DOI:
10.1039/C6RA03950B
(Paper)
RSC Adv., 2016,
6, 39926-39932
Surface-functionalizable amphiphilic nanoparticles for pickering emulsions with designer fluid–fluid interfaces†
Received
12th February 2016
, Accepted 14th April 2016
First published on 15th April 2016
Abstract
This work describes the synthesis of amphiphilic silica nanoparticles with functionalizable surfaces for the stabilization of aqueous drops in fluorinated solvents. State-of-the-art droplet microfluidics technology has relied on a single type of surfactant consisting of perfluorinated polyether and polyethylene glycol (PFPE–PEG). This type of surfactant, however, is known to have limitations including synthetic complexity and the cross-contaminations of droplet content caused by the inter-drop transport of small hydrophobic molecules. Previously we have overcome these limitations by replacing the surfactant with partially fluorinated silica nanoparticles (referred to as “F-SiO2 NPs”) as droplet stabilizers. Nonetheless, neither surfactants nor F-SiO2 NPs can provide additional functionalities at fluid–fluid interfaces due to the lack of reactive functional groups for further modifications. Here we introduce reactive groups on the surface of F-SiO2 NPs by the co-hydrolysis of different silane precursors, and demonstrate the successful conjugation of particle surfaces with fluorescent molecules, biomolecules and polymers. Our particles serve dual functions: they are amphiphilic for stabilizing water-in-oil drops, and they possess functional surface groups for presenting desired surface chemistries at the droplet interface. These particles offer the following advantages: (1) their synthesis is simple and scalable (up to grams); (2) they are effective in stabilizing droplets against coalescence under typical droplet manipulation conditions; (3) they prevent inter-drop molecular transport; and (4) their surfaces contain reactive groups such as amines that are capable of further conjugation with various “designer” molecules. We believe that the particles described in this work will open up opportunities in creating emulsions with tailored interfacial properties for new applications requiring customized fluid–fluid interfaces.
Introduction
Droplet microfluidics, where droplets act as individual reactors, has accelerated a range of high-throughput applications,1,2 including digital polymerase chain reactions (dPCR)3 and the directed evolution of enzymes.4 State-of-the-art droplet based assays have relied critically on a single type of surfactant: a tri-block copolymer consisting of fluoroalkyl groups as the fluorophilic tail and polyethylene glycol (PEG) domains as the hydrophilic headgroup (PFPE–PEG, or “EA-surfactants”).5 The fluorinated tails are necessary for solubilizing the surfactant in fluorinated solvents, which are often used as the continuous phase due to their gas permeability and biocompatibility.5 The PEG headgroup renders the droplet surface biocompatible with assays occurring inside the drops.5 Despite its widespread use in the field of droplet microfluidics, this type of surfactant is known to have three limitations: (1) the synthesis and purification tend to be complicated and costly. (2) Previous studies have shown that reverse micelles formed by these surfactants can transport small hydrophobic molecules in the droplets.6–11 Since these molecules are often used as fluorogenic substrates, this inter-drop transport (referred to as “leakage” hereafter) compromises the accuracy of the assay.12 (3) The PEG domain lacks reactive functional groups or side chains for the covalent immobilization of desired molecules onto droplet inner surfaces. A simple strategy that facilitates the modification of the surface chemistry of droplet interfaces can be particularly useful for studies such as interfacial rheology,13,14 and the use of drops as artificial cells which often require the reconstitution of proteins at droplet surfaces.15–20 Polyglycerol-based surfactants with side chains containing either hydroxyl or methoxy groups have been recently described.21 Further modification of the side chains, however, was not demonstrated.
In our previous work, we have addressed the first two limitations of surfactants by replacing them with partially fluorinated silica nanoparticles (referred to as “F-SiO2 NPs”) to stabilize water-in-fluorocarbon emulsions, or, “fluorinated pickering emulsions”.22,23 We have developed a simple method to synthesize F-SiO2 NPs in small scales at relatively low cost (see cost estimation in our previous paper23). These particles were effective in eliminating leakage and in maintaining assay accuracy, as the particles were irreversibly adsorbed at the droplet interface and did not form reverse micelles.22,23 In addition, the particles were compatible with the growth of bacteria, as well as anchorage-dependent mammalian cells. The latter was not possible with surfactant–laden interfaces as they could not provide sufficient rigidity for the adhesion and growth of such cells.22
Nevertheless, the F-SiO2 NPs we have reported thus far were intended primarily as inert droplet stabilizers consisting of silanol and fluoroalkyl surface groups only. While simple, this surface chemistry was not amenable to further conjugation, and could limit the range of applications of these NPs especially those requiring specific surface functional groups on the surface of the drops. Furthermore, the NPs have been synthesized in small batches at laboratory scale (typically 10–100 mg). While sufficient for initial characterizations, the ability to scale up the synthesis is necessary for extending the practical use of these particles.
The objective of this work is, therefore, to develop a strategy that allows: (1) the large-scale of synthesis of NPs in the gram scale, and (2) the covalent conjugation of the NPs with desirable functional molecules without compromising droplet stability and the ability of the NPs to prevent leakage of droplet contents. We show that these requirements can be satisfied by functionalizing NPs with surface amine groups (referred to as “NH2–F-SiO2 NPs” hereafter). Unlike previous work on designer particles which are typically dispersed in a single phase rather than at a fluid–fluid interface,24–28 our particles are “multi-functional” and serve at least two purposes: (1) they are amphiphilic and stabilize water-in-oil emulsions, and (2) they possess functional groups to present desired surface chemistries at the droplet interface. In the rest of the paper, we first describe the large-scale synthesis of the NPs. Second, we describe three examples of the conjugation of our particles with other molecules.
Experimental design
Materials
All chemicals were used as purchased without purification. Absolute ethanol (99%), anhydrous dimethylformamide (DMF) (98%), tetraethyl orthosilicate (TEOS) (98%), ammonium hydroxide solution (28%), (3-aminopropyl) triethoxysilane (APTES) (99%), rhodamine B (97%), rhodamine B isothiocyanate (RBITC) (98%), fluorescein isothiocyanate (FITC) (98%), (+)-biotin N-hydroxysuccinimide ester (biotin-NHS) (98%) were purchased from Sigma-Aldrich. 1H,1H,2H,2H-Perfluorooctyltriethoxysilane (FAS) (97%) was purchased from Fisher Scientific. Methoxyl PEG isothiocyanate (mPEG-ITC, MW = 5000) was purchased from Nanocs Inc. Fluorescein-labelled streptavidin (streptavidin–FITC) (1 mg mL−1) was purchased from Vector Laboratories Inc.
Large-scale synthesis of F-SiO2 NPs
F-SiO2 NPs were synthesized in large scale by modifying our previously published work.22 7.14 mL of TEOS was added to a solution mixture containing 100 mL of ethanol (EtOH), 2 mL of deionized water, and 2.86 mL of NH4OH (28 wt%). The solution was stirred vigorously (∼800 rpm) at room temperature for 12 hours. 6 mL of neat FAS was then added directly to every 60 mL of the synthesized SiO2 NPs dispersion obtained above, followed by vigorous stirring (∼800 rpm) at room temperature for 40 min. EtOH was added to dilute the reacting solution to terminate the reaction (with a dilution factor of 5 before starting the first washing cycle). Three washing cycles were performed. Each cycle included the collection of particles by centrifugation (Sorvall LEGEND X1R) at 5000 rpm for 60 min, removal of the supernatant and replenishment of EtOH (except for the last cycle where EtOH replenishment was not necessary). The solid particles were isolated by desiccation overnight.
Large-scale synthesis of NH2–F-SiO2 NPs
F-SiO2 NPs with surface amine groups (referred to as “NH2–F-SiO2 NPs”) were synthesized using the same method as that for F-SiO2 NPs, except either 1.6 mL neat APTES or 1.6 mL 2% APTES (v/v) in EtOH were pre-mixed with 6 mL neat FAS before the addition to 60 mL of the as-synthesized SiO2 NPs dispersion. The initial APTES concentrations in the reaction mixture were [APTES] = 101 mM for synthesizing NH2–F-SiO2 NPs with high surface amine density, and [APTES] = 2.02 mM for synthesizing NH2–F-SiO2 NPs with low surface amine density.
Synthesis of 2 μm F-SiO2 MPs and NH2–F-SiO2 MPs for Auger Electron Spectroscopy (AES) analysis
2 μm F-SiO2 microparticles (MPs) were synthesized following our previous work.22 The synthesis of 2 μm NH2–F-SiO2 MPs was identical to that for 2 μm F-SiO2 MPs, except every 120 μL of neat FAS was replaced by a mixture of 120 μL FAS and 20 μL APTES.
Conjugation of NH2–F-SiO2 NPs with other molecules
NPs to be conjugated (in solid form obtained from 10 mL original dispersion after removing EtOH by desiccation) were redispersed in 1 mL EtOH under sonication. 0.5 mL conjugating agent in anhydrous DMF was then added dropwise (see details in Table S1†). The resulting mixture was stirred at room temperature for 1 hour. The conjugated product was purified by three cycles of centrifugation-washing (with EtOH) to remove excess conjugating agent. The NPs were dried under desiccation overnight. We refer to these particles as “X-N–F-SiO2 NPs”, where X is the molecule to be conjugated on the NPs.
Physical adsorption of rhodamine B (or fluorescein) onto NH2–F-SiO2 NPs
Excess solid rhodamine B (or fluorescein) (10 mg) was added to 1 mL of 1.5% (wt/wt) NH2–F-SiO2 NPs in HFE-7500. The mixture was sonicated for 30 minutes. The resulting mixture was filtered by a syringe filter (pore size: 5 μm) to remove undispersed rhodamine B (or fluorescein) solid. The filtrate was collected in an Eppendorf tube.
Droplet generation
Monodisperse microdroplets were generated from flow-focusing devices. The continuous phase contained 1.5% (w/w) NPs dispersed in HFE-7500. The choice of NPs for the continuous phase and the composition of the dispersed phase are specified in Table S2.† The flow rates of the continuous phase and the dispersed phase were fixed at 0.8 mL h−1 and 0.2 mL h−1 respectively. The droplets were collected in Eppendorf tubes. In some cases, excess NPs in the continuous phase were removed by washing with HFE-7500 three times for subsequent fluorescence imaging (see details on Table S2†).
Leaching tests
Monodisperse droplets stabilized by either fluo–N–F-SiO2 NPs or RBads–F-SiO2 NPs were collected in Eppendorf tubes. Here “fluo” refers to fluorescein molecules that are covalently attached to the particles, and “RBads” refers to rhodamine B molecules that are physically adsorbed to the particles. Excess NPs in the continuous phase were removed by washing with 0.5 mL of HFE-7500 three times. The droplets were then transferred to an Eppendorf tube containing 1.5% (wt/wt) NH2–F-SiO2 NPs in HFE-7500. The resulting mixture was incubated at room temperature for 3 hours.
E. coli cell culture
E. coli was cultured following the protocol in our previous publication.22 In this work, the growth of E. coli inside droplets stabilized by NH2–F-SiO2 NPs was also tested.
Results and discussions
Gram-scale synthesis of particles
The F-SiO2 NPs synthesized in small scales following our previous protocol22,23 were stable for at least 6 months. No precipitation was observed over the storage period after NPs were dispersed in HFE-7500. Compared with freshly prepared NPs, these NPs performed equally well in terms of: (1) the ability to generate monodisperse drops from flow-focusing devices, and (2) the stability of the drops under typical droplet manipulation conditions (Fig. S1†). This long-term stability of our NPs indicates that they can be potentially synthesized and stored in large volumes for future use. We are thus motivated to refine our protocol to make it compatible with the production of NPs in large scales.
To increase the scale of synthesis, we used the same basic approach as described in our previous work.22 Hydrophilic SiO2 NPs were first synthesized by Stober method, followed by their fluorination in the presence of fluoroalkyl silane (FAS) (Fig. 1a). It is known that the nucleation and growth of silica nanoparticles in the Stober process depend on multiple factors.29 We chose experimental parameters including the concentration of reactants and stirring rates to obtain particles with good dispersibility in HFE-7500 (see Experimental design for details). For a typical large-scale synthesis, we have obtained ∼2 g of solid F-SiO2 NPs. The scale was currently limited by the volume of reagents our centrifuge could process (see Experimental). To prepare a stock dispersion, we dispersed these NPs in ∼35 mL HFE-7500 to form a dispersion with a NPs concentration of ∼3.3% (wt/wt) (Fig. 1b(i)). This volume was more than 20 times larger than the scale obtained in our previous work at the same NPs concentration (∼1.5 mL, Fig. 1b(ii)). The stock dispersion was further diluted to a concentration of 1.5% (wt/wt) for experiments in the rest of the paper.
 |
| Fig. 1 (a) Schemes of F-SiO2 NPs and NH2–F-SiO2 NPs synthesis. Note that fluoroalkyl groups and silanol groups are not drawn on the NPs surface. (b) Photograph showing the contrast in scale between F-SiO2 NPs from (i) current large-scale synthesis and (ii) previous small-scale synthesis. These two samples have the same concentration of NPs (3.3% by weight) in HFE-7500. (c) XPS spectra of NH2–F-SiO2 NPs with initial [APTES] = 101 mM. | |
To extend our synthesis to create F-SiO2 NPs with additional surface functional groups, we performed co-hydrolysis of FAS and a silane precursor containing desired functional groups. In this work, we chose APTES as a model silane to synthesize fluorinated SiO2 NPs with surface amine groups (“NH2–F-SiO2 NPs”), since amine groups are known to be compatible with various conjugation techniques.30 X-ray photoelectron spectroscopy (XPS) of synthesized NH2–F-SiO2 NPs showed both F and N peaks (Fig. 1c), which confirmed that the presence of both FAS and APTES during the co-hydrolysis process allowed the growth of silica shells containing fluoroalkyl and amine groups on NPs surface. In our previous work,23 we showed that the presence of excess hydrophilic groups (such as PEG) could compromise NPs dispersibility in fluorinated solvents. The poor dispersibility of NPs resulted in the clogging of microfluidic devices during droplet generation. To maintain the dispersibility of our NH2–F-SiO2 NPs in HFE-7500, we reduced the amount of amine groups on NPs surface by reducing [APTES] during the co-hydrolysis step (Fig. S2†). These NH2–F-SiO2 NPs maintained good dispersibility in HFE-7500 for at least a month, as evidenced by the lack of change in the size of the particles measured by dynamic light scattering (from 120 nm to 126 nm).
To show that our NH2–F-SiO2 NPs were di-functionalized particles rather than a simple mixture of F-SiO2 NPs and aminated particles (NH2–SiO2 NPs), we used Auger Electron Spectroscopy (AES) to probe the elemental composition on the surface of individual particles. For this experiment only, we used 2 μm-NH2–F-SiO2 and F-SiO2 microparticles (MPs) instead of nanoparticles to ensure the AES signals were collected from a single particle, as charging of the surface of nanoparticles under SEM made it difficult to identify the interface between the nanoparticles and the substrate. For both NH2–F-SiO2 and F-SiO2 MPs, we randomly chose regions on the surface of individual particles to collect AES signals from. In the case of F-SiO2 MPs, there was no nitrogen peak observed as expected (Fig. S3,† blue spectrum). In contrast, in the case of NH2–F-SiO2 MPs, a nitrogen peak (∼400 eV) was observed, indicating the presence of surface amine groups. The presence of a fluorine peak (∼680 eV) indicated the presence of surface fluoroalkyl groups, which further confirmed the di-functionality of our synthesized NH2–F-SiO2 MPs (Fig. S3,† red spectrum).
To estimate the relative abundance of both amine groups and fluoroalkyl groups on the surface of NH2–F-SiO2 NPs, we used surface elemental composition from XPS to measure the ratio between nitrogen content and fluorine content from our particles. Table S3† shows that the ratio between surface amine and fluoroalkyl groups was approximately 1.35
:
1 and 0.18
:
1 for NH2–F-SiO2 NPs with high and low surface amine densities, respectively.
Fig. S4–S8† verify that our method for scaling up the synthesis and for introducing surface amine groups did not compromise the desired properties of F-SiO2 NPs demonstrated in our previous work. The NPs synthesized here were able to generate monodisperse microdroplets from flow-focusing devices (Fig. S4†). Drops stabilized by these NPs were also stable under typical droplet manipulation conditions. In addition, these NPs were effective in preventing the leakage of small fluorescent molecules (Fig. S5 and S6†) and were compatible with the growth of bacteria cells inside the droplets (Fig. S7 and S8†).
Examples of surface functionalization
Here we demonstrate three examples of surface functionalization of NH2–F-SiO2 NPs with: (1) fluorescent molecules, (2) biomolecules (biotin), and (3) polymer (PEG).
Example 1: conjugation with fluorescent molecules. We conjugated NH2–F-SiO2 NPs (the sample corresponding to Fig. S2†) with fluorescein by reacting these NPs with fluorescein isothiocyanate (FITC) under anhydrous environment (Fig. 2a). Isothiocyanate forms covalent bond with primary amines, and is commonly used for fluorophore conjugation of proteins.30 Fig. 2c shows a fluorescence image of droplets stabilized by NH2–F-SiO2 NPs conjugated with fluorescein (referred to as “fluo–N–F-SiO2 NPs”. The notation “–N” stands for the covalent bond formed between the target molecule and nitrogen atom of the amine group). We measured the fluorescence intensity distribution across a droplet using line-scan in imageJ. The intensity was then normalized to that in the continuous phase. As can be seen, the center of the droplet was much brighter than the continuous phase. We can identify the presence of a fluorescent rim either directly from the fluorescence image or from the line-scan profile, where the separation of the two fluorescence peaks corresponded to the diameter of the droplet (∼100 μm). To show that the fluorescent rim observed was not due to physical adsorption of FITC onto NPs, we used F-SiO2 NPs (without surface amine groups) as a negative control (Fig. 2b). Excess FITC molecules were removed by centrifugation before the NPs were redispersed to neat HFE-7500 and used for imaging. Fig. 2c shows that droplets stabilized by F-SiO2 NPs prepared using the scheme shown in Fig. 2b were not fluorescent (see line-scan in Fig. 2c and S9†). As expected, the absence of amine groups prevented the covalent conjugation of FITC on F-SiO2 NPs surface. Combined, these results indicate the fluorescent rim in Fig. 2c was due to the conjugation between FITC and NH2–F-SiO2 NPs rather than the non-specific physical adsorption of FITC on the particles. As aminated particles with no fluoroalkyl groups could not be dispersed in HFE-7500, the fluorescent rim here could not have been formed from a simple mixture of fluo–N-SiO2 NPs and F-SiO2 NPs. This result indirectly supports the di-functionality of our NH2–F-SiO2 NPs, which is critical for both their dispersibility in fluorinated solvents and the successful covalent conjugation of molecules.
 |
| Fig. 2 Schemes showing the conjugation of fluorescent molecules with fluorescein isothiocyanate (FITC) on the surface of (a) NH2–F-SiO2 NPs and (b) F-SiO2 NPs, respectively. Note that fluoroalkyl groups and silanol groups are not drawn on the NPs surface. (c) Fluorescence image of droplets stabilized by NH2–F-SiO2 NPs conjugated with fluorescein (fluo–N–F-SiO2 NPs). (d) Line scans of fluorescence intensity of a typical droplet stabilized by fluo–N–F-SiO2 NPs (red, product from scheme in (a)) and F-SiO2 NPs (blue, product from scheme in (b)), respectively. The line scan started at the continuous phase about 10 μm away from the droplet surface, and fluorescence intensity was normalized to that of the continuous phase. The insets show the fluorescence image of the drop across which the line scans were obtained. The contrast in the right inset was enhanced to show the presence of the droplet. | |
Previously, we have rendered our F-SiO2 NPs fluorescent by the physical adsorption of rhodamine B (the resulting NPs were referred to as “RBads–F-SiO2 NPs”. See Experimental). Nonetheless, only a limited range of fluorescent molecules can be adsorbed onto F-SiO2 NPs for fluorescence imaging. For example, fluorescein cannot be physically adsorbed on F-SiO2 NPs using the same method (Fig. S10†). An advantage of the covalent approach described here is that it expands the range of fluorescent molecules that can be conjugated onto NPs surface, as long as the molecules contain reactive groups that can react with amines (see Fig. S11† for covalent conjugation of rhodamine B on NH2–F-SiO2 NPs). Another advantage of covalent conjugation over physical adsorption is that the former prevents the leaching of fluorescent molecules from the particles. Fig. 3b and c compare the leaching of fluorescent molecules from fluo–N–F-SiO2 NPs with that from RBads–F-SiO2 NPs. In both cases, when the droplets were suspended in a continuous phase containing neat HFE-7500, fluorescence was observed on the droplet surface only (Fig. S12†). There was no leaching of rhodamine B from RBads–F-SiO2 NPs into HFE-7500, likely due to the low solubility of small hydrophobic molecules in fluorinated solvents.9 After suspending the droplets in a continuous phase containing 1.5% (wt/wt) NH2–F-SiO2 NPs in HFE-7500, only fluo–N–F-SiO2 NPs stabilized drops maintained their original fluorescence intensity distribution across the droplet (Fig. 3b and d). For RBads–F-SiO2 NPs stabilized drops, the continuous phase became fluorescent, with a fluorescence intensity exceeding that in the central region of the droplet (Fig. 3c and d). We note that the continuous phase in Fig. 3c appeared uniformly fluorescent as the size of the NH2–F-SiO2 NPs (about 100 nm) was much smaller than the resolution of our imaging system, and the concentration of the NPs was sufficiently high that the NPs could not be imaged as individual fluorescent dots. Overall, our results demonstrate that the leaching of fluorescein was negligible when fluorescein was covalently bonded to the surface of the particles. On the other hand, physically adsorbed rhodamine B diffused from RBads–F-SiO2 NPs to otherwise non-fluorescent NH2–F-SiO2 NPs in the continuous phase.
 |
| Fig. 3 (a) Schemes showing the leaching tests for droplets stabilized by fluo–N–F-SiO2 NPs and RBads–F-SiO2 NPs, respectively. (b and c) Fluorescence images showing the resulting droplets after incubation in HFE-7500 containing 1.5% NH2–F-SiO2 NPs, where the droplets were stabilized by (b) fluo–N–F-SiO2 NPs and (c) RBads–F-SiO2 NPs, respectively. (d) Line scans of fluorescence intensity of a typical fluo–N–F-SiO2 NPs stabilized droplet (green) and RBads–F-SiO2 NPs stabilized droplet (red). The line scans started at the continuous phase about 10 μm away from the droplet surface, and fluorescence intensity was normalized to that of the continuous phase. | |
The covalently-conjugated fluorescent NPs described here can be useful for measuring the composition of particles at the liquid–liquid interface. As a simple proof of concept, we mixed fluo–N–F-SiO2 NPs and NH2–F-SiO2 NPs at different ratios while fixing the total amount of NPs in HFE-7500. These NPs mixtures were then used as the continuous phase to generate aqueous drops. By examining the fluorescence intensity of the resulting droplets after removing excess un-adsorbed NPs in the continuous phase, we found a linear correlation between the percentage of fluorescent NPs (i.e., fluo–N–F-SiO2 NPs) and the average fluorescence of a drop (Fig. S13†). This result suggests these two types of NPs possess similar adsorption properties onto the droplet surface, likely due to their similar size and surface chemistries. The fluorescence intensity of the drop can thus be used as a simple, indirect measure for the composition of the particles on the droplet surface.
Example 2: conjugation with biomolecules. Thus far, most droplet stabilizers are designed to offer biocompatibility without participating in the reactions occurring inside the drop.5,21 The presence of amine groups on our NPs opens up opportunities for the conjugation of biomolecules on the surface of the particles. Drops stabilized by these bio-functionalized NPs can be potentially used for new types of assays. Since fluorinated solvents are known to be benign to biomolecules, including proteins and DNA,31 it should be possible to disperse these NPs in fluorinated solvents without compromising the functions of the biomolecules prior to their adsorption at the droplet surface. As a proof of concept, we demonstrate the covalent conjugation of biotin on NH2–F-SiO2 NPs using an approach similar to that for the conjugation of fluorescent molecules. Biotin-linked N-hydroxysuccinimide ester (biotin-NHS) was used to link its biotin unit to surface amine groups of NH2–F-SiO2 NPs via amide bonds (Fig. 4a, the product is referred to as “biotin–N–F-SiO2 NPs” hereafter). Fig. 4b shows the XPS spectrum of biotin–N–F-SiO2 NPs synthesized from NH2–F-SiO2 NPs as characterized in Fig. 1c. The successful conjugation of biotin was indicated by (i) the presence of sulfur peak at ∼165 eV; and (ii) an increase in carbon-to-fluorine content ratio of the product (from ∼0.75 to ∼1 compared with NH2–F-SiO2 NPs in Fig. 1c) after the removal of unreacted biotin-NHS. We were able to obtain good dispersibility of biotin–N–F-SiO2 NPs in HFE-7500 by using NH2–F-SiO2 NPs with lower surface amine density. Fig. 4c shows that this dispersion was able to generate stable monodisperse aqueous drops from flow-focusing device.
 |
| Fig. 4 (a) Scheme of biotin conjugation on NH2–F-SiO2 NPs. Note that fluoroalkyl groups and silanol groups are not drawn on the NPs surface. (b) XPS spectra of biotin–N–F-SiO2 NPs obtained from the biotinylation of NH2–F-SiO2 NPs with initial [APTES] = 101 mM. (c) Optical image of monodisperse droplets stabilized by biotin–N–F-SiO2 NPs obtained from biotinylation of NH2–F-SiO2 NPs with initial [APTES] = 2.02 mM. | |
Example 3: conjugation with polymers. Previously, we showed that the non-specific adsorption of proteins onto NPs surface was prevented if PEG was on the surface of NPs.23 Fig. 5a shows the covalent conjugation of PEG on NH2–F-SiO2 NPs by using PEG-isothiocyanate (PEG-ITC, Fig. 5a). To test if PEG was conjugated onto NPs surface, we introduced fluorescein-conjugated streptavidin (referred to as “streptavidin–FITC”) in the disperse phase. The fluorescence intensity profile across the droplets will indicate whether streptavidin–FITC adsorbs onto the NPs surface. When the drops were stabilized by NH2–F-SiO2 NPs without PEG conjugation, the fluorescence was highly intense on the droplet surface (thick bright fluorescent rims in Fig. 5d, red curve in Fig. 5f), indicating significant adsorption of streptavidin–FITC onto the droplet surface (also see the left sketch in Fig. 5c). In contrast, when the drops were stabilized by PEG–N–F-SiO2 NPs with low PEG density (corresponding to the blue XPS spectrum in Fig. S14,† with a surface oxygen-to-fluorine ratio of 1.17), the intensity of the fluorescence rim was significantly diminished (Fig. 5e, green curve in Fig. 5f). In addition, fluorescence was partially recovered in the central regions of the droplet, indicating the mitigation of streptavidin–FITC adsorption onto the droplet surface. Such adsorption, however, was not completely eliminated due to the relatively low PEG density required to maintain good dispersibility of PEG–N–F-SiO2 NPs in HFE-7500. This observation is consistent with the trade-off between NPs dispersibility in fluorinated solvents and the density of surface hydrophilic groups discussed in our previous work.23 When the surface PEG density on PEG–N–F-SiO2 NPs was further increased (corresponding to the red XPS spectrum in Fig. S14,† with a surface oxygen-to-fluorine ratio of 2), the resulting droplets did not show any streptavidin–FITC adsorption. The dispersibility of the particles in HFE-7500 was compromised, however (Fig. S15†).
 |
| Fig. 5 (a) Scheme of PEG conjugation with NH2–F-SiO2 NPs. Note that fluoroalkyl groups and silanol groups are not drawn on the NPs surface. (b) Scheme showing the generation of aqueous drops stabilized by either NH2–F-SiO2 NPs or PEG–N–F-SiO2 NPs. The continuous phase contained 2% (w/w) NPs (either NH2–F-SiO2 NPs or PEG–N–F-SiO2 NPs) in HFE-7500 and the disperse phase contained 0.1 mg mL−1 fluorescein-conjugated streptavidin (streptavidin–fluorescein isothiocyanate, streptavidin–FITC). (c) Sketch of the resulting droplets obtained from the droplet generation process shown in (b). The droplets were stabilized by NH2–F-SiO2 NPs (left) and PEG–N–F-SiO2 NPs (right), respectively. (d and e) Fluorescence image of droplets obtained from the droplet generation process shown in (b). The droplets were stabilized by (d) NH2–F-SiO2 NPs and (e) PEG–N–F-SiO2 NPs, respectively. (f) Line scans showing fluorescence intensity distribution of droplets from (d) and (e), respectively. | |
Conclusions
Table 1 summarizes the advantages of our surface-functionalizable NPs compared with other state-of-the-art droplet stabilizers: (1) our method enables the scaling-up of the synthesis of fluorinated NPs at a lower cost compared with surfactants that are commercially available so far.23 The current scale is limited only by the volume that our centrifuge can process (∼400 mL per round). (2) Our method allows simple introduction of additional functional groups on NPs surface by co-hydrolysis. It expands the selection of surface functional groups, as a wide range of silane precursors are commercially available. These NPs can be further functionalized with molecules of interest via subsequent conjugation chemistry with consistent conjugation density and small batch-to-batch variations (Fig. S16†). We believe that the multi-functional NPs demonstrated in this work can be useful for creating droplet interfaces with tailored properties for new applications requiring customized fluid–fluid interfaces.
Table 1 Comparison of this work with other droplet stabilizers
|
Synthetic complexity |
Cost on scaling up |
Biocompatibility |
Accuracy (no leakage) |
Further surface functionalization |
Krytox-NH4 |
Simple |
Low |
Poor to moderate |
Low |
No |
Krytox-NH4 + jeffamine32 |
Simple |
Low |
Good |
Low |
No |
EA-surfactant |
Complicated |
High |
Good |
Low |
No |
Polyglycerol based fluoro-surfactant21 |
Complicated |
High |
Good |
Not demonstrated |
Not demonstrated |
F-SiO2 NPs (our previous work) |
Simple |
Low |
Good |
High |
No |
NH2–F-SiO2 NPs (this work) |
Simple |
Low |
Good |
High |
Yes |
Acknowledgements
We acknowledge funding from the Stanford Nano Shared Facilities Bio/Medical Mini Seed Grant, Stanford CHeM-H, Stanford Bio-X, and the Stanford Woods Institute for the Environment. S. T. acknowledges additional support from 3M Untenured Faculty Award.
References
- S.-Y. Teh, R. Lin, L.-H. Hung and A. P. Lee, Lab Chip, 2008, 8, 198–220 RSC.
- H. Song, D. L. Chen and R. F. Ismagilov, Angew. Chem., Int. Ed., 2006, 45, 7336–7356 CrossRef CAS PubMed.
- M. M. Kiss, L. Ortoleva-Donnelly, N. R. Beer, J. Warner, C. G. Bailey, B. W. Colston, J. M. Rothberg, D. R. Link and J. H. Leamon, Anal. Chem., 2008, 80, 8975–8981 CrossRef CAS PubMed.
- J. J. Agresti, E. Antipov, A. R. Abate, K. Ahn, A. C. Rowat, J.-C. Baret, M. Marquez, A. M. Klibanov, A. D. Griffiths and D. A. Weitz, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 4004–4009 CrossRef CAS PubMed.
- C. Holtze, A. C. Rowat, J. J. Agresti, J. B. Hutchison, F. E. Angile, C. H. J. Schmitz, S. Koster, H. Duan, K. J. Humphry, R. A. Scanga, J. S. Johnson, D. Pisignano and D. A. Weitz, Lab Chip, 2008, 8, 1632–1639 RSC.
- F. Courtois, L. F. Olguin, G. Whyte, A. B. Theberge, W. T. S. Huck, F. Hollfelder and C. Abell, Anal. Chem., 2009, 81, 3008–3016 CrossRef CAS PubMed.
- Y. Chen, A. W. Gani and S. K. Y. Tang, Lab Chip, 2012, 12, 5093–5103 RSC.
- P. A. Sandoz, A. J. Chung, W. M. Weaver and D. Di Carlo, Langmuir, 2014, 30, 6637–6643 CrossRef CAS PubMed.
- Y. Skhiri, P. Gruner, B. Semin, Q. Brosseau, D. Pekin, L. Mazutis, V. Goust, F. Kleinschmidt, A. El Harrak, J. B. Hutchison, E. Mayot, J.-F. Bartolo, A. D. Griffiths, V. Taly and J.-C. Baret, Soft Matter, 2012, 8, 10618–10627 RSC.
- G. Woronoff, A. El Harrak, E. Mayot, O. Schicke, O. J. Miller, P. Soumillion, A. D. Griffiths and M. Ryckelynck, Anal. Chem., 2011, 83, 2852–2857 CrossRef CAS PubMed.
- L. Mazutis, J.-C. Baret, P. Treacy, Y. Skhiri, A. F. Araghi, M. Ryckelynck, V. Taly and A. D. Griffiths, Lab Chip, 2009, 9, 2902–2908 RSC.
- L. Rosenfeld, T. Lin, R. Derda and S. K. Y. Tang, Microfluid. Nanofluid., 2014, 16, 921–939 CrossRef.
- A. J. Mendoza, E. Guzman, F. Martinez-Pedrero, H. Ritacco, R. G. Rubio, F. Ortega, V. M. Starov and R. Miller, Adv. Colloid Interface Sci., 2014, 206, 303–319 CrossRef CAS PubMed.
- N. J. Alvarez, S. L. Anna, T. Saigal, R. D. Tilton and L. M. Walker, Langmuir, 2012, 28, 8052–8063 CrossRef CAS PubMed.
- M. Takinoue and S. Takeuchi, Anal. Bioanal. Chem., 2011, 400, 1705–1716 CrossRef CAS PubMed.
- V. Noireaux and A. Libchaber, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 17669–17674 CrossRef CAS PubMed.
- E. M. Schmid, D. L. Richmond and D. A. Fletcher, Methods Cell Biol., 2015, 128, 319–338 Search PubMed.
- E. Karzbrun, A. M. Tayar, V. Noireaux and R. H. Bar-Ziv, Science, 2014, 345, 829–832 CrossRef CAS PubMed.
- D. Needham, Nat. Nanotechnol., 2011, 6, 761–762 CrossRef CAS PubMed.
- G. Villar, A. J. Heron and H. Bayley, Nat. Nanotechnol., 2011, 6, 803–808 CrossRef CAS PubMed.
- O. Wagner, J. Thiele, M. Weinhart, L. Mazutis, D. A. Weitz, W. T. S. Huck and R. Haag, Lab Chip, 2016, 16, 65–69 RSC.
- M. Pan, L. Rosenfeld, M. Kim, M. Xu, E. Lin, R. Derda and S. K. Y. Tang, ACS Appl. Mater. Interfaces, 2014, 6, 21446–21453 CAS.
- M. Pan, F. Lyu and S. K. Y. Tang, Anal. Chem., 2015, 87, 7938–7943 CrossRef CAS PubMed.
- M. Caldorera-Moore, N. Guimard, L. Shi and K. Roy, Expert Opin. Drug Delivery, 2010, 7, 479–495 CrossRef CAS PubMed.
- N. Doshi and S. Mitragotri, Adv. Funct. Mater., 2009, 19, 3843–3854 CrossRef CAS.
- S. Mitragotri and J. Lahann, Nat. Mater., 2009, 8, 15–23 CrossRef CAS PubMed.
- Z. Peng and H. Yang, Nano Today, 2009, 4, 143–164 CrossRef CAS.
- E. Brown, H. Zhang, N. A. Forman, B. W. Maynor, D. E. Betts, J. M. DeSimone and H. M. Jaeger, Phys. Rev. E, 2011, 84, 031408 CrossRef PubMed.
- K. Nozawa, H. Gailhanou, L. Raison, P. Panizza, H. Ushiki, E. Sellier, J. P. Delville and M. H. Delville, Langmuir, 2005, 21, 1516–1523 CrossRef CAS PubMed.
- H. Sahoo, RSC Adv., 2012, 2, 7017–7029 RSC.
- K. M. Midthun, P. G. Taylor, C. Newby, M. Chatzichristidi, P. S. Petrou, J.-K. Lee, S. E. Kakabakos, B. A. Baird and C. K. Ober, Biomacromolecules, 2013, 14, 993–1002 CrossRef CAS PubMed.
- C. J. DeJournette, J. Kim, H. Medlen, X. Li, L. J. Vincent and C. J. Easley, Anal. Chem., 2013, 85, 10556–10564 CrossRef CAS PubMed.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra03950b |
|
This journal is © The Royal Society of Chemistry 2016 |
Click here to see how this site uses Cookies. View our privacy policy here.