Room temperature living cationic polymerization of styrene with HX-styrenic monomer adduct/FeCl3 systems in the presence of tetrabutylammonium halide and tetraalkylphosphonium bromide salts

Article abstract of DOI:10.1016/j.polymer.2010.01.051

Living cationic polymerization of styrene was achieved with a series of initiating systems consisting of a HX-styrenic monomer adduct (X?=?Br, Cl) and ferric chloride (FeCl₃) in conjunction with added salts such as tetrabutylammonium halides (nBu₄N⁺Y^-; Y^-?=?Br^-, Cl^-, I^-) or tetraalkylphosphonium bromides [nR′₄PBr; R′?=?CH₃CH₂-, CH₃(CH₂)₂CH₂-, CH₃(CH₂)₆CH₂-] or tetraphenylphosphonium bromide [(C₆H₅)₄PBr] in dichloromethane (CH₂Cl₂) and in toluene. Comparison of the molecular weight distributions (MWDs) of the polystyrenes prepared at different temperatures (e.g., -25?°C, 0?°C and 25?°C) showed that the polymerization is better controlled at ambient temperature (25?°C). The polymerization was almost instantaneous (completed within 1?min) and quantitative (yield ～100%) in CH₂Cl₂. In CH₂Cl₂, polystyrenes with moderately narrow (M_w/M_n?～?1.33-1.40) and broad (M_w/M_n?～?1.5-2.4) MWDs were obtained respectively with and without nBu₄N⁺Y^-. However, in toluene, the MWDs of the polystyrenes obtained respectively with and without nBu₄N⁺Y^-/nR′₄P⁺Br^- were moderately narrow (M_w/M_n?=?1.33-1.5) and extremely narrow (M_w/M_n?=?1.05-1.17). Livingness of this polymerization in CH₂Cl₂ was confirmed via monomer-addition experiment as well as from the study of molecular weights of obtained polystyrenes prepared simply by varying monomer to initiator ratio. A possible mechanistic pathway for this polymerization was suggested based on the results of the ¹H NMR spectroscopic analysis of the model reactions as well as the end group analysis of the obtained polymer.

Full text of DOI:10.1016/j.polymer.2010.01.051

Polymer 51 (2010) 1258–1269
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Room temperature living cationic polymerization of styrene with HX-styrenic
monomer adduct/FeCl3 systems in the presence of tetrabutylammonium halide
and tetraalkylphosphonium bromide salts
*
Sanjib Banerjee, Tapas K. Paira, Atanu Kotal, Tarun K. Mandal
Polymer Science Unit, Indian Association for the Cultivation of Science, Jadavpur, Kolkata 700 032, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Living cationic polymerization of styrene was achieved with a series of initiating systems consisting of
a HX-styrenic monomer adduct (X ¼ Br, Cl) and ferric chloride (FeCl3) in conjunction with added salts
such as tetrabutylammonium halides (nBu4NþYꢀ; Yꢀ ¼ Brꢀ, Clꢀ, Iꢀ) or tetraalkylphosphonium bromides
Received 10 November 2009
Received in revised form
11 January 2010
Accepted 20 January 2010
Available online 2 February 2010
[nR04PBr; R0
¼
CH3CH2–, CH3(CH2)2CH2–, CH3(CH2)6CH2–] or tetraphenylphosphonium bromide
[(C6H5)4PBr] in dichloromethane (CH2Cl2) and in toluene. Comparison of the molecular weight distri-
butions (MWDs) of the polystyrenes prepared at different temperatures (e.g., ꢀ25 ꢁC, 0 ꢁC and 25 ꢁC)
showed that the polymerization is better controlled at ambient temperature (25 ꢁC). The polymerization
was almost instantaneous (completed within 1 min) and quantitative (yield w100%) in CH2Cl2. In CH2Cl2,
polystyrenes with moderately narrow (Mw/Mn w 1.33–1.40) and broad (Mw/Mn w 1.5–2.4) MWDs were
obtained respectively with and without nBu4NþYꢀ. However, in toluene, the MWDs of the polystyrenes
obtained respectively with and without nBu4NþYꢀ/nR04PþBrꢀ were moderately narrow (Mw/Mn ¼ 1.33–
1.5) and extremely narrow (Mw/Mn ¼ 1.05–1.17). Livingness of this polymerization in CH2Cl2 was
confirmed via monomer-addition experiment as well as from the study of molecular weights of obtained
polystyrenes prepared simply by varying monomer to initiator ratio. A possible mechanistic pathway for
this polymerization was suggested based on the results of the 1H NMR spectroscopic analysis of the
model reactions as well as the end group analysis of the obtained polymer.
Keywords:
Cationic polymerization
Ferric chloride
Styrene
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
side reactions [21] leading to the formation of polystyrenes with
broad molecular weight distributions (Mw/Mn). As a result, at least
Ever since their discovery in the 1950s living polymers have
played a central role in polymer science and technology because of
their wide range of applications [1–3]. In particular, the ionic
‘‘living’’ polymerization process unifies a number of attractive
features, such as predictable polymer length combined with high
monodispersity, end group functionality and stereoregularity.
Among ‘‘living’’ polymerization techniques, cationic polymeriza-
tion is limited to only certain vinyl monomers such as styrene [4–8],
substituted styrenes [9–13], vinyl ethers [14–16] and isobutene
[17–20]. Strongly electron-donating groups present on these vinyl
monomers stabilize the propagating carbocationic species and
enhance their reactivity to undergo cationic polymerization reac-
tion. But styrene, when polymerized with conventional cationic
initiators, forms an unstable growing carbocation that is prone to
undergo chain transfer, chain termination and other undesirable
theoretically, cationic polymerization of styrene with conventional
cationic initiators has been considered difficult [22]. Therefore,
specially designed initiating systems have been used to achieve
living cationic polymerization of styrene as reported by many
researchers [5,7,8,23–27]. For example, Higashimura and his
coworkers were the first to report cationic polymerization of
different vinyl monomers that occurs in a controlled manner [23].
They were also the first to report ‘‘living’’ cationic polymerization of
styrene using methanesulfonic acid/tin tetrachloride (CH3SO3H/
SnCl4) initiating system with tetrabutylammonium chloride
(nBu4NCl) in CH2Cl2 [24]. Subsequently, they have also reported
cationic polymerization of styrene using other initiating systems
[5,25]. But, none of the initiating systems contain ferric chloride
(FeCl3). Following these works, several other research groups have
reported cationic polymerization of styrene using TiCl4 at ꢀ80 ꢁC
[7,26] and at ꢀ70 ꢁC [8,27]. However, in all of these cases, the
molecular weight distributions (MWDs) of the obtained polymers
are quite broad [polydispersity index (PDI) ¼ Mw/Mn ¼ 1.8, [7] 1.37,
[26] 1.31 [8] and 1.46 [27]]. Matyjaszewski and his associates used
* Corresponding author. Fax: þ91 33 2473 2805.
E-mail address: psutkm@mahendra.iacs.res.in (T.K. Mandal).
0032-3861/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2010.01.051

