Gold Colloids from Cationic Surfactant Solutions. 1.

Gold Colloids from Cationic Surfactant Solutions.1.

Mechanisms That Control Particle Morphology

Epameinondas Leontidis,*,?Konstantina Kleitou,?Tasoula Kyprianidou-Leodidou,?Vlasoula Bekiari,?and Panagiotis Lianos?

Department of Chemistry,University of Cyprus,P.O.Box20537,1678Nicosia,Cyprus,and Engineering Science Department,University of Patras,GR-26500Patras,Greece

Received August29,2001.In Final Form:January23,2002

The mechanism of formation of gold particles by reduction of Au III in solutions of alcyltrimethylammonium chloride surfactants was studied in the absence and in the presence of NaCl.Au III anions interact strongly with trimethylammonium cations forming insoluble ion pairs(Torigoe et https://www.360docs.net/doc/fe14688218.html,ngmuir1992,8,59).Above the surfactant critical micelle concentration,the ion pairs are solubilized in the micelles returning to the solution.Gold particles were produced by photochemical reduction of the clear micellar solutions.The coupling between surfactant aggregation and inorganic crystallization phenomena in these systems was investigated using transmission electron microscopy(TEM),UV-vis,and time-resolved fluorescence spectroscopy.At concentrations close to the phase boundary of the L1phase with the lyotropic liquid crystalline phases many gold particles have a threadlike morphology,as previously noted by Esumi et al. (Langmuir1995,11,3285).The presence of NaCl modifies the micellar size and affects the gold crystallization process in surprising and unexpected ways,as evidenced by intermediate structures observed by TEM. Our observations support the idea that the formation of threadlike gold particles occurs primarily through a combination of crystal aggregation and specific crystal face stabilization and not through templating mechanisms.

I.Introduction

Gold colloids have fascinated science from the times of the alchemists.Although the modern era of research on gold colloids was initiated by Michael Faraday150years ago,1the scientific interest has always been strong.In recent years,the emergence of nanotechnology and the evolution of modern analytical methods that allow easy observation and manipulation of nanoparticles have created an explosion of interest in the synthesis and study of gold clusters and colloids.2A major current research theme is to tailor the size and shape of the particles at will using colloid chemistry methods.While size-control and stabilization have been demonstrated in numerous cases, and ingenious new methods for the production and stabilization of gold nanoparticles have been reported,3 only a limited number of papers dealing with effective shape control of gold particles exists in the literature. Controlling the shape of nanoparticles is technologically important,since the optical,electronic,magnetic,and catalytic properties of a nanomaterial often depend critically not only on particle size but also on particle shape.4In the case of gold,for example,it was recently demonstrated that the longitudinal plasmon resonance shifts to higher wavelengths with increasing aspect ratio of the particles5and also that nanoparticles with a high aspect ratio exhibit a strong fluorescence enhancement.6 Forcing colloidal particles to acquire a nonspherical, elongated shape is no easy matter as it involves a fight against thermodynamics,which dictates that the mini-mum free energy structure is that with the minimum specific area,hence a sphere,a cube,or some other compact shape.Forcing gold particles to acquire shapes with high aspect ratios using wet chemical routes has been dem-onstrated only in a handful of experimental systems to date.In all these cases it is either certain or possible that a templating mechanism is operating.The rodlike gold particles,prepared in the pores of alumina membranes by electrochemical reduction,are a clear demonstration of the success of a hard template.7Formation of rodlike gold particles by an electrochemical reduction method in a surfactant solution was demonstrated by the group of

*To whom all correspondence should be addressed.?University of Cyprus.

?University of Patras.

(1)Faraday,M.Philos.Trans.R.Soc.London,Ser.A1857,147,145.

(2)(a)Henglein,A.Top.Curr.Chem.1988,143,113.(b)Henglein,

A.Chem.Rev.1989,89,1861.(c)Schmid,G.Chem.Rev.1992,92,1709.

(d)Wang,Z.-L.Adv.Mater.1998,10,13.(e)Templeton,A.C.;Wuelfing, P.W.;Murray,R.W.Acc.Chem.Res.2000,33,27.

(3)(a)Brust,M.;Walker,M.;Bethell,D.;Schiffrin,D.J.;Whyman, https://www.360docs.net/doc/fe14688218.html,mun.1994,801.(b)Yee,C.K.;Jordan,R.;Ulman,A.; White,H.;King,A.;Rafailovich,M.;Sokolov,https://www.360docs.net/doc/fe14688218.html,ngmuir1999,15, 3486.(c)Templeton,A.C.;Chen,S.;Gross,M.;Murray,https://www.360docs.net/doc/fe14688218.html,ngmuir 1999,15,66.(d)Han,M.Y.;Quek,C.H.;Huang,W.;Chew,C.H.;Gan, L.M.Chem.Mater.1999,11,1144.(e)Warner,M.G.;Reed,S.M.; Hutchison,J.E.Chem.Mater.2000,12,3316.(f)Bronstein,L.M.; Chernyshov,D.M.;Valetsky,P.M.;Wilder,E.A.;Spontak,R.J. Langmuir2000,16,8221.(g)Brown,K.R.;Walter,D.G.;Natan,M. J.Chem.Mater.2000,12,306.(h)Zhou,Y.;Itoh,H.;Uemura,T.;Naka, K.;Chujo,https://www.360docs.net/doc/fe14688218.html,mun.2001,613.(i)Yonezawa,T.;Yasui,K.; Kimizuka,https://www.360docs.net/doc/fe14688218.html,ngmuir2001,17,271.(j)Shon,Y.-S.;Wuelfing,P.W.; Murray,https://www.360docs.net/doc/fe14688218.html,ngmuir2001,17,1255.

(4)(a)Creighton,J.A.;Eadon,D.G.J.Chem.Soc.,Faraday Trans. 1991,87,3881.(b)Dirix,Y.;Bastiaansen,C.;Caseri,W.;Smith,P.Adv. Mater.1999,11,223.(c)Dickson,R.M.;Lyon,L.A.J.Phys.Chem.B 2000,104,6095.

(5)(a)Van der Zande,B.M.I.;Bo¨hmer,M.R.;Fokkink,L.G.J.; Scho¨nenberger,C.J.Phys.Chem.1997,101,852.(b)Link,S.;Mohamed, M.B.;El-Sayed,M.A.J.Phys.Chem.B1999,103,3073.(c)Link,S.; El-Sayed,M.A.J.Phys.Chem.B1999,103,8410.

(6)Mohamed,M.B.;Volkov,V.;Link,S.;El-Sayed,M.A.Chem. Phys.Lett.2000,317,517.

(7)(a)Tierney,M.J.;Martin,C.R.J.Phys.Chem.1989,93,2878.

(b)Foss,C.A.;Tierney,M.J.,Martin,C.R.J.Phys.Chem.1992,96, 9001.(c)Foss,C.A.;Hornyak,G.L.;Stockert,J.A.;Martin,C.R.J. Phys.Chem.1994,98,2963.(d)Hornyak,G.L.;Patrissi,C.J.;Martin, C.R.J.Phys.Chem.B1997,101,1548.(e)Al-Rawashdeh,N.A.F.; Sandrock,M.L.;Seugling,C.J.;Foss,C.A.J.Phys.Chem.B1998,102, 361.(f)Van der Zande,B.M.I.;Bo¨hmer,M.R.;Fokkink,L.G.J.; Scho¨nenberger,https://www.360docs.net/doc/fe14688218.html,ngmuir2000,16,451.

3659

Langmuir2002,18,3659-3668

10.1021/la011368s CCC:$22.00?2002American Chemical Society

Published on Web04/02/2002

Wang8and postulated to depend on some unidentified templating mechanism.Gold particles formed with the previous two methods have been used in very interesting experiments,including thermal and laser-shape anneal-ing9and examination of the crystal faces by high-resolution transmission electron microscopy(TEM),10while they have shown interesting light propagation,4c absorption,5fluo-rescence,6and self-assembly11properties.A third method to produce long,threadlike gold particles s although not exclusively s was advanced by Esumi and Torigoe in a series of publications.12In their method,gold particles are produced by photochemical reduction in the presence of cationic surfactant micelles.The authors expressed the opinion that the templating effect of rodlike micelles is essential for the formation of particles with very large aspect ratios in these systems.12b,e

Despite the reported successful attempts to modify gold crystal shapes in surfactant solutions,8,12little is known about the actual mechanism behind shape control.Rela-tively few recent studies are concerned with the formation mechanism of gold colloids,13even though such studies have appeared in much older pioneering work.14While important investigations using theγ-irradiation method have focused on the reduction steps of the reaction AuCl4-f Au0f Au n,15gold particle formation by photoreduction in the presence of surfactants and electrolytes is little understood.

We have decided to further investigate the formation of gold particles in trimethylammonium surfactant solu-tions to achieve a better understanding of the crystal-lization process in the presence of surfactant aggregates. The solution of cationic trimethylammonium surfactants is simpler than other systems that have been used so far. In addition,these surfactants appear to play a rather unique role in crystallizing systems,as they are involved in a variety of cases,in which rodlike particles are formed. The most notable example is the synthesis of a large range of mesoporous zeolites,which is thought to involve either cylindrical micelles or the organization of micelles into bundles during the gelation process.16Other cases include the electrochemical method of Wang discussed before,8 the production of silver through reduction by ascorbic acid,17and the formation of nanowires of CdS and Mo3Se3-.18We would like to understand the formation mechanism of the rodlike or threadlike particles.Is this really a templating mechanism,requiring the presence of long micelles or not?There is much discussion about“soft”templating phenomena in surfactant solutions in the literature.Important work by the group of Pileni,which examines the formation of rodlike Cu particles in sur-factant lyotropic phases,is a related example.19Recent evidence from Pileni’s group suggests that templating is not the only and sometimes probably not the dominant force operating in the particle production process.19c,d A second issue that we address for the first time in this work is the effect of additional electrolyte on gold crystallization.Ions present in such solutions may not only affect micellar sizes and shapes and gold particle interactions but also interfere with the photochemical reduction process and affect the activity and solubility of various components.Finally,ions affect the coordination chemistry of Au III,a fact that is overlooked in recent literature.To examine electrolyte effects on gold particle growth,we have conducted parallel experiments in the presence and absence of excess NaCl.

Wishing to keep the crystallizing system as simple as possible,we have opted to avoid stabilizers or chemical reducing agents and have used photochemical reduction almost exclusively in this work.Photochemical reduction has been used as a method for the preparation of gold colloids in several recent publications.12,20Its main disadvantage is that it does not permit efficient nucleation control:nuclei are continuously formed,as long as gold precursor ions exist in solution.

We make extensive use of TEM to investigate gold particle formation.TEM has been used to study gold crystal formation already from the early years of its invention14and is a necessary tool in gold nanoparticle research.21The main disadvantage of TEM is that one must remove the solvent before viewing the particles, which sometimes leads to aggregate formation on the TEM grid.22Finally,time-resolved fluorescence spectroscopy with pyrene as a fluorescent probe is used to examine potential changes in the surfactant aggregates present in our solutions.Fluorescence was preferred over the more

(8)(a)Yu,Y.-Y.;Chang,S.-S.;Lee,C.-L.;Wang,C.R.J.Phys.Chem. B1997,101,6661.(b)Chang,S.-S.;Shih,C.-W.;Chen,C.-D.;Lai,W.-

C.;Wang,https://www.360docs.net/doc/fe14688218.html,ngmuir1999,15,701.

(9)(a)Mohamed,M.B.;Ismail,K.Z.;Link,S.;El-Sayed,M.A.J. Phys.Chem.B1998,102,9370.(b)Link,S.;Burda,C.;Mohamed,M.

B.;Nikoobakht,B.;El-Sayed,M.A.J.Phys.Chem.A1999,103,1165.

(c)Mohamed,M.B.;Wang,Z.L.;El-Sayed,M.A.J.Phys.Chem.B 1999,103,10255.(d)Takami,A.;Kurita,H.;Koda,S.J.Phys.Chem. B1999,103,1226.(e)Maye,M.M.;Zheng,W.;Leibowitz,F.L.;Ly,N.

K.;Zhong,https://www.360docs.net/doc/fe14688218.html,ngmuir2000,16,490.

(10)(a)Wang,Z.L.;Mohamed,M.B.;Link,S.;El-Sayed,M.A.Surf. Sci.1999,440,L809.(b)Wang,Z.L.J.Phys.Chem.B2000,104,1153.

(c)Wang,Z.L.;Gao,R.P.;Nikoobakht,B.;El-Sayed,M.A.J.Phys. Chem.B2000,104,5417.

(11)(a)Nikoobakht,B.;Wang,Z.L.;El-Sayed,M.A.J.Phys.Chem. B2000,104,8635.(b)Dujardin,E.;Hsin,L.-B.;Wang,C.R.C.;Mann, https://www.360docs.net/doc/fe14688218.html,mun.2001,1264.

(12)(a)Torigoe,K.;Esumi,https://www.360docs.net/doc/fe14688218.html,ngmuir1992,8,59.(b)Esumi,K.; Matsuhisa,K.;Torigoe,https://www.360docs.net/doc/fe14688218.html,ngmuir1995,11,3285.(c)Esumi,K.; Megumi,N.;Aihara,N.;Usui,K.New J.Chem.1998,22,719.(d)Esumi, K.;Hara,J.;Aihara,N.;Usui,K.;Torigoe,K.J.Colloid Interface Sci. 1998,208,578.(e)Kameo,A.;Suzuki,A.;Torigoe,K.;Esumi,K.J. Colloid Interface Sci.2001,241,289.

(13)(a)Chen,S.;Templeton,A.C.;Murray,https://www.360docs.net/doc/fe14688218.html,ngmuir2000, 16,3543.(b)Jana,N.R.;Gearheart,L.;Murphy,C.J.Chem.Mater. 2001,13,2313.(c)Okitsu,K.;Yue,A.;Tanabe,S.;Matsumoto,H.; Yobiko,https://www.360docs.net/doc/fe14688218.html,ngmuir2001,17,7717.

(14)Turkevich,J.;Stevenson,P.C.;Hillier,J.Discuss.Faraday Soc. 1951,11,55.

(15)(a)Kurihara,K.;Kizling,J.;Stenius,P.;Fendler,J.H.J.Am. Chem.Soc.1983,105,2574.(b)Mosseri,S.;Henglein,A.;Janata,E. J.Phys.Chem.1989,93,6791.(c)Gachard,E.;Remita,H.;Khatouri, J.;Keita,B.;Nadjo,L.;Belloni,J.New J.Chem.1998,22,1257.(d) Henglein,A.;Meisel,https://www.360docs.net/doc/fe14688218.html,ngmuir1998,14,7392.(e)Henglein,A. Langmuir1999,15,6738.(f)Bronstein,L.;Chernyshov,D.;Valetsky, P.;Tkachenko,N.;Lemmetyinen,H.;Hartmann,J.;Fo¨rster,S. Langmuir1999,15,83.

(16)(a)Beck,J.S.;Vartuli,J.C.Curr.Opin.Solid State Mater.Sci. 1996,1,76.(b)Ying,J.Y.;Mehnert,C.P.;Wong,M.S.Angew.Chem., Int.Ed.1999,38,56.

(17)(a)Jana,N.R.;Gearheart,L.;Murphy,https://www.360docs.net/doc/fe14688218.html,mun. 2001,617.(b)Jana,N.R.;Gearheart,L.;Murphy.C.Adv.Mater.2001, 13,1389.

(18)(a)Chen,C.-C.;Chao,C.-Y.,Lang,Z.-H.Chem.Mater.2000,12, 1516.(b)Messer,B.;Song,J.H.;Huang,M.;Wu,Y.;Kim,F.;Yang,P. Adv.Mater.2000,12,1526.

(19)(a)Pileni,M.-P.;Gulik-Krzywicki,T.;Tanori,J.;Filankembo,

A.;Dedieu,https://www.360docs.net/doc/fe14688218.html,ngmuir1998,14,7359.(b)Pileni,M.-P.;Ninham,

B.W.;Gulik-Krzywicki,T.;Tanori,J.;Lisiecki,I.;Filankembo,A.Adv. Mater.1999,11,1358.(c)Filankembo,A.;Pileni,M.-P.J.Phys.Chem. B2000,104,5865.(d)Pileni,https://www.360docs.net/doc/fe14688218.html,ngmuir2001,17,7476.

(20)(a)Yonezawa,Y.;Sato,T.;Ohno,M.;Hada,H.J.Chem.Soc., Faraday Trans.11987,83,1559.(b)Itakura,T.;Torigoe,K.;Esumi, https://www.360docs.net/doc/fe14688218.html,ngmuir1995,11,4129.(c)Sato,T.;Ito,T.;Iwabuchi,H.;Yonezawa, Y.J.Mater.Chem.1997,7,1837.(d)Moriguchi,I.;Fujiyoshi,N.; Sakamoto,R.;Teraoka,Y.;Kagawa,S.Colloids Surf.,A1997,126,159.

(e)Ravaine,S.;Fanucci,G.E.;Seip,C.T.;Adair,J.H.;Talham,D.R. Langmuir1998,14,708.(f)Zhou,Y.;Wang,C.Y.;Zhu,Y.R.;Chen, Z.Y.Chem.Mater.1999,11,2310.(g)Han,M.Y.;Quek,https://www.360docs.net/doc/fe14688218.html,ngmuir 2000,16,362.

(21)Wang,Z.L.J.Phys.Chem.B2000,104,1153.

(22)(a)Ohara,P.C.;Heath,J.R.;Gelbart,W.M.Angew.Chem.,Int. Ed.Engl.1997,36,1078.(b)Haidara,H.;Mougin,K.;Schultz,J. Langmuir2001,17,659.(c)Maillard,M.;Motte,L.;Pileni,M.-P.Adv. Mater.2001,13,200.

3660Langmuir,Vol.18,No.9,2002Leontidis et al.

conventional scattering methods for reasons that will become apparent below.

II.Experimental Section

Chemicals.Tetrachloroauric acid(HAuCl4),its sodium salt (NaAuCl4),pyrene,and dodecyl-(DTAC)and hexadecyltri-methylammonium chloride(CTAC)were obtained from Fluka. All other chemicals were obtained from Sigma,Aldrich,or Fluka. All were of analytical grade and used without further purification.

Solution Preparation.Dilute solutions were simply made by mixing appropriate volumes of separate aqueous solutions of AuCl4-(5mmol/L),surfactant(usually0.1M),and electrolyte. For concentrated solutions(25-45wt%surfactant),we dissolve the appropriate amount of solid surfactant within an aqueous solution containing the gold salt and the electrolyte.Concentrated DTAC solutions were found to have a density close to that of pure water.For DTAC,which was used much more extensively, a solution of35wt%is roughly1.3M.For the concentrated surfactant solutions,we have kept the molar ratio,r g,of gold to surfactant equal to1/60,which we prove below to be sufficient for complete solubilization of the gold-surfactant ion pair in the micelles.Fresh solutions were used for reduction;they were protected to avoid partial gold reduction by ambient light.

Gold(III)Reduction.The photochemical reduction of the gold solutions was carried out under air,using a high-pressure 400-W Hg lamp.Although some oxidation of newly formed gold nanocrystals is possible and has been reported in some cases,15e,23 gold is much less sensitive than other metals.Previous photo-chemical work with gold was also made mostly under air.12,20 Three to four milliliters of the solutions were placed in small glass vials with a wide opening.The vials were held inside a jacketed beaker that contained water thermostated at25°C. This precaution was found necessary,since the powerful Hg lamp can heat up the samples significantly.Loss of water by evapora-tion during the reduction process can be particularly troublesome, especially for concentrated surfactant solutions that are close to the phase boundaries with lyotropic phases.After the prescribed reduction time,each vial was removed,capped with Parafilm, covered with aluminum foil,and stored in a dark place,prior to further analysis.We have also occasionally reduced Au III with sodium borohydride or hydrazine,but we invariably found that, in the absence of specific stabilizers,chemical reduction proceeds always in a fast,uncontrollable way and produces large ag-gregates.

Spectroscopic Measurements.UV-vis absorption spectra of gold-containing surfactant solutions were obtained on a Shimadzu UV-160A or on a UV-1601spectrometer.FTIR spectra of the ion pairs that precipitate upon mixing equal amounts of surfactant cations and AuCl4-were measured with the KBr pellet method on a Shimadzu FTIR-8900spectrometer.Time-resolved fluorescence measurements were made by registering fluores-cence decay profiles with the photon-counting technique,using a homemade nanosecond hydrogen flash lamp and ORTEC electronics.

Electron Microscopy.TEM was performed at the Institute of Neurology and Genetics,Cyprus,on a JEOL-1010A instrument, with an acceleration voltage of80kV.Drops of the solutions were put on Formvar-coated grids and left there a few minutes for water evaporation.Concentrated solutions(25-45wt% surfactant)were diluted with sufficient amounts of water,before being examined by TEM.

III.Results

III.1.On Ion-Pair Formation between DTAC and AuCl4-and the Coordination of Au III.To understand the gold reduction results presented below,we must first consider two important points.The first is the formation of insoluble1:1ion pairs between AuCl4-and cationic surfactants,12,24a general phenomenon exhibited by other complex ions of heavy metals,25which has been effectively used for the production of mesostructured materials from alkylamines.26

Esumi et al.have observed that the ion pairs formed between C n TAC and AuCl4-can be solubilized in the micelles formed in excess surfactant.Because we would like to work with clear solutions and avoid precipitates that would hinder our mechanistic investigation,we identified the surfactant-to-gold ratio,beyond which complete dissolution of the precipitated ion-pairs occurs, in the absence and in the presence of1M NaCl.We have prepared solutions with a AuCl4-concentration of1 mmol/L and surfactant concentrations smaller than0.1 M.For small surfactant concentrations the final solutions were turbid as expected.Precipitation of the formed ion pairs was enhanced by centrifugation.The FTIR spectrum of the precipitate clearly showed all the expected bands of the dodelyltrimethylammonium ion.Increasing sur-factant concentration above the critical micelle concen-tration(cmc),we observed a gradual dissolution of the precipitate in the supernatant after centrifugation,with a corresponding increase of AuCl4-absorbance.The positions of the two principal gold bands are originally located at221and295nm in the absence of NaCl and at 228and312nm in the presence of NaCl.This is a proof that AuCl4-exists in its hydrolyzed form in the absence of NaCl in dilute DTAC solutions.However,with increas-ing DTAC concentration the gold-surfactant adducts return to the solution and the gold peaks are now found at235and332nm(Figure1a).This implies that the gold-surfactant adduct is solubilized intact in the micelles formed above the cmc,as already discussed by Esumi et al.12b,e In Figure1b we plot the absorbance of the solution at the new band maximum(332nm)and observe that(a) the absorbance starts increasing above the cmc(which is equal to20mmol/L for DTAC at25°C),27and(b)levels off at a surfactant concentration of60-70mmol/L. Observation(a)is a strong indication that we are dealing with a solubilization phenomenon involving the micelles. Observation(b)indicates that the ratio of micellized surfactant to Au III required for complete solubilization of the ion pairs is roughly equal to([DTAC]plateau-cmc)/ [AuCl4-]≈50.Given the aggregation numbers of DTAC micelles(50-100,see below),we anticipate that each micelle contains on average one to two gold-surfactant adducts.In the ensuing experiments we have kept the ratio r g)[Au III]/[DTAC]equal to1/60,with two goals in mind:(a)that all Au III species be associated with the surfactant micelles,and(b)that the solutions be clear and do not contain precipitates that would hinder the mechanistic investigation.

The second important point to consider is that AuCl4-is partly hydrolyzed through the following reaction:28-30 The equilibrium constant for this reaction at298K was reported to be equal to2.4×10-6M2in dilute aqueous

(23)Weaver,S.;Taylor,D.;Mills,https://www.360docs.net/doc/fe14688218.html,ngmuir1996,12,4618.

(24)(a)Buslaeva,T.M.;Sinitsyn,N.M.;Samarova,L.V.;Koteneva, N.A.Rus.J.Inorg.Chem.1989,34,882.(b)Moriguchi,I.;Fujiyoshi, N.;Sakamoto,R.;Teraoka,Y.;Kagawa,S.Colloids Surf.,A1997,126, 159.

(25)(a)Yonezawa,T.;Tominaga,T.;Richard,D.J.Chem.Soc.,Dalton Trans.1996,783.(b)Yonezawa,T.;Toshima,N.;Wakai,C.;Nakahara, M.;Nishinaka,M.;Tominaga,T.;Nomura,H.Colloids Surf.,A2000, 169,35.

(26)(a)Antonelli,D.M.;Ying,J.Y.Chem.Mater.1996,8,874.(b) Antonelli,D.M.;Ying,J.Y.Angew.Chem.,Int.Ed.Engl.1996,35,426.

(c)Antonelli,D.M.;Nakahira,A.;Ying,J.Y.Inorg.Chem.1996,35, 3126.(d)Liu,P.;Moudrakovski,I.L.;Liu,J.;Sayari,A.Chem.Mater. 1997,9,2513.(e)Do,J.;Jacobson,A.J.Chem.Mater.2001,13,2436.

(27)Malliaris,A.;Lang,J.;Zana,R.J.Colloid Interface Sci.1986, 110,237.

AuCl

4

-+H

2

O h AuCl

3

OH-+H++Cl-

Gold Colloids from Cationic Surfactant Solutions Langmuir,Vol.18,No.9,20023661

solutions,29although a smaller value (7×10-7M 2)has also been reported.30Further chloride substitution by hydroxide ions in the coordination sphere of gold is possible at very dilute solutions,while water may also play the role of ligand.Chemical equilibrium calculations show that in the absence of electrolyte and at small surfactant concentrations,a significant percentage of Au III must be in the form of AuCl 3OH -or of even more strongly hydrolyzed complexes,a fact corroborated by the spectra in Figure 1a.This may have an impact on the photore-duction process,since each complex species is character-ized by a different reduction potential,while a different reduction pathway may be followed in each case.However,at the high DTAC and NaCl concentrations used in the experiments presented in this work,there is always abundant chloride in the system,which should force most of the Au III to be in the AuCl 4-form.We have measured the pH of the 35%DTAC solutions and found it to be 2.37in the absence of NaCl and 2.45in the presence of NaCl.The overall AuCl 4-concentration being 0.022M,these pH values imply that the degree of hydrolysis of AuCl 4-before reduction may be as high as 20%.31

III.2.Time Evolution of Gold Particle Formation in the Absence and Presence of NaCl.According to recent results by Esumi’s group,12e threadlike gold par-ticles are obtained by irradiation in DTAC solutions,provided that the surfactant concentration is in excess of 20wt %.To examine the formation mechanism of threadlike gold particles in concentrated DTAC solutions,we have undertaken experiments,in which identical samples containing 35wt %DTAC and having ratio r g )1/60were irradiated for different periods of time.A second set of samples identical to the first,but also containing 0.65M NaCl,was investigated in a similar way.The sequence of TEM pictures presented in Figure 2was obtained for the salt-free case.Initially,small irregular aggregates of gold nuclei are observed.The individual nuclei are smaller than 5nm in size,their aggregates being of the order of 50nm (Figure 2a).After 10min of irradiation,the aggregates have undergone a sintering process (as manifested by their irregular contours)and continue growing to sizes of about 30nm (Figure 2b).One can observe numerous such compact particles with a relatively narrow size distribution.At 15min (picture available in the supporting material)a distribution of spherical particles with sizes ranging from 15to 75nm is observed.A wider size distribution is indeed expected as new nuclei are continuously formed in this system,because of the continuous irradiation.The TEM picture obtained from a sample after 35min of irradiation is revealing.In a collection of spherical particles,we now find some particles with rodlike shape and with morphol-ogy strongly suggestive of linear particle aggregation (pointed by arrows in Figure 2c).At the same time,we observe new nuclei continuously forming in the system.After 90min the system has evolved a number of particle morphologies ranging from spherical to rodlike,with a few large,thin,triangular or polygonal particles that are often observed in gold reduction work and are known to result from Ostwald ripening (Figure 2d).32For longer irradiation times the percentage of rodlike particles increases.

In the presence of excess NaCl,the TEM pictures are quite remarkable and present a completely different evolution of the crystallizing system (Figure 3).At times smaller than 15min we observe the formation of thin,irregular surfactant-based structures,as evidenced by their low electron contrast (Figure 3a).These structures appear to evolve into delicate dendritic forms after 30min of irradiation,which break up on the microscope grid (Figure 3b).We believe that these dendrites are genuine structures in the solution and not artifacts obtained upon drying of the solution on the TEM grids.Dendritic structures in similar systems are usually associated with diffusion-limited particle aggregation processes.33After 1h of irradiation time we observe numerous,surprising,large,crosslike structures,with segments as long as 1μm (Figure 3c).Nucleation of small spherical gold particles goes on at the same time.Figure 4is a close-up of one of these remarkable structures,the formation of which has not been observed before to our knowledge.The structures contain gold atoms (hence the contrast under the TEM),but they also contain surfactant,as evidenced by the fact

(28)Elding,L.I.;Gro ¨ning,A.-B.Acta Chem.Scand.A 1978,32,867.(29)Carlsson,L.;Lundgren,G.Acta Chem.Scand.1967,21,819.(30)Chateau,H.;Gadet,M.-C.;Pouradier,J.J.Chim.Phys.1966,63,269.

(31)AuCl 4-and AuCl 3OH -in these systems are expected to be bound on DTA +and do not exist free in solution.However,assuming that the equilibrium constant for the reaction in water (2.4×10-6M 2)still holds and neglecting ionic activity coefficients or the fact that the water concentration is not 55.5M,we obtain a hydrolysis degree for AuCl 4-equal to only 1.0%in the presence of 1M DTAC and without excess NaCl.A much higher hydrolysis degree is implied by the pH measure-ments (≈10-2.4/0.022)18%).

(32)(a)Bruche,B.Kolloid-Z .1960,170,97.(b)Milligan,W.O.;Morriss,R.H.J.Am.Chem.Soc.1964,86,3461.

(33)(a)Servan,https://www.360docs.net/doc/fe14688218.html,mun.1998,351.(b)Zhou,Y.;Yu,S.H.;Wang,C.Y.;Li,X.G.;Zhu,Y.R.;Chen,Z.Y.Adv.Mater.1999,11,

850.

Figure 1.UV absorption investigation of ion-pair formation between DTA +and AuCl 4.All solutions contained 1mmol/L AuCl 4-.(a)UV -vis spectra of solutions containing 0.2mmol/L DTAC (dashed line),0.2mmol/L DTAC and 1M NaCl (dashed-and-dotted line),and 80mmol/L (solid line)DTAC.(b)Solution absorbance at 332nm as a function of DTAC concentration for [DTAC]>cmc.A solubilization plateau can be seen above 60mmol/L.

3662Langmuir,Vol.18,No.9,2002Leontidis et al.

that they disintegrate under electron-beam focus (see Supporting Information).These structures are reproduc-ible and always appear in this system,although sometimes at shorter or longer irradiation times,depending on the amount of material in the sample,fluctuations in lamp intensity,etc.For longer irradiation times we observe the formation of very long threadlike gold particles,along with a smaller amount of particles of other morphologies,as was also observed in the absence of NaCl.When NaCl is present,however,the percentage of threadlike particles is much larger,and their length increases dramatically (Figure 3d).The sequence of structures in Figure 3and the dramatic differences in the crystallization evolution with and without NaCl strongly contradict the idea of gold particle templating by rodlike micelles .In fact,we have here the proof that surfactant structures other than micelles are present and may contribute to the reduction mechanism.

III.3.Photochemical Gold Reduction Experiments in CTAC Solutions.Esumi et al.12b,e have

performed

Figure 2.TEM pictures showing the time evolution of gold-particle formation in salt-free solutions containing 35%DTAC and AuCl 4-with r g )1/60:(a)5min;(b)10min;(c)35min;(d)90min.

Gold Colloids from Cationic Surfactant Solutions Langmuir,Vol.18,No.9,20023663

extensive gold photoreduction experiments in CTAC solutions;therefore we will not present our own results here.Esumi et al.attributed the formation of threadlike particles to a templating effect of the large micelles formed in this system above a surfactant concentration of roughly 1.2M,12,34while a sphere-to-rod transition for CTAC is already documented at 7.5wt %.34c In the presence of

high NaCl concentrations,CTAC has been reported to form flexible,wormlike micelles.34b,c However,DTAC does not form large rodlike micelles,even at high external NaCl concentration,34a since it has a much shorter alkyl chain.34

(34)(a)Ozeki,S.;Ikeda,S.Bull.Chem.Soc.Jpn.1981,54,552.(b)Imae,T.;Ikeda,S.Colloid Polym.Sci.1987,265,1090.(c)Lee,Y.S.;Surjadi,D.;Rathman,https://www.360docs.net/doc/fe14688218.html,ngmuir 1996,12,

6202.

Figure 3.TEM pictures showing the time evolution of gold-particle formation in solutions containing 0.65M NaCl,35%DTAC,and AuCl 4-with r g )1/60:(a)15min;(b)30min;(c)1h;(d)8h.

3664Langmuir,Vol.18,No.9,2002Leontidis et al.

In our experiments with CTAC we have found that rodlike gold particles are formed only after considerably longer irradiation times than is the case with DTAC,even in the presence of excess NaCl.This relative difficulty of forming rodlike gold particles in CTAC systems contradicts the micellar templating model .To further assess the micellar templating idea,we have performed dynamic fluorescence measurements to measure how the DTAC and CTAC micelles develop in the presence of NaCl and AuCl 4-and also upon gold particle formation.To our knowledge such extensive measurements have not been performed before in the presence of Au III anions and gold particles.We have preferred the dynamic fluorescence method to more conventional light scattering,because the complexity of the structures observed in the presence of gold particles (Figures 2and 3)would render the light scattering results impossible to interpret.

III.4.Time-Resolved Pyrene Fluorescence Prob-ing.The number of micelles in a given aqueous micellar solution can be counted by time-resolved fluorescence probing.A micelle-bound fluorophore is introduced in the micellar solution,together with a micelle-bound quencher.By analyzing the intramicellar quenching kinetics using standard models,one can deduce the number of micelles,given the number of quencher molecules and provided that a certain statistics of quencher distribution is adopted.A standard probe of the aqueous micellar environment,used for a long time by numerous researchers,is pyrene,a hydrophobic molecule that binds to the dispersed phase in aqueous micellar solutions.36-38In the case of aqueous micelles,pyrene is both the fluorophore and the quencher,

since it can quench monomer fluorescence by excimer https://www.360docs.net/doc/fe14688218.html,ually,pyrene is solubilized at concentrations comparable with the concentration of micelles,and it is accepted that its distribution among micelles follows Poisson statistics.36We have used pyrene decay kinetics to count micelles in solution by analyzing monomer fluorescence decay profiles with the following model:39

A 2)1/τ0,where τ0is the decay time in the absence of quenching;A 3)[P]/[M],where [P]is the pyrene concen-tration and [M]is the micellar concentration;A 4)k q ,which is the pseudo-first-order intramicellar quenching rate constant.The above parameters are valid if we accept that there is no quencher migration among micelles.39Given the pyrene concentration [P],the micellar concen-tration [M]can be directly obtained by fitting the model of eq 1to the decay profile.The micellar aggregation number N ,a measure of the micellar size,can also be calculated by the following formula for a specific surfactant concentration c s

where cmc is the critical micelle concentration and [M]is the micellar concentration.Another important parameter to be deduced from eq 1is the intramicellar quenching rate,k q .Increase or decrease of k q is an index of structural variations in the https://www.360docs.net/doc/fe14688218.html,bination of k q and N can offer another useful parameter that characterizes micellar structure,the so-called microviscosity of the micellar environment,given by

V being the effective micellar volume,V )(n +1)N ,where n is the number of carbon atoms in the surfactant alkyl chain.40This correlation holds true for spherical or spheroidal micelles.40

The advantage of pyrene fluorescence in the systems studied here is that it focuses on the surfactant aggregates,while scattering methods would be strongly affected by the presence of the gold particles and would not yield useful information about micellar shape and size.We have verified that pyrene fluorescence is not affected by the presence of gold salts and particles and that quenching occurs only through excimer formation.We have calculated [M],k q ,N ,and ηi by pyrene time-resolved fluorescence analysis,and the results are tabulated in Tables 1and 2,for DTAC and CTAC,respectively.Inspection of Table 1reveals that the variation of DTAC concentration in the range 0.05-1.0M has a small effect on micellar size,which increases with surfactant concentration.Indeed,N was found to vary from 55to 68.These values are close to those found previously by others.41Addition of NaCl,in the range 0.1-1.0M caused a further increase of N ,which in the case of 1.0M DTAC and 1.0M NaCl was 96.The increase of the micellar size is accompanied by an analogous decrease of the intramicellar quenching rate,which is expected,since reaction efficiency decreases when

(35)(a)Israelachvili,J.N.;Mitchell,D.J.;Ninham,B.W.J.Chem.Soc.,Faraday Trans.21976,72,1525.(b)Gelbart,W.M.,Ben-Shaul,A.,Roux,D.,Eds.Micelles,Membranes,Microemulsions,and Mono-layers ;Springer-Verlag:New York,1994.

(36)Infelta,P.P.;Gratzel,M.J.Chem.Phys.1979,70,179.(37)Lianos,P.;Zana,R.J.Colloid Interface Sci.1981,84,100.

(38)Alargova,R.G.,Kochijashky,I.I.;Sierra,M.L.;Zana,https://www.360docs.net/doc/fe14688218.html,ngmuir 1988,14,5412.

(39)Zana,R.Surfactant Solutions :New Methods for Investigation ;Marcel Dekker:New York,1987.

(40)Anthony,O.;Zana,https://www.360docs.net/doc/fe14688218.html,ngmuir 1996,12,3590.

(41)Wikander,G.;Eriksson,P.-O.;Burnell,E.E.;Lindblom,G.J.Phys.Chem.1990,94,

5964.

Figure 4.Close-up of one of the crosslike particles of Figure 3c.

I (t ))I 0exp {-A 2t -A 3[1-exp(-A 4t )]}

(1)

N )

c s -cmc [M]

(2)

ηi ≈1/Vk q

(3)

Gold Colloids from Cationic Surfactant Solutions Langmuir,Vol.18,No.9,20023665

the effective reaction volume increases.On the contrary, the effective microviscosity of the micellar phase increased only slightly in going from the pure micellar phase to the one containing salt but remained,otherwise,unaffected by the change in the micellar size.These results indicate that the present micellar system does not suffer any important structural variations,either by changing sur-factant concentration or by introducing additional ionic strength.The micelles do increase in size and decrease in number,but they remain in the same structural domain, that of the small spheroidal micelles.In addition,the small micellar-charge neutralization effected by the DTA+AuCl4-pairs and the additional ionic strength induces a further small increase in the micellar size,without changing the micellar structural domain,i.e.,the micelles remain small and spheroidal.Finally,after UV irradiation of the solutions and the formation of gold colloidal particles,we have detected a further decrease in the number of micelles, which gives even higher N values.However,the un-changedηi values suggest that the micellar structural domain did not change even after formation of colloidal gold.Part of the surfactant may participate in the stabilization of the metal particles,but most of it is still expended in the formation of small spheroidal micelles. The case of CTAC micelles is very similar,as seen in Table2.CTAC micelles keep their size practically un-changed in the presence of AuCl4-or reduced gold,even when NaCl is present.Micellar size increases gradually with NaCl concentration,in agreement with the litera-ture.41,42The large wormlike CTAC micelles reported in other works34were obtained at conditions of very high external salt concentrations(2-4M NaCl),used appar-ently because they are relevant to the production of mesoporous zeolites.We have observed by TEM that high NaCl concentrations lead to significant salting-out of the surfactant,while NaCl itself is hard to dissolve in concentrated surfactant solutions(see TEM picture in Supporting Information),and we have opted to avoid such high salt concentrations.

The general conclusion of the fluorescence experiments is that in the range of system compositions that we have worked there exists no well-defined transition to rodlike micelles.The micellar aggregation number shows a slow, gradual increase with surfactant and gold concentration, but the most elongated micelles observed in these systems cannot have aspect ratios larger than2-2.5.43

IV.Final Discussion and Conclusions

The formation of gold particles in solutions containing trialkylammonium surfactants is a very complex process. We have focused on the formation of threadlike gold particles,initially reported in the interesting series of papers by Torigoe and Esumi.12Most of the evidence,which we have obtained from TEM and fluorescence spectros-copy,suggests that the rodlike or threadlike particle morphologies that dominate the distribution at long irradiation times do not originate from a templating mechanism.Time-resolved fluorescence results show that the surfactant micelles remain spheroidal even at35%

(42)(a)Hayter,J.B.;Penfold,J.Colloid Polym.Sci.1983,261,1022.

(b)Magid,L.J.;Han,Z.;Warr,G.G.;Cassidy,M.A.;Butler,P.D.; Hamilton,W.A.J.Phys.Chem.B1997,101,7919.

(43)Assuming spherocylinders with area per surfactant head equal to40?2,one deduces a micellar radius of13?.The aggregation number increases from55(for spheres)to125(for spheroids).From these data one obtains a cylinder length of34?,which yields an overall aspect ratio of2.3.

Table1.Data Obtained by Analysis of Pyrene Decay Profiles in Aqueous DTAC Micelles in the Absence and

Presence of Additives

surfactant concn

(M)NaCl

concn

(M)

AuCl4-

concn

(mM)

micelle

concn

(mM)

micellar

aggregation

no.(N)a

intramicellar

quenching

rate

constant

(107s-1)

micellar

micro-

viscosity

n i

No Additives

0.05000.5455 1.80.08

0.07000.9155 1.70.08

0.4000 6.8056 1.60.09

0.60009.4062 1.50.08

1.000014.468 1.40.08

Addition of NaCl

0.070.100.8658 1.50.09

0.07 1.000.6676 1.10.10

1.00.1014.866 1.20.10

1.00.3014.269 1.10.10

1.00.6011.1880.970.09

1.0 1.0010.2960.840.10

Addition of AuCl4-

0.07 1.00.070.6478 1.10.09

0.07 1.00.200.6083 1.00.10

0.07 1.00.600.5985 1.00.10

1.00.6 6.09.11080.750.09

1.0 1.00.0710.2960.840.09

1.0 1.00.109.71010.840.10

1.0 1.00.209.41040.760.09

1.0 1.00.409.11080.750.10

1.0 1.00.608.81110.750.09

1.0 1.0 6.08.41170.740.10

UV Treatment

0.07 1.00.070.7467 1.200.10

0.07 1.00.200.6676 1.100.09

0.07 1.00.600.63790.970.10

1.00.6 6.07.81250.620.10

1.0 1.00.078.11210.710.09

1.0 1.00.107.91240.650.10

1.0 1.00.207.91240.620.10

1.0 1.00.407.71270.610.10

1.0 1.00.607.41320.580.10

1.0 1.0 6.00.71380.560.10

a Micellar aggregation numbers,N,have been calculated by adopting that the cmc is20×10-3M.

Table2.Data Obtained by Analysis of Pyrene Decay Profiles in Aqueous CTAC Micelles in the Absence and

Presence of Additives

surfactant

concn

(M)

NaCl

concn

(M)

AuCl4-

concn

(mM)

micelle

concn

(mM)

micellar

aggregation

no.(N)a

intramicellar

quenching

rate

constant

(107s-1)

micellar

micro-

viscosity

n i

No Additives

0.07000.82840.610.16

0.4000 4.3930.590.14

0.75007.9950.540.15

Addition of NaCl

0.750.107.8960.570.14

0.750.307.6980.570.14

0.75 1.007.41010.510.15

Addition of AuCl4-

0.75 1.00.077.41010.510.14

0.75 1.00.107.41020.540.14

0.75 1.00.407.41020.500.15

0.75 1.00.607.41020.500.15

UV Treatment

0.75 1.00.077.41020.510.15

0.75 1.00.107.41020.540.15

0.75 1.00.407.41020.500.15

0.75 1.00.607.41020.500.15

a Micellar aggregation numbers,N,have been calculated by adopting that the cmc is1.6×10-3M.

3666Langmuir,Vol.18,No.9,2002Leontidis et al.

surfactant and in the presence of0.6-1M NaCl,in systems that produce a large percentage of threadlike gold particles.Some of the TEM evidence(see,e.g.,Figure2c) strongly suggests that threadlike particles evolve through a one-dimensional aggregation process.In fact,Figure2 suggests that a hierarchy of aggregation phenomena occurs:First,nuclei are aggregated and sintered into compact primary particles,then these particles themselves aggregate to form rodlike structures.Such two-level colloid aggregation processes have recently been observed and studied in the case of gold spheroids,44although aggrega-tion appears to be strongly directional in the present system.The remarkable sequence of structures observed in the presence of NaCl further weakens the micellar templating argument.The experimental fact that gold and surfactant concentration must be above certain thresholds for the threadlike particles to form12is a further argument in favor of the aggregation hypothesis.Why would such an aggregation pattern be favored in these systems?The phenomenon of crystal shape(or“habit”) modification by tailor-made substances is well-known and has been studied for many years by the crystal-growth community.45Recent work by the group of Alivisatos46 suggests that rodlike particle shapes can be obtained when a surfactant present in the system adsorbs specifically on a particular face of a growing crystal and stabilizes it, allowing the crystal to grow in different directions only. Surfactants are thus postulated to act as tailor-made additives.Production of silver nanowires in the presence of additives was also recently reported and assumed to occur because of preferential stabilization of specific crystal faces.47To generalize the previous ideas,one might propose that growth by aggregation occurs also under a similar limitation.Faces on which the surfactant adsorbs strongly would be better stabilized than other faces.The aggrega-tion would proceed by joining particles along faces,which are not properly stabilized.48,49d The resulting aggregates would then show defects,which however can be efficiently

healed,as was demonstrated in fundamental work by the group of El-Sayed.10c The rodlike particles produced by the Wang et al.method8were shown to be dominated by the theoretically unstable{110}faces,their axial growth direction being?112?.10a It may well be that we are looking at a similar growth phenomenon,but detailed high-resolution TEM work would be needed to clarify this issue. There has been much discussion in recent literature about the growth of crystals by primary crystal aggregation.49 It has been shown in several systems that the unique particle morphologies observed by electron microscopy are a result of particle aggregation.48,49In our system, surfactant molecules are used to stabilize an ever-increasing number of gold particles.The total percentage of surfactant expended to stabilize the particles is, however,rather small.We have made a rough estimate of the amount of surfactant adsorbed on gold particles for the35%DTAC and0.022M AuCl4-solution mostly used in this work.Assuming that all gold atoms can be found in particles with25nm radius and that the surfactant (with molecular length15?)forms bilayers on the particles50,51with average cross-sectional area40?2/ molecule,we find that only0.1%of the surfactant molecules adsorbs on the particles.The surfactant micelles do not change their size or shape in a significant way as the surfactant and gold concentrations in the system increase,even at1M NaCl,as indicated by the fluorescence decay results.

The most remarkable results of the present investigation were obtained in the presence of a high NaCl concentration in the surfactant solution.The obvious effect of NaCl is to increase the ionic strength and decrease electrostatic repulsion between micelles and particles,thus enhancing aggregation phenomena.However,this cannot be so important,since the ionic strength due to DTAC is already high.The unexpected evolution of the system through the crosslike structures of Figures3c and4implies that the surfactant may be salted out to a certain extent,because of the high salt concentration,forming a sequence of structures.The development of the gold crystals,through

(44)(a)Adachi,https://www.360docs.net/doc/fe14688218.html,ngmuir2000,16,6460.(b)Adachi,https://www.360docs.net/doc/fe14688218.html,ngmuir 2001,17,3863.

(45)(a)Addadi,L.;Berkovitch-Yellin,Z.;Weissbuch,I.;van Mil,J.; Shimon,L.J.W.;Lahav,M.;Leiserowitz,L.Angew.Chem.,Int.Ed. Engl.1985,24,466.(b)Sugimoto,T.;Itoh,H.;Mochida,T.J.Colloid Interface Sci.1998,205,42.

(46)Puntes,V.F.;Krishnan,K.M.;Alivisatos,A.P.Science2001, 291,215.

(47)Liu,S.,Yue,J.;Gedanken,A.Adv.Mater.2001,13,656.

(48)(a)Penn.R.L.;Banfield,J.F.Geochim.Cosmochim.Acta1999, 63,1549.(b)Banfield,J.F.;Welch,S.A.;Zhang,H.;Ebert,T.T.;Penn, R.L.Science2000,289,751.(c)Penn,R.L.;Oskam,G.;Strathmann, T.J.;Searson,P.C.;Stone,A.T.;Veblen,D.R.J.Phys.Chem.B2001, 105,2177.

(49)(a)OcaZa,M.;Rodriguez-Clemente,R.;Serna,C.J.Adv.Mater. 1995,7,212.(b)Goia,D.V.;Matijevic′,E.New J.Chem.1998,22,1203.

(c)Privman,V.;Goia,D.V.;Park,J.;Matijevic′,E.J.Colloid Interface Sci.1999,213,36.(d)Adair,J.H.;Suvaci,E.Curr.Opin.Colloid Interface Sci.2000,5,160.(e)van Hyning,D.L.;Klemperer,W.G.;Zukoski,C.

https://www.360docs.net/doc/fe14688218.html,ngmuir2001,17,3128.

(50)(a)Patil,V.;Mayya,K.S.;Pradhan,S.D.;Sastry,M.J.Am. Chem.Soc.1997,119,9281.(b)Shen,L.;Laibinis,P.E.;Hatton,T.A. Langmuir1999,15,447.

(51)Nikoobakht,B.;El-Sayed,https://www.360docs.net/doc/fe14688218.html,ngmuir2001,17,

6368. Figure5.Schematic of the processes leading to the production of threadlike gold particles:(a)ion-pair precipitation and subsequent solubilization in micelles;(b)formation of gold particles by UV irradiation,their stabilization by surfactant bilayers,and one-dimensional particle growth by aggregation of gold particles.

Gold Colloids from Cationic Surfactant Solutions Langmuir,Vol.18,No.9,20023667

continuous nucleation,disrupts and eventually terminates the crystal growth of the surfactant.Some of the significant NaCl effect must then be due to the drop in surfactant solubility.NaCl also slows down the photochemical formation of zerovalent gold,since it provides additional excess of Cl-ions,which affects some of the photochemical reactions.23A question mark still remains regarding the potential effect of Cl-concentration on Au III coordination. For the present system this effect should not be very significant,since the dissociation of DTAC already pro-vides a high Cl-concentration.

Our current understanding of the sequence of events during gold-particle production in this system is pre-sented schematically in Figure5.There we depict the initial precipitation of the gold-surfactant ion pair,its solubilization in the micelles,the gradual micellar growth observed by fluorescence,the initial formation of secondary gold particles,and their directional aggregation under the influence of preferential surfactant binding,to form long linear aggregates.The present work demonstrates once more that extensive mechanistic investigations are required to clarify inorganic crystallization phenomena in the complex surfactant systems increasingly used today in materials science applications.

Acknowledgment.We are grateful to Professor Paul Smith and Dr.Walter Caseri(Institut fu¨r Polymere,ETH, Zu¨rich)for their strong interest in this project,for their support in the early stages(initial experiments were performed in their laboratory),and for some preliminary TEM measurements.We are grateful to Ms.Konstantina Kapnisi,who performed exploratory experiments for her Undergraduate Diploma Thesis.We are finally grateful to Dr.Kyriacos Kyriacou and Mr.Andreas Zenios of the Department of Microscopy of the Institute of Neurology and Genetics(Nicosia,Cyprus)for support with the TEM measurements.This work was partly supported by research grants from the University of Cyprus to E.L. and by the Greek-French Cooperative Research Program (Platon),where Cyprus participated as a Mediterranean Partner.

Supporting Information Available:Images of gold particles produced after irradiation in surfactant and both salt and salt-free solutions and melting of crosslike structures by electron beams.This material is available free of charge via the Internet at https://www.360docs.net/doc/fe14688218.html,.

LA011368S

3668Langmuir,Vol.18,No.9,2002Leontidis et al.