- Email: [email protected]
Single-column method of ion chromatography for the determination of common cations and some transition metals
Journal of Chromatography A, 1118 (2006) 68–72
Single-column method of ion chromatography for the determination of common cations and some transition metals Wenfang Zeng a , Yongxin Chen a , Hairong Cui b , Feiyan Wu a , Yan Zhu a,∗ , James S. Fritz c a
Department of Chemistry, Xixi Campus of Zhejiang University, Hangzhou, Zhejiang 310028, China Technology Center of Hubei Entry-Exit Inspection and Quarantine Bureau, Wuhan 430022, China c Ames Laboratory, US Department of Energy and Department of Chemistry, Iowa State University, Ames, IA 5001, USA b
Available online 14 February 2006
Abstract A single-column method for the simultaneous determination of common cations and transition metals in real samples is proposed in this paper. Eleven cations (copper, lithium, sodium, ammonium, potassium, cobalt, nickel, magnesium, calcium, strontium and zinc) were separated and analyzed by means of ion chromatography using an isocratic elution with 2.5 mM methane sulfonic acid and 0.8 mM oxalic acid as mobile phase, IonPac SCS1 (250 mm × 4 mm I.D.) as the separation column and non-suppressed conductor detection. Optimized analytical conditions were further validated in terms of accuracy, precision and total uncertainty and the results showed the reliability of the IC method. The relative standard deviations (RSDs) of the retention time and peak area were less than 0.04 and 1.30%, respectively. The coefficients of determination for cations ranged from 0.9988 to 1.000. The method developed was successfully applied to determination of cations in samples of beer and bottled mineral water. The spiked recoveries for the cations were 94–106%. The method was applied to beer and beverage without interferences. © 2006 Elsevier B.V. All rights reserved. Keywords: Non-suppressed; Ion chromatography; Transition metals; Cations; Single-column
1. Introduction Due to a strong environmental impact, trace metal ion determination and speciation have received particular attention in the last years. The determination of trace metals in complex matrices remains one of the most complicated areas of analytical chemistry. The determination of alkali, alkaline earth and transition metals in beverage is of interest for quality control. High levels of metals are undesirable because of their known or supposed toxicity so that, for instance, a limit of 0.215 mg kg−1 for lead was proposed and actually 1 mg kg−1 is a limit in some countries [1]. Metals in beverage are of interest not only for quality control, but can be used also as an investigation for its taste. Numerous analytical techniques have been employed in the determination of heavy and transition metals. The most widely used instruments (e.g. atomic absorption and inductively cou-
∗
Corresponding author. Tel.: +86 571 88273637; fax: +86 571 88273637. E-mail address: [email protected] (Y. Zhu).
0021-9673/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2006.01.065
pled plasma atomic emission spectrometers) used for metal determination suffer from both spectral and chemical interferences and, in some cases, are unsuitable for direct trace analysis in complex matrices or for studies on metal speciation [2,3]. It is typically a necessity when using AAS or ICP-MS methods to remove sample matrix interferences and/or preconcentrate metals of interest using low efficiency large particle size chelating or ion exchange sorbents, often on-line prior to analysis, and many have been synthesized and evaluated for specific metals and applications, as illustrated in a number of fundamental reviews by Shaw and Haddad [4], Torre and Marina [5], Nickson et al. [6] and Garg et al. [7]. Chelation ion chromatography (CIC), using high performance stationary phases containing chelating functional groups, is a technique which has recently found wide application in the separation and detection of trace metal ions. However, most studies using CIC involve the preconcentration of trace metals on a pre-column packed with chelating ion exchanger, followed by another matrix isolation column and finally ion chromatographic determination using a standard high-performance anionor cation-exchange column, depending on the type of complexes
W. Zeng et al. / J. Chromatogr. A 1118 (2006) 68–72
formed in the eluent. Clearly, the equipment required is rather complex, involving a combination of several concentration and separation columns, pumps and switching valves under computer control. This significantly complicates the analysis and increases its cost [2,8–14]. So it is an important task to establish a rapid and convenient method for the simultaneous determination of these metals by IC. Some papers describe simultaneous separation of alkali, alkaline earth [15–22]. However, not many papers [2] describe the simultaneous determination of alkali, alkaline earth and some transition metals with conductivity detection using SCIC. This paper presents another simple IC method for the determination of 11 cations in one injection. It also describes the development and use of a simplified mode of non-suppressed IC as a tool for the analysis of beer samples with complex matrices. The retention characteristics of these metal ions on IonPac SCS1 (250 mm × 4 mm I.D.) column was investigated for a variety of eluents including methane sulfonic acid, citric acid, tartaric acid and oxalic acid. 2. Experimental 2.1. Chromatographic system All systems and components for ionic analysis were from Dionex (Sunnyvale, CA, USA). The hardware used for these analyses consisted of an ICS-2000 ion chromatograph equipped with an isocratic pump, a conductivity detector and a column heater. Eluent flow rates were set at 1 ml/min with an injection volume of 25 l. An IonPac SCG 1(50 mm × 4 mm I.D.) guard column and an IonPac SCS1 (250 mm × 4 mm I.D., 318 q/column, weak cation exchanger functionalized with carboxylic acid groups) analytical column were used to separate the cations. All measurements were made at 30 ◦ C. To display positive peaks, the “Conductivity Polarity” in the Detector screen should be set to “INVERTED”.
69
Table 1 Retention times (min) of metal ions in organic acid eluents Metals
Oxalic acida
Tartaric acid
Citric acid
Cu2+
4.15 6.42 8.11 9.09 11.96 13.76 16.45 17.82 22.65 28.38 34.253
>90 7.02 8.94 10.07 13.35 35.70 32.20 32.31 29.28 36.69 44.355
>90 7.00 8.91 10.03 14.30 35.50 31.80 31.94 29.09 36.37 44.02
Li+ Na+ NH4 + K+ Ni2+ Zn2+ Co2+ Mg2+ Ca2+ Sr2+ a
Eluent: 2.5 mM MSA + 0.8 mM organic acid.
3. Results and discussion 3.1. Selection of the eluent for the separation of metals At first, increasing MSA concentration decreased retention time of the metals. However, when its concentration was larger than 3.0 mM, the background conductance would be more than 920 us. This resulted in increased signal-to-noise ratios and unsatisfactory separation results. Thus, 2.5 mM was found to be appropriate. The elution order was lithium, sodium, ammonium, potassium, cobalt≈zinc, magnesium, nickel≈calcium and strontium. However, the peak of nickel tailed seriously. The use of organic acid in the eluent reduced the tailing and improved the peak shape considerably. The retention characteristics of three different organic acids were tested (Table 1), but only oxalic acid provided selectivity for the separation of transition metals. In the eluents of 2.5 mM MSA + 0.8 mM tartaric acid or 2.5 mM MSA + 0.8 mM citric acid, copper could not be eluted within 90 min and transition metals did not separated well.
2.2. Reagents and samples Deionised water, generated by a Milli-Q deionised water unit which had a resistance better than 18.2 M cm, was used for the preparation of all the solutions. All reagents were of analytical reagent grade unless otherwise specified. Working standard solutions were prepared daily by serial dilution of stock standard solution of each metal containing 1000 mg l−1 (National Research Center for Certified Reference Materials, Beijing, China). 2.3. Sample preparation Approximately 4 ml of sample A and sample B was added into 100 ml volumetric flask, then dilution to volume with deionized water. Bottled mineral water was injected without dilution. Sample solutions were filtered through a 0.45 m membrane filter before sample injection.
Fig. 1. Plot of measured retention times vs. added oxalic acid concentration in the mobile phase.
70
W. Zeng et al. / J. Chromatogr. A 1118 (2006) 68–72 Table 3 Linear range and detection limits data of metal cations Cations
Linear range (mg l−1 )
Coefficient of correlation (r2 )
Detection limita (S/N = 3, mg l−1 )
Cu2+ Li+ Na+ NH4 + K+ Ni2+ Zn2+ Co2+ Mg2+ Ca2+ Sr2+
0.8–60 0.04–10 0.4–40 0.16–16 0.4–40 0.8–60 0.8–60 0.8–60 0.4–40 0.8–60 0.8–60
0.9992 0.9990 0.9988 0.9998 0.9996 0.9999 0.9997 0.9998 0.9999 1.0000 0.9999
0.028 0.001 0.002 0.002 0.009 0.032 0.010 0.012 0.005 0.012 0.024
a
The data were obtained for standards in reagent water.
Fig. 2. Chromatogram of standard cations. Peaks (mg l−1 ): 1 = copper (1.6); 2 = lithium (0.08); 3 = sodium (0.8); 4 = ammonium (0.32); 5 = potassium (0.8); 6 = magnesium (0.8); 7 = zinc (1.6); 8 = cobalt (1.6); 9 = nickel (1.6); 10 = calcium (1.6); 11 = strontium (1.6).
