Removal of the copper catalyst from atom transfer radical polymerization mixtures by chemical reduction with zinc powder

Article abstract of DOI:10.1002/pola.24789

Simple mixing of an atom transfer radical polymerization (ATRP) mixture with zinc powder was demonstrated to result in rapid decolorizing of the solution and precipitation of elemental copper, using small amounts of silica gel as seeding material. The experiments revealed that the chemical reduction of copper by wetted zinc powder (i.e., 0.325 g/mmol copper) is fast and completed within less than 5 min. UV spectra of the filtered polymer solution showed no any trace of copper. Terminal bromoalkyl groups of the polymers in the ATRP solution were determined to be unchanged by short-term contact with zinc powder at room temperature and a nearly complete reductive dehalogenation takes place only after 24 h of interaction, as evidenced by reaction of elemental zinc with a model compound, ethyl bromoacetate. Indeed, poly(methyl methacrylate) (PMMA) sample (M_n: 7900, polydispersity index: 1.09) isolated from ATRP mixture after the copper removal a by short contact with zinc powder (i.e., 15 min) was determined "still living" as confirmed by chain extension with styrene, ethyl acrylate, and t-butyl acrylate monomers to give block copolymers. The presence of acetic acid was demonstrated to accelerate reductive dehalogenation of PMMA end-groups by zinc and yields nonliving polymer within 2 h.

Full text of DOI:10.1002/pola.24789

Removal of the Copper Catalyst from Atom Transfer Radical
Polymerization Mixtures by Chemical Reduction with Zinc Powder
FATMA CANTURK, BUNYAMIN KARAGOZ, NIYAZI BICAK
Department of Chemistry, Istanbul Technical University, Maslak, Istanbul 34469, Turkey
Received 23 December 2010; accepted 21 May 2011
DOI: 10.1002/pola.24789
Published online 8 June 2011 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: Simple mixing of an atom transfer radical polymer-
ization (ATRP) mixture with zinc powder was demonstrated to
result in rapid decolorizing of the solution and precipitation of
elemental copper, using small amounts of silica gel as seeding
material. The experiments revealed that the chemical reduction
of copper by wetted zinc powder (i.e., 0.325 g/mmol copper) is
fast and completed within less than 5 min. UV spectra of the
filtered polymer solution showed no any trace of copper.
Terminal bromoalkyl groups of the polymers in the ATRP solu-
tion were determined to be unchanged by short-term contact
with zinc powder at room temperature and a nearly complete
reductive dehalogenation takes place only after 24 h of interac-
tion, as evidenced by reaction of elemental zinc with a model
compound, ethyl bromoacetate. Indeed, poly(methyl methacry-
late) (PMMA) sample (Mn: 7900, polydispersity index: 1.09) iso-
lated from ATRP mixture after the copper removal a by short
contact with zinc powder (i.e., 15 min) was determined ‘‘still
living’’ as confirmed by chain extension with styrene, ethyl ac-
rylate, and t-butyl acrylate monomers to give block copoly-
mers. The presence of acetic acid was demonstrated to
accelerate reductive dehalogenation of PMMA end-groups by
C
V
zinc and yields nonliving polymer within
2
h.
2011
Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 49:
3536–3542, 2011
KEYWORDS: atom transfer radical polymerization (ATRP); block
copolymer; chemical reduction; copper catalyst; copper re-
moval; separation of polymers
INTRODUCTION Copper mediated atom transfer radical poly-
merization (ATRP) has been widely used for creating numerous
polymer architectures such as block, multiblock, star-block, and
graft copolymers with well-defined structures, from styrenic
and acrylic monomers.1 Comparing with other controlled/living
polymerization techniques such as reversible addition fragmen-
tation chain transfer (RAFT) and nitroxy mediated polymeriza-
tion (NMP) operating with chain transfer mechanism, ATRP is
much more efficient in grafting from solid surfaces and provides
better chain growth control.2 Despite its great versatility, ATRP
has not found extensive use in mass production of polymers.