S. Banerjee et al. / Polymer 51 (2010) 1258–1269
1259
boron trichloride (BCl3) to carry out cationic polymerization of
styrene at ꢀ75 ꢁC, but the obtained polystyrenes had broad PDIs
(ꢂ1.3) [28]. However, polystyrene with narrow MWD (PDI w 1.1)
has been synthesized through cationic polymerization of styrene
using TiCl3(OiPr) [titanium(IV) chloride substituted with an
electron-donating isopropoxy group] at ꢀ78 ꢁC [29].
To explore the nature of the propagating species, model reactions
were also designed by mimicking the reaction conditions of the
polymerization reactions, but in the absence of the monomer.
A possible mechanism of this ‘‘living’’ cationic polymerization is
discussed based on the results of the 1H NMR spectroscopic
analysis of these model reactions.
A detailed survey of the literature shows that there have been
very few reports regarding the use of FeCl3 as an activator or as one
of the components of the initiating systems in the cationic poly-
merization of vinyl monomers. These are cationic polymerization of
1,3-pentadiene with FeCl3 in n-hexane at 0 ꢁC and at room
temperature [30], cationic polymerization of isobutyl vinyl ether
using ‘‘FeCl3/added base’’ initiating system at 0 ꢁC in toluene [31],
cationic polymerization of styrene with FeCl3 in liquid SO2 at 0 ꢁC
[32] and electroinitiated cationic polymerization of styrene using
FeCl3 in acetone at 0 ꢁC [33]. According to these reports, cationic
polymerization of styrene with FeCl3 was poorly controlled.
Although, there have been some reports that described the cationic
polymerization of styrene and other vinyl monomers using ‘‘SnCl4/
added salt’’, [5,6,25] ‘‘TiCl4/added salt’’ [7,34] and ‘‘TiCl3(OiPr)/
added salt’’ [29], but, the system comprising of FeCl3 and an added
halide salt has never been used to initiate cationic polymerization
of styrenic monomers. Although, one recent report shows the use of
initiating system comprising of FeCl3 and nBu4NCl for cationic
polymerization of isobutyl vinyl ether in CH2Cl2 at 0 ꢁC. But, in this
case, the obtained polymers showed broad MWDs (PDI ꢂ 1.67) [31].
Again, to the best of our knowledge, there are no reports of cationic
polymerization of styrene at or near room temperature (25 ꢁC).
The objective of this work is to develop new FeCl3 based
initiating systems for truly ‘‘living’’ cationic polymerization of
styrene at room temperature (25 ꢁC). Thus, herein, we report the
systematic study on the cationic polymerization of styrene
employing a series of initiating systems consisting of a HX-styrenic
monomer adduct [e.g., C6H5(CH3)CHBr (I) or C6H5(CH3)CHCl (II) or
C6H5C(CH3)2Cl (III)] as the alkyl halide type initiators coupled with
a strong Lewis acid (FeCl3) as the activator and an externally added
2. Experimental
2.1. Materials
Styrene (St) (Aldrich, purity
ꢂ
99%) and a-Methylstyrene
(a-MeSt) (Aldrich, purity > 99%) were washed with 5 wt% aqueous
NaOH solution to remove the inhibitor, dried overnight with calcium
chloride(CaCl2) andthendistilled overcalciumhydride(CaH2)under
reduced pressure prior to use. (1-Bromoethyl)benzene (Aldrich,
purity 97%) was used as received. Anhydrous ferric chloride (FeCl3)
(Merck, India, purity 96%) was used as received. Tetrabutylammo-
nium bromide, [CH3(CH2)3]4NBr (Aldrich, purity ꢂ 99%), tetrabuty-
lammonium chloride, [CH3(CH2)3]4NCl (Aldrich, purity ꢂ 99%),
tetrabutylammonium iodide, [CH3(CH2)3]4NI (BDH laboratory
reagents, England, purity 98%), tetraethylphosphonium bromide,
(CH3CH2)4PBr (Aldrich, purity 97%), tetrabutylphosphonium
bromide, [CH3(CH2)3]4PBr (Aldrich, purity 98%), tetraoctylphospho-
nium bromide, [CH3(CH2)7]4PBr (Aldrich, purity 97%), tetraphenyl-
phosphonium bromide, (C6H5)4PBr (Aldrich, purity 97%) were used
without any purification, but after drying in vacuum at 35 ꢁC for 6 h
before use.
All the solvents were of reagent grade and were purchased from
Merck, India. First, dichloromethane (CH2Cl2) (Merck, India) was
washed with concentrated H2SO4 (thrice, 30 mL acid per 100 mL),
then successively with water (once), 5 wt% aqueous NaOH solution
(twice) and then again with distilled water (thrice), dried overnight
over anhydrous CaCl2 and distilled over phosphorous pentoxide
(P2O5) just before use in the polymerization process. Toluene was
purified by shaking twice with concentrated H2SO4 (100 mL of acid
per litre of toluene), once with water, once with aqueous 5% NaOH
solution and then with water. It was then refluxed over sodium/
benzophenone under nitrogen atmosphere and distilled prior to
use in the polymerization reaction.
salt [tetrabutylammonium halide, nBu4NY;
Y
¼
Br, Cl,
I or
tetraalkylphosphonium bromide, nR04PBr; R0
¼
CH3CH2–,
CH3(CH2)2CH2–, CH3(CH2)6CH2– or tetraphenylphosphonium
bromide, (C6H5)4PBr] at room temperature (25 ꢁC) in CH2Cl2 and
in toluene. The chemical structures of all the initiators and
polymerization reaction are schematically shown in Scheme. 1.
2.2. Synthesis of HX-styrenic monomer adducts
The HBr-styrene (HBr-St) adduct (I) was obtained from
commercial source (see Scheme 1 for chemical structure). The HCl-
styrene adduct (II) and HCl-
Scheme 1) were synthesized by bubbling dry HCl gas into a 1.0 M
CH2Cl2 solution of styrene and -methylstyrene respectively for 5 h
a-methylstyrene adduct (III) (see
a
at 0 ꢁC [8,26,35]. HCl gas was generated by dropping concentrated
sulphuric acid into powdery sodium chloride and was passed
through a column packed with anhydrous CaCl2 for drying. After 5 h
of reaction, the excess HCl gas was removed by bubbling dry
nitrogen gas into the reaction mixture. The clean and quantitative
formation of adducts II and III were confirmed via 1H NMR and 13
C
NMR spectroscopy (see Fig. S1–S4 in ESM). The adducts were iso-
lated by evaporating the solvents in a rotary evaporator. Finally, the
isolated adducts were vacuum distilled from CaH2 to obtain the
pure HCl adducts (II and III) and were stored under nitrogen in
a freezer that were used in the subsequent polymerization reaction.
2.2.1. HCl-styrene adduct (II)
1H NMR (300 MHz, CDCl3,
d
ppm): 1.86–1.88 (d, 3H, CH3), 5.08–
5.14 (q, 1H, CH), 7.29–7.45 (m, 5H, aromatic CH); 13C NMR
(300 MHz, CDCl3, ppm): 26.64 (CH3), 58.89 (C–Cl), 126.62, 128.36,
128.75 (aromatic CH), 142.95 (aromatic C).
d
Scheme 1. Living cationic polymerization of styrene using HX-styrenic monomer
adduct/FeCl3 in the presence of added salt.