The influence of oxalic acid on the chromatogram was also studied. Oxalic acid is not only a strong complexing agent but also a weak acid. Its concentration effectively influenced the retention times of 11 metals as shown in Fig. 1. It found that the retention time of Ni2+ , Zn2+ , Sr2+ , Ca2+ , Mg2+ and Co2+ decreased with the increasing concentration of oxalic acid. Fig. 1 also shows that oxalic acid slightly influenced the retention times of Li+ , NH4 + , Na+ and K+ . If the concentration of oxalic acid was larger than 1.0 mM, the copper peak would fall into the large negative peak of water. If the concentration was lower than 0.6 mM, zinc and cobalt would not separate from each other. The experiment showed that the optimum concentration for MSA and oxalic acid was 2.5 and 0.8 mM, respectively. A typical chromatogram of a synthetic standard solution is shown in Fig. 2. The elution order is Cu2+ , Li+ , Na+ , NH4 + , K+ , Ni2+ , Zn2+ , Co2+ , Mg2+ , Ca2+ and Sr2+ . All individual metal peaks are well separated.
Fig. 3. Chromatogram of beer A. Peaks: 1 = copper; 2 = sodium; 3 = ammonium; 4 = potassium; 5 = nickel; 6 = zinc; 7 = cobalt; 8 = magnesium; 9 = calcium; 10 = strontium.
The linearity and detection limits data obtained for standards in reagent water are summarized in Table 3. The linear calibration curves were obtained in each concentration range. All metals had good linearity with correlation coefficients that were greater than 0.9988. The detection limit (signal-to-noise ratio of 3:1) of this method was at the ppb level. The data confirmed that the precision of this method was good.
3.2. Method validation According to Mehta [23] requirements the selectivity of the proposed method is right because the peaks showed resolutions ≥1.5 for all the determined compounds. To check the precision of the method, eleven replicate analysis of a standard solution were performed. The relative standard deviations (RSDs) of the retention time and peak area were less than 0.04 and 1.30%, respectively, as shown in Table 2.
3.3. Sample determination Samples of four kinds of beers and two kinds of bottled mineral water were analyzed.
Table 2 RSD data of analyzing standard cations (n = 11) Metalsa
Cu2+
Li+
Na+
NH4 +
K+
Ni2+
Zn2+
Co2+
Mg2+
Ca2+
Sr2+
Retention time (%) Peak area (%) Peak height (%)
0.035 1.282 0.958
0.023 0.675 0.298
0.022 0.437 0.234
0.016 1.044 0.320
0.012 1.039 0.337
0.036 1.115 0.418
0.025 0.769 0.354
0.023 0.382 0.221
0.018 0.289 0.244
0.023 0.930 0.300
0.016 0.923 0.436
a
The concentration of the analytes (mg l−1 ): Cu2+ (4), Li+ (0.2), Na+ (2), NH4 + (0.8), K+ (2), Ni2+ (4), Zn2+ (4), Co2+ (4), Mg2+ (2), Ca2+ (4) and Sr2+ (4).
W. Zeng et al. / J. Chromatogr. A 1118 (2006) 68–72
71
Table 4 Concentrations and spiked recoveries of cations in beer A (25 times dilution) Metals
Cu2+
Li+
Na+
NH4 +
K+
Ni2+
Zn2+
Co2+
Mg2+
Ca2+
Sr2+
Concentration (mg l−1 ) detected RSD (%, n = 5) of unspiked sample Spiked (mg l−1 ) Found (mg l−1 ) Recovery (%)a RSD (%, n = 7) of spiked sample Detection limitc (S/N = 3, mg l−1 )
0.044 2.083 0.10 0.150 106 1.787 0.042
0.001 3.007 0.10 0.105 99 3.110 0.001
0.938 0.556 1.0 1.940 100.2 0.332 0.004
0.900 1.951 1.0 1.870 97 1.509 0.003
9.610 0.879 1.0 10.56 95 0.588 0.013
0.082 2.433 0.10 0.185 103 1.385 0.015
0.021 1.766 0.10 0.124 103 1.220 0.015
2.730 1.274 1.0 3.705 97.5 1.358 0.006
1.611 1.200 1.0 2.600 98.9 0.995 0.012
0.421 0.550 1.0 1.450 103 2.034 0.031
a b c
Spiked recoveries were obtained by adding standard solution to the sample, and then determined. Detection limit. The data was obtained for standards in diluted beer.
Table 5 Determination of metals in real samples Samples
Mineral water A Mineral water B Beer Bb Beer Cb Beer Db a b
Contents (mg l−1 ) Cu2+
Li+
Na+
NH4 +
K+
Ni2+
Zn2+
Co2+
Mg2+
Ca2+
Sr2+
0.041 0.020
62.14 70.30 2.32 2.28 3.58
29.19 32.88 10.33 11.23 9.86
12.79 13.63 2.57 2.64 2.55
41.44 42.72 1.20 1.31 1.14
0.52 0.37
Detection limit. Twenty-five times dilution.
All the samples were diluted with deionized water and directly injected into the IC system through a 0.45 m filter. The presence of these compounds was confirmed by comparing their retention times with those of standard solutions. Table 4 summarized the determination and reproducibility of metals (n = 5) in the real samples. The chromatograms of samples A and bottled mineral water are shown in Figs. 3 and 4, respectively. The data for these samples are summarized in Table 5. The concentration of Ni2+ in sample A was below the detection limit, and therefore, could not be detected. The concentrations of NH4 + , Ni2+ , Zn2+ and Co2+ in bottled mineral water were
below the detection limits and, therefore, could not be determined. Good recovery rates were obtained for all of the metals. The accuracy of the method was evaluated from recovery assays, preparing spiked samples of beer. The obtained values ranged from 95 to 106%. One of the advantages was that the analytical media could be injected directly after dilution. It seems this IC system would be suitable for practical use. 4. Conclusion The proposed techniques show satisfactory sensitivity, detection limits and standard deviation for alkaline, alkaline earth and some transition metals (copper, zinc, nickel, cobalt) determination in beer and mineral water. The determination of these eleven metals on an IonPac SCS1 column is described using ion chromatography with acidified aqueous mobile phases (2.5 mM MSA + 0.8 mM oxalic acid) and non-suppressed conductor detection. Ion chromatography of a mixture of these metals is optimized to give baseline resolution. Acknowledgements This is a project (No. 20375035) supported by National Natural Science Foundation of China and by Zhejiang Provincial Natural Science Foundation of China (No. Z404105). References
Fig. 4. Chromatogram of bottled mineral water A. Peaks: 1 = lithium; 2 = sodium; 3 = potassium; 4 = magnesium; 5 = calcium; 6 = strontium.
[1] P.L. Buldini, S. Cavalli, A. Mevoli, J.L. Sharma, Food chem. 73 (2001) 487.
72
W. Zeng et al. / J. Chromatogr. A 1118 (2006) 68–72
[2] P.N. Nesterenko, P. Jones, J. Chromatogr. A 770 (1997) 129. [3] M.R. Cave, O. Butler, R.N. Chenery, J.M. Cook, M.S. Cresser, D.L. Miles, J. Anal. At. Spectrom. 16 (2001) 194. [4] M.J. Shaw, P.R. Haddad, Env. Int. 30 (2004) 403. [5] M. Torre, M.L. Marina, Crit. Rev. Anal. Chem. 24 (1994) 327. [6] R.A. Nickson, S.J. Hill, P.J. Worsfold, Anal. Proc. 32 (1995) 387. [7] B.S. Garg, R.K. Sharma, N. Bhojak, S. Mittal, Microchem. J. 61 (1999) 94. [8] W. Shotyk, I. Immenhauser-Potthast, J. Chromatogr. A 706 (1995) 167. [9] X. Ding, S. Mou, K. Liu, Y. Yan, J. Chromatogr. A 883 (2000) 127. [10] X. Ding, S. Mou, K. Liu, A. Siriraks, J. Riviello, Anal. Chim. Acta 407 (2000) 319. [11] S. Motellier, H. Pitsch, J. Chromatogr. A 739 (1996) 119.
[12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
P. Jones, P.N. Nesterenko, J. Chromatogr. A 789 (1997) 413. P.N. Nesterenko, P. Jones, J. Chromatogr. A 804 (1998) 223. M.J. Shaw, P. Jones, P.N. Nesterenko, J. Chromatogr. A 953 (2002) 141. C. Sarzanini, J. Chromatogr. A 850 (1999) 213. G.J. Sevenich, J.S. Fritz, J. Chromatogr. 347 (1985) 147. H. Saitoh, K. Oikawa, J. Chromatogr. 329 (1985) 247. L.M. Nair, R. Saari-Nordhaus, J.M. Anderson, J. Chromatogr. A 671 (1994) 43. Y. Zhu, Y. Chen, M. Ye, J.S. Fritz, J. Chromatogr. A 1085 (2005) 18. K. Ohta, K. Tanaka, B. Paull, P.R. Haddad, J. Chromatogr. A 770 (1997) 219. S. Kwon, K. Lee, K. Tanaka, K. Ohta, J. Chromatogr. A 850 (1999) 79. J. Morris, J.S. Fritz, J. Chromatogr. A 602 (1992) 111. A.C. Mehta, Analyst 122 (1997) 83R.
Single-column method of ion chromatography for the determination of common cations and some transition metals Wenfang Zeng a , Yongxin Chen a , Hairong Cui b , Feiyan Wu a , Yan Zhu a,∗ , James S. Fritz c a
Department of Chemistry, Xixi Campus of Zhejiang University, Hangzhou, Zhejiang 310028, China Technology Center of Hubei Entry-Exit Inspection and Quarantine Bureau, Wuhan 430022, China c Ames Laboratory, US Department of Energy and Department of Chemistry, Iowa State University, Ames, IA 5001, USA b
Available online 14 February 2006
Abstract A single-column method for the simultaneous determination of common cations and transition metals in real samples is proposed in this paper. Eleven cations (copper, lithium, sodium, ammonium, potassium, cobalt, nickel, magnesium, calcium, strontium and zinc) were separated and analyzed by means of ion chromatography using an isocratic elution with 2.5 mM methane sulfonic acid and 0.8 mM oxalic acid as mobile phase, IonPac SCS1 (250 mm × 4 mm I.D.) as the separation column and non-suppressed conductor detection. Optimized analytical conditions were further validated in terms of accuracy, precision and total uncertainty and the results showed the reliability of the IC method. The relative standard deviations (RSDs) of the retention time and peak area were less than 0.04 and 1.30%, respectively. The coefficients of determination for cations ranged from 0.9988 to 1.000. The method developed was successfully applied to determination of cations in samples of beer and bottled mineral water. The spiked recoveries for the cations were 94–106%. The method was applied to beer and beverage without interferences. © 2006 Elsevier B.V. All rights reserved. Keywords: Non-suppressed; Ion chromatography; Transition metals; Cations; Single-column
1. Introduction Due to a strong environmental impact, trace metal ion determination and speciation have received particular attention in the last years. The determination of trace metals in complex matrices remains one of the most complicated areas of analytical chemistry. The determination of alkali, alkaline earth and transition metals in beverage is of interest for quality control. High levels of metals are undesirable because of their known or supposed toxicity so that, for instance, a limit of 0.215 mg kg−1 for lead was proposed and actually 1 mg kg−1 is a limit in some countries [1]. Metals in beverage are of interest not only for quality control, but can be used also as an investigation for its taste. Numerous analytical techniques have been employed in the determination of heavy and transition metals. The most widely used instruments (e.g. atomic absorption and inductively cou-
∗
Corresponding author. Tel.: +86 571 88273637; fax: +86 571 88273637. E-mail address: [email protected] (Y. Zhu).
0021-9673/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2006.01.065
pled plasma atomic emission spectrometers) used for metal determination suffer from both spectral and chemical interferences and, in some cases, are unsuitable for direct trace analysis in complex matrices or for studies on metal speciation [2,3]. It is typically a necessity when using AAS or ICP-MS methods to remove sample matrix interferences and/or preconcentrate metals of interest using low efficiency large particle size chelating or ion exchange sorbents, often on-line prior to analysis, and many have been synthesized and evaluated for specific metals and applications, as illustrated in a number of fundamental reviews by Shaw and Haddad [4], Torre and Marina [5], Nickson et al. [6] and Garg et al. [7]. Chelation ion chromatography (CIC), using high performance stationary phases containing chelating functional groups, is a technique which has recently found wide application in the separation and detection of trace metal ions. However, most studies using CIC involve the preconcentration of trace metals on a pre-column packed with chelating ion exchanger, followed by another matrix isolation column and finally ion chromatographic determination using a standard high-performance anionor cation-exchange column, depending on the type of complexes
W. Zeng et al. / J. Chromatogr. A 1118 (2006) 68–72
formed in the eluent. Clearly, the equipment required is rather complex, involving a combination of several concentration and separation columns, pumps and switching valves under computer control. This significantly complicates the analysis and increases its cost [2,8–14]. So it is an important task to establish a rapid and convenient method for the simultaneous determination of these metals by IC. Some papers describe simultaneous separation of alkali, alkaline earth [15–22]. However, not many papers [2] describe the simultaneous determination of alkali, alkaline earth and some transition metals with conductivity detection using SCIC. This paper presents another simple IC method for the determination of 11 cations in one injection. It also describes the development and use of a simplified mode of non-suppressed IC as a tool for the analysis of beer samples with complex matrices. The retention characteristics of these metal ions on IonPac SCS1 (250 mm × 4 mm I.D.) column was investigated for a variety of eluents including methane sulfonic acid, citric acid, tartaric acid and oxalic acid. 2. Experimental 2.1. Chromatographic system All systems and components for ionic analysis were from Dionex (Sunnyvale, CA, USA). The hardware used for these analyses consisted of an ICS-2000 ion chromatograph equipped with an isocratic pump, a conductivity detector and a column heater. Eluent flow rates were set at 1 ml/min with an injection volume of 25 l. An IonPac SCG 1(50 mm × 4 mm I.D.) guard column and an IonPac SCS1 (250 mm × 4 mm I.D., 318 q/column, weak cation exchanger functionalized with carboxylic acid groups) analytical column were used to separate the cations. All measurements were made at 30 ◦ C. To display positive peaks, the “Conductivity Polarity” in the Detector screen should be set to “INVERTED”.
69
Table 1 Retention times (min) of metal ions in organic acid eluents Metals
Oxalic acida
Tartaric acid
Citric acid
Cu2+
4.15 6.42 8.11 9.09 11.96 13.76 16.45 17.82 22.65 28.38 34.253
>90 7.02 8.94 10.07 13.35 35.70 32.20 32.31 29.28 36.69 44.355
>90 7.00 8.91 10.03 14.30 35.50 31.80 31.94 29.09 36.37 44.02
Li+ Na+ NH4 + K+ Ni2+ Zn2+ Co2+ Mg2+ Ca2+ Sr2+ a
Eluent: 2.5 mM MSA + 0.8 mM organic acid.
3. Results and discussion 3.1. Selection of the eluent for the separation of metals At first, increasing MSA concentration decreased retention time of the metals. However, when its concentration was larger than 3.0 mM, the background conductance would be more than 920 us. This resulted in increased signal-to-noise ratios and unsatisfactory separation results. Thus, 2.5 mM was found to be appropriate. The elution order was lithium, sodium, ammonium, potassium, cobalt≈zinc, magnesium, nickel≈calcium and strontium. However, the peak of nickel tailed seriously. The use of organic acid in the eluent reduced the tailing and improved the peak shape considerably. The retention characteristics of three different organic acids were tested (Table 1), but only oxalic acid provided selectivity for the separation of transition metals. In the eluents of 2.5 mM MSA + 0.8 mM tartaric acid or 2.5 mM MSA + 0.8 mM citric acid, copper could not be eluted within 90 min and transition metals did not separated well.
2.2. Reagents and samples Deionised water, generated by a Milli-Q deionised water unit which had a resistance better than 18.2 M cm, was used for the preparation of all the solutions. All reagents were of analytical reagent grade unless otherwise specified. Working standard solutions were prepared daily by serial dilution of stock standard solution of each metal containing 1000 mg l−1 (National Research Center for Certified Reference Materials, Beijing, China). 2.3. Sample preparation Approximately 4 ml of sample A and sample B was added into 100 ml volumetric flask, then dilution to volume with deionized water. Bottled mineral water was injected without dilution. Sample solutions were filtered through a 0.45 m membrane filter before sample injection.
Fig. 1. Plot of measured retention times vs. added oxalic acid concentration in the mobile phase.
70
W. Zeng et al. / J. Chromatogr. A 1118 (2006) 68–72 Table 3 Linear range and detection limits data of metal cations Cations
Linear range (mg l−1 )
Coefficient of correlation (r2 )
Detection limita (S/N = 3, mg l−1 )
Cu2+ Li+ Na+ NH4 + K+ Ni2+ Zn2+ Co2+ Mg2+ Ca2+ Sr2+
0.8–60 0.04–10 0.4–40 0.16–16 0.4–40 0.8–60 0.8–60 0.8–60 0.4–40 0.8–60 0.8–60
0.9992 0.9990 0.9988 0.9998 0.9996 0.9999 0.9997 0.9998 0.9999 1.0000 0.9999
0.028 0.001 0.002 0.002 0.009 0.032 0.010 0.012 0.005 0.012 0.024
a
The data were obtained for standards in reagent water.
Fig. 2. Chromatogram of standard cations. Peaks (mg l−1 ): 1 = copper (1.6); 2 = lithium (0.08); 3 = sodium (0.8); 4 = ammonium (0.32); 5 = potassium (0.8); 6 = magnesium (0.8); 7 = zinc (1.6); 8 = cobalt (1.6); 9 = nickel (1.6); 10 = calcium (1.6); 11 = strontium (1.6).