This is largely due to greenish color of the polymer products
arising from the copper complex contaminants.
ligands on silica supports were determined to be effective
and able to reduce the copper concentrations down to 35–
40 ppm levels. However, molecular weight distributions of
the polymers were broad, owing to heterogeneity of the
process.5 Haddleton’s group reported a Schiff base ligand
covalently linked to silica support6 yielding copper-free poly-
mers with relatively high polydispersities [polydispersity
index (PDI): 1.4–1.6]. Shen et al. introduced amine ligands
covalently attached to silica gel support through oligoethyle-
neoxide spacer chains.7 Moderate chain length of the spacer
group was demonstrated to provide better control of the
molecular weights and low polydispersities as low as 1.2.
Shen’s group also reported an ATRP catalyst supported on mag-
netite nanoparticles (20–30 nm) easily recoverable by a mag-
netic bar from the reaction medium.8 Although narrow molecu-
lar weight distributions (PDI < 1.2) were attained, significant
activity loss of the recycled catalyst was the main drawback.
Honigfort and Brittain described copper complex of an amine
functionalized Janda-gel resin as recoverable catalyst.9 How-
ever, the polydispersities were relatively high (PDI: 1.40–1.85).
Removal of the copper from laboratory-scale samples can be
achieved by flash chromatography using alumina or silica gel
columns.3 However, this is not practical for industrial processes.
Considerable efforts aiming at preparing copper-free polymers
by ATRP have been proceeded in two directions:4 (i) design of
solid supported or special ligands forming easily isolable cop-
per complexes and (ii) removal of the copper complexes from
ordinary ATRP mixtures by chemical or physical means.
Synthesis of new special ligands forming isolable and soluble
complexes has been alternative pathway to attain homogeneity
and better control of the chain growth in ATRP. In the following
studies, Honigfort et al. reported amine ligands bearing stilbene
The use of solid supported copper catalyst has been consid-
ered as convenient way because of easy recovery of the cata-
lyst by filtration. Copper complexes of physically adsorbed
Correspondence to: N. Bicak (E-mail: bicak@itu.edu.tr)
C
V
Journal of Polymer Science Part A: Polymer Chemistry, Vol. 49, 3536–3542 (2011) 2011 Wiley Periodicals, Inc.
3536
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
units.10 Copper complexes of those ligands become soluble or
insoluble according to the cis–trans isomerization of the stil-
bene moieties upon irradiation with UV lamp for 2 h. An inter-
pentamethylated diethylene triamine (PMDETA, Merck), 2,20-
Bipyridine (J.T. Baker), and all the other chemicals were
used as purchased.
sting ligand, perluoro alkylated diethylenetriamine (AC8F17
)
Activation of Zinc Powder
has been exploited together with a fluorous solvent immiscple
with toluene at room temperature. Copper complex of the
ligandꢀin the mixed solvent provides nearly homogeneous ATRP
at 90 C. A phase separation occurs upon cooling, the copper
complex being retained in the fluorous phase.11
Commercial zinc powder (2 g) was treated with 10 mL NaOH
solution (30%) for 5 min to dissolve the oxide layer at the par-
ticle surfaces, and the aqueous phase was decanted. This pro-
cess was repeated twice, and the residue was washed several
times with distilled water (5 ꢂ 30 mL). The activated zinc
powder was obtained after successive washings with metha-
nol (2 ꢂ 10 mL), and diethyl ether (2 ꢂ 10 mL). The resulting
active powder was stored in a tightly closed bottle, without
further drying. Vapor of the residual ether provides protective
atmosphere against air oxygen. This is useful to avoid rapid
oxidation of the activated zinc powder, while handling.
Postpurification of ATRP mixtures by chemical or physical
sorption of the copper catalyst is another pathway to cop-
per-free polymers. Commercial ion exchange resin in acid or
Naþ forms has been demonstrated to be useful in the copper
removal, but the process is rather slow and greatly affects
from solvent polarity.12 Hydrated clay has also been used to
remove copper catalyst by physical sorption.13 Ding et al.
reported a reversibly immobilized catalyst system, in which
a thymine anchored silica reversibly binds the ligand via
hydrogen bonding with amido groups of the multiamine
functional ligand.14 Direct electro-deposition of the copper
catalyst from ATRP mixtures using tetrabutylammonium
hydroxide as supporting electrolyte has been used for
preparing colorless polymers.15
For safety reasons, however, storage of large amounts of acti-
vated powder is not advisable owing to its potential flamma-
bility, especially when contacting with paper.
