ANALYST, MAY 1992. VOL. 117 873 Determination of Trace Amounts of Copper With Extraction-Photoacoustic Spectrometry Y. Deng and M. Ye The Centre of Analysis and Testing, Wuhan University, Wuhan, Hubei 430072, People’s Republic of China A sensitive method for the determination of copper in aqueous solution by pulsed laser-induced photoacoustic spectrometry after extraction is described. The copper is chelated with 1,5-diphenylcarbazide in chloroform and extracted into the chloroform phase. The optimum extraction conditions were found experimentally. By using the stepwise dilution method, the minimum absorbance that could be measured was 8.5 x 10-6 at 6.2 mJ per pulse; this corresponds t o 6.9 x 10-3 pg I-’ of copper in chloroform. The relative standard deviation for ten measurements of an absorbance of 7.8 x 10-3 at 0.7 mJ is 1.5%.Application of the method t o the determination of copper in biological samples is also described. The results obtained were compared with those given by atomic absorption spectrometry. Keywords: Extraction; photoacoustic spectrometry; copper determination; 1,5-diphenylcarbazide The ultra-high sensitivity of photoacoustic (PA) measure- ments for liquid samples has been demonstrated.’-3 In recent years, the PA determination of trace amounts of several solutes, such as cobalt, holmium, neodymium, benzene and cyclohexane, has been reported, and detection limits of 1 x 10-7-1 x 10-8 cm-1 were reached.- The determination of 1 x 10-4 pg 1 - 1 of cobalt by extraction-PA spectrometry has also been described.’ Copper is a common element that occurs widely in organisms as a necessary trace element possessing specific physiological functions.The determination of trace amounts of copper in biological material is receiving increasing attention in nutrition, and in medicinal and physiological studies. 1,5-Diphenylcarbazide (DPC) is a reagent mainly used for the selective colorimetric determination of chromium(v1). Recently, the use of this reagent for the colorimetric determi- nation of copper in basic medium by extracting the metal into chloroform as the Cu-DPC chelate was reported.8 However, no detailed studies of the extraction conditions were carried out. This paper describes the determination of sub-nanomolar amounts of copper in aqueous solution by PA measurement after extraction of the metal with DPC into chloroform.It is also demonstrated that the proposed method can be used for the determination of copper in biological samples such as grass carp and pig kidney. Experimental Apparatus A schematic diagram of the PA measurement system used is shown in Fig. 1. The excitation beam was the frequency- doubled (532 nm) output of an Nd:YAG laser (YG 581, Quantel) operating at 10 Hz with a duration of 8 ns. The power incident into the PA cell was monitored by a laser energy/ power meter (LPE-lA, Chinese Academy of Sciences). The beam radius, determined by an aperture, was 2 mm, unless stated otherwise. The PA cell for liquid PA measurements was laboratory-built and is shown schematically in Fig. 2. It was constructed by using a fused-quartz cuvette and two PZT-SH discs, which were fixed separately onto the outside walls of the cell.In order to enhance the PA signal response, the PZT discs were connected in series. In some experiments, a post-cell mirror was used for back-reflecting the pump light so that it passed through the sample again along the incident direction [Fig. 2(6)]. The cell with the post-cell mirror will be referred to as cell A, and the cell without the post-cell mirror as cell B. The PA signal was amplified ten-fold with a preamplifier (M115, PAR), from which the output was fed into a boxcar averager (M162/165, PAR) with a 20 ps aperture and 28% delay. A dual-pen x-y recorder was used to record the PA and power signals synchronously. A spectrophotometer (UV-240, Shimadzu) with a 1 cm cuvette was used for the absorbance measurements.Reagents A stock solution of copper(i1) (1.0 mg ml-1 in 1 mol dm-3 HCI) was prepared from high-purity copper wire. Working 532 nm - Iris I -I- I J Recorder Boxcar Pre-amplifier Fig. 1 Schematic diagram of the PA measurement set-up Aluminium foil Cell POdy Laser beam I c Laser beam874 ANALYST, MAY 1992, VOL. 117 solutions of copper at the ng 1-1 level were prepared daily by appropriate dilution of the stock solution with water o r the desired buffer solution. Analytical-reagent grade DPC was purified by three re- crystallizations from absolute ethanol, after which white crystals of DPC were obtained. A stock solution of DPC in chloroform (0.1Y0) was prepared by dissolving 100 mg of DPC in 3 ml of hot absolute ethanol and diluting to 100 ml with chloroform.The solution was stored in a brown flask. The DPC extractant used was obtained by further diluting the stock solution with chloroform. All other chemicals used were of analytical-reagent grade. Doubly distilled water was used throughout. Prior to solvent extraction, the buffer solution