Three-Dimensional Clustered Nanostructures for ... - ACS Publications

Sep 1, 2016 - Rhodamine 6G (R6G) and 4-MBA as target organic pollutants to evaluate the ... KEYWORDS: microfluidic technology, fluid construction, SER...
0 downloads 7 Views 6MB Size
Download PDF
Research Article www.acsami.org

Three-Dimensional Clustered Nanostructures for Microfluidic Surface-Enhanced Raman Detection Gang Wang,†,‡ Kerui Li,† Francis J. Purcell, De Zhao,† Wei Zhang,† Zhongyuan He,† Shuai Tan,‡ Zhenguan Tang,‡ Hongzhi Wang,*,† and Elsa Reichmanis*,‡ †

State Key Laboratory for Modification of Chemical Fibers and Polymer Materials, College of Materials Science and Engineering, Donghua University, Shanghai 201602, People’s Republic of China ‡ School of Chemical and Biomolecular Engineering, School of Chemistry and Biochemistry, School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States S Supporting Information *

ABSTRACT: A materials fabrication concept based on a fluidconstruction strategy to create three-dimensional (3D) ZnO@ ZnS-Ag active nanostructures at arbitrary position within confined microchannels to form an integrated microfluidic surfaceenhanced Raman spectroscopy (SERS) system is presented. The fluid-construction process allowed facile construction of the nanostructured substrates, which were shown to possess a substantial number of integrated hot spots that support SERS activity. Finite-difference time-domain (FDTD) analysis suggested that the 3D clustered geometry facilitated hot spot formation. High sensitivity and good recycle performance were demonstrated using 4-mercaptobenzoic acid (4-MBA) and a mixture of Rhodamine 6G (R6G) and 4-MBA as target organic pollutants to evaluate the SERS microfluidic device performance. The 3D clustered nanostructures were also effective in the detection of a representative nerve agent and biomolecule. The results of this investigation provide a materials and process approach to the fabrication of requisite nanostructures for the online detection of organic pollutants, devices for real-time observation of environmental hazards, and personal-health monitoring. KEYWORDS: microfluidic technology, fluid construction, SERS detection, clustered nanostructures, biomolecule detection

■

INTRODUCTION Three-dimensional (3D) nanostructured materials have significant potential in catalytic processes, detection of environmental pollutants and biomolecular species, and energy applications.1,2 Such materials are often used in electronic and photonic applications; they also facilitate surface-enhanced Raman spectroscopy (SERS) detection of many analytes.3,4 Recent synthetic advances allow access to a wide range of 3D morphologies that can be tuned for desired performance attributes.5,6 For example, 3D Ag-ZnO nanostructures have been fabricated via optothermal effects for online biodetection,5 and 3D gold nanofingers have been produced by imprinting strategies and applied to molecular trapping and detection.6 However, the application space for integrated 3D nanoscale materials is limited because of high fabrication costs associated with incorporation into micro/nanodevices.7,8 Thus, alternative, low-cost construction strategies are desirable in order to fully exploit the unique properties of 3D nanostructured materials and expand the application space. One application, vitally important in diverse fields, is the label-free detection of analytes from small-volume, highly diluted, multicomponent samples.9,10 For instance, trace © XXXX American Chemical Society

detection of toxic organic molecules is significant in monitoring environmental pollutants;11 and in cancer diagnostics, probing single-stranded DNA and its hybridization process is beneficial in the design of DNA markers.12 An alternative promising approach to trace detection involves microfluidics,13 where small volumes (e.g., micro/nanoliter scale) of fluids can be readily confined and liquids, suspended particles, and biological samples can be easily manipulated.14,15 Currently, many such microanalyses use glass capillaries because of their low-cost, facile fabrication, and ability to limit leakage of fluids.16,17 Because of capillary forces, many solutions can readily flow into a capillary, affording a “self-filling” mechanism.18 Further, spectroscopic analysis techniques are appealing because they offer the possibility for direct, sensitive, nondestructive, labelfree, and real-time collection of quantitative chemical information.19−22 SERS, where optical scattering and light collection occur within a 3D focal volume, is particularly attractive. Compared with conventional two-dimensional SERS Received: August 22, 2016 Accepted: September 1, 2016

A

DOI: 10.1021/acsami.6b10542 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

by two pinheads to a “Y” pattern tube made of polytetrafluoroethylene (PTFE). The “Y” pattern PTFE tubes were connected to the capillary microchannel with epoxy resin (E44, Shanghai Resin Co., LTD) in advance. Preparation of ZnO@ZnS Clustered Nanostructures. ZnO nanorod arrays were prepared on the inner wall of the capillaries as previously reported.34 The ZnO surfaces were then exposed to thioacetamide (TAA) by pumping 0.2 M aqueous TAA from the two syringes into a capillary microchannel containing the prefabricated ZnO nanorod arrays at 25 μL/min in an oven at 90 °C. Varying the TAA flow duration from 3 to 30 h yielded capillaries with ZnO@ZnS clustered nanostructures (denoted as ZnO@ZnS-x, where x is the flow duration). The final products were washed with water and dried at 70 °C before characterization and/or subsequent reaction. Preparation of ZnO@ZnS-Ag Clustered Nanostructures. Sodium thioglycollate (ST)-capped ZnO@ZnS clustered nanostructures were prepared on the inner capillary walls, by pumping 0.2 M ST from two syringes into the capillary microchannel containing ZnO@ ZnS clustered nanostructures, at 25 μL/min for 2 h, followed by washing with deionized water and drying at 100 °C. Colloidal Ag was prepared by reducing AgNO3 with NaBH4. The Ag colloid was then introduced into both syringes and pumped into the microchannel containing the ST-capped ZnO@ZnS clustered nanostructures at 10 μL/min for 1 h, after which the capillary was dried at 60 °C. Characterization of Various Nanostructures on the Inner Wall of Capillary Microchannels. Transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) images were obtained using a JEM-2100F microscope (JEOL, Ltd., Japan), operating at 200 kV. Samples were prepared by scraping nanostructures from the inner capillary walls on which they had been grown. Nanostructure morphologies were characterized using field emission-scanning electron microscopic (FE-SEM) images obtained on an S-4800 FE-SEM microscope (Hitachi, Ltd., Japan) and an Ultra-60 FE-SEM microscope (Carl Zeiss AG, Germany). Energy dispersive X-ray (EDX) microanalyses were also recorded by the FE-SEM apparatus. Samples were prepared by cutting subject capillaries into small pieces. Photocatalytic and SERS Detection of R6G, 4-MBA, and Mixtures Thereof. The outer polymer coating was removed from the capillaries by heating. They were then used as microreactors to photodegrade R6G. The capillary microchannels were oriented parallel under a 200 W mercury lamp, with a central wavelength of 365 nm. Aqueous R6G (5 mg/L) was pumped into the microchannel and collected in a dark container. The residence time (RT) in the modified channels was calculated by

substrates, 3D geometries have exhibited notably improved SERS detection performance.5,23,24 To maximize the quantity of generated and detected light, SERS substrates should contain so-called “hot spots” in a large 3D volume.17,25 The integration of 3D active SERS substrates into microfluidic devices could combine the respective advantages of both, thereby offering an attractive approach for many chemical and biological sensing applications.26−29 While some steps in that direction have been reported,30,31 microstructure fabrication and SERS sensitivity remain problematic, limiting widespread adoption. Fabrication typically requires access to expensive lithographic or thermal evaporation facilities, and process steps can be complex.32,33 It is almost impossible to incorporate requisite 3D nanostructures with appropriately controlled gap dimensions into confined microchannels. To enable low-cost, accessible microfluidic SERS detection of a range of analytes, new approaches to the fabrication of requisite materials and devices are required. Here, we present a concept based on a fluid-construction strategy to create 3D ZnO@ZnS-Ag SERS active nanostructures at arbitrary position inside a confined microchannel. This approach offers several attractive attributes: (1) facile, low cost fabrication; (2) 3D nanoclustered morphologies; and (3) high SERS sensitivity owing to large effective surface area that can contribute to SERS enhancement. ZnO nanorods were used to template the growth of ZnO@ ZnS core/shell clustered nanostructures. The in situ chemical conversion process afforded clustered assemblies on the inner wall of capillary microchannels. Ag nanoparticles were then loaded onto the structures in a flow process to afford ZnO@ ZnS-Ag clustered nanostructures. The method developed here obviates the need and expense of microlithographic processing or micromachining. Finite-difference time-domain (FDTD) calculations provided insight into the SERS mechanism. Both 4mercaptobenzoic acid (4-MBA) and a mixture of Rhodamine 6G (R6G) and 4-MBA were used as target organic pollutants to evaluate SERS microfluidic device performance; high sensitivity and good recycle performance were achieved. The SERS enhancement factor for optimized 3D ZnO@ZnS-Ag nanostructures was estimated to be about 2 × 109. The 3D clustered nanostructures were also effective in the detection of a representative nerve agent and biomolecule. The results of this investigation provide a materials and process approach to the fabrication of requisite nanostructures for the online detection of organic pollutants, devices for real-time observation of environmental hazards, and personal-health monitoring.

■

RT = VCM /v

(I)

where VCM is the total internal volume of the microchannel and v is the R6G flow rate. Different RTs were achieved by adjusting the flow rate. To evaluate the continuous photocatalytic performance of the capillary microchannel recycled at certain RTs, samples of the same RT were removed after a fixed time interval. The R6G solution concentration was monitored by measuring the characteristic R6G 526 nm absorption band with a UV−vis absorption spectrophotometer (Agilent 8510, USA). The detection capability of the modified capillary microchannels was tested by introducing target molecular solution with various concentrations into the microchannels containing ZnO@ZnS-Ag clustered nanostructures and then analyzing the microchannel by SERS. A Raman spectrometer (RXN1, Kaiser Optical Systems, Inc., Ann Arbor, MI) equipped with a 785 nm laser source was used to obtain Raman spectra. The laser power was set at 15 mW for the SERS detection, and Raman spectra were collected from seven random points on one device. SERS Detection of Nerve Agent Simulant and BSA Molecules. In addition to the detection of organic pollutants, another important application is nerve agent and biomolecule (e.g., protein) detection. To explore this facet, the ability of the ZnO@ZnS-Ag clustered nanorod array to detect/diagnose the presence of bovine serum albumin (BSA) was investigated. MP (a nerve agent simulant, 10−8 M) and BSA (10−8 M) were used to evaluate the SERS

EXPERIMENTAL SECTION

R e agen t s and M a ter ia ls. Z n( C H 3 C O O ) 2 , N a OH , hexamethylenetetramine, Zn(NO3)2, sodium citrate, and silver nitrate were purchased from Alfa Aesar Co., Inc. (analytically pure), and used without further purification. R6G, 4-MBA, and methylparathion (MP) were purchased from Tokyo Chemical Industry Co., Ltd. In all cases, water was produced by distillation followed by deionization to a resistance of 18 MΩ cm (Barnstead E-Pure system). Glass capillaries with an internal diameter of 530 μm and a polyimide outer coating were purchased from Nanjing Jianuo Apparatus (China) and used as the microchannel. The inner surface of the capillary was treated by previously reported methods, which are briefly described here.34 Capillary microchannels were sequentially rinsed with concentrated sulfuric acid/hydrogen peroxide/water (v/v/v: 4/1/20; 130 °C), water, ammonia/hydrogen peroxide/water (v/v/v: 1/4/20; 70 °C), followed by water, and were then dried in an oven at 100 °C. Two reaction solutions were filled individually into two syringes (10 mL) which were attached to a dual-syringe infusion pump and connected B

DOI: 10.1021/acsami.6b10542 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces performance of the 3D ZnO@ZnS-Ag clustered nanostructures. The analytes with desired concentrations were injected into the channel composed of 3D ZnO@ZnS-Ag nanostructures, keeping them in a liquid state and a static flow condition. The Raman spectra were recorded with a confocal microprobe Raman system with the excitation wavelength of 785 nm; the laser power was about 15 mW. The accumulation time was 30 s. SERS measurements were repeated five times on the nanostructures and then averaged. Simulation of Electromagnetic Field Enhancement. The FDTD method was used to simulate the electromagnetic field enhancement of ZnO@ZnS-Ag nanostructures and 3D cluster nanostructures. The ZnO@ZnS-Ag nanostructures were simplified as ZnS cylinders covered by randomly distributed Ag nanospheres. ZnS nanorods with a diameter of 120 nm and length of 1500 nm, revealed by SEM observation, were used in the model for the FDTD simulations. Silver nanospheres with a diameter ca. 20 nm were randomly distributed on the ZnS nanorods to imitate the nanostructures and 3D clustered structures. Two identical ZnS nanorods decorated with Ag nanoparticles were separated by a distance of 40 nm to investigate the influence of the internanorod distance on the electromagnetic field distributions.

Scheme 1. Schematic Representation of the Fabrication Process and Reaction Sequence for the Preparation of 3D Clustered ZnO@ZnS-Ag Nanostructures: (a) 3D Network Scheme in the Confined Microchannels; (b) Involve Process of the ZnO@ZnS-Ag Nanostructures

■

RESULTS AND DISCUSSION ZnO Nanorod Arrays and ZnO@ZnS Clustered Nanostructures. Uniform ZnO nanorod arrays, oriented approximately perpendicular to the inner capillary wall, were fabricated by continuously pumping the reactive hexamethylenetetramine and Zn(NO3)2 precursor solutions through a 12 cm long capillary as reported previously.34 The successful growth of nanorod assemblies was confirmed using FE-SEM imaging (Figure 1). The apparently smooth nanorods were 100−150 nm wide and 1−2 μm long.

nanoparticles onto the clustered nanorods via SH moieties. When optimized, the original 3D clustered nanostructure can be retained in the final Ag-decorated network (ZnO@ZnS-Ag). As shown in Figure 2, the surface coverage and resultant ZnO@ZnS nanostructure morphology depended upon TAA

Figure 1. Typical FE-SEM images of pristine ZnO nanorod arrays on an inner capillary wall: (a) top and (b) tilted views.

The steps associated with the fabrication of the multilayered 3D clustered ZnO@ZnS-Ag nanostructures are presented in Scheme 1. The as-prepared ZnO nanorod arrays served as an in situ template and source of zinc for the subsequent ZnS growth. TAA solution was passed through the channel at 90 °C and reacted with H2O to form S2−, which then reacted with Zn2+ residing at the ZnO network surface. Mechanistically, TAA lone pair electrons are available to interact with empty Zn2+orbitals to form ZnHS+ coordination compounds. ZnHS+, in turn, is expected to enhance the activity of Zn2+ within the parent, facilitating nucleation and growth of ZnS on the surface. As the ZnS layer is formed, neighboring nanorods interact to form a clustered structure. The chemical transformations involved in fabrication of the ZnO@ZnS clustered nanostructures are detailed below: Zn 2 + + CH3CSNH 2 ↔ ZnHS+ + CH3CSNH+

(1)

2ZnHS+ + S2 − ↔ 2ZnS + H 2S

(2)

Figure 2. FE-SEM images of ZnO@ZnS nanostructures of (a) ZnO@ ZnS-3 h, (b) ZnO@ZnS-6 h, (c) ZnO@ZnS-8 h, (d) ZnO@ZnS-10 h, (e) ZnO@ZnS-20 h, and (f) ZnO@ZnS-30 h. The scale bar is 1 μm.

treatment time. In each case, the structures exhibited larger diameters and rougher surfaces than the parent template (Figure 1). Exposure to TAA for 3 h was insufficient to effect complete coverage of the oxide; the nanorod top surfaces did not visibly coat with ZnS (Figure 2a). After 6 h, ZnS appeared to cover the entire nanorod surface, the morphology was retained, and the diameter increased (Figure 2b). During the course of the in situ conversion process, ZnS-coated nanorods appeared to aggregate and seemingly “connect” with nearby structures to form clustered assemblies. Increased TAA exposure time led to progressively rougher nanorod surfaces, with concomitant decreased inter-rod spacing (Figure 2c,d). Extended exposure to the reagent led to significantly decreased surface area and near elimination of the rodlike morphology

In situ chemical conversion is required to develop functional ZnS surfaces that integrate function with structural and chemical stability. After formation of ZnO@ZnS, introduction of sodium thioglycolate (ST) facilitated binding of Ag C

DOI: 10.1021/acsami.6b10542 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces (Figure 2e,f). The optimum TAA exposure time to fabricate well-formed ZnO@ZnS clustered nanostructures on the capillary inner wall was judged to be 6 h. In the reaction system, TAA, which serves as the S source, hydrolyzes at 90 °C and then reacts with ZnO at the surface of the nanowires to form ZnS. If the sulfidation time was very short, some ZnS nanoparticles were observed on the ZnO nanostructures because of ion exchange that occurred as S2− reacted with Zn2+ that slowly dissolved from the ZnO surface to form initial ZnS shells. With increased sulfidation time, ZnO will be dissolved from beneath the surface and Zn2+ will react with the S2− present on the initial ZnS shells. ZnS shells become thicker as more ZnS nanoparticles form and deposit on the initial ZnS shells. Nanorod diameter will thereby be increased significantly more than would be expected by the simple replacement of O with S within the parent lattice. FE-SEM images of ZnO@ZnS after 6 h of exposure to TAA (ZnO@ZnS-6h) are shown in Figure 3a. Nanorods with

Figure 4. (a) FE-SEM, (b) TEM, (c) HRTEM, and (d) EDX images of ZnO@ZnS-Ag-6 h.

were ∼0.31 and ∼0.23 nm, which correspond to lattice spacings for the (002) wurtzite ZnS face and (111) cubic Ag face, respectively. The EDX spectrum confirmed the presence of Zn, O, Ag, and S, consistent with the presence of ZnO, ZnS, and Ag (Figure 4d). Because of the dimensions of the microchannel, it is almost impossible to directly perform XRD characterization. It is also impossible to scratch the microstructures from the inner wall, owing to the interactions and integrity between the in situ fabricated microstructures and inner wall. In future work, synchrotron radiation will be utilized to gain more detailed insight into the crystallization process. The microfluidic materials growth process developed here to prepare ZnO@ZnS-Ag nanorod assemblies afforded structures with a range dimensions that depended upon reagent residence time and could be optimized to attain plasma structures which exhibited maximum electric field and thus SERS effectiveness. The EF obtained from arrays prepared with TAA residence times of 3, 6, and 8 h were determined as shown in Figure 5.

Figure 3. (a) FE-SEM, (b) TEM, (c) HRTEM, and (d) EDX spectrum of ZnO@ZnS-6 h.

wrinkled surfaces and a diameter of 100−200 nm were arranged in an array. Differences in contrast along all ZnO@ZnS-6 h nanorods were seen upon TEM analysis (Figure 3b). Dark inner centers and lighter edges indicated a core/shell structure. HRTEM imaging of the core/shell boundary revealed a polycrystalline shell and single-crystalline core (Figure 3c). The calculated interplanar spacings of ∼0.31 and ∼0.26 nm corresponded to the lattice spacings of the (002) faces of wurtzite ZnS and ZnO, respectively.35 The EDX spectrum in Figure 3d confirmed nanorod composition, namely Zn, O, and S, as expected for ZnO and ZnS. The TEM and EDX results were consistent with ZnO@ZnS core/shell nanorods. ZnO@ZnS-Ag Clustered Nanostructures. ZnO@ZnS-Ag clustered nanostructures were produced by pumping freshly prepared colloidal Ag into the modified microchannels. It is known that thiols, alkanethiols, or sulfhydryl compounds are usually absorbed on Ag surfaces due to the easily formed Ag−S bond.36 As a sulfhydryl compound, ST should also be an efficient Ag capping agent. Thus, Ag-loaded ST-capped ZnO@ ZnS was produced by driving freshly prepared colloidal Ag through microchannels with ST-capped ZnO@ZnS nanostructures. A FE-SEM image of the ZnO@ZnS-Ag product is shown in Figure 4a, demonstrating that the pristine nanorod array structures were maintained. TEM imaging of a single ZnO@ ZnS-Ag nanorod revealed the presence of a core/shell structured nanorod with dark dots indicative of Ag nanoparticles. HRTEM analysis showed that the shell was polycrystalline (Figure 4c). The calculated interplanar spacings

Figure 5. Enhancement factor (EF) data calculated from ZnO@ZnSAg clustered nanorod arrays with different TAA loading time.

ZnO@ZnS-Ag-6 h structures retained distinct nanorod features, exhibited complete coverage of the template with ZnS, small gap size, and exhibited the largest EF. The small gap size allows for more “hot spots” with enhanced electric field, which leads to a larger EF. With extended in situ conversion time, the ZnO@ZnS nanorods grew and became connected, thereby eliminating some fraction of gaps with concomitant D

DOI: 10.1021/acsami.6b10542 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces decreased EF. The in situ conversion time is readily tuned in order to optimize the clustered nanorod array geometry to achieve the highest SERS sensitivity. In addition, the control experiment with various Ag loading time can be found in Figure S1, which suggested the importance of Ag amounts. To further investigate the SERS effect, the intensity (|E|2) distributions were simulated by FDTD calculations at the excitation wavelength of 785 nm, as shown in Figure 6. The

previous work, the dimensions of the metal nanoparticles ca. 10−20 nm also exhibited high SERS detection performance.5 The analysis showed obvious electromagnetic field enhancement from the clustered array. The narrow gaps provided hot spots in field intensity, which are expected to facilitate strong SERS signal enhancement. Examination of Figure 6c, where the observed clustered nanostructures are outlined with red ovals, reveals that several adjacent rods appear to lean toward each other to form clustered structures. These clusters are likely to produce a large number of “hot spots”. An increased number of “hot spots” is the key to high SERS sensitivity. The 3D structure of the ZnO@ZnS-Ag clusters induced numerous SERS “hot spots” located between neighboring Ag nanoparticles on both the same and/or adjacent nanorods, which suggests that the ZnO@ZnS-Ag clustered nanostructures fabricated here will be good candidates for SERS sensing applications. Owing to the increased surface area and level of Ag nanoparticle loading, enhanced photocatalytic performance can also be anticipated. SERS spectra collected from solutions of 4-MBA (1 × 10−7− 1 × 10−9 M) held within the modified microchannels are presented in Figure 7a. The strong bands at approximately 1594 and 1078 cm−1 are assigned to the ν8a (a1) and ν12 (a1) characteristic aromatic ring breathing vibrations, respectively. The less intense peak at 1420 cm−1 was attributed to the ν (COO−) stretching mode. Other weak bands at 1138 cm−1 (ν15, b2) and 1184 cm−1 (ν9, a1) correspond to C−H deformation modes. These results are consistent with reported data27 and confirmed the ZnO@ZnS-Ag clustered nanostructure effectiveness as a highly sensitive SERS sensor. The results described above suggest that upon incorporation into a continuous flow system, the modified microchannel has potential to function as an in-line, real-time sensor for the detection of organic pollutants. Invariably, such functionality will require the ability to identify individual species from mixtures of compounds. Thus, identification of individual components within a mixture via dynamic detection is critical. To investigate the capability of the ZnO@ZnS-Ag structure to interrogate multicomponent systems, mixtures of R6G and 4MBA were prepared in aqueous solution. Figure 7b presents SERS spectra obtained from the individual probe molecules and a mixture of the two components. Each substance is readily discerned in the mixed analyte solution, suggesting that the capillary-based microfluidic sensor is a versatile platform for the detection of molecular materials in the presence of other species. Using the common method for calculating the EF, a

Figure 6. Electromagnetic field enhancement of ZnO@ZnS-Ag nanostructures using FDTD simulations: (a) nanorod array structure; (b) clustered nanorod structure. (c) Schematic diagrams of the SERS mechanism of the clustered nanostructures. The insets of the FE-SEM images (i, ii) are the corresponding structures for the modeling units.

SERS signal is approximately proportional to the square of the electric field at the excitation wavelength. It is speculated that SERS “hot spots” are generated between neighboring Ag nanoparticles on the same or clustered nanorods. Figure 6a depicts the distribution of the electric field corresponding to the vertical nanostructures. The distribution of the electric field generated by the clusters is depicted in Figure 6b. The corresponding FE-SEM images for the model units are depicted in the insets (i, ii). The size of the Ag nanoparticles was ca. 10− 20 nm, which can be identified in the TEM images. From the

Figure 7. (a) Raman spectra of 4-MBA absorbed onto a ZnO@ZnS-Ag clustered nanostructure modified microchannel from solutions with concentrations ranging from 1 × 10−7 to 1 × 10−9 M. (b) Raman spectra of R6G, 4-MBA, and a mixture of R6G and 4-MBA. The integration time of the spectra is 60 s. E

DOI: 10.1021/acsami.6b10542 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces conservative value of EF was estimated to be 5 × 109 (a more detailed description can be found in Figure S2). The ability to reuse a SERS substrate is attractive for many low-cost applications. Thus, the effectiveness of the ZnO@ZnSAg system for analyte detection after recycling was explored. Aqueous R6G at a concentration of 10−9 M was used as a representative analyte. After the initial SERS measurement, the microchannel was cleaned with deionized water and simultaneously subjected to UV irradiation for 1 min at a water flow rate of 25 μL/min. The samples were then rinsed with water and dried. In total, the cycle was performed 10 times. Figure 8

Figure 9. (a) Photograph of self-filling performance of microfluidic SERS sensor. (b) Microfluidic SERS detection of MP (nerve agent simulant) in ethanol. The integration time of the spectra is 30 s. (c) Microfluidic SERS detection of BSA in water. The concentration is 10−8 M, and the integration time of the spectra is 30 s.

nanoparticles.37−39 The Raman band assignments are discussed in detail (see Table S1) in the Supporting Information. Moreover, bovine serum albumin (BSA) was used to further demonstrate the detection sensitivity of the 3D clustered substrates. BSA is a common protein for biochemical applications such as enzyme-linked immunosorbent assays, immunoblots, and immunohistochemistry. The BSA spectrum obtained from a 10−8 M aqueous solution is presented in Figure 9c; the BSA SERS peak positions are in good agreement with literature results.40 Briefly, the 845 cm−1 SERS peak is associated with the tyrosine group; the peaks located at 940 cm−1 arise from the skeletal stretching of α-helix; the peak at 1041 cm−1 originates from the Phe amino acids; the C−N bond vibration in the protein structure gave rise to the appearance of the 1112 cm−1 peak; the peak at 1265 cm−1 is originated from amide III; and the peaks at 1362 and 1434 cm−1 can be assigned to the vibrations of the CH, CH2, and CH3 groups in the proteins. The observed band positions are slightly different from those reported in the literature, which indicates that there is a slight change in the secondary structure of BSA because of the different interactions between BSA molecules and ZnO@ ZnS-Ag 3D clustered nanostructures.

Figure 8. Recycling performance for SERS detection of R6G after cycling up to 10 times. The integration time of the spectra is 60 s.

depicts the original R6G Raman spectrum along with those obtained after the 5th and 10th cycle. The ratios of the two bands between 1300 and 1400 cm−1 are 1.76:1, 1.78:1, and 1.78:1 for pristine, cycle 5, and cycle 10, which suggests that the 3D ZnO@ZnS-Ag nanorod clustered structures can be effectively recycled. Moreover, Figure S3 shows the Raman spectra recorded from five random spots on the substrate, which demonstrate the good uniformity and reproducibility of the substrate as the relative standard deviation (RSD) of the band intensity of R6G is less than 20%. The photocatalysis and durability of the ZnO@ZnS-Ag structures for photodegradation were also investigated (Figures S4 and S5). In a continuous photocatalytic test of the modified microchannels, the composite ZnO@ZnS-Ag nanorod array exhibited good structural and chemical stability. The photocatalytic response suggests the possibility that the devices could be integrated to allow for continuous high-throughput removal of organic pollutants, coupled with their detection by SERS. SERS Detection of Nerve Agent Simulant and Proteins. Examples of realistic analytes include nerve agents, pesticides, and biomolecules, where it is desirable to minimize sample handling. As seen in Figure 9a, the microfluidic channel fabricated here is readily filled by capillary action. To demonstrate the versatility of the capillary-based system, the spectral response of MP, a nerve agent simulant and pesticide, was evaluated. A SERS spectrum was readily obtained from MP at a concentration of only 10−8 M (Figure 9b). This represents at least a 1 order of magnitude increase in sensitivity vs previously reported SERS sensors for MP detection. Reported sensitivities ranged from 10−5 to 10−7 M using various nanostructured substrates, including Ag nanorods and Ag

■

CONCLUSIONS In summary, a materials design and fabrication concept based on a fluid-construction strategy to create 3D ZnO@ZnS-Ag nanostructures at arbitrary position inside confined microchannels to form an integrated microfluidic SERS system was demonstrated. The facile fluid-construction process provides for 3D SERS active substrates via in situ interfacial conversion. Substantial numbers of hot spots were integrated into the 3D clustered nanostructures, thereby creating a framework for SERS sensitivity, which was further investigated by FDTD analysis. Microfluidic SERS active device performance was evaluated using 4-MBA and a mixture of R6G and 4-MBA as target organic pollutants; high sensitivity and good recycle performance were achieved. The 3D clustered nanostructures F

DOI: 10.1021/acsami.6b10542 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

(7) Choi, K.; Ng, A. H. C.; Fobel, R.; Wheeler, A. R. Digital Microfluidics. Annu. Rev. Anal. Chem. 2012, 5 (1), 413−440. (8) Ren, K.; Zhou, J.; Wu, H. Materials for Microfluidic Chip Fabrication. Acc. Chem. Res. 2013, 46 (11), 2396−2406. (9) Zhu, J.; Ozdemir, S. K.; Xiao, Y.-F.; Li, L.; He, L.; Chen, D.-R.; Yang, L. On-Chip Single Nanoparticle Detection and Sizing by Mode Splitting in an Ultrahigh-Q Microresonator. Nat. Photonics 2010, 4 (1), 46−49. (10) Im, H.; Shao, H.; Park, Y. I.; Peterson, V. M.; Castro, C. M.; Weissleder, R.; Lee, H. Label-Free Detection and Molecular Profiling of Exosomes with a Nano-Plasmonic Sensor. Nat. Biotechnol. 2014, 32 (5), 490−495. (11) Cao, Q.; Che, R. Tailoring Au-Ag-S Composite Microstructures in One-Pot for Both SERS Detection and Photocatalytic Degradation of Plasticizers DEHA and DEHP. ACS Appl. Mater. Interfaces 2014, 6 (10), 7020−7027. (12) Ngo, H. T.; Wang, H. N.; Fales, A. M.; Vo-Dinh, T. Label-Free DNA Biosensor Based on SERS Molecular Sentinel on Nanowave Chip. Anal. Chem. 2013, 85 (13), 6378−6383. (13) Li, C. G.; Lee, K.; Lee, C. Y.; Dangol, M.; Jung, H. A Minimally Invasive Blood-Extraction System: Elastic Self-Recovery Actuator Integrated with an Ultrahigh- Aspect-Ratio Microneedle. Adv. Mater. 2012, 24 (33), 4583−4586. (14) Kang, J.-H.; Reichmanis, E. Low-Threshold Photon Upconversion Capsules Obtained by Photoinduced Interfacial Polymerization. Angew. Chem., Int. Ed. 2012, 51 (47), 11841−11844. (15) Park, C. P.; Kim, D.-P. A Microchemical System with Continuous Recovery and Recirculation of Catalyst-Immobilized Magnetic Particles. Angew. Chem., Int. Ed. 2010, 49 (38), 6825−6829. (16) Wang, G.; He, Z.; Shi, G.; Wang, H.; Zhang, Q.; Li, Y. Controllable Construction of Titanium Dioxide-Zirconium Dioxide@ zinc Hydroxyfluoride Networks in Micro-Capillaries for Bio-Analysis. J. Colloid Interface Sci. 2015, 446, 290−297. (17) Wang, G.; Shi, G.; Wang, H.; Zhang, Q.; Li, Y. In Situ Functionalization of Stable 3D Nest-Like Networks in Confined Channels for Microfluidic Enrichment and Detection. Adv. Funct. Mater. 2014, 24 (7), 1017−1026. (18) Sainiemi, L.; Sikanen, T.; Kostiainen, R. Integration of Fully Microfabricated, Three-Dimensionally Sharp Electrospray Ionization Tips with Microfluidic Glass Chips. Anal. Chem. 2012, 84 (21), 8973− 8979. (19) Nie, S. Probing Single Molecules and Single Nanoparticles by Surface-Enhanced Raman Scattering. Science 1997, 275, 1102−1106. (20) Schlücker, S. Surface-Enhanced Raman Spectroscopy: Concepts and Chemical Applications. Angew. Chem., Int. Ed. 2014, 53 (19), 4756−4795. (21) Gómez, M.; Lazzari, M. Reliable and Cheap SERS Active Substrates. Mater. Today 2014, 17 (7), 358−359. (22) Zhang, X.; Young, M. A.; Lyandres, O.; Van Duyne, R. P. Rapid Detection of an Anthrax Biomarker by Surface-Enhanced Raman Spectroscopy. J. Am. Chem. Soc. 2005, 127 (12), 4484−4489. (23) Huan, T. N.; Ganesh, T.; Kim, K. S.; Kim, S.; Han, S. H.; Chung, H. A Three-Dimensional Gold Nanodendrite Network Porous Structure and Its Application for an Electrochemical Sensing. Biosens. Bioelectron. 2011, 27 (1), 183−186. (24) Huang, D.; Bai, X.; Zheng, L. Ultrafast Preparation of ThreeDimensional Dendritic Gold Nanostructures in Aqueous Solution and Their Applications in Catalysis and SERS. J. Phys. Chem. C 2011, 115 (30), 14641−14647. (25) Lee, S.; Hahm, M. G.; Vajtai, R.; Hashim, D. P.; Thurakitseree, T.; Chipara, A. C.; Ajayan, P. M.; Hafner, J. H. Utilizing 3D SERS Active Volumes in Aligned Carbon Nanotube Scaffold Substrates. Adv. Mater. 2012, 24 (38), 5261−5266. (26) Tong, L.; Righini, M.; Gonzalez, M. U.; Quidant, R.; Käll, M. Optical Aggregation of Metal Nanoparticles in a Microfluidic Channel for Surface-Enhanced Raman Scattering Analysis. Lab Chip 2009, 9, 193−195. (27) Mao, H.; Wu, W.; She, D.; Sun, G.; Lv, P.; Xu, J. Microfluidic Surface-Enhanced Raman Scattering Sensors Based on Nanopillar

were also effective in the detection of a representative nerve agent and biomolecule. Overall, the fluid construction and in situ conversion concept provide an alternative approach to obtain desired 3D nanostructures, and the integrated microfluidic-SERS system may be valuable in many chemical and biological analyses.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b10542. Details about the enhancement factor comparison, photodegradation, uniformity of SERS detection, SERS detection performance with various Ag loading time, assignment of SERS bands for MP and BSA (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (H.W.). *E-mail: [email protected] (E.R.). Author Contributions

G.W. and K.L. made equal contributions. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is supported by NSF of China (51172042), MOST of China (2012AA030309), MOE of China (IRT1221, No.1112-04), STC of Shanghai (12 nm0503900, 13JC1400200), SRFDP (20110075130001), Donghua University (CUSF-DHD-2013002) and Eastern Scholar. This work was also supported by the Georgia Institute of Technology and the Brook Byers Institute for Sustainable Systems. G. Wang also thanks the China Scholarship Council for Fellowship support.

■

ABBREVIATIONS SERS, surface-enhanced Raman scattering; RT, residence time; EF, electric field; ST, sodium thioglycollate; TAA, thioacetamide; R6G, rhodamine 6G; MP, methylparathion.

■

REFERENCES

(1) Ray, P. Size and Shape Dependent Second Order Nonlinear Optical Properties of Nanomaterials and Their Application in Biological and Chemical Sensing. Chem. Rev. 2010, 110 (9), 5332− 5365. (2) Zhang, Q.; Uchaker, E.; Candelaria, S. L.; Cao, G. Nanomaterials for Energy Conversion and Storage. Chem. Soc. Rev. 2013, 42 (7), 3127−3171. (3) Lim, D.-K.; Jeon, K.-S.; Hwang, J.-H.; Kim, H.; Kwon, S.; Suh, Y. D.; Nam, J.-M. Highly Uniform and Reproducible Surface-Enhanced Raman Scattering from DNA-Tailorable Nanoparticles with 1-nm Interior Gap. Nat. Nanotechnol. 2011, 6 (7), 452−460. (4) Tan, E.; Yin, P.; You, T.; Wang, H.; Guo, L. Three Dimensional Design of Large-Scale TiO2 Nanorods Scaffold Decorated by Silver Nanoparticles as SERS Sensor for Ultrasensitive Malachite Green Detection. ACS Appl. Mater. Interfaces 2012, 4 (7), 3432−3437. (5) Xie, Y.; Yang, S.; Mao, Z.; Li, P.; Zhao, C.; Cohick, Z.; Huang, P.H.; Huang, T. J. In Situ Fabrication of 3D Ag @ ZnO Nanostructures for Microfluidic. ACS Nano 2014, 8 (12), 12175−12184. (6) Hu, M.; Ou, F. S.; Wu, W.; Naumov, I.; Li, X.; Bratkovsky, A. M.; Williams, R. S.; Li, Z. Gold Nanofingers for Molecule Trapping and Detection. J. Am. Chem. Soc. 2010, 132 (37), 12820−12822. G

DOI: 10.1021/acsami.6b10542 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces Forests Realized by an Oxygen-Plasma-Stripping-of-Photoresist Technique. Small 2014, 10 (1), 127−134. (28) Chrimes, A. F.; Khoshmanesh, K.; Stoddart, P. R.; Mitchell, A.; Kalantar-zadeh, K. Microfluidics and Raman Microscopy: Current Applications and Future Challenges. Chem. Soc. Rev. 2013, 42 (13), 5880−5906. (29) Sun, J.; Xianyu, Y.; Jiang, X. Point-of-Care Biochemical Assays Using Gold Nanoparticle-Implemented Microfluidics. Chem. Soc. Rev. 2014, 43 (17), 6239−6253. (30) Lee, M.; Lee, K.; Kim, K. H.; Oh, K. W.; Choo, J. SERS-Based Immunoassay Using a Gold Array-Embedded Gradient Microfluidic Chip. Lab Chip 2012, 12 (19), 3720−3727. (31) Chung, E.; Gao, R.; Ko, J.; Choi, N.; Lim, D. W.; Lee, E. K.; Chang, S.-I.; Choo, J. Trace Analysis of Mercury Ions Using AptamerModified Au/Ag Core−shell Nanoparticles and SERS Spectroscopy in a Microdroplet Channel. Lab Chip 2013, 13, 260−266. (32) Sackmann, E. K.; Fulton, A. L.; Beebe, D. J. The Present and Future Role of Microfluidics in Biomedical Research. Nature 2014, 507 (7491), 181−189. (33) Piorek, B. D.; Lee, S. J.; Moskovits, M.; Meinhart, C. D. FreeSurface Microfluidics/Surface-Enhanced Raman Spectroscopy for Real-Time Trace Vapor Detection of Explosives. Anal. Chem. 2012, 84, 9700−9705. (34) He, Z.; Li, Y.; Zhang, Q.; Wang, H. Capillary MicrochannelBased Microreactors with Highly Durable ZnO/TiO2 Nanorod Arrays for Rapid, High Efficiency and Continuous-Flow Photocatalysis. Appl. Catal., B 2010, 93 (3−4), 376−382. (35) Shuai, X. M.; Shen, W. Z. A Facile Chemical Conversion Synthesis of ZnO/ZnS Core/shell Nanorods and Diverse Metal Sulfide Nanotubes. J. Phys. Chem. C 2011, 115 (14), 6415−6422. (36) Tripathi, G. N. R. Adsorption of 2-Mercaptopyrimidine on Silver Nanoparticles in Water. J. Phys. Chem. B 2003, 107 (40), 11125−11132. (37) Kemna, E. W. M.; Segerink, L. I.; Wolbers, F.; Vermes, I.; van den Berg, A. Label-Free, High-Throughput, Electrical Detection of Cells in Droplets. Analyst 2013, 138 (16), 4585−4592. (38) Yang, L.; Liu, H.; Wang, J.; Zhou, F.; Tian, Z.; Liu, J. Metastable State Nanoparticle-Enhanced Raman Spectroscopy for Highly Sensitive Detection. Chem. Commun. 2011, 47 (12), 3583−3585. (39) Huang, Z.; Meng, G.; Huang, Q.; Chen, B.; Zhou, F.; Hu, X.; Qian, Y.; Tang, H.; Han, F.; Chu, Z. Polyacrylic Acid Sodium Salt Film Entrapped Ag-Nanocubes as Molecule Traps for SERS Detection. Nano Res. 2014, 7 (8), 1177−1187. (40) Blum, C.; Schmid, T.; Opilik, L.; Weidmann, S.; Fagerer, S. R.; Zenobi, R. Understanding Tip-Enhanced Raman Spectra of Biological Molecules: A Combined Raman, SERS and TERS Study. J. Raman Spectrosc. 2012, 43 (12), 1895−1904.

H

DOI: 10.1021/acsami.6b10542 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX