Cationic Spherical Polyelectrolyte Brushes as Nanoreactors for the Generation of Gold Particles

Cationic Spherical Polyelectrolyte Brushes as

Nanoreactors for the Generation of Gold Particles

Geeta Sharma, Matthias Ballauff *a

Polymer-Institut, Universitat Karlsruhe, Kaiserstrasse 12, 76128 Karlsruhe, Germany

Received: September 10, 2003; Revised: December 2, 2003; Accepted: December 2, 2003; DOI: 10.1002/marc.200300107

Keywords: ion exchangers; nanoparticles; nanoreactor; nanotechnology; polyelectrolytes

Introduction

Nanometric gold particles attached to colloidal spheres

have been a field of intense research during the recent

years.[1–11] The motivation for a large number of studies

derives from possible applications related, for example, to

the unique optical properties of gold particles with dia-

meters in the range of a few nanometers.[12–16] Moreover,

latex or silica particles covered by a dense layer of gold

particles may be used for the fabrication of photonic

crystals.[17–19] Most of the systems prepared so far have

been generated by the adsorption of colloidal gold nano-

particles onto latexes or silica spheres having a suitably

modified surface. Covering the surface of latex particles by

a cationic polyelectrolyte leads to a net positive surface

charge that attracts the much smaller gold particles bearing

a negative charge. The recent literature on this subject is

enormous and the making of such a composite particles

seems well established by now.

Much less is known of the alternative route, namely,

creating the metallic gold directly at the surface of colloid

particles by reduction of a suitable gold salt. In this way

the surface of the latex or silica sphere becomes a ‘‘nano-

reactor’’ in which the reduction is done. This way of

generating Au-nanoparticles is comparable to the use of

micelles[20,21] or dendrimers[22–26] as ‘‘nanoreactors’’.

Evidently, the gold compound must be firmly immobilized

in this nanoreactor to prevent the exchange of the gold

ions into the continuous phase. Hence, it is important to

elucidate the interactions of suitable gold compounds with

model colloids having a well-defined surface structure.

Summary: If long polyelectrolyte chains are attacheddensely to colloidal latex particles, a spherical polyelec-trolyte brush results. These spherical polyelectrolytes aredispersed in water and carry a high charge. We demonstratethat these systems can be used to immobilize ions of heavymetals, such as gold, as counter-ions. Reduction of these ionsleads to metallic nanoparticles. In this way the brush layerattached to the surface of the particles becomes a ‘‘nanore-actor’’ that may be used for chemical conversions of themetal ions. We show that the reduction of AuCl4

� ionswithin these nanoreactors leads to well-defined and rathermonodisperse gold nanoparticles that are attached to the sur-face of the core. A stable dispersion of polymeric core parti-cles with attached nanoparticles results. All results reportedhere suggest that chemical reactions of ions immobilized inspherical polyelectrolyte brushes provide a new route tocomposite particles of inorganic and organic materials.

Transmission electron micrograph of gold particles on a core-shell system.

Macromol. Rapid Commun. 2004, 25, 547–552 DOI: 10.1002/marc.200300107 � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication 547

a Present address: Lehrstuhl Physikalische Chemie I, UniversitaetBayreuth, 95440 Bayreuth, Germany.

To the authors’ best knowledge, only a few studies are

available in which the surface of colloidal systems have

been used for the generation of Au particles. Schmidt and

co-workers used bottlebrush polymers to bind AuCl4�

ions.[27] Subsequent reduction lead to Au nanowires.

Armes et al. used micelles of triblock copolymers to bind

the AuCl4� ions from aqueous solution.[28] Here, too, the

formation of gold nanoparticles was observed. A similar

approach to bind AuCl4� ions by polyallylamine grafted

on the surface of poly(methyl methacrylate) spheres was

used by Youk to produce gold particles of average size

around 12 nm.[29]

Here we show that latex particles bearing long cationic

polyelectrolyte chains may be used as well-defined nano-

reactors. Figure 1 shows our method in a schematic fashion:

In step I, long chains of poly(aminoethyl methacrylate

hydrochloride) (PAEMH) are grown on the surface of a

polystyrene core by photo-emulsion polymerization.[30]