1260
S. Banerjee et al. / Polymer 51 (2010) 1258–1269
2.2.2. HCl-
a
-methylstyrene adduct (III)
, ppm): 2.04 (s, 6H, CH3), 7.26–7.65
(m, 5H, aromatic CH); 13C NMR (300 MHz, CDCl3,
, ppm): 34.38
standard polystyrene samples with peak molecular weights (Mp) of
860, 1800, 3600, 8500, 19 100, 43 400, 50 000, and 100 000.
1H NMR (300 MHz, CDCl3,
d
d
(CH3), 69.68 (C–Cl), 125.53, 127.63, 128.32 (aromatic CH), 146.37
(aromatic C). These data matches with those reported for the same
compound published elsewhere [36].
3. Results and discussion
3.1. Cationic polymerization of styrene using FeCl3
Cationic polymerization of styrene was carried out using
a series of new initiating systems consisting of various combi-
nations of HX-styrenic monomer adducts [HBr-St (I):
2.3. Polymerization procedure
All polymerizations were performed under dry nitrogen in
a 25 mL two-necked round-bottomed flask equipped with a three-
way stopcock attached to a nitrogen gas filled balloon in one neck
and a silicone rubber septum in the other. The concentrations of
monomer (styrene) and the initiator adducts (I–III) were kept
constant at 1.0 M and 20 mM respectively. The parameters that
were varied are the concentrations of the Lewis acid activator
(FeCl3) and the added halide salts to examine the effect of their
concentrations on the molecular weights and molecular weight
distributions (MWDs) of the obtained polystyrenes. Polymerization
reactions were performed in two different solvents, dry CH2Cl2 and
dry toluene. The following is a typical polymerization procedure:
nBu4PBr (108.59 mg, 0.32 mmol) was taken in a double necked
reaction flask and was evacuated through a high vacuum pump for
6 h and was sealed with a silicone rubber septum after back filling
with nitrogen. Dry CH2Cl2 (7.0 mL), previously purged with ultra
high pure nitrogen gas, was then injected into the reactor by
a syringe. The reaction mixture was further purged with nitrogen
for 10 min to remove any dissolved gas present in CH2Cl2. The
reactor containing the mixture was placed into a water bath made
up of double walled water jacket glass vessel maintained at
a temperature of 25 ꢁC and the mixture was stirred magnetically.
C6H5CH(CH3)Br, HCl-St (II): C6H5CH(CH3)Cl, and HCl-a-MeSt (III):
C6H5C(CH3)2Cl] as the alkyl halide type initiators, a strong Lewis
acid (FeCl3) as the activator and externally added salts such as
tetrabutylammonium halides (nBu4NY; Y ¼ –Br, –Cl, –I) or
tetraalkylphosphonium bromides (nR04PBr: R0
CH3(CH2)2CH2–, CH3(CH2)6CH2–) or tetraphenylphosphonium
bromide, [(C6H5)4PBr] as shown in Scheme 1.
¼
CH3CH2–,
Polarity of the solvent always plays a key role in the living
cationic polymerization of any vinyl monomers. In cationic poly-
merization, with the increase of dielectric constant of the poly-
merization solvent, the rate of chain transfer to monomer and chain
transfer to solvent decreases compared to the rate of propagation
[37]. Generally, in a cationic polymerization reaction MWDs of the
obtained polymers became progressively narrower with decreasing
solvent polarity as reported by many researchers [38–40]. As usual
in cationic polymerization, the reactions were faster in a more polar
solvent. Thus, to examine the role of the solvent polarity, we per-
formed polymerization in two different solvents, CH2Cl2, a rela-
tively polar solvent and toluene, a non-polar solvent. The results of
the cationic polymerization of styrene in these two solvents are
discussed separately below.
HBr-styrene adduct (I) i.e., (1-Bromoethyl)benzene (28
mL,
3.1.1. Polymerization in CH2Cl2
0.205 mmol) and styrene (1.0 mL, 8.728 mmol) were then injected
separately into the reactor with a dry syringe through the septum.
Finally, the polymerization was initiated by rapidly injecting
a solution of anhydrous FeCl3 (129.9 mg, 0.8 mmol) in CH2Cl2
(1.0 mL). During the polymerization, the reaction mixture was
maintained stirring magnetically.
Polymerization reaction was terminated by addition of excess
prechilled methanol containing a small amount of ammonia (0.1%
v/v). The quenched reaction mixture was sequentially washed with
2 N hydrochloric acid, 10% aqueous NaOH solution and then with
triple distilled water to remove any FeCl3 present. The conversion of
styrene was measured by gravimetry after isolating the obtained
polystyrenes via solvent evaporation under reduced pressure and
overnight drying in vacuum oven at 60 ꢁC.
3.1.1.1. Effect of initiator. FeCl3 alone has been used very early as
a catalyst for the cationic polymerization of styrene in liquid SO2
medium at 0 ꢁC [32]. However, they have not explored anything
about the livingness of this polymerization. Later on, it has been
reported that FeCl3 alone is incapable of initiating polymerization
of styrene in acetone [33]. These reports prompted us to study the
cationic polymerization of styrene using FeCl3 in more detail. To
understand the system better, we investigated the effect of HX-
styrenic monomer adducts (I, II and III see Scheme 1 for their
chemical structure) as initiators and the added halide salt (AþYꢀ)
on the kinetics and molecular weight distributions (MWDs) of the
obtained polymers. To start with, first, we have conducted cationic
polymerization of styrene with FeCl3 in CH2Cl2 at 25 ꢁC in absence
of any initiator and added salts. The polymerization was instanta-
neous with a yield of 100% within 1 min. But, the MWD of the
obtained polystyrene was very high (PDI ¼ 2.62) (see entry 1 in
Table 1). Thus, to control the MWDs of the obtained polystyrenes,
we then introduced HX-styrenic monomer adducts (e.g., I or II or
III) in the polymerization system as initiator. The reason for
choosing such initiators is that they are structurally similar to the
propagating species of the polymer end group. An important
advantage of use of HCl-monomer adduct over other HX-monomer
adducts (X ¼ Br, I) is that the carbon-chlorine bond in the former
adduct is relatively more stable compared to that of C–X bond
present in later two adducts in inert atmosphere at room temper-
ature. This makes the preparation, characterization and storage
processes of well-defined multifunctional initiators much easier
prior to the polymerization [41]. The polymerization result with
these adducts are depicted in Table 1 and the corresponding GPC
traces are presented in Fig. S5 of the ESM. In these cases, we also
observed quantitative polymerization, w100% within 1 min (see
Table 1). GPC traces of all the obtained polystyrenes using I, II and
2.4. Characterization
2.4.1. NMR spectroscopy
lH NMR and 13C NMR spectra of all the synthesized initiators
(adducts II and III), products of the model reactions and the
obtained polystyrenes were recorded at 25 ꢁC on a Bruker DPX
300 MHz spectrometer using CDC13 as the solvent and TMS as the
internal reference.
2.4.2. GPC measurements
The number average molecular weight (Mn) and polydispersity
index (PDI) of the purified polystyrenes were measured by size-
exclusion chromatography using a Waters 1515 isocratic HPLC
pump connected to three Waters Styragel HR1, HR3 and HR4
columns and a Waters 2414 Refractive Index Detector at room
temperature (25 ꢁC). HPLC grade THF was used as the eluent with
a flow rate of 1 mL/min. The columns were calibrated against eight

S. Banerjee et al. / Polymer 51 (2010) 1258–1269
1261
Table 1
Polymerization data for styrene with FeCl3 using different HX-adducts in presence and in absence of added salt.
Entry
HX-adduct
Added salt
Dichloromethane
Toluene
Mn, GPC
PDI
Mn, GPC
PDI
1
2
3
4
5
6
7
–
I
–
–
–
–
10504
7696
3952
2600
13 208
2600
2.62
2.41
2.25
1.46
1.48
1.42
1.40
2496
1352
3536
1560
1220
1040
1248
1.47
1.33
1.35
1.31
1.21
1.19
1.17
II
III
I
II
III
nBu4NBr
nBu4NBr
nBu4NBr
8112
Polymerization condition: [M]0 ¼ 1.0 M; [I–III]0 ¼ 20 mM; [FeCl3]0 ¼ 100 mM; [added salt]0 ¼ 40 mM; Temperature ¼ 25 ꢁC; Time w1 min (dichloromethane) and 48 h
(toluene); Conversion w100% (dichloromethane) and 72% (toluene).
III, are unimodal in nature (see Fig. S5 in ESM) with a PDI of 2.41,
2.25 and 1.46 respectively (see entries 2–4 in Table 1). It is inter-
esting to note that these results clearly indicate that among all
the three initiators, the initiator III produced polystyrene with the
narrowest PDI (1.46, see entry 4 in Table 1). Such low PDI of the
obtained polystyrenes (using III) may be attributed to the genera-
tion of a stable tertiary carbocation compared to that obtained
using I and II, which generates secondary carbocation. But the exact
reason of this observation is still under investigation.
Again, it has been reported that in the cases of cationic poly-
merization of vinyl monomers using metal halide, the added salts
such as tetraalkylammonium halide can reduce the PDI of the
obtained polymer [5,6,25,31,34]. This study encouraged us to
explore the effect of addition of salts such as tetraalkylammonium
halide or tetraalkylphosphonium halide in this polymerization
medium as added salts. For example, when nBu4NBr was added in
the above-mentioned systems as mentioned in entries 2–4 in Table
1, we found that the PDIs of the obtained polystyrenes reduced
further (see entries 5–7 in Table 1). It is worth mentioning that the
effect of the added salt, nBu4NBr is prominent only in presence of
the HX-monomer adduct. Note that the PDI of the polystyrene
produced using FeCl3/added salt was close to that obtained using
FeCl3 alone (2.62, see entry 1 in Table 1). From these results, we may
conclude that the cationic polymerization of styrene with FeCl3 can
be better controlled using a combination of both HX-styrenic
monomer adduct and nBu4NBr than that using either initiator or
salt. Since, adduct, III produced polymer with lowest PDI, we
carried out rest of the polymerizations using III to examine the
effect of the nature of the added salt and temperature.
are shown in Fig. S6 in the ESM, which clearly shows unimodal
molecular weight distribution in all cases. In all the three cases, the
PDI of the obtained polystyrene were little bit lower compared to
that obtained in absence of any salt (compare entry 1 with entries
2–4 in Table 2). The reason for this low PDI of the obtained poly-
styrenes using quaternary halide salt is as follows: It has been well
accepted that the addition of quaternary salts reduce the PDIs of the
polystyrenes obtained from cationic polymerization of styrene
using tetrabutylammonium chloride (nBu4NCl) [5–7,24,29,34,42]
and also using different tetrabutylammonium halides (nBu4NþYꢀ;
Y– ¼ I–, Br–, Cl–) [25]. According to their proposition, the high
nucleophilicity of the halide anions appears to be the primary
factor that controls the polymerization towards a truly living
system [13,34,43]. Consequently, in our case, the low PDIs of the
obtained polymers are explained schematically using the above-
mentioned concepts (see Scheme 2). It is well known that FeCl3 is
a strong Lewis acid. Thus, it usually generates a binary counteranion
–ClFeClꢀ3 in non-aqueous medium by abstracting Cl from the C–Cl
bond of the HCl-a-MeSt adduct, III. Nucleophilicity of this anion is
so low that it cannot stabilize the growing carbocation. Thus,
a highly dissociated and nonliving species (see structure 1 in
Scheme 2) is generated in absence of any externally added salt.
However, in presence of an added salt, the counteranion of the salt
(Yꢀ) binds to FeCl3. As a result, the Lewis acidity of FeCl3 is
decreased due to the formation of a comparatively weakly acidic
species FeCl3Yꢀ. Thus, FeCl3Yꢀ induced a controlled dissociation of
the C–Cl bond of the HCl-a-MeSt adduct, III, which may not be
possible in absence of such species. Again, Yꢀ can also bind with the
propagating carbocationic species 1 and 2 and converts them to
a ‘‘Dormant’’ species 3 (see Scheme 2). In this way, the uncontrolled
propagation via the nonliving, dissociated species 1 is eliminated in
presence of salt and a living polymerization occurs selectively via
3.1.1.2. Effect of added halide salts (nBu4NY/nR04PBr). Polymeriza-
tion reactions were further carried out using III/FeCl3 in conjunc-
tion with three different tetrabutylammonium halide salts of
varying counteranion (nBu4NþYꢀ; Yꢀ ¼ Brꢀ, Clꢀ, Iꢀ) to examine
their effect on the MWDs of the obtained polymers. The obtained
results are summarized in Table 2 and the corresponding GPC traces
Table 2
Polymerization data for styrene using III/FeCl3 system with different added salts.
Entry
Added salt
Dichloromethane
Toluene
Mn, GPC
PDI
Mn, GPC
PDI
1
2
3
4
5
6
7
8
–
2600
8112
1560
2912
3328
3224
1872
4992
1.46
1.40
1.38
1.33
1.83
2.00
1.41
1.78
1560
1248
1040
1144
1210
1144
1352
1280
1.32
1.17
1.05
1.08
1.07
1.10
1.16
1.11
nBu4NBr
nBu4NCl
nBu4NI
nEt4PBr
nBu4PBr
nOct4PBr
Ph4PBr
Polymerization condition: [M]0 ¼ 1.0 M; [III]0 ¼ 20 mM; [FeCl3]0 ¼ 100 mM; [added
salt]0 ¼ 40 mM; Temperature ¼ 25 ꢁC; Time w1 min (dichloromethane) and 48 h
(toluene); Conversion w100% (dichloromethane) and 72% (toluene).
Scheme 2. Possibility of formation of different living species in the cationic poly-
merization of styrene using HX-styrenic monomer adduct/FeCl3 in the presence of
added salt.

1262
S. Banerjee et al. / Polymer 51 (2010) 1258–1269
Table 4
the non-dissociated dormant species 3 [13]. So, the polymers with
narrow MWDs, as obtained in the presence of added quaternary
halide salts, are due to the suppression of the uncontrolled prop-
agation, stabilization of the carbocationic centre and thus
suppression of the chain transfer reactions [7]. It has been reported
previously that addition of electron donors eliminates side reac-
tions such as chain transfer, intramolecular alkylation by the
growing cations and irreversible chain termination reaction [44].
It should be noted from the results depicted in entries 5–8 in
Table 2 that the presence of quaternary phosphonium bromide salts
instead of tetrabutylammonium halides does not improve the
livingness of this polymerization system when all other reaction
parameters are similar.
Polymerization data for styrene using III and nBu4NBr system at different FeCl3
concentrations.
Entry [FeCl3]0 (mM) Dichloromethane
Conv. (%) Mn, GPC PDI
Toluene
Conv. (%) Mn, GPC PDI
1
2
3
4
0
40
100
150
0
45
100
70
–
–
0
–
–
17 600 1.46 40
1040
1248
1310
1.22
1.17
1.31
8112
2720
1.40 72
1.44 65
Polymerization condition: [M]0 ¼ 1.0 M; [III]0 ¼ 20 mM; [added salt]0 ¼ 40 mM;
Time w1 min (dichloromethane) and 48 h (toluene).
compared to that obtained for polystyrenes prepared at 25 ꢁC
(compare entry 3 with entry 1 and 2 and entry 6 with entry 4 and 5
in Table 5). The reason for such unusual broad MWDs of the
obtained polymers is explained as follows: It has been mentioned
that the chain transfer processes characteristically limit the
molecular weight of the obtained polymers via cationic polymeri-
zation [40]. For styrenic monomers, alkylative attack on an
aromatic group is thought to be the dominant chain transfers
process [45,46]. In any polymerization system, the rate of chain
transfer is usually higher at high temperature that resulted in
broadening of the MWDs of the obtained polymers [47]. Therefore,
in our case, one should expect to obtain polymers with broad
MWDs at higher temperature (ca. 25 ꢁC). But, we obtained poly-
mers with broad MWDs at lower temperature (ca. ꢀ25 ꢁC) and low
MWDs at higher temperature (ca. 25 ꢁC). Again, it has been
reported for any cationic polymerization that the equilibrium
between the dormant and the active species usually controls the
MWDs of the obtained polymers. Faster equilibrium produced
polymers with narrow MWDs [31]. On this basis, we propose that
the broad MWDs are not due to the chain transfer reaction of the
propagating carbocationic species, but possibly due to the slower
interconversion between the dormant and the activated species at
low temperature (ca. ꢀ25 ꢁC) [6]. This is the reason why we
conducted all the polymerization of styrene at room temperature
(ca. 25 ꢁC).
3.1.1.3. Effect of added halide salt concentration. Polymerization of
styrene (1.0 M) with III/FeCl3 (20 mM/100 mM) system was carried
out at varying concentrations of added salt, nBu4NBr (ca. 0, 20, 40,
60 mM) to achieve the optimum reaction condition. The Mn and
MWD values of the obtained polystyrenes are given in Table 3 (the
corresponding GPC traces are shown in Fig. S7 in ESM), which
suggested that the polymerization with III/FeCl3 system produces
polystyrene with narrowest MWD when an added salt (nBu4NBr)
concentration of 40 mM was used (compare entry 3 with entries 1,
2 and 4 in Table 3). Hence, we conducted all the polymerizations
using this salt concentration.
3.1.1.4. Effect of Lewis acid activator (FeCl3) concentration. To check
the effect of activator concentration on the polymerization of
styrene in dichloromethane at 25 ꢁC, four different sets of poly-
merization reactions (see Table 4) were also carried out using
a fixed ratio of styrene/III/nBu4NBr ¼ 1000/20/40 mM, but varying
the FeCl3 concentrations (ca. 0, 40, 100 and 150 mM). Mn and MWD
values of the obtained polystyrenes are given in Table 4 (the cor-
responding GPC traces are shown in Fig. S8 in ESM), which sug-
gested that the polymerization with III/nBu4NBr system produces
polystyrene with narrowest MWD only at [FeCl3] ¼ 100 mM
(compare entries 3 with 2 and 4 in Table 4). This is the reason why
we have conducted all the polymerizations at this FeCl3 concen-
tration. Note that we did not observe any polymerization in absence
of the Lewis acid activator, FeCl3 (see entry 1 in Table 4).
3.1.1.6. Polymer molecular weight control. It should be noted, that
the polymerization of styrene in CH2Cl2 at 25 ꢁC was very fast and
the complete conversion was obtained within 1 min. Therefore,
sampling or kinetic measurement during the polymerization had
not been possible. Thus, to get further insight into the polymeri-
zation system, we carried out a series of polymerization reactions
with III/FeCl3/nBu4NI system simply by varying monomer to initi-
ator ratio ([M]o/[III]o) from 50 to 200 and keeping the concentra-
tions of FeCl3 and nBu4NI constant at 100 mM and 40 mM
respectively. The results of these polymerization systems are
summarized in Fig. 1, which shows the plot of [M]0/[III]0 vs. Mn of
the obtained polymers along with their corresponding GPC traces.
It clearly reveals that Mn values of the obtained polystyrenes are in
close agreement with that of the calculated values. Also, the Mn
values of the polystyrenes increase almost linearly with increase of
the feed ratio of monomer to initiator ([M]0/[III]0). The MWDs of
the obtained polystyrenes (1.32–1.38) remains almost constant
with increase of the feed ratio of monomer to initiator ([M]0/[III]0).
Thus, we were able to produce polystyrenes of controlled molecular
weight and MWDs by cationic polymerization of styrene using our
initiating systems.
3.1.1.5. Effect of temperature. According to the literature reports,
cationic polymerization of styrene is usually better controlled at
low temperature [4–8]. Thus, to check this low temperature effect
in our case, we carried out polymerization reactions at ꢀ25 ꢁC and
0 ꢁC keeping all the other reaction parameters similar to that used
at 25 ꢁC. The molecular weights and MWDs of the obtained poly-
styrenes are given in Table 5 and the corresponding GPC traces of
the polymers are presented in Fig. S9 in ESM. Again, the GPC traces
are unimodal in nature. However, to our surprise, the MWDs of the
polystyrenes obtained at ꢀ25 ꢁC and 0 ꢁC were relatively broader
Table 3
Polymerization data for styrene using FeCl3 at varying concentration of added salt,
nBu4NBr.
Entry
[nBu4NBr]0 (mM)
Dichloromethane
Toluene
Mn, GPC
PDI
Mn, GPC
PDI
1
2
3
4
0
20
40
60
2600
2460
8112
1.46
1.43
1.40
1.45
1560
1300
1248
1460
1.32
1.22
1.17
1.24
3.1.1.7. Monomer-addition
experiment. Two-stage
sequential
17 200
monomer addition experiment for chain extension was carried out
to establish the living nature of this styrene polymerization. The
GPC traces of the polystyrenes obtained by chain extension
Polymerization condition: [M]0 ¼ 1.0 M; [III]0 ¼ 20 mM; [FeCl3]0 ¼ 100 mM; Time
w1 min (dichloromethane) and 48 h (toluene); Conversion w100% (dichloro-
methane) and 72% (toluene).

S. Banerjee et al. / Polymer 51 (2010) 1258–1269
1263
Table 5
Polymerization data for styrene using III/FeCl3/added salt system at different polymerization temperatures.
Entry
Added salt
Temp. (ꢁC)
Dichloromethane
Toluene
Mn, GPC
PDI
Mn, GPC
PDI
1
2
3
4
5
6
nBu4NBr
nBu4NBr
nBu4NBr
nBu4NI
nBu4NI
nBu4NI
ꢀ25
0
25
ꢀ25
0
12 896
2490
8112
3341
2680
2860
1.47
1.45
1.40
1.69
1.44
1.33
1210
1040
1248
1280
1060
1176
1.24
1.20
1.17
1.18
1.13
1.08
25
Polymerization condition: [M]0 ¼ 1.0 M; [III]0 ¼ 20 mM; [FeCl3]0 ¼ 100 mM; [added salt]0 ¼ 40 mM; Time w1 min (dichloromethane) and 48 h (toluene); Conversion w100%
(dichloromethane) and 72% (toluene).
experiment are shown in Fig. 2. A fresh feed of monomer
([M]0 ¼ [M]add ¼ 1.0 M) was added just before the initial charge of
the styrene was completely polymerized to examine the livingness
of the styrene polymerization by III/FeCl3/nBu4NI initiating system
in CH2Cl2 at 25 ꢁC. In the first stage, a conversion of 96% was
reached after 1 min of polymerization and after the second stage
the overall conversion was 194%. The Mn and PDI values for the two
stages are Mn,(1st stage) ¼ 2080, PDI ¼ 1.33 and Mn (2nd
stage) ¼ 6240, PDI ¼ 1.41 respectively. Both the GPC traces (Fig. 2)
show unimodal distribution along with a clear lateral shift of
maxima towards higher molecular weight region on going from
stage-1 to stage-2. These results clearly indicate that the styrene
polymerization with III/FeCl3/nBu4NI initiating system is living in
a relatively polar solvent like CH2Cl2. However, the PDI of the
polystyrene slightly increases after the addition of the second
monomer feed, probably due to the slow initiation and slow
exchange between reversibly terminated and propagating
species [48].
controlled. As a result, the MWDs of the obtained polystyrenes
were broader (PDI ꢂ 1.33) compared to that usually obtained in any
well-controlled cationic polymerization system [5,6,24,25,29].
However, these reported polymerization systems are conducted
either at 0 ꢁC or even less up to ꢀ78 ꢁC whereas we performed all
the our polymerization reactions at room temperature (ca. 25 ꢁC).
3.1.2. Polymerization in toluene
In order to check whether the use of low polarity solvent can
have better control over this polymerization, we performed all the
above-mentioned polymerization reactions in a less polar solvent
(such as toluene) instead of CH2Cl2.
3.1.2.1. Effect of initiator. As mentioned above, in the case of poly-
merization in CH2Cl2, we were unable to study the kinetics because
of the fact that this polymerization was almost instantaneous (yield
w100% within 1 min) in presence or in absence of any added salt.
However, we try to study the kinetics of this cationic polymeriza-
tion using our newly developed initiating systems in toluene with
the expectation that the rate of polymerization will be much slower
compared to that in CH2Cl2. Fig. 3A shows the time-conversion
profiles of the cationic polymerization of styrene in toluene using
From the above-mentioned results, it can be summarized that
the cationic polymerization of styrene in CH2Cl2 with (I–III)/FeCl3
system was instantaneous both in presence and in absence of added
salt. Although the polymerization is living in nature, but is not well
Fig. 1. Plot showing the variation of Mn and PDI with varying initial molar ratio of monomer to III i.e. [M]0/[III]0 and the corresponding GPC traces of polystyrenes obtained with III/
FeCl3/nBu4NI in CH2Cl2 at 25 ꢁC: [M]0 ¼ 1.0–4.0 M; [III]0 ¼ 20 mM; [FeCl3]0 ¼ 100 mM; [nBu4NI]0 ¼ 40 mM. The line labeled calcd indicates the calculated Mn assuming the
formation of one living polymer per adduct (III) molecule.

