Electrochemistry Communications 8 (2006) 1369–1374 www.elsevier.com/locate/elecom

Trace analysis of hydrogen sulfide by monitoring As(III) at a poly(L-lactide) stabilized gold nanoparticles modified electrode Yue-Shain Song, Govindan Muthuraman, Jyh-Myng Zen

*

Department of Chemistry, National Chung Hsing University, 250 Kuo-Kuang Road, Taichung 402, Taiwan Received 13 May 2006; received in revised form 19 June 2006; accepted 19 June 2006 Available online 26 July 2006

Abstract An indirect electrochemical approach for the determination of sulfide by measuring the inhibited oxidation current of As(III) in HCl medium using a poly(L-lactide) stabilized gold nanoparticles modified screen-printed electrode (designated as PLA-AuNP/SPE) is described in this work. The PLA-AuNP/SPE is applied effectively to detect As(III) in HCl medium and can be tolerable from the interference of Cu, Cd, Fe, Zn, Mn and Ni and hence provides a direct and selective detection method for As(III) in natural waters. An substantial decrease in the oxidation peak current of As(III) is observed in the presence of sulfide due to the formation of As2S3 complex. Surface analysis by XPS together with electrochemical and Raman spectroscopic characterization were used to elucidate the detection mechanism. Under the optimal experimental conditions, a calibration plot with a wide linear range up to 700 lM was achieved with a detection limit (S/N = 3) of 0.04 lM. The sensitivity observed in present study is good enough to detect sulfide at levels lower than the current EPA standard. The proposed electrochemical approach was successfully used for the determination of sulfide in acid rain and hot spring water. Ó 2006 Elsevier B.V. All rights reserved. Keywords: Sulfide; Arsenite; Differential pulse voltammetry; Screen-printed electrode

1. Introduction The toxicity of sulfide in its liberated hydrogen sulfide (H2S) form is of widespread awareness and the corresponding risk factors are manifested with exposure level of working atmosphere [1–4]. In many clinical cases sulfide poisoning generally starts at levels from 30 to 3000 g/L and lethal doses upon exposure of hydrogen sulfide are in the range of 300–1000 ppm. Since such a low level of sulfide is enough to cause physiological distress, there has been an increasing need for portable devices for monitoring trace levels of sulfide in a variety of environmental and industrial applications [5–9]. Although various detection methods have been developed to detect sulfide, electrochemical method is compara-

*

Corresponding author. Fax: +886 4 22854007. E-mail address: [email protected] (J.-M. Zen).

1388-2481/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.elecom.2006.06.014

tively sensitive, fast and simple [5–14]. Electroanalysis can also permit miniature assemblies/chips with disposable type of low-cost screen-printed electrodes (SPEs) [15]. Conventional bare macro-electrodes such as Au, Pd, Ag and Pt [16–20] were initially used and various chemically modified electrodes (CMEs) based on different modifiers were further reported for the sulfide assays [14,21–24]. Our recent reports discussed mainly the disposable type CMEs with modifiers such as cinder hexacyano cobalt hybrid [25], cinder/tetracyano nikelate [26] and ferricyanide incorporated Tosflex [27] for trace level detection of sulfide. In this study, an attempt is made for the indirect determination of sulfide based on its inhibited effect on the As(III) signal at the PLA-AuNP/SPE. It is well-documented that sulfide has the tendency to form monolayer adsorption on Au electrode and also the possibility of forming complex with H2S in the form of As2S3 [28,29]. It is thus expected that the substantial decrease in the oxidation peak current of As(III) in the presence of sulfide at the

1370

Y.-S. Song et al. / Electrochemistry Communications 8 (2006) 1369–1374

PLA-AuNP/SPE may be developed as an indirect electrochemical approach for the determination of sulfide. Factors that influence the sensitivity and selectivity of the PLA-AuNP/SPE were optimized in this study. Surface analysis (XPS), Raman spectral analysis and electrochemical studies have been conducted to elucidate the detection mechanism. Finally the proposed approach was applied to natural water analysis with satisfactory results. We believe this novel approach can meet the new challenge in ultratrace sulfide detection.