was extracted three times with chloroform containing DPC and then washed three times with chloroform to remove all the extractable impurities. Procedure Under optimum extraction conditions, an aliquot of an aqueous solution containing 10-80 ng of copper was trans- ferred into a 60 ml separating funnel and buffer solution of the desired pH was added so that the final volume of the aqueous phase was 20 ml. The solution was then shaken vigorously with 5 ml of chloroform containing 0.02% DPC for 3 min, after which the mixture was allowed to stand for 20 min so that the two phases could separate.After extraction, an aliquot of the extract (about 1.4 ml) was placed directly in the PA cell from the funnel for PA measurement. In preparing the absorption curve and for carrying out the experiments into the effect of pH, a solution containing microgram levels of copper was extracted and the absorbance measurements were performed with a spectrophotometer. Treatment of the Samples A known amount of the fresh sample (pig kidney, grass carp) was dried at 80 "C in an air oven for 8 h and then ground into a powder. A 1 g amount of the dried powder was digested with a mixture of nitric acid and hydrogen peroxide in a flask.After the organic matter had been destroyed, the residue was dissolved in 50 ml of 0.1 mol dm-3 HCI, transferred into a 100 ml calibrated flask and diluted to the mark with water. The sample solutions thus prepared were used for the determina- tion of copper. Results and Discussion Extraction of Copper(r1) With DPC The reaction of copper(i1) with DPC in basic aqueous solution forms a brown chelate which can be extracted into chloroform; in addition, copper(i1) in basic aqueous solution is also directly extractable as the Cu-DPC chelate by using chloroform containing DPC. The extraction was performed in various buffer systems, such as phosphate, tris(hydroxymethy1)- methylamine-hydrochloric acid and ammonium chloride- ammonia. It was found that the ammonium chloride buffer was superior to the other buffers examined for the extraction of copper with chloroform containing DPC.The dependence of the absorbance of the extract on pH is shown in Fig. 3; a pH of 8.6 was chosen. The effect of varying the DPC concentration over the range 0.002-0.1% on the degree of extraction, for 80 ng of copper in 20 ml of ammonium chloride buffer solution with 5 ml of DPC solution in chloroform, was examined. The degree of extrac- tion was found to be constant; hence, a DPC concentration of 0.02% was used as the optimum. Similar studies revealed that a shaking time of 1 min was sufficient to achieve quantitative extraction. Hence, for extracting less than 80 ng of copper in an aqueous phase volume of 20 ml, the optimum extraction 0.8 r 1 a, C 0.71 m 0.4 1 I I I I I I I 7.8 8.0 8.2 8.4 8.6 8.8 9.0 9.2 PH Fig.3 Effect of pH on the absorbance of the extract. The extraction was carried out in 10 ml of 0.1 mol dm-3 NH3-NH4Cl buffer containing 4 pg of Cu" with 10 ml of 0.1% DPC in chloroform 1.2 1 I g 1.0 C $ 0.8 $ 0.6 0 0.4 0.2 0 400 450 500 550 600 650 700 Wavelengthlnm Fig. 4 Absorption spectra of the copper chelate and the DPC reagent blank. A, Cu-DPC chelate against reagent blank; and B, DPC in chloroform against solvent conditions are as follows: sufficient 0.01 mol dm-3 ammonium chloride buffer of pH 8.6 to bring the final volume of the aqueous phase to 20 ml; 5 ml of 0.02% DPC in chloroform; shaking time, 3 min. Interference experiments were carried out under the conditions described above.Except for cobalt and nickel, none of the elements examined, i.e., Fe"', Cr"', Mn" 7 7 Zn" Cd", Hg", Mg" and Cr"', produced an extractable coloured chelate. Chelate of Copper With DPC The absorption spectra of the Cu-DPC chelate and of the reagent blank in chloroform are shown in Fig. 4. The wavelength of the absorption peak of the chelate at 545 nm closely matches that of the pump beam (532 nm) used here. The molar absorptivity of the Cu-DPC chelate in chloroform was found to be 7.8 x 1041 mol-1 cm-1 at 532 nm, whereas the absorbance of the DPC extractant at the same wavelength was negligible. Hence, this provides a low blank value in the absorption measurements, which is important for highly sensitive pulsed PA measurements. However, DPC is easily oxidized by air to produce a pink compound, which causes the blank to increase.Therefore, it was noted that careful purification of the extractant to obtain a sufficiently constant and low background is a prerequisite for highly sensitive analysis with the pulsed PA method. The colour system, after extraction, was found to be stable for at least 2 h. However, a gradual decrease in the PA signal was observed during irradiation with the pulsed laser. Curve A in Fig. 5 represents the situation in which the concentration of the chelate is slightly high; the PA signal decreases exponen-ANALYST, MAY 1992, VOL. 117 100 r I 875 I I I I I 1 0 4 8 12 16 20 4.25 1 1 Tim e/m i n 4.00 U s 3 3.75 3.50 3.25 0 2 4 6 8 1 0 Time/min Fig. 5 ( a ) Stability curve for the extracted Cu-DPC chelate during irradiation with a pulsed laser.Pulsed energy: 6.2 mJ. A. Absorbance = 4.4 x 10-2; and B , absorbance = 1.9 X 10-4. ( b ) Variation of In(&*) with irradiation time for curve A Time - Fig. 6 chelate. For details, see text PA signal generated from pulsed excitation of the Cu-DPC tially with irradiation time at 6.2 mJ per pulse. However, no decay of the PA signal is apparent at a low concentration of the chelate during the same irradiation time (curve B). In practice, an irradiation time of 1-2 min is sufficient for a single PA measurement. The deviation in the detection caused by the decay of the PA signal due to laser light irradiation is negligible, particularly for a weak absorbance or a low laser energy irradiation. PA Measurement The pulsed PA waveform of the extract is shown in Fig.6. The magnitude of the peak marked with an asterisk is directly proportional to the absorbance of the extract at a constant incident radiation power. It was observed that the waveforms were the same with both cell A and cell B. However, there is an almost 50% enhancement of the magnitude of the PA signal for cell A compared with cell B. It is known that the magnitude of the PA signal is linearly dependent on the pulsed energy. In order to increase the sensitivity for trace analysis, the pulsed laser energy applied 0 20 40 60 80 100 120 Po we r/m W Dependence of the magnitude of the PA signal on the incident Fig. 7 laser power Table 1 Application of the proposed method to the determination of copper in biological samples and comparison of thc results obtaincd with those given by AAS Cu foundlpg g- 1 Proposed Sample" method? AAS Pig kidney 25.0 26.5 Grass carp 6.93 7.50 * Dried powder.t Average of three determinations. should be as high as possible. However, a high pulsed energy will always induce many non-linear effects such as photodissociation and breakdown. The optimum pulsed energy value is that which not only maintains high sensitivity for PA detection, but also allows Beer's law to be obeyed over the concentration range to be examined. The graph of PA signal intensity versus incident laser power is shown in Fig. 7 . By using cell A, the optimum energy value found experimentally was about 6.5 mJ per pulse in a 3 mm beam radius for an absorbance of 0.04.The minimum absorbance that could be measured, which was obtained by using a series of solutions prepared by stepwise dilution of the Cu-DPC chelate with chloroform, was 8.5 x 10-6 at 6.2 mJ per pulse. This corresponds to 6.9 x 10-3 pg 1-1 of copper in chloroform solution. The relative standard deviation for ten replicate measurements of a chloroform solution of the chelate with an absorbance of 7.8 x 10-3 was 1.5% at 0.7 mJ per pulse. Because of the high content of copper in the biological samples examined here, a calibration graph was constructed in the range 0.04-0.4 pg 1-1 of copper in aqueous solution. The detection limit, defined as a signal-to-noise ratio of 3 at 1.9 mJ per pulse, was 0.022 pg 1- in aqueous solution. Applications of the Method The proposed method was applied to the determination of trace amounts of copper in pig kidney and fish flesh (grass carp).The results obtained with the proposed method were compared with those given by atomic absorption spectrometry (AAS) and are shown in Table 1. The relative standard deviation for six replicate extraction- PA measurements of 0.4 ml of a prepared solution was 7.8% for the grass carp sample. The recovery of copper from a sample (fish) was found to be 98.5% by extraction and PA measurement.876 ANALYST, MAY 1992, VOL. 117 Conclusion A sensitive method for the determination of copper, based on extraction and PA spectrometry, has been developed. The combination of PA measurement and separation by extraction considerably improves the selectivity of the PA measurement; further, the use of an organic solvent enhances the magnitude of the PA signal. The proposed method allows a Cu-DPC extract with an absorbance of 8.5 X 10-6 to be detected, and can also be applied to the determination of copper in biological samples containing little or no cobalt and nickel. This project was supported by the National Natural Science Foundation of China. References 1 Lahmann, W., Ludewig, H . , and Welling, H., Anal. Chem.. 1977. 49, 549. 2 Oda. S., Sawada, T., and Kamada, H., Anal. Chem., 1978.50, 865. 3 Kitamori, T.. Fujii, M., Sawada. T., and Gohshi, Y., J. Appl. Phys., 1985, 58, 268. 4 Yan, H.. Deng. Y . , and Zeng, Y., Chem. J. Chin. Univ., 1988, 4 , 25. 5 Yan, H.. Deng, Y., and Zeng, Y . , Kexue Tongbao (Engl. Trunsl.), 1989, 34. 790. 6 Zuo, B., Deng, Y . , and Zeng, Y., Chem. J. Chin. Univ., 1990, 11, 15. 7 Kitamori, T., Suzuki, K., Sawada, T., Gohshi. Y.. and Motojima, K., Anal. Chem., 1986.58, 2275. 8 Huang, X., Fenxi Huaxue, 1990, 18, 304. Paper I I0451 4H Received August 29, I991 Accepted December 4, I991