Preparation of ATRP Stock Solution of PMMA
Methylmethacrylate (MMA) (40 mL, 0.4 mol) was polymer-
ized by ATRP method in toluene (60 mL), using ethyl bro-
moacetate (1.34 g, 8 mmol) as initiator, CuBr (1.148 g, 8
mmol) and one of the ligands selected from PMDETA, 2,20-
Bipyridine, and HTETA. The initial composition of the poly-
merization mixtures was chosen as [MMA]/[R-Br]/[CuBr]/
[Ligand]: 25/1/1/1 molar ratio, and the polymerizations
were conducted for 3 h at 70 C. The resulting viscous solu-
tion was cooled and stored as stock solution for the copper
removal experiments.
Despite those significant contributions, there is yet no practi-
cal solution for the copper removal from ATRP mixtures.
Herein, we describe a simple and quick method of the
copper removal from ATRP mixtures, based on chemical
reduction of the copper by zinc powder. The standard redox
potential of Cu/Zn couple is,
ꢀ
Cu0 ꢁ Zn0 ¼ 0:337 V ꢁ ðꢁ0:763 VÞ ¼ 1:1 Volt
The Copper Removal Experiments
A sample of ATRP solution (10.0 mL) containing 0.8 mmol
copper complex was thoroughly mixed with 0.26 g of acti-
vated zinc powder, 0.2 g of silica gel, and 0.1 g of water,
without dilution. The mixture was stirred until disappear-
ance of the color of ATRP solution and filtered. The colorless
polymer in filtrate was isolated by precipitation in metha-
nol–acetic acid mixture (95/5, v/v).
The spontaneous electron transfer from zinc metal to Cu (II)
in aqueous medium furnishes well-known ‘‘Daniel Battery,’’
which has been widely studied in electrochemistry.
Commercial zinc powder contains oxide film on its surface.
Activation of the zinc powder by removal of the oxide layer
is especially crucial in organometallic reactions.
Test for the Reductive Dehalogenation of Ethyl
Bromoacetate by Zinc Powder
In this work, commercial zinc powder was activated by con-
centrated NaOH solution before use in the copper removal.
The chemical reduction of the copper complex was followed
by disappearance of the color of ATRP mixtures containing
some common ligands and effects of contact time, and the
ingredients were studied.
To inspect reductive dehalogenating effect of wetted zinc
powder on polymers in the ATRP mixtures, a model com-
pound, ethyl bromoacetate (EBA) was reacted with toluene
in the absence of the copper complex. For this purpose,
activated zinc powder (3 g), water (0.9 g, 0.05 mol), toluene
(20 mL), and ethyl bromoacetate (5.6 g, 0.04 mol) were
mixed in a canonical flask and stirred at room temperature
for 24 h. Progress of the reductive dehalogenation was sim-
ply monitored by 1H-NMR spectra of the aliquots taken at
various time intervals (15, 30, 60, 100 min, and 24 h) from
the mixtures. Debromination of EBA was assayed by decreas-
ing integral of the singlet of ACH2Br group at 3.4 ppm.
EXPERIMENTAL
Materials
Methyl methacrylate (Aldrich), styrene (E. Merck), ethyl
bromoacetate (Fluka), and toluene (E. Merck) were redis-
tilled before use. Cuprous bromide was freshly prepared by
the method given in the literature.16 The ligand, hexacis-
1,1,4,7,10,10-hexyl-1,4,7,10-tetraaza decane, forming entirely
soluble copper complexes was prepared by action of
1-bromo hexane on triethylene tetramine (TETA) as reported
before.17 Zinc powder (J.T. Baker), silica gel (E. Merck),
Chain Extension Test by ATRP for the Presence of
Terminal Bromines in PMMA
To investigate dehalogenating effect of zinc powder on the
polymers, an ATRP solution of PMMA was prepared by
COPPER CATALYST REMOVAL FROM ATRP MIXTURES, CANTURK, KARAGOZ, AND BICAK
3537

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
similar procedure described above, using initial composition
with [MMA]/[Ethyl-2-bromoisobutyrate]/[CuBr]/[PMDETA]:
50/1/1/1 molar ratio. The reaction was conducted in tolu-
ene (1/1) for 1 h. Forty milliliters of this solution was freed
from copper by immediate zinc powder treatment as
described above. Gel permeation chromatogram (GPC) traces
of the isolated polymer by precipitation in 150 mL metha-
nol–acetic acid mixture (9/1) revealed a molecular weight of
Mn: 7.900 (PDI: 1.09).