1264
S. Banerjee et al. / Polymer 51 (2010) 1258–1269
Fig. 2. Mn and PDI of polystyrenes obtained with III/FeCl3/nBu4NI in a monomer-
addition experiment in CH2Cl2 at 25 ꢁC: [M]0 ¼ [M]add ¼ 1.0 M; [III]0 ¼ 20 mM;
[FeCl3]0 ¼ 100 mM; [nBu4NI]0 ¼ 40 mM. The molar amount of the second styrene feed
was the same as in the first.
(I–III)/FeCl3/nBu4NBr systems at 25 ꢁC. These results clearly indi-
cate that the rate of polymerization in toluene is indeed much
slower compared to that in CH2Cl2, although we used same initi-
ating systems. For example, the yield was 72% in 48 h of polymer-
ization in toluene, whereas that was 100% in less than 1 min in
CH2Cl2. The result also shows that the overall rate of polymerization
(Rp) depends on the nature of the initiators and it follows the order:
Rp, I > Rp, III > Rp, II. The higher reactivity of the C–Br bond compared
to that of C–Cl bond might be responsible for such order of prop-
agation rate as also reported elsewhere [49]. Fig. 3B shows the
semilogarithmic plot of ln{[M]0/[M]} vs time of polymerization
reactions carried out using I, II and III. Theoretically, for a truly
living polymerization system, this semilogarithmic plot should be
linear. However, in our case, these plots deviate from linearity for all
the three initiators. Usually, the curvature in semilogarithmic plot
indicates that the concentration of propagating and reversibly
terminated chains decreases more rapidly than the formation of
new chains by the slow initiation as in the case of cationic poly-
merization of styrene with 2-chloro-2,4,4-trimethylpentane/TiCl4/
dimethyl acetamide system [48].
Table 1 shows the Mn and MWDs of the polystyrenes thus
obtained with (I–III)/FeCl3 system both in presence and in absence
of added salts in toluene. The corresponding GPC traces are shown
in Fig. S10 of ESM. According to Table 1, the MWDs of the poly-
styrenes, obtained with III either in presence or in absence of salt,
are lower than that obtained using the other two initiators, I and II
(see entries 2–4 and 5–7 in Table 1). From these results, we may
consider that III is the best initiator among the three initiators used
in this polymerization of styrene with FeCl3 in toluene. In the case
of polymerization in CH2Cl2, we also observed that the III is best
among all three initiators as mentioned above.
Fig. 3. Time-conversion curves (A) and semilogarithmic plot of monomer concentra-
tion vs time (B) for the polymerizations of styrene ([M]0 ¼ 1.0 M) with I or II or III
(20 mM)/FeCl3 (100 mM) in presence of nBu4NBr (40 mM) at 25 ꢁC in toluene.
using various tetrabutylammonium halide salts (nBu4NþYꢀ;
Yꢀ ¼ Brꢀ, Clꢀ, Iꢀ) and quaternary phosphonium bromide salts
(nR04PBr; R0 ¼ CH3CH2–, CH3(CH2)2CH2–, CH3(CH2)6CH2–, C6H5–).
The corresponding GPC traces are shown in Fig. 4. All the GPC traces
are unimodal in nature. Note that GPC trace (see 2nd curve of
Fig. 4b) of one sample showed one additional peak in the lower
molecular weight region. This additional peak arises due to the
presence of air in the THF eluent, which are usually coming out
after elution of the polymer through the column. In fact, we
observed this peak with almost all samples, if we degassed the
eluent for long time then its intensity decreases. These peaks are
not due to polymers. MWD of the obtained polystyrene without
added salt is 1.32, whereas the MWDs of the polystyrenes obtained
using added salts such as nBu4NBr, nBu4NCl and nBu4NI are 1.17,
1.05 and 1.08 respectively (compare entry 1 with entries 2–4 in
Table 2). It should be noted that MWDs of the polystyrenes
obtained using III/FeCl3/nBu4NY in toluene are much lower than
that obtained using the same initiating systems in CH2Cl2 (compare
the results of polymerizations in dichloromethane and toluene in
Table 2). Also, in our case, the PDIs of the polystyrenes obtained
3.1.2.2. Effect of added halide salts (nBu4NY/nR04PBr). Table 2 shows
the Mn and MWDs values of the polystyrenes that were prepared

S. Banerjee et al. / Polymer 51 (2010) 1258–1269
1265
can only be obtained if the polymerization of styrene using III/
FeCl3/added salt systems behaves like a truly living system.
Although, we are unable to make such conclusion from the kinetic
study of the same systems in toluene.
We also performed the polymerization of styrene with III/FeCl3
in conjunction tetraalkylphosphonium bromide salts instead of
quaternary ammonium halides with the expectation that the
system would be better controlled. The PDIs of the polystyrenes
obtained using tetraalkylphosphonium bromide salts are also low
(PDI ¼ 1.07–1.16) as that obtained using tetraalkylammonium
halide salts (compare entries 2–4 with entries 5–8 in Table 2).
Notably, with increase in the chain length of the alkyl group of
quaternary phosphonium bromide salts from C2 to C4 and C8, MWDs
of the obtained polystyrenes increases (entries 5–7 in Table 2).
3.1.2.3. Effect of added halide salt concentration. To optimize the
reaction condition, polymerization of styrene (1.0 M) with III/FeCl3
(20 mM/100 mM) system was carried out at varying concentrations
of added salt, nBu4NBr (ca. 0, 20, 40, 60 mM). The Mn and MWD
values of the obtained polystyrenes are given in Table 3 (the cor-
responding GPC traces are shown in Fig. S11 in ESM), which sug-
gested that the polymerization with III/FeCl3 system produces
polystyrene with narrowest molecular weight distribution at
a nBu4NBr concentration of 40 mM (compare entry 3 with entries 1,
2 and 4 in Table 3). This is the reason why we have conducted all the
polymerizations at an added salt concentration of 40 mM.
Fig. 4. Mn and PDI of the polystyrenes with their GPC traces ([M]0 ¼ 1.0 M) obtained
with III (20 mM)/FeCl3 (100 mM) in presence of added salt (40 mM) at 25 ꢁC in toluene
at a conversion of >72%. (a) tetrabutylammonium halide salts; (b) tetraalkylphos-
phonium bromide salts.
3.1.2.4. Effect of Lewis acid activator (FeCl3) concentration. Four
different sets of polymerization reactions (see Table 4) were carried
out at styrene/III/nBu4NBr ¼ 1000/20/40 mM, but using anhydrous
FeCl3 of varying concentrations (ca. 0, 40, 100 and 150 mM) in
toluene at 25 ꢁC. Mn and MWD values of the obtained polystyrenes
are given in Table 4 (the corresponding GPC traces are shown in
Fig. S12 in ESM), which suggested that the polymerization with III/
nBu4NBr system produces polystyrene with narrowest molecular
using nBu4NCl and nBu4NI are far less compared to those reported
for polystyrenes obtained from cationic polymerization of styrene
using other initiating systems [4,5,24,25,27,48,50]. From these
results; one may conclude that the polystyrenes with such low PDIs
Fig. 5. 1H NMR spectrum of the mixture obtained from the model reaction mimicking the reaction condition of the polymerization reactions involving III and nBu4NþBrꢀ (A) and
involving III, anhydrous FeCl3 and nBu4NþBrꢀ (B). Polymerization condition: [III]0 ¼ 20 mM; [nBu4NþBrꢀ
]
¼ 40 mM; [FeCl3]0 ¼ 100 mM; Temperature ¼ 25 ꢁC; Solvent ¼ CH2Cl2
0
(the peak X is due to presence of CH2Cl2 in the polymer).

