Combination of a Sample Pretreatment Microfluidic ... - ACS Publications

Subscriber access provided by UQ Library

Article

Combination of a Sample Pretreatment Microfluidic Device with a Photoluminescent Graphene Oxide Quantum Dot Sensor for Trace Lead Detection Minsu Park, Hyun Dong Ha, Yong Tae Kim, Jae Hwan Jung, Shin-Hyun Kim, Do Hyun Kim, and Tae Seok Seo Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b02907 • Publication Date (Web): 12 Oct 2015 Downloaded from http://pubs.acs.org on October 19, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Combination of a Sample Pretreatment Microfluidic Device with a Photoluminescent Graphene Oxide Quantum Dot Sensor for Trace Lead Detection Minsu Park, Hyun Dong Ha, Yong Tae Kim, Jae Hwan Jung, Shin-Hyun Kim, Do Hyun Kim, and Tae Seok Seo* Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-gu, Daejeon 305-701, Republic of Korea ABSTRACT: A novel trace lead ion (Pb2+) detection platform by combining a microfluidic sample pretreatment device with a DNA aptamer linked photoluminescent graphene oxide quantum dot (GOQD) sensor was proposed. The multilayered microdevice included a microchamber which was packed with cation exchange resins for preconcentrating metal ions. The sample loading and recovery was automatically actuated by a peristaltic polydimethylsiloxane (PDMS) micropump with a flow rate of 84 µL/min. Effects of the micropump actuation time, metal ion concentration, pH, and the volume of the sample and eluent on the metal ion capture and preconcentration efficiency were investigated on a chip. The Pb2+ samples whose concentration ranged from 0.48 nM to 1.2 µM were successfully recovered with preconcentration factor (PF) value between 4 and 5. Then, the preconcentrated metal ions were quantitatively analyzed with a DNA aptamer modified GOQD. The DNA aptamer on the GOQD specifically captured the target Pb2+ which can induce electron transfer from GOQD to Pb2+ upon UV irradiation, thereby resulting in the fluorescence (FL) quenching of the GOQD. The disturbing effect of foreign anions on the Pb2+ detection, and the spiked Pb2+ real samples were also analyzed. The proposed GOQD metal ion sensor exhibited highly sensitive Pb2+ detection with a detection limit of 0.64 nM and a dynamic range from 1 nM to 1000 nM. The on-chip preconcentration of the trace metal ions from a large-volume sample followed by the metal ion detection by the fluorescent GOQD sensor can provide an advanced platform for on-site water pollution screening.

Since the growing environmental contamination issues can threaten human health, the sensitive and accurate analysis of heavy metal ions even at trace level is of importance.1 Conventional methods require a sample pretreatment process prior to the metal ion detection. Sample pretreatment should be preceded for the following reasons. i) Raw samples may contain any disturbing substances or matrices that can cause inaccurate metal ion analysis. ii) Some trace metal ion concentration is too low to be detected by the analytical instruments.2 A traditional method for preconcentrating and separating trace metal ions is based on the solid-phase extraction (SPE) due to its high preconcentration factor.3-5 Then, the pretreated metal ion samples were analyzed by bulky instrumentations such as flame atomic absorption spectroscopy (FAAS), electrothermal atomic absorption spectrometry (ETAAS), inductively coupled plasma optical emission spectrometer (ICP-OES), and inductively coupled plasma mass spectrometer (ICP-MS). However, the conventional technologies for metal ion preconcentration are typically performed in a massive column-based SPE which is a timeconsuming process and requires a large amount of samples.

Despite the high sensitivity of the conventional instruments such as FAAS, ICP-OES, and ICP-MS, the accurate analysis of trace metal ions below sub-ppb level still remains as a challenge.6 Furthermore, the metal ion analyzers are complicated, expensive, and bulky, which are not adequate for the on-site metal ion detection and the integration with the sample pretreatment unit. To address these issues, the concept of micro-scale preconcentration was widely explored using a microcolumn SPE with chemically modified adsorbents. For example, Yin et al. presented a flow injection (FI) based preconcentration method with a nanometer-sized alumina packed micro-column.7 Huang et al. utilized a mesoporous titanium dioxide as a SPE matrix for micro-column preconcentration.8 But, these attempts still need tedious manual operation, and the concentrated sample analysis depends on the conventional large-scale system such as ICP-MS and ICP-OES. On the other hand, the microfluidic system or lab-on-a-chip (LOC) technology has demonstrated its great capabilities to overcome the drawbacks of the micro-column based sample pretreatment system due to the merits of automation,

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 15

Figure 1. (A) Schematic illustration of a five-layered sample pretreatment microdevice. The microdevice consisted of a glass manifold, a PDMS thin film, a glass microchannel layer, a PDMS chamber layer, and a glass bottom layer. (B) A digital image of the cation exchange resin incorporated microdevice (top), and the side view of an assembled microdevice (down). (C) Experimental scheme for Pb2+ detection on a GOQD sensor through electron transfer mediated fluorescence quenching. rapidity, portability, and the potential for further integration with in-situ heavy metal ion detector. In this regards, Lafleur et al. established a centrifugal microfluidic SPE device for on-site trace metal preconcentration of water sample.9 Shih et al. suggested a simple open-channel poly(methyl methacrylate) chip for trace metal preconcentration.10 They used microliter-volume of a metal ion sample and determined the metal ion concentration by ICP-MS and laser ablation ICP-MS. In reality, the use of such a small volume of samples is not acceptable to analyze the trace amount of metal ions in real samples, and on-site metal ion detection is not available due to the use of the commercial analyzer at the end.

In this study, we developed an advanced metal ion preconcentrating microdevice as well as a novel metal ion sensor for detecting the preconcentrated trace metal ions from a large volume of samples. We designed the microfluidic sample pretreatment device which is equipped with a large volume of the SPE chamber and a peristaltic pneumatic micropump for automatic sample loading and recovery, enabling the metal ion preconcentration with milliliter-scale samples. Regarding the metal ion sensors, photoluminescent GOQD was employed for simple and sensitive metal ion detection. Recently, graphene quantum dot (GQD), GOQD, and fewlayered nanometer-sized graphene derivatives have garnered great attention due to unique optical properties, facile surface grafting, good chemical stability, and


Page 3 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


photostability of graphene and its derivatives allow them to be utilized as a fluorescent tag in the fields of biological applications.11 Owing to these characteristics, the GQDs and GOQDs based heavy metal sensors have been explored. The GQD based Fe3+ detection using the FL quenching, and the FL chemosensing for Hg2+ ion have been reported.12,13 The GQD-DMA-tryptophan conjugates were also used for Pb2+ detection.14 Herein, we utilized the DNA aptamerlinked GOQDs as Pb2+ sensing materials by measuring the FL quenching of GOQDs. By combining the microfluidic pretreatment device with the GOQD based Pb2+ sensor, we can provide a novel trace metal ion detection platform with high speed, simplicity, sensitivity and selectivity.

EXPERIMENTAL SECTION Chemicals and Materials. Graphite nanoparticles (Avg. diameter: 4 nm, 93%) were purchased from SkySpring Nanomaterials (USA). All metal and anion salts (NaCN, NaF, NaBr, Na2HPO4, NaHCO3, NaNO3, NaNO2, PbN2O6, MnN2O6, CdN2O6, CuN2O6, ZnN2O6, MgCl2, CaCl2, NaCl, NH4Cl, KCl, CoN2O6, AgNO3, HgN2O6, FeN3O9, FeCl2, NiN2O6), 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC), N-hydroxysulfosuccinimide (Sulfo-NHS), and hexamethyldisilazane (HMDS) were obtained from Sigma Aldrich (USA). Hydrochloric acid (HCl) (35%), sodium hydroxide (NaOH) were purchased from JUNSEI (Japan), and deionized (DI) Milli-Q water from Millipore purification system (USA). The sequence of the Pb2+ specific DNA aptamer (5’-GTG GGT AGG GCG GGT TGG3’) was from Bioneer (Korea), and Ambersep GT74 was ordered from SUPELCO (USA). Operation of sample pretreatment on a chip. The design of the heavy metal ion pretreatment microfluidic chip (Figure 1A), the digital image and the cross-sectional view of the assembled microdevice (Figure 1B) is presented. The microdevice was composed of five layers with a glassPDMS-glass-PDMS-glass sandwich structure (from top to bottom), and contained a SPE chamber which was packed with cation exchange resins to capture and release the target metal ions. Details of fabrication procedure are described in supporting information. The assembled microdevice was operated with an external pneumatic pump. The top glass manifold, the second PDMS thin film and the third glass fluidic channel layer functioned as a three-layer microvalve system.15 A fluidic control was accomplished by sequentially actuating the input (I), diaphragm (D) and output (O) valves. The vacuum/pressure was applied to each valve on the manifold sequentially to actuate the micropump with 3 step scheme (I-D-O) (Figure S1). The peristaltic pumping operation could manipulate the sample flow and the pumping actuation time was tuned automatically by inhouse LabVIEW software (National Instruments, USA). Ambersep GT74, one of the commercialized weakly acidic cation exchange resins, was used for selective capture of Pb2+ (Table S1). The ion exchange mechanism of a thiol group of a copolymer resin with metal ions is wellestablished in the previous reports.16 We estimated the exchange capacity of the SPE chamber to preconcentrate

Pb2+. The cation exchange resins were loaded into the SPE chamber by pipetting them through the resin inlet as illustrated in Figure 1B. A Pb2+ sample solution was processed through the SPE chamber, and the waste solution was collected at the outlet reservoir. The Pb2+ concentration of the sample and the waste solution was measured by an ICP-MS (Perkin-Elmer, USA), and the difference of the two concentrations between the initial solution and the recovered sample was used for calculating the capture yield of the metal ion in the cation exchange resins. To elute the captured Pb2+ ion in SPE chamber, a certain amount of 7% HCl solution was injected six times into the sample inlet, and each fraction of the recovered Pb2+ solution which is flew out from the outlet reservoir, was analyzed to measure the Pb2+ concentration depending on the elution time. Detection of Pb2+ on the photoluminescent GOQD sensor. One mL of a DNA aptamer linked GOQD solution (0.01 mg/mL) was transferred to a quartz cuvette, and the photoluminescence intensity was recorded from 400 nm to 650 nm with excitation at 320 nm wavelength (RF-5301 PC spectrophotometer, SHIMADJU, Japan). To obtain the calibration curves (Figure 4C), 1 mM of a Pb2+ stock solution was prepared by dissolving 6.6 mg Pb(NO3)2 in 20 mL deionized water (DI-water), and serially diluted to 100 nM. For sensing the Pb2+ in the GOQD solution, Pb2+ concentration was tuned from 1 nM to 1 μM, and then the photoluminescence intensity was measured. The synthetic method and characterization for GOQD were described in Figure S2 and S3. The selectivity test was performed under the identical experimental conditions as above with other metal ions (Figure 4D). For a negative control, the photoluminescence intensity of the pristine GOQDs, which were not conjugated with the DNA aptamer, was measured in the presence of 0.2 μM Pb2+. The real water sample was collected from drinking water, tap water, and lake water. After spiking a certain amount of target metal ions (Pb2+) in addition to 1 µM of foreign metal ions (Ag+, Hg2+, Fe3+) into the real sample, the sample pretreatment on a chip was performed as described in the section of “Operation of sample pretreatment on a chip”. Then, the eluted sample was neutralized and the obtained Pb2+ was analyzed by the fluorescent DNA aptamer linked GOQD solution as well as ICP-MS.

RESULTS AND DISCUSSION Sample flow rate control and Pb2+ capture efficiency on a chip. Figure 1C shows the overall procedure for the sample pretreatment on a chip and the metal ion detection by the GOQD sensor. First, we evaluated the on-chip Pb2+ preconcentration using 50 nM Pb2+ sample under the optimized micropumping scheme. The pump actuation time, which determines the time to complete one cycle of I-D-O micropump, is an important factor to control the sample flow rate and the metal ion capture yield.17 Thus, we tested different pump actuation time (150, 200, 250, and 300 ms) to optimize the sample flow rate and to maximize the metal ion capture efficiency. As shown in Figure 2A, the sample flow rate proportionally decreased with increase of



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 15

resin on a chip has enough capture capability of Pb2+. As shown in Table S2, the total exchange capacity of Pb2+ on a chip was calculated as 107.2 mg. Since 0.5 M of a Pb2+ solution is equivalent to 103.6 mg/mL of Pb2+, the total exchange capacity of the proposed microdevice is enough to pretreat the Pb2+ sample even at high concentration. Since the effect of pH is one of the important factors for metal ion pretreatment,18 we also studied the capture efficiency of Pb2+ by injecting 3 mL of metal ion samples whose pH was 2-10 into the SPE chamber at a flow rate of 84 μL/min. Figure 2B inset shows that the Pb2+ capture efficiencies in the SPE chamber were almost 94% under acidic conditions. However, the capture efficiencies significantly decreased under alkaline conditions. This phenomenon can be interpreted based on the hydroxyl complexes with the metal ions. Pb2+ exists in various forms at different pH. The dominant species were Pb2+ (> 80%) until pH 7, but the composition of the lead complexes was changed to Pb2+ (~50%), Pb(OH)+ (~45%), Pb3(OH)42+ (~3.0%), and Pb(OH)2 (~1.0%) at pH 8.19 The variety of the Pb2+ forms over pH 8 resulted in the reduced Pb2+ capture efficiency in SPE chamber. At pH 7.0, the highest metal ion capture efficiency was obtained (98.9 ± 1%), so pH 7 was chosen as an optimized pH value for further experiments.

Figure 2. (A) Sample flow rate and Pb2+ capture efficiency versus an actuation time of a micropump. (B) On-chip capture efficiency versus the concentration of Pb2+. Inset: the capture efficiency of Pb2+ at different pH value (Concentration of sample was fixed to 10 nM). pump actuation time (green symbols), indicating that the increased actuation time resulted in retarding the vacuum/pressure cycle. In order to treat a large volume of a sample, fast actuation time would be ideal. In case of 150 ms actuation time, a sample could be processed with a rate of 84 μL/min. On the other hand, the Pb2+ capture yield in the SPE chamber was almost 90% regardless of the actuation time (grey bars), which means that the pump actuation time influences the sample flow rate, not the metal ion capture efficiency. We also performed the same experiments with different concentration of the Pb2+ solution, and Figure S4 shows that the metal ion capture efficiency was almost equivalent as 90% regardless of flow rate and sample concentrations. Therefore, a flow rate of 84 μL/min with 150 ms actuation time was chosen for sample injection and elution in all further experiments. Figure 2B shows the Pb2+ capture efficiency depending on the concentration and the pH (inset) of a sample. The effect of the sample concentration was investigated by varying the Pb2+ concentration from 0.5 nM to 1.2 μM. The capture efficiency reached 92-99% for the entire range, which indicates that the incorporated cation exchange

Preconcentration on a chip. In order to evaluate the on-chip preconcentration ability for the trace metal ions from a large volume sample, the captured Pb2+ in the SPE chamber was eluted, and then analyzed by ICP-MS. 7% HCl was selected as an eluent according to the specification of Ambersep GT74. We eluted the captured Pb2+ in the SPE chamber by using 100 μL 7% HCl, and repeated the elution process six times. The preconcentration factor (PF) is defined as PF = RF × (Va/Ve), where the recovery factor (RF) is the released Pb2+ quantities divided by the captured Pb2+ quantities, Va is the volume of the loaded sample, and Ve is the volume of an eluent solution. To maximize the PF, the volume of the sample solution should be increased or the volume of the eluting solution should be minimized.20,21 Thus, we assessed the PF values with variation of the volume of the sample and eluent, while considering the consumed process time. The effect of the eluent volume on the PF was shown in Figure 3A. With the sample volume fixed at 3 mL, we measured the RF, PF, and the process time by changing the eluent volume to 120, 300, and 600 μL. The Va/Ve value was 25, 10, and 5 with the eluent volume of 120, 300, and 600 μL, and the corresponding RF value was 0.13, 0.23, and 0.91, which resulted in the PF value of 3.13, 2.32, and 4.49, respectively. Notably, the RF value with the eluent volume of 600 μL reached 91%, meaning that the sufficient amount of the eluent would be necessary to quantitatively recover the captured Pb2+ on the cation exchange resin. Besides the eluent volume, the sample volume also plays an important role to determine the PF on a chip. We measured the RF, PF, and the process time by changing the sample volume to 3, 6, and 9 mL, while the eluent volume was fixed at 600 μL. At the sample volume of 3, 6, and 9 mL, the RF value was 0.97, 0.87, and 1.11, the PF value was 4.97, 8.65, and 14.9, and the total processing time was 45.7, 81.4,


Page 5 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. (A) Effect of the eluent volume (120, 300, and 600 µL) on the PF. (B) Effect of the sample volume (3, 6, and 9 mL) on the PF. The concentration of the samples was 200 nM. (C) RF of Pb2+ versus the eluent volume at nano-molar range of concentration. (D) RF of Pb2+ versus the eluent volume at micro-molar range of concentration. and 117 min, respectively. Although the processing time took around 2 h, the high PF value of 14.9 was obtained on a chip even with a 9 mL sample solution, demonstrating the efficient sample pretreatment capability with a large volume of a metal ion sample. Furthermore, we evaluated the PF depending on the sample concentration with the sample volume of 3 mL, and the eluent volume of 600 μL. The tested concentration of the samples ranged from the nanomolar (0.48, 1.45, 13, 28 nM) (Figure 3C) to the micromolar concentration (0.2, 1.2, 2.4 μM) (Figure 3D). At the concentration of 0.48 and 1.45 nM, the RF value was 0.82 and 0.86, and the PF value was 4.06 and 4.18, respectively. In case of 13 and 28 nM, the RF value was almost 1, and the PF value thus slightly increased to 4.94 and 5.09, respectively. On the other hand, above the concentration of 1 μM, the PF values became decreased. At the concentration of 0.2 μM, the RF value shows 1.04, but at 1.2 and 2.4 μM, the RF value decreased to 0.8 and 0.36. Accordingly, the PF value was reduced to 3.87 and 1.72. The reason for the low RF value at relatively high concentrations might be attributed to the reduced releasing capacity with the 600 μL eluent solution.

However, since the low concentration of the metal ion could be properly concentrated on the proposed microdevice, it is adequate for the accurate analysis of the trace metal ion. Thus, through the optimization of the volume of the sample, the eluent, and the micropumping operation, we could automatically preconcentrate the trace metal ion with high PF on a chip, which enables the sensitive Pb2+ detection with the GOQD sensor in the downstream process. Detection of Pb2+ on the photoluminescent GOQD sensor. The amino-modified Pb2+ specific DNA aptamer (5’-NH2-(CH2)6-GTG GGT AGG GCG GGT TGG) was conjugated on the GOQDs through a carbodiimide chemistry to produce the DNA aptamer linked GOQD conjugates. Figure 4A shows the UV-vis absorption spectra and N 1s XPS (inset) for the pristine GOQDs and the DNA aptamer conjugated GOQDs. In comparison with the pristine GOQDs (black line), a 260 nm peak was observed from DNA aptamer-GOQD conjugates due to the DNA molecules (blue line). The distinct peak of N 1s XPS at 399.7 eV (red line) confirms the successful DNA aptamer linkage on the GOQD.22 Figure 4B displays the FL spectra of the DNA-GOQD conjugates depending on the Pb2+



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 15

Figure 4. (A) UV-vis absorption spectra for the pristine GOQD (black) and the DNA aptamer conjugated GOQDs (blue). Inset: N 1s XPS for the pristine GOQD (black) and the DNA aptamer conjugated GOQDs (red). (B) Fluorescence spectra for the DNA aptamer conjugated GOQD upon the addition of Pb2+ from 0 to 1000 nM with excitation at 320 nm. (C) rQE (%) depending on the Pb2+ concentration. Inset: a linear plot of the F0/F versus [Pb2+] according to the Stern-Volmer equation. (D) Selectivity test of the DNA aptamer linked-GOQD for sensing Pb2+. The blue bar indicates the rQE of the GOQD without DNA aptamer in the presence of Pb2+. The concentration of all used metal ions was 200 nM. concentration (1, 2.5, 5, 10, 25, 50, 100, 250, 500, 1000 nM) with excitation of 320 nm by monochromatic diffraction light of a Xenon lamp. When the Pb2+ was added, it was specifically captured in the sequence of DNA aptamer and then G-quadruplex complex was formed.23 Then, the FL intensity of GOQD was quenched due to electron transfer from the GOQD to the Pb2+. The electron transfer mediated FL quenching mechanism of the GOQD with metal ions has been investigated in our previous work.24 Thus, as the concentration of Pb2+ increased from 0 to 1 μM, the FL intensity of GOQD gradually decreased. The FL intensity was maintained without significant change at least for 2 h (Figure S5). In Figure 4C, we plotted the relative quenching efficiency (rQE, %) versus the Pb2+ concentration, and the rQE was calculated from the following equation. rQE (%) = (1–F/F0)×100

where F is the integrated FL intensity in the presence of Pb2+ and F0 is that of the pristine GOQD. As shown in Figure 4C, the rQE values increased in proportional to the Pb2+ concentration. The inset of Figure 4C shows a linear relationship between the FL intensity ratio (F0/F) and the Pb2+ concentration according to the Stern-Volmer equation over a broad range of Pb2+ concentration. The Stern-Volmer plot is one of the well-established principles for explaining the relation between the fluorophore and the quencher.25 The dynamic range of the GOQD based Pb2+ sensor was from 1 to 1000 nM and the calibration curve was described as F0/F=2.6×10-3×[Pb2+]+1.09 with the corresponding regression coefficient (R) of 0.972. The limit of detection (LOD), calculated by 3×(SD/S), was 0.64 nM, where SD is the standard deviation of a blank solution and S is the slope of the calibration curve. The high sensitivity of the GOQD sensor satisfies the detection criterion defined by the U.S. Environmental Protection Agency (EPA) for Pb2+ in drink water (72 nM),26 and the detection


Page 7 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


sensitivity is comparable to other graphene based Pb2+ sensors.14,27-32 Since the sample pretreatment microfluidic device could preconcentrate Pb2+ and remove foreign anions, the sensitive and selective Pb2+ detection on the GOQD sensor was significantly improved. To evaluate the selectivity of the GOQD based Pb2+ sensor for sensing Pb2+, we measured the rQE (%) of the DNA aptamer-GOQD in the presence of other metal ions including 200 nM Mn2+, Mg2+, Cd2+, Ca2+, Na+, Cu2+, NH4+, K+, Zn2+, Hg2+, Ag+, Co2+, Ni2+, Fe2+, and Fe3+. As a negative control, we measured the rQE of the pristine GOQDs, which were not conjugated with DNA aptamers, by adding 200 nM of Pb2+ (Figure 4D, a blue bar). The rQE toward the Pb2+ of the pristine GOQDs showed only 3%, while that of the DNA aptamer modified GOQDs showed 31% which indicates the specific capture ability of DNA aptamer for Pb2+. These results suggest that the conjugation of the DNA aptamer on the GOQD is critical to induce significant quenching of the GOQD with high selectivity. In addition, the linkage of the DNA on the GOQDs not only reduces the possibility of the non-specific metal ion interaction with the oxygen functional groups of GOQDs, but also imparts steric hindrance to prevent the metal ions from approaching the GOQD surface. Thus, our DNA aptamer linked GOQDs based fluorescent sensor can be utilized for Pb2+ detection with high selectivity. The disturbing side effect of foreign ions on the FL signal of the GOQD. One of the important advantages of our microdevice is to remove foreign anions using cation exchange resins during the pretreatment step. Figure 5 shows that the anions could hinder the GOQD based fluorescent detection of Pb2+ due to broad quenching variance. The same experimental procedure was followed with addition of certain anions (CN-, F-, Br-, PO42-, HCO3-, NO3-, NO2-). The double-headed arrows indicate the fluctuation range of the quenched FL signals of the GOQD. The presence of the anions led to the FL quenching of the pristine GOQD more or less (the largest deviation from the original quenching was calculated as 13.2% in the presence of PO42-) and the broad signal variance. These results could be ascribed to the unexpected π-interaction between the surface of aromatic group and anions,33 and such an interaction might disturb the precise sensing of our GOQD based Pb2+ sensor. To solve this problem, the foreign anions should be removed in advance. Because the cation exchange resins were used in the SPE chamber, the anions in the sample could be eliminated. As mentioned above, the Ambersep GT74 resin possesses only cation capture property, not for anion. As shown in Table S3, the anion concentration of the recovered solution was almost zero, because the anions were flushed out during the washing step on a chip. Thus, any anions that hinder the fluorescent detection of Pb2+ were successfully removed, so we could perform the accurate and reproducible detection of Pb2+ using the DNA aptamer linked GOQD. Measurement of Pb2+ in the real samples. We validated the applicability of the GOQD based Pb2+ sensor using real samples. The real aqueous samples were collected from drinking water, tap water and Duck Lake in

Figure 5. The disturbing effect of foreign ions. Each dot indicates the maximum FL intensity of the GOQDs depending on the types of anions, and the double-headed arrow shows the variance width of the quenched FL signal in the presence of anions. All experiments were performed for three times. The concentration of F-, Br-, PO42-, NO3- , and NO2- was 0.43, 1.36, 1.53, 1.09, and 0.98 ppm, while Pb2+ was 200 nM. Table 1. Recovery of Pb2+ after on-chip pretreatment using real water samples in the presence of foreign metal ionsc. Type

Drinking Water

Tap Water

Lake Water a Mean

Found (nM)

Recovery (%)

ICP-MS (nM)

25

26a ± 0.7b

104

25.1 ± 0.2

50

47.7 ± 1.8

95.4

43.8 ± 0.3

100

97.7 ± 1.6

97.7

103.6 ± 0.7

5

4.1 ± 1.8

82

4.6 ± 0.3

50

58.2 ± 1.9

116.4

53.7 ± 0.3

100

103.5 ± 1.9

103.5

101.2 ± 0.4

25

26.1 ± 1.9

104.4

24.2 ± 0.7

100

95.4 ± 4.3

95.4

99.3 ± 0.1

values for triplicate experiments.

b Standard c

Spiked (nM)

deviation

Each real sample contains 1 μM of Ag+, Hg2+, and Fe3+.

KAIST (Daejeon, Korea), and then were briefly filtered using centrifuge and spiked with Pb2+. Considering the potential disturbing effect of other metal ions on the fluorescence quenching of the DNA aptamer linked GOQD sensor, the other transition metal ions such as Ag+, Hg2+, and Fe3+ were also added in the real samples. For comparison, we also analyzed the same sample using ICPMS. As shown in Table 1, the recovery of Pb2+ in real samples was in the range of 82-116% despite the presence of interferential metal ions, and SD was less than 5, which implies that our GOQD sensor coupled with microfluidic sample pretreatment has great potential for practical applications for detecting Pb2+ in real samples. Compared



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

with the values of the ICP-MS, the Pb2+ concentration measured by our sensing system revealed only 2.3% deviation, which implies that the GOQD sensor for metal ion detection is quite reliable.

CONCLUSIONS In conclusions, we demonstrated a novel microfluidic based metal ion preconcentrating device and a metal ion sensor using a DNA aptamer linked photoluminescent GOQD. During the sample pretreatment process on a chip, a trace level of Pb2+ were preconcentrated and impurities such as foreign anions that disturb the fluorescent sensing of GOQD were successfully removed. Then the recovered Pb2+ was quantitatively analyzed by the DNA aptamer modified photoluminescent GOQD without using any extra reagent. The combination of these two functional units allows us to perform the trace metal ion analysis with high selectivity, selectivity and speed. The integration of the GOQD sensor into the sample pretreatment microdevice will provide us with an advanced water pollution screening tool for on-site metal ion detection with sample-in answer-out capability, which is under way in our laboratory.

ASSOCIATED CONTENT Supporting Information. Additional information as noted in text. The Supporting Information is available free of charge on the ACS publications website at DOI:

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]. Fax: +82-42-350-3910.

ACKNOWLEDGMENT This work was supported by Converging Technology Project funded by the Korean Ministry of Environment (M112-000610002-0) and the Engineering Research Center of Excellence Program of Korea Ministry of Science, ICT & Future Planning (MSIP) / National Research Foundation of Korea (NRF) (Grant NRF-2014R1A5A1009799).

REFERENCES (1) Liu, Y.; Chang, X.; Guo, Y.; Meng, S. J. Hazard. Mater. 2006, B135, 389-394. (2) Daorattanachai, P.; Unob, F.; Imyim, A. Talanta 2005, 67, 59-64. (3) Korn, M. G. A.; Andrade J. B.; Jesus D. S.; Lemos, V. A.; Bandeira M. L. S. F.; Santos W. N. L.; Bezerra, M. A.; Amorim, F. A. C.; Souza, A. S.; Ferreira, S. L. C. Talanta 2006, 69, 16-24. (4) Sharma, R. K.; Mittal, S.; Koel, M. Crit. Rev. Anal. Chem. 2003, 33 (3), 183-197. (5) Camel, V. Spectrochim. Acta Part B 2003, 58, 1177-1233. (6) Lum, T. S.; Tsoi, Y. K.; Leung, K. S. Y. J. Anal. At. Spectrom. 2014, 29, 234-241. (7) Yin, J.; Jiang, Z.; Chang, G.; Hu, B. Anal. Chim. Acta 2005, 540, 333-339. (8) Huang, C.; Jiang, Z.; Hu, B. Talanta 2007, 73, 274-281. (9) Lafleur, J. P.; Salin, E. D. J. Anal. At. Spectrom. 2009, 24, 15111516. (10) Shih, T. T.; Chen, W. Y.; Sun, Y. C. J. Chromatogr. A 2011, 1218, 2342-2348.

Page 8 of 15

(11) Liu, Q.; Guo, B.; Rao, Z.; Zhang, B.; Gong, J. R. Nano Lett. 2013, 13, 2436-2441. (12) Ananthanarayanan, A.; Wang, X.; Routh, P.; Sana, B.; Lim, S.; Kim, D. H.; Lim, K. H.; Li, J.; Chen, P. Adv. Funct. Mater. 2014, 24, 3021-3026. (13) Chakraborti, H.; Sinha, S.; Ghosh, S.; Pal, S. K. Mater. Lett. 2013, 97, 78-80. (14) Qi, Y. X.; Zhang, M.; Fu, Q. Q.; Liu, R.; Shi, G. Y. Chem. Commun. 2013, 49, 10599-10601. (15) Grover, W. H.; Skelley, A. M.; Liu, C. N.; Lagally, E. T.; Mathies, R. A. Sens. Actuators, B 2003, 89, 315-323. (16) Podkościelna, B.; Kołodyńska, D. Polym. Adv. Technol. 2013, 24, 866-872. (17) Lee, C. J.; Jung, J. H.; Seo, T. S. Anal. Chem. 2012, 84, 49284934. (18) Mendil, D.; Kiris, T.; Tuzen, M.; Soylak, M. Int. J. Food Sci. Technol. 2013, 48, 1201-1207. (19) Baes Jr. C. F.; Mesmer, R. E. The Hydrolysis of Cations 1976, p. 364 (20) Sharma, R. K.; Pant, P. J. Hazard. Mater. 2009, 163, 295-301. (21) Ersöz, A.; Say, R.; Denizli, A. Anal. Chim. Acta 2004, 502, 91-97. (22) Tetsuka, H.; Asahi, R.; Nagoya, A.; Okamoto, K.; Tajima, I.; Ohta, R.; Okamota, A. Adv. Mater. 2012, 24, 5333-5338. (23) Li, T.; Dong, S.; Wang, E. J. Am. Chem., Soc. 2010, 192, 1315613157. (24) Ha, H. D.; Jang, M. H.; Liu, F.; Cho, Y. H.; Seo, T. S. Carbon 2015, 81, 367-375. (25) Lankowicz, J. R. Principles of Fluorescence SpectroscopyThird Edition 2006, p. 277-289. (26) National Primary Drinking Water Regulations: National Report; Publication No. EPA816F09004; United States Environmental Protection Agency (US EPA), 2009 (27) Zhao, X. H.; Kong, R. M.; Zhang, X. B.; Meng, H. M.; Liu, W. N.; Tan, W.; Shen, G. L.; Yu, R. Q. Anal. Chem. 2011, 83, 50625066. (28) Wen, Y.; Peng, C.; Li, D.; Zhuo, L.; He, S.; Wang, L.; Huang, Q.; Xu, Q. H.; Fan, C. Chem. Commun. 2011, 47, 6278-6280. (29) Li, X.; Wang, G.; Ding, X.; Chen, Y.; Gou, Y.; Lu, Y. Phys. Chem. Chem. Phys. 2013, 15, 12800-12804. (30) Li, M.; Zhou, X.; Guo, S.; Wu, N. Biosens. Bioelectron. 2013, 43, 69-74. (31) Bai, Y.; Zhao, L.; Chen, Z.; Wang, H.; Feng, F. Anal. Method 2014, 6, 8120-8123. (32) Fu, X.; Lou, T.; Chen, Z.; Lin, M.; Feng, W.; Chen, L. ACS Appl. Mater. Inter. 2012, 4, 1080-1086. (33) Shi, G.; Ding, Y.; Fang, H. J. Comput. Chem. 2012, 33, 13281337.


Page 9 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table of Contents

KEWORDS: microfluidics, sample pretreatment, graphene quantum dots, heavy metal detection


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. (A) Schematic illustration of a five-layered sample pretreatment microdevice. The microdevice consisted of a glass manifold, a PDMS thin film, a glass microchannel layer, a PDMS chamber layer, and a glass bottom layer. (B) A digital image of the cation exchange resin incorporated microdevice (top), and the side view of an assembled microde-vice (down). (C) Experimental scheme for Pb2+ detection on a GOQD sensor through electron transfer mediated fluorescence quenching. 175x156mm (300 x 300 DPI)


Page 10 of 15

Page 11 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. (A) Sample flow rate and Pb2+ capture efficiency versus an actuation time of a micropump. (B) On-chip capture efficiency versus the concentration of Pb2+. Inset: the capture efficiency of Pb2+ at different pH value (Concentration of sample was fixed to 10 nM). 80x125mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. (A) Effect of the eluent volume (120, 300, and 600 µL) on the PF. (B) Effect of the sample volume (3, 6, and 9 mL) on the PF. The concentration of the samples was 200 nM. (C) RF of Pb2+ versus the eluent volume at nano-molar range of concentration. (D) RF of Pb2+ versus the eluent volume at micro-molar range of concentration. 160x125mm (300 x 300 DPI)


Page 12 of 15

Page 13 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. (A) UV-vis absorption spectra for the pristine GOQD (black) and the DNA aptamer conjugated GOQDs (blue). Inset: N 1s XPS for the pristine GOQD (black) and the DNA aptamer conjugated GOQDs (red). (B) Fluorescence spectra for the DNA aptamer conjugated GOQD upon the addition of Pb2+ from 0 to 1000 nM with excitation at 320 nm. (C) rQE (%) depending on the Pb2+ concentration. Inset: a linear plot of the F0/F versus [Pb2+] according to the Stern-Volmer equation. (D) Selectivity test of the DNA aptamer linked-GOQD for sensing Pb2+. The blue bar indicates the rQE of the GOQD without DNA aptamer in the presence of Pb2+. The concentration of all used metal ions was 200 nM. 160x132mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. The disturbing effect of foreign ions. Each dot indicates the maximum FL intensity of the GOQDs depending on the types of anions, and the double-headed arrow shows the variance width of the quenched FL signal in the presence of anions. All experiments were performed for three times. The concentration of F-, Br-, PO42-, NO3- , and NO2- was 0.43, 1.36, 1.53, 1.09, and 0.98 ppm, while Pb2+ was 200 nM. 80x59mm (300 x 300 DPI)


Page 14 of 15

Page 15 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table of Contents 85x45mm (300 x 300 DPI)


Combination of a Sample Pretreatment Microfluidic ... - ACS Publications

Recommend Documents