SCHEME 1 Chemical reduction of the copper complex by zero-
valent zinc.
Characterization and Analysis of the Polymer Solutions
1H-NMR spectra of the polymers were recorded by a Bruker
250 MHz NMR spectrometer, using CDCl3 as solvent.
GPCs were taken in tetrahydrofuran (THF) solutions with a
flow rate of 0.3 mL/min using Agillant 1100 series instru-
ment consisting of a pump, a refractive index-detector, and
Waters Styrogel (HR4, HR3, and HR2) columns. Polystyrene
was used as standard.
To inspect the presence of bromine end groups in the poly-
mer, it was subjected to chain extension with styrene, ethyl
acrylate, and t-butyl acrylate by ATRP method. In these reac-
tions, the molar composition, [monomer]/[R-Br]/[CuBr]/
[PMDETA]: 300/1/1/1, was kept constant. A typical proce-
dure is as follows: PMMA sample (0.5 g) was dissolved in a
mixture of 4.4 mL toluene–styrene (1/1) under nitrogen
atmosphere. PMDETA (11 mg, 6.3 ꢂ 10ꢁ5 mol) and CuBr
(9 mg, 6.3 ꢂ 10ꢁ5 mol)
UV–visible spectra were obtained by a Chebios Optimum-One
UV–visible spectrophotometer.
Atomic absorption spectrometer (AAS) was used for analyses
of zinc and copper ions in the polymer solutions, using a
High-resolution continuum source atomic absorption spec-
trometer (HR-CS AAS; ContrAA 700) comprising a flame and
a graphite furnace in tandem.
The reaction was conducted at 90 ꢀC for 24 h, and the
product was isolated as described above. This procedure
was repeated for the chain extensions with ethyl acrylate
and t-butyl acrylateꢀ monomers, except the polymerization
temperature was 80 C. The polymers isolated similarly after
copper removal by treating with zinc powder. The GPC traces
of the resulting polymers obtained by chain extension with
styrene, ethyl acrylate, and t-butyl acrylate revealed number
average molecular weights of Mn: 35.900 (PDI: 1.07), 47.800
(PDI: 1.15), and 21.400 (PDI: 1.03), respectively.
RESULTS AND DISCUSSION
Chemical Reduction of the Copper Complex
Addition of small amounts of wetted zinc powder (0.325 g/
mmol copper) to ATRP solution and stirring for 5 min
resulted in disappearance of the greenish color of the solu-
tion, indicating removal of the copper catalyst by chemical
reduction.
The remaining portion of the ATRP mixture (ꢃ10 mL) was
stirred with 0.52 g (10 mmol), 0.1 g H2O, and 0.2 g silica gel
for 24 h. The resulting polymer was isolated in conventional
manner. A sample of this product (0.5 g) was subjected to
chain extension with styrene using exactly the same proce-
dure given above, to inspect the presence of reactivable bro-
mine terminal groups. However, no mass increase was
observed in this case. Its GPC trace did not show significant
increase in the molecular weight (Mn: 8100) in comparison
with that of the prepolymer (Mn: 7900).
Without trace water, however, the time for the disappearance
of the color was about 1 h. Most probably, the role of water
in this process is to facilitate the electron transfer in the
organic medium. Stoichiometric equation of the chemical
reduction of the copper (Scheme 1) suggests 1 mol of ele-
mental zinc per 1 mol of copper complex in ATRP mixture.
However, the use of equimolar zinc powder did not result in
colorless solution, and no copper precipitate was observed
in the ATRP mixture of PMMA. Considering with the use of
3- to 5-fold excess of activated zinc in organic reactions, that
is, Reformatsky reaction,16 only small portion of zinc (20–
30%) is estimated to involve in the reactions. The rest part
remains unreacted.