The influence of oxalic acid on the chromatogram was also studied. Oxalic acid is not only a strong complexing agent but also a weak acid. Its concentration effectively influenced the retention times of 11 metals as shown in Fig. 1. It found that the retention time of Ni2+ , Zn2+ , Sr2+ , Ca2+ , Mg2+ and Co2+ decreased with the increasing concentration of oxalic acid. Fig. 1 also shows that oxalic acid slightly influenced the retention times of Li+ , NH4 + , Na+ and K+ . If the concentration of oxalic acid was larger than 1.0 mM, the copper peak would fall into the large negative peak of water. If the concentration was lower than 0.6 mM, zinc and cobalt would not separate from each other. The experiment showed that the optimum concentration for MSA and oxalic acid was 2.5 and 0.8 mM, respectively. A typical chromatogram of a synthetic standard solution is shown in Fig. 2. The elution order is Cu2+ , Li+ , Na+ , NH4 + , K+ , Ni2+ , Zn2+ , Co2+ , Mg2+ , Ca2+ and Sr2+ . All individual metal peaks are well separated.
Fig. 3. Chromatogram of beer A. Peaks: 1 = copper; 2 = sodium; 3 = ammonium; 4 = potassium; 5 = nickel; 6 = zinc; 7 = cobalt; 8 = magnesium; 9 = calcium; 10 = strontium.
The linearity and detection limits data obtained for standards in reagent water are summarized in Table 3. The linear calibration curves were obtained in each concentration range. All metals had good linearity with correlation coefficients that were greater than 0.9988. The detection limit (signal-to-noise ratio of 3:1) of this method was at the ppb level. The data confirmed that the precision of this method was good.
3.2. Method validation According to Mehta [23] requirements the selectivity of the proposed method is right because the peaks showed resolutions ≥1.5 for all the determined compounds. To check the precision of the method, eleven replicate analysis of a standard solution were performed. The relative standard deviations (RSDs) of the retention time and peak area were less than 0.04 and 1.30%, respectively, as shown in Table 2.
3.3. Sample determination Samples of four kinds of beers and two kinds of bottled mineral water were analyzed.
Table 2 RSD data of analyzing standard cations (n = 11) Metalsa
Cu2+
Li+
Na+
NH4 +
K+
Ni2+
Zn2+
Co2+
Mg2+
Ca2+
Sr2+
Retention time (%) Peak area (%) Peak height (%)
0.035 1.282 0.958
0.023 0.675 0.298
0.022 0.437 0.234
0.016 1.044 0.320
0.012 1.039 0.337
0.036 1.115 0.418
0.025 0.769 0.354
0.023 0.382 0.221
0.018 0.289 0.244
0.023 0.930 0.300
0.016 0.923 0.436
a
The concentration of the analytes (mg l−1 ): Cu2+ (4), Li+ (0.2), Na+ (2), NH4 + (0.8), K+ (2), Ni2+ (4), Zn2+ (4), Co2+ (4), Mg2+ (2), Ca2+ (4) and Sr2+ (4).
W. Zeng et al. / J. Chromatogr. A 1118 (2006) 68–72
71
Table 4 Concentrations and spiked recoveries of cations in beer A (25 times dilution) Metals
Cu2+
Li+
Na+
NH4 +
K+
Ni2+
Zn2+
Co2+
Mg2+
Ca2+
Sr2+
Concentration (mg l−1 ) detected RSD (%, n = 5) of unspiked sample Spiked (mg l−1 ) Found (mg l−1 ) Recovery (%)a RSD (%, n = 7) of spiked sample Detection limitc (S/N = 3, mg l−1 )
0.044 2.083 0.10 0.150 106 1.787 0.042
0.001 3.007 0.10 0.105 99 3.110 0.001
0.938 0.556 1.0 1.940 100.2 0.332 0.004
0.900 1.951 1.0 1.870 97 1.509 0.003
9.610 0.879 1.0 10.56 95 0.588 0.013
0.082 2.433 0.10 0.185 103 1.385 0.015
0.021 1.766 0.10 0.124 103 1.220 0.015
2.730 1.274 1.0 3.705 97.5 1.358 0.006
1.611 1.200 1.0 2.600 98.9 0.995 0.012
0.421 0.550 1.0 1.450 103 2.034 0.031
a b c
Spiked recoveries were obtained by adding standard solution to the sample, and then determined. Detection limit. The data was obtained for standards in diluted beer.
Table 5 Determination of metals in real samples Samples
Mineral water A Mineral water B Beer Bb Beer Cb Beer Db a b
Contents (mg l−1 ) Cu2+
Li+
Na+
NH4 +
K+
Ni2+
Zn2+
Co2+
Mg2+
Ca2+
Sr2+
0.041 0.020
62.14 70.30 2.32 2.28 3.58
29.19 32.88 10.33 11.23 9.86
12.79 13.63 2.57 2.64 2.55
41.44 42.72 1.20 1.31 1.14
0.52 0.37
Detection limit. Twenty-five times dilution.
