Journal of Colloid and Interface Science 421 (2014) 64–70
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Anisotropic colloidal crystal particles from microfluidics Yao Cheng, Cun Zhu, Zhuoying Xie, Hongcheng Gu, Tian Tian, Yuanjin Zhao ⇑, Zhongze Gu ⇑ State Key Laboratory of Bioelectronics, School of Biological Science and Medical Engineering, Southeast University, Nanjing 210096, China
a r t i c l e
i n f o
Article history: Received 6 November 2013 Accepted 27 January 2014 Available online 5 February 2014 Keywords: Anisotropy Colloidal crystal Silica nanoparticles Hydrogel Barcoding Microfluidics Capillary
a b s t r a c t Anisotropic colloidal crystal particles (CCPs) have showed their great potential in biotechnology and structural materials due to their anisotropic shapes and tunable optical property. However, their controllable generation is still a challenge. Here, a novel microfluidic approach is developed to generate anisotropic CCPs. The microfluidic device is composed of an injection capillary and a collection capillary with available size and shape. Based on the device, the anisotropic particles with non-close-packed colloidal crystal structures are achieved by photo-polymerizing droplet templates in a confined collection capillary with different shapes and sizes. Moreover, anisotropic close-packed CCPs can be made from non-close-packed CCPs through a thermal process. It is demonstrated that the anisotropic CCPs in different sizes, structural colors and shapes (rods, cuboids and disks) can be generated. These distinguishable features of resultant particles make them ideal barcodes for high-throughput bioassays. In order to prove it, DNA multiplex detection is carried out. The experimental results indicate that achieved particles have a great encoding capacity and are highly practical for multiplex coding bioassays. Therefore, we believe that the anisotropic CCPs would be highly promising barcodes in biomedical applications, including highthroughput bioassays and cell culture research where multiplexing is needed. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction The self-assembly of colloidal nanoparticles provides a simple and cheap approach to create three-dimensional structure materials [1–6]. The achieved colloidal crystal materials are with long-range ordered array of the nanoparticles and thus display a structure of periodic variation in the refractive index. The structure imparts the colloidal crystals an interesting optical property of photonic band gaps (PBGs), which is of great significance for photonic crystal devices [7–9], biological and chemical sensors [10–16], tunable lasers [17] and intelligent interfacial materials [18]. A major obstacle to produce the colloidal crystal materials is lack of reliable approaches to assemble nanoparticles into well controllable shapes. Virtually most colloidal-scale particles are naturally spherical because the interfacial tension between two immiscible phases tends to minimize the surface area, leading to the formation of spherical droplets. However, non-spherically anisotropic particles are also desired due to their great potential in biotechnology, structural materials, and other fields [19–22]. Traditional methods of fabricating non-spherically anisotropic particles include template molding [23], stretching of spherical particles [24,25], self-assembly of particles [26–28] and imprint ⇑ Corresponding authors. Fax: +86 25 83795635. E-mail addresses: [email protected] (Y. Zhao), [email protected] (Z. Gu). http://dx.doi.org/10.1016/j.jcis.2014.01.041 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
lithography technique [29]. It is usually difficult to use these methods to produce large quantities of monodisperse anisotropic particles of tunable geometries and adjustable sizes. As an alternative, microfluidic technique has emerged as a promising and versatile technique for generating non-spherically anisotropic particles due to their capabilities in manipulating fluids in controlled environments [30–35]. By confining the photo-precursor droplets in the microfluidic channels of different sizes and shapes, anisotropic particles can be generated in a flash of ultraviolet light [36]. In addition to droplet microfluidics, the anisotropic particles can also be produced by microfluidic lithography technology [37,38] which combines lithography and photo-chemistry into laminar flow microfluidics. Although many progresses have been achieved, it is still a challenge to fabricate CCPs with the feature of controllable anisotropic shapes and tunable optical properties. In this paper, we report a novel microfluidic approach to generate a series of anisotropic CCPs. A microfluidic device was assembled by aligning an injection capillary and a collection capillary coaxially inside a square capillary. When the size of droplets was larger than the inner diameter of collection capillary, they would be squeezed into non-spherical shape. Thus, by employing collection capillaries with different sizes and shapes, the monodisperse droplet templates with corresponding geometries appeared and flowed through the capillaries. In order to introduce optical properties into the droplet templates, the dispersed phase used for
65
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
droplet templates generation was the photo-precursor solution that contained non-close-packed colloidal nanoparticles array in the poly (ethylene glycol) diacrylate (PEG-DA) medium. By photo-polymerizing the droplet templates in situ of the confined collection capillaries, anisotropic particles with the same ordered colloidal crystal structures could be generated. These non-closepacked CCPs could be transformed into anisotropic close-packed CCPs by a simply thermal process. It was demonstrated that the anisotropic CCPs in different sizes, structural colors and shapes (rods, cuboids and disks) could be generated. As a typical application, the angle-independent rod-like CCPs were employed for multiplex biomolecule assay. Spectroscopic measurements and DNA hybridization experiments indicate that CCPs are highly practical for multiplex coding bioassays. We anticipate that this technology may find important application in multiplexed high-throughput bioassays 2. Materials and methods 2.1. Material The material used as outer phase was hexadecane (Sigma–Aldrich Co.). The inner phase was composed of a silica nanoparticle colloidal photonic crystal aqueous solution. To prepare the inner phase, monodisperse charged silica nanoparticles were first synthesized by the Stöber method [39]. The purified silica nanoparticles were dispersed in the pre-gel solution composed of 10% (v/v) poly (ethylene glycol) diacrylate (PEG-DA, MW 700, Sigma–Aldrich Co.) and 1% (v/v) photoinitiator (2-hydroxy-2-methylpropiophenone, Sigma–Aldrich Co.), and subsequently were shaken with an ion-exchange resin (Bio-Rad AG501-X8(D)) until strong diffraction became visually evident (20 min). After removing the resin by centrifugation, the colloidal crystal aqueous solution was achieved and sonicated for 30 min before use. 2.2. Methods 2.2.1. Microfluidics The capillary microfluidic device consisted of coaxially assembled glass capillaries with different cross-sectional shapes on glass slides. To construct the microfluidic device for fabricating rodshaped particles, a round capillary with an outer diameter/inner diameter of 1/0.58 mm (World Precision Instrument, Inc., Shanghai, China), tapered to achieve an orifice diameter of approximately 100 lm, was used for the inner tube. Another round capillary with an outer diameter of 1 mm and an inner diameter of 200 lm, was used as collection tube and coated with a hydrophobic reagent (trimethoxy (octadecyl) silane, Sigma–Aldrich Co.). The above capillaries were then coaxially assembled in a square capillary with a side length of 1.05 mm (AIT Glass, Rockaway, NJ, USA). A transparent epoxy resin was used to seal the tubes where required. To fabricate the cuboid-shaped particles, typically the orifice diameter of inner tube was adjusted to 200 lm, while the collection tube was changed into a square capillary with an inner diameter of 500 lm. For the disk-shaped particles, flat capillaries with an inner width/height of 1/0.2 mm, were used as inner tube and collection tube instead of round capillaries. By heating and pulling, the inner flat tube had a tapered tip with an inner width diameter of about 50 lm. All the collection tubes were treated with the hydrophobic reagent. In this work, inner phase and outer phase were pumped into the capillary microfluidic device using syringe pumps (PHD 2000 series, Harvard, Plymouth Meeting, PA, USA). By controlling the flow rates, droplet templates were stably generated in the collection tube and in situ photo-polymerized by UV-light (365 nm, 100w).
Finally, the particles were thoroughly washed with hexane, ethanol and pure water, respectively. 2.2.2. Thermal processing Representative non-close-packed colloidal crystal hydrogel particles were first dried at 65 °C for 6 h, and then sintered in the muffle at a rate of 5–800 °C for hold time of 3 h. After natural cooling to ambient temperature, the particles were taken out. 2.2.3. Biomolecule assay For further study of the anisotropic CCPs in biomolecule assays, oligonucleotides (Table 1) were bought from Invitrogen Biotechnology Co., Ltd. Hybridization buffer (750 mmol L1 NaCl, 150 mmol L1 sodium citrate, pH 7.4) and wash buffer (PBS, 10 mmol L1 phosphate sodium buffer solution, pH 7.4, 100 mmol L1 NaCl) were obtained from Shanghai Your Sun Biological Technology Co., Ltd. For the immobilization of probes, the particles were decorated with 5% (v/v) (3-Glycidyloxypropyl) trimethoxysilane (GPTMS) in toluene solution for 24 h, washed respectively by toluene, ethyl alcohol and pure water for three times, and then incubated with the amino-functionalized probe DNA solution (10 lM) at 4 °C overnight. In order to block residual reactive sites, the particles were immersed in 5% (w/v) bovine serum albumin (BSA) at room temperature for 1 h. All reagents were measured by 2 lL per bead. After being washed with wash buffer, the particles were used as carriers for DNA sequence detection in microwells. For multiplexed assays, rod-like particles in three colors were immobilized with oligonucleotide probes, named probe A, probe B, and probe C, correspondingly. The particles were incubated in hybridization buffer with 10 lM AMC-labeled target DNAs under continuous shaking at 37 °C for 1 h. Then, the particles were rinsed away three times with wash buffer in turns to remove unbound target DNA. Finally, the fluorescence spectra and images were taken after the reaction. 2.2.4. Characterization Photographs of photonic crystal hydrogel particles were taken by an optical microscope (OLYMPUS BX51) equipped with a cool CCD camera (Media-Cybernetics Evolution MP 5.0). Reflection spectra were recorded by an optical microscope equipped with a fiber optic spectrometer (Ocean Optics, USB2000-FLG). Fluorescence spectra were recorded by a microscope (OLYMPUS IX71) equipped with a fiber optic spectrometer (Ocean Optics, QE65000). The microstructures were characterized by a scanning electron microscope (SEM, HITACHI, S-3000N). 3. Results and discussion 3.1. Preparation of colloidal nanoparticles solution as dispersed phase To introduce optical properties into the anisotropic particles, we conducted the fabrication by using a PEG-DA aqueous solution containing 1% (v/v) photoinitiator and charged monodisperse silica nanoparticles as dispersed phase. When the concentration of
Table 1 Sequences of oligonucleotides used in this work. Base sequence
Name
50 -NH2-C6-GCG GCC TTA ATC ATT TCG CTT TCA GAA CTG-30 50 -NH2-C6-TGATCG CGG TGT CAG TTC TTT-30 50 -NH2-C6-GTGGAA TTG AGC AGC GTT GGT-30 50 -AMC-CAG TTC TGA AAG CGA AAT GAT GAA GGC CGC-30 50 -AMC-AAA GAA CTG ACA CCG CGA TCA-30 50 -AMC-ACC AAC GCT GCT CAA TTC CAC-30
Probe DNA A Probe DNA B Probe DNA C Target DNA A Target DNA B Target DNA C
66
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
charged silica nanoparticles reached a certain level, these nanoparticles tended to self-assemble into non-close-packed face-centered cubic (FCC) or body-centered cubic (BCC) ordered structures in solution owing to the interparticle repulsion and energy configuration. Thanks to the periodic arrangement of silica nanoparticles in the PEG-DA aqueous solution, the resultant mixture generated a PBG to prohibit the propagation of light with certain wavelength or frequency located in the PBG. As a result, the PEG-DA solution would exhibit vivid iridescent color and characteristic reflection peak. The position of main reflection peak could be estimated by Bragg’s equation:
k ¼ 2Dðn2eff cos2 hÞ
1=2
ð1Þ
where neff is the average refractive index of the solution, D is the diffracting plane spacing, and h is the Bragg angle of incidence of the light. When the incident light was perpendicular (h = 90°) to the (1 1 1) plane of the FCC structures, Eq. (1) could be simplified as k = 2Dneff. It is not difficult to find that the optical properties of the PEG-DA solution could be manipulated via changing either diffracting plane spacing or average refractive index of the solution. Therefore, the emerged PBG of the PEG-DA solution could be conveniently controlled by changing the size or concentration of silica nanoparticles that dispersed in the PEG-DA solution. It is worth pointing out that the concentration should be controlled within a proper region, since PEG-DA solutions containing silica nanoparticles with relatively high concentrations would result in the generation of droplets with poor monodispersity in the collection capillary due to high viscosity, while those with relatively low concentrations would affect the assembly of silica nanoparticles and thus quality of the products. In this case, we prepared the dispersed phases by dispersing charged silica nanoparticles with the diameter of 150 nm in the PEG-DA aqueous solution. As a result, PEG-DA solutions with controllable PBGs covering the whole visible range could be achieved by simply varying the concentrations of silica nanoparticles in the range from 10% to 30% (w/v), as shown in Fig. 1. 3.2. Microfluidic generation of anisotropic CCPs Anisotropic CCPs were evolved from the non-spherical colloidal crystal droplets that served as templates. Fig. 2 shows the schematic illustration of a typical glass capillary based microfluidic device for the fabrication of anisotropic particles with a series of geometries, which was assembled by coaxially aligning a cylindrical injection capillary and collection capillaries with various shapes
Fig. 1. (a) Digital photographs and (b) corresponding characteristic normalized reflection peaks of five kinds of colloidal crystal PEG-DA solutions. These solutions were prepared by dispersing charged silica nanoparticles with the size of 150 nm at different concentrations (from left to right: 27%, 23%, 21%, 19%, and 15% (w/v), respectively) in the PEG-DA aqueous solution.
Fig. 2. Schematic illustration of the microfluidic device used for the generation of anisotropic particles with various geometries. This device was assembled by a cylindrical injection capillary, a square capillary, and a collection capillary that aligned coaxially. The collection capillaries with various sizes and shapes were used to control the morphologies of generated template droplets in the channel. The resultant droplets were subsequently solidified in situ using UV irradiation.
inside a square capillary. The dispersed phase was pumped through the injection capillary, while the continuous phase flowed through the square capillary into the collection capillary in the same direction. In this case, the PEG-DA solution was continuously sheared by hexadecane at the tip of the injection capillary to form monodisperse W/O droplets. When the flow rate of hexadecane was not fast enough, PEG-DA solution would be enriched at the tip of injection capillary, eventually leading to the generation of non-spherical droplets in the confined collection capillary. After in situ solidification of these generated droplets with the UV irradiation, the CCPs with corresponding non-spherical shape were obtained immediately. It was more simple and efficient method for the CCPs generation than traditional evaporation induced crystallization which required impractically long times for complete consolidation and inevitably yields crystals with severe cracks. However, in the present method, the concentration of the nanoparticles should be well optimized to impart the non-spherical CCPs with vivid structural colors. The anisotropic morphology of resultant non-spherical CCPs was mainly determined by the droplets generated in microfluidic devices, which was closely related to the geometry of collection capillary. By controlling the size and shape of the collection capillary, non-spherical droplets thus CCPs could be conveniently fabricated. When a regular cylindrical capillary was applied as the collection capillary, rod-like droplets could be obtained. Replacing the collection capillary with square or flat capillary, cuboid-like or disk-like droplets would be formed in the device, respectively. We consider the formation of these non-spherical droplets should be attributed to the enrichment of aqueous droplets at the tip of the injection capillary owing to the relatively slow shearing, as well as the confinement of the collection capillary. Fig. 3 shows the representative digital photographs, SEM images and statistic geometric characterization of these three kinds of anisotropic CCPs. Thanks to the non-close-packed ordered structures formed via the self-assembly of silica nanoparticles on both surface and bulk of the anisotropic CCPs, they exhibited brilliant structural colors after the in situ solidification. However, the three kinds of resultant particles showed different optical properties when they were viewed at different directions and angles, as shown in Fig. 4. The cuboid-like and disk-like CCPs all displayed obvious spectra-shift no matter along the minor axes or major axes. As for the rod-like CCPs, the spectra showed similar shift when they were detected along the major axes, while the spectra showed unobservable shift along the minor axes. This angle-independent optical property should be due to the cylindrical symmetry of the rod-like particles, which endowed them with identical photonic responses independent of the rotation round the major axes. Thus, it is anticipated that the cod-like CCPs will be ideal materials for encoded multiplex bioassay, because the feature of angle-independent will make their decoding easy and accurate.
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
67
Fig. 3. (a–c) Optical image of anisotropic CCPs with various geometries: (a) rods, (b) cuboids, (c) disks; (d–f) their corresponding SEM images; and (g–i) their corresponding statistic geometric characterization along the minor axes and major axes respectively. Scale bars: (a, d) 200 lm, (b, c, e, f) 500 lm.
Besides the geometry of collection capillary, the relative flow rates between the dispersed phase and continuous phase would also affect the morphology of resultant non-spherical droplets and thus the CCPs. Therefore, we systematically investigated the effect of flow rates on the morphology of generated droplets by fixing the flow rate of one phase while varying the other. Take the rod-like particles generated in the cylindrical collection capillary for example, the outcomes of these trials are demonstrated in Fig. 5. When the flow rate of oil phase was fixed at 0.5 mL h1, the aspect ratio of resultant rod-like droplets dramatically increased with the increase in the flow rate of PEG-DA aqueous phase. However, when the flow rate of oil phase was boosted to 2 mL h1, no significant changes were found on the aspect ratio of resultant droplets even increasing the flow rate of the PEG-DA aqueous phase from 0.5 mL h1 to 5 mL h1 (Fig. 5a). On the other side, when the flow rate of PEG-DA aqueous phase was fixed at 1 mL h1 or 2 mL h1, the aspect ratio of obtained droplets decreased together with the increase in flow rate of oil phase (Fig. 5b). Hence, increasing the flow rate of PEG-DA aqueous phase and/or decreasing that of oil phase would be beneficial for the fabrication of anisotropic particles with larger aspect ratios. Moreover, by using PEG-DA solutions containing silica nanoparticles with different concentrations, an array of anisotropic CCPs with controllable aspect ratios and various structural colors could be achieved (Fig. 5c).
3.3. Transform non-close-packed structures into close-packed structures The non-close-packed lattice structures in the resultant anisotropic CCPs could be easily transformed into close-packed ordered structures via a thermal process. We conducted the calcination utilizing rod-like CCPs that assembled from the silica nanoparticles with the diameter of 210 nm in a muffle furnace at 800 °C for 3 h. Fig. 6 shows the digital photographs and corresponding SEM images of these particles before and after the calcination. The reflection peak of rod-like particles underwent a significant blueshift after the calcination, from the original position located at 630 nm to 470 nm, and the color of particles simultaneously changed from red to bluish violet after the thermal treatment (Fig. 6a and d). We consider that the shrink of non-close-packed ordered structures and reduction in average refractive index should be responsible for the blue-shift of reflection peak and structural color. During the calcination, PEG-DA hydrogels filled in the spaces among neighboring silica nanoparticles were gradually decomposed into gases. As a result, silica nanoparticles entrapped in the PEG-DA hydrogel were released from the hydrogel networks and driven to each other to form close-packed hexagonal ordered structures, eventually leading to the decrease in both inter-particle distance and average refractive index. SEM images taken from the surface and cross-section of the rod-like CCPs further confirmed
68
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
Fig. 4. Characteristic reflection peaks of three kinds of CCPs: (a, c, e) recorded at different angles along the major axes; (b, d, f) recorded at different angles along the minor axes.
Fig. 5. (a, b) Relationships between the aspect ratios of rod-like droplets and the flow rates that obtained by (a) fixing the flow rate of oil phase at 0.5 mL h1 and 2 mL h1 while varying that of PEG-DA aqueous phase, and (b) fixing the flow rate of PEG-DA aqueous phase at 2 mL h1 and 1 mL h1 while changing that of oil phase. (c) Anisotropic particles with various aspect ratios and structural colours that obtained by using PEG-DA solutions containing silica nanoparticles with different concentrations as the dispersed phases. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
our hypothesis. Compared with original structures of the particles, PEG-DA hydrogels could rarely be found on the surface and bulk of particles after calcination (Fig. 6b, c, e, and f). The calcinated particles still maintained bright structural colors and ordered lattice structures, indicating that close-packed anisotropic CCPs with high-quality could be obtained via the programmed thermal treatment. 3.4. Biomolecule assay As a typical application, the angle-independent rod-like CCPs were employed for biological multiplexed assays, in which a model DNA hybridization experiment was carried out. Three rod-like particles in blue, green and orange were immobilized with
different oligonucleotides, named probe A, probe B, and probe C, respectively. The probes were amine-functionalized oligonucleotides, whose corresponding targets were labeled with 7-amino-4methylcoumarin (AMC). All the particles were incubated in microwells each containing stochastic combination of target A, target B and target C. Due to the specific hybridization between the probe DNA and its corresponding fluorescently labeled target DNA, it was expected that the fluorescence signals would be detected only on the particles which their corresponding targets were present. Fig. 7 shows a typical result of the multiplex DNA detection in the solution of target A. It can be observed that only the CCPs in blue color exhibited fluorescence signal. The result was consistent with the content of the analyte solution to which the encoded rodlike particles were exposed. Thus, it is feasible to use the structural colors as the encoded elements of the particles for multiplexed assays. Moreover, as the particles could also be distinguished by their aspect ratios (as demonstrated in Fig. 5c), the identification codes of the particles could be further increased by using the joint coding of the structural colors and the aspect ratios.
4. Conclusions We fabricated a series of anisotropic CCPs by microfluidics. A microfluidic device was assembled by aligning a cylindrical injection capillary and a collection capillary with available size and shape coaxially inside a square capillary. Based on these devices, pre-gel droplet templates that containing non-close-packed colloidal nanoparticles array and with a series of geometries were generated and flowed through the collection capillaries. By polymerizing the droplet templates in the capillaries, the nonspherical particles with ordered colloidal crystal structures could be achieved. These non-close-packing colloidal crystal particles could also be transformed into close-packing colloidal crystal particles with the corresponding geometries by a simply thermal processing. The achieved particles were with simple identification from their sizes, structural colors or shapes. In order to demonstrate the feasibility of anisotropic CCPs in the biomolecule assay, multiplex DNA detection was carried out. The experiments show
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
69
Fig. 6. (a) Optical image of rod-like particles before calcination; SEM images taken from (b) surface, and (c) cross-section. (d) Optical image, and (e, f) SEM images after calcining at 800 °C for 3 h. Scale bars: (a, d) 200 lm, (b, c, e, f) 200 nm.
Fig. 7. Optical image (a) and fluorescence microscope images (b) of rod-shaped particles after DNA hybridization. Scale bar indicates 200 lm.
that particles are highly practical for multiplex coding bioassays. This microfluidic approach provides a novel method to create anisotropic CCPs and will be of greatly important in potential application for colloidal science, microfluidics, optical materials and biosensors, as well as further understanding transport and dynamic behavior of anisotropic CCPs with non-edged shape (e.g. rod and disk) or sharp-edged shape (cuboids) in many fields of application. Acknowledgments We are grateful to the support of the National Science Foundation of China (Grant Nos. 21105011, 51073034 and 50925309), the National Science Foundation of Jiangsu (Grant No. BK2012735), the Science and Technology Development Program of Suzhou (Grant No. ZXG2012021), the Postdoctoral Science Foundation of China (Grants 2012M520052 and 2013T60484), the Program for Changjiang Scholars and Innovative Research Team in University (IRT1222), and the 333 Talent Project Foundation. References [1] J. Ge, Y. Yin, Angew. Chem. Int. Ed. 50 (2011) 1492–1522. [2] Y.J. Zhao, Z.Y. Xie, H.C. Gu, C. Zhu, Z.Z. Gu, Chem. Soc. Rev. 41 (2012) 3297– 3317. [3] Z.W. Mao, H.L. Xu, D.Y. Wang, Adv. Funct. Mater. 20 (2010) 1053–1074. [4] O.D. Velev, S. Gupta, Adv. Mater. 21 (2009) 1897–1905. [5] G.V. Freymann, V. Kitaev, B.V. Lotsch, G.A. Ozin, Chem. Soc. Rev. 42 (2013) 2528–2554. [6] D. Xu, W. Zhu, Q. An, W.N. Li, X.S. Li, H.W. Yang, J.X. Yin, G.T. Li, Chem. Commun. 48 (2012) 3494–3496. [7] M.A. Haque, G. Kamita, T. Kurokawa, K. Tsujii, J.P. Gong, Adv. Mater. 22 (2010) 5110–5114.
[8] S.H. Kim, H. Hwang, S.M. Yang, Angew. Chem. Int. Ed. 51 (2012) 3601–3605. [9] X.B. Hu, Q. An, G.T. Li, S.Y. Tao, J. Liu, Angew. Chem. Int. Ed. 45 (2006) 8145– 8148. [10] Y.J. Zhao, X.W. Zhao, J. Hu, J. Li, W.Y. Xu, Z.Z. Gu, Angew. Chem. Int. Ed. 48 (2009) 7350–7352. [11] L. Zhang, C.L. Lu, Y.F. Li, Z. Lin, Z.H. Wang, H.P. Dong, T.Q. Wang, X.M. Zhang, X. Li, J.H. Zhang, B. Yang, J. Colloid Interface Sci. 374 (2012) 89–95. [12] X.S. Li, L.H. Peng, J.C. Cui, W.N. Li, C.X. Lin, D. Xu, T. Tian, G.X. Zhang, D.Q. Zhang, G.T. Li, Small 8 (2012) 612–618. [13] L.R. Shang, F.Q. Shangguan, Y. Cheng, J. Lu, Z.Y. Xie, Y.J. Zhao, Z.Z. Gu, Nanoscale 5 (2013) 9553–9557. [14] J.T. Bacaa, D.N. Finegoldb, S.A. Asher, Analyst 133 (2008) 385–390. [15] D. Nakayama, Y. Takeoka, M. Watanabe, K. Kataoka, Angew. Chem. Int. Ed. 42 (2003) 4197–4200. [16] H. Li, J.X. Wang, F. Liu, Y.L. Song, R.M. Wang, J. Colloid Interface Sci. 356 (2011) 63–68. [17] R.L. Justin, Y.R. Ying, P. Jiang, H.F. Stephen, Adv. Mater. 18 (2006) 300–303. [18] Y.J. Zhao, H.C. Gu, Z.Y. Xie, H.C. Shum, B.P. Wang, Z.Z. Gu, J. Am. Chem. Soc. 135 (2013) 54–57. [19] Y.J. Zhao, Z.Y. Xie, H.C. Gu, L. Jin, X.W. Zhao, B.P. Wang, Z.Z. Gu, NPG Asia Mater. 4 (2012) e25. [20] A.H.L. Chow, H.H.Y. Tong, P. Chattopadhyay, B.Y. Shekunov, Pharm. Res. 24 (2007) 411–437. [21] S.C. Glotzer, M.J. Solomon, Nat. Mater. 6 (2007) 557–562. [22] S. Elingarami, X.L. Li, N.Y. He, J. Nanosci. Nanotechnol. 13 (2013) 4539–4551. [23] B.F. Ye, Y.J. Zhao, T.T. Li, Z.Y. Xie, Z.Z. Gu, J. Mater. Chem. 21 (2011) 18659– 18664. [24] P.P. Lele, E.M. Furst, Langmuir 25 (2009) 8875–8878. [25] J.A. Champion, S. Mitragotri, Proc. Natl. Acad. Sci. 103 (2006) 4930–4934. [26] Z.Y. Yu, C.F. Wang, L.T. Ling, L. Chen, S. Chen, Angew. Chem. Int. Ed. 51 (2012) 2375–2378. [27] J.R. Millman, K.H. Bhatt, B.G. Prevo, O.D. Velev, Nat. Mater. 4 (2005) 98–102. [28] D.J. Kraft, W.S. Vlug, C.M. van Kats, A. van Blaaderen, A. Imhof, W.K. Kegel, J. Am. Chem. Soc. 131 (2009) 1182–1186. [29] J.Y. Wang, Y.P. Wang, S.S. Sheiko, D.E. Betts, J.M. DeSimone, J. Am. Chem. Soc. 134 (2012) 5801–5806. [30] C.H. Chen, A.R. Abate, D. Lee, E.M. Terentjev, D.A. Weitz, Adv. Mater. 21 (2009) 3201–3204. [31] T. Hutter, W.A.C. Bauer, S.R. Elliott, W.T.S. Huck, Adv. Funct. Mater. 22 (2012) 2624–2631.
70
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
[32] A.R. Studart, H.C. Shum, D.A. Weitz, J. Phys. Chem. B 113 (2009) 3914–3919. [33] H.C. Shum, A.R. Abate, D. Lee, A.R. Studart, B. Wang, C.H. Chen, J. Thiele, R.K. Shah, D.A. Weitz, Macromol. Rapid Commun. 31 (2010) 108–118. [34] E. Kumacheva, P. Garstecki, 10. Microfluidic Synthesis of Polymer Particles with Non-Conventional Shapes, Microfluidic Reactors for Polymer Particles, John Wiley & Sons, Ltd., Chichester, UK, 2011. [35] S. Seiffert, Macromol. Rapid Commun. 33 (2012) 1286–1293.
[36] S.Q. Xu, Z.H. Nie, M. Seo, P. Lewis, E. Kumacheva, H.A. Stone, P. Garstecki, D.B. Weibel, I. Gitlin, G.M. Whitesides, Angew. Chem. Int. Ed. 44 (2005) 734–738. [37] S.K. Suh, K. Yuet, D.K. Hwang, K.W. Bong, P.S. Doyle, T.A. Hatton, J. Am. Chem. Soc. 134 (2012) 7337–7343. [38] D. Dendukuri, D.C. Pregibon, J. Collins, T.A. Hatton, P.S. Doyle, Nat. Mater. 5 (2006) 365–369. [39] W. Stober, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1968) 62–69.
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Anisotropic colloidal crystal particles from microfluidics Yao Cheng, Cun Zhu, Zhuoying Xie, Hongcheng Gu, Tian Tian, Yuanjin Zhao ⇑, Zhongze Gu ⇑ State Key Laboratory of Bioelectronics, School of Biological Science and Medical Engineering, Southeast University, Nanjing 210096, China
a r t i c l e
i n f o
Article history: Received 6 November 2013 Accepted 27 January 2014 Available online 5 February 2014 Keywords: Anisotropy Colloidal crystal Silica nanoparticles Hydrogel Barcoding Microfluidics Capillary
a b s t r a c t Anisotropic colloidal crystal particles (CCPs) have showed their great potential in biotechnology and structural materials due to their anisotropic shapes and tunable optical property. However, their controllable generation is still a challenge. Here, a novel microfluidic approach is developed to generate anisotropic CCPs. The microfluidic device is composed of an injection capillary and a collection capillary with available size and shape. Based on the device, the anisotropic particles with non-close-packed colloidal crystal structures are achieved by photo-polymerizing droplet templates in a confined collection capillary with different shapes and sizes. Moreover, anisotropic close-packed CCPs can be made from non-close-packed CCPs through a thermal process. It is demonstrated that the anisotropic CCPs in different sizes, structural colors and shapes (rods, cuboids and disks) can be generated. These distinguishable features of resultant particles make them ideal barcodes for high-throughput bioassays. In order to prove it, DNA multiplex detection is carried out. The experimental results indicate that achieved particles have a great encoding capacity and are highly practical for multiplex coding bioassays. Therefore, we believe that the anisotropic CCPs would be highly promising barcodes in biomedical applications, including highthroughput bioassays and cell culture research where multiplexing is needed. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction The self-assembly of colloidal nanoparticles provides a simple and cheap approach to create three-dimensional structure materials [1–6]. The achieved colloidal crystal materials are with long-range ordered array of the nanoparticles and thus display a structure of periodic variation in the refractive index. The structure imparts the colloidal crystals an interesting optical property of photonic band gaps (PBGs), which is of great significance for photonic crystal devices [7–9], biological and chemical sensors [10–16], tunable lasers [17] and intelligent interfacial materials [18]. A major obstacle to produce the colloidal crystal materials is lack of reliable approaches to assemble nanoparticles into well controllable shapes. Virtually most colloidal-scale particles are naturally spherical because the interfacial tension between two immiscible phases tends to minimize the surface area, leading to the formation of spherical droplets. However, non-spherically anisotropic particles are also desired due to their great potential in biotechnology, structural materials, and other fields [19–22]. Traditional methods of fabricating non-spherically anisotropic particles include template molding [23], stretching of spherical particles [24,25], self-assembly of particles [26–28] and imprint ⇑ Corresponding authors. Fax: +86 25 83795635. E-mail addresses: [email protected] (Y. Zhao), [email protected] (Z. Gu). http://dx.doi.org/10.1016/j.jcis.2014.01.041 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
lithography technique [29]. It is usually difficult to use these methods to produce large quantities of monodisperse anisotropic particles of tunable geometries and adjustable sizes. As an alternative, microfluidic technique has emerged as a promising and versatile technique for generating non-spherically anisotropic particles due to their capabilities in manipulating fluids in controlled environments [30–35]. By confining the photo-precursor droplets in the microfluidic channels of different sizes and shapes, anisotropic particles can be generated in a flash of ultraviolet light [36]. In addition to droplet microfluidics, the anisotropic particles can also be produced by microfluidic lithography technology [37,38] which combines lithography and photo-chemistry into laminar flow microfluidics. Although many progresses have been achieved, it is still a challenge to fabricate CCPs with the feature of controllable anisotropic shapes and tunable optical properties. In this paper, we report a novel microfluidic approach to generate a series of anisotropic CCPs. A microfluidic device was assembled by aligning an injection capillary and a collection capillary coaxially inside a square capillary. When the size of droplets was larger than the inner diameter of collection capillary, they would be squeezed into non-spherical shape. Thus, by employing collection capillaries with different sizes and shapes, the monodisperse droplet templates with corresponding geometries appeared and flowed through the capillaries. In order to introduce optical properties into the droplet templates, the dispersed phase used for
65
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
droplet templates generation was the photo-precursor solution that contained non-close-packed colloidal nanoparticles array in the poly (ethylene glycol) diacrylate (PEG-DA) medium. By photo-polymerizing the droplet templates in situ of the confined collection capillaries, anisotropic particles with the same ordered colloidal crystal structures could be generated. These non-closepacked CCPs could be transformed into anisotropic close-packed CCPs by a simply thermal process. It was demonstrated that the anisotropic CCPs in different sizes, structural colors and shapes (rods, cuboids and disks) could be generated. As a typical application, the angle-independent rod-like CCPs were employed for multiplex biomolecule assay. Spectroscopic measurements and DNA hybridization experiments indicate that CCPs are highly practical for multiplex coding bioassays. We anticipate that this technology may find important application in multiplexed high-throughput bioassays 2. Materials and methods 2.1. Material The material used as outer phase was hexadecane (Sigma–Aldrich Co.). The inner phase was composed of a silica nanoparticle colloidal photonic crystal aqueous solution. To prepare the inner phase, monodisperse charged silica nanoparticles were first synthesized by the Stöber method [39]. The purified silica nanoparticles were dispersed in the pre-gel solution composed of 10% (v/v) poly (ethylene glycol) diacrylate (PEG-DA, MW 700, Sigma–Aldrich Co.) and 1% (v/v) photoinitiator (2-hydroxy-2-methylpropiophenone, Sigma–Aldrich Co.), and subsequently were shaken with an ion-exchange resin (Bio-Rad AG501-X8(D)) until strong diffraction became visually evident (20 min). After removing the resin by centrifugation, the colloidal crystal aqueous solution was achieved and sonicated for 30 min before use. 2.2. Methods 2.2.1. Microfluidics The capillary microfluidic device consisted of coaxially assembled glass capillaries with different cross-sectional shapes on glass slides. To construct the microfluidic device for fabricating rodshaped particles, a round capillary with an outer diameter/inner diameter of 1/0.58 mm (World Precision Instrument, Inc., Shanghai, China), tapered to achieve an orifice diameter of approximately 100 lm, was used for the inner tube. Another round capillary with an outer diameter of 1 mm and an inner diameter of 200 lm, was used as collection tube and coated with a hydrophobic reagent (trimethoxy (octadecyl) silane, Sigma–Aldrich Co.). The above capillaries were then coaxially assembled in a square capillary with a side length of 1.05 mm (AIT Glass, Rockaway, NJ, USA). A transparent epoxy resin was used to seal the tubes where required. To fabricate the cuboid-shaped particles, typically the orifice diameter of inner tube was adjusted to 200 lm, while the collection tube was changed into a square capillary with an inner diameter of 500 lm. For the disk-shaped particles, flat capillaries with an inner width/height of 1/0.2 mm, were used as inner tube and collection tube instead of round capillaries. By heating and pulling, the inner flat tube had a tapered tip with an inner width diameter of about 50 lm. All the collection tubes were treated with the hydrophobic reagent. In this work, inner phase and outer phase were pumped into the capillary microfluidic device using syringe pumps (PHD 2000 series, Harvard, Plymouth Meeting, PA, USA). By controlling the flow rates, droplet templates were stably generated in the collection tube and in situ photo-polymerized by UV-light (365 nm, 100w).
Finally, the particles were thoroughly washed with hexane, ethanol and pure water, respectively. 2.2.2. Thermal processing Representative non-close-packed colloidal crystal hydrogel particles were first dried at 65 °C for 6 h, and then sintered in the muffle at a rate of 5–800 °C for hold time of 3 h. After natural cooling to ambient temperature, the particles were taken out. 2.2.3. Biomolecule assay For further study of the anisotropic CCPs in biomolecule assays, oligonucleotides (Table 1) were bought from Invitrogen Biotechnology Co., Ltd. Hybridization buffer (750 mmol L1 NaCl, 150 mmol L1 sodium citrate, pH 7.4) and wash buffer (PBS, 10 mmol L1 phosphate sodium buffer solution, pH 7.4, 100 mmol L1 NaCl) were obtained from Shanghai Your Sun Biological Technology Co., Ltd. For the immobilization of probes, the particles were decorated with 5% (v/v) (3-Glycidyloxypropyl) trimethoxysilane (GPTMS) in toluene solution for 24 h, washed respectively by toluene, ethyl alcohol and pure water for three times, and then incubated with the amino-functionalized probe DNA solution (10 lM) at 4 °C overnight. In order to block residual reactive sites, the particles were immersed in 5% (w/v) bovine serum albumin (BSA) at room temperature for 1 h. All reagents were measured by 2 lL per bead. After being washed with wash buffer, the particles were used as carriers for DNA sequence detection in microwells. For multiplexed assays, rod-like particles in three colors were immobilized with oligonucleotide probes, named probe A, probe B, and probe C, correspondingly. The particles were incubated in hybridization buffer with 10 lM AMC-labeled target DNAs under continuous shaking at 37 °C for 1 h. Then, the particles were rinsed away three times with wash buffer in turns to remove unbound target DNA. Finally, the fluorescence spectra and images were taken after the reaction. 2.2.4. Characterization Photographs of photonic crystal hydrogel particles were taken by an optical microscope (OLYMPUS BX51) equipped with a cool CCD camera (Media-Cybernetics Evolution MP 5.0). Reflection spectra were recorded by an optical microscope equipped with a fiber optic spectrometer (Ocean Optics, USB2000-FLG). Fluorescence spectra were recorded by a microscope (OLYMPUS IX71) equipped with a fiber optic spectrometer (Ocean Optics, QE65000). The microstructures were characterized by a scanning electron microscope (SEM, HITACHI, S-3000N). 3. Results and discussion 3.1. Preparation of colloidal nanoparticles solution as dispersed phase To introduce optical properties into the anisotropic particles, we conducted the fabrication by using a PEG-DA aqueous solution containing 1% (v/v) photoinitiator and charged monodisperse silica nanoparticles as dispersed phase. When the concentration of
Table 1 Sequences of oligonucleotides used in this work. Base sequence
Name
50 -NH2-C6-GCG GCC TTA ATC ATT TCG CTT TCA GAA CTG-30 50 -NH2-C6-TGATCG CGG TGT CAG TTC TTT-30 50 -NH2-C6-GTGGAA TTG AGC AGC GTT GGT-30 50 -AMC-CAG TTC TGA AAG CGA AAT GAT GAA GGC CGC-30 50 -AMC-AAA GAA CTG ACA CCG CGA TCA-30 50 -AMC-ACC AAC GCT GCT CAA TTC CAC-30
Probe DNA A Probe DNA B Probe DNA C Target DNA A Target DNA B Target DNA C
66
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
charged silica nanoparticles reached a certain level, these nanoparticles tended to self-assemble into non-close-packed face-centered cubic (FCC) or body-centered cubic (BCC) ordered structures in solution owing to the interparticle repulsion and energy configuration. Thanks to the periodic arrangement of silica nanoparticles in the PEG-DA aqueous solution, the resultant mixture generated a PBG to prohibit the propagation of light with certain wavelength or frequency located in the PBG. As a result, the PEG-DA solution would exhibit vivid iridescent color and characteristic reflection peak. The position of main reflection peak could be estimated by Bragg’s equation:
k ¼ 2Dðn2eff cos2 hÞ
1=2
ð1Þ
where neff is the average refractive index of the solution, D is the diffracting plane spacing, and h is the Bragg angle of incidence of the light. When the incident light was perpendicular (h = 90°) to the (1 1 1) plane of the FCC structures, Eq. (1) could be simplified as k = 2Dneff. It is not difficult to find that the optical properties of the PEG-DA solution could be manipulated via changing either diffracting plane spacing or average refractive index of the solution. Therefore, the emerged PBG of the PEG-DA solution could be conveniently controlled by changing the size or concentration of silica nanoparticles that dispersed in the PEG-DA solution. It is worth pointing out that the concentration should be controlled within a proper region, since PEG-DA solutions containing silica nanoparticles with relatively high concentrations would result in the generation of droplets with poor monodispersity in the collection capillary due to high viscosity, while those with relatively low concentrations would affect the assembly of silica nanoparticles and thus quality of the products. In this case, we prepared the dispersed phases by dispersing charged silica nanoparticles with the diameter of 150 nm in the PEG-DA aqueous solution. As a result, PEG-DA solutions with controllable PBGs covering the whole visible range could be achieved by simply varying the concentrations of silica nanoparticles in the range from 10% to 30% (w/v), as shown in Fig. 1. 3.2. Microfluidic generation of anisotropic CCPs Anisotropic CCPs were evolved from the non-spherical colloidal crystal droplets that served as templates. Fig. 2 shows the schematic illustration of a typical glass capillary based microfluidic device for the fabrication of anisotropic particles with a series of geometries, which was assembled by coaxially aligning a cylindrical injection capillary and collection capillaries with various shapes
Fig. 1. (a) Digital photographs and (b) corresponding characteristic normalized reflection peaks of five kinds of colloidal crystal PEG-DA solutions. These solutions were prepared by dispersing charged silica nanoparticles with the size of 150 nm at different concentrations (from left to right: 27%, 23%, 21%, 19%, and 15% (w/v), respectively) in the PEG-DA aqueous solution.
Fig. 2. Schematic illustration of the microfluidic device used for the generation of anisotropic particles with various geometries. This device was assembled by a cylindrical injection capillary, a square capillary, and a collection capillary that aligned coaxially. The collection capillaries with various sizes and shapes were used to control the morphologies of generated template droplets in the channel. The resultant droplets were subsequently solidified in situ using UV irradiation.
inside a square capillary. The dispersed phase was pumped through the injection capillary, while the continuous phase flowed through the square capillary into the collection capillary in the same direction. In this case, the PEG-DA solution was continuously sheared by hexadecane at the tip of the injection capillary to form monodisperse W/O droplets. When the flow rate of hexadecane was not fast enough, PEG-DA solution would be enriched at the tip of injection capillary, eventually leading to the generation of non-spherical droplets in the confined collection capillary. After in situ solidification of these generated droplets with the UV irradiation, the CCPs with corresponding non-spherical shape were obtained immediately. It was more simple and efficient method for the CCPs generation than traditional evaporation induced crystallization which required impractically long times for complete consolidation and inevitably yields crystals with severe cracks. However, in the present method, the concentration of the nanoparticles should be well optimized to impart the non-spherical CCPs with vivid structural colors. The anisotropic morphology of resultant non-spherical CCPs was mainly determined by the droplets generated in microfluidic devices, which was closely related to the geometry of collection capillary. By controlling the size and shape of the collection capillary, non-spherical droplets thus CCPs could be conveniently fabricated. When a regular cylindrical capillary was applied as the collection capillary, rod-like droplets could be obtained. Replacing the collection capillary with square or flat capillary, cuboid-like or disk-like droplets would be formed in the device, respectively. We consider the formation of these non-spherical droplets should be attributed to the enrichment of aqueous droplets at the tip of the injection capillary owing to the relatively slow shearing, as well as the confinement of the collection capillary. Fig. 3 shows the representative digital photographs, SEM images and statistic geometric characterization of these three kinds of anisotropic CCPs. Thanks to the non-close-packed ordered structures formed via the self-assembly of silica nanoparticles on both surface and bulk of the anisotropic CCPs, they exhibited brilliant structural colors after the in situ solidification. However, the three kinds of resultant particles showed different optical properties when they were viewed at different directions and angles, as shown in Fig. 4. The cuboid-like and disk-like CCPs all displayed obvious spectra-shift no matter along the minor axes or major axes. As for the rod-like CCPs, the spectra showed similar shift when they were detected along the major axes, while the spectra showed unobservable shift along the minor axes. This angle-independent optical property should be due to the cylindrical symmetry of the rod-like particles, which endowed them with identical photonic responses independent of the rotation round the major axes. Thus, it is anticipated that the cod-like CCPs will be ideal materials for encoded multiplex bioassay, because the feature of angle-independent will make their decoding easy and accurate.
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
67
Fig. 3. (a–c) Optical image of anisotropic CCPs with various geometries: (a) rods, (b) cuboids, (c) disks; (d–f) their corresponding SEM images; and (g–i) their corresponding statistic geometric characterization along the minor axes and major axes respectively. Scale bars: (a, d) 200 lm, (b, c, e, f) 500 lm.
Besides the geometry of collection capillary, the relative flow rates between the dispersed phase and continuous phase would also affect the morphology of resultant non-spherical droplets and thus the CCPs. Therefore, we systematically investigated the effect of flow rates on the morphology of generated droplets by fixing the flow rate of one phase while varying the other. Take the rod-like particles generated in the cylindrical collection capillary for example, the outcomes of these trials are demonstrated in Fig. 5. When the flow rate of oil phase was fixed at 0.5 mL h1, the aspect ratio of resultant rod-like droplets dramatically increased with the increase in the flow rate of PEG-DA aqueous phase. However, when the flow rate of oil phase was boosted to 2 mL h1, no significant changes were found on the aspect ratio of resultant droplets even increasing the flow rate of the PEG-DA aqueous phase from 0.5 mL h1 to 5 mL h1 (Fig. 5a). On the other side, when the flow rate of PEG-DA aqueous phase was fixed at 1 mL h1 or 2 mL h1, the aspect ratio of obtained droplets decreased together with the increase in flow rate of oil phase (Fig. 5b). Hence, increasing the flow rate of PEG-DA aqueous phase and/or decreasing that of oil phase would be beneficial for the fabrication of anisotropic particles with larger aspect ratios. Moreover, by using PEG-DA solutions containing silica nanoparticles with different concentrations, an array of anisotropic CCPs with controllable aspect ratios and various structural colors could be achieved (Fig. 5c).
3.3. Transform non-close-packed structures into close-packed structures The non-close-packed lattice structures in the resultant anisotropic CCPs could be easily transformed into close-packed ordered structures via a thermal process. We conducted the calcination utilizing rod-like CCPs that assembled from the silica nanoparticles with the diameter of 210 nm in a muffle furnace at 800 °C for 3 h. Fig. 6 shows the digital photographs and corresponding SEM images of these particles before and after the calcination. The reflection peak of rod-like particles underwent a significant blueshift after the calcination, from the original position located at 630 nm to 470 nm, and the color of particles simultaneously changed from red to bluish violet after the thermal treatment (Fig. 6a and d). We consider that the shrink of non-close-packed ordered structures and reduction in average refractive index should be responsible for the blue-shift of reflection peak and structural color. During the calcination, PEG-DA hydrogels filled in the spaces among neighboring silica nanoparticles were gradually decomposed into gases. As a result, silica nanoparticles entrapped in the PEG-DA hydrogel were released from the hydrogel networks and driven to each other to form close-packed hexagonal ordered structures, eventually leading to the decrease in both inter-particle distance and average refractive index. SEM images taken from the surface and cross-section of the rod-like CCPs further confirmed
68
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
Fig. 4. Characteristic reflection peaks of three kinds of CCPs: (a, c, e) recorded at different angles along the major axes; (b, d, f) recorded at different angles along the minor axes.
Fig. 5. (a, b) Relationships between the aspect ratios of rod-like droplets and the flow rates that obtained by (a) fixing the flow rate of oil phase at 0.5 mL h1 and 2 mL h1 while varying that of PEG-DA aqueous phase, and (b) fixing the flow rate of PEG-DA aqueous phase at 2 mL h1 and 1 mL h1 while changing that of oil phase. (c) Anisotropic particles with various aspect ratios and structural colours that obtained by using PEG-DA solutions containing silica nanoparticles with different concentrations as the dispersed phases. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
our hypothesis. Compared with original structures of the particles, PEG-DA hydrogels could rarely be found on the surface and bulk of particles after calcination (Fig. 6b, c, e, and f). The calcinated particles still maintained bright structural colors and ordered lattice structures, indicating that close-packed anisotropic CCPs with high-quality could be obtained via the programmed thermal treatment. 3.4. Biomolecule assay As a typical application, the angle-independent rod-like CCPs were employed for biological multiplexed assays, in which a model DNA hybridization experiment was carried out. Three rod-like particles in blue, green and orange were immobilized with
different oligonucleotides, named probe A, probe B, and probe C, respectively. The probes were amine-functionalized oligonucleotides, whose corresponding targets were labeled with 7-amino-4methylcoumarin (AMC). All the particles were incubated in microwells each containing stochastic combination of target A, target B and target C. Due to the specific hybridization between the probe DNA and its corresponding fluorescently labeled target DNA, it was expected that the fluorescence signals would be detected only on the particles which their corresponding targets were present. Fig. 7 shows a typical result of the multiplex DNA detection in the solution of target A. It can be observed that only the CCPs in blue color exhibited fluorescence signal. The result was consistent with the content of the analyte solution to which the encoded rodlike particles were exposed. Thus, it is feasible to use the structural colors as the encoded elements of the particles for multiplexed assays. Moreover, as the particles could also be distinguished by their aspect ratios (as demonstrated in Fig. 5c), the identification codes of the particles could be further increased by using the joint coding of the structural colors and the aspect ratios.
4. Conclusions We fabricated a series of anisotropic CCPs by microfluidics. A microfluidic device was assembled by aligning a cylindrical injection capillary and a collection capillary with available size and shape coaxially inside a square capillary. Based on these devices, pre-gel droplet templates that containing non-close-packed colloidal nanoparticles array and with a series of geometries were generated and flowed through the collection capillaries. By polymerizing the droplet templates in the capillaries, the nonspherical particles with ordered colloidal crystal structures could be achieved. These non-close-packing colloidal crystal particles could also be transformed into close-packing colloidal crystal particles with the corresponding geometries by a simply thermal processing. The achieved particles were with simple identification from their sizes, structural colors or shapes. In order to demonstrate the feasibility of anisotropic CCPs in the biomolecule assay, multiplex DNA detection was carried out. The experiments show
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
69
Fig. 6. (a) Optical image of rod-like particles before calcination; SEM images taken from (b) surface, and (c) cross-section. (d) Optical image, and (e, f) SEM images after calcining at 800 °C for 3 h. Scale bars: (a, d) 200 lm, (b, c, e, f) 200 nm.
Fig. 7. Optical image (a) and fluorescence microscope images (b) of rod-shaped particles after DNA hybridization. Scale bar indicates 200 lm.
that particles are highly practical for multiplex coding bioassays. This microfluidic approach provides a novel method to create anisotropic CCPs and will be of greatly important in potential application for colloidal science, microfluidics, optical materials and biosensors, as well as further understanding transport and dynamic behavior of anisotropic CCPs with non-edged shape (e.g. rod and disk) or sharp-edged shape (cuboids) in many fields of application. Acknowledgments We are grateful to the support of the National Science Foundation of China (Grant Nos. 21105011, 51073034 and 50925309), the National Science Foundation of Jiangsu (Grant No. BK2012735), the Science and Technology Development Program of Suzhou (Grant No. ZXG2012021), the Postdoctoral Science Foundation of China (Grants 2012M520052 and 2013T60484), the Program for Changjiang Scholars and Innovative Research Team in University (IRT1222), and the 333 Talent Project Foundation. References [1] J. Ge, Y. Yin, Angew. Chem. Int. Ed. 50 (2011) 1492–1522. [2] Y.J. Zhao, Z.Y. Xie, H.C. Gu, C. Zhu, Z.Z. Gu, Chem. Soc. Rev. 41 (2012) 3297– 3317. [3] Z.W. Mao, H.L. Xu, D.Y. Wang, Adv. Funct. Mater. 20 (2010) 1053–1074. [4] O.D. Velev, S. Gupta, Adv. Mater. 21 (2009) 1897–1905. [5] G.V. Freymann, V. Kitaev, B.V. Lotsch, G.A. Ozin, Chem. Soc. Rev. 42 (2013) 2528–2554. [6] D. Xu, W. Zhu, Q. An, W.N. Li, X.S. Li, H.W. Yang, J.X. Yin, G.T. Li, Chem. Commun. 48 (2012) 3494–3496. [7] M.A. Haque, G. Kamita, T. Kurokawa, K. Tsujii, J.P. Gong, Adv. Mater. 22 (2010) 5110–5114.
[8] S.H. Kim, H. Hwang, S.M. Yang, Angew. Chem. Int. Ed. 51 (2012) 3601–3605. [9] X.B. Hu, Q. An, G.T. Li, S.Y. Tao, J. Liu, Angew. Chem. Int. Ed. 45 (2006) 8145– 8148. [10] Y.J. Zhao, X.W. Zhao, J. Hu, J. Li, W.Y. Xu, Z.Z. Gu, Angew. Chem. Int. Ed. 48 (2009) 7350–7352. [11] L. Zhang, C.L. Lu, Y.F. Li, Z. Lin, Z.H. Wang, H.P. Dong, T.Q. Wang, X.M. Zhang, X. Li, J.H. Zhang, B. Yang, J. Colloid Interface Sci. 374 (2012) 89–95. [12] X.S. Li, L.H. Peng, J.C. Cui, W.N. Li, C.X. Lin, D. Xu, T. Tian, G.X. Zhang, D.Q. Zhang, G.T. Li, Small 8 (2012) 612–618. [13] L.R. Shang, F.Q. Shangguan, Y. Cheng, J. Lu, Z.Y. Xie, Y.J. Zhao, Z.Z. Gu, Nanoscale 5 (2013) 9553–9557. [14] J.T. Bacaa, D.N. Finegoldb, S.A. Asher, Analyst 133 (2008) 385–390. [15] D. Nakayama, Y. Takeoka, M. Watanabe, K. Kataoka, Angew. Chem. Int. Ed. 42 (2003) 4197–4200. [16] H. Li, J.X. Wang, F. Liu, Y.L. Song, R.M. Wang, J. Colloid Interface Sci. 356 (2011) 63–68. [17] R.L. Justin, Y.R. Ying, P. Jiang, H.F. Stephen, Adv. Mater. 18 (2006) 300–303. [18] Y.J. Zhao, H.C. Gu, Z.Y. Xie, H.C. Shum, B.P. Wang, Z.Z. Gu, J. Am. Chem. Soc. 135 (2013) 54–57. [19] Y.J. Zhao, Z.Y. Xie, H.C. Gu, L. Jin, X.W. Zhao, B.P. Wang, Z.Z. Gu, NPG Asia Mater. 4 (2012) e25. [20] A.H.L. Chow, H.H.Y. Tong, P. Chattopadhyay, B.Y. Shekunov, Pharm. Res. 24 (2007) 411–437. [21] S.C. Glotzer, M.J. Solomon, Nat. Mater. 6 (2007) 557–562. [22] S. Elingarami, X.L. Li, N.Y. He, J. Nanosci. Nanotechnol. 13 (2013) 4539–4551. [23] B.F. Ye, Y.J. Zhao, T.T. Li, Z.Y. Xie, Z.Z. Gu, J. Mater. Chem. 21 (2011) 18659– 18664. [24] P.P. Lele, E.M. Furst, Langmuir 25 (2009) 8875–8878. [25] J.A. Champion, S. Mitragotri, Proc. Natl. Acad. Sci. 103 (2006) 4930–4934. [26] Z.Y. Yu, C.F. Wang, L.T. Ling, L. Chen, S. Chen, Angew. Chem. Int. Ed. 51 (2012) 2375–2378. [27] J.R. Millman, K.H. Bhatt, B.G. Prevo, O.D. Velev, Nat. Mater. 4 (2005) 98–102. [28] D.J. Kraft, W.S. Vlug, C.M. van Kats, A. van Blaaderen, A. Imhof, W.K. Kegel, J. Am. Chem. Soc. 131 (2009) 1182–1186. [29] J.Y. Wang, Y.P. Wang, S.S. Sheiko, D.E. Betts, J.M. DeSimone, J. Am. Chem. Soc. 134 (2012) 5801–5806. [30] C.H. Chen, A.R. Abate, D. Lee, E.M. Terentjev, D.A. Weitz, Adv. Mater. 21 (2009) 3201–3204. [31] T. Hutter, W.A.C. Bauer, S.R. Elliott, W.T.S. Huck, Adv. Funct. Mater. 22 (2012) 2624–2631.
70
Y. Cheng et al. / Journal of Colloid and Interface Science 421 (2014) 64–70
[32] A.R. Studart, H.C. Shum, D.A. Weitz, J. Phys. Chem. B 113 (2009) 3914–3919. [33] H.C. Shum, A.R. Abate, D. Lee, A.R. Studart, B. Wang, C.H. Chen, J. Thiele, R.K. Shah, D.A. Weitz, Macromol. Rapid Commun. 31 (2010) 108–118. [34] E. Kumacheva, P. Garstecki, 10. Microfluidic Synthesis of Polymer Particles with Non-Conventional Shapes, Microfluidic Reactors for Polymer Particles, John Wiley & Sons, Ltd., Chichester, UK, 2011. [35] S. Seiffert, Macromol. Rapid Commun. 33 (2012) 1286–1293.
[36] S.Q. Xu, Z.H. Nie, M. Seo, P. Lewis, E. Kumacheva, H.A. Stone, P. Garstecki, D.B. Weibel, I. Gitlin, G.M. Whitesides, Angew. Chem. Int. Ed. 44 (2005) 734–738. [37] S.K. Suh, K. Yuet, D.K. Hwang, K.W. Bong, P.S. Doyle, T.A. Hatton, J. Am. Chem. Soc. 134 (2012) 7337–7343. [38] D. Dendukuri, D.C. Pregibon, J. Collins, T.A. Hatton, P.S. Doyle, Nat. Mater. 5 (2006) 365–369. [39] W. Stober, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1968) 62–69.
