Highly Selective Colorimetric Detection of Hydrochloric Acid Using Unlabeled Gold Nanoparticles and an Oxidizing Agent

Published: November 10, 2011

r 2011 American Chemical Society 9206 dx.doi.org/10.1021/ac202500m |Anal. Chem. 2011, 83, 9206–9212

LETTER

pubs.acs.org/ac

Highly Selective Colorimetric Detection of Hydrochloric Acid UsingUnlabeled Gold Nanoparticles and an Oxidizing AgentSuraj Kumar Tripathy, Ju Yeon Woo, and Chang-Soo Han*

School of Mechanical Engineering, Korea University, Seoul, 136713, South Korea

bS Supporting Information

Water pollution has become one of the most challengingissues for mankind. As rapid industrialization has increased

the release of toxic chemicals into natural water systems,1�6

both the illegal dumping and accidental leakage of wastes fromvarious industries have raised concern among environmentalists.Chlorine-containing inorganic acids in general, and hydrochloricacid (HCl) in particular, are among the most abundant waterpollutants.7,8 Because of its extensive use, excellent stability, andhigh mobility in aqueous environments, HCl has the poten-tial to acidify wetlands and water resources.9 However, a rapidand selective detection technique for HCl has not yet beendeveloped.

Here, we describe a colorimetric system for the detection ofHCl in aqueous environments using unlabeled gold nanoparticle(AuNP) probes. This nonaggregation-based detection systemexploits the fact that aqueous chloro species induce the rapidleaching of AuNPs in an aqueous dispersion containing a strongoxidizing agent, such as HNO3 or H2O2.

10 This process leads tothe marked damping of the surface plasmon resonance (SPR)peak of the AuNP dispersion. The simple and cost-effectivetechnique is highly selective for HCl over several commonmineral acids, salts, and anions and is useful for the rapiddetection of HCl at concentrations as low as 500 ppm, far lessthan that deemed the hazardous limit in the near or long-term(3000 ppm and 1000 ppm, respectively).11

Colorimetric sensors are molecular or ionic sensors that signalanalyte interaction through a change in color. They offer a simpleand convenient platform for analyte detection, and AuNPsensors have received much interest as unique and tunablecolorimetric sensors.12 The optical properties of AuNPs aredominated by SPR, the collective oscillation of electrons at their

surfaces, in resonance with the incident electromagnetic radia-tion. The SPR absorption wavelengths of AuNPs are highlydependent upon their size, shape, and refractive index, and thusany minor perturbation in their chemical environment, surfacestructure, or aggregation may lead to irreversible or reversiblecolorimetric changes of their dispersions.12 Appropriately func-tionalized ligands on the AuNP surface can induce a shift in theabsorbance peak or can quench fluorescence in the presence ofan analyte or environmental response.13,14 Current AuNP sensorapplications include the detection of glucose, protein modifica-tions, or metal ions.15�17 Although these optical chemosensorshave been of great value in ionic and molecular sensing, furtherresearch is needed to enable their practical application as environ-mental sensors.

The simplicity and cost-effectiveness of colorimetric, nonag-gregation-based sensors using unlabeled metal nanoparticlesmakes them very user-friendly.18 In one of the earliest attemptsto develop a colorimetric sensor, Chen et al.19 developed asystem for the detection of lead ions based on the leaching ofunlabeled AuNPs. This idea was further extended by Wu et al.,20

who used the leaching kinetics of unlabeled AuNPs in thepresence of pyrophosphate to develop an Fe3+ sensor. Becausethey are free of the requirements for complex labeling agents andspecial working environments, similar techniques should be aprimary focus of sensor research. However, the development ofmolecular sensors for environmental systems is complicated dueto the presence of multiple ions in the analytical system and the

Received: September 21, 2011Accepted: November 10, 2011

ABSTRACT: We report a colorimetric system for the detection of HCl inaqueous environments using unlabeled gold nanoparticle (AuNP) probes. Thisnonaggregation-based detection system relies on the ability of chloro species tocause rapid leaching of AuNPs in an aqueous dispersion containing a strongoxidizing agent, such as HNO3 orH2O2. The leaching process leads to remarkabledamping of the surface plasmon resonance peak of the AuNP dispersion. Thismethod works only with AuNPs of a particular size (∼30 nm diameter). It ishighly selective for HCl over several common mineral acids, salts, and anions.This simple and cost-effective sensing system provides rapid and simple detectionof HCl at concentrations as low as 500 ppm (far below the hazard limit) in naturalwater systems.

9207 dx.doi.org/10.1021/ac202500m |Anal. Chem. 2011, 83, 9206–9212

Analytical Chemistry LETTER

possibility of various interfering reactions with common molec-ular species, such as salts and minerals. In the present study, weexplored the utility of unlabeled AuNPs for the selective colori-metric detection of HCl in aqueous environmental systems.

AuNPs dispersed in an aqueous medium were procured fromMijitech (Korea). HCl, HF, and H2S were from Junshi chemicals.Zn chloride, Sn chloride, Na perchlorate, Zn nitrate, andNa phos-phate were from Sigma. All chemicals were of analytical grade andusedwithout further purification. Deionized water (18.2MΩ cm)was prepared using a Milli-Q water system. Ultraviolet�visible(UV�vis) spectra were recorded using a 1 mL quartz cuvette in aUV�vis spectrophotometer (Scinco-SD 1000).

The sensing process was performed as follows: 200 μL of theAuNP dispersion was placed in a 5 mL vial and diluted with 1 mLof deionized water. A known amount of the oxidizing agent(HNO3 or H2O2) was added, and the mixture was incubated for30 min. UV�vis spectra were taken to confirm that the additionof the oxidizing agent did not affect the SPR spectrum. Standardsolutions of various mineral acids and other salts (100, 200, 500,800, 1000, 2000, and 5000 ppm) were prepared and preserved

for further use. For each sample tested, a 1 mL aliquot was addedto each vial containing the AuNP dispersion. The vials wereshaken well and then incubated for 30min without manipulation.The colorimetric change was assessed visually and then byUV�vis spectroscopy. The relative sensitivity is expressed bythe following equation:

S ¼ ðAbs0 � AbsAÞ=Abs0where the absorbance intensity at 530 nm (A530) of the oxidant-containing AuNP dispersion in the absence and presence of theanalyte is represented by Abs0 and AbsA, respectively.

A schematic view of the reaction was expected to occurbetween AuNPs and HCl in the presence of a strong oxidizingagent as shown in Figure 1a. As a noblemetal, gold is invulnerableto attack by any strong acid or base, but in the presence of astrong oxidant and chloride ions, it can be dissolved (leached) asthe gold chloride species [Au(Cl)2]h or [Au(Cl)4]h in the follow-ing reaction:13

Au0ðaqÞ þ nHþ þ nCl� f ½AuðClÞx��

where x = 2 or 4 depending upon the concentration of chlorideion, pH, and nature of the oxidizing agent. The stability of thegold chloride ions depends upon the Au and HCl concentrationsand the electrochemical potential (Eh) of the system.13

The above leaching reaction is precisely what occurs when Auis leached in aqua regia, which contains concentratedHNO3. Theuse of chlorine as an oxidant is well documented, with chlorina-tion being a common method of recovering gold prior to theinvention of cyanidation.21 However, this simple technique hasnot been previously exploited for the development of a chlorineor HCl sensor, possibly because AuNPs are stable at higher pH;the technique suggested by Chen et al.21 is applicable only atalkaline pH (pH 7�12). Furthermore, aqueous dispersions ofAuNPs in the size range normally used for chemosensor applica-tions (2�12 nm in diameter) are unstable at low pH; they arerapidly destroyed by strong acids.

As shown in Figure 1S (see the Supporting Information),AuNPs (∼12 nm diameter) are readily dissolved by any of themineral acids even in the absence of an oxidizing agent. Expectingthat the stability of the AuNPs would be enhanced by increasingtheir size, we examined the stability of AuNP dispersions(20�30 nm diameter) in the presence of different mineral acids(including HCl) and oxidizing agents and found that they wereindeed more stable (Figure 1S in the Supporting Information).However, larger AuNPs (∼90 nm in diameter) yielded a lessintense color, limiting their utility as sensing probes.

The intensity of color due to the SPR effect of a AuNPdispersion decreases sharply as the particle size increases.22,23 Asshown in Figure 1S in the Supporting Information, a dispersioncontaining∼90 nm particles is light pink, whereas those contain-ing∼12 or∼30 nm particles are much deeper in color, appearingwine-red. In addition, the larger particles do not disperse stably;they show a tendency to settle as the pH of the mediumchanges.24 Addition of analyte molecules to the larger AuNPsgreatly reduces the intensity of the dispersion color, either becauseof the decreased intensity of the SPR phenomenon or because thelarger particles aggregate at the lower pH.22�24 Taking thesefindings into consideration, we selected AuNPs ∼30 nm indiameter as the basis of our HCl sensing probe.

The colorimetric test for HCl detection was carried out atroom temperature under atmospheric pressure. As shown in

Figure 1. (a) Leaching of Au NPs by HCl in the presence of strongoxidizing agents, (b) effect of addition of HCl and possible interferingchemicals on the AuNPs dispersion in presence of HNO3, and (c) effectof addition of HCl and possible interfering chemicals on the Au NPsdispersion in presence of H2O2.



Figure 1a, the intense wine-red color of an aqueous dispersion of30 nm AuNPs gradually decreased in intensity as increasingamounts of HCl were added in the presence of HNO3. The colorchange also occurred more rapidly as the HCl concentration

increased. At HCl concentrations greater than 2000 ppm, thedispersion became colorless within 15�20 min. Experimentsusing H2O2 rather than HNO3 as the strong oxidizer yieldedsimilar observations (Figure 1b), although the magnitude of the

Figure 2. UV�visible absorbance spectrum of AuNPs aqueous dispersion in the presence of a different concentration of HCl (a) 100, (b) 200, (c) 500,(d) 1000, (e) 2000, (f) 5000 ppm), challenged with (a) HNO3 and (b) H2O2. Effect of the the addition of different concentrations of HCl on theabsorbance of Au NPs and relative sensitivity (c) in the presence of HNO3 and (d) in the presence of H2O2.



color change was slightly decreased, and the color change processwas slower, requiring up to 60 min for completion. At very lowHCl concentrations, the process might be even further delayed.

The reddish color of the AuNP dispersion in the HNO3

oxidant system was unaffected by the presence of mineral acidsother than HCl or commonly encountered salts (Figure 1a);none of the other chemical species tested induced a color changecomparable to that caused by HCl. These results demonstratethat the presence of the aqueous hydrochloric acid species leadsto specific and rapid leaching of the AuNPs in the presence of astrong oxidant.

To quantitatively evaluate the colorimetric change induced byHCl, we investigated the UV�vis spectrum of the AuNP dis-persion. As shown in Figure 2a, an aqueous dispersion of AuNPsdisplayed a distinct SPR absorbance peak at 530 nm. The addi-tion of HCl red-shifted the absorbance peak to 535 nm andsharply decreased its intensity. We attribute the sharp decrease in

peak intensity to the leaching of the metallic AuNPs by HCl.With an increase in the concentration of this acid, the reductionin the peak intensity is more pronounced. A similar result wasobtained when HNO3 was replaced with H2O2 (Figure 2b), butthe peak was red-shifted to 540 nm rather than 535 nm becausethe H2O2 oxidant system exhibits less SPR damping.

Liu et al.25 reported similar observations for a detection systemusing protein-capped, fluorescent AuNPs. In the presence ofcyanide ions, the fluorescence spectrum was quenched by theleaching of the AuNPs in the aqueous system.25 In addition to arapid decrease in the fluorescence intensity, they also observeda slight red-shift in the fluorescence spectrum, similar to ourpresent findings.

Comparative plots of acid concentration vs absorbance (andrelative sensitivity) for our system using HNO3 or H2O2 as theoxidizing agent are shown in Figure 2c,d. The data were well fit byHill’s sigmoid equation and yielded sensitivity values of 0.08, 0.27,

Figure 3. (a) XRD pattern of test Au NPs before and after adding HCl, respectively. (b) XPS spectrum of test Au NPs with and without HCl,respectively.

Figure 4. Parts a, c, and e show the TEM images of test AuNPs and test AuNPs with 1000 and 2000 ppm of HCl, respectively. Parts b, d, and e show thecorresponding particle size distribution graph.



and 0.98 for HCl concentrations of 500, 1000, and 5000 ppm,respectively. The lower limit for the detection of HCl is about500 ppm, which is less than the hazardous limit in environmentalwater sources.11

To better understand the sensing mechanism of our system,we used X-ray diffraction (XRD) to examine the structural chan-ges associated with the sensing reaction. After HNO3-containingAuNP dispersions were carefully prepared with and without HCl,equal amounts were deposited on a quartz slide for analysis. Asshown in Figure 3a, the virgin AuNPs showed three clear peaks at2θ=37.8, 44.51, and64.60 corresponding to (111), (200), and (220)planes of metallic Au with a face-centered cubic structure. Analysisusing Scherrer’s formula yielded amean crystallite diameter of 27 nm.As clearly shown in Figure 3a, the XRD pattern of the sampleschanged dramatically with the sensing reaction. The disappearance of

the peaks corresponding to metallic Au suggests the leaching of themetal in the presence of HCl and a strong oxidizing agent.

To confirm these findings, we used X-ray photoelectronspectroscopy to examine the interaction between HCl and theAuNPs. The Au 4f7/2 spectrum of the virgin AuNPs (in thepresence of HNO3) could be deconvoluted into two peaks cen-tered at 81.0 and 84.5 eV, corresponding to the binding energiesof Au0 and Au+, respectively;25 the addition of 5000 ppm HClwith HNO3 caused the intensity of both peaks to dramaticallydecrease and shifted them to 81.2 and 86.5 eV, respectively. Thisresult indicates that HCl induced the oxidation of the AuNPsfrom the Au0 to the Aun+ state in the presence of the HNO3

oxidant (Figure 3b). Wang et al.26 observed a similar change inpeak position during the formation of Au3+ ions.

We investigated the morphological change associated with thesensing reaction using transmission electron microscopy. As shownin Figure 4 (boxes), before the addition of HCl, near sphericalAuNPs ranging in size from 25 to 35 nm in diameter were clearlyvisible. However, after the addition of 1000 or 2000 ppm HCl,both the size and number of the AuNPs decreased significantly.All the above experimental results are in good agreement with thesuggested leaching mechanism.

Two common approaches are employed to detect HCl in theaqueous environment, neither of which can discriminate amongchloride ion sources.27,28 One approach isMohr’s test, a two-stepprocess in which the pH of the system is first adjusted to an acidicpH and then AgNO3 is added. In the presence of HCl, theAgNO3 forms a AgCl precipitate in the following reaction:

AgNO3 þ MxClyðaqÞ f AgClðprecipitateÞ

where M can be H, Sn, Zn, Cu, or another metal and x and yrepresent the valences of elements M and Cl, respectively. Theother common approach toHCl detection is quantitative analysisby acid�base titration.29,30 The time-consuming and nonspecificnature of both of these traditional detection techniques justifiesthe development of alternative, nanoparticle-based detectiontechniques.

Most optical chemosensors are readily disturbed by thepresence of contaminating salts or ions (anions, in particular).To investigate the specificity of our system for HCl, we measuredits absorbance response to three other acids (H2S, HF, andCH3COOH), five common salts (Zn chloride, Sn chloride,Na chloride, Fe3+ chloride, and Na iodide), and three anions(phosphate, chlorate, and nitrate). As clearly shown in Figure 5a,none of these ions or salts induced remarkable change in theintensity and/or position of the SPR peak AuNP dispersion.Only HCl induced a notable damping of the absorbance spec-trum (Figure 5b), whereas no significant effects were noted fromthe presence of any of the other chemical species. However, weshould note that the SPR-based spectral damping event wasslightly affected by the presence of Fe3+ chloride. A similar resultwas observed by Zhang et al.31 during their detection of iodideusing Cu�Au core/shell nanoparticles. The color change inducedby Fe3+ ions was easily detectable by the naked eye.

To evaluate whether the AuNP-based sensing system dis-cussed here is applicable to natural systems, we collected localsamples of ground, tap, river, and seawater for analysis with thissystem. In the absence of added HCl, the addition of these real-world water samples had no significant effect on the signal(Figure 5a,b); thus, like deionized water, these water samplesdid not appear to affect the performance of the sensing system.

Figure 5. (a) Effects of different concentrations of HCl and othercommonly encountered chemicals on the relative damping of the SPspectrum of Au NPs dispersion (b) Effects of a concentration of HCl(5000 ppm) and other commonly encountered chemicals on the relativedamping of the SP spectrum in various water samples.



On the other hand, the addition of real-world water samplescharged with 5000 ppm of HCl led to a significant change in theabsorbance value; interestingly, all of the water samples yieldedcomparable absorbance changes in this experiment. (SP dampingdata for several water samples are shown in Figure 2S in theSupporting Information.)

In further investigations, we attempted to demonstrate thequantitative analysis of HCl in artificial wastewater samplesprepared by mixing together 500 ppm of each of the variouschemical species investigated in this paper. As shown inFigure 6a, addition of this artificial effluent did not change theSPR peak of the AuNPs even after 30 min. We then used oursensing system to analyze two different artificial effluent samplescontaining an unknown amount of HCl (blind HCl samples).The addition of the blind HCl samples damped the SPRabsorbance peaks, as expected. A fit of the absorbance valuesto Hill’s sigmoid curve (Figure 6b) showed the concentrations ofHCl to be 900 and 1050 ppm.

In summary, we have demonstrated a technique for theselective detection of HCl in an aqueous environment basedupon the damping of AuNP SPR absorbance by HCl-triggeredleaching. As this technique does not involve any labeling, it issimple and user-friendly. It demonstrates good selectivity in thepresence of commonly encountered salts and anions and can beused to directly analyze natural water systems. These featuresmake this system a potentially powerful tool for the investigation

of HCl in marine ecosystems prone to industrial pollution.Furthermore, this technique could be extended to the detectionof HNO3 orH2O2 in aqueous environments (see Figure 3S in theSupporting Information). We are currently developing systemsfor the selective detection of other mineral acids and commonanions.

’ASSOCIATED CONTENT

bS Supporting Information. Additional information as notedin text. This material is available free of charge via the Internet athttp://pubs.acs.org.

’AUTHOR INFORMATION

Corresponding Author*E-mail: [email protected].

’ACKNOWLEDGMENT

This research was supported by Basic Research Fund of KoreaInstitute of Machinery & Materials (KIMM), and partially sup-ported by Center for Advanced Soft Electronics in Global FrontierProgram and Center for Nanoscale Mechatronics and Manufactur-ing in Frontier Program from MEST, Korea.

’REFERENCES

(1) Burgess, W. G.; Hoque, M. A.; Michel, H. A.; Voss, C. I.; Breit,G. N.; Ahmed, K. M. Nat. Geosci. 2010, 3, 83.

(2) Editorial. Save our cities. Nature 2010, 467, 883.(3) Vorosmarty, C. J.; Mclntyre, P. B.; Gessner, M. O.; Dudgeon, D.;

Prusevich, A.; Green, P.; Glidden, S.; Bunn, S. E.; Sullivan, C. A.;Liermann, C. R.; Davies, P. M. Nature 2010, 467, 551.

(4) Gilbert, N. Nature 2010, 466, 806.(5) Qiu, J. Nature 2010, 466, 308.(6) Chakraborti, D.; Rahman, M.; Paul, K.; Chowdhury, U. K.;

Sengupta, M. K.; Lodh, D.; Chandra, C. R.; Saha, K. C.; Mukharjee,S. C. Talanta 2002, 58, 3.

(7) Alwell, C.; Manderscheid, B.; Meesenburg, H.; Bittersohl, J.Nature 2000, 407, 856.

(8) Meron, D.; Atias, E.; Kruh, L. I.; Elifantz, H.; Minz, D.; Finel, M.;Banin, E. Inter. Soc. Microbial Ecol. J. 2011, 5, 51.

(9) Evans, C. D.; Monteith, D. T.; Fowler, D.; Cape, J. N.; Brayshaw,S. Environ. Sci. Technol. 2011, 45, 1887.

(10) Sardar, R.; Funston, A. M.; Mulvaney, P; Murray, R. W.Langmuir 2009, 25, 13840.

(11) http://www.lycos.com/info/hydrochloric-acid--water.html(12) Mulvaney, P. Langmuir 1996, 12, 788.(13) Gupta, C. K. Chemical Metallurgy: Principles & Practice;

Wiley-VCH Verlag GmbH: Weinheim, Germany, 2003.(14) Lin, Y. W.; Huang, C. C.; Chang, H. T. Analyst 2011, 136, 863.(15) Xia, F.; Zuo, X.; Yang, R.; Xiao, Y.; Kang, D.; Vall�ee-B�elisle, A.;

Gong, X.; Yuen, J. D.; Hsu, B. Y.; Heeger, A. J.; Plaxco, K. W. Proc. Natl.Acad. Sci. U.S.A. 2010, 107, 10837.

(16) Radhakumary, C.; Sreenivasan, K. Anal. Chem. 2011, 83, 2829.(17) Lisowski, C. E.; Hutchison, J. E. Anal. Chem. 2009, 81, 10246.(18) Choi, Y.; Park, Y; Kang, T; Lee, L. P. Nat. Nanotechnol. 2009,

4, 742.(19) Chen, Y. Y.; Chang, H. T.; Shiang, Y. C.; Hung, Y. L.; Chiang,

C. K.; Huang, C. C. Anal. Chem. 2009, 81, 9433.(20) Wu, S. P.; Chen, Y. P.; Sung, Y. M. Analyst 2011, 136, 1887.(21) Hilson, G.; Monhemius, A. J. J. Clean. Prod. 2006, 14, 1158.(22) Jain, P. K.; Lee, K. S.; El-Sayed, I. H.; El-Sayed, M. A. J. Phys.

Chem. B 2006, 110, 7238.

Figure 6. (a) Effect of the addition of artificial effluent to the test AuNPs followed by the addition of blind HCl samples and (b) determina-tion of the amount of HCl from the graph.



(23) Amendola, V.; Meneghetti, M. J. Phys. Chem. C 2009, 113,4277.(24) Nam, J.; Won, N.; Jin, H.; Chung, H.; Kim, S. J. Am. Chem. Soc.

2009, 131, 13639.(25) Wang, S.; Qian, K.; Bi, X. Z.; Huang, W. J. Phys. Chem. C 2009,

113, 6505.(26) Wang, C. Y.; Liu, C. Y.; Zheng, X. Colloid Surf., A 1998,

131, 271.(27) Liu, Y.; Ai, K.; Cheng, X.; Huo, L.; Lu, L. Adv. Funct. Mater.

2010, 20, 951.(28) http://msds.chem.ox.ac.uk/HY/hydrochloric_acid.html.(29) Jeffery, G. H.; Bassett, J.; Mendham, J.; Denny, R. C. Vogel’s

Textbook of Quantitative Chemical Analysis, 5th ed.; Longman Scientificand Technical: Harlow, Essex, UK, 1989.(30) Irons, G. P.; Greenway, G. M. Anal. Proc. 1994, 31, 91.(31) Zhang, J.; Xu, X.; Yang, C.; Yang, F.; Yang, X.Anal. Chem. 2011,

83, 3911–3917.

Documents

Highly Selective Colorimetric Detection of Hydrochloric Acid Using Unlabeled Gold Nanoparticles and an Oxidizing Agent