By this ‘‘grafting-from technique’’ a cationic spherical

polyelectrolyte brush is generated similar to a system under

recent consideration.[31] The grafting of the PAEMH chains

achieved by photo-emulsion polymerization is so dense

that the brush limit is reached in these systems, that is, the

average distance of the chains on the surface is much smaller

than their average dimensions.[32] The chloride counter-

ions are subsequently exchanged by AuCl4� ions (step III,

Figure 1)). As an alternative route, the hydrochloride can

be converted into the free base first and then reacted with

HAuCl4 (step II; Figure 1). The gold compound thus immo-

bilized can be reduced in the final step IV to yield metallic

gold particles on the surface.

The use of the cationic spherical polyelectrolyte brushes

as nanoreactors has a number of clear advantages. The

surface of the nearly monodisperse core particles is well

defined[30] and the extension of the surface layer consisting

of polyelectrolyte chains can be studied precisely by dyna-

mic light scattering (DLS), as shown recently for anionic

spherical polyelectrolyte brushes.[33,34] Moreover, it can

be demonstrated that approximately 95% of the counter-

ions are confined in the brush layer.[35] This confinement

was predicted some time ago by Pincus.[32] It ‘‘closes’’ the

nanoreactor and prevents the formation of Au particles in

the aqueous phase. In this way, the well-established physics

of the polyelectrolyte brush layer can be used for chemical

reactions on the surface of colloidal particles.

Experimental Part

Materials and Methods

All chemicals used for the synthesis are of analytical gradeand were used as received. Styrene (BASF) was distilledunder reduced pressure and stored at �4 8C until used. Hexa-decyl trimethyl ammonium bromide (CTAB) was used as

Figure 1. Schematic representation of the formation of gold particles on the surface of thecore-shell system. The core-shell system having a shell of poly(aminoethyl methacrylatehydrochloride) [I] is deprotonated reversibly at high pH to give amine shell system [II]. I andII can be counter-ion exchanged/complexed with HAuCl4 to give III. Reduction of III withNaBH4 forms IV with nanosized gold particles.

548 G. Sharma, M. Ballauff

Macromol. Rapid Commun. 2004, 25, 547–552 www.mrc-journal.de � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

surfactant and 2,20-azodiisobutyramidine dihydrochloride(V-50; Fluka) was used as initiator in the synthesis of poly-styrene particles. Ethanolamine hydrochloride, absoluteethanol, methacryloyl chloride, and 1,2-dichloroethane wereused as received.

Tetrachlorogold(III) acid trihydrate was used to intro-duce gold ions into the brushes and NaBH4 was used to reducethe ions. Distilled and deionized water was used in the entiresynthesis.

The compound 2-[p-(2-hydroxy-2-methylpropiophenone)]-ethyleneglycol methacrylate (HMEM) was used as the photo-initiator whose synthesis has been described previously. Themonomer aminoethyl methacrylate hydrochloride (AEMH)was synthesized according to reference [36]. Characterizationwas performed by NMR spectroscopy using a Bruker AMC400 spectrometer.

Photoemulsion polymerization was done in a UV-reactor(Heraeus TQ 150 Z3).[30,33] The hydrodynamic radius wasdetermined by dynamic light scattering using a Peters ALV4000 goniometer.[33,34] Molecular-weight determination ofthe polyelectrolyte brushes grafted onto the styrene core wasdetermined using a Ubbelohde viscosimeter.[30] The pH andconductometric measurements were carried out by a Q pH70 pH meter and a Q cond 2200 conductometer, respectively.

The transmission electron microscopy (TEM) measure-ments were performed using a Philips (FEI) CM 200 FEGequipped with an energy dispersive X-ray system.

Synthesis of Polystyrene Core Particles

The polystyrene core was synthesized by the emulsion poly-merization of styrene.[30] One mole of styrene was emulsifiedby CTAB (7.29 mmoles) in 420 mL of water. The mixture wasthen heated at 80 8C under nitrogen atmosphere. A solutionof thermal initiator V-50 (1.617 mmoles) in 50 mL of waterwas introduced in the reactor. The polymerization was conti-nued for 20 min after which the photoinitiator (HMEM) wasadded under starved conditions (0.5 mL �min�1) to obtain athin shell of photoinitiator on the core particles. The amountsof the photoinitiator varied from 0.5 to 4 mol-% of the styr-ene used. After the addition of the photoinitiator, the reactionwas continued for 6 h. The latex obtained was cooled andpurified by ultrafiltration until the conductivity of the serumwas around 20–30 mS � cm�1.

Synthesis of the Core-Shell Particles byPhotoemulsion Polymerization

The amount of monomer AEMH was calculated by the mol-%of the core used in the photoemulsion polymerization, whichvaried from (10–60%). The core used was diluted to 2.5 wt.-%of water. The mixture of diluted core and the monomerwas charged into the reactor. The mixture was evacuated andflushed with nitrogen several times. Photopolymerization wascarried out by UV-Vis irradiation with constant stirring.Samples were drawn regularly from the reaction mixture andthe increase in hydrodynamic radii was followed by dynamiclight scattering (DLS). The core-shell polyelectrolyte brushesthus obtained were cleaned by ultrafiltration using distilledand deionized water (Millipore) under a pressure of 1 bar

until the conductivity of the outgoing serum dropped under2 mS � cm�1.

Formation of Gold Particles in the Brush

In a typical run, 3 mL of the polyelectrolyte brush system(3.10 wt.-%) was placed in an ultrafiltration cell, ion exchangewas then carried out by serum replacement against 5� 10�5

M

of a 2 L HAuCl4 solution. The remaining co-ions were cleanedby ultrafiltration with pure water until the conductivity of theserum was reduced to 2 mS � cm�1. The white polyelectrolytebrush system turns pale yellow after treatment with HAuCl4solution. This system bearing HAuCl4 ions was reduced usinga 5� 10�3

M solution of NaBH4. Reduction with NaBH4

changed the color of the latex to brown.

Results and Discussion

The synthesis of polystyrene core particles by emulsion

polymerization, and polyelectrolyte brushes by photo-

emulsion polymerization has been recently described in

detail by Guo et al.[30] We carried out the synthesis in a

similar way with the exception that positively charged

surfactant (CTAB) and a thermal initiator were used to give

the latex a cationic charge.[31] Photoemulsion polymeriza-

tion of these cationic latexes to obtain a brush of the

mentioned monomers proceeded exactly as described

before.[31] The formation of the brush was followed by

dynamic light scattering by observing the increase in

hydrodynamic radii. Since the radius of the nearly

monodisperse core particles is known, DLS leads directly

to the thickness L of the brush layer.[30,33,34]

As already found for the anionic systems discussed in

reference [30], the brush thickness L was found to depend on

the amount of photoinitiator used and the amount of

monomer. Increasing the amount of monomer increased the

brush thickness. Using more than 30 mol-% of monomer,

however, increased the viscosity of the latex, which may be

caused by the formation of a gel structure in the latex. The

photoemulsion polymerization needs approximately 100–

150 min to reach sufficient conversion.[30]

Purification of the core latex was highly important to

achieve a desired thickness of the brush layer. The presence

of superfluous surfactant reduced the grafting density and

increased the brush thickness. Latexes not cleansed

beforehand gave a gel even with small quantities of

monomer were used at low effective volume fractions,

whereas the cleaning of the same core to different extents

gave different brush lengths.

The spherical polyelectrolyte thus generated could be

fully characterized with regard to the contour length of the

attached chains and the grafting density.[30] The amount of

polymer grafted onto the core is determined by titration

with AgNO3 after exhaustive cleaning by ultrafiltration.

The contour length of the brush polymer was determined by

cleaving the brushes from the core surface initiated by base

Cationic Spherical Polyelectrolyte Brushes as Nanoreactors for the Generation of Gold Particles 549


hydrolysis.[30] The strong base breaks the ester bond

between the vinyl bond of the polyelectrolyte chains and

the photoinitiator. The free PAEMH chains are saponifi-

cated under these conditions to give poly(methacrylic

acid) which can be subsequently analyzed. In this way

the contour length Lc and the grafting density s can be

determined in a similar way to the methods employed

previously.[30]

Figure 2 displays the thickness L of the brush layer

attached to the surface of the particles as a function of

the concentration of added salt. There is a pronounced

shrinking of the polyelectrolyte layer with increasing

ionic strength. The same phenomenon was observed re-

cently when studying anionic spherical polyelectrolyte

brushes.[33,34] The explanation of the strong swelling at low

ionic strength is given by the confinement of the counter-

ions within the brush layer, as discussed above. In the limit

of salt-free solutions the high osmotic pressure stretches

the polyelectrolyte chains to nearly full length (‘‘osmotic

limit’’). If the ionic strength is raised, the electrostatic

interaction within the brush layer is screened more and

more. In the limit of high concentrations of added salt

(‘‘salted brush’’), the surface layer may be compared to

uncharged systems, as was recently shown.[34] These

investigations also showed that divalent ions lead to a

much more pronounced shrinking.

Figure 2 also shows that addition of AuCl4� ions leads

to a much stronger shrinking than expected for a mono-

valent salt. We assign this behavior to the complexation

of at least two �NH2 groups by one AuCl4� ion. Hence,

the strong binding of this gold compound is much enhanc-

ed through complexation as expected from data in the

literature.[29,37–40] The interaction of protonated and

neutral amines with HAuCl4 leading to the formation of

gold nanoparticles has been exploited quite often, but

the nature of the exact species formed is not very clear.

Formation of a complex between cetyltrimethyl ammonium

bromide and HAuCl4 is used for the formation of gold

nanorods.[41] The protonation of primary alkylamines by

HAuCl4 in organic solvents has also been used for the

formation of gold nanoparticles and for metal-ion extrac-

tion.[42] The interaction of heterocyclic amines and HAuCl4has also been studied. RþAuCl4

� species are formed with

AuCl4� as counter-ions at low pH whereas complexes are

formed at neutral pH.

The PAEMH chains could be converted into the

uncharged amino-form by treatment with strong base and

subsequent purification. In this way a dense surface layer of

poly(aminoethyl methacrylate) chains is generated that can

be used for complexation of other metal ions as well.

Figure 3 demonstrates that the conversion of the poly(-

aminoethyl methacrylate hydrochloride) chains into an

uncharged polymer is accompanied by a strong decrease of

the layer thickness L as expected. The latex is still stable

against coagulation, however, and may be used for the up-

take of ions again. Hence, the neutral amino form of the

brush is titrated with HAuCl4. No salt is added in this

procedure to ensure the full uptake of AuCl4� ions into the

brush. Since the total number of charges is known with good

precision, this method allows the calculation of the amount

of the precious metal ions necessary for full conversion.

Because of the high affinity of the brush particles for AuCl4�

ions, no ions are left in the solution. This can be shown by

DLS: Titration with HAuCl4 does not lead to the strong

swelling as expected for a titration with HCl, but to the same

layer thickness L as observed previously (see the discussion

of Figure 2).

Having shown that aurochlorate ions are fully confined

within the brush layer, we now turn to the reduction of these

ions within the ‘‘nanoreactor’’. This is achieved by the

addition of dilute solutions of NaBH4 to the latex. The re-

duction is immediately obvious from a change in the color

Figure 2. Brush thickness L as a function of concentration ca ofadded salt: (*) NaCl, and (&) HAuCl4 (core radius: 42 nm, brushthickness: 80 nm).

Figure 3. The brush thickness L is plotted against pH of thesuspension. The parameter in the different curves is the ionicstrength (core radius: 40 nm, brush thickness: 160 nm):~: 0.0001M; *: 0.001 M; &: 0.01 M; �: 1 M.



of the suspension from yellow to brown. No change in color

is observed afterwards suggesting that gold particles do not

aggregate. A shallow maximum becomes visible in the UV/

Vis spectra, which has a maximum around a wavelength of

502–512 nm. This is to be expected for Au nanoparticles of

less than 3 nm in size. The particles having diameter of more

than 3 nm absorb at 520 nm and exhibit a red color.

Uncovered gold particles, made by radiolysis, with an

average diameter of less than 1.5–2.5 nm show an

absorbance between 490–500 nm.[43] These particles are

orange–yellow in color. The presence of impurities or any

capping agents attached to the surface of the particles that

absorb in the 200–800 nm range modifies the spectrum and

gives a different maximum for the same particle sizes.[44]

For the present system, however, the strong scattering of the

latex itself precludes further analysis of the UV/Vis spectra.

The resonance plasmon of gold appears only as a shoulder

in the UV/Vis spectrum of the latex.

Transmission electron microscopy (TEM) gives direct

proof that Au nanoparticles have been formed on the

surface, that is, in the ‘‘nanoreactors’’. Figure 4a displays

TEM pictures of several particles while Figure 4b shows an

enlarged view of the nanoparticles. It is evident that the Au

particles have a diameter of ca. 2 nm and exhibit a rather

narrow size distribution. Moreover, Figure 4b demonstrates

that the particles are fully crystalline as expected.

Obviously, the nucleation of the Au particles takes place

on many spots on the surface. This leads to particles that are

considerably smaller than the Au particles grown in

solution under the same conditions. No further aggregation

Figure 4. a) TEM pictures of gold particles on core-shell system; b) TEM picture at highmagnification showing the lattice fringes of the crystalline gold particles.

Cationic Spherical Polyelectrolyte Brushes as Nanoreactors for the Generation of Gold Particles 551


takes place under these conditions. This observation sug-

gests that nucleation of nanoparticles can be controlled in

these systems. Further studies are under way to investigate

whether this finding can be generalized for other metal

nanoparticles as well.

During the process of reduction, the latex remains stable.

The onset of a slow aggregation is only observed after

several days. The factors affecting the stability of the latex

covered by gold particles must still be investigated in more

detail. For the present investigation it suffices to note that

the colloidal stability is not impeded by the chemical

reaction taking place in the nanoreactors.

It should be noted in this context that no reagents have

been used for the stabilization of the Au particles. Hence,

their surface could be used for immobilizing further func-

tional groups on the surface of the Au particles like sulfur

or phosphorus containing reagents.

Conclusion

We have shown that cationic spherical polyelectrolyte

brushes can be used as nanoreactors for the formation of Au

particles. The brushes immobilize metal ions by counter-

ion exchange and complexation. The localized metal ions

can be reduced by a suitable reagent to yield nanosized

metal particles which are nearly monodisperse and crystal-

line in nature.

[1] F. Caruso, H. Lichtenfeld, M. Giersig, H. Mohwald, J. Am.Chem. Soc. 1998, 120, 8523.

[2] F. Caruso, H. Mohwald, Langmuir 1999, 15, 8276.[3] A. Dokoutchaev, J. T. James, S. C. Koene, S. Parthak, C. K. S.

Prasak, M. E. Thomson, Chem. Mater. 1999, 11, 2389.[4] O. Siiman, A. Burshteyn, J. Phys. Chem. B 2000, 104, 9795.[5] T. Ji, V. G. Lirtsman, Y. Avny, D. Davidov, Adv. Mater. 2001,

13, 1253.[6] M. A. Khan, Ch. Perruchot, S. P. Armes, D. P. Randall,

J. Mater. Chem. 2001, 11, 2363.[7] F. Caruso, M. Spasova, V. Salgueirino-Maceira, L. M.

Liz-Marzan, Adv. Mater. 2001, 13, 1090.[8] D. I. Gittins, A. S. Susha, B. Schoeler, F. Caruso, Adv. Mater.

2002, 14, 508.[9] Ch. Graf, A. van Blaaderen, Langmuir 2002, 18, 524.

[10] V. G. Pol, A. Gedanken, J. Calderon-Moreno, Chem. Mater.2003, 15, 1111.

[11] A. Ghosh, C. R. Patra, P. Mukherjee, M. Sastry, R. Kumar,Microporous Mesoporous Mater. 2003, 58, 201.

[12] A. P. Alivisatos, Science 1996, 271, 933.

[13] M. M. Alvarez, J. T. Khoury, T. G. Schaaf, M. N. Shafigullin,I. Vezmar, R. L. Whetten, J. Phys. Chem. B 1997, 101, 3706.

[14] W. P. McConnell, J. P. Novak, L. C. Brousseau III, R. R.Fuierer, R. C. Tenent, D. L. Feldheim, J. Phys. Chem. B 2000,104, 8925.

[15] T. P. Bigioni, R. L. Whetten, O. Dag, J. Phys. Chem. B 2000,104, 6983.

[16] S. Link, M. El-Sayed, J. Phys. Chem. B 1999, 103, 4212[17] B. Gates, Y. Xia, Appl. Phys. Lett. 2001, 78, 3178.[18] S. Lal, S. L. Westcott, R. N. Taylor, J. B. Jackson,

P. Nordlander, N. J. Halas, J. Phys. Chem. B 2002, 106,5609.

[19] Z. Liang, A. S. Susha, F. Caruso, Adv. Mater. 2002, 14, 1160.[20] J. P. Spatz, S. Moßmer, M. Moller, Chem. Eur. J. 1996, 2,

1552.[21] J. P. Spatz, S. Moßmer, Ch. Hartmann, M. Moller, T. Herzog,

M. Krieger, H.-G. Boyen, P. Ziemann, B. Kabius, Langmuir2000, 16, 407.

[22] M. Zhao, R. M. Crooks, Angew. Chem. 1999, 111, 375;Angew. Chem. Int. Ed. 1999, 38, 364.

[23] M. Zhao, R. M. Crooks, Chem. Mater. 1999, 11, 3379.[24] F. Grohn, B. J. Bauer, Y. A. Akpalu, C. L. Jackson, E. J. Amis,

Macromolecules 2000, 33, 6042.[25] Y. Li, M. A. El-Sayed, J. Phys. Chem. B 2001, 105, 8938.[26] R. West, Y. Wang, T. Goodson III, J. Phys. Chem. B 2003,

107, 3419.[27] R. Djalali, S.-Y. Li, M. Schmidt, Macromolecules 2002, 35,

4282.[28] S. Liu, J. V. M. Weaver, M. Save, S. P. Armes, Langmuir

2002, 18, 8350.[29] J. H. Youk, Polymer 2003, 44, 5053.[30] X. Guo, A. Weiss, M. Ballauff, Macromolecules 1999, 32,

6043.[31] Y. Mei, A. Wittemann, G. Sharma, M. Ballauff, Th. Koch, H.

Gliemann, J. Horbach, Th. Schimmel, Macromolecules2003, 36, 3452.

[32] P. Pincus, Macromolecules 1991, 24, 2912.[33] X. Guo, M. Ballauff, Langmuir 2000, 16, 8719.[34] X. Guo, M. Ballauff, Phys. Rev. E 2001, 64, 051406.[35] B. Das, X. Guo, M. Ballauff,Prog. Colloid Polym. Sci. 2002,

121, 34.[36] I. M. Martinez, J. Forcada, J. Polym. Sci. Part. A: Polym.

Chem. 2000, 38, 4230.[37] A. Kumar, S. Mandal, P. R. Selvakannan, R. Pasricha, A. B.

Mandale, M. Sastry, Langmuir 2003, 19, 6277.[38] S. Bharathi, O. Lev, Chem. Commun. 1997, 2303.[39] G. V. Myasoedov, N. I. Shcherbinina, E. A. Zakhartchenko,

S. S. Kolobov, L. V. Lileeva, P. N. Komozine, I. N. Marov,V. K. Belyaeva, Sol. Extraction Ion. Exchange 1997, 15,1107.

[40] D. V. Leff, L. Brandt, J. R. Heath, Langmuir 1996, 12, 4723.[41] B. Nikoobakht, M. A. El-Sayed, Chem. Mater. 2003, 15,

1957.[42] F. J. Alguacil, S. Martinez, A. M. Sastre, J. Chem. Res. 2001,

384.[43] A. Hengelein, Langmuir 1999, 15, 6738.[44] D. G. Duff, A. Baiker, Langmuir 1993, 9, 2301.



Documents

Cationic Spherical Polyelectrolyte Brushes as Nanoreactors for the Generation of Gold Particles