1266
S. Banerjee et al. / Polymer 51 (2010) 1258–1269
weight distribution only at FeCl3 concentration of 100 mM
(compare entries 3 with 2 and 4 in Table 4). This is the reason why
we have conducted all the polymerizations at a fixed FeCl3
concentration of 100 mM. Interestingly, we did not observe any
polymerization in absence of the Lewis acid activator, FeCl3 (see
entry 1 in Table 4).
3.1.2.5. Effect of polymerization temperature. We also conducted
polymerization reactions at ꢀ25 ꢁC and 0 ꢁC keeping all the other
reaction parameters similar to that used at 25 ꢁC to check its effect
on the polydispersities of the obtained polymers. Polystyrenes
obtained at 25 ꢁC showed narrower MWDs (for GPC traces see
Fig. S13 in ESM) compared to that obtained at ꢀ25 ꢁC and 0 ꢁC
(compare entry 3 with entries 1–2 and entry 6 with entries 4–5 in
Table 5). These results indicate that polymerization carried out at
lower temperature is poorly controlled than that carried out at
25 ꢁC, provided the other reaction parameters remain unchanged.
Similar results were also observed in the case of polymerizations in
CH2Cl2.
Scheme 3. Possible homohalogen exchange pathway of the model reaction using HX-
styrenic monomer adduct/FeCl3 in the presence of added salt.
mixture (see Fig. 5A). But, we did not observe any change of the
characteristic peaks of III in the mixture in comparison to that of
neat III. The result indicates that there is no structural change of III
due to this reaction, i.e. there is no halogen exchange between III
and nBu4NþBrꢀ. However, the 1H NMR spectrum of the mixture
after adding anhydrous FeCl3 shows a clear shift in the signals of the
3.1.3. Possible polymerization pathway
To explore a possible polymerization pathway for this poly-
merization of styrene using III/FeCl3 system in presence of added
salts, first of all, we carried out some model reactions. We then
analyzed these reactions via 1H NMR spectroscopy. The model
reactions are the actual polymerization reactions, but were carried
out in absence of monomer (here, styrene). The reaction recipes
were as follows: HX-adduct (I or III) 20 mM; FeCl3 100 mM; added
salt 40 mM.
methyl protons of III from
d 1.99 ppm to d 1.86 ppm (compare
Fig. 5A with Fig. 5B). Such shift is probably due to the generation of
a new dissociated species 5 in the mixture by the dynamic halogen
exchange between chlorine atom of FeCl3 with that of III (see
Scheme 3).
Similar experiments were also conducted with initiator I and
The first model reaction was the reaction of III with a solution of
nBu4NþI, but we did not observe any halogen exchange via 1H NMR
nBu4NþBrꢀ. We acquired the 1H NMR spectrum of this reaction
Fig. 6. 1H NMR spectrum of the mixture obtained from the model reaction mimicking the reaction condition of the polymerization reactions involving I and nBu4Nþ ꢀ
I (A) and
involving I, anhydrous FeCl3 and nBu4Nþ ꢀ
peak X is an unidentified peak).
I
(B). Polymerization condition: [I]0 ¼ 20 mM; [nBu4Nþ ꢀ
I
]
¼ 40 mM; [FeCl3]0 ¼ 100 mM; Temperature ¼ 25 ꢁC; Solvent ¼ CH2Cl2. (the
0

S. Banerjee et al. / Polymer 51 (2010) 1258–1269
1267
irrespective of the conversion and the types of the halogens X and Y
in the initiator and in the added halide salt. These spectra are not
shown here for brevity. Similar results were also observed by other
researchers for their system [42]. In general, living cationic poly-
merizations of styrene/vinyl monomers are terminated by adding
prechilled methanol containing a little amount of NH3 (0.1%).
Therefore, the polymers obtained via such kind of polymerization/
termination usually have four possible end groups (see Scheme 5).
These are: (i) a methoxide (–OCH3) end group formed by the
covalent attachment of –OCH3 to the propagating carbocationic end
[43], (ii) halogen atom as the end group [6,42,44,48], (iii) an olefinic
end group formed by the chain termination reaction through b-
hydrogen elimination reaction [52] and (iv) an indanyl ring as the
end group [4,12,40]. However, no other characteristic peaks were
observed for other possible end groups such as an indan or an olefin
or a methoxide (though the polymerization was quenched with
methanol).
Scheme 4. Possible heterohalogen exchange pathway of the model reaction using HX-
styrenic monomer adduct/FeCl3 in the presence of added salt.
spectroscopy (see Fig. 6A). When anhydrous FeCl3 was added to this
mixture, the 1H NMR spectrum of the mixture exhibits a shift of the
The absence of the terminal methoxide group indicated a strong
interaction between the terminal carbon and halogen. The living
nature of this polymerization of styrene by (I–III)/FeCl3/added salt
system was justified by the absence of the indan and the olefinic
terminals as the end group of the polystyrene chain. This polymer-
ization of styrene using FeCl3 is free from chain transfer and chain
termination reactions, because if that be the case, indanic or olefinic
end group would be present, instead of halide as the end group.
From the above-mentioned polymerization results, we observed
that the polymerization of styrene at room temperature (25 ꢁC) is
better controlled only in presence of all the three components of
the initiating system i.e. HX-adduct, FeCl3 and the added salt. Also,
end group analysis of the obtained polystyrene shows the presence
of halogen atom at the end. Based on these results as well as the
study of the model reactions, herein we propose the following
reaction sequence (see Scheme 6) as a possible pathway for this
polymerization of styrene initiated with the HX-adduct/FeCl3/
added salt system. At first, the metal halide (FeCl3) activates the C–
X bond of the initiator and generates a dissociated species, which
interacted with styrene through its carbocationic center and starts
the propagation. The role of the counteranion (Yꢀ) of the added salt
is to stabilize the propagating end of the carbocationic species and
thus to induce livingness to the polymerization system. The
cationic part of the added salt of AY i.e. Aþ has a role to play for this
substitution process. It actually forms an ion pair with [FeCl3X]ꢀ
and thus provides a driving force for the substitution of X by Y and
this substitution facilitates the propagation of the polymerization.
The above-mentioned result of the end group analysis clearly
showed the presence of counteranion (Yꢀ) at the end of the poly-
styrene chain.
characteristic methine and methyl signals (CH/CH3;
d, ppm) from
d
5.32/2.03 ppm to 4.42/2.40 ppm (Fig. 6B). In this case, this shift is
d
probably due to the rapid conversion of PhCH(Br) to PhCH(Cl) via
a dissociated species 6 by the dynamic halogen exchange between
the chlorine atom of FeCl3 with the bromine atom of adduct, I (see
Scheme 4).
Therefore, from these NMR results, we may conclude that there
is a rapid halogen exchange between the growing end and the FeCl3
(activator) no matter whether the reaction is initiated either with
the alkyl chloride (ca. III) or the alkyl bromide (ca. I). Also, as there
is no dissociated species (see 5 or 6 in Schemes 3,4) formation in
absence of FeCl3, one should not expect any polymerization upon
addition of styrene to this system. Indeed, we did not obtain any
polymer in absence of FeCl3 (entry 1 in Table 4) as mentioned
above. Similar halogen exchange reactions have also been studied
earlier to propose the possible mechanistic pathway for their
polymerization system [25,51].
To ascertain the possible mechanism, it is also necessary to
know the end group of the polystyrene, obtained from the poly-
merization of styrene with III/FeCl3/added salt system. Thus, 1H
NMR spectroscopic characterization was performed on some of the
representative polystyrene samples. 1H NMR spectrum (Fig. 7) of
one of such samples shows a broad signal around
d 3.48 ppm that
can be assigned to the methine proton of the halide terminal
[–CH2CH(Ph)–Y]. Peaks for the –CH3 protons of the polystyrene
were observed at d 1.03 ppm. All other characteristic of polystyrene
comes at their respective positions [42]. The spectra of other
samples also exhibit similar broad signal at around 3.48 ppm,
Fig. 7. 1H NMR spectrum of polystyrene obtained with III/FeCl3/nBu4NI in CH2Cl2 at 25 ꢁC after quenching the polymerization with ammonical methanol at a monomer conversion
of >90%. Polymerization condition: [M]0 ¼ 1.0 M; [III]0 ¼ 20 mM; [FeCl3]0 ¼ 100 mM; [nBu4NI]0 ¼ 40 mM.

1268
S. Banerjee et al. / Polymer 51 (2010) 1258–1269
Scheme 5. Chemical structures of four possible end groups attached to polystyrene chains.
Scheme 6. Possible polymerization pathway of living cationic polymerization of styrene using FeCl3 in presence of added salt.
4. Conclusions
polarity of the solvent used for the polymerization, and poly-
merization temperature were needed to obtain polystyrene with
narrow molecular weight distribution. The polymerization was
instantaneous (within 1 min) and quantitative (yield w100%) in
CH2Cl2. Living nature of this polymerization in CH2Cl2 was
confirmed by the successful monomer addition experiment as
well as from the study of molecular weights of obtained poly-
styrenes prepared simply by varying monomer to initiator ratio.
The polymerization in CH2Cl2 produces polystyrenes with poly-
dispersities ranging from 1.33 to 1.38. But, the polystyrene
obtained from toluene showed extremely narrow PDIs (1.05–
1.17). A possible polymerization pathway was suggested based on
the results of the 1H NMR spectroscopic analysis of the model
reactions as well as the end group analysis of the obtained
polymers.
In summary, we have systematically investigated the cationic
polymerization of styrene using a newly designed initiating
systems consisting of HX-styrenic monomer adduct (see I–III in
Scheme 1)/FeCl3 system in presence of added halide salts. A
particular advantage of this initiating system over those reported
previously is that it enables the synthesis of well-defined living
polystyrenes even at room temperature (ca. 25 ꢁC). The poly-
merization was poorly controlled in absence of the added halide
salt. That is the use of the added salt (tetrabutylammonium
halide or tetraalkylphosphonium bromide or tetraphenylphos-
phonium bromide) was mandatory to achieve controlled living
polymerization. The appropriate combination of the concentra-
tions of the components of the FeCl3 based initiating system,

S. Banerjee et al. / Polymer 51 (2010) 1258–1269
[20] Bochmann M, Dawson DM. Angew Chem Int Ed Engl 1996;35:2226–8.
1269
Acknowledgements
[21] Matyaszewski K. Comprehensive polymer science. Oxford: Pergamon; 1989.
[22] Pepper DC. J Polym Sci Polym Symp 1976;60:51.
S.B. and T.K.P thank CSIR, Govt. of India, for providing Fellow-
ships. This research was supported by the grants from CSIR, New
Delhi and DST, New Delhi, under the Nano Science and Technology
Initiative.
[23] Miyamoto M, Sawamoto M, Higashimura T. Macromolecules 1984;17:
2228–30.
[24] Ishihama Y, Sawamoto M, Higashimura T. Polym Bull 1990;23:361–6.
[25] Higashimura T, Ishihama Y, Sawamoto M. Macromolecules 1993;26:744–51.
[26] De P, Faust R. Macromolecules 2004;37:4422–33.
[27] Storey RF, Thomas QA. Macromolecules 2003;36:5065–71.
[28] Matyjaszewski K, Lin C-H. J Polym Sci Part A: Polym Chem 1991;29:1439–46.
[29] Hasebe T, Kamigaito M, Sawamoto M. Macromolecules 1996;29:6100–3.
[30] Santarella J-M, Rousset E, Randriamahefa S, Macedo A, Cheradame H. Eur
Polym J 2000;36:2715–27.
Appendix. Supplementary data
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.polymer.2010.01.051.
[31] Kanazawa A, Kanaoka S, Aoshima S. Macromolecules 2009;42:3965–72.
[32] Asami R, Tokura N. J Polym Sci 1960;42:553–60.
[33] Murty VVV, Nayak B. Eur Polym J 1981;17:151–5.
[34] Kamigaito M, Maeda Y, Sawamoto M, Higashimura T. Macromolecules
1993;26:1643–9.
References
[35] Kostjuk SV, Dubovik AY, Vasilenko IV, Frolov AN, Kaputsky FN. Eur Polym J
2007;43:968–79.
[36] Russell R, Moreau M, Charleux B, Vairon J-P, Matyjaszewski K. Macromolecules
1998;31:3775–82.
[37] Sakurada Y, Higashimura T, Okamura S. J Polym Sci 1958;33:496–500.
[38] Aoki S. Kobunshi Kagaku 1963;20:726–9.
[39] Higashimura T, Okamura S. Kobunshi Kagaku 1956;13:338–42.
[40] Wood KB, Stannett VT. Polym Bull 1991;25:199–204.
[41] Schappacher M, Deffieux A. Macromolecules 1991;24:2140–2.
[42] Miyashita K, Kamigaito M, Sawamoto M, Higashimura T. J Polym Sci Part A:
Polym Chem 1994;32:2531–42.
[1] Erdodi G, Ivan B. Chem Mater 2004;16:959–62.
[2] Puskas JE, Chen Y, Dahman Y, Padavan D. J Polym Sci Part A: Polym Chem
2004;42:3091–109.
[3] Kennedy JP. J Polym Sci Part A: Polym Chem 2005;43:2951–63.
[4] Faust R, Kennedy JP. Polym Bull 1988;19:21–8.
[5] Ishihama Y, Sawamoto M, Higashimura T. Polym Bull 1990;24:201–6.
[6] Miyashita K, Kamigaito M, Sawamoto M, Higashimura T. Macromolecules
1994;27:1093–8.
[7] Nagy A, Majoros I, Kennedy JP. J Polym Sci Part A: Polym Chem 1997;35:3341–7.
[8] Kostjuk SV, Kapytsky FN, Mardykin VP, Gaponik LV, Antipin LM. Polym Bull
2002;49:251–6.
[43] Sawamoto M, Enoki T, Higashimura T. Macromolecules 1987;20:1–6.
[44] Kaszas G, Puskas JE, Chen CC, Kennedy JP. Polym Bull 1988;20:413–9.
[45] Kennedy J, Marechal E. Carbocationic polymerization. New York: Wiley &
Sons; 1982.
[46] Imanishi Y. Transfer reactions in cationic polymerization. Kyoto, Japan: Dept.
Polym., Kyoto University; 1964.
[9] Higashimura T, Kojima K, Sawamoto M. Polym Bull 1988;19:7–11.
[10] Kojima K, Sawamoto M, Higashimura T. J Polym Sci Part A: Polym Chem
1990;28:3007–17.
[11] Faust R, Kennedy JP. Polym Bull 1988;19:35–41.
[12] Higashimura T, Kamigaito M, Kato M, Hasebe T, Sawamoto M. Macromolecules
1993;26:2670–3.
[47] Okamura S, Higashimura T. Chem High Polymers (Japan) 1956;13:262–7.
[48] Kostjuk SV, Dubovik AY, Vasilenko IV, Mardykin VP, Gaponik LV, Kaputsky FN,
et al. Polym Bull 2004;52:227–34.
[49] Schappacher M, Deffieux A. Macromolecules 1991;24:4221–3.
[50] Puskas JE, Luo W. Macromolecules 2003;36:6942–4.
[51] Kamigaito M, Sawamoto M, Higashimura T. Macromolecules 1992;25:
2587–91.
[13] Kojima K, Sawamoto M, Higashimura T. Macromolecules 1990;23:948–53.
[14] Miyamoto M, Sawamoto M, Higashimura T. Macromolecules 1984;17:265–8.
[15] Nuyken O, Kroner H. Makmmol Chem 1990;191:1–16.
[16] Yonezumi M, Takaku R, Kanaoka S, Aoshima S. J Polym Sci Part A: Polym Chem
2008;46:2202–11.
[17] Plesch PH, Austin JC. J Polym Sci Part A: Polym Chem 2008;46:4265–84.
[18] Xu X, Wu Y, Qiu Y, Dong C, Gy Wu, Wang H. Eur Polym J 2006;42:2791–800.
[19] Mitu S, Baird MC. Eur Polym J 2006;42:2039–44.
[52] Zsuga M, Faust R, Kennedy JP. Polym Bull 1989;21:273–80.