All the samples were diluted with deionized water and directly injected into the IC system through a 0.45 m filter. The presence of these compounds was confirmed by comparing their retention times with those of standard solutions. Table 4 summarized the determination and reproducibility of metals (n = 5) in the real samples. The chromatograms of samples A and bottled mineral water are shown in Figs. 3 and 4, respectively. The data for these samples are summarized in Table 5. The concentration of Ni2+ in sample A was below the detection limit, and therefore, could not be detected. The concentrations of NH4 + , Ni2+ , Zn2+ and Co2+ in bottled mineral water were
below the detection limits and, therefore, could not be determined. Good recovery rates were obtained for all of the metals. The accuracy of the method was evaluated from recovery assays, preparing spiked samples of beer. The obtained values ranged from 95 to 106%. One of the advantages was that the analytical media could be injected directly after dilution. It seems this IC system would be suitable for practical use. 4. Conclusion The proposed techniques show satisfactory sensitivity, detection limits and standard deviation for alkaline, alkaline earth and some transition metals (copper, zinc, nickel, cobalt) determination in beer and mineral water. The determination of these eleven metals on an IonPac SCS1 column is described using ion chromatography with acidified aqueous mobile phases (2.5 mM MSA + 0.8 mM oxalic acid) and non-suppressed conductor detection. Ion chromatography of a mixture of these metals is optimized to give baseline resolution. Acknowledgements This is a project (No. 20375035) supported by National Natural Science Foundation of China and by Zhejiang Provincial Natural Science Foundation of China (No. Z404105). References
Fig. 4. Chromatogram of bottled mineral water A. Peaks: 1 = lithium; 2 = sodium; 3 = potassium; 4 = magnesium; 5 = calcium; 6 = strontium.
[1] P.L. Buldini, S. Cavalli, A. Mevoli, J.L. Sharma, Food chem. 73 (2001) 487.
72
W. Zeng et al. / J. Chromatogr. A 1118 (2006) 68–72
[2] P.N. Nesterenko, P. Jones, J. Chromatogr. A 770 (1997) 129. [3] M.R. Cave, O. Butler, R.N. Chenery, J.M. Cook, M.S. Cresser, D.L. Miles, J. Anal. At. Spectrom. 16 (2001) 194. [4] M.J. Shaw, P.R. Haddad, Env. Int. 30 (2004) 403. [5] M. Torre, M.L. Marina, Crit. Rev. Anal. Chem. 24 (1994) 327. [6] R.A. Nickson, S.J. Hill, P.J. Worsfold, Anal. Proc. 32 (1995) 387. [7] B.S. Garg, R.K. Sharma, N. Bhojak, S. Mittal, Microchem. J. 61 (1999) 94. [8] W. Shotyk, I. Immenhauser-Potthast, J. Chromatogr. A 706 (1995) 167. [9] X. Ding, S. Mou, K. Liu, Y. Yan, J. Chromatogr. A 883 (2000) 127. [10] X. Ding, S. Mou, K. Liu, A. Siriraks, J. Riviello, Anal. Chim. Acta 407 (2000) 319. [11] S. Motellier, H. Pitsch, J. Chromatogr. A 739 (1996) 119.
[12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
P. Jones, P.N. Nesterenko, J. Chromatogr. A 789 (1997) 413. P.N. Nesterenko, P. Jones, J. Chromatogr. A 804 (1998) 223. M.J. Shaw, P. Jones, P.N. Nesterenko, J. Chromatogr. A 953 (2002) 141. C. Sarzanini, J. Chromatogr. A 850 (1999) 213. G.J. Sevenich, J.S. Fritz, J. Chromatogr. 347 (1985) 147. H. Saitoh, K. Oikawa, J. Chromatogr. 329 (1985) 247. L.M. Nair, R. Saari-Nordhaus, J.M. Anderson, J. Chromatogr. A 671 (1994) 43. Y. Zhu, Y. Chen, M. Ye, J.S. Fritz, J. Chromatogr. A 1085 (2005) 18. K. Ohta, K. Tanaka, B. Paull, P.R. Haddad, J. Chromatogr. A 770 (1997) 219. S. Kwon, K. Lee, K. Tanaka, K. Ohta, J. Chromatogr. A 850 (1999) 79. J. Morris, J.S. Fritz, J. Chromatogr. A 602 (1992) 111. A.C. Mehta, Analyst 122 (1997) 83R.