Accelerated Dehalogenation of PMMA Chain-Ends
by Zinc and Acetic Acid
To obtain dehalogenated polymer, the copper removal experi-
ment described above was repeated in the same conditions,
except acetic acid was used instead of water. For this pur-
pose, 0.4 g (ꢃ6 mmol) zinc powder, 0.2 g silica gel, and
1 mL acetic acid were added to 5 mL of colored ATRP solu-
tion of PMMA and stirred for 2 h at room temperature. The
green color the ATRP mixture disappeared in 2–3 min and
gray zinc particles turned to white within 10 min, because of
oxidation of elemental zinc. After stirring for additional 1 h,
the polymer sample (Mn: 7.900, PDI: 1.09) in the mixture
was subjected to chain extension with ATRP of styrene (8h)
and isolated by precipitation in methanol–acetic acid mixture
(95/5, v/v) and dried overnight at 40 ꢀC under vacuum).
1H-NMR spectrum of the polymer did not show the presence
of the polystyrene block, indicating a complete dehalogena-
tion of the prepolymer, PMMA by zinc and acetic acid.
Taking this into account, nearly 4-fold excess of activated
zinc (0.325 g/mmol of the copper complex) was used for the
copper removal in this study. This procedure resulted in dis-
appearance of the green color of the ATRP solution in less
than 5 min. It is noteworthy that, when zinc powder-ATRP
mixture was left to stand for 1 h without agitation, surfaces
of the zinc powder turned to reddish-brown indicating
accumulation of elemental copper, and the mixture was still
colored. Further stirring did not give a colorless solution.
This is evidence for the fact that, surfaces of the zinc powder
are covered by zero-valent copper and not able to reduce
the cupric ions any more. To avoid this problem, small
amount of silica gel (0.2 g) was introduced together with
zinc powder to create seeding centers for the precipitation
3538
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
The polymer, PMMA isolated by precipitation in methanol
with 5% acetic acid was colorless, and AAS of its toluene so-
lution, however, did not show any detectible zinc or copper.
Most probably, the zinc impurities in the previous solution
have been retained in the methanol solution as acetate salt.
The Side Reaction
Simplicity of the wetted-zinc powder process is very attrac-
tive to obtain copper-free polymers in ATRP. However, a pos-
sible side reaction, dehalogenation of halo alkyl end-groups
of ATRP growth-polymers can be considered as important
limitation of this procedure. The reaction of activated zinc
with alkyl halides is well documented in organic chemistry.
In the presence of proton donors such as acids and water,18
the reaction results in replacement of halogen with hydrogen
atom, and the process is termed as ‘‘reductive dehalogena-
tion.’’ The reaction of zinc with alkyl halides in neutral or
basic conditions results in Wurtz-type of coupling to alkyl
dimers alongside with the reductive dehalogenation.19 Such
a side reaction of zinc metal with the haloalkyl end-groups
of the polymers prepared by ATRP may destroy their
bromoalkyl end groups as well.
FIGURE 1 UV–visible spectrum of diluted (1/250) ATRP mixture
of PMMA in toluene, before (a) and after (b) copper removal
with zinc powder.
of elemental copper. This was helpful to reduce the copper
complex in short times by moderate stirring.
UV–visible spectrum of the colorless filtrate (Fig. 1) does not
show absorption bands of the original ATRP solution, located
at 330 and 780 nm.
Extent of possible reductive dehalogenation with zinc pow-
der while removing the copper catalyst was investigated
using ethyl bromoacetate (EBA) as model compound. This
was carried out by action of 3 g of zinc powder on ethyl
bromoacetate solution (5.6 g, 0.04 mol) in toluene (20 mL)
in the absence of copper complex.
A weak band appeared at 315 nm might be attributed to
residual copper complex. However, this band having a
maximum at 315 nm was also observed in UV spectrum of
toluene solution of ZnCl2–ligand mixture (10ꢁ4 M of each).
Therefore, this band must be due to zinc complex formed.
AAS of the filtrate revealed 4.1 ppm of zinc, but no detectible
copper. Considering the detection limit of the AAS apparatus,
final copper concentration of the solution must be lower
than 0.25 ppm.
By adding trace amount water (0.9 g, 0.05 mol), the mixture
1
was stirred for 24 h at room temperature. H-NMR spectrum
of the filtered solution [Fig. 2(b)] did not show typical sin-
glet of ACH2ABr group at 3.9 ppm.
FIGURE 2 1H-NMR spectrum of ethyl bromoacetate–toluene mixtures before (a) and after (b) treatment with zinc powder.
COPPER CATALYST REMOVAL FROM ATRP MIXTURES, CANTURK, KARAGOZ, AND BICAK
3539

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 2 Reductive dehalogenation of the model compound, EBA with zinc powder.
Instead, a new singlet emerged at 2.2 ppm revealed a quanti-
tative reductive dehalogenation of the bromomethyl group
(Scheme 2). Except those of toluene solvent, the three peaks
in the 1H-NMR spectrum of the resulting mixture [Fig. 3(b)]
well match with those of simple ethyl acetate.
To examine whether the bromo end groups of PMMA in the
mixture were affected within short periods, 40 mL of ATRP
solution was freed from copper by zinc powder treatment
(for 15 min), and the polymer isolated from the colorless
solution by precipitation in methanol–acetic acid mixture
(9/1) was subjected to chain extension with styrene, ethyl
acrylate, and t-butyl acrylate monomers by ATRP method
(see ‘‘Experimental’’ section).
Rapid disappearance of the greenish color of the ATRP solu-
tion implies a fast copper reduction in comparison with the
reductive dehalogenation. To compare the reaction rates, the
model reaction was repeated also for 15, 30, 60, and 100
min. Time depended dehalogenation yields of EBA were sim-
ply estimated based on 1H-NMR spectra, using the following
relationship.
GPC traces of the resulting products (Fig. 3) showed signifi-
cant mass increases in the molecular weights. Thus, number
average molecular weights and polydispersity indexes of the
chain extension products were Mn: 35.900 (PDI: 1.07),
47.800 (PDI: 1.15), and 21.400 (PDI: 1.03) for styrene, ethyl
acrylate, and t-butyl acrylate, respectively. These results
imply formation of the block copolymers, PMMA-b-PS,
PMMA-b-PEA and PMMA-b-P(t-BA), and the presence of bro-
moalkyl end groups of the prepolymer used in the chain
extension as well. The narrow molecular weight distributions
and unimodal curves in their GPC traces can be ascribed to
the absence of dead end in PMMA used in the chain exten-
sions performed by ATRP.
À
Á
I10=I20 ꢁ ðI1=I2Þ
À
Á
Percentage debromination yield ¼
ꢂ 100
I10=I20
where I1 and I2 represent integrals of the proton signals of
ACH2ABr and AOCH2A groups, respectively. The superscript
zero denotes initial values of the corresponding peak inte-
grals. The results showed that debromination of EBA within
30 min is almost zero and stirring for longer than 1 h causes
to significant rise in the debromination as shown in Table 1.
1H-NMR spectrum of the product obtained by chain exten-
sion with styrene (Fig. 4) exhibits aromatic proton signals of
the polystyrene block in 7.2–7.4 ppm range in addition to
the typical signals of PMMA block. The strong signal at
1.6 ppm is associated with CH2 protons of this block. A weak
singlet arising from OCH3 of PMMA block appears at
3.58 ppm. Weakness of PMMA proton signals is due to low
percentage of this block.
This result revealed possible loss of the bromine atom at the
chain ends of ATRP-growth polymers and disposal of termi-
nal bromoalkyl groups by long-term interaction with zinc
powder.
Apparently, reduction of copper in ATRP solution by zinc
powder is reasonably fast in comparison with the reductive
debromination of the polymer end-groups.
1H-NMR spectrum of the block copolymer with PEA shows
OCH2 signals at 4.1 ppm. New peaks appearing at 2.27 and
1.62 ppm are associated with CHACO and CH2 protons of
1
PEA, respectively. The H-NMR spectrum of the block copoly-
mer with t-BA exhibits characteristic methyl protons of
t-butyl group of P(t-BA) component at 1.42 ppm as a sharp
singlet.