drop (18 lL) of PLA-AuNP solution was spread evenly on the SPE surface and allowed to dry at 70 °C for 5 min. Note that the PLA-AuNP solution can be coated on SPE effectively only after the treatment with Triton X-100 surfactant [31,32]. Without the Triton X-100 pretreatment, a drop of PLA-AuNP remained as a drop-like nature on the electrode surface presumably due to the effect of the nonpolar groups on the surface [33]. The pretreatment assures the good reproducibility for the preparation of the PLA-AuNP/SPE. 3. Results and discussion

2. Experimental

2.2. Instrumentation Electrochemical measurements were performed with a CHI 703 electrochemical workstation (Austin, TX, USA) in a three-electrode cell assembly. A complete cell setup consists of either a bare SPE or a chemically modified SPE as working electrode, an Ag/AgCl as reference electrode and a platinum disk as auxiliary electrode. The working solution used in this investigation was not degassed since oxygen did not interfere in the analysis. The SPE with a working area of 0.2 cm2 and a conductive track radius of 2.5 mm was purchased from Zensor R&D (Taichung, Taiwan). The measured average resistance was 85.64 ± 2.10 X/ cm. A gold disk electrode with 1.6 mm dia. (BAS) was used for comparison in experiments with the PLA-AuNP/SPE. 2.3. Preparation of the PLA-AuNP/SPE PLA (average molecular weight = 8000, PDI = 1.07) was prepared by a procedure as reported earlier [30]. In brief, certain amount of PLA (0.66 mg) was first dissolved in a THF (0.3 mL) containing HAuCl4 (0.83 mM, 10 mL) solution. The reaction mixture was stirred for 10 min at room temperature. Aqueous solution of NaBH4 (4.5 mM, 30 mL) was freshly prepared and added into the stirred solution. The reaction mixture was found to rapidly change color from yellow to brownish red, indicating the formation of gold nanoparticles. To prepare the PLA-AuNP/SPE, a

Fig. 1A shows the cyclic voltammetric response of 30 lM As(III) in the presence and absence of sulfide at

A

60 50 40 30

Current / μA

Sodium sulfide (Na2S Æ 9H2O), hydrogen tetrachloroauric acid (HAuCl4), sodium borohydrate (NaBH4) and L-lactide were obtained from Aldrich. Sodium arsenite (NaAsO2) (Fluka), THF (Merck) and all other compounds and reagents (ACS-certified reagent grade) were used as received. Millipore de-ionized water (18 MX/cm) was used throughout this investigation. Sulfur-containing samples of hot spring water and acid rain were collected from Yangmingshan National Park (Taipei, Taiwan) and area nearby a local thermal power plant (Taichung, Taiwan), respectively. Thorough filtration process was adopted (Millipore 0.2 lM white Nylon diameter 47 mm) prior to routine analysis.

3.1. Electrochemical detection of sulfide by monitoring As(III) signal

20 10 0

c

-10

b

-2 0 -30 0.5

a 0.4

0.3

0.2

0.1

0.0

-0.1

-0.2

-0.3

-0.4

-0.5

Potential / V

B

0

c

-25 -50

Current / μA

2.1. Chemical and reagents

-75 -100 -125

b

-150 -175 -200 -225 0.5

a 0.4

0.3

0.2

0.1

0.0

-0.1

-0.2

-0.3

Potential / V Fig. 1. Cyclic voltammetric (A) and differential pulse anodic stripping voltammetric (B) responses of 30 lM H3AsO3 at the PLA-AuNP/SPE in 1 M HCl in the absence (a) and presence (b) of 20 lM sulfide and (c) 50 lM sulfide. Scan rate was 100 mV/s for (A) and deposition time was 130 s at 0.5 V vs. Ag/AgCl for (B).

Y.-S. Song et al. / Electrochemistry Communications 8 (2006) 1369–1374

3.2. XPS and Raman spectroscopic characterization If it is not due to the adsorption of sulfide on the electrode surface, what could be the possible cause for the decrease in peak current in the presence of sulfide at the PLA-AuNP/SPE? The most proper answer must be the formation of arsenic sulfide (As2S3) in working solution. Surface analysis by XPS together with Raman spectroscopic characterization was applied to elucidate the detection mechanism. In order to identify the species responsible for the attenuation of peak current, XPS analysis was first performed at the PLA-AuNP/SPE. After deposition in solution containing H3AsO3 with/without Na2S at 0.5 V for 130 s, XPS spectra show no defined peak for sulfide and arsenic in either condition. In other words, when adding

A

-2 -4

Current / μA

-6 -8 -10

b

-12 -14

a -16 0.5

0.4

0.3

0.2

0.1

0.0

-0.1

-0.2

-0.3

-0.1

-0.2

-0.3

Potential / V

B

0 -10 -20

Current / μA

the PLA-AuNP/SPE in 1 M HCl at a scan rate of 100 mV/s. As can be seen, in the presence of As(III), a well-defined anodic peak at 0.14 V with a corresponding cathodic peak at 0.08 V was observed at the PLA-AuNP/SPE [34]. It is reported earlier by Compton’s group that oxidation and reduction follows multi-electron transfer and single electron transfer, respectively [35]. After the addition of 20 lM sulfide, an inhibited peak response can be clearly seen as shown in curve b. As the sulfide concentration increased to 50 lM (curve c), a completely diminished baseline like CV response was observed. In other words, the decrease in the peak response of As(III) may be taken as an indirect method for determination of sulfide. Note that, as concluded in our previous study [36], DPASV showed the highest resolution with the lowest noise As(III) signal and it was thus chosen for sulfide detection in subsequent studies. As shown in Fig. 1B, after the deposition at 0.5 V (vs. Ag/AgCl) for 130 s, the stripping peak of As(III) was found to decrease correspondingly with the addition of 10 lM Na2S (curve b) compared to the signal of 30 lM H3AsO3 alone (curve a). Again, a completely diminished baseline like response was observed (curve c) upon the addition of 50 lM Na2S at the PLA-AuNP/SPE. To validate the electroanalytical performance on sulfide detection, the same experiment was also done on a bare Au disk electrode for comparison. Fig. 2 depicts the comparative DPSV responses of both electrodes in 10 lM As(III). As shown in Fig. 2A, a clear passivity of sulfide was observed on the Au disk electrode surface for two continuous detection runs. This is as expected, as sulfide was reported to have the tendency of forming monolayer adsorption on Au electrode [28]. In contrast, good reproducibility between two runs indicates efficient regeneration of electrode surface with no sign of passivity on the PLAAuNP/SPE, as shown in Fig. 2B. Somehow, the adsorption of sulfide did not occur on the electrode surface of PLAAuNP/SPE. It is believed that the PLA polymer should have something to do with the observation. Nevertheless the unique property of the PLA-AuNP/SPE is essential for the success of sulfide analysis as will be discussed in later section.

1371

-30 -40 -50 -60 -70

a,b -80 0.5

0.4

0.3

0.2

0.1

0.0

Potential / V Fig. 2. Differential pulse anodic stripping voltammetric responses of (A) Au disk electrode and (B) PLA-AuNP/SPE when placed in 10 lM H3AsO3 solution both in the absence (a) and presence (b) of sulfide. Other conditions are the same as in Fig. 1B.

sodium sulfide and H3AsO3 in 1 M HCl medium, the As2S3 did not adsorb on the electrode surface. Note that As2S3 is soluble only in hot HCl and highly basic condition and sulfide ion can exist in various forms of H2S, HS and S2 at different pH (pKa1 = 6.96 and pKa2 = 11) [37]. Since sodium sulfide forms H2S in acidic medium and As2S3 is not soluble in HCl at room temperature [37,38], the asformed product may thus either dissolve in the reaction solution or precipitate as colloids in the form of insoluble complex, as illustrated in Scheme 1. By considering that arsenite exists in the form of H3AsO3 in acidic medium [39], our observation leads to the following proposed equations. Na2 S þ 2HCl ! H2 S þ 2NaCl

ð1Þ

2H3 AsO3 þ 3H2 S ! As2 S3 þ 6H2 O

ð2Þ

As mentioned earlier, the substantial decrease in the oxidation peak current of As(III) in the presence of sulfide at the PLA-AuNP/SPE can be used as an indirect

1372

Y.-S. Song et al. / Electrochemistry Communications 8 (2006) 1369–1374

A

200 180

Inhibition current / μA

160 140 120 100 80 60 40 20 0

0

10

20

30

40

50

60

70

80

[Na 2S] / μM

B 12000 (b)

10000

Scheme 1.

electrochemical approach for the determination of sulfide. Instead of reacting with H3AsO3, we have yet to account for the possibility of H2S to escape from the working solution. It is expected that the possibility of un-reacted H2S to escape from solution is low as H3AsO3 can react rapidly with H2S to form As2S3 with a high formation constant (log K = 24.0) [29]. To confirm this, the following experiments were performed. Sodium sulfide was step by step titrated into 30 lM H3AsO3 and the stripping peak current was continuously monitored. As shown in Fig. 3A, the inhibition current increases constantly with the titration of sulfide and remains constant after 45 lM of addition. This result is quantitatively consistent with Eq. (2), in which 2 moles of H3AsO3 can react with 3 moles of H2S to form 1 mole of As2S3. Most importantly, it clearly indicates the As2S3 complex formation in solution with no free H3AsO3 present in solution after the Na2S concentration in excess. To further identify the formation of As2S3 in solution, Raman spectroscopy was performed. As shown in Fig. 3B, the observation of a strong band at 343.2 cm1 (the characteristic band of As2S3 [29]) only at the (H3AsO3 + sodium sulfide) solution confirms the presence of As2S3 species in solution phase. 3.3. Optimization of analytical parameters After understanding the mechanism for the detection of sulfide, we turned our attention to optimizing the analytical response of the system. In order to get maximum sensitivity of As(III) at the PLA-AuNP/SPE, experimental conditions such as deposition potential, deposition time, and DPSV

Current

8000 6000 4000 2000 0

(a,c)

0

200

400

600

800

1000 1200 1400 1600

Relative cm-1 (Delta wave number) Fig. 3. (A) A plot of inhibition current against [Na2S] in the presence of 30 lM H3AsO3. (B) Raman spectra of (a) H3AsO3 alone, (b) H3AsO3 + Na2S and (c) Na2S alone in 1 M HCl.

parameters were optimized. As to the deposition potential and deposition time on As(III) detection, the peak current increases as the potential of the electrode becomes more negative from 0.1 to 0.6 V. However, the peak current drops rapidly as the potential is more negative than 0.6 V caused by the residual oxygen reduction and H2 evolution on the electrode surface. Therefore, a deposition potential of 0.5 V was selected for further studies. For higher concentration (5 lM) of As(III), the peak current increases as the deposition time increases and starts to level off around 100 s. For a lower concentration (0.1 lM) of As(III), it takes about 500 s for the peak current to level off. Therefore, in order to use all the AuNP sites available, a longer time is needed for the lower concentration of As(III). In this study, a deposition time of 130 s was used in subsequent studies. The DPSV parameters that were investigated were the pulse period, the amplitude, the pulse width, and the sample width. These parameters are interre-

Y.-S. Song et al. / Electrochemistry Communications 8 (2006) 1369–1374 0 140

Inhibition current / μA

A Current / μA

-50 -100 2-

100 μM S

-150

120 100 80 60 40 20

2-

80 μM S

0

20

40

60

80

100

120

80

100

120

2-

[S ] / μ M

2-

60 μM S 240 μM S

-200

2-

+ real sample + 20 μM S

-250

+ real sample 3+

800 μM As

-300 0.5 0.4 0.3 0.2

0.1 0.0 -0.1 -0.2 -0.3 -0.4

Potential / V 0 160

-50 -100 2-

100 μM S

-150

2-

80 μM S

2-

60 μM S

Inhibition current / μA

B Current / μA

lated and have a combined effect on the response, but here only the general trends will be examined. A series of pulse period was studied and the signal increased initially with pulse period and remained constant after a pulse period of 0.05 s. Similarly, increase in the amplitude causes an increase in the As(III) peak up to 0.08 V. The peak potential shifts to the negative direction with increasing amplitude. The effects of pulse width and the sample width were not significant in the system. Overall, the best instrument settings for DPSV were as follows: pulse period, 0.05 s; pulse amplitude, 0.08 V; pulse width, 0.025 s; sample width, 0.0167 s. Calibration experiments were carried out by measuring the inhibition current of sulfide using DPSV with 130 s of deposition time at 0.5 V vs. Ag/AgCl in the presence of 800 lM H3AsO3. As shown in Fig. 4A, the peak current decreases accordingly upon addition of sodium sulfide. Under the optimized experimental conditions, the calibration plot is linear with a wide range up to 90.94 ppm (i.e., 700 lM) and a detection limit of 0.04 lM (S/

1373

140 120 100 80 60 40

0

20

2-

-200

40

60 2-

[S ] / μ M

40 μM S 2+ real sample + 20 μM S + real sample

-250 3+

800 μM As

A

-300 0.5 0.4

0 2-

700 μM S

Current / μA

0.1 0.0 -0.1 -0.2 -0.3 -0.4

Potential / V

-50

Fig. 5. Differential pulse anodic stripping voltammetric responses of sulfide and the obtained standard addition plots in real water samples of (A) acid rain and (B) hot spring water at the PA-Au NP/SPE. Other conditions are the same as in Fig. 1B.

2-

500 μM S

-100

0.3 0.2

-150

N = 3). Such a performance characteristics along with high stability was found to compare favorably to those of our previous reported methods for detecting sulfide [25–27].

2-

300 μM S

-200 2-

100 μM S

-250

3.4. Real sample analysis 2-

0 μM S

-300 0.5

0.4

0.3

0.2

0.1

0.0

-0.1

-0.2

-0.3

B

300

Inhibition current / μ A

Potential / V

250 200 150 100 50 0

0

100

200

300

400

500

600

700

800

The applicability of the approach was demonstrated through the determination of sulfide in hot spring water and acid rain. As can be seen in Fig. 5, both real water samples show an As(III) stripping peak at 0.17 V. By applying the standard addition method, linear plots of inhibited current were obtained for both samples and used for the calculation of sulfide concentration. The sulfide values measured from the present work were 25.50 lM (acid rain) and 41.78 lM (hot spring water), respectively. To validate the accuracy of the proposed method, both water samples were also subjected to the analysis by the standard Caro’s reaction method [40]. The fact that very close values of 24.67 lM (acid rain), and 39.85 lM (hot spring water) of sulfide observed indicates good correlation and efficiency of the proposed electrochemical method. 4. Conclusions

2-

[S ] / μ M Fig. 4. (A) Typical differential pulse anodic stripping voltammetric responses at the PLA-AuNP/SPE with various concentration of Na2S in 800 lM of As(III) in 1 M HCl and (B) calibration plot of inhibition current against [Na2S]. Other conditions are the same as in Fig. 1B.

An indirect approach by monitoring the As(III) signal in HCl medium at the PLA-AuNP/SPE was successfully demonstrated for detection of sulfide. Compared to the performance at a traditional macro-sized Au disk electrode, the

1374

Y.-S. Song et al. / Electrochemistry Communications 8 (2006) 1369–1374

PLA-AuNP/SPE has the advantage of unique surface regeneration property. Both XPS and Raman spectroscopic studies confirm that the formation of solution phase As2S3 is responsible for the inhibited oxidation current of As(III) in HCl medium. Such a performance characteristics along with high sensitivity compares favorably to the existing methods and shows the usefulness of the proposed method for detecting sulfide. Extended practical assay for acid rain and hot spring water concludes promising results and can be further applied into other real samples. Acknowledgements The authors gratefully acknowledge financial support from the National Science Council of Taiwan. Helpful suggestion from Dr. Chu-Chieh Lin and Dr. Shu-Hua Chien is also greatly appreciated. References [1] R.P. Pohanish, S.A. Greene, Hazardous Materials Handbook, Van Nostrandeinhold, New York, 1996. [2] P. Patnaik, A Comprehensive Guide to the Hazardous Properties of Chemical Substances, second ed., Wiley, New York, 1999. [3] R.G. Hendrickson, A. Chang, R.J. Hamilton, Am. J. Ind. Med. 45 (2004) 346. [4] S. Gao, K.K. Tanji, S.C. Scardaci, Agron. J. 96 (2004) 70. [5] N.S. Lawrence, J. Davis, R.G. Compton, Talanta 52 (2000) 771. [6] J.-R. Fernandez, J.M. Costa, R. Pereiro, A. Sanz-Medel, Anal. Chim. Acta 398 (1999) 23. [7] W. Puacz, W. Szahun, Analyst 120 (1995) 939. [8] M.A. Spaziani, J.L. Davis, M. Tinani, M.K. Carroll, Analyst 122 (1997) 1555. [9] N.S. Lawrence, P.D. Randhir, J. Wang, Anal. Chim. Acta 517 (2004) 131. [10] J.P. Hart, A.K. Abass, Anal. Chim. Acta 342 (1997) 199. [11] P. Jeroschewski, S. Steuchart, M. Kuhl, Anal. Chem. 68 (1996) 4351. [12] V. Stanic, T.H. Estell, A.C. Pierre, R.J. Mikula, Electrochim. Acta 43 (1998) 2639. [13] F. Hissner, J. Mattusch, K. Heining, J. Chromatogr. A 848 (1998) 503.

[14] N.S. Lawrence, L. Jiang, T.G.J. Jones, R.G. Compton, Anal. Chem. 75 (2003) 2054. [15] J. Wang, Acc. Chem. Res. 35 (2002) 811. [16] R.O. Lezna, N.R. De Tacconi, A.J. Arvia, J. Electroanal. Chem. 283 (1990) 319. [17] I.G. Casella, M.R. Guascito, E. Desimoni, Anal. Chim. Acta 409 (2000) 27. [18] R.D. Rocklin, E.L. Johnson, Anal. Chem. 55 (1983) 4. [19] G. Roman, A.Ch. Pappas, D.D. Kovala, M.I. Prodromidis, Anal. Chim. Acta 523 (2004) 201. [20] M. Thompson, N.S. Lawrence, J. Davis, L. Jiang, T.G.J. Jones, R.G. Compton, Sensor. Actuat. B 87 (2002) 33. [21] Y.-H. Tse, P. Janda, H. Lam, A.B.P. Lever, Anal. Chem. 67 (1995) 981. [22] M.I. Prodromidis, P.G. Veltsistas, M.I. Karayannis, Anal. Chem. 72 (2000) 3995. [23] D. Giovanelli, N.S. Lawrence, L. Jiang, T.G.J. Jones, R.G. Compton, Analyst 128 (2003) 173. [24] I.G. Casella, M. Contursi, E. Desimoni, Analyst 127 (2002) 647. [25] J.-M. Zen, P.-Y. Chen, A. Senthil Kumar, Electroanalysis 14 (2002) 513. [26] J.-M. Zen, J.-L. Chang, P.-Y. Chen, R. Ohara, K.-C. Pan, Electroanalysis 17 (2005) 739. [27] D.-M. Tsai, A. Senthil Kumar, J.-M. Zen, Anal. Chim. Acta 556 (2006) 145. [28] M.D. Lay, K. Varazo, J.L. Stickney, Langmuir 19 (2003) 8416. [29] E.A. Rochette, B.C. Bostick, G. Li, S. Fendorf, Environ. Sci. Technol. 34 (2000) 4714. [30] H.-Y. Chen, B.-H. Huang, C.-C. Lin, Macromolecules 38 (2005) 5400. [31] J.-M. Zen, M.-J. Chung, Anal. Chem. 67 (1995) 3571. [32] D. Myers, Surfactant Science and Technology, VCH, New York, 1988. [33] J. Berg, Wettability, Marcel Dekker, New York, 1993. [34] Y.-C. Sun, J. Mierzwa, M.-H. Yang, Talanta 44 (1997) 1379. [35] Z. Jia, A.O. Simm, X. Dai, R.G. Compton, J. Electroanal. Chem. 587 (2006) 247. [36] Y.-S. Song, G. Muthuraman, Y.-Z. Chen, C.-C. Lin, J.-M. Zen, Electroanalysis, in press. [37] S. Budaveri (Ed.), The Merck Index, 11th ed., 834, Merck & Co, USA, 1989, p. 127, Number 834. [38] A.I. Vogel, A Textbook of Macro- and Semimacro-qualitative Inorganic Analysis, Longman, London, 1965. [39] T.M. Loehr, R.A. Plane, Inorg. Chem. 7 (1968) 1708. [40] S.S.M. Hassan, S.A.M. Marzouk, H.E.M. Sayour, Anal. Chim. Acta 466 (2002) 47.