1260 DOI: 10.1021/la103718v Langmuir 2011, 27(4), 1260–1268Published on Web 11/04/2010

pubs.acs.org/Langmuir

© 2010 American Chemical Society

Gold Nanoparticle Self-Assembly in Two-Component Lipid

Langmuir Monolayers†

Alina Mogilevsky and Raz Jelinek*

Ilse Katz Institute for Nanotechnology and Department of Chemistry, Ben Gurion University,Beer Sheva 84105, Israel

Received September 16, 2010. Revised Manuscript Received October 17, 2010

Self-assembly processes are considered to be fundamental factors in supramolecular chemistry. Langmuir mono-layers of surfactants or lipids have been shown to constitute effective 2D “templates” for self-assembled nanoparticlesand colloids. Here we show that alkyl-coated gold nanoparticles (Au NPs) adopt distinct configurations whenincorporated within Langmuir monolayers comprising two lipid components at different mole ratios. Thermodynamicand microscopy analyses reveal that the organization of the Au NP aggregates is governed by both lipid components.In particular, we show that the configurations of theNP assemblies were significantly affected by the extent ofmolecularinteractions between the two lipid components within the monolayer and the monolayer phases formed by eachindividual lipid. This study demonstrates that multicomponent Langmuir monolayers significantly modulate the self-assembly properties of embedded Au NPs and that parameters such as the monolayer composition, surface pressure,and temperature significantly affect the 2D nanoparticle organization.

Introduction

Langmuir monolayers have traditionally been employed tostudy the behavior of amphiphilic molecules, such as lipids andfatty acids, on water surfaces.1 In particular, the 2D organizationof monolayers makes such systems amenable to comprehensivephysicochemical investigation using varied surface-characteriza-tion techniques.1 Numerous reports have illuminated the thermo-dynamic profiles, phase transitions, and structural features ofLangmuir monolayers and films of surfactant molecules usingvaried experimental methods.

Langmuir monolayers constitute promising platforms forbottom-up self-assembly processes, particularly those involvingnanoparticles and colloids.2-5 Monolayers have been employedas versatile templates for the in situ growth of crystalline partic-ulate films because their favorable properties allow for theoccurrence of diverse self-assembly processes. Specifically,numerous natural and synthetic surfactants have been shown toform Langmuir monolayers, and their 2D phase behavior andstructures are well understood.6 Hydrophobic colloids also adoptmonolayer organization at the air/water interface.7 In particular,the structure, morphology, orientation, and total film thicknesscan be readily controlled via the selection of the monolayer and

experimental conditions such as temperature, concentration,mole ratio, surface pressure, and subphase composition.8-12

Additionally important from a practical/applied standpoint isthe fact that monolayer films can be conveniently transferred tosolid substrates without disrupting the film structure.

Recent studies have demonstrated that Langmuir monolayersof surfactants and lipids could form “templates” for the deposi-tion of varied types of nanoparticles (NPs).4,13-16 Gold nanopar-ticles (Au NPs) coated with hydrophobic substances allow theNPs to “float” upon the aqueous subphase.17 Recent studies haveshown that Langmuir monolayers of amphiphilic molecules andlipids constrain hydrophobically coated Au NPs into diverse 2Dpatterns.4 We have demonstrated that a careful selection of sur-factant constituents and external parameters, such as temperatureand surface pressure, could produce varied AuNP configurationsat the air/water interface, including nanowires, nanoislets, andnanorods.18,19

Identification and optimization of the factors affecting theorganization of gold (and other types of) NPs at the air/waterinterfaces are critical to attaining a better understanding of their2D assembly properties and for potential utilization of the filmassembly approach for practical purposes. The types and proper-ties of surfactants generally constitute primary factors affectingthe NP film patterns.20-22 The capping agents of the NPs addi-tionally exert significant effects upon NP organization because

† Part of the Supramolecular Chemistry at Interfaces special issue.*Corresponding author. Tel: þ972-8-6461747. Fax: þ972-8-6472943.

E-mail: razj@bgu.ac.il.(1) Zasadzinski, J. A.; Viswanathan, R.; Madsen, L.; Garnaes, J.; Schwartz,

D. K. Science 1994, 263, 1726–1733.(2) Grzelczak, M.; Vermant, J.; Furst, E. M.; Liz-Marzan, L. M. ACS Nano

2010, 4, 3591–3605.(3) Wang, Y.; Xia, Y. Nano Lett. 2004, 4, 2047–2050.(4) Hassenkam, T.; Norgaard, K.; Iversen, L.; Kiely, C. J.; Brust, M.; Bjornholm,

T. Adv. Mater. 2002, 14, 1126–1130.(5) Gracias, D. H.; Tien, J.; Breen, T. L.; Hsu, C.; Whitesides, G. M. Science

2000, 289, 1170–1172.(6) Fendler, J. H.; Meldrum, F. C. Adv. Mater. 1995, 7, 607–632.(7) Wickman, H. H.; Korley, J. N. Nature 1998, 393, 445–447.(8) Finer, E. G.; Phillips, M. C. Chem. Phys. Lipids 1973, 10, 237–252.(9) Moriguchi, I.; Teraoka, Y.; Kagawa, S. Langmuir 1992, 8, 1431–1434.(10) Lo Nostro, P.; Gabrielli, G. Langmuir 1993, 9, 3132–3137.(11) Kawai, T.; Kamio, H.; Kondo, T.; Kon-No, K. J. Phys. Chem. B 2005, 109,

4497–4500.

(12) Petty, M. C. Langmuir Blodgett Films; Cambridge University Press;New York, 1996.

(13) Bai, X.; Ma, H.; Li, X.; Dong, B.; Zheng, L. Langmuir 2010, 26, 14970–14974.(14) Santhanam, V.; Liu, J.; Agarwal, R.; Andres, R. P. Langmuir 2003, 19,

7881–7887.(15) Wang, Z.; Li, X.; Yang, S. Langmuir 2009, 25, 12968–12973.(16) Harishchandra, R. K.; Saleem, M.; Galla, H. J. J. R. Soc. Interface 2010, 7,

S15–S26.(17) Lin, B.; Schultz, D. G.; Lin, X.; Li, D.; Gebhardt, J.; Meron, M.; Viccaro,

P. J. Thin Solid Films 2007, 10, 2615–2619.(18) Markovich, N.; Volinsky, R.; Jelinek, R. J. Am. Chem. Soc. 2009, 131,

2430–2431.(19) Volinsky, R.; Jelinek, R. Angew. Chem., Int. Ed. 2009, 48, 4540–4542.(20) Hansen, C. R.; Westerlund, F.; Moth-Poulsen, K.; Ravindranath, R.;

Valiyaveettil, S.; Bjornholm, T. Langmuir 2008, 24, 3905–3910.(21) Khomutov, G. B. Adv. Colloid Interface Sci. 2004, 111, 79–116.

DOI: 10.1021/la103718v 1261Langmuir 2011, 27(4), 1260–1268

Mogilevsky and Jelinek Article

these molecules modulate the interactions both between the NPsand the surrounding surfactants and among the NPs themselvesin the monolayer environment.13,23,24

Multicomponent Langmuir monolayers offer intriguing possi-bilities for modulating nanoparticle self-assembly because theinteractions between the monolayer constituents add anotherparameter that could significantly modulate a film’s physicalproperties.25-29 In particular, the distinct sensitivity ofmonolayerphases to molecular composition30,31 might affect self-assemblyphenomena involving embedded nanoparticles. Here we presentan investigation of Au NP organization within two-componentlipid Langmuir monolayers. Our analysis reveals the distinctdependence of Au NP self-assembly upon the interactions be-tween the two lipid components. The experimental results point tothe intimate relationship between nanoparticle film configura-tions at the air/water interface and compositions of multicompo-nent lipid monolayers.

Materials and Methods

Materials. HAuCl4, tetradecylammonium bromide, sodiumborohydride, and dodecanethiol were purchased from Sigma-Aldrich and used as provided. Lipids including 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), 1,2-diarachidoyl-sn-glyc-ero-3-phosphocholine (DAPC), and C18 ceramide (brain) wereprovided in powder form from Avanti Polar Lipids (Alabaster,AL) and used as provided. Chloroform (CHCl3) and methanol(CH3OH) were HPLC grade (Frutarom Ltd., Haifa, Israel).

Synthesis of Dodecanethiol-Capped Au NPs. Synthesis ofthe dodecanthiol-capped Au NPs was carried out as previouslydescribed.32 Briefly,we used a slightlymodified Brust33 procedurebased on a two-phase (toluene/water) reduction ofHAuCl4 in thepresence of a stabilizing ligand.HAuCl4 (100mg)was dissolved inwater (10 mL) and transferred to toluene (40 mL) by mixing withtetradecylammonium bromide (500 mg). Dodecanethiol (0.0147mL) was added to the vigorously stirred solution. Finally,NaBH4

(200 mg) in water (1 mL) was added. After 10 min, the solutionwas dark brown, indicating the formation of Au NPs. Thesolution was stirred overnight to ensure the completion of thereaction. The organic phase was washed with 2 M H2SO4 andwater in a separation funnel and subsequently evaporated invacuum to near dryness. The resultant dodecanethiol-cappedAu NPs were precipitated by the addition of ethanol (60 mL).After separation by filtration, the precipitate was washed threetimeswith ethanol andoncewith 2-propanol anddried invacuo toyield the Au NPs (34.9 mg) as a black wax, which was easilydissolved in chloroform (final concentration 9.00 mg/mL). TheAuNPs were characterized by TEM, and the average diameter ofthe particles was approximately 3 nm.

Surface-Pressure/Area Isotherms. Monolayers were as-sembled and compressed, at 18 and 35 �C, through the followingprocedure: parent solutions of the lipids were prepared by dis-solution in chloroform to a concentration of 2 mM (for each lipid

component). For each isotherm experiment, lipid/Au NP solu-tions were prepared in chloroform by mixing 20 μL of Au NPs(9mg/mL) and 80μLof total lipids (ceramide andDMPC/DAPCat the desired mole ratios) with the mixture subsequently spreadon the water subphase. After an equilibration time of 15 min toallow for solvent evaporation prior to compression, the mono-layers were compressed at a rate of 8 cm2/min. Surface-pressure/area isotherms were recorded using a computerized Langmuirtrough (model 622/D1, Nima Technology Ltd., Coventry, U.K.).The water subphase used in the Langmuir trough was doublypurified with a Barnstead D7382 water purification system(Barnstead Thermolyne Corporation, Dubuque, IA), yielding18.3 mΩ resistivity. The surface pressure was monitored using1-cm-wide filter paper as a Wilhelmy plate. Each isothermrepresents three experimental runs, which were reproducible towithin an error of 1.0 A2/molecule.

Transmission Electron Microscopy (TEM). Samples formicroscopyanalyseswere extractedata surfacepressureof 10mN/musing horizontal transfer onto 400 mesh copper Formvar/carbongrids (Electron Microscopy Sciences, Hatfield, PA). TEM imageswere recordedona Jeol JEM-1230 transmission electronmicroscope(JEOL LTD, Tokyo, Japan) operating at 120 kV.

Results

Au NP Organization within Single-Lipid Monolayers.

This study examines the assembly properties of spherical AuNPs within two-component lipid monolayers specifically com-prising ceramide/dimyristoylphosphatidylcholine (DMPC) andceramide/diarachidoylphosphatidylcholine (DAPC). Such mixedmonolayers exhibit distinct thermodynamic properties that aredependent upon the lipid molecules and the interactions betweenthem. Figure 1 depicts the surface-pressure/area isotherms of thethree lipids studied.DMPCexhibits a liquid-expandedmonolayerphase over almost the entire pressure range at the temperaturesexamined (solid isotherms, Figure 1).34 DAPC, however, formsa liquid-condensed monolayer instantly upon deposition on thewater subphase (short-dashed-line isotherms, Figure 1).35

(Liquid-condensed/liquid-condensed phase transitions are ob-served at around 20mN/m (Figure 1).) Ceramide, the N-acylatedderivative of sphingosine, also adopts a condensed monolayerphase upon deposition at the air/water interface at both tempera-tures examined (long-dashed-line isotherms, Figure 1).36

To examine the organization of the Au NPs in binary lipidmonolayers and the impact of mixing the two lipids, we firstexamined the gold assemblies within single-component lipidmonolayers (Figure 2). Figure 2 depicts transmission electronmicroscopy (TEM) images of Au NP/lipid films compressed to asurface pressure of 10 mN/m at subphase temperatures of 18 and35 �C, respectively. The TEM images in Figure 2 demonstratesignificant differences in Au NP organization within monolayersof the three lipids. In addition, a temperature effect uponNP self-assembly is apparent.

Specifically, Figure 2A shows that the Au NPs formed irregu-larly shaped thick stripes in DMPC monolayers.32 The thickstripes,which comprisedpartially fused individualAuNPs (therebylosing their spherical shape4), were similar at both temperaturesexamined. The formation of randomly shaped Au NP aggregatesin DMPC monolayers has been ascribed to the fluidity of themonolayer.34 Indeed, Figure 2B shows that Au NPs embeddedwithin DAPC monolayers adopted more ordered configurations,

(22) Norgaard, K.; Bjornholm, T. Chem. Commun. (Cambridge, U.K.) 2005,1812–1823.(23) Giersig, M.; Mulvaney, P. Langmuir 1993, 9, 3408–3413.(24) Fu, Y. Z.; Li, J. R.; Du, Y. K.; Jiang, L. Surf. Sci. 2006, 835–840.(25) Maggio, B.; Lucy, J. A. Biochem. J. 1976, 155, 353–364.(26) Yang, Z.; Engquist, I.; Wirde, M.; Kauffmann, J. M.; Gelius, U.;

Bo Liedberg, B. Langmuir 1997, 13, 3210–3218.(27) Zhang, L.; Lu, T.;W.Gokel, G.W.; Kaifer, A. E.Langmuir 1993, 9, 786–791.(28) Ulman, A. Chem. Rev. 1996, 96, 1533–1554.(29) Birdi, K. S. Self-Assembly Monolayer Structures of Lipids and Macromole-

cules at Interfaces; Kluwer Academics/Plenum Publishers: New York, 1999.(30) Stranick, S. J.; Parikh, A. N.; Tao, Y. T.; Allara, D. L.; Weiss, P. S. J. Phys.

Chem. 1994, 98, 7636–7646.(31) Scheffer, L.; Solomonov, I.; Weygand, M. J.; Kjaer, K.; Leiserowitz, L.;

Addadi, L. Biophys. J. 2005, 88, 3381–3391.(32) Mogilevsky, A.; Volinsky, R.; Dayagi, Y.; Markovich, N.; Jelinek, R.

Langmuir 2010, 26, 7893–7898.(33) Brust, M.; et al. J. Chem. Soc., Chem. Commun. 1994, 801–802.

(34) Gaboriaud, F.; Golan, R.; Volinsky, R.; Berman, A.; Jelinek, R. Langmuir2001, 17, 3651–3657.

(35) Dynarowicz-Latka, P.; Rosilio, V.; Boullanger, P.; Fontaine, P.; Goldmann,M.; Baszkin, A. Langmuir 2005, 21, 11941–11948.

(36) Vaknin, D.; Kelley, M. S. Biophys. J. 2000, 79, 2616–2623.

1262 DOI: 10.1021/la103718v Langmuir 2011, 27(4), 1260–1268

Article Mogilevsky and Jelinek

most likely because of the liquid-condensed phase of the DAPCmonolayer at both 18 and 35 �C.32 In particular, Figure 2B featuresthin columns of individual Au NPs assembled around condensedDAPC domains at 18 �Cwhereas somewhat thicker Au assembliesaccumulated around the circularly shaped condensed domainsat 35 �C (Figure 2B). Different nanoparticle configurations wereobserved in Au NP/ceramide monolayers (Figure 2C). Thesemonolayers yielded isolated ring-shaped Au NP islets at 18 �Cwhereasmerging of the isolated gold assemblies into long branchedstripes was apparent at 35 �C (Figure 2C).AuNPs/Ceramide/DMPCMonolayers.Whereas Figure 2

underscores the close relationship between the Au NP self-assembled structures and lipid environments at the air/waterinterface, we further aimed to assess whether and to which extentthe modulation of monolayer properties in two-component lipidmixtures affects the organization of embedded nanoparticles.Figures 3-6 present TEM results obtained from mixed mono-layers comprisingAuNPs, ceramide, andDMPC (Figures 3 and 4)or Au NPs, ceramide, and DAPC (Figures 5 and 6). Figures 3-6also display graphs correlating monolayer compositions with theaverage molecular areas of the two-component monolayers, cal-culated from the respective isotherms at 10 mN/m.37 These graphswere designed to assess the deviations of the two-lipid monolayersfrom ideal mixtures, thus evaluating the interactions between thelipid molecules comprising the monolayers and the impact of suchinteractions upon the Au NP assemblies.38

Figure 3A depicts a graph correlating the average molecularareas of Au NP/ceramide/DMPC/monolayers as a function ofceramide mole fraction, extracted from the isotherms acquired in18 �C at a surface pressure of 10 mN/m. The straight broken linein Figure 3A corresponds to values calculated according to theadditivity rule, which essentially predicts physical parameters inideal mixtures.38 According to the additivity rule, in a monolayercomprising an ideal mixture of noninteracting components(ceramide and phospholipids in this study) the mean moleculararea (Amean) at a given surface pressure will be essentially equalthe weighted average

Amean ¼ XceramideAceramide þð1-XceramideÞAphospholipid

in which Xceramide and Aceramide correspond to the mole fractionand molecular area of ceramide, respectively. (1 - Xceramide) andAphospholipid similarly correspond to the phospholipid mole frac-tion and molecular area, respectively, in a pure phospholipidmonolayer. An important point that needs to be emphasized isthat the straight line in Figure 3A (and other additivity ruleanalysis graphs in this article) was calculated for monolayers thatcomprised AuNPs in addition to lipids. The concentration of AuNPs in allmonolayers containing different ceramide/phospholipidratios, however, was identical, and the only variable in the experi-ments was the ratio between the lipid components. Indeed, acomparison between the additivity rule results with and withoutAuNPs (Supporting Information) indicates that the AuNPs hada minimal effect upon the interactions between the two lipidcomponents.

Figure 3A demonstrates that in all ceramide mole ratios above0.2 the average molecular areas were lower than the valuescalculated according to the additivity rule. This result indicatesa predominance of attractive forces between ceramide andDMPC at 18 �C, leading to the reduction of the average molec-ular areas as shown in Figure 3A.37,38 Importantly, the graph inFigure 3A was qualitatively similar to the results obtained for aceramide/DMPC monolayer without Au NPs (Supporting In-formation), suggesting that the incorporation of Au NPs in themonolayers did not alter the physical properties of the mono-layers and the interactions between the lipid components.

The TEM data in Figure 3B,C attest to the changes in Au NPorganization induced by the two distinctmonolayer compositionsexamined. Figure 3B shows the formation of elongated Au NP“necklaces” within a ceramide/DMPC (3:7mol ratio) monolayer.The Au NP features in Figure 3B appear to emanate frommerging, or superimposition, of the long Au NP stripes recordedin the monolayer of pure DMPC (Figure 2A) and the ring-shapeAu NP assemblies observed in the monolayer of pure ceramide(Figure 2C). This superimposition is consistent with the ideal-mixture formation of the monolayer at a ceramide mole ratio of0.3 (left arrow, Figure 3A).

In contrast to the organized Au NPs in the ceramide/DMPCmonolayer comprising a 3:7 mole ratio (Figure 3B), no orderingof the Au NPs occurred when the ratio between ceramide andDMPCwas inverted (ceramide/DMPC 7:3mol ratio, Figure 3C).The TEM image acquired at this composition showed long, thickaggregates of Au NPs and no evidence of ring-shape film

Figure 1. Surface pressure/area isotherms of the lipids examined. The isotherms were recorded at the two temperatures indicated.(-) DMPC, (---) DAPC, and (---) ceramide.

(37) Maget-Dana, R. Biochim. Biophys. Acta 1999, 1462, 109–140.(38) George, L.; Gaines, J. Insoluble Monolayers at Liquid-Gas Interfaces.

Wiley-Interscience: New York, 1966.

DOI: 10.1021/la103718v 1263Langmuir 2011, 27(4), 1260–1268

Mogilevsky and Jelinek Article

structures that were abundant in the ceramide/DMPC (3:7 ratio)film (Figure 3B). Indeed, an examination of the averagemoleculararea graph in Figure 3A reveals that at a 7:3 ceramide/DMPCratio the mixed monolayer exhibited significant deviation froman ideal mixture, specifically, a lower molecular area resultingfrom the apparent attraction between the ceramide and DMPC

molecules (right arrow, Figure 3A). The significant interactionsbetween ceramide andDMPCcould explain the reorganizationofthe Au NPs in the monolayer, which leads to a distinct con-figuration that was significantly different than the monolayers ofthe pure lipids (Figure 2) or the Au NP/ceramide/DMPC mono-layer at the 3:7 ceramide/DMPC ratio (Figure 3B).

Figure 2. Au NP organization in monolayers of the pure lipids. TEM images of (A) Au-NP/DMPC, (B) Au-NP/DAPC, and (C) Au NP/ceramide films. All films were transferred from the air/water interface at a surface pressure of 10 mN/m. The scale bars in the imagescorrespond to 100 nm. (The scale bars in the two insets at (18 �C) corresponds to 25 nm.)

1264 DOI: 10.1021/la103718v Langmuir 2011, 27(4), 1260–1268

Article Mogilevsky and Jelinek

Figure 4 features the thermodynamic and TEM analysis of AuNPs incorporated within ceramide/DMPC monolayers at 35 �C.The graph depicting the averagemolecular area versusmonolayercomposition in Figure 4A indicates different lipid interactionscompared to the data acquired at 18 �C (Figure 3A). In particular,slight repulsion between the two lipid components appeared todominate in monolayers comprising a small percentage of cer-amide (left arrow in Figure 4A). Importantly, attraction forceswere clearly absent in the high-mole-ratio ceramide samples(Figure 4A), in contrast to the scenario of the mixed monolayerat 18 �C (Figure 3A).

The TEM results in Figure 4B,C show representative nano-particle assemblies within Au NPs/ceramide/DMPC monolayersat 35 �C. Specifically, the images reveal necklace assemblies ofthe AuNPs at both a 3:7 mol ratio between ceramide and DMPC(Figure 4B) and at a 7:3 ceramide/DMPC ratio (Figure 4C).However, Au NPs in the 3:7 ceramide/DMPC monolayer fea-tured a narrower necklace configuration (Figure 4B) compared tothe corresponding arrangement within the 7:3 ceramide/DMPCmonolayer (Figure 4C). Although the necklace structures can beascribed to the superimposed strip structure of AuNPs in DMPC

at 35 �C (Figure 2A) and Au NPs in ceramide (Figure 2C), themuch broader ring configurations in Figure 4C cannot be easilyexplained according to the additivity rule results. We speculatethat the fluidity of DMPC allows the aggregation of Au NPs inlarger monolayer areas, resulting in the thicker ring structuresin monolayers of the 7:3 ceramide/DMPC ratio. Indeed, it waspreviously reported that ceramide displaying long, unsaturatedN-acyl chains was very miscible within DMPC monolayers.39

Au NP/Ceramide/DAPC Monolayers. A similar interplaybetween Au NP configurations and interactions among the lipidmolecules within the binary monolayers was apparent in Au NP/ceramide/DAPC monolayers (Figures 5 and 6). Figure 5 presentsthermodynamic and TEM analyses of Au NP/ceramide/DAPCmonolayers at 18 �C. The graph in Figure 5A depicting the relation-ship between the mean molecular areas and ceramide mole fractiondemonstrates that in the entire composition range themolecular areavalueswere localized in close proximity to the straight line calculatedaccording to the additivity rule (e.g., ideal mixtures). This result

Figure 3. Au NP organization in mixed ceramide/DMPC monolayers at 18 �C. (A) Graph showing the mean molecular areas recorded at10 mN/m for different monolayer compositions. The broken line corresponds to values calculated according to the additivity rule. (See thetext.) The two arrows indicate the compositions at which the TEM images were obtained. (B) TEM images of a film comprising AuNPs andceramide/DMPC in a 3:7mole ratio. (C) TEM image of a film comprisingAuNPs and ceramide/DMPC in a 7:3mole ratio. The scale bars inthe images correspond to 100 nm. (The scale bar in the inset corresponds to 50 nm.)

(39) Holopainen, J. M.; Brockman, H. L.; Brown, R. E.; Kinnunen, P. K.Biophys. J. 2001, 80, 765–775.

DOI: 10.1021/la103718v 1265Langmuir 2011, 27(4), 1260–1268

Mogilevsky and Jelinek Article

indicates that ceramide and DAPC exhibit minimal interactionswithin the mixed monolayers.

TheTEMdata inFigure 5B,Chighlight theAuNPdistributionwithin the ceramide/DAPC monolayers at 18 �C at two differentmole ratios. In a monolayer comprising ceramide/DAPC in a3:7 mole ratio (Figure 5B), the Au NPs produced thick ringsaround circular spaces, most likely corresponding to condensedlipid domains. Similar to the analysis of the ceramide/DMPCmonolayers above, this configuration probably arises from asuperimposition of theAuNP columns around the abundant con-densed lipid domains in monolayers of pure DAPC (Figure 2B)and the ring-shapedAuNPaggregates observed inmonolayers ofpure ceramide (Figure 2C). This description is consistent with theapparent ideal mixture behavior of the ceramide/DAPC mono-layer inferred fromFigure 5A (left arrow). The apparent retentionof the spherical shape of the individual Au NPs (inset inFigure 5B) is particularly noteworthy. This result stands in sharpcontrast to the pronounced fusion of Au NPs into irregularlyshaped particles in the ceramide/DMPC monolayer (insets inFigure 3B) and is evidence of the influence of the condensedDAPC monolayer phase.32

The TEM image in Figure 5C of a Au NP/ceramide/DAPCmonolayer (ceramide/DAPC 7:3 mol ratio) depicts long stripsof Au NP aggregates, within which gold-free circular domainsappear. This AuNP configuration seems different from the struc-tures formed in the monolayers of the single-lipid components(Figure 2) and cannot be easily explained according to the meanmolecular area result in Figure 5A (which pointed to an almostideal mixture at this mole ratio, right arrow). However, the longstrips in Figure 5C could be affected by the DAPC componentwithin the mixed monolayer because Au NPs in pure DAPCmonolayers indeed formed interconnected networks (Figure 2B).Such gold stripes could be possibly modulated further by thebroader ring-shapedAuNPaggregates observed inpure ceramidefilms (Figure 2C). The higher abundance of ceramide in thesample examined in Figure 5C consequently leads to a spreadingout of the gold-free domain structure, which is depicted in theTEM image.

Figure 6 presents the thermodynamic and TEManalyses of AuNPs/ceramide/DAPCmonolayers examined at 35 �C. In contrastto the seemingly ideal mixture of the two lipid components at18 �C (mean molecular area graph in Figure 5A), Figure 6A

Figure 4. Au NP organization in mixed ceramide/DMPC monolayers at 35 �C. (A) Graph showing the mean molecular areas recorded at10 mN/m for different monolayer compositions. The broken line corresponds to values calculated according to the additivity rule. The twoarrows indicate the compositions at which the TEM images were obtained. (B) TEM images of a film comprising Au NPs and ceramide/DMPC in a 3:7mole ratio. (C) TEM image of a film comprisingAuNPs and ceramide/DMPC in a 7:3mole ratio. The scale bars in the imagescorrespond to 100 nm. (The scale bar in the inset corresponds to 50 nm.)

1266 DOI: 10.1021/la103718v Langmuir 2011, 27(4), 1260–1268

Article Mogilevsky and Jelinek

indicates the existence of significant attractive forces betweenceramide and DAPC at 35 �C, giving rise to the lower averagemolecular areas recorded in monolayers comprising between 20and 70% ceramide.

The TEM results in Figure 6B,C reflect the impact of the lipidinteractions upon the Au NP organization within the two-component monolayers at 35 �C. At a ceramide/DAPC ratio of3:7 (Figure 6B), the TEM image reveals elongated, irregularlyshaped strips of Au NPs. Importantly, no encircled gold-freedomains were observed in the TEM image in Figure 6B, in con-trast to the AuNP configurations within pureDAPC (Figure 2B)or pure ceramide (Figure 2C) monolayers and also different fromthe Au NPs/ceramide/DAPC monolayer at 18 �C (Figure 5B).The distinct, seemingly disordered Au NP assembly shown inFigure 6B ismost likely due to the significant interactions betweenceramide and DAPC (left arrow in Figure 6A), leading to amonolayer exhibiting different physical properties compared toan ideal mixture within Langmuir monolayers of the two lipids.

Similar to the TEM result in Figure 6B, the image in Figure 6Ccorresponding to Au NPs assembled in ceramide/DAPC mono-layers in a 7:3 mole ratio shows elongated Au NP aggregates that

do not exhibit specific shapes. However, unlike the ceramide/DAPC monolayer in the 3:7 mol ratio (Figure 6B), Figure 6Cshows that some encircled domains did appear within the mono-layer. This observation might be related to the fact that thedeviation from an ideal mixture of the Au NP/ceramide/DAPCmonolayer at the ceramide/DAPC 7:3 mol ratio (right arrow inFigure 6A) was less pronounced than for the correspondingmonolayer in a 3:7 ratio (Figure 6A). Accordingly, the Au NPaggregates in the Au NP/ceramide/DAPC monolayer (7:3 ratiobetween ceramide and DAPC) retain the gold-free domainstructure, as recorded in the TEM experiment in Figure 6C.

Discussion

Self-assembly phenomena involving hydrophobically coatednanoparticles at the air/water interface have attracted interest aspossible platforms for bottom-up structural techniques. Variedstudies have shown that surfactant and lipid monolayers consti-tute effective templates for the 2D organization of Au NPs.4,32

In particular, it has been shown that AuNPs preferentially aggre-gate at interfaces between different monolayer phases, partic-ularly the interface between liquid-expanded and liquid-condensed

Figure 5. Au NP organization in mixed ceramide/DAPC monolayers at 18 �C. (A) Graph showing the mean molecular areas recorded at10 mN/m for different monolayer compositions. The broken line corresponds to values calculated according to the additivity rule. The twoarrows indicate the compositions in which the TEM images were obtained. (B) TEM images of a film comprising Au NPs and ceramide/DAPC in a 3:7mole ratio. (C) TEM image of a film comprisingAuNPs and ceramide/DAPC in a 7:3mole ratio. The scale bars in the imagescorrespond to 100 nm. (The scale bar in the inset corresponds to 50 nm.)

DOI: 10.1021/la103718v 1267Langmuir 2011, 27(4), 1260–1268

Mogilevsky and Jelinek Article

surfactant phases. Here we show that the nanoparticle templatingapproach can be expanded through the incorporation of hydro-phobically coated Au NPs in multicomponent lipid monolayers.In particular, we demonstrate that the extent of mixing andconcomitant molecular interactions in binary lipid monolayersexert significant effects upon the aggregation and configurationsof nanoparticle assemblies.

The thermodynamic and TEM experiments carried out clearlyshow that Au NP organization within two-component lipidmonolayers is dependent upon the mole ratios between the lipidsin themixedmonolayers. Specifically, the additivity rule analyses,in conjunction with TEM images extracted at specific ratiosbetween the two lipid components, indicate that the deviationsfrom ideal mixtures observed in specific ceramide/phospholipidsratios were a primary factor affecting the Au NP configurations.In particular, the TEM results demonstrated that, in the majorityof monolayers examined, when the monolayers appeared tocontain noninteracting lipid components, the Au NP assembliescould be characterized as weight-averaged superimpositions be-tween the respective structures formed within the monolayersof individual lipids. This interpretation explains, for example, theremarkable “necklace” structures obtained in the Au NPs/ceramide/DMPC monolayers at 18 �C (Figure 3) and 35 �C

(Figure 4)most likely corresponding to superimpositions betweenthe ring-shaped Au NP aggregates in monolayers of pure cer-amide (Figure 2C) and the elongated Au NP “stripes” assembledin DMPC monolayers (Figure 2A).

Beside the degree ofmixing andmolecular interactions betweenthe lipid constituents forming the films, the data indicate that thephases of the specific lipids in the Langmuir monolayer environ-ment exert substantial effects upon film configurations. Indeed,the degree in which one or both lipids display a liquid-expandedphase in the pure monolayer appeared to be an importantparameter for promoting superimposition between the Au NPconfigurations recorded in monolayers of the individual lipids.Thus, superimposed Au NP assemblies appeared more abundantand visually distinctive in monolayers comprising ceramide andDMPC (Figures 3 and 4) ascribed to the liquid-expanded mono-layer phase of DMPC. In comparison, in the binary monolayerscontaining ceramide and DAPC, which both exhibit liquid-condensed phases, the Au NP structures were less ordered andexhibited lesser uniformity. These results suggest that the con-tributions of fluid lipid components within binarymonolayers aidthe reorganization of the nanoparticles into the distinct config-urations shaped by the second lipid component that exhibits acondensed phase.

Figure 6. Au NP organization in mixed ceramide/DAPC monolayers at 35 �C. (A) Graph showing the mean molecular areas recorded at10 mN/m for different monolayer compositions. The broken line corresponds to values calculated according to the additivity rule. The twoarrows indicate the compositions at which the TEM images were obtained. (B) TEM images of a film comprising Au NPs and ceramide/DAPC in a 3:7mole ratio. (C) TEM image of a film comprisingAuNPs and ceramide/DAPC in a 7:3mole ratio. The scale bars in the imagescorrespond to 100 nm. (The scale bar in the inset corresponds to 50 nm.)

1268 DOI: 10.1021/la103718v Langmuir 2011, 27(4), 1260–1268

Article Mogilevsky and Jelinek

Although the experimental data demonstrate the distinctvariations of Au NPs organization within two-component lipidmonolayers, this study also underscores the fact that Au NPsessentially constitute a sensitive measure of the extent and impactof intermolecular forces in mixed monolayers. Essentially, theTEM analysis reveals that the configurations of the Au NPaggregates, in particular the extent of superimposition betweenthe single-component lipid monolayers, compared to the Au NPstructures observed in the binary mixtures, were dependent upondeviations of the films from ideal mixtures. This observationcould open the way to using hydrophobically coated Au NPs(or other monolayer-forming colloidal systems) as a vehicle forinvestigating multicomponent, complex monolayers.

Conclusions

This study presents an intriguing new approach to modulatingthe organization of supramolecular Au NP assemblies at the air/water interface, namely, through exploiting bimolecular interac-tions between different lipid molecules in two-component mono-layers. The experimental data presented in this work highlight the

dramatic effects of the molecular interactions between the lipidmolecules upon the Au NP organization. In particular, thedistinct gold configurations were dependent upon the mixing ofthe two lipid molecules within the monolayers, and the degreesof their thermodynamic properties were affected through themixing. Overall, this work constitutes an interesting contributionto the diverse “toolbox” for constructing colloidal patterns at theair/water interface.

Acknowledgment.Weare grateful toR. Jeger for helpwith theTEM experiments.

Supporting Information Available: Isotherms of the AuNP/individual lipids, additivity rule graphs of the mixed-lipid monolayers without Au NPs, Brewster angle micros-copy (BAM) images of representative Au NP/lipids mono-layers, and atomic force microscopy (AFM) images ofrepresentative Au NP/lipids films transferred to solid sub-strates. This material is available free of charge via theInternet at http://pubs.acs.org.