Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2021 Jan 8;6(3):2052–2059. doi: 10.1021/acsomega.0c05020

Speciation Analysis of Cr(VI) and Cr(III) in Water with Surface-Enhanced Raman Spectroscopy

Olga Dvoynenko †, Shih-Lin Lo ‡, Yi-Ju Chen ‡, Guan Wei Chen †, Hsin-Mei Tsai †, Yuh-Lin Wang †,#, Juen-Kai Wang †,§,*
PMCID: PMC7841920  PMID: 33521444

Abstract

graphic file with name ao0c05020_0007.jpg

Identifying and quantifying chromium in water are important for the protection of precious water resources from chromium pollution. Standard methods however are unable to easily distinguish toxic hexavalent chromium, Cr(VI), from innocuous trivalent chromium, Cr(III), are time-consuming, or require large sample quantity. We show in this report that Cr(VI) and Cr(III) in water can be differentiated based on their distinct spectral features of surface-enhanced Raman scattering (SERS). Their SERS signals exhibit different pH dependences: the SERS features of Cr(VI) and Cr(III) are most prominent at pH values of 10 and 5.5, respectively. The obtained limit of detection of Cr(VI) in water is below 0.1 mg/L. Both concentration curves of their SERS signals show Langmuir sorption isotherm behavior. A procedure was developed to quantify Cr(VI) concentration based on the direct retrieval or addition method with an error of 10%. Finally, the SERS detection of Cr(VI) is shown to be insensitive to co-present Cr(III). The developed SERS procedure offers potential to monitor toxic chromium in fields.

Introduction

Chromium (Cr) has many oxidation states—the common ones are +3, Cr(III), and + 6, Cr(VI)—and is widely used in industry, particularly chrome plating.1 The popular plating method is hexavalent chromium plating, while the trivalent chromium plating is less common because of its uncontrollable process and poorer outcomes. Chromate however can enter cells and cause mutagenic damage.2−4 The International Agency of Research Cancer (IARC) has classified Cr(VI) compounds under Group 1 since 1990.5 In contrast, Cr(III) was considered an essential nutrient6 and no toxicity has been observed.7 Because of the difficulty in differentiating chromium species with current analytical methods, a guideline value of total chromium issued by WHO is 0.05 mg/L,5 whose health risk is however questionable. Many countries adopt higher regulated concentrations. As an example, the United States has a national primary drinking water regulation of the maximum concentration level for total chromium of 0.1 mg/L. Consequently, differentiating Cr(VI) from Cr(III) species is crucial in the protection of water resources from chromium pollution. In water, Cr(VI) is present as oxyanions: both Cr2O72– and HCrO4 dominate at pH values between 0 and ∼6, while CrO42– emerges at a pH value of ∼4.5, reaches its maximum at a pH value of ≥8, and remains so at higher pH values.1,8,9 On the other hand, at least five species of Cr(III) exist in water depending on pH values: Cr3+, CrOH2+, Cr(OH)2, Cr(OH)30, and Cr(OH)4.1,9,10 The knowledge above provides the fundamental understanding of equilibrated chromium species in aqueous solution.

There are now several well-developed analytical methods used in environmental laboratories as standards for determining chromium concentration. They can be separated into two types. The first type is elemental analysis methods that characterize chromium according to its characteristic atomic propensities: (1) core-level absorption and emission energy—atomic adsorption spectrometry (AAS) (APHA Method 3111A, 1992)11 with a method detection limit (MDL) of 0.02 μg/L and inductively coupled plasma-atomic emission spectrometry (USEPA 200.7, Rev. 5.0, 2007)12 with an MDL of 4 μg/L— and (2) mass—inductively coupled plasma-mass spectrometry (ICP-MS) (USEPA Method 200.8, Rev. 5.5, 1999)13 with an MDL of 0.08 μg/L. Only the total chromium content in aqueous samples is determined by these methods. The second type is spectrophotometric methods that can determine Cr(VI) concentration. In the standard method (APHA 3500-Cr-D-1992),14 the concentration of Cr(VI) is determined by UV–vis absorption of a colored complex compound of Cr(VI) with 1,5-diphenyl-carbazide. The MDL of this method is from 0.1 mg/L. The disadvantages of the second-type methods are two-fold: (1) the use of additional reagents in sample preparation could alter the chromium species composition; (2) interference from sample matrices might affect the quantitative accuracy and MDL. In addition, modern hyphenated techniques, such as high-performance liquid chromatograph in combination with ICP-MS or liquid chromatography with AAS detection, can avoid those interferences and separate chromium types in the sample. However, they require expensive equipment, controlled laboratory conditions, complicated sample preparation procedures, and long analytical time that inhibit their wide applications. Most importantly, these methods cannot be used in fields to provide in-time information to quickly identify the contaminated water resource. Accordingly, there is an urgent need for a fast screening method to discriminate different chromium species—particularly, Cr(III) and Cr(VI)—in fields.

Infrared absorption spectroscopy and Raman spectroscopy are two powerful tools to identify molecular species according to their characteristic vibrational spectra. The weak absorption of the vibrational modes of the analyte is often overwhelmed by a huge absorption background from water, while Raman spectroscopy, providing background-free detection, suffers from its weak signal. Thanks to the discovery of surface-enhanced Raman scattering (SERS),15,16 Raman scattering probability can be enhanced by several orders of magnitude by enhancing the local field of nanostructures that interacts with analytes in proximity.17 It has emerged as a new paradigmatic approach to assay molecules of extremely low concentration.18 Despite its application to numerous chemical compounds, only few studies were devoted to the detection of chromium in aqueous solutions. Two types of SERS enhancers were used in these previous studies. The first type is metal colloids19−23 that suffer from time-variant sensitivity. The second type is metal nanostructures on substrates24−26 that often manifest poor reproducibility. Particularly, Mosier-Boss and Lieberman demonstrated the ppb sensitivity of chromate and well-behaved Frumkin isotherm with use of cation-coated roughened metal surfaces.24,25 Recently, Ji et al.27 and Bu et al.28 separately developed functionalized nanoparticles to capture Cr(VI) species in water. Only few groups devoted research efforts to the SERS detection of Cr(III) but not quantitative analysis. For examples, the SERS spectra of Cr(III) were obtained on silver SERS-activated plates29 and from silver-coated corrosive chromium without offering the concentration curve of the SERS signal Cr(III).30,31 Ye et al. obtained the curve based on the dependence of the Au nanoparticle aggregation on the Cr(III) concentration.32

In spite of these previous SERS studies of Cr(VI) and Cr(III), their application to identifying and quantifying Cr(VI) and Cr(III) species in actual water samples is still lacking, owing to two major hurdles: (1) poor reproducibility associated with metal colloids and roughened metal substrates and (2) perplexed influences from coated SERS enhancers. Finally, given the facts that the species of Cr(III) and Cr(VI) present in water depend on pH values,1,8−10 there is no systematic study to show how the pH values of the chromium samples affect the SERS spectra and how the acquired relationship can be used to develop protocols for analytical applications. Two possible reasons of lack of such studies are as follows: (1) the aggregation of metal colloids could be highly sensitive to the pH value of the samples; (2) the adhesion of the functionalized linkers and their analyte-capturing capability greatly depend on the pH condition.

In this work, two approaches were adopted to overcome the two hurdles encountered in the previous SERS studies of chromium, respectively: (1) well-controlled Ag nanoparticle array to enable highly repeatable, uniform SERS enhancement and (2) direct SERS detection of Cr species in water. Instead of metal colloidal nanoparticles, we have used an SERS substrate based on Ag nanoparticle array imbedded within nanochannels in anodic aluminum oxide (dubbed AgNP/AAO) developed by our group.33 Hot spots thus created at the gap between adjacent Ag nanoparticles by plasmonic coupling firmly enhance the local optical field and amplify Raman radiation. The far-field34 and near-field35 optical propensities of such hot spots investigated experimentally agree with high-precision electrodynamic simulation.36 We have applied this substrate successfully to clinical microbiology37 and food safety38 for their respective semiquantitative analysis. Second, the direct SERS detection of Cr species in water has two advantages. The first one is that the species is identified with characteristic SERS spectral signature, rather than relying on capturing agents. The second one is stable detection enabled by adsorption–desorption equilibrium of analytes occurring at the SERS substrate. Since the SERS signal reflects the analytes adhered on the surface of the SERS enhancer, a sorption isotherm behavior is expected to occur between the Cr species and the SERS substrate. In fact, several groups studied the sorption isotherm of species on different SERS enhancers. For example, Zhang et al.39 showed the adsorption isotherm of an anthrax biomarker on a silver film over nanosphere substrates. Altun et al.40 characterized the sorption isotherms of SERS signals of organic species on the Ag/HfO2/CNT SERS substrate. Note that the sorption isotherms obtained with conventional methods (such as volumetric analysis, gravimetric analysis, voltammetry, and calorimetry) are different from those obtained with SERS because the SERS signal is attributed to only the analytes adsorbed on the surface of the metal nanoparticles on the SERS substrate that may contain other nonmetal surfaces. In this work, a systematic study was performed to reveal how the SERS spectra of Cr(III) and Cr(VI) in water depend on pH values under such equilibrium, yielding their characteristic spectral signatures for identification and the respective concentration curves of their SERS signals for quantitative analysis. In the end, an SERS-based procedure was developed to quantify Cr(VI) in water even in the presence of Cr(III).

Materials and Methods

Chemicals and Sample Preparation

Chromium (Cr3+ of 1000 μg/mL in 2% HNO3, High-Purity Standard) was used for preparing Cr(III) samples, while potassium chromate powder (K2CrO4, 99.0%, Alfa Aesar) was used for Cr(VI) sample preparation. The pH values of alkaline Cr samples were adjusted with sodium hydroxide (Shimakyu’s Pure Chemicals), while those of acid samples were adjusted with nitric acid (HNO3, 65% solution in water, ACROS Organic). All solutions were prepared with deionized water. To search the optimal pH values for the SERS detection of Cr species, Cr(VI) and Cr(III) aqueous solutions with respective concentrations of 10–5 and 10–4 M of a series of pH values were prepared. The pH value was measured with a pH meter (Seven Excellence, Mettler Toledo) with an InLab Routine Pro combined electrode and KCl solution (3.0 M) as an internal electrolyte. The electrode was calibrated with buffer solutions. The pH values were cross-checked using pH test strips (pH Fix, Macherey-Nagel). Owing to the possible dissolution of CO2 in air into the sample solutions, the prepared samples were kept in sealed vases with minimal air containment. The pH values of the sample solutions were monitored before and after the SERS measurement. The concentrations of the prepared samples were verified if the absorption was linearly proportional to the concentration (Beer–Lambert law).41,42 For the experiments of mixed Cr(VI) and Cr(III), the final pH values of the mixtures were verified to be around 10.

SERS Substrate

The SERS-active substrates used in the experiments were made of arrays of Ag nanoparticles imbedded in the nanochannels of anodic aluminum oxide—named as AgNP/AAO. Its fabrication was reported previously33 and is briefed here. An aluminum-coated glass slide prepared by sputtering was anodized and then chemically etched to create two-dimensional hexagonally packed arrays of nanochannels. Silver electrochemical plating produced Ag nanoparticles in the nanochannels, yielding two-dimensional Ag-nanosphere array with an average sphere diameter of 50 nm and a mean gap of 7 nm. Individual SERS-active substrates were then cleaned by rinsing with deionized water and then vacuum-sealed in plastic bags. Each substrate was freshly used in the SERS measurement. A significant advantage of the SERS substrates is its large active area of 2.5 cm × 5 cm with a uniform enhancement factor.

SERS Measurement

The Raman measurements were performed with a commercial Raman microscope system. A HeNe laser served as the excitation source. The laser beam was focused using a 10× micro-objective lens onto the sample; the scattered radiation was collected backward through the same lens, dispersed using an 80-cm spectrograph, and recorded using a liquid-nitrogen cooled charge-coupled device. The spectral resolution and error were calibrated to be 7 and 0.1 cm–1, respectively. In addition to the use of highly uniform SERS substrates, four special measures were taken to enhance the repeatability of SERS measurements. First, all the SERS measurements were performed with a designated aluminum trough where the substrate with the sample droplets was positioned. A groove at the inner bottom of the trough was filled with 0.5 mL of water. The flat crest of the trough was covered with a low-fluorescence glass plate such that the interior humidity was maintained at its saturation level. The invariance of the sample concentration was confirmed by the constant SERS signal of adenine for at least 1 h. Second, the sample droplet on the SERS substrate touched the bottom surface of the glass plate, resulting in only flat optical interfaces for the optical path between the SERS substrate and the glass plate. Third, the uniformity of the SERS substrate was checked with the SERS measurement of adenine (10–4 M). A substrate was considered uniform as the SERS signals of four adenine droplets, located at the four corners of a 2 cm × 4 cm rectangle, varied less than 10%. The uniformity test result of an SERS substrate is shown in Figure S8 of the Supporting Information. The result shows that the standard deviation of the SERS signals of adenine (10–4 M) at 15 different sites over the central 2 cm × 4 cm region is 15.3%, while that of the signals except the three obtained from the rightmost column is 11%. The pass rate of the substrate that has a standard deviation of less than 10% is more than 50%. The subsequent SERS measurements of the samples were performed within that rectangular region. Fourth, the sample droplets with different characteristics (concentration or pH) and three water droplets with a designated pH value were applied randomly on the substrate, further reducing the influence of the gradual variation in the enhancement factor over the selected region. Raman measurements were repeated at several randomly chosen positions within each sample droplet. Special attention was paid to check if no abnormal Raman feature appeared and the signal strength was linearly proportional to the irradiating laser power and the integration time.

Spectral Analysis

The inevitable SERS continuum background43,44 of each measured SERS spectrum was removed by the following two steps. The first step is to subtract the measured SERS spectrum of a nearby droplet of aqueous solution with a designated pH value from that of the Cr sample with the same pH value, removing the major portion of the background. The residual background was eliminated by a de-baseline procedure (the second step): the resultant spectral data outside the frequency range of the characteristic Raman feature of interest were fit with least-order polynomial functions such that the difference between the experimental spectrum and the fitted function showed random propensity. The characteristic feature in the spectrum after the background removal was fit with multiple Gaussian peaks. The extracted area of the spectral feature was considered to reflect the amount of species of interest adsorbed on the substrate surface for the quantitative analysis.

Results and Discussion

The SERS measurements in this work were performed as the analyte species undertook adsorption–desorption equilibrium45 —sorption isotherm—between the ones dissolved in water and the ones adsorbed on the surface of the SERS substrate. It is therefore crucial to understand the relationship between the equilibrated species in water and the observed SERS spectrum at a specific pH value and how this relationship is varied with the analyte concentration. In the following, the SERS spectra of Cr(VI) at different pH values are shown first. The observed spectra are interpreted with the present Cr(VI) species, resulting in an optimal pH value for the SERS speciation of Cr(VI) in water. Comparing the SERS spectrum and the normal Raman spectrum acquired under that pH condition helps identify the species responsible for the SERS spectrum and designate the spectral feature for quantitative analysis. Under such pH conditions, the extracted concentration curve of the signal area of the SERS feature is revealed. Fitting the curve with the Langmuir sorption isotherm model yields the affinity of that Cr(VI) species, providing the scientific foundation for the quantitative analysis procedure of Cr(VI) in water. The SERS analysis of Cr(III) then follows. In the end, the result of the SERS detection of Cr(VI) in the presence of Cr(III) shows the capability of this speciation protocol in identifying and quantifying Cr(VI) in water in the presence of Cr(III).

pH-Dependent SERS Spectra of Cr(VI) in Water

As reported previously,1,8,9 Cr(VI) exists as Cr2O72– and HCrO4 in the pH range from 0 to 6, while it appears predominantly as CrO42– at pH ≥ 8. The SERS spectrum of Cr(VI) in water thus expectantly varies with the pH condition of water, as these oxyanions would confer different vibrational modes. Figure 1a shows the typical SERS spectra of Cr(VI) in water (10–5 M ≈ 0.5 mg/L) at pH = 3.5, 5.5, and 10 as well as the normal Raman spectrum of Cr(VI) in water (0.05 M ≈ 2.6 g/L) at pH = 10. Note that Cr(VI) in water under the three different pH conditions shows distinct SERS spectra. Moreover, the SERS spectrum at pH = 10 is almost identical to the normal Raman spectrum of Cr(VI) in water at the same pH value: one feature centered at 350 cm–1 and one compounded feature ranging from 700 to 1000 cm–1. Since CrO4 dominates at pH = 10, the agreement between the SERS spectrum and the normal spectrum indicates that CrO42– species is physisorbed on the surface of the SERS substrate. The 350 cm–1 feature was attributed to symmetric and asymmetric bending modes, while the feature ranging from 700 to 1000 cm–1 was attributed to symmetric and asymmetric stretching modes.8 The broadened features of the SERS spectrum compared with the corresponding features of the normal Raman spectrum at pH = 10 suggest that the adsorbed CrO4 on the surface of the SERS substrate takes a range of different configurations.

Figure 1.

Figure 1

Open in a new tab

(a) SERS spectra of Cr(VI) in water (10–5 M ≈ 0.5 mg/L) at pH = 3.5 (red curve), 5.5 (green), and 10 (blue curve) as well as the normal Raman spectrum (black curve) of Cr(VI) in water (0.05 M ≈ 2.6 g/L) at pH = 10; (b) normalized integrated SERS signal area from 700 to 1000 cm–1 of Cr(VI) species in water, AVI, vs pH value, where the maximal average AVI value is set to one. The line is a guide to the eye.

The feature between 700 and 1000 cm–1 in the SERS spectra of Cr(VI) in water was tentatively taken as the marker of CrO42–, and its signal area, AVI, was extracted with the background removal-plus-Gaussian fitting procedure described above to reflect the coverage of CrO4 on the surface of the SERS substrate according to the first-layer theory of SERS.46Figure 1b shows that AVI increases with the pH value as pH > 4, reaches a maximum at pH = 9.5, and decreases afterward. The increase of AVI as the pH value is increased from 4 to 8 follows approximately with the increase of CrO42– in this pH range. On the other hand, although CrO4 remains the single dominant species in water as pH > 8,1,8,9AVI decreases as pH > 9.5. The first possible reason for the disagreement is that the adsorption capability (affinity) of CrO42– on the SERS substrate might be sensitive to the pH value of the sample. The second reason is that the SERS substrate could deteriorate for pH > 10. Nevertheless, the optimal pH value for the speciation analysis of Cr(VI) is between 9 and 10. Finally, in the context of the quantitative analysis, it is imperative to understand whether CrO4 remains the dominant species in water for the concentration range of interest (from 10–7 to 10–3 M). The fractions of all the Cr(VI) species in water are governed by their equilibrium reaction equations.47 The equations to obtain the fractions of these species (fH2CrO4, fHCrO4–, fCr2O72–, and fCrO42–) and the calculated results are shown in Figure S1 of the Supporting Information. At pH = 10, fCrO42– is dominant, while the total fraction of the other three species is less than 0.001 for CVI. Accordingly, at pH = 10, the only detectable Cr(VI) species is CrO42– for the concentration range of interest, thus greatly simplifying the quantitative analysis.

Quantitative Analysis of Cr(VI) in Water

Before using the SERS method to quantify Cr(VI), the concentration curve of its SERS signal needs to be acquired and analyzed with a sorption isotherm model so that a quantitative analysis procedure can be developed. As described above, the SERS signal of the Cr(VI) species dissolved in water reflects its coverage on the surface of the SERS substrate, θVI. If θVI follows a Langmuir sorption isotherm,48 the signal area, AVI, of the spectral marker of that species is given by

graphic file with name ao0c05020_m001.jpg 1

where AVI∞ is AVI at infinity concentration and LVI is the affinity. Representative SERS spectra of Cr(VI) in water of different concentrations at pH = 10 are shown in Figure S2 of the Supporting Information. Figure 2 shows the concentration curve of AVI at pH = 10. Note that the curve fits eq 1 very well and the extracted LVI is 1.47 ± 0.04 L/mg. A similar concentration curve was obtained for the signal area of the 350 cm–1 feature, and the extracted affinity is 1.45 ± 0.12 L/mg. The good correspondence between the two extracted affinities confirms that the two spectral features stem from the same species CrO4 and the SERS substrate is uniform. Since AVI between 700 and 1000 cm–1 is larger than that centered at 350 cm–1, the high-frequency AVI is used as the marker to identify Cr(VI) in water and its concentration curve, Figure 2, serves as the basis in the quantitative analysis of Cr(VI) species.

Figure 2.

Figure 2

Open in a new tab

Normalized integrated SERS signal area from 700 to 1000 cm–1 of Cr(VI) species in water, AVI, as a function of Cr(VI) concentration, CVI. The solid line is the fit to eq 1. AVI∞ is the signal area with the infinite CVI.

The acquired concentration curve of AVI confers two messages. First, since AVI is not proportional to CVI, as shown in Figure 2, traditional methods are not applicable49 for the quantitative analysis of Cr(VI) in water based on its detected SERS signal. Second, the effective concentration range of SERS detection is roughly between 0.1 and 2 mg/L. Based on these two messages, we propose a two-step quantification procedure to determine the concentration of Cr(VI): (1) assessment of SERS-effective concentration and (2) determination of the concentration. In the first step, the sample with an unknown concentration CVI* is diluted by 10 and 100 times, yielding three samples with concentrations of CVI – j = 10–jCVI*, j = 0, 1, and 2; the SERS measurements performed on the three samples then determine the one of the lowest concentration, CVI – J, that is detectable; if none of them is detectable, CVI* is below the minimal limit of the SERS detection. In the second step, two quantification methods can be used to determine CVI: direct retrieval method and addition method. In the direct retrieval method, the SERS measurements of the diluted unknown sample with a concentration of CVI – J* as well as the standard Cr(VI) samples with concentrations from 0.1 to 10 mg/L are performed on the same SERS substrate to obtain their respective AVI values; the obtained AVI values of the standard samples are fit to eq 1 to extract LVI and AVI; the AVI value of the diluted unknown sample, AVI – J, is then used to determine CVI – J* and thus CVI. In the addition method, the SERS measurements of the diluted unknown sample with a concentration of CVI – J* mixed with the standard samples with concentrations from 0.1 to 10 mg/L are performed on the same SERS substrate; the extracted AVI values are fit to a modified Langmuir model

graphic file with name ao0c05020_m002.jpg 2

to obtain CVI – J*, where CVI is the concentration of one of the standard samples. The flow chart of this procedure is illustrated in Figure S5 of the Supporting Information. The procedure was tested on two samples with their respective concentrations of 0.52 and 10.4 mg/L. The direct retrieval method was applied to the first case to obtain a concentration of 0.586 ± 0.078 mg/L, while the addition method was applied to the second case to obtain a concentration of 12.8 ± 2.3 mg/L. The measured concentrations for the both cases are within the error bars of the measurements. The data of these two tests are shown in Figure S6 of the Supporting Information. The results demonstrate that the SERS-based procedure obtained reasonably accurate concentrations of the unknown samples. As a final note, such accurate quantification results are made possible by our stable and uniform SERS substrate (without colloidal aggregation) and the SERS detection of the CrO4 species directly adsorbed on the substrate (without any linker coating).

SERS Propensities of Cr(III) in Water

Previous studies1,9,10 showed that there are at least seven possible Cr(III) species existing in water—Cr3+, CrOH2+, Cr(OH)2+, Cr(OH)3, Cr(OH)4–, Cr2(OH)2, and Cr3(OH)45+— and their fractions also depend on the pH condition. More than two species are present simultaneously in water, except that Cr3+ and Cr(OH)3 dominate at pH < 3.6 and pH = 9.5, respectively. Figure 3 shows three representative SERS spectra of Cr(III) in water (10–4 M = 5.2 mg/L) at pH = 3.8, 5.5, and 6.8. At pH = 3.8 and 6.8, there is barely any spectral feature, while at pH = 5.5, a single feature appears at around 600 cm–1. Since CrOH2+ and Cr(OH)2+ predominate at pH ∼5, this 600 cm–1 feature would be attributed to the two species though the normal Raman feature was not observed in this study. Finally, at pH = 6.8∼10, there is also no spectral feature observable probably because the precipitated Cr(OH)3 has weak affinity on the SERS substrate. Nevertheless, it seems that the 600 cm–1 feature can serve as the marker of Cr(III) species. The main Cr(III) species present in water are governed by seven equilibrium reaction equations with their respective reaction constants.50 The equations to obtain the fractions of these species (fCr3+, fCrOH2+, fCr(OH)2+, fCr(OH)30, fCr(OH)4–, fCr2(OH)24+, and fCr3(OH)45+) at pH = 5.5 and 10 and the results are shown in Figure S3 of the Supporting Information. Since there is no observable feature in the SERS spectrum of Cr(III) at pH > 6.8, the dominant species Cr(OH)30 (fCr(OH)30 > 0.95) do not contribute to the SERS signal, suggesting that its affinity to the SERS substrate is very small probably owing to its charge neutrality. On the other hand, at pH = 5.5, fCrOH2+ and fCr(OH)2+ are higher than 0.1 and particularly fCrOH2+ > 0.8. The larger positive charge of CrOH2+ than that of Cr(OH)2 suggests that its affinity to the SERS substrate is higher, and therefore, the single spectral feature at 600 cm–1 could be attributed to the Cr–OH stretch.51 More evidence is needed to support this assignment. The spectral feature centered at 600 cm–1 was taken as the marker of CrOH2+, and its signal area, AIII, was extracted. Figure 3b shows AIII as the pH value is varied from 3.8 to 6.8. Note that AIII increases with the pH value, reaches a maximum at pH = 6, and decreases afterward. In comparison, the calculated fraction of CrOH2+ increases from 0.1 at pH = 2.7 to the maximum of 0.89 at pH = 4.8 and decreases to 0.1 at pH = 6.6. Therefore, the pH dependence of AIII agrees approximately with that of fCrOH2+. The possible reasons for the discrepancy are two-fold. First, the affinity of CrOH2+ on the surface of the SERS substrate could be sensitive to the pH condition, similar to the case of CrO42– described above. Second, CrOH2+ might not be the only species adsorbed on the SERS substrate to confer the 600 cm–1 feature. Nevertheless, according to Figure 3b, the optimal pH value to perform speciation analysis of Cr(III) based on SERS is 5.5–6. Similar to Cr(VI) in water, the signal area of the 600 cm–1 feature, AIII, would follow a Langmuir sorption isotherm model:

graphic file with name ao0c05020_m003.jpg 3

where AIII∞ is AIII at infinity concentration, CIII is the concentration of Cr(III), and LIII is the affinity. Representative SERS spectra of Cr(III) in water of different concentrations at pH = 5.5 are shown in Figure S4 of the Supporting Information. The experimentally obtained AIII as a function of CIII is shown in Figure 4. Note that the experimental data fit eq 3 well and the extracted LIII is 0.14 ± 0.04 L/mg. Presumably, the same quantitative analysis procedure for Cr(III) can be developed based on the results of Figure 3b and Figure 4. Since Cr(III) compounds are benign to water and soil in the environment and are not toxic, the need to detect Cr(III) on site is not pressing and the procedure is not deliberated here.

Figure 3.

Figure 3

Open in a new tab

(a) SERS spectra of Cr(III) in water (10–4 M = 5.2 mg/L) at pH = 3.8 (red curve), 5.5 (green curve), and 6.8 (blue curve); (b) normalized integrated SERS signal area at 600 cm–1 of Cr(III) species in water, AIII, vs pH value where the maximal average AIII value is set to one. The line is a guide to the eye.

Figure 4.

Figure 4

Open in a new tab

Normalized integrated SERS signal area at 600 cm–1 of Cr(III) species in water, AIII, as a function of Cr(III) concentration, CIII. The solid line is the fit to eq 3. AIII∞ is the signal area with the infinite CIII.

Quantifying Cr(VI) in the Presence of Cr(III) in Water

The two-step procedure to quantify the amount of Cr(VI) species is developed based on the dependence of the SERS signal of Cr(VI) in water on its concentration and pH value. However, will the identification and the quantification of Cr(VI) species in water be affected by the presence of Cr(III) species? Finding the appropriate procedure to perform such an analysis is crucial to the use of SERS in analyzing Cr(VI) species in water. The respective dependences of the SERS signals of Cr(VI) and Cr(III) species on the pH value in water, Figure 1b and Figure 3b, indicate that the best pH condition to identify and quantify Cr(VI) in water is at pH ∼10, where the SERS signal of Cr(VI) is most pronounced, while that of Cr(III) is diminished. This is so because the anion CrO42– appears predominantly at pH > 8, while neutral Cr(OH)3 prevails at pH > 7. The SERS measurements of two Cr(VI) concentrations, 0.4 and 4 mg/L, mixed with Cr(III) of a series of concentrations were performed to examine the possible interference by the co-present Cr(III). Figure 5 shows AVI/⟨AVI⟩0 mixed with different Cr(III) concentrations, where ⟨AVI⟩0 is the average value of AVI without the mixed Cr(III). Three representative SERS spectra of Cr(VI) at 4 mg/L mixed with Cr(III) at 10 mg/L (red line), 20 mg/L (green line), and 40 mg/L (blue line) at pH = 10 are shown in Figure S7 of the Supporting information. Note that the variations of these two cases are within 20%, indicating that the present Cr(III) species virtually do not influence the SERS detection of Cr(VI) species in water if the pH value is set at 10. The slight decay in AVI/⟨AVI⟩0 with the Cr(III) concentration could be caused by the small amount of precipitated Cr(OH)3s remaining after centrifugation that partially covers the SERS substrate. Additional filtering would effectively remove it.

Figure 5.

Figure 5

Open in a new tab

Normalized integrated SERS signal areas from 700 to 1000 cm–1 of Cr(VI) species in water, AVI/⟨AVI⟩0, with concentrations of 0.4 mg/L (gray columns) and 4 mg/L (black columns) mixed with Cr(III) species of five different concentrations: CIII = 2.6, 10.4, 20, 30, and 40 mg/L. ⟨AVI⟩0 is the average value of AVI without mixed Cr(III).

Conclusions

This study shows systematic speciation analysis of Cr(VI) and Cr(III) species in water. The pH dependence of the SERS spectra of Cr(VI) in water were examined. The equilibrium reactions among different possible Cr(VI) species provide the foundation to understand the SERS spectra under different pH conditions. Particularly, the SERS spectrum at pH = 10 is the most prominent and is consistent with its normal Raman spectrum, indicating that the dominant anion CrO42– under this pH condition is physisorbed on the SERS substrate. The portrayed concentration curve of Cr(VI) follows a Langmuir sorption isotherm. A procedure was then developed to quantify Cr(VI) in water based on direct retrieval or addition methods. Similarly, the SERS spectra of Cr(III) in water were obtained at different pH values. In comparison, the equilibrium reactions of different Cr(III) species in water show that more than one species is present, thus complicating the assignment of the observed SERS spectra. Nevertheless, the prominent feature at 600 cm–1 exhibits a simple Langmuir sorption isotherm. Finally, the SERS experiments of mixed Cr(VI) and Cr(III) species show that the detected SERS signal of Cr(VI) at pH = 10 is virtually not affected by the co-present Cr(III) species, indicating that the identification and the quantification of Cr(VI) in water can be performed readily. The detection limit of this procedure is below 0.1 mg/L, making it suitable for most regulated concentrations of Cr(VI) among different countries. Although our preliminary study of applying this procedure to a water sample discharged from a factory gave an error of about 10% in the extracted concentration of Cr(VI), more studies are still needed in future to address the issue of various co-existing ions in actual effluent water. The most exciting prospect of our study is that, in principle, similar SERS-based analytic methods can be developed to perform quantitative speciation analysis of aqueous solutions of transition metal species with known equilibrium reactions among different oxyanions and oxycations.52,53

Acknowledgments

The authors acknowledge the financial support from the Ministry of Science and Technology (MOST 108-2112-M-002-010-MY3 and MOST 108-2639-M-001-003-ASP) and Academia Sinica in Taiwan.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.0c05020.

  • Hydrolysis reactions of Cr(VI) and Cr(III) species in water; SERS spectra of Cr(VI) and Cr(III) with different concentrations; flow chart of quantification procedure; results of determining Cr(VI) concentrations based on direct retrieval and addition methods; SERS spectra of mixed solutions of Cr(VI) and Cr(III); uniformity test result of SERS substrate (PDF)

The authors declare no competing financial interest.

Supplementary Material

ao0c05020_si_001.pdf (1MB, pdf)

References

  1. Kotaś J.; Stasicka Z. Chromium occurrence in the environment and methods of its speciation. Environ. Pollut. 2000, 107, 263–283. 10.1016/S0269-7491(99)00168-2. [DOI] [PubMed] [Google Scholar]
  2. Zhitkovich A. Importance of Chromium-DNA Adducts in Mutagenicity and Toxicity of Chromium(VI): Review. Chem. Res. Toxicol. 2005, 18, 3–11. 10.1021/tx049774+. [DOI] [PubMed] [Google Scholar]
  3. Cieslak-Golonka M. Toxic and mutagenic effects of chromium (VI). A review. Polyhedron 1996, 15, 3667–3689. 10.1016/0277-5387(96)00141-6. [DOI] [Google Scholar]
  4. Saha R.; Nandi R.; Saha B. Sources and toxicity of hexavalent chromium. J. Coordinat. Chem. 2011, 64, 1782–1806. 10.1080/00958972.2011.583646. [DOI] [Google Scholar]
  5. Guidelines for Drinking Water Quality, fourth ed., WHO, Geneva, 2011. [Google Scholar]
  6. Anderson R. A. Chromium as an Essential Nutrient for Humans. Regul. Toxicol. Pharmacol. 1997, 26, S35–S41. 10.1006/rtph.1997.1136. [DOI] [PubMed] [Google Scholar]
  7. Bagchi D.; Stohs S. J.; Downs B. W.; Bagchi M.; Preuss H. G. Cytotoxicity and oxidative mechanisms of different forms of chromium. Toxicology 2002, 180, 5–22. 10.1016/S0300-483X(02)00378-5. [DOI] [PubMed] [Google Scholar]
  8. Ramsey J. D.; Xia L.; Kendig M. W.; McCreery R. L. Raman spectroscopic analysis of the speciation of dilute chromate solution. Corros. Sci. 2001, 43, 1557–1572. 10.1016/S0010-938X(00)00145-1. [DOI] [Google Scholar]
  9. Rakhunde R.; Deshpande L.; Juneja H. D. Chemical Speciation of Chromium in Water: A Review. Crit. Rev. Environ. Sci. Technol. 2012, 42, 776–810. 10.1080/10643389.2010.534029. [DOI] [Google Scholar]
  10. Richard F. C.; Bourg A. C. M. Aqueous geochemistry of chromium: A review. Water Res. 1991, 25, 807–816. 10.1016/0043-1354(91)90160-R. [DOI] [Google Scholar]
  11. Standard Methods for the Examination of Water and Wastewater, Method 3111: Metals by flame atomic absorption spectrometry, AWWA, APHA, 18th Edition, 1992.
  12. U. S. Environmental Protection Agency . Method 200.7. Trace elements in water, solids, and biosolids by inductively coupled plasma-atomic emission spectrometry. Revision 5.0. EPA 821-R-01-010, 2001.
  13. U.S. Environmental Protection Agency . Method 200.8, Revision 5.5. Determination of Trace Elements in Waters and Wastes by Inductively Coupled Plasma-Mass Spectrometry, EPA-821/R-99-017, 1999.
  14. Standard Methods for the Examination of Water and Wastewater, Method 3500-Cr-D: Colorimetric method, AWWA, APHA, 18th Edition, 1992.
  15. Fleischmann M.; Hendra P. J.; McQuillan A. J. Raman spectra of pyridine adsorbed at a silver electrode. Chem. Phys. Lett. 1974, 26, 163–166. 10.1016/0009-2614(74)85388-1. [DOI] [Google Scholar]
  16. Jeanmaire D. L.; Van Duyne R. P. Surface Raman electrochemistry. Part 1. Heterocyclic, aromatic and aliphatic amines adsorbed on the anodized silver electrode. J. Electroanal. Chem. 1977, 84, 1–20. 10.1016/S0022-0728(77)80224-6. [DOI] [Google Scholar]
  17. Kneipp K.; Moskovits M.; Kneipp H.. Surface-Enhanced Raman Scattering: Physics and Application, Springer, Berlin; 2006. [Google Scholar]
  18. Nie S.; Emory S. R. Probing Single Molecules and Single Nanoparticles by Surface-Enhanced Raman Scattering. Science 1997, 275, 1102–1106. 10.1126/science.275.5303.1102. [DOI] [PubMed] [Google Scholar]
  19. Feilchenfeld H.; Siiman O. Surface Raman Excitation and Enhancement Profiles for Chromate, Molybdate, and Tungstate on Colloidal Silver. J. Phys. Chem. 1986, 90, 2163–2168. 10.1021/j100401a035. [DOI] [Google Scholar]
  20. Feilchenfeld H.; Siiman O. Adsorption and Aggregation Kinetics and Its Fractal Description for Chromate, Molybdate, and Tungstate Ions on Colloidal Silver from Surface Raman Spectra. J. Phys. Chem. 1986, 90, 4590–4599. 10.1021/j100410a024. [DOI] [Google Scholar]
  21. Siiman O.; Feilchenfeld H. Internal Fractal Structure of Aggregates of Silver Particles and Its Consequences on Surface-Enhanced Raman Scattering Intensities. J. Phys. Chem. 1988, 92, 453–464. 10.1021/j100313a042. [DOI] [Google Scholar]
  22. Du J.; Jing C. Preparation of Fe3O4@Ag SERS substrate and its application in environmental Cr (VI) analysis. J. Colloid Interface Sci. 2011, 358, 54–61. 10.1016/j.jcis.2011.02.044. [DOI] [PubMed] [Google Scholar]
  23. Xiao J.; Meng Y. Y.; Zhang P. L.; Wen W.; Liu Z. M.; Zhang T. Quantitative Analysis of Chromate (CrVI) by Normal Raman Spectroscopy and Surface-Enhanced Raman Spectroscopy Using poly(diallyldimethylammonium) Chloride–Capped Gold Nanoparticles. Laser Phys. 2012, 22, 1481–1488. 10.1134/S1054660X12100258. [DOI] [Google Scholar]
  24. Mosier-Boss P. A.; Lieberman S. H. Detection of Anions by Normal Raman Spectroscopy and Surface-Enhanced Raman Spectroscopy of Cationic-Coated Substrates. Appl. Spectrosc. 2003, 57, 1129–1137. 10.1366/00037020360695991. [DOI] [PubMed] [Google Scholar]
  25. Mosier-Boss P. A.; Lieberman S. H. Surface-Enhanced Raman Spectroscopy (SERS) and Molecular Modelling of the Chromate Interaction with 4-(2-Mercaptoethyl) pyridinium. Langmuir 2003, 19, 6826–6836. 10.1021/la030052c. [DOI] [Google Scholar]
  26. Hurley B. L.; McCreery R. L. Raman Spectroscopy of Monolayers Formed from Chromate Corrosion Inhibitor on Copper Surfaces. J. Electrochem. Soc. 2003, 150s, B367–B373. [Google Scholar]
  27. Ji W.; Wang Y.; Tanabe I.; Han X.; Zhao B.; Ozaki Y. Semiconductor-driven “turn-off” surface enhanced Raman scattering spectroscopy: application in selective determination of chromium(VI) in water. Chem. Sci. 2015, 6, 342–348. 10.1039/C4SC02618G. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Bu X.; Zhang Z.; Zhang L.; Li P.; Wu J.; Zhang H.; Tian Y. Highly sensitive SERS determination of chromium(VI) in water based on carbimazole functionalized alginate-protected silver nanoparticles. Sens. Actuators B: Chem. 2018, 273, 1519–1524. 10.1016/j.snb.2018.07.058. [DOI] [Google Scholar]
  29. Li Y.-S.; Wang K.; He P.; Huang B. X.; Kovacs P. Surface-enhanced Raman Spectroelectrochemical Studies of Corrosion Films on Implant Co-Cr-Mo Alloy in Biosimulating Solutions. J. Raman Spectrosc. 1999, 30, 97–103. . [DOI] [Google Scholar]
  30. Melendres C. A.; Pankuch M.; Li Y. S.; Knight R. L. Surface enhanced Raman spectroelectrochemical studies of the corrosion films on iron and chromium in aqueous solution environments. Electrochim. Acta 1992, 37, 2747–2754. 10.1016/0013-4686(92)85202-V. [DOI] [Google Scholar]
  31. Oblonsky L. J.; Devine T. M. A surface enhanced Raman spectroscopic study of the passive films formed in borate buffer on iron, nickel, chromium and stainless steel. Corros. Sci. 1995, 37, 17–41. 10.1016/0010-938X(94)00102-C. [DOI] [Google Scholar]
  32. Ye Y.; Liu H.; Yang L.; Liu J. Sensitive and selective SERS probe for trivalent chromium detection using citrate attached gold nanoparticles. Nanoscale 2012, 4, 6442–6448. 10.1039/c2nr31985c. [DOI] [PubMed] [Google Scholar]
  33. Wang H. H.; Liu C. Y.; Wu S. B.; Liu N. W.; Peng C. Y.; Chan T. H.; Hsu C. F.; Wang J. K.; Wang Y. L. Highly Raman-Enhancing Substrates Based on Silver Nanoparticle Arrays with Tunable Sub-10 nm Gaps. Adv. Mater. 2006, 18, 491–495. 10.1002/adma.200501875. [DOI] [Google Scholar]
  34. Biring S.; Wang H.-H.; Wang J.-K.; Wang Y.-L. Light scattering from 2D arrays of mono-dispersed Ag-nanoparticles separated by tunable nano-gaps: spectral evolution and analytical analysis of plasmonic coupling. Opt. Express 2008, 16, 15312–15324. 10.1364/OE.16.015312. [DOI] [PubMed] [Google Scholar]
  35. Cheng T.-Y.; Wang H.-H.; Chang S. H.; Chu J.-Y.; Lee J.-H.; Wang Y.-L.; Wang J.-K. Revealing local, enhanced optical field characteristics of Au nanoparticle arrays with 10 nm gap using scattering-type scanning near-field optical microscopy. Phys. Chem. Chem. Phys. 2013, 15, 4275–4282. 10.1039/c3cp43270j. [DOI] [PubMed] [Google Scholar]
  36. Lin B.-Y.; Hsu H.-C.; Teng C.-H.; Chang H.-C.; Wang J.-K.; Wang Y.-L. Unraveling near-field origin of electromagnetic waves scattered from silver nanorod arrays using pseudo-spectral time-domain calculation. Opt. Express 2009, 17, 14211–14228. 10.1364/OE.17.014211. [DOI] [PubMed] [Google Scholar]
  37. Han Y.-Y.; Lin Y.-C.; Cheng W.-C.; Lin Y.-Z.; Teng L.-J.; Wang J.-K.; Wang Y.-L. Rapid Antibiotic Susceptibility Testing of Bacteria from Patients’ Blood via Assaying Bacterial Metabolic Response with Surface-Enhanced Raman Spectroscopy. Sci. Rep. 2020, 10, 12538. 10.1038/s41598-020-68855-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Lian W.-N.; Shiue J.; Wang H.-H.; Hong W.-C.; Shih P.-H.; Hsu C.-K.; Huang C.-Y.; Hsing C.-R.; Wei C.-M.; Wang J.-K.; Wang Y.-L. Rapid detection of copper chlorophyll in vegetable oils based on surface-enhanced Raman spectroscopy. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess 2015, 32, 627–634. 10.1080/19440049.2015.1014867. [DOI] [PubMed] [Google Scholar]
  39. Zhang X.; Young M. A.; Lyandres O.; Van Duyne R. P. Rapid Detection of an Anthrax Biomarker by Surface-Enhanced Raman Spectroscopy. J. Am. Chem. Soc. 2005, 127, 4484–4489. 10.1021/ja043623b. [DOI] [PubMed] [Google Scholar]
  40. Altun A. O.; Bond T.; Pronk W.; Park H.-G. Sensitive Detection of Competitive Molecular Adsorption by Surface-Enhanced Raman Spectroscopy. Langmuir 2017, 33, 6999–7006. 10.1021/acs.langmuir.7b01186. [DOI] [PubMed] [Google Scholar]
  41. Xia L.; Akiyama E.; Frankel G.; McCreery R. Storage and Release of Soluble Hexavalent Chromium from Chromate Conversion Coatings Equilibrium Aspects of CrVI Concentration. J. Electrochem. Soc. 2000, 147, 2256–2562. [Google Scholar]
  42. Stunzi H.; Marty W. Early Stages of the Hydrolysis of Chromium (III) in Aqueous Solution. 1. Characterization of a Tetrameric Species. Inorg. Chem. 1983, 22, 2145–2150. 10.1021/ic00157a012. [DOI] [Google Scholar]
  43. Dvoynenko M. M.; Wang H.-H.; Hsiao H.-H.; Wang Y.-L.; Wang J.-K. Study of Signal-to-Background Ratio of Surface-Enhanced Raman Scattering: Dependences on Excitation Wavelength and Hot-Spot Gap. J. Phys. Chem. C 2017, 121, 26438–26445. 10.1021/acs.jpcc.7b08362. [DOI] [Google Scholar]
  44. Dvoynenko M. M.; Wang J.-K. Can electrodynamic interaction between a molecule and metal dominate a continuum background in surface-enhanced Raman scattering?. Phys. Chem. Chem. Phys. 2015, 17, 27258–27263. 10.1039/C5CP04633E. [DOI] [PubMed] [Google Scholar]
  45. Foo K. Y.; Hameed B. H. Insights into the modeling of adsorption isotherm systems. Chem. Eng. J. 2010, 156, 2–10. 10.1016/j.cej.2009.09.013. [DOI] [Google Scholar]
  46. Otto A. Theory of First Layer and Single Molecule Surface Enhanced Raman Scattering (SERS). Phys. Stat. Sol. A 2001, 188, 1455–1470. . [DOI] [Google Scholar]
  47. Tong S.-Y.; Li K.-A. The distribution of chromium(VI) species in solution as a function of pH and concentration. Talanta 1986, 33, 775–777. 10.1016/0039-9140(86)80187-4. [DOI] [PubMed] [Google Scholar]
  48. Limousin G.; Gaudet J.-P.; Charlet L.; Szenknect S.; Barthès V.; Krimissa M. Sorption isotherms: A review on physical bases, modeling and measurement. Appl. Geochem. 2007, 22, 249–275. 10.1016/j.apgeochem.2006.09.010. [DOI] [Google Scholar]
  49. Moosavi S.M.; Ghassabian S. Linearity of Calibration Curves for Analytical Methods: A Review of Criteria for Assessment of Method Reliability. In Calibration and Validation of Analytical Methods: A Sampling of Current Approaches; IntechOpen Limited: London, UK, 2018; 109–127, 10.5772/intechopen.72932. [DOI] [Google Scholar]
  50. Rai D.; Sass B. M.; Moore D. A. Chromium(III) Hydrolysis Constants and Solubility of Chromium(III) Hydroxide. Inorg. Chem. 1987, 26, 345–349. 10.1021/ic00250a002. [DOI] [Google Scholar]
  51. Rai D.; Moore D. A.; Hess N. J.; Rao L.; Clark S. B. Chromium(III) Hydroxide Solubility in The Aqueous Na+-OH–-H2PO4–-HPO24–-PO43–-H2O System: A Thermodynamic Model. J. Solution Chem. 2004, 33, 1213–1242. 10.1007/s10953-004-7137-z. [DOI] [Google Scholar]
  52. Carrington A.; Symons M. C. R. Structure and Reactivity of the Oxyanions of Transition Metals. Chem. Rev. 1963, 63, 443–460. 10.1021/cr60225a001. [DOI] [Google Scholar]
  53. Brown P. L.; Ekberg C. Hydrolysis of Metal Ions; Wiley-VCH: Weinheim, 2016. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao0c05020_si_001.pdf (1MB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES