Article pubs.acs.org/Langmuir

Model for the Phase Transfer of Nanoparticles Using Ionic Surfactants Chakra P. Joshi and Terry P. Bigioni* Department of Chemistry, The University of Toledo, Toledo, Ohio 43606, United States S Supporting Information *

ABSTRACT: Ionic surfactants are widely used for the phase transfer of nanoparticles from aqueous to organic phases; however, a model that can be used to select ionic surfactants based on the nanoparticle solution properties has yet to be established. Here, we have studied the phase transfer of a variety of nanoparticles and have identified hydrophobicity, steric repulsion, and interfacial tension as key factors in determining whether or not phase transfer will occur. Based on these studies, we have developed a simple model for phase transfer wherein the success of the surfactant depends only on three criteria. The phase transfer agents must (i) efficiently load onto or cross the interface, (ii) solubilize the nanoparticles in the receiving phase, and (iii) sterically stabilize the nanoparticles to prevent aggregation due to van der Waals forces between the inorganic cores. Using these criteria, the effectiveness of ionic surfactants could be predicted based on their molecular geometry and the properties of the nanoparticle solutions. These rules provide a basis for choosing surfactants for phase transfer of spherical nanoparticles up to 16 nm in diameter and advances the development of a general model of nanoparticle phase transfer, which would include all nanoparticle shapes, sizes, and solvents.

1. INTRODUCTION Organic molecules are widely used to impart solubility to nanoparticles, generally as ligands covalently bound to the surfaces of nanoparticles (NPs).1 It is possible to change the NP solubility from polar to nonpolar by exchanging ligands;2,3 however, this poses a technical challenge due to the incompatibility of the new ligands with the original solvent. Ionic surfactants are particularly useful for satisfying the solubility criteria in either solvent, due to their amphiphilic nature, allowing phase transfer into the new solvent by temporarily modifying the surface of the NPs.2−5 Once the NPs are transferred into the new phase, the ionic surfactants may be easily displaced in favor of a covalently bound ligand shell with the appropriate functionality for the new solvent.3,6,7 While this is a good strategy, guidance for choosing surfactants is quite limited. In fact, despite nearly two decades of employment,2,3 no theoretical model of this strategy is available for making rational surfactant choices for phase transfer based on NP and solvent properties. Despite the lack of guidance, many examples of phase transfer exist in the literature.2,3,5,8−10 For example, ionic surfactants were used for the phase transfer of sodium hexametaphosphate stabilized aqueous CdS quantum dots (∼5 nm) in 1996 by Tian et al.,2 using dioctadecyldimethylammonium bromide (DODAB). Cheng et al. demonstrated the complete phase transfer of 2−5 nm aqueous citrate Au NPs using tetraoctylammonium bromide (TOAB).5 Recently, Muhammed and Pradeep demonstrated phase transfer of aqueous Au25SG18 clusters into toluene using TOAB.8 In this © 2014 American Chemical Society

case, they argued that the electrostatic interactions between the glutathione ligands and the ionic surfactants were required to impart the hydrophobicity required for successful phase transfer. While phase transfer using ionic surfactants has enjoyed some success, there appear to be limits to the sizes of NPs that can be transferred.5 This presents a technical barrier to progress, especially since it is not always possible to synthesize NPs with a desired size and shape in any solvent. For example, with few exceptions,11 large monodispersed Au NPs that are >10 nm in diameter have been difficult to synthesize in organic solvents,12,13 while they are routinely made in water with a wide variety of sizes and shapes.14−16 Although phase transfer with ionic surfactants appears to be a viable solution to this problem, success has been limited to the phase transfer of NPs ≤ 5 nm.4,5 Several unsuccessful attempts have been made to surpass the current 5 nm limit. For example, Cheng et al.5 successfully phase transferred NPs of size 5 nm; however, they were unsuccessful for larger Au NPs. This failure was attributed to insufficient hydrophobicity of TOAB for the solubilization of larger Au NPs. It is unclear how to overcome these limits due to the lack of a fundamental understanding of the phase transfer process. The partial phase transfer of NPs > 5 nm in diameter was recently demonstrated by chemically modifying the NP surface Received: September 8, 2014 Revised: October 20, 2014 Published: October 27, 2014 13837

dx.doi.org/10.1021/la503574s | Langmuir 2014, 30, 13837−13843

Langmuir

Article

Figure 1. Phase transfer from water (lower phase) into toluene (upper phase) using both TOAB and C16TAB surfactants, with (a) 13.1 nm citratecapped Au NPs, (b) 13.5 nm citrate-capped Ag NPs, (c) Na4Ag44(p-MBA)30 clusters, and (d) 5 nm citrate-capped CdSe quantum dots. (e) TOAB and C18TAB surfactants were used to phase transfer 16.1 nm citrate-capped Au NPs.

with a thiol-containing ligand;9 however, Zhu et al. modified the surface of 20 nm citrate-protected Au NPs with thioglycolic acid before the dilute solution of NPs was phase transferred using CTAB.9 They showed that the success of the phase transfer depended on the chain length of the ionic surfactant, with longer chains being more successful due to their greater hydrophobicity.9 Zhao and Kang used DODAB to partially phase transfer 12.5 nm aqueous Au NPs to organic solvents with 63% transfer efficiency.10 In this case, the hydrophobicity of the octadecyl chain helped promote phase transfer.10 While these partial successes are encouraging, a better understanding is still needed. Herein we have studied in detail ionic-surfactant mediated NP phase transfer from water to toluene and propose a model that has successfully predicted how to overcome current size limitations. By systematically studying a wide range of ionic surfactants and NP systems, we have developed a simple theory that defines the surfactant properties required for phase transfer. We have determined that only three things need to be considered to successfully predict which ionic surfactants succeed and which fail at phase transfer: (i) the length of the carbon chains, for suppression of NP aggregation, (ii) the bulkiness of the carbon chains, for solubilizing the NPs in the organic phase, and (iii) the ability to reduce the surface tension of the liquid−liquid interface, for loading onto and crossing the interface. Using this theory, ionic surfactants were identified for the phase transfer of spherical NPs up to 16 nm, significantly surpassing the existing limit of 5 nm and thereby demonstrating the utility of the theory. Once phase transferred, these NPs can be ligated based on the final nanoparticle requirements.

Next, 10 mL of 38.8 mM aqueous trisodium citrate was added to the yellow HAuCl4 solution, immediately turning it colorless. The solution darkened in ∼45 s and then turned red in ∼200 s. The procedure for Ag NPs was similar; however, the reaction was carried out on ice as it used NaBH4 as the reducing agent. Small-sized citrate Au NPs were synthesized using Jana et al.14 and Ojea-Jimenez et al.16 methods. Larger sizes of citrate-stabilized Au NPs were synthesized according to the Zhao et al. method.15 The CdSe quantum dots were synthesized according to the protocol of Ingole et al.19 and the aqueous Na4Ag44(p-MBA)30 clusters were synthesized according to the protocol of Desireddy et al.20 For phase transfer, 5 mL of toluene was added to 5 mL of aqueous Au NPs. Taking one case as an example, 2.41 mg of TOAB (4.4 μmol) and 0.51 mg C16TAB (1.4 μmol) solid powders were added to the vial, mixed, and allowed to phase separate. In this case, phase transfer succeeded and the toluene phase appeared red in color. Phase transfer of Ag NPs, Ag clusters, and CdSe quantum dots was evaluated in the same way. The extent of phase transfer was measured with absorption spectroscopy. It is worth noting here that surfactant was added as a solid powder rather than as a solution as is commonly reported in the literature.5,8−10 This strategy was necessary to deliver a larger quantity of surfactant to the liquid−liquid interface and into the aqueous phase than is possible in solution form, due to the limited solubility of the ionic surfactants in toluene. As a result, phase transfer efficiency was found to be higher for solid surfactant compared with saturated solutions of surfactants.

3. RESULTS AND DISCUSSION The complete phase transfer of NPs up to 5 nm in diameter has been established using ionic surfactants, typically TOAB.5 In the case of gold, the plasmonic red color acts as an indicator of successful phase transfer. For Au NPs larger than 5 nm, however, the color of the receiving solvent is not red, but is instead purple.5 This color is characteristic of aggregated Au NPs, indicating that the particle−particle interactions were too strong for the NPs to be completely solvated by TOAB. To weaken these forces, we have employed longer-chain surfactants to serve as more effective spacers between the larger NPs. When the long-chain surfactant CTAB (hereafter referred to as C16TAB, to denote its 16 carbon atom chain length) was introduced into the purple toluene phase, the Au NP aggregates dissolved to produce a bright red solution. The use of both TOAB and C16TAB enabled the successful phase transfer of 13 nm Au and Ag NPs, as shown in Figure 1a,b. Note that the colors of the final solutions were the same as those of the initial solutions, indicating that no aggregation occurred upon phase transfer. This was also quantitatively verified by absorption spectroscopy. Further, the addition of C16TAB did not disrupt the ability to phase transfer smaller NPs, as was demonstrated using Na4Ag44(p-MBA)30 clusters (containing methanol cosolvent) and 5 nm CdSe quantum dots (containing DMF cosolvent), as shown in Figure 1c,d. The

2. MATERIALS AND METHODS Chloroauric acid trihydrate (99.999%), doceyldimethylammonium bromide (DDAB, 98%), 1-hexadecyltriphenylphosphonium bromide (HTPB, 98+%), tetraoctadecylammonium bromide (4C18TAB, 98%), hexadecyltrimethylammonium bromide (CTAB or C16TAB, 98%), tetradecyltrimethylammonium bromide (C14TAB, >98%) were purchased from Alfa Aesar. Octyltrimethylammonium bromide (C8TAB, >98%), octadecyltrimethylammonium bromide (C18TAB, >98%), and sodium hexadecyl sulfate with ∼40% sodium octadecyl sulfate (SHS, >98%) were purchased from TCI America. Sodium borohydride (99%), 1-dodecaenthiol (DDT, ≥98%,), dihexadecyldimethylammonium bromide (DHAB, 97%), tetrahexylammonium iodide (THAI, ≥99%), p-mercaptobenzoic acid (p-MBA, 90%) and sodium dodecyl sulfate (SDS, ≥99%) were purchased from SigmaAldrich. Tetraoctylammonium bromide (TOAB, 98%), dimethylformamide (DMF), ethanol, methanol, acetone, toluene, trisodium citrate dehydrate (granular), and chloroform were purchased from Fisher. Millipore water (18.2 MΩ cm @ 25 °C) was used for aqueous syntheses. All materials were used without further purification. Syntheses for aqueous Au and Ag NPs are described in the literature.17,18 Briefly, for citrate-stabilized Au NPs, 100 mL of 1 mM aqueous HAuCl4 was boiled in a beaker while stirring at 700 rpm. 13838

dx.doi.org/10.1021/la503574s | Langmuir 2014, 30, 13837−13843

Langmuir

Article

hydrophobicity to effectively solubilize the NPs in toluene. Addition of a second surfactant containing only a single long alkyl chain (CnTAB, where n = 8, 14, 16, 18) was equally as ineffective, regardless of the length of the alkyl chains. In all cases tested, combinations of CnTAB surfactants were unable to transfer the 13 nm Au NPs (see Table 1).

largest NP successfully phase transferred with the TOAB +C18TAB combination was 16 nm in diameter, as shown in Figure 1e. Anionic surfactants were also used, in place of the cationic CnTAB surfactants, to attempt to solubilize the TOAB NP aggregates. In this case, the long-chain anionic surfactants SDS (C12) and SHS (C16) failed to produce a red solution of Au NPs in toluene. This is consistent with the expectation of repulsive interactions between the like charges of the anionic surfactants and the citrate-protected metal NPs. In the case of cationic surfactants, however, the surfactants interact favorably with the anionic NPs by charge pairing. In principle, it is possible for the cationic surfactants to reduce the net charge on the NPs to zero before entering the toluene receiving phase; however, this is not required since alkali metal cations are also available in the solution and may contribute to achieving charge neutrality. Charge pairing can be used to estimate the density of cationic surfactants on the anionic Au NPs if the coverage of citrate molecules on Au surfaces is known. STM imaging of citrate monolayers on Au(111) shows a surface packing density of 1 molecule/1.15 nm2 under conditions where each carboxylate is expected to be deprotonated (as is the case for our experiments).21 It is reasonable to assume that the citrate surface packing density on the Au NPs is similar to its packing density on Au(111), since it should not vary dramatically between facets on the spherical NPs and since the Au NP synthesis conditions involved more than enough citrate to fully saturate the surfaces of the NPs once the Au3+ ion reduction was completed. Based on ion pairing with the citrate trianions, the packing density of cationic surfactants is expected to be approximately one molecule per 38 Å2. The surface packing density of alkanethiols on Au(111) is approximately double this estimate;22 therefore, it is possible to accommodate enough molecules to saturate the surface of the NPs with either CnTAB surfactants or a mixture of TOAB and CnTAB surfactants. This is not the case if only TOAB is used, however, due to the bulkiness of its four carbon chains. In this case, other counterions would be needed, e.g., alkali metal cations. In experiments with mixtures of TOAB and CnTAB, the amount of surfactant used was always more than enough to fully saturate the surface of the NPs. It is important to note that the cationic surfactants were added as solids such that they immediately settled to the water−toluene interface where they dissolved readily into the water subphase. Using this delivery method, the solubility of the surfactants in toluene did not limit their transport into the water phase; therefore, there was no kinetic impediment to fully saturating the surface of the NPs with TOAB and CnTAB surfactants. The success of this combination of surfactants points toward a simple strategy for phase transfer with ionic surfactants. By screening a large number of combinations of ionic surfactant pairs, the general principles for successful phase transfer emerged. Namely, both bulkiness and length were found to be important for phase transfer, as bulkiness imparted solubility to the NPs in the receiving phase and long chain lengths prevented their aggregation. For example, when only C16TAB was used with 13 nm Au NPs, the phase transfer failed. This was indicated by the clear and colorless toluene receiving phase. Since C16TAB contains a single long alkyl chain, it does not provide enough

Table 1. Phase Transfer of 13.1 nm Au Nanoparticles as a Function of Surfactant Bulkiness and Length surfactants used

receiving phase

comment

C16TAB C8TAB+C14TAB C8TAB+C16TAB C14TAB+C16TAB C16TAB+C18TAB

colorless colorless colorless colorless colorless

linear + linear (incr. length)

TOAB TOAB+C8TAB TOAB+C14TAB TOAB+C16TAB TOAB+C18TAB

purple purple purple/red red red

bulky + linear (incr. length)

TOAB+THAI TOAB+DDAB TOAB+HTPB DDAB+DHAB

purple purple red red

bulky + bulky (incr. length)

The addition of TOAB, a bulky surfactant, provided sufficient hydrophobicity to effectively solubilize the 13 nm Au NPs in toluene, imparting color to that phase. The color of the toluene layer depended on the alkyl chain length of the CnTAB. For either C8TAB or C14TAB with TOAB, the color of the toluene layer was purple, due to the aggregation of Au NPs. It is worth noting, however, that the toluene phase has a slight reddish color in the case of the TOAB and C14TAB combination, indicating some suppression of aggregation. For the case of either C16TAB or C18TAB with TOAB, the color of the toluene layer was red, indicating a complete suppression of Au NP aggregation upon phase transfer into toluene. This shows that a second surfactant can be used with TOAB for successful phase transfer as long as that surfactant is sufficiently long to prevent aggregation once the NPs enter the organic phase (see Table 1). A schematic showing the complementary roles of the TOAB and CnTAB surfactants is shown in Figure 2. The role of TOAB is simply to provide hydrophobicity to the surface of the metal NPs, which enables them to be soluble in toluene. This is independent of particle size; all particle sizes tested were transferred into the organic phase with TOAB alone, although all but the smallest sizes remained aggregated in the organic phase as indicated by a purple color (see Table 2). In contrast,

Figure 2. Effective role of each surfactant molecule. (a) TOAB is short and bulky and provides hydrophobicity while (b) CnTAB is long and keeps NPs apart to suppress aggregation (not to scale). 13839

dx.doi.org/10.1021/la503574s | Langmuir 2014, 30, 13837−13843

Langmuir

Article

Table 2. Phase Transfer of Aqueous Au Nanoparticles into Toluene as a Function of Core Sizea

a

Red: completely dissolved; purple: aggregated; (purple/red): partially aggregated.

Figure 3. Three models for Au NP interactions in toluene. (a) Surfactant chains do not interpenetrate; (b) chains partially interpenetrate, possibly limited by an adsorbed layer of TOAB (red); and (c) chains completely interpenetrate to span the cores.

interpenetration. The surface coverage of CnTAB would have to be very high for bundling to occur, however, which is unlikely with TOAB also on the NP surface. Indeed, the interaction energies for the case of no interpenetration appear to be too low to be consistent with the experimental observations of NP solubility (see Supporting Information for details); therefore, CnTAB does not appear to form bundles. For the case of complete interpenetration of ligand chains, the CnTAB chains would presumably extend to the surface of the other NP (Figure 3c). This ability would be impeded by the presence of TOAB on the NP surface, however. Indeed, the interaction energies for the case of complete interpenetration appear to be too high to be consistent with the experimental observations of NP solubility (see Supporting Information); therefore, CnTAB does not appear to extend to the surface of the other NP. The best comparison with experimental solubilities was for the partial interpenetration model (Figure 3b). In this case, an impenetrable layer of hydrocarbon was added to the surface of each NP, which could be representative of an adsorbed layer of TOAB. The CnTAB chains could otherwise completely interpenetrate. For this model, the pattern of interaction energies was consistent with the pattern of nanoparticle solubilities; however, neither our measurements nor our model are detailed enough to determine the extent of interpenetration, i.e., the thickness of the impenetrable layer. We simply conclude that surfactant chain interpenetration is likely significant, which is consistent with expectations for a mixture of two surfactants with different chain lengths. Although the phase transfer of larger NPs can be achieved by pairing together surfactants with the necessary character to solubilize and stabilize the NPs in the receiving phase, the kinetics related to each of the surfactants can cause problems. If the rate at which one surfactant arrives at the NPs is very different than that of the other surfactant, failure can occur despite their individual properties. For example, if only C16TAB were added to aqueous 13 nm Au NPs, then the NPs would irreversibly aggregate such that the subsequent addition of TOAB would have no effect. To avoid these sorts of

CnTAB acts as a spacer to keep the NPs far enough apart to suppress the attractive van der Waals forces that lead to aggregation. The effectiveness of the CnTAB spacer ought to depend on the NP diameter, since the attractive forces scale with both interparticle separation and core diameter.23 This was indeed the case. For example, 7.2 nm Au NPs were successfully phase transferred using C14TAB as the second surfactant, whereas 13 nm Au NPs required C16TAB and 16 nm Au NPs required C18TAB for successful phase transfer. Phase transfer of 20 nm Au NPs was only partly successful with TOAB and C18TAB (see Table 2 for complete results). A simple model was constructed to better understand the dependence of solubility on chain length given in Table 2 and the geometry of the chain−chain interactions. Luedtke and Landman24,25 showed that the alkyl chains on ligand-passivated NPs can either organize into bundles when interchain interactions are strong or interdigitate when the interactions are weaker and the ligand shell is disordered. While the surfactants used in Table 2 have a mixture of chain lengths, these two possibilities provide very useful limits for the core− core separation, i.e., complete bundling with no interpenetration of ligand chains and complete interdigitation of ligand chains. It is possible to model the interaction between two Au NPs as only an interaction between their metal cores at a separation determined by their ligands,23,26 although more detailed models exist.25,27,28 In this case, however, the difference between solvating the ligand chains with a good solvent or with the ligand chains on the other NP could be considered negligible; i.e., it was not necessary to account for solvation in order to explain the improved solubility of NPs with increased chain length. By comparing these calculated interactions to the pattern of experimental solubilities as a function of NP size and surfactant chain length (see Table 2), the role of bundling or interdigitation can be ascertained. For the case of no interpenetration of ligand chains (Figure 3a), complete bundling would be expected since the conformational flexibility of individual chains ought to facilitate 13840

dx.doi.org/10.1021/la503574s | Langmuir 2014, 30, 13837−13843

Langmuir

Article

Figure 4. Phase transfer of aqueous ∼13 nm Au NPs (lower phase) into toluene (upper phase) using different ionic surfactants. The surfactant structure and geometry is shown below each vial.

Figure 5. Phase transfer of aqueous ∼13 nm Au NPs using a single ionic surfactant DHAB. (a) Surfactant-induced Au NP aggregate, (b) NP aggregate near the liquid−liquid interface, (c) jet of NPs entering the toluene phase, and (d) NP diffusing in toluene.

two alkyl chains with 12 or 16 carbon atoms, when 13 nm Au NPs were phase transferred the receiving toluene phase was red. These surfactants were able to impart enough hydrophobicity to the NPs and keep them sufficiently well separated to enable successful phase transfer. A surfactant containing four 18-carbon-atom chains was also successful, producing a red toluene phase, for the same reasons. The phase transfer efficiency was highest for DHAB (two 16carbon-atom chains), which was able to transfer nearly all of the NPs. In the case of the shorter chain DDAB (two 12-carbonatom chains), a significant amount of material remained aggregated. This is consistent with other experiments wherein the chain lengths were too short to suppress the attractive van der Waals forces that lead to aggregation. In the case of 4C18TAB (four 18-carbon-atom chains), the aqueous phase remained red in color. This indicated that some NPs were left behind due to the difficulty with which this surfactant entered the aqueous phase. It is noteworthy that the curvature of the interface did not change upon addition of this surfactant, in contrast to all of the other surfactants. The kinetics for phase transfer with 4C18TAB were also quite slow. These observations indicate that the hydrophobicity of the octadecyl chains impeded the movement of this surfactant onto and across the toluene−water interface. Clearly a barrier to

problems, it would be advantageous to combine the desirable properties of each surfactant in a single molecule. A series of ionic surfactants was selected with varying chain lengths and bulkiness. By comparing the effectiveness of each surfactant in the phase transfer of 13 nm Au NPs, the general principles behind successful phase transfer were found. Once again, both bulkiness and length were found to be important for phase transfer, as shown in Figure 4. The CnTAB surfactants with a single long chain all failed to transfer the NPs since they lacked the bulkiness required to solubilize the NPs in the toluene receiving phase. In this case, no NPs crossed the water−toluene phase boundary. The addition of three phenyl rings did not increase the bulkiness of the surfactants enough to improve the phase transfer. When the bulkiness was increased by using four alkyl chains instead of only one, the NPs became hydrophobic enough to leave the water phase. In the case of four hexyl chains, the NPs were not hydrophobic enough to enter the toluene phase and therefore remained on the interface. In the case of four octyl chains, the NPs were hydrophobic enough to enter the toluene phase but the chain length was not sufficient to prevent aggregation of the 13 nm Au NPs. Surfactants with two or more chains that contained 12 or more carbon atoms were in general successful. In the case of 13841

dx.doi.org/10.1021/la503574s | Langmuir 2014, 30, 13837−13843

Langmuir

Article

successful NP phase transfer; namely, the surfactant must be able to (i) get onto or cross the liquid−liquid interface, (ii) provide enough hydrophobicity to solubilize the NPs in the receiving phase, and (iii) act as a spacer to suppress the attractive van der Waals forces that lead to irreversible aggregation. DHAB was found to have the best balance of bulkiness (for NP solubility) and length (for steric repulsion) among the various surfactants screened, giving a deep red solution with nearly all of the material successfully transferred. The role of surfactant chain bundling and interdigitation was examined using an offset-interdigitation model for NP interactions. Mechanistic details of phase transfer were also presented. This improved understanding enabled the phase transfer of NPs up to 16 nm in diameter; however, this work provides guidance for selecting surfactants for the phase transfer of larger NPs as well.

adsorbing onto the interface would prevent molecules from passing through the interface in order to interact with the NPs that are the target of the phase transfer. From this result, an additional requirement can be defined for successful phase transfer. The three requirements can be summarized as follows. The surfactant must be able to (i) get onto or cross the liquid−liquid interface, (ii) provide enough hydrophobicity to solubilize the NPs in the receiving phase, and (iii) separate NPs enough to suppress the attractive van der Waals forces that lead to irreversible aggregation. These studies were also able to show the mechanism by which the single surfactant phase transfer works. When the surfactant was added to the water phase in the absence of a toluene receiving phase, the NPs aggregated to give the water phase a purple color. Upon addition of the toluene, the NP aggregates crossed the interface and dissolved to form a red solution. In this way, the Au NPs were delivered into the receiving phase in packets of aggregated material. This process was observed using optical microscopy, as shown in Figure 5. When a NP aggregate touched the water−toluene interface, a jet of Au NPs was observed to enter the toluene phase. The NPs within the aggregate were rapidly injected into the toluene phase due to the capillary force created by the hydrophobicity of the NPs in the aqueous phase. Once in the toluene phase, the dissolved NPs diffused to form a uniform red solution. This single-surfactant strategy also works for the phase transfer of aqueous NPs into other organic receiving phases with relatively high surface tensions, such as chloroform and xylene. Organic solvents such as hexane and cyclohexane are not suitable receiving phases, however, due to their low surface tensions. This can be rationalized by recognizing the requirement that a surfactant must be able to get onto or cross the liquid−liquid interface in order to be effective for phase transfer, as stated above. If a surfactant can reduce the interfacial tension, this requirement will be satisfied. Low surface tension solvents tend to also have a low interfacial tension, however, which makes it challenging for a surfactant to lower the tension of the interface further. It should be noted that this study has focused only on spherical NPs, due to the simplicity of their geometry and interparticle interactions. The concepts found in our study should in principle be applicable to NPs of different shapes; however, it must be recognized that the interactions between nonspherical NPs tend to be stronger since more mass is able to approach and interact over a given distance.23 For instance, rod-shaped NPs can interact along their length, which can lead to strong attractive forces and aggregation at modest separations. Further, surfactant-passivated NPs can themselves act as surfactants and reduce the interfacial tension by an amount that is proportional to their area.29,30 A nonspherical NP occupies a larger area on the interface and would therefore interact more strongly with the interface than a spherical particle of the same volume. In this case, binding of nonspherical NPs to the interface could be strong enough to interfere with phase transfer. In any of these cases, however, our model should serve as a starting point for identifying surfactants that can overcome the limitations introduced when considering NPs of different shapes.



ASSOCIATED CONTENT

S Supporting Information *

Details of the particle interaction model and the surfactant solubilities are available free of charge via the Internet at http:// pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel.: +1 419 530 4095. Fax: +1 419 530 4033. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors wish to thank Prof. Jacques Amar for useful discussions and the National Science Foundation for support under award number CHE-1012896.



REFERENCES

(1) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. Synthesis of Thiol-derivatised Gold Nanoparticles in a Two-Phase Liquid-Liquid System. J. Chem. Soc., Chem. Commun. 1994, 1994 (7), 801−802. (2) Tian, Y.; Fendler, J. H. Langmuir−Blodgett Film Formation from Fluorescence-Activated, Surfactant-Capped, Size-Selected CdS Nanoparticles Spread on Water Surfaces. Chem. Mater. 1996, 8 (4), 969− 974. (3) Sarathy, K. V.; Kulkarni, G. U.; Rao, C. N. R. A Novel Method of Preparing Thiol-Derivatised Nanoparticles of Gold, Platinum and Silver Forming Superstructures. Chem. Commun. 1997, 1997 (6), 537−538. (4) Chen, S.; Yao, H.; Kimura, K. Reversible Transference of Au Nanoparticles Across the Water and Toluene Interface: A Langmuir Type Adsorption Mechanism. Langmuir 2001, 17 (3), 733−739. (5) Cheng, W.; Wang, E. Size-Dependent Phase Transfer of Gold Nanoparticles from Water into Toluene by Tetraoctylammonium Cations: A Wholly Electrostatic Interaction. J. Phys. Chem. B 2004, 108 (1), 24−26. (6) Waters, C. A.; Mills, A. J.; Johnson, K. A.; Schiffrin, D. J. Purification of Dodecanethiol Derivatised Gold Nanoparticles. Chem. Commun. 2003, 2003 (4), 540−541. (7) Mayya, K. S.; Caruso, F. Phase Transfer of Surface-Modified Gold Nanoparticles by Hydrophobization with Alkylamines. Langmuir 2003, 19 (17), 6987−6993. (8) Muhammed, M. A. H.; Pradeep, T. Aqueous to Organic Phase Transfer of Au25 Clusters. J. Cluster Sci. 2009, 20 (2), 365−373. (9) Zhu, H.; Tao, C.; Zheng, S.; Wu, S.; Li, J. Effect of Alkyl Chain Length on Phase Transfer of Surfactant Capped Au Nanoparticles

4. CONCLUSIONS We have studied the role of the molecular structure of ionic surfactants in the aqueous-to-organic phase transfer of NPs. We have found that the surfactant must satisfy three criteria for 13842

dx.doi.org/10.1021/la503574s | Langmuir 2014, 30, 13837−13843

Langmuir

Article

Across the Water/Toluene Interface. Colloids Surf., A 2005, 256 (1), 17−20. (10) Zhao, S.; Kang, Y. Phase Transfer of Au Nanoparticles Using One Chemical Inducer: DDAB. J. Nanopart. Res. 2011, 13 (6), 2399− 2406. (11) Hiramatsu, H.; Osterloh, F. E. A Simple Large-Scale Synthesis of Nearly Monodisperse Gold and Silver Nanoparticles with Adjustable Sizes and with Exchangeable Surfactants. Chem. Mater. 2004, 16 (13), 2509−2511. (12) Lin, X. M.; Sorensen, C. M.; Klabunde, K. J. Digestive Ripening, Nanophase Segregation and Superlattice Formation in Gold Nanocrystal Colloids. J. Nanopart. Res. 2000, 2, 157−164. (13) Shankar, R.; Wu, B. B.; Bigioni, T. P. Wet Chemical Synthesis of Monodisperse Colloidal Silver Nanocrystals Using Digestive Ripening. J. Phys. Chem. C 2010, 114 (38), 15916−15923. (14) Jana, N. R.; Gearheart, L.; Murphy, C. J. Seeding Growth for Size Control of 5-40 nm Diameter Gold Nanoparticles. Langmuir 2001, 17 (22), 6782−6786. (15) Zhao, L.; Jiang, D.; Cai, Y.; Ji, X.; Xie, R.; Yang, W. Tuning the Size of Gold Nanoparticles in the Citrate Reduction by Chloride Ions. Nanoscale 2012, 2012 (4), 5071−5076. (16) Ojea-Jiménez, I.; Romero, F. M.; Bastús, N. G.; Puntes, V. Small Gold Nanoparticles Synthesized with Sodium Citrate and Heavy Water: Insights into the Reaction Mechanism. J. Phys. Chem. C 2010, 114 (4), 1800−1804. (17) Turkevich, J.; Stevenson, P. C.; Hillier, J. A Study of the Nucleation and Growth Processes in the Synthesis of Colloidal Gold. Discuss. Faraday Soc. 1951, 11, 55−75. (18) Creighton, J. A.; Blatchford, C. G.; Albrecht, M. G. Plasma Resonance Enhancement of Raman Scattering by Pyridine Adsorbed on Silver or Gold Sol Particles of Size Comparable to the Excitation Wavelength. J. Chem. Soc., Faraday Trans. 2 1979, 75, 790−798. (19) Ingole, P. P.; Abhyankar, R. M.; Prasad, B. L. V.; Haram, S. K. Citrate-Capped Quantum Dots of Cdse for the Selective Photometric Detection of Silver Ions in Aqueous Solutions. Mater. Sci. Eng., B 2010, 168 (1-3), 60−65. (20) Desireddy, A.; Conn, B. E.; Guo, J.; Yoon, B.; Barnett, R. N.; Monahan, B. M.; Kirschbaum, K.; Griffith, W. P.; Whetten, R. L.; Landman, U.; Bigioni, T. P. Ultrastable Silver Nanoparticles. Nature 2013, 501 (7467), 399−402. (21) Lin, Y.; Pan, G.-B.; Su, G.-J.; Fang, X.-H.; Wan, L.-J.; Bai, C.-L. Study of Citrate Adsorbed on the Au(111) Surface by Scanning Probe Microscopy. Langmuir 2003, 19 (24), 10000−10003. (22) Sellers, H.; Ulman, A.; Shnidman, Y.; Eilers, J. E. Structure and Binding of Alkanethiolates on Gold and Silver Surfaces: Implications for Self-Assembled Monolayers. J. Am. Chem. Soc. 1993, 115 (21), 9389−9401. (23) Israelachvili, J. Intermolecular and Surface Forces; Academic Press: New York, 1992. (24) Luedtke, W. D.; Landman, U. Structure, Dynamics, And Thermodynamics of Passivated Gold Nanocrystallites. J. Phys. Chem. 1996, 100 (32), 13323−13329. (25) Landman, U.; Luedtke, W. D. Small is Different: Energetic, Structural, Thermal, And Mechanical Properties. Faraday Discuss. 2004, 125, 1−22. (26) Ohara, P. C.; Leff, D. V.; Heath, J. R.; Gelbart, W. M. Crystallization of Opals from Polydisperse Nanoparticles. Phys. Rev. Lett. 1995, 75 (19), 3466−3469. (27) Khan, S. J.; Pierce, F.; Sorensen, C. M.; Chakrabarti, A. SelfAssembly of Ligated Gold Nanoparticles: Phenomenological Modeling and Computer Simulations. Langmuir 2009, 25 (24), 13861−13868. (28) Yan, H.; Cingarapu, S.; Klabunde, K. J.; Chakrabarti, A.; Sorensen, C. M. Nucleation of Gold Nanoparticle Superclusters from Solution. Phys. Rev. Lett. 2009, 102 (9), 095501. (29) Lin, Y.; Skaff, H.; Emrick, T.; Dinsmore, A. D.; Russell, T. P. Nanoparticle Assembly and Transport at Liquid-Liquid Interfaces. Science 2003, 299 (5604), 226−229.

(30) Bigioni, T. P.; Lin, X. M.; Nguyen, T. T.; Corwin, E. I.; Witten, T. A.; Jaeger, H. M. Kinetically Driven Self Assembly of Highly Ordered Nanoparticle Monolayers. Nat. Mater. 2006, 5 (4), 265−270.

13843

dx.doi.org/10.1021/la503574s | Langmuir 2014, 30, 13837−13843

Model for the phase transfer of nanoparticles using ionic surfactants.

Ionic surfactants are widely used for the phase transfer of nanoparticles from aqueous to organic phases; however, a model that can be used to select ...
5MB Sizes 1 Downloads 4 Views