TABLE 1 Reductive Debromination of EBA with Zinc Powder at
Room Temperature
[Zn]/[EBA]
Reaction Time
Debrominationa (%)
1.37
1.37
1.37
1.37
1.37
15 min
30 min
60 min
100 min
24 h
0
0
FIGURE 3 GPC traces of PMMA isolated from ATRP solution af-
ter removal of the copper by zinc powder treatment for 15 min
and the polymers obtained by chain extension of this polymer
with styrene, ethyl acrylate, and t-butyl acrylate using ATRP
method.
<1
ꢃ3.4
ꢃ100
Assigned by 1H-NMR.
a
3540
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
and 3.3 min, for the cases of PMDETA and bipyridyl ligand,
respectively (Table 2). Obviously the difference is due to
greater hydrophilicity of those ligands comparing with that
of HTETA. Although THF and acetone are not common sol-
vents in ATRP, we have studied effects of these solvents on
the copper removal with zinc powder.
Colors of the solutions diluted with equal volumes of THF
and acetone disappeared within 10.2 and 13.1 min, respec-
tively. Most probably these water-miscible solvents cause to
partial precipitation of the polymer on the wet particle surfa-
ces, thereby the copper reduction by zinc particles becomes
slow (Table 2). Slightly shorter time (8.8 min) to fade the
color of ATRP solution diluted with MMA must be due to its
immiscibility with water.
FIGURE 4 1H-NMR spectrum of the block copolymers, PMMA-
b-PS, PMMA-b-PEA, and PMMA-b-P(t-BA) obtained by chain
extension of PMMA isolated from ATRP solution, freed from
copper by zinc powder treatment for 15 min. [Color figure can
be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
Dehalogenation of the Polymer by Zinc Powder and
Acetic Acid
Replacement of terminal halogen atoms of the polymers pro-
duced by ATRP might be of interest in some applications, as
reported by Coessens and Matyjaszewski.20
Taking this into account, the last part of the study was
addressed to replacement of the haloalkyl end groups of
PMMA in ATRP solution, while removing the copper. To
accelerate the reductive dehalogenation, acetic acid was cho-
sen as proton donor, because of its miscibility with organic
solvents. Addition of 1 mL acetic acid to the zinc powder–
ATRP mixture gave white precipitate of zinc acetate within
10–15 min. The stirring was continued for 2 h at room tem-
perature. To prove the absence of the bromoalkyl groups at
the chain ends, isolated PMMA (Mn: 7.900, PDI: 1.09) was
subjected to chain extension by ARTP of styrene at 90 ꢀC
similarly (see ‘‘Experimental’’ section). 1H-NMR spectrum of
the resulting product did not show aromatic proton signals
of poly(styrene), indicating a complete reduction of the
terminal bromoalkyl groups of PMMA by zinc powder
treatment in the presence of acetic acid. It was found that
preactivation of zinc powder is not necessary for the dehalo-
genation, because the oxide layer on the particle surfaces is
readily removed by action acetic acid.
All these results are clear evidences for the formation of
block copolymers and retaining of bromoalkyl groups of the
precursor polymer, PMMA isolated after short-term zinc
treatment. By contrast, GPC trace of the polymer isolated
from the ATRP mixture treated with zinc powder for 24 h
did not show a significant increase in molecular mass
(Mn ¼ 8.100).
1H-NMR spectrum of this product did not show any aromatic
signal, indicating the absence of polystyrene segments and
debromination of the prepolymer PMMA by zinc powder in
24 h. In another words, stirring with zinc powder for 24 h
results in almost quantitative removal of the bromine at the
chain ends, and the resulting polymer is no longer living.
To investigate ligand effect on the copper reduction, ATRP
solutions with PMDETA and bipyridyl ligands were treated
with zinc powder similarly. The reaction was even faster
than before, so that colors of the solutions fade within 1.5
TABLE 2 Ligand and Solvent Effects on the Chemical Reduction of Copper by Zinc Powder in ATRP Solution
Contact Time for
Decolorization
[Zn]/[Cu]
Entrya
(mol/mol)
Ligand
Solvent
with Zn Powder (min)
1
2
3
4
5
6
a
5
5
5
5
5
5
HTETA
PMDETA
Bipyridyl
HTETA
HTETA
HTETA
Toluene
3.5
1.5
Toluene
Toluene
3.3
Toluene þ THFb
Toluene þ acetoneb
Toluene þ MMAb
10.2
13.1
8.8
b
Conditions: A mixture of 7.5 mL ATRP solution with 0.133 M copper
complex, 0.3 g activated zinc powder, 0.2 g silica gel, 0.1 g water was
stirred until disappearance of the color.
The solutions were diluted by equal volume of the solvents.
COPPER CATALYST REMOVAL FROM ATRP MIXTURES, CANTURK, KARAGOZ, AND BICAK
3541

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
CONCLUSIONS
9 Honigfort, M. E.; Brittain, W. J. Macromolecules 2003, 36,
3111–3114.
The wetted zinc powder process presented offers an easy
access to copper-free ATRP solutions within minutes, while
retaining the bromoalkyl end-groups of the polymers. Also,
the process can be used for dehalogenation of the polymers,
10 Honigfort, M. E.; Brittain, W. J.; Bosanac, T.; Wilcox C. S.
Macromolecules 2002, 35, 4849–4851.
11 Haddleton, D. M.; Jackson, S. G.; Bon, S. A. F. J Am Chem
using small amounts of acetic acid. Considering with 10ꢁ3
M
Soc 2000, 122, 1542–1543.
average copper concentration of typical ATRP solutions, zinc
powder to be used for the chemical reduction of copper in 1
L solution is around 3 g, together with 2 g of silica gel. Re-
covery of the copper in elemental form may be considered
as beneficial for large-scale processing of ATRP mixtures.
12 Matyjaszewski, K.; Pintauer, T.; Gaynor, S. Macromolecules
2000, 33, 1476–1478.
13 Munirasu, S.; Deshpande, A.; Baskaran, D. Macromol Rapid
Commun 2008, 29, 1538–1543.
14 Ding, S.; Yang, J.; Radosz, M.; Shen, Y. J Polym Sci A:
Polym Chem 2004, 42, 22–30.
REFERENCES AND NOTES
15 Nasser-Eddine, M.; Delaite, C.; Dumas, P.; Vataj, R.; Louati,
1 Matyjaszewski, K.; Xia, J. Chem Rev 2001, 101, 2921–2990.
A. Macromol Mater Eng 2004, 289, 204–207.
2 Tsarevsky, N. V.; Matyjaszewski, K. Chem Rev 2007, 107,
16 Practical organic chemistry, 5th ed. In Vogel’s Textbook of
Practical Organic Chemistry; Vogel, A. I.; Tatchell, A. R.; Fur-
nis, B. S.; Hannaford, A. J.; Smith, P. W. G., Eds.; Prentice
Hall: London, 1989.
2270–2299.
3 Matyjaszewski, K.; Pintauer, T.; Gaynor, S. Macromolecules
2000, 33, 1476–1478.
4 Shen, Y.; Tang, H.; Ding, S. Prog Polym Sci 2004, 29,
17 Acar, M.; Bicak, N. J Polym Sci A: Polym Chem 2003, 41,
1053–1078.
1677–1680.
5 Nguyen, J. V.; Jones, C. W. J Catal 2005, 232, 276–294.
18 Zhang, Y. P.; Liao, S. J.; Xu, Y. Tetrahedron Lett 1994, 35,
6 Haddleton, D. M.; Kukulj, D.; Radigue, A. P. Chem Commun
1999, 99–100.
4599–4602.
19 Von Gyldenfeldt, F.; Marton, D.; Tagliavini, G. Organometal-
7 Shen, Y.; Zhu, S.; Pelton, R. Macromolecules 2001, 34,
lics 1994, 13, 906–913.
5812–5818.
20 Coessens, V.; Matyjaszewski, K. Macromol Rapid Commun
8 Ding, S.; Xing, Y.; Radosz, M.; Shen, Y. Macromolecules
1999, 20, 66–70.
2006, 39, 6399–6405.
3542
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA