. Cubbon, R. C. B.; Margerison, D. Proc Roy Soc 1962, 268A, 260. Johnson, A. F.; Worsfold, D. J. J Polym Sci Part A-3, 1965, 449. Morton, M.; Bostick, E. E.; Livigni, R. A.; Fetters, L. J. J Polym Sci Part A 1963, 1, 1735. Francois, B.; Sinn, V.; Parrod, J. J Polym Sci Part C 1963, 4, 375. Guyot, A.; Vialle, J. J Macromol Sci Chem 1970, A4, 107. Worsfold, D. J.; Bywater, S. Can J Chem 1964, 42, 2884. Alvarino, J. M.; Bello, A.; Guzman, G. M. Eur Polym J 1972, 8, 53. Fetters, L. J.; Balsara, N. P.; Huang, J. S.; Jeon, H. S.; Almdal, K.; Lin, M. Y. Macromolecules 1995, 28, 4996. Pitsikalis, M.; Sioula, S.; Pispas, S.; Hadjichristidis, N.; Cook, D. C.; Li, J.; Mays, J. W. J Polym Sci Part A: Polym Chem 1999, 37, 4337. Allen, R. D.; Long, T. E.; McGrath, J. E. Polym Bull 1986, 15, 127. Batra, U.; Russel, W. B.; Pitsikalis, M.; Sioula, S.; Mays, J. W.; Huang, J. S. Macromolecules 1997, 30, 6120. Zilliox, J. G.; Roovers, J. E. L.; Bywater, S. Macromolecules 1975, 8, 573. Ziegler, K.; Gellert, H. G. Annales 1950, 567, 179. Urwin, J. R.; Reed, P. J. J Organomet Chem 1972, 39, 1.

3234

J. POLYM. SCI. PART A: POLYM. CHEM.: VOL. 38 (2000) 87. Tselikas, Y.; Iatrou, H.; Hadjichristidis, N.; Liang, K. S.; Mohanty, K.; Lohse, D. J. J Chem Phys 1996, 105, 2456. 88. Iatrou, H.; Hadjichristidis, N. Macromolecules 1992, 25, 4649. 89. Iatrou, H.; Hadjichristidis, N. Macromolecules 1993, 26, 2479. 90. Avgeropoulos, A.; Poulos, Y.; Hadjichristidis, N. Macromolecules 1996, 29, 6076. 91. Sioula, S.; Tselikas, Y.; Hadjichristidis, N. Macromolecules 1997, 30, 1518.

83. Gilman, H.; Haubein, A. H. J Am Chem Soc 1944, 66, 1515. 84. Gilman, H.; Cartlidge, K. F. J Organomet Chem 1964, 2, 447. 85. Gido, S. P.; Chin, L.; Pochan, D. J.; Pispas, S.; Mays, J. M.; Hadjichristidis, N. Macromolecules 1996, 29, 7022. 86. (a) Mays, J. M. Polym Bull 1990, 23, 249; (b) Iatrou, H.; Siakali-Kioulafa, E.; Hadjichristidis, N.; Roovers, J.; Mays, J. W. J Polym Sci Part B: Polym Phys 1995, 33, 1925.