ANALYTICAL SCIENCES MAY 2016, VOL. 32 491

Introduction

Rapid screening of pollutants in an ambulatory environment is one of the most pressing needs for pollution control. Increasing threats of water contaminants (biological and chemical) have urged researchers to develop rapid tests or methods for their quantification.1 Inorganic heavy metal ions are a major source of chemical pollutants and have led to fatal disorders in living species. For their detection, various analytical tools2–6 are available that are not only expensive but also involve complex operational protocols, thus limiting their application to laboratory usage only. Hence, an upsurge of interest in the development of low cost, easy to operate, highly sensitive and reliable methods for routine detection of heavy metal ions has been reported in recent years.7

Recently, efforts have been being made to develop rapid assays using low-cost cellulose based platforms.7–10 A classic example is the pregnancy test designed on a nitrocellulose substrate. Cellulose based “filter paper” substrates have several advantages as they are economical and readily available as compared to nitrocellulose membranes. The high capillary action of the filter paper makes them an excellent platform for rapid assays in analytical and bioanalytical applications.10–13 Arciuli et al.7 had demonstrated colorimetric detection of phenolic compounds in liquid beverage where the LOD of the paper sensor was 5 μM and a linear response was up to 0.5 mM.

Earlier, Nie et al.10 also designed a microfluidic paper-based electrochemical device (μPEDs) for the detection of heavy metal ions using anodic strip voltammetry. But fabrication of microchips or microfluidic channels onto the substrate surface requires a substantial amount of knowledge. Chao et al.8 reported a gold nanoparticle based lateral flow assay for the detection of Hg2+ in water, which was easy to use and read. β-Galactosidase based sensor has been reported for simultaneous detection of six different heavy metals on the paper platform.9 Even though they fulfilled the necessity of a multiplexed sensor, the use of any form of biomolecules increases the cost per test.

It has been observed that the hydrophilic nature of filter paper serves as a poor barrier for water vapor and this limits its application. Thus, improving the hydrophobicity of the hydrophilic substrate represents an interesting area of research where researchers have come up with different modification mechanisms using HMDS,14,15 wax16–18 and polystyrene.19,20 Wax and polystyrene alter the physical properties of cellulose material, and their application in producing hydrophobic barriers. On the other hand, the application of HMDS modified filter paper has been limited for phase separator applications. Hence, the present article is focused on: A) the application of HMDS modified paper as a platform for rapid detection of Hg2+ using 6-hydroxy-3-(2-oxoindolin-3-ylideneamino)-2-thioxo-2H-1,3-thiazin-4(3H)-one (HOTT), B) digitalizing the color intensity of the assay spots by flatbed scanning and scoring the pixel gray volumes21,22 (PGV) using image analysis software, and C) demonstrating the application of the developed method for detection of the tested metal ions in water samples.

We also confirmed the feasibility of HMDS modified paper

2016 © The Japan Society for Analytical Chemistry

† To whom correspondence should be addressed.E-mail: lmujawar@kau.edu.sa

Hexamethyldisilazane Modified Paper as an Ultra-sensitive Platform for Visual Detection of Hg2+, Co2+, Zn2+ and the Application to Semi-quantitative Determination of Hg2+ in Wastewater

Liyakat Hamid MUJAWAR,*† Adel Abdulaziz FELEMBAN,** and Mohammad Soror EL-SHAHAWI**

*Center of Excellence in Environmental Studies, King Abdulaziz University, Jeddah, Saudi Arabia ** Department of Chemistry, Faculty of Science, King Abdulaziz University, P. O. Box 80203, Jeddah 21589,

Saudi Arabia

The present article reports the application of hexamethylsilazane (HMDS) modified filter paper for ultrasensitive detection of Hg2+, Co2+ and Zn2+. By chemical vapor deposition of HMDS, a highly hydrophilic filter paper was fabricated to a low wetting (hydrophobic) substrate. The water contact angle (θ) of modified paper was ~128°, whereas scanning electron and atomic force microscopy confirmed the surface modification. Using chromogenic reagents, a one-step assay for aforementioned ions was demonstrated onto pristine as well as hydrophobic paper. The assay was completed in less than 10 min and the end-result was in form of a color change that could be easily read by the naked eye. The limit of detection on modified paper was 0.5 ppb, which was 5-order of magnitude superior to that observed on pristine paper. The proposed method was successfully applied for semi-quantitative determination of Hg2+ ions in real wastewater samples.

Keywords Hexamethyldisilazane, hydrophobic, filter paper, surface modification, one-step assay, detection

(Received August 31, 2015; Accepted October 20, 2015; Published May 10, 2016)

Original Papers

492 ANALYTICAL SCIENCES MAY 2016, VOL. 32

for the detection of Co2+ and Zn2+ using other chromogenic reagents such as 4-(5-bromo-2-pyridylazo)-N,N-diethyl-3-hydroxy-aniline (5-Br-PADAP) and dithizone (H2DZ), respectively.

Experimental

Reagents and materialsThe reagent 6-hydroxy-3-(2-oxoindolin-3-ylideneamino)-2-

thioxo-2H-1,3-thiazin-4(3H)-one (HOTT) (Fig. 1A) was prepared as reported.23 Dithizone (H2DZ) and 4-(5-bromo-2-pyridylazo)-N,N-diethyl-3-hydroxyaniline (5-Br-PADAP) (Figs. 1B and 1C) were purchased from BDH (Poole, England) and Fluka AG (St. Gallen, Switzerland), respectively. HgCl2, Co(NO3)2·6H2O and Zn(NO3)2·6H2O were also obtained from BDH (Poole, England). Whatman filter paper (Grade 4) was obtained from GE Whatman (Buckinghamshire, UK) and standard plain micro glass slides were obtained from HiMedia (Mumbai, India). Paraffin wax was delivered by Techno Pharchem (Haryana, India) and styrofoam sheets were purchased from a local vendor. A series of Britton–Robinson buffer (pH 3 – 9) was prepared as described earlier.23 Also, 0.25% (w/v) 5-Br-PADAP and H2DZ were prepared individually in ethanol and were used for the detection of trace levels of Co2+ and Zn2+ ions, respectively.

EquipmentsNon-contact mode atomic force microscope (AFM) and a

scanning electron microscope (SEM) (JEOL-JSM6301-F) (Peabody, MA) were used for characterization of the surface topography and roughness of pristine and modified filter paper substrates. For AFM analysis, an area of 1 × 1 μm was scanned (256 lines at a frequency of 0.4 Hz) using Park Systems NX10 AFM (Suwon, South Korea). Double distilled water was obtained from a Milli-Q Plus system (Millipore, MA) and was used for the preparation of standard solutions. A  Jenway pH meter (Model 3505, UK) and digital-micro-pipette (Volac) were used for pH measurements. Contact angle measuring system (Krüss DSA30S, Hamburg, Germany) was used for measuring the contact angle (θ) of a sessile water droplet on pristine and modified surfaces. Static contact angle and hysteresis measurements were also performed.

Modification of hydrophobic substratesWhatman filter paper was cut into strips (12 × 1.5 cm). Then,

10 mL HMDS was placed onto a petri plate enclosed into a desiccator and allowed to saturate with silane vapors for 30 min. Paper strips were placed onto a perforated plate of the desiccator and allowed to incubate in the desiccator for 24 h (SI-1A, Supporting Information). Simultaneously, glass slides were cleaned with ethanol and silanized by HMDS as demonstrated earlier by Mujawar et al.24 Preparation of the hydrophobic paper from polystyrene (PS) and wax was performed by the protocol discussed earlier.17,25

One-step assayDiluted reagents (HOTT, 5-Br-PADAP, H2DZ) were manually

arrayed on pristine and HMDS modified paper using Capp pipette (Odense S, Denmark). Dispensing was executed at room temperature (~21°C) where the volume of the supplied ligand droplet was ~600 nL. Printed paper substrates were air-dried at room temperature for 15 min to ensure complete evaporation of the solvent. Simultaneously, stock solutions (1000 ppm) were prepared in Milli-Q water from HgCl2, Co(NO3)2·6H2O, Zn(NO3)2·6H2O. A series of diluted samples (102, 101, 100, 10–1, 10–2, 10–3 ppm) were also prepared in Milli-Q water from the stock solutions. On pristine paper, 3 μL of heavy metal ion solution was supplied to the reagent spots whereas on hydrophobic paper the supplied volume was 10 μL. The hydrophobic paper was incubated at 45°C for 10 min in order to accelerate the drying process (SI-1B, Supporting Information).

Contact angle studiesStatic: To evaluate the static contact angle (θ), an accurate

volume (1 μL) of Milli-Q water droplet was placed at three different locations on each paper substrate. The static contact angle was measured by Tangent-1 (T-1) method using drop analysis software (DSA-4). The average contact angle (θ) was calculated and data were presented along with standard deviation (±SD).

Hysteresis: In order to estimate surface in-homogeneity, a  contact angle hysteresis experiment was performed using Milli-Q water. A  goniometer was programmed to dispense Milli-Q water at the flow rate of 20 μL/min starting with an initial volume of 1 μL, advancing up to 30 μL and receding back to 1 μL. The change in contact angle per second was measured by Tangent (T-1) method. Three independent sets of experiments were performed at various locations on each substrate surface in order to obtain average advancing and receding contact angles.

Scanning and image analysisThe color intensity of the ligand spots on pristine and

hydrophobic papers was digitalized by scanning them on a flatbed scanner (Fujitsu, fi-6230Z, Japan) at a resolution of 1200 dpi and 24-bit color scale. The intensity of each spot was calculated by uploading the scanned images in image analysis software (Image J). The original scanned image file (RGB) was transformed to 8-bit grayscale format since the pixel gray volume (PGV) obtained from a grayscale image is linearly scaled from min-max to 0 – 255. A  circular region of interest was placed at the circumference of the colored area and mean PGV was calculated per spot. A  set of three individual experiments was performed and average PGV was calculated at respective concentrations ± standard deviation (SD, n = 3).

Analytical applicationsThe proposed HMDS modified paper was applied for semi-

quantitative detection of heavy metal ions in tap- and municipal wastewater samples. Tap water was spiked with a known

Fig. 1　Chemical structures of reagents: HOTT (A), dithizone (B) and 5-Br-PADAP (C).

ANALYTICAL SCIENCES MAY 2016, VOL. 32 493

concentration (1 ppm) of Hg2+ and analyzed following the recommended experimental procedures described in section One-step assay. The details on the collection of wastewater samples, their analysis by ICP-OES and IC are presented in SI-9 (Supporting Information). These samples were analyzed on HMDS modified paper in order to semi-quantitatively determine the presence of aforementioned heavy metal ions by color formation and PGV analysis.

Results and Discussion

Goniometer measurementsThe hydrophobicities of modified filter papers (PS, wax and

HMDS) and HMDS silanized glass were characterized by determining the static contact angle (θ) of a water droplet. The average contact angles along with their standard deviations are listed in Table 1. The water contact angle (θ) on polystyrene modified paper was 110.9 ± 0.5°, which was higher than reported on a standard HTA polystyrene slide (θ ~92°).24,26 Even though Abe et al.25 already demonstrated the application of PS in producing barriers for hydrophilic channels, the consequence of PS modification on the cellulose surface was not reported. A significant difference in advancing and receding contact angle was observed on PS modified filter paper (Table 1). The observed hysteresis may be to partial wetting of PS modified surface and that may influence the shape of the receding droplet thus causing higher contact angle hysteresis (37.5 ± 4.4°).

The application of paraffin wax for producing hydrophobic paper has already been demonstrated by Songjaroen et al.17 We followed a similar protocol for producing wax-modified paper where the average contact angle was 117.7 ± 1.2°. In this case, too, hysteresis was observed as the receding contact angle value was lower than the advancing contact angle, thus showing high hysteresis of 30.3 ± 5.6° (Table 1). On the other hand, we found that the wettability of HMDS modified filter paper was less as compared to other modified paper substrates. The water contact angle (θ) on HMDS modified paper was 127.6 ± 0.9° whereas on HMDS silanized glass the value was 99.9 ± 0.8°. The application of HMDS in improving the hydrophobicity of cellulose-based material had already been discussed in detail by Chinga-Carrasco et al.27 According to them, silane vapor reacts with free OH-groups of cellulose fibers and incorporate hydrophobic methyl groups on them which in turn impart low wettability to the cellulose material. HMDS vapor adsorbs on the substrate surface as uniform layers, thus acts as a potential agent in improving the substrate’s hydrophobicity. Earlier, Mujawar et al.24 successfully demonstrated the application of HMDS to produce hydrophobic glass slides, which were used as a platform for producing high quality microarrays. We observed low hysteresis on HMDS modified paper as compared to other surfaces (26.5 ± 3.6°) (Table 1).

Atomic force microscopy (AFM) analysisTo understand the influence of surface modification on the

topography and roughness of the substrate, AFM analysis was performed on pristine filter paper and compared to modified (HMDS, PS and wax) substrates. The optical images of all four paper substrates did not highlight any significant differences in surface topography as shown in Fig. 2 (A – D). Upon scanning, no dissimilarity in the surface morphology of pristine and HMDS modified paper was observed (Figs. 2E and 2F). This indicates that the adsorption of silane did not exert any morphological changes on the cellulose fibers. Even though PS modification also did not prompt any visible changes (via the naked eye), but AFM analysis revealed partial coverage of cellulose fibers with the polymer (Fig. 2G). On the other hand, wax modification resulted in a complete coverage of the organic compound (i.e., wax) on the cellulose substrate (Fig. 2H).

The influence of various modifications on the surface roughness of filter paper substrates was critically investigated and is listed in SI-2 (Supporting Information). The topography of cellulose fibers for pristine and HMDS modified paper was similar, but upon surface modification, a significant change in the roughness was noticed. The average (Ra) and RMS (Rq) roughness of HMDS modified paper was 16 and 13 μm, respectively. These values were ~60% less than what was observed for pristine filter paper. The curtailed roughness with minimum morphological changes of the cellulose surface may cause low hysteresis on HMDS modified paper. The Ra and Rq

Table 1　Water contact angle (θ) on various surfacesa

Whatman filter paper Glass

Polystyrene Paraffin wax HMDS HMDS

Static CA 110.9 ± 0.5°117.7 ± 1.2°127.6 ± 0.9° 99.9 ± 0.8°Advancing CA 105.1 ± 4.7°118.1 ± 5.9°132.6 ± 1.9° 99.8 ± 0.8°Receding CA 67.6 ± 18.0°87.8 ± 21.3°101.8 ± 19.9°71.9 ± 12.6°Hysteresis 37.5 ± 4.4° 30.3 ± 5.6° 26.5 ± 3.6° 27.9 ± 1.2°

a. Values ± standard deviation (n = 3).

Fig. 2　(Left) Optical and (right) AFM images for pristine (A, E), HMDS modified (B, F), polystyrene modified (C, G) and wax modified (D, H) filter paper. The red area on the optical images (left pane) represents the region of interest (ROI) used for AFM imaging. The dimensions of optical and AFM images were 480 × 360 μm and 1 × 1 μm, respectively.

494 ANALYTICAL SCIENCES MAY 2016, VOL. 32

values for the PS modified surface were higher than observed for silane modified paper, whereas wax modified surface showed the least roughness value. HMDS modification retained the original flexibility of the filter paper, which may allow simple and functional elements (e.g., filters or separators) to be built onto the platform.

Scanning electron microscopy (SEM) analysisSEM images for pristine filter paper are shown in the left hand

side of Fig. 3 and those on the right hand side represent HMDS paper. One can see that pristine filter paper exhibits a fibrous structure and some fibers form straight parallel aggregates. The surface morphology of modified paper (Figs. 3C and 3D) is an obvious difference with that of unmodified filter paper (Figs. 3A and 3B). The number of voids in the pristine filter paper was found to be higher than on HMDS modified filter paper, whereas no significant changes were noticed on the size of voids in the modified and unmodified filter papers. The wettability of HMDS modified and non modified filter papers may account for the observed trend. The HMDS modified paper showed a tighter surface, most likely attributed to the introduction of silane molecules onto the cellulose surface.

One-step assay for the detection of heavy metal ionsThe application of HMDS modified paper in improving the

LOD of one-step assay for detection of Hg2+ was demonstrated and successfully compared with pristine filter paper. HOTT reagent is known to develop a brick-red color upon interaction with Hg2+ ions.23 On pristine paper, the color formation was observed only for 102 and 103 ppm Hg2+ ion solution (Fig. 4A, inset). However, on hydrophobic paper, yellow colored HOTT spots were transformed into brick-red even at extremely low Hg2+ concentrations (i.e., 5 × 10–4 ppm). The color intensity of Hg2+ spots can be easily read via the naked eye and was found to surge with increasing concentration of Hg2+ (Fig. 4B, inset).

The reagent HOTT has numerous coordination sites in complex formation which gave variable bonding modes with metal ions (Fig. 1A). A  comparison of the IR spectra of the HOTT reagent and its Hg(II) complex has revealed that the reagent participated to Hg(II) in a bi-dentate mono negative

fashion through the thione sulfur ν(–N–C=S) as indicated from the observed shift of ν(–N–C=S) to a lower wave number with the appearance of a new band at 395 cm–1 due to ν(Hg–S).28 Participation of the oxime oxygen of the reagent was also noticed from the disappearance of ν–OH and the appearance of ν(–Hg–O) at 1100 cm–1 and ν(Hg–O) at 549 cm–1. Thus, it can be concluded that the reagent HOTT coordinated to Hg(II) via hydroxyl oxygen of the oxime and sulfur of thione forming a six-membered ring chelate in 1:2, Hg:HOTT, molar ratio, i.e. Hg(HOTT)2.23 The UV-vis studies for HOTT and the HOTT-Hg complex were individually performed and are discussed in detail in SI-4 (Supporting Information).

The performance and reproducibility of the one-step assay on pristine and hydrophobic paper were performed by recording the data of three sets of experiments on both substrates. On pristine paper, spots of HOTT reagent developed color only for 102 and 103 ppm concentrations of Hg2+, hence no PGV was calculated for Hg2+ concentrations below 102 ppm (Fig. 4A, data plot). As the image analysis software is equipped to capture the presence of any available pixels, failure to calculate any PGV of assay spots on pristine paper would mean the incumbency of the substrate for rapid detection of Hg2+. On hydrophobic paper, PGV increased proportionally from 5 × 10–4 ppm (i.e., 0.5 ppb) to 103 ppm (Fig. 4B, data plot). Hence, the quantitative data for all Hg2+ concentrations was available.

The use of HOTT in dual wave β-correction spectrophotometry for determination and detection of Hg2+ has been reported by Hamza et al.23 Thus, the influence of pH on the color intensity of the developed colored Hg-HOTT complex was critically investigated by the developed method using Britton–Robinson buffer (pH 3 – 9) and Hg2+ (100 ppm) solution. As shown in Fig. 5, the PGV values for the HOTT-Hg complex increased gradually up to pH 4 – 6, at the same time the PGV of blank spot (negative control) also increased. The dissociation of the oxime hydroxyl group at this pH (4 – 6) may account for the observed change. At solution pH 7 – 9, the PGVBlank value negatively affects the signal-to-noise ratio of the one-step assay. This behavior is most likely attributed to the hydrolysis of the Hg(HOTT)2 complex at pH >7. The normalized PGV (PGVComplex–PGVBlank) value with respect to pH is in a form of a

Fig. 3　SEM images for pristine filter paper (A, B) and HMDS modified filter paper (C, D) scanned at (left) 400× and (right) 1600×.

ANALYTICAL SCIENCES MAY 2016, VOL. 32 495

bell-shaped curve (Fig. 5, inset) which clearly indicates the optimum pH (4 – 6) range. The results are in good agreement with the data reported earlier.23 However, in our developed method we did not make any pH adjustment as Hg2+ solution was prepared in Milli-Q water (pH 5.4).

We also confirmed the application of our hydrophobic paper in the enhanced detection of Co2+ and Zn2+, respectively. The former metal ion was detected using 5-Br-PADAP and for the detection of the latter H2DZ was used. The outcome of the assay on HMDS modified paper was similar to that observed for the Hg2+ assay. As shown in Fig.6, the colored complex formed on 5-Br-PADAP and H2DZ spots with respective metal ions was visible via the naked eye from 5 × 10–4 ppm (i.e., 0.5 ppb) onwards. In both cases, the LOD on HMDS modified paper was also 5-order of magnitude lower to that observed on pristine paper platform. A detailed explanation of the one-step assay for Co2+, Zn2+, UV-vis and IR spectra is described in SI-3 and SI-4 (Supporting Information).

The difference observed in the color intensities and PGV values also explains the influence of surface property (i.e., wettability) in improving the sensitivity of the assay. As observed in Fig. 4 (inset) and Fig. 6, the analyte solution on hydrophobic paper was confined to a limited space, whereas on the pristine surface, a similar solution spread immediately and entirely covered the ligand spots. The overlay of metal ion droplet on the ligand spot represents coverage, which may help us to understand the influence of surface wettability on signal density. The ratio of the surface areas (analyte/ligand) on pristine and HMDS modified filter paper is listed in Table 2 as coverage (%).

It was observed that substrate hydrophobicity had no influence on the spreading of the ligand drop as it was prepared in ethanol.

However, a considerable difference in the surface area of metal ion spots was observed; on pristine paper the droplet for [Hg2+] = 1000 ppm spread evenly throughout the ligand spot, showing 100% coverage, whereas on hydrophobic paper, the maximum coverage was only 53%. Similarly at 100 ppm analyte concentration, the coverage of the reagent spot on pristine filter paper was 63% and was three times higher (i.e., ~20%) than observed on HMDS modified paper (Table 2). Coverage of 5-Br-PADAP and H2DZ spots with aqueous Co2+ and Zn2+ ions showed similar observations (SI-5, Supporting Information).

On pristine paper, the metal ion droplet spreads immediately in the XY (horizontal) as well as in the Z (vertical) direction, due to which the analyte molecules are dispersed throughout the (ligand) spot area. During this process, the colored (ligand-analyte) complex also diffuses, thereby reducing the density of colored complex formation. At high analyte concentrations the dispersion of analyte molecules would have the least effect on color formation, but at low concentrations no visible color would be developed (SI-6 A, B, Supporting Information). On the other hand, the spreading of an aqueous droplet on hydrophobic paper is restricted in the XY direction only, due to which the diffusion of the analyte molecules is also confined to a limited space. The increased density of the analyte molecules reacts with the ligand molecules that in turn may enhance the signal density. This phenomenon may be less effective at high analyte concentrations where the signal density is already strong, whereas at low concentrations the amassing of analyte molecules in the limited spot area may enhance the signal density (SI-6 C, D, Supporting Information). Hence, on hydrophobic paper the development of a visible colored complex is more prominent even at low analyte concentrations, which is not the case on pristine filter paper. Increasing signal density with reduced spot area is in accordance to Ekin’s “Ambient analyte theory”.29,30 This theory relates the signal and signal density in a microspot and states that the signal (total intensity) increases with an increasing amount of capture molecules and growing spot size, whereas the signal density (signal/area) increases with decreasing amount of capture molecules (decreasing spot size).

Earlier, Mujawar et al. had demonstrated the application of hydrophobic glass slides in diagnostic immunoassays. They observed higher signal density and improved signal-to-noise ratio on high hydrophobic glass slides (θ >90°) as compared to pristine glass slides.24 Recently, Feng et al.31 demonstrated that by supplying higher volumes of analyte solution, the sensitivity

0 5×10–4 10–3 10–2 10–1 100 101 102 103

0 5×10–4 10–3 10–2 10–1 100 101 102 103

Fig. 4　(Inset) Scanned images of Hg2+ assay on (A) pristine and (B) HMDS modified hydrophobic paper. Data plot represents PGV values for various Hg2+ concentrations as calculated by ImageJ analysis.

Fig. 5　Data plot showing the effect of pH on the PGV of the HOTT-Hg complex and blank spots. Inset shows the normalized PGV (PGVComplex-PGVBlank) value at various solution pH.

496 ANALYTICAL SCIENCES MAY 2016, VOL. 32

of the paper-based assay can be improved to μM range. On the other hand, Takahashi et al.32 also showed that by selective filtration-enrichment of Hg2+ ions through dithizone nanofiber-coated membrane, the limit of detection could be lowered to 10 ppb. However, as compared to reported work on paper-based assays for detection of Hg2+ ions, our developed method was found to be the most sensitive with easy colorimetric read-out (Table 3). Our method does not involve any complicated surface fabrication31 or synthesis steps33 nor requires expensive conjugated nanomaterial34,35 or biomolecules9 for sensitivity enhancement. The HMDS modified hydrophobic paper can act as a versatile platform not only for detection of heavy metal ions but also in miniaturized assays where the loss of expensive analytes/reagents can be reduced considerably.

Interference studyThe performance of the developed method for detection of

Hg2+ in the presence of potential interfering species was also

tested. The detection of the tested metal ions at a concentration of 1.0 ppm in the presence of a relatively high excess (10 – 100 fold) of Mg2+, Ba2+, Ca2+, Fe, Si, Ni2+, Pb2+, Mn2+, Cu2+, Al3+, Fe3+, Si4+, SO4

2–, PO43–, Cl–, NO3

–, AsO2–, VO3

–, SbO2– and

SeO3–, which are the most concurrent ions in water, was tested

individually by the developed method. The tolerance limit was defined as the concentration of foreign ion added causing a relative deviation within ±5.0% in the magnitude of the color intensity. Most of these ions did not interfere with the analyte (i.e., Hg2+). Interference of Cu2+ was masked by adding one crystal of Na2SO3. The data plot on interference and tolerance limit for interfering species are presented in SI-7 and SI-8 (Supporting Information), respectively.

Analytical applicationsThe developed HMDS modified hydrophobic paper platform

was validated by detection and semi-quantitative analysis of Hg2+ in tap water (spiked with 1 ppm of Hg2+) and municipal wastewater samples. The component analysis of the wastewater samples was performed by ICP-OES and IC, listed in SI-9 (Supporting Information). As shown in Table 4, the PGV values of spiked tap water sample showed a close proximity with standard Hg2+ (1 ppm) solution. The variation of ±1% confirms the accuracy and precision of the proposed method for trace analysis of Hg2+ in complex matrices. Based on the calculated PGV values for domestic wastewater samples, the concentration of the aforementioned ion was found to be between 0.1 – 1.0 ppm (Table 4).

Conclusions

A one-step assay for ultra-sensitive detection of Hg2+, Co2+ and Zn2+ was demonstrated on HMDS modified filter paper. Due to low wettability of the modified paper, the metal ion droplet localizes on the reagent spot and evaporates in a confined area. The colored complex formed from the accumulation of metal ions on the ligand spot results in enhancement of signal density

Table 2　Spreading of HOTT reagent and aqueous Hg2+ spots on pristine and hydrophobic papera

Filter paperHOTT Hg2+ HOTT Hg2+ HOTT Hg2+ HOTT Hg2+

Surface area (mm2) at 1000 ppm Coverage (%) at 1000 ppm Surface area (mm2) at 100 ppm Coverage (%) at 100 ppm

Pristine 29.5 ± 2.3 29.5 ± 2.3 100 31.8 ± 2.3 20.2 ± 1.2 63.4 ± 1.3Hydrophobic 30.6 ± 1.3 16.2 ± 1.0 53.2 ± 5.0 28.3 ± 1.0 5.7 ± 0.7 20.1 ± 0.5

a. Values ± standard deviation (n = 3).

Fig. 6　Scanned image showing performance of one-step assay for (A) Co2+ and (B) Zn2+ on pristine and hydrophobic modified pristine paper.

Table 3　Analytical comparison of developed method with some reported paper-based assays on the detection of Hg2+ ions in water

ReferenceLOD, ppm

Remarks

8 1 × 10–3 Lateral flow assay, requires AuNPs 9 1 × 10–3 Use of enzyme (β-galactosidase)31 50 × 10–3 Fabricating of well-using waxprinting32 2 × 10–3 Dithizone nanofibre coated membrane33 2 × 10–3 Synthesis of polymer membrane required,

assay time ~25 min34 10 × 10–3 Oligonucleotide sequence coupled with

AuNP needed35 3 × 10–3 Two-color ratiometeric method; requires

AuNPs and aptamers36 1 × 10–3 Use of aptamers

This work

0.5 × 10–3 Ultrasensitive, rapid, easy to fabricate, simple colorimetric read-out via the naked eye

ANALYTICAL SCIENCES MAY 2016, VOL. 32 497

per spot, which in turn enhanced the LOD. On HMDS modified paper, the assay sensitivity was improved by 5-order of magnitude which allowed visual detection of the aforementioned ions in sub-ppm range (i.e., 0.5 ppb). The proposed method was further extended for semi-quantitative determination of the Hg2+ ion using the metal specific HOTT reagent. The observed LOD was also lower than the maximum allowable level (MAL) in water for the tested metal ions as set by the World Health Organization (WHO). The present method is not only cost-effective but also offers a simple system coupled with good reproducibility and accuracy.

Acknowledgements

The authors are thankful for the support from the Center of Excellence in Environmental Studies (CEES), King Abdulaziz University and Ministry of Higher Education, Kingdom of Saudi Arabia. We would like to thank Mr. Mohammed Adnan Javeed and Ms. Cathy Lee (from Park Systems) for helping us with AFM analysis. We are also grateful to Dr. Aslam Khan who helped us with ICP-OES and IC analysis of wastewater samples. And lastly, we would like to thank Mrs. Amina Mira for editing the linguistic and grammatical errors in the revised manuscript.

Supporting Information

SI-1: Illustration on surface modification and one-step assay. SI-2: Roughness of pristine and modified papers. SI-3: One-step assay for Co2+ and Zn2+ ions. SI-4: Electronic spectra of reagents and their complexes with metal ions. SI-5: Coverage of ligand and metal ion spots on pristine and hydrophobic paper. SI-6: Influence of surface wettability on signal density and color formation. SI-7: Interference study. SI-8: Tolerance limits of interfering species. SI-9: Analysis of wastewater samples by ICP-OES and IC.

References

1. N. Jalbani and M. Soylak, Ecotoxicol. Environ. Saf., 2014, 102, 174.

2. H. Bagheri, A. Afkhami, A. Shirzadmehr, H. Khoshsafar, and H. Ghaedi, Int. J. Environ. Anal. Chem., 2013, 93, 578.

3. S. Su, M. He, N. Zhang, and C. Cui, Anal. Methods, 2014, 6, 1182.

4. J. E. O’Sullivan, R. J. Watson, and E. C. V. Butler, Talanta, 2013, 115, 999.

5. R. Verma and B. D. Gupta, Food Chem., 2015, 166, 568. 6. H. Lou, Y. Zhang, Q. Xiang, J. Xu, H. Li, P. Xu, and X. Li,

Sens. Actuators, B, 2012, 166–167, 246.

7. M. Arciuli, G. Palazzo, A. Gallone, and A. Mallardi, Sens. Actuators, B, 2013, 186, 557.

8. C. H. Chao, C. S. Wu, C. C. Huang, J. C. Liang, H. T. Wang, P. T. Tang, L. Y. Lin, and F. H. Ko, Microelectron. Eng., 2012, 97, 294.

9. S. M. Z. Hossain and J. D. Brennan, Anal. Chem., 2011, 83, 8772.

10. Z. Nie, C. A. Nijhuis, J. Gong, X. Chen, A. Kumachev, A. W. Martinez, M. Narovlyansky, and G. M. Whitesides, Lab Chip, 2010, 10, 477.

11. P. Stenstad, M. Andresen, B. Tanem, and P. Stenius, Cellulose, 2008, 15, 35.

12. K. F. Lei, S.-I. Yang, S.-W. Tsai, and H.-T. Hsu, Talanta, 2015, 134, 264.

13. A. Kumar, A. Hens, R. K. Arun, M. Chatterjee, K. Mahato, K. Layek, and N. Chanda, Analyst, 2015, 140, 1817.

14. M. da Silva, N. Demarquette, and I. Tan, Cellulose, 2003, 10, 171.

15. I. H. Tan, M. L. P. da Silva, and N. R. Demarquette, J. Mater. Chem., 2001, 11, 1019.

16. Y. Lu, W. Shi, J. Qin, and B. Lin, Anal. Chem., 2009, 82, 329.

17. T. Songjaroen, W. Dungchai, O. Chailapakul, and W. Laiwattanapaisal, Talanta, 2011, 85, 2587.

18. L. Yu and Z. Z. Shi, Lab Chip, 2015, 15, 1642. 19. L. Zhang, X. Cao, L. Wang, X. Zhao, S. Zhang, and P.

Wang, Analyst, 2015, 140, 4105. 20. Y. Sameenoi, P. N. Nongkai, S. Nouanthavong, C. S. Henry,

and D. Nacapricha, Analyst, 2014, 139, 6580. 21. C. Suárez-Pantaleón, J. Wichers, A. Abad-Somovilla,

A. van Amerongen, and A. Abad-Fuentes, Biosens. Bioelectron., 2013, 42, 170.

22. L. H. Mujawar, A. Moers, W. Norde, and A. Van Amerongen, Anal. Bioanal. Chem., 2013, 405, 7469.

23. A. Hamza, A. S. Bashammakh, A. A. Al-Sibaai, H. M. Al-Saidi, and M. S. El-Shahawi, J. Hazard. Mater., 2010, 178, 287.

24. L. H. Mujawar, W. Norde, and A. Van Amerongen, Analyst, 2013, 138, 518.

25. K. Abe, K. Kotera, K. Suzuki, and D. Citterio, Anal. Bioanal. Chem., 2010, 398, 885.

26. L. H. Mujawar, A. Van Amerongen, and W. Norde, Talanta, 2012, 98, 1.

27. G. Chinga-Carrasco, N. Kuznetsova, M. Garaeva, I. Leirset, G. Galiullina, A. Kostochko, and K. Syverud, J. Nanopart. Res., 2012, 14, 1.

28. K. Nakamoto, “Infrared and Raman Spectra of Inorganic and Coordination Compounds”, 1971, Wiley Interscience, New York.

29. R. Ekins and F. W. Chu, Trends Biotechnol., 1999, 17, 217. 30. M. F. Templin, D. Stoll, M. Schrenk, P. C. Traub, C. F.

Vöhringer, and T. O. Joos, Trends Biotechnol., 2002, 20, 160.

31. L. Feng, X. Li, H. Li, W. Yang, L. Chen, and Y. Guan, Anal. Chim. Acta, 2013, 780, 74.

32. Y. Takahashi, S. Danwittayakul, and T. M. Suzuki, Analyst, 2009, 134, 1380.

33. H. El Kaoutit, P. Estevez, F. C. Garcia, F. Serna, and J. M. Garcia, Anal. Methods, 2013, 5, 54.

34. G.-H. Chen, W.-Y. Chen, Y.-C. Yen, C.-W. Wang, H.-T. Chang, and C.-F. Chen, Anal. Chem., 2014, 86, 6843.

35. Q. Wei, R. Nagi, K. Sadeghi, S. Feng, E. Yan, S. J. Ki, R. Caire, D. Tseng, and A. Ozcan, ACS Nano, 2014, 8, 1121.

36. G. G. Lewis, J. S. Robbins, and S. T. Phillips, Chem. Commun., 2014, 50, 5352.

Table 4　PGV and approximate concentration of Hg2+ in tap and domestic (samples 1 – 4) water samples

Water sample PGV Concentration, ppm

Tap water (1 ppm) 15.36 ~1.0Sample 1 13.64 0.1 – 1.0Sample 2 13.64 0.1 – 1.0Sample 3 10.76 0.1 – 1.0Sample 4 10.34 0.1 – 1.0