Trace Analysis and Chemical Identification on Cellulose Nanofibers

Article pubs.acs.org/acssensors

Trace Analysis and Chemical Identification on Cellulose NanofibersTextured SERS Substrates Using the “Coffee Ring” Effect Ruoyang Chen,†,# Liyuan Zhang,*,†,‡,# Xu Li,§,⊥ Lydia Ong,§,⊥ Ye Gaung Soe,† Neil Sinsua,† Sally L. Gras,§,⊥ Rico F. Tabor,∥ Xungai Wang,‡ and Wei Shen*,† †

Department of Chemical Engineering and ∥School of Chemistry, Monash University, Clayton, VIC 3800, Australia ‡ Institute for Frontier Materials, Deakin University, Waurn Ponds, VIC 3217, Australia § The Department of Chemical and Biomolecular Engineering and ⊥The Bio21 Molecular Science and Biotechnology Institute, The University of Melbourne, Parkville, VIC 3010, Australia S Supporting Information *

ABSTRACT: Surface-enhanced Raman scattering (SERS) has the theoretical possibility of detecting chemicals at the single molecular level. This potential is frequently limited, however, by the critical requirements of the surface morphology and mechanical stability of SERS substrates. In this paper, we report a new method for fabricating a SERS substrate with a significantly improved mechanical stability and analytical sensitivity, using cellulose nanofibers (CNFs) and gold nanoparticles (AuNPs). We constructed a uniformly CNFs-textured substrate on a glass surface by means of suppressing the “coffee ring” effect of the CNF sessile drop and then introduced an AuNP suspension onto the CNFs-textured substrate by taking advantage of the “coffee ring” effect. A widened detection zone is formed by AuNPs on the CNFs-textured glass, producing a stable SERS substrate for trace analysis and chemical identification. Microscopic and spectroscopic characterizations of the CNF-AuNPs SERS substrate show that the CNFs enhance the stability of both the AuNP clusters and the SERS activity. The CNF-AuNPs SERS substrate is significantly more stable and sensitive than the SERS substrate fabricated by directly depositing the AuNP suspension on a smooth glass surface. KEYWORDS: “coffee ring” effect, surface-enhanced Raman scattering (SERS), cellulose nanofibers (CNFs), gold nanoparticles (AuNPs), trace analysis, chemical identification

S

fabrication of such high performance SERS substrates has attracted increasing research interest.18 Recently, the “coffee ring” effect of sessile drops containing noble metal nanoparticles has been employed to fabricate the SERS substrate, as the aggregated metal nanoparticles in the ring region form many “hot spots”.19,20 The formation of a “coffee ring” in a dried colloidal sessile drop is a ubiquitous phenomenon.21 It is a spontaneous hydrodynamic process attributed to the local flow field within the sessile drop during drying. The enhanced evaporation rate at the pinned threephase contact line of the sessile drop builds up the horizontal outward flow to replenish the faster liquid loss at the edge; this hydrodynamic flow carries suspended particles to the drop edge. The particles accumulated at the edge increase the energy barrier for the recession of the contact line, resulting in the perpetual pinning of the contact line. After drying, the redistributed particles are concentrated at the perimeter of the sessile drop, forming a “coffee ring”.22−25 Although aggregated metal nanoparticles provide many “hot spots” for

urface-enhanced Raman scattering (SERS) is a surfacesensitive spectroscopic phenomenon that enhances the intensity of Raman scattered signals of analyte molecules absorbed on roughened metallic surfaces by a million-fold compared to Raman signals generated from nonabsorbed analyte molecules.1−3 SERS has numerous unique advantages, such as molecular specificity (based on the distinguishable vibrational characteristics of analyte molecules), high sensitivity (down to single molecular level), simplicity of use, and minimal photobleaching.4−6 These advantages mean that the SERS is an excellent platform for rapid and nondestructive detection of chemical and biological substances.7−13 The application of SERS is often limited, however, by the availability of distinctively structured SERS substrates.14,15 Typically, a highly SERS-active substrate possesses many “hot spots” that are junctions and gaps between two adjacent noble metal nanoparticles.16 Within these “hot spots”, the electric field intensity of the incident laser is concentrated to excite the localized surface plasmon, leading to the enhancement of the Raman scattering of adsorbed analyte molecules. A SERS substrate with a critical nanoscale surface roughness (Ra: 50− 200 nm) has more effective “hot spots”.17 Therefore, © XXXX American Chemical Society

Received: June 15, 2017 Accepted: June 27, 2017 Published: June 27, 2017 A

DOI: 10.1021/acssensors.7b00403 ACS Sens. XXXX, XXX, XXX−XXX

Article

ACS Sensors

Figure 1. Fabrication of the CNF-AuNPs SERS substrate: (a) deposition of a CNF sessile drop onto a clean and smooth glass surface; (b) formation of the uniform CNFs-textured substrate after drying; (c) introduction of a AuNP sessile drop onto the CNFs-textured substrate (b); (d) the obtained CNF-AuNPs SERS substrate. Raman detection: (e) deposition of the analyte solution; (f) Raman detection of the analyte samples.

Figure 2. (a) Raman intensity of the 4-aminothiophenol (4-ATP) at the characteristic peak of 1583 cm−1 along the radii of AuNPs (left) and CNFAuNPs (right) SERS substrates, with the detection zones indicated by red vertical lines, respectively; the insets show the optical and AFM images of AuNPs (left) and CNF-AuNPs (right) SERS substrates; (b) multi SERS spectra acquisition of 4-ATP in the detection zone of CNF-AuNPs SERS substrate, clearly showing the 100% frequency of the major characteristic peaks.

could interfere with the Raman signal of analyte samples.25 More importantly, none of them could overcome the problem of the redispersion of the “coffee ring” when a liquid sample is introduced onto such a SERS substrate. These methods are therefore not suitable for the fabrication of SERS substrates. In this research, we aimed to use the “coffee ring” effect to fabricate a novel SERS substrate with cellulose nanofibers (CNFs) and gold nanoparticles (AuNPs). CNFs have a large aspect ratio with a longitudinal dimension of microns and a lateral dimension of one-tenth of a micron.29 The dried CNF sessile drop can form a uniform deposit of CNF networks, which provides the critical roughness for the generation of “hot spots” and barriers for mitigating the transport of AuNPs to form a wide ring and to minimize the redispersion of AuNPs.30 The fabricated SERS substrate was tested with 4-aminothiophenol (4-ATP) to assess SERS performance. To demonstrate the feasibility of trace analysis and chemical identification of CNF-AuNPs SERS substrates, we used the

Raman enhancement, they do suffer from some limitations in regard to practical use. A major limitation of such SERS substrates is the instability of the “hot spots”; the aggregated metal nanoparticles could be easily redistributed, leading to unwanted changes in the number and quality of “hot spots”.17 It is also difficult to control the difference in Raman spectra sampled on such a SERS substrate.17 Furthermore, the “coffee ring” formed from a diluted metal nanoparticle suspension results in a very narrow band of particle aggregation for SERS detection. Given the above limitations, fabrication of a stable “coffee ring” effect-based SERS substrate with a large detection zone and a high sensitivity becomes necessary. Several strategies have been proposed to control the “coffee ring” effect of a colloidal sessile drop, e.g., temperature control of the substrate,26 application of an electrical potential,27 and addition of organic mixtures.28 However, most of those methods are laborious and require either extra energy or organic solvent, which could be flammable and/or toxic, or B


Article

ACS Sensors

particle-induced interfacial deformation of a drying sessile drop and indicated that the deposition pattern in such a dried drop was related to the ratio of capillary attraction to hydrodynamic F force acting on the particles ( γ ).38 They considered that the

new SERS substrate to quantify forbidden food additives, i.e., Rhodamine B (RhB) and Dopamine (DA). RhB is a synthetic dye that has been widely used as a staining fluorescent dye in biology and a colorant in food.31,32 However, RhB is suspected to be carcinogenic and have reproductive and developmental toxicity, as well as neuro- and other chronic toxicities toward human beings and animals.32 Consequently, RhB has been banned as a food additive by the China Food and Drug Admission (CFDA) and the European Food Safety Authority (EFSA).33 DA is an organic chemical of the catecholamine family, which is extensively used as a potent drug for treatment of heart failure and hypotension.34 Recently, DA has been reported to be added into animal feeds as a “lean meat powder” to promote the muscle growth of domestic animals.35 However, DA has been proven to be a dominant etiological contributor to several forms of neurological disorders, e.g., schizophrenia, Parkinson’s disease, and Huntington’s disease.36,37 Food containing DA may have serious side effects for anyone with such neurological disorders. Therefore, the use of DA in food has been banned in most countries.35 The serious safety concerns of RhB and DA as food additives require a sensitive and stable analytical method to rapidly detect these compounds. Here we use the CNF-AuNPs SERS substrate to detect and quantify the simulating solutions containing RhB, DA, or both. The schematic of the fabrication of the CNFAuNPs SERS substrate and the Raman detection is given in Figure 1.

Fμ

uniform deposition pattern can be formed when

Fγ Fμ

> 1 and

estimated it by Fγ Fμ

≈

8γwHp2b3k 4/3 μvsLm5

(1)

where γw and μ are the surface tension and dynamic viscosity of water, respectively; Lm = (4πkb3/3Φ)1/3 is the distance between two ellipsoidal particles, in which b, k, and Φ are the semiminor axis, aspect ratio, and initial volume fraction of ellipsoidal particles, respectively; s and Hp are the shape-dependent correction factor and interface distortion of the particles, respectively; and v is the initial radial velocity of the outward flow within the sessile drop. The CNF sessile drop in our present paper is a complicated system, if we assume CNFs are ellipsoidal particles with a semiminor axis (b) of 50 nm and an aspect ratio (k) of 10, the shape-dependent correction factor (s) and interface distortion (Hp) of such particles can be estimated as 3.8 and 600 nm, respectively.38,39 The CNFs are assumed to disperse uniformly at the liquid−air interface during the initial drying stage. Also, the profile of the sessile drop features a spherical cap due to the small Bond number Bo = ρgh2/γ ≪ 1, where ρ, h, and γ are the density, height, and surface tension of the sessile drop, respectively.40 Thus, the ratio of capillary attraction (Fγ) to hydrodynamic force (Fμ) acting on CNFs can

■

RESULTS AND DISCUSSION Characterization of CNF-AuNPs SERS Substrate. Raman mappings of 4-ATP at its major characteristic peak of 1583 cm−1 along the radii of the AuNPs (without CNFs) and CNF-AuNPs SERS substrates are displayed in Figure 2a. The AuNPs SERS substrate (left) shows a narrow band of aggregated AuNPs (LD ≤ 30 μm) in the peripheral region of the dried AuNP sessile drop. This is an expected result of the “coffee ring” effect induced by drying the colloidal sessile drop on a bare and smooth glass surface (Movie S-1); most of the AuNPs aggregate at the contact line of the sessile drop and generate a large number of “hot spots” for SERS detection, while there is a depletion of AuNPs and “hot spots” in the regions immediately away from the drop edge, where no Raman signal could be detected. Compared with the AuNPs SERS substrate, the CNF-AuNPs SERS substrate has a wider SERS detection zone (LD ≥ 300 μm). The underlying mechanisms of such widened bands are related to the two-step fabrication process of the CNF-AuNPs SERS substrates: preparation of the CNFs-textured substrate and introduction of the AuNPs suspension. When a CNF sessile drop is placed on a glass surface, most of the CNFs are affected by the nonuniform evaporation-induced hydrodynamic force (Fμ) and are transported toward the drop edge. When they reach the liquid−air interface of the drop, CNFs will tend to deform the interface, activating strong capillary attraction (Fγ) toward one another, and forming the CNF clusters that resist their outward motion.25,38 Therefore, the CNFs distribute almost uniformly in the dried sessile drop (Figure S-1a). To verify this hypothesis, surfactant (sodium dodecyl sulfate, SDS, 0.1% w/w) was added to the CNF sessile drop. Surfactant lowers the surface tension of the drop, reducing the capillary attraction between CNFs at the interface.25 As a result, the “coffee ring” is restored after drying (Figure S-1b). Recently, Kim et al. studied the ellipsoidal

be estimated to be larger than 1, i.e.,

Fγ

Fμ

≫ 1. Kim’s ideal model

has the quantitative implication of the uniform deposition of the dried CNF sessile drop. In a practical situation, however, the irregular shape of CNFs could produce more complex interfacial deformation, activating stronger capillary attraction. Furthermore, the entanglement of CNFs could promote the formation of a uniform CNFs-textured substrate. On the CNFs-textured substrate, the AuNP sessile drop experiences two drying stages (Figure S-2). In the first stage, CNFs in the peripheral region provide barriers for the drop to recede, resulting in the temporary pinning of the contact line. During this period, the “coffee ring” effect drives the transport of AuNPs toward the peripheral region. These outward migrating AuNPs are, however, easily captured by the CNFs substrate (Movie S-2), resulting in the retardation of the AuNP migration and allowing a smaller number of AuNPs to aggregate at the contact line. As the evaporation propagates to the second stage, the receding contact line overcomes the barriers of the CNFs substrate and recedes toward the central region. During this recession stage, AuNPs can be deposited onto the CNF deposits, forming a wide band of AuNP deposits at the peripheral region of the sessile drop. These AuNPs provide the suitable “hot spots” for SERS detection, generating a widened detection zone on the CNFs-textured SERS substrate. Meanwhile, some AuNPs can be trapped by the CNF deposits in the central region, resulting in the occasional occurrence of the Raman signal of analyte samples in the central region. As for the Raman intensity of analyte samples on the AuNPs and CNF-AuNPs SERS substrates in Figure 2a, the latter shows a stronger intensity. This phenomenon is related to the C


Article

ACS Sensors enhanced stability of “hot spots” on the CNFs-textured substrate to the disposition of the analyte solution. The immobilization of AuNPs by the CNF networks significantly reduces their redispersion during the liquid sample introduction; this minimizes the loss of “hot spots” (Figure S-3a,b). By contrast, many “hot spots” on the AuNPs SERS substrate are lost by the redispersion of AuNPs (Figure S-3c,d); the SERS signal from the AuNPs SERS substrate is therefore reduced.17 To demonstrate the reproducibility of the CNF-AuNPs SERS substrate, the multi SERS spectra of 4-ATP were acquired from eight different locations in the detection zone, as displayed in Figure 2b. The main characteristic peaks at 1081, 1143, 1179, 1393, 1433, and 1583 cm−1 are clearly identifiable in the random multispectra acquisition. These results indicate that the CNF-AuNPs SERS substrate has a high reproducibility that can be used for rapid chemical identification. SEM and optical micrographs of the AuNPs and CNFAuNPs SERS substrates are presented in Figure 3, respectively,

ISERS = IeN

∂σ ΩEeEs ∂Ω

(2)

where Ie is the intensity of the exciting light; ∂σ is the ∂Ω differential Raman cross section per adsorbed analyte molecule; and Ω is the solid angle of the collection optics; Ee and Es are the electromagnetic surface-averaged intensity enhancement factors of the exciting and scattering light, respectively. In the detection zone of the CNF-AuNPs SERS substrate, the major characteristic Raman peaks of 4-ATP were captured with high reproducibility for samples of all the investigated analyte concentrations, ranging from 1 × 10−4 to 1 × 10−12 M (Figure 4a). When the 4-ATP is reduced to the trace amount (1 × 10−12 M), the Raman signal can still be clearly detected (Figure 4b), showing the characteristic peaks at 1008, 1081, and 1583 cm−1 and a signal-to-noise ratio larger than 2.41,42 This result, together with the approximation in the Supporting Information, indicates that the detection zone of the CNFAuNPs SERS substrate is appropriate for trace analysis. By contrast, the central region of the CNF-AuNPs SERS substrate shows a much lower Raman signal intensity (Figure 4c), and there is no Raman signal of the analyte when the sample concentration reaches 1 × 10−12 M (Figure 4d). These results are also in agreement with the SERS detection results in Figure 2a. Accordingly, the detection zone of the CNF-AuNPs SERS substrate was used for the subsequent trace analysis of Rhodamine B (RhB) and Dopamine (DA). Figure 5 shows the Raman spectra of both RhB and DA solutions of different investigated sample concentrations, ranging from 1 × 10−4 to 1 × 10−12 M, on the detection zone of the CNF-AuNPs SERS substrate. The spectroscopic assignments of major characteristic Raman peaks are listed in Tables S-2 and S-3. Figure 5a,b shows that the characteristic peaks of RhB at 621, 1525, and 1648 cm−1 can always be detected for samples of all the investigated concentrations (even down to 1 × 10−12 M), indicating high reproducibility.43 Meanwhile, Figure 5b shows the incremental relationship between the signal intensity of the characteristic peak and the concentration of RhB. Similar Raman sensitivity and reproducibility are achieved with the detection of the DA solutions of different investigated concentrations (Figure 5c). Trace amounts of DA (1 × 10−12 M) can also be easily detected (Figure 5d), clearly showing the major characteristic Raman peaks at 1492, 1580, and 1609 cm−1.34 The incremental relationship between the signal intensity of the characteristic peaks and the concentration of DA is also similar to RhB, as shown in Figure 5d. These results demonstrate that the CNFAuNPs SERS substrate not only can provide accurate and reproducible Raman signals of the probe analyte samples, but also can be a highly effective tool for trace analysis of the forbidden chemical additives. Raman spectra of a mixed solution containing both RhB and DA were collected on the CNF-AuNPs SERS substrate, as indicated in Figure 6. The major characteristic peaks of each compound are labeled with different symbols (red sun for RhB and blue moon for DA). The key characteristic peaks of these chemicals can always be distinguished clearly, even when the concentration of the mixture decreases from 1 × 10−7 to 1 × 10−9 M. These characteristic peaks can potentially be used to rapidly screen for mixed chemical additives in complex food matrices. Hence, the CNF-AuNPs SERS substrate has a potential application in the rapid and convenient identification of toxic organic additives in food mixtures.

Figure 3. SEM images of different SERS substrates: (a,b) the detection zone and central region of the AuNPs SERS substrate, respectively; (c,d) the detection zone and central region of the CNF-AuNPs SERS substrate, respectively. These regions are indicated by the red squares in the insets of the overview of the SERS substrates.

for characterization of the AuNP distribution on different substrates. The SEM images show the aggregation (Figure 3a) and depletion (Figure 3b) of AuNPs in the peripheral and central regions of the AuNPs SERS substrate. By contrast, the CNF-AuNPs SERS substrate presents a wider peripheral region, where the CNFs are covered by dense arrangements of AuNPs (Figure 3c); in the central region, there are still a significant number of AuNPs adhering to CNFs (Figure 3d). These observations are in agreement with the SERS detection results in Figure 2. Trace Chemical Analysis Using the CNF-AuNPs SERS Substrate. To evaluate the sensitivity of the CNF-AuNPs SERS substrate, Raman spectra of analyte samples in different concentrations were acquired and shown in Figure 4. The analysis of major characteristic peaks and their assignments are listed in Table S-1. The SERS intensity (ISERS) of analyte samples on the CNF-AuNPs SERS substrate decreases with the decrease in analyte concentration. This can be related to the number of analyte molecules (N) trapped by the “hot spots” as follows:3 D


Article

ACS Sensors

Figure 4. Raman spectra of (a) different concentrations of 4-ATP and (b) 1 × 10−12 M of 4-ATP solution detected in the detection zone of the CNF-AuNPs SERS substrate. Raman spectra of (c) different concentrations of 4-ATP and (d) 1 × 10−12 M of 4-ATP solution detected in the central region. The red star marks in the insets show the selected detection regions and laser-excited zones.

Figure 5. Raman spectra of analyte solutions with (a) different concentrations of Rhodamine B (RhB), (b) 1 × 10−12 M of RhB, (c) different concentrations of Dopamine (DA), and (d) 1 × 10−12 M of DA in the detection zone of the CNF-AuNPs SERS substrate. The relation between the Raman intensity and the concentration of analyte samples are shown in the insets.

E


Article

ACS Sensors

of 1% (w/v). Twenty-five grams of the fibril slurries was mixed with 45 g of zirconia balls and 20 mL of Millipore water. The mixture was then subjected to ball milling for 1 h. After the zirconia balls and any large remaining fibers were removed by filtration, the CNF suspension was obtained and stored under 4 °C. The CNFs suspension was diluted to 0.1% (v/v) for texturing the glass substrates. Fabrication of SERS Substrates. A sessile drop of CNF suspension (1.50 ± 0.05 μL) was placed onto the glass microscope slide and dried under ambient conditions to construct a CNFstextured substrate with a diameter of ca. 3 mm. Subsequently, a 1.50 ± 0.05 μL of AuNP suspension was deposited onto the CNFs-textured substrate under ambient conditions. After drying, the CNF-AuNPs SERS substrates were fabricated. To investigate the SERS effect of the CNF-AuNPs substrate, a AuNPs SERS substrate was set as the control. It was fabricated by drying of a sessile drop of AuNP suspension (1.50 ± 0.05 μL) on a glass microscope slide. Preparation of Analyte Samples. 4-ATP was dissolved in ethanol and diluted to different concentrations with Millipore water, ranging from 10−4 to 10−12 M. The solution and mixture of analyte samples were prepared by dissolving the RhB and DA powders in Millipore water separately. All of the above analyte solutions were used within 3 h of preparation. Sessile drops of analyte samples (1.50 ± 0.05 μL) were gently deposited onto the SERS substrates and dried under ambient conditions. Characterization and Detection. A BX-60 microscope (Olympus, Japan) was employed to observe the SERS substrates and record the drying process of AuNP sessile drops. An OCAH-230 contact angle apparatus (Dataphysics, Germany) was used to capture the profile evolution of the AuNPs sessile drop drying on the CNFstextured substrate. An FEI Quanta 200 FEG scanning electron microscopy (SEM) instrument was used to observe the microscopic morphologies of SERS substrates. The atomic force microscope (AFM) measurements were performed on a JPK NanoWizard 3 AFM machine in AC (intermittent contact) mode. SERS detection was carried out with a Renishaw inVia Raman microscope (Russia) equipped with a 633 nm laser. The laser spot size was around 2 μm in diameter. The laser beam, with intensity of 10%, was positioned through a Leica imaging microscope objective lens (X 50).

Figure 6. Raman spectra of mixtures containing different concentrations of RhB and DA in the detection zone of CNF-AuNPs SERS substrate; the major characteristic peaks of RhB and DA are labeled with the red sun and blue moon, respectively.

■

CONCLUSION This research presented a simple, facile, and convenient method for the fabrication of a CNF-AuNPs SERS substrate using the “coffee ring” effect. Such SERS substrates have been demonstrated to be highly effective and sensitive, as well as having a large detection zone and a high spectral reproducibility. The low limit of detection (LOD) of such SERS substrates indicates the possibility for trace analysis of forbidden chemicals in materials such as food. Additionally, this highly SERS-active substrate can be used to rapidly identify toxic organic chemicals of low concentration in mixtures of two molecules. Further work is required to extend these findings to more complicated chemical analytes and test the applications to food materials.

■

■

EXPERIMENTAL SECTION

Materials. All chemicals were of analytical grade and used without further purification. Gold(III) chloride trihydrate (HAuCl4·3H2O), sodium dodecyl sulfate (SDS), 4-aminothiophenol (4-ATP), and dopamine hydrochloride (DA) were purchased from Sigma-Aldrich (U.S.A.). Rhodamine B (RhB) and trisodium citrate (Na3Ct·2H2O) were purchased from Ciba (Switzerland) and Ajax Finechem Pty Ltd. (Australia), respectively. Glass microscope slides were obtained from Objektträger (Germany) and thoroughly rinsed with ethanol and Millipore water (18.2 MΩ·cm) successively prior to use. Dry sheets of commercial Northern Bleached Softwood Kraft were manufactured by Grande Prairie Pulp Mill (Canada) for making CNFs, and Zirconox cerium-doped zirconia balls with a diameter of 500 μm (Klausen Pty Ltd., Australia) were employed for ball milling. Synthesis of Gold Nanoparticles. The gold nanoparticles (AuNPs) were synthesized using a modified citrate reduction method reported in our previous paper.17 90 mL of Millipore water was brought to boil and 10 mL of 2.5 mM HAuCl4·3H2O was added with vigorous continual stirring. After the solution was brought to boil again, 4 mL of 0.25% (w/v) Na3Ct·2H2O was added. Under the condition of continuous boiling and vigorous stirring, the color of the solution changed from pale yellow to colorless and then to almost black and finally to red-purple. After an additional 30 min, a AuNPs colloidal suspension of 40 nm in diameter was synthesized and stored under 4 °C in the dark. Preparation of Cellulose Nanofibers. The cellulose nanofibers (CNFs) were prepared from softwood pulp by ball milling.29 Small pieces of softwood pulp sheets were soaked in Millipore water with a solid content of 10% (w/v) under ambient conditions for 24 h to obtain the swollen pulp fibrils. These fibrils were disintegrated by a kitchen blender and then diluted into fibril slurries with a solid content

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acssensors.7b00403. Morphologies of the dried CNF sessile drops with and without the addition of SDS; morphological evolution of the AuNP sessile drop drying on the CNFs-textured substrate; stability of SERS substrates to the deposition of analyte solution; main Raman peaks of analyte samples and their assignments; approximation of the average number of excited analyte molecules (PDF) Movie for the “coffee ring” effect of the AuNP sessile drop drying on a bare glass substrate (MPG) Movie for the capture of AuNPs by CNF deposits during drying (MPG)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Sally L. Gras: 0000-0002-4660-1245 Rico F. Tabor: 0000-0003-2926-0095 Wei Shen: 0000-0002-9991-2100 F


Article

ACS Sensors Author Contributions

(15) Dai, P. C.; Xue, Y. M.; Wang, X. B.; Weng, Q. H.; Zhang, C.; Jiang, X. F.; Tang, D. M.; Wang, X.; Kawamoto, N.; Ide, Y.; Mitome, M.; Golberg, D.; Bando, Y. Pollutant Capturing SERS Substrate: Porous Boron Nitride Microfibers with Uniform Silver Nanoparticle Decoration. Nanoscale 2015, 7, 18992−18997. (16) Moskovits, M. Persistent Misconceptions Regarding SERS. Phys. Chem. Chem. Phys. 2013, 15, 5301−5311. (17) Zhang, L. Y.; Li, X.; Ong, L.; Tabor, R. F.; Bowen, B. A.; Fernando, A. I.; Nilghaz, A.; Garnier, G.; Gras, S. L.; Wang, X. G.; Shen, W. Cellulose Nanofibre Textured SERS Substrate. Colloids Surf., A 2015, 468, 309−314. (18) Betz, J. F.; Yu, W. W.; Cheng, Y.; White, I. M.; Rubloff, G. W. Simple SERS Substrates: Powerful, Portable, and Full of Potential. Phys. Chem. Chem. Phys. 2014, 16, 2224−2239. (19) Wang, W. D.; Yin, Y. Q.; Tan, Z. Q.; Liu, J. F. Coffee-Ring Effect-Based Simultaneous SERS Substrate Fabrication and Analyte Enrichment for Trace Analysis. Nanoscale 2014, 6, 9588−9593. (20) Xu, J. W.; Du, J. J.; Jing, C. Y.; Zhang, Y. L.; Cui, J. L. Facile Detection of Polycyclic Aromatic Hydrocarbons by a SurfaceEnhanced Raman Scattering Sensor Based on the Au Coffee Ring Effect. ACS Appl. Mater. Interfaces 2014, 6, 6891−6897. (21) Chen, R. Y.; Zhang, L. Y.; Zang, D. Y.; Shen, W. Blood Drop Patterns: Formation and Applications. Adv. Colloid Interface Sci. 2016, 231, 1−14. (22) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S. R.; Witten, T. A. Capillary Flow as the Cause of Ring Stains from Dried Liquid Drops. Nature 1997, 389, 827−829. (23) Marín, Á . G.; Gelderblom, H.; Lohse, D.; Snoeijer, J. H. Orderto-Disorder Transition in Ring-Shaped Colloidal Stains. Phys. Rev. Lett. 2011, 107, 085502. (24) Anyfantakis, M.; Baigl, D. Dynamic Photocontrol of the CoffeeRing Effect with Optically Tunable Particle Stickiness. Angew. Chem., Int. Ed. 2014, 53, 14077−14081. (25) Yunker, P. J.; Still, T.; Lohr, M. A.; Yodh, A. G. Suppression of the Coffee-Ring Effect by Shape-Dependent Capillary Interactions. Nature 2011, 476, 308−311. (26) Soltman, D.; Subramanian, V. Inkjet-Printed Line Morphologies and Temperature Control of the Coffee Ring Effect. Langmuir 2008, 24, 2224−2231. (27) Eral, H. B.; Augustine, D. M.; Duits, M. H. G.; Mugele, F. Suppressing the Coffee Stain Effect: How to Control Colloidal SelfAssembly in Evaporating Drops Using Electrowetting. Soft Matter 2011, 7, 4954−4958. (28) Hu, H.; Larson, R. G. Marangoni Effect Reverses Coffee-Ring Depositions. J. Phys. Chem. B 2006, 110, 7090−7094. (29) Zhang, L. Y.; Tsuzuki, T.; Wang, X. G. Preparation of Cellulose Nanofiber from Softwood Pulp by Ball Milling. Cellulose 2015, 22, 1729−1741. (30) Tian, L.; Luan, J.; Liu, K. K.; Jiang, Q.; Tadepalli, S.; Gupta, M. K.; Naik, R. R.; Singamaneni, S. Plasmonic Biofoam: A Versatile Optically Active Material. Nano Lett. 2016, 16, 609−616. (31) Su, X. M.; Li, X. Y.; Li, J. J.; Liu, M.; Lei, F. H.; Tan, X. C.; Li, P. F.; Luo, W. Q. Synthesis and Characterization of Core-Shell Magnetic Molecularly Imprinted Polymers for Solid-Phase Extraction and Determination of Rhodamine B in Food. Food Chem. 2015, 171, 292−297. (32) Jain, R.; Mathur, M.; Sikarwar, S.; Mittal, A. Removal of the Hazardous Dye Rhodamine B Through Photocatalytic and Adsorption Treatments. J. Environ. Manage. 2007, 85, 956−964. (33) Qi, P.; Lin, Z. H.; Li, J. X.; Wang, C. L.; Meng, W. W.; Hong, H.; Zhang, X. W. Development of a Rapid, Simple and Sensitive HPLC-FLD Method for Determination of Rhodamine B in ChiliContaining Products. Food Chem. 2014, 164, 98−103. (34) Pande, S.; Jana, S.; Sinha, A. K.; Sarkar, S.; Basu, M.; Pradhan, M.; Pal, A.; Chowdhury, J.; Pal, T. Dopamine Molecules on AucoreAgshell Bimetallic Nanocolloids: Fourier Transform Infrared, Raman, and Surface-Enhanced Raman Spectroscopy Study Aided by Density Functional Theory. J. Phys. Chem. C 2009, 113, 6989−7002.

#

Ruoyang Chen and Liyuan Zhang contributed equally as cofirst authors to this work. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS

■

REFERENCES

Authors thank Kalar Contreras, Hui He, Finlay Shanks, and Dashen Dong for their help with experiments, and Hansen Shen for proof reading. Ruoyang Chen thanks Monash Graduate Education and the Faculty of Engineering for postgraduate research scholarships. Liyuan Zhang and Wei Shen acknowledge Australian Research Council Discovery Project (DP1094179 and 140100052) and Zhejiang International Science and Technology Cooperation Project (2015C34014).

(1) Fleischmann, M.; Hendra, P. J.; McQuillan, A. J. Raman Spectra of Pyridine Adsorbed at a Silver Electrode. Chem. Phys. Lett. 1974, 26, 163−166. (2) Albrecht, M. G.; Creighton, J. A. Anomalously Intense Raman Spectra of Pyridine at a Silver Electrode. J. Am. Chem. Soc. 1977, 99, 5215−5217. (3) Kruszewski, S. Dependence of SERS Signal on Surface Roughness. Surf. Interface Anal. 1994, 21, 830−838. (4) Vendrell, M.; Maiti, K.; Dhaliwal, K.; Chang, Y. T. SurfaceEnhanced Raman Scattering in Cancer Detection and Imaging. Trends Biotechnol. 2013, 31, 249−257. (5) Willets, K. A. Super-Resolution Imaging of SERS Hot Spots. Chem. Soc. Rev. 2014, 43, 3854−3864. (6) Indrasekara, A. S. D. S.; Meyers, S.; Shubeita, S.; Feldman, L. C.; Gustafsson, T.; Fabris, L. Gold Nanostar Substrates for SERS-Based Chemical Sensing in the Femtomolar Regime. Nanoscale 2014, 6, 8891−8899. (7) Doering, W. E.; Piotti, M. E.; Natan, M. J.; Freeman, R. G. SERS as a Foundation for Nanoscale, Optically Detected Biological Labels. Adv. Mater. 2007, 19, 3100−3108. (8) Xie, J.; Zhang, Q.; Lee, J. Y.; Wang, D. I. C. The Synthesis of SERS-Active Gold Nanoflower Tags for In Vivo Applications. ACS Nano 2008, 2, 2473−2480. (9) Samanta, A.; Maiti, K. K.; Soh, K. S.; Liao, X.; Vendrell, M.; Dinish, U. S.; Yun, S. W.; Bhuvaneswari, R.; Kim, H.; Rautela, S.; Chung, J.; Olivo, M.; Chang, Y. T. Ultrasensitive Near-Infrared Raman Reporters for SERS-Based In Vivo Cancer Detection. Angew. Chem., Int. Ed. 2011, 50, 6089−6092. (10) Guerrini, L.; Krpetić, Ž .; van Lierop, D.; Alvarez-Puebla, R. A.; Graham, D. Direct Surface-Enhanced Raman Scattering Analysis of DNA Duplexes. Angew. Chem., Int. Ed. 2015, 54, 1144−1148. (11) Andreou, C.; Neuschmelting, V.; Tschaharganeh, D. F.; Huang, C. H.; Oseledchyk, A.; Iacono, P.; Karabeber, H.; Colen, R. R.; Mannelli, L.; Lowe, S. W.; Kircher, M. F. Imaging of Liver Tumors Using Surface-Enhanced Raman Scattering Nanoparticles. ACS Nano 2016, 10, 5015−5026. (12) Huang, J. A.; Zhang, Y. L.; Zhao, Y.; Zhang, X. L.; Sun, M. L.; Zhang, W. Superhydrophobic SERS Chip Based on a Ag Coated Natural Taro-Leaf. Nanoscale 2016, 8, 11487−11493. (13) Song, C. Y.; Yang, Y. J.; Yang, B. Y.; Sun, Y. Z.; Zhao, Y. P.; Wang, L. H. An Ultrasensitive SERS Sensor for Simultaneous Detection of Multiple Cancer-Related MiRNAs. Nanoscale 2016, 8, 17365−17373. (14) López-Tocón, I.; Otero, J. C.; Arenas, J. F.; Garcia-Ramos, J. V.; Sanchez-Cortes, S. Multicomponent Direct Detection of Polycyclic Aromatic Hydrocarbons by Surface-Enhanced Raman Spectroscopy Using Silver Nanoparticles Functionalized with the Viologen Host Lucigenin. Anal. Chem. 2011, 83, 2518−2525. G


Article

ACS Sensors (35) Yang, J.; Strickler, J. R.; Gunasekaran, S. Indium Tin OxideCoated Glass Modified with Reduced Graphene Oxide Sheets and Gold Nanoparticles as Disposable Working Electrodes for Dopamine Sensing in Meat Samples. Nanoscale 2012, 4, 4594−4602. (36) Ali, S. R.; Ma, Y.; Parajuli, R. R.; Balogun, Y.; Lai, W. Y. C.; He, H. A Nonoxidative Sensor Based on a Self-Doped Polyaniline/Carbon Nanotube Composite for Sensitive and Selective Detection of the Neurotransmitter Dopamine. Anal. Chem. 2007, 79, 2583−2587. (37) Ben-Shachar, D.; Zuk, R.; Gazawi, H.; Ljubuncic, P. Dopamine Toxicity Involves Mitochondrial Complex I Inhibition: Implications to Dopamine-Related Neuropsychiatric Disorders. Biochem. Pharmacol. 2004, 67, 1965−1974. (38) Kim, D. O.; Pack, M.; Hu, H.; Kim, H.; Sun, Y. Deposition of Colloidal Drops Containing Ellipsoidal Particles: Competition between Capillary and Hydrodynamic Forces. Langmuir 2016, 32, 11899−11906. (39) Loudet, J. C.; Yodh, A. G.; Pouligny, B. Wetting and Contact Lines of Micrometer-Sized Ellipsoids. Phys. Rev. Lett. 2006, 97, 018304. (40) Chen, R. Y.; Zhang, L. Y.; Zang, D. Y.; Shen, W. Wetting and Drying of Colloidal Droplets: Physics and Pattern Formation. In Advances in Colloid Science, Rahman, M. M.; Asiri, A. M., Eds.; InTech, 2016; Chapter 1, pp 3−25. (41) Hu, X. G.; Wang, T.; Wang, L.; Dong, S. J. Surface-Enhanced Raman Scattering of 4-Aminothiophenol Self-Assembled Monolayers in Sandwich Structure with Nanoparticle Shape Dependence: OffSurface Plasmon Resonance Condition. J. Phys. Chem. C 2007, 111, 6962−6969. (42) Ballerini, D. R.; Ngo, Y. H.; Garnier, G.; Ladewig, B. P.; Shen, W.; Jarujamrus, P. Gold Nanoparticle-Functionalized Thread as a Substrate for SERS Study of Analytes Both Bound and Unbound to Gold. AIChE J. 2014, 60, 1598−1605. (43) Liu, X.; Shao, Y.; Tang, Y.; Yao, K. F. Highly Uniform and Reproducible Surface Enhanced Raman Scattering on Air-stable Metallic Glassy Nanowire Array. Sci. Rep. 2015, 4, 5835.

H


Trace Analysis and Chemical Identification on Cellulose Nanofibers

Recommend Documents