- Email: [email protected]
Direct determination of trace amounts of sodium in water-soluble organic pharmaceuticals by microwave induced plasma atomic emission spectrometry
Talanta 54 (2001) 855– 862 www.elsevier.com/locate/talanta
Direct determination of trace amounts of sodium in water-soluble organic pharmaceuticals by microwave induced plasma atomic emission spectrometry Krzysztof Jankowski * Department of Analytical Chemistry, Faculty of Chemistry, Warsaw Uni6ersity of Technology, ul. Naokowskiego 3, 00 -664 Warsaw, Poland Received 19 October 2000; received in revised form 10 January 2001; accepted 24 January 2001
Abstract The direct determination of trace sodium by microwave induced plasma atomic emission spectrometry (MIP-AES) in water-soluble organic substances utilized in pharmaceutical preparations was developed. No decomposition of the organic constituents was required. Samples were dissolved with water and introduced to the plasma after ultrasonic nebulization without desolvation. A limit of detection (3s) of 0.91– 3.0 ng ml − 1 was obtained under experimental conditions. The quantitative MIP-AES procedure involved the standard addition method. The sodium content determined in reference material NIST SRM 1568A Rice Flour agreed with the certified value (6.6 90.8 mg g − 1). Physical and chemical interferences were investigated. It was found for the microwave plasma that it is possible to introduce organic substances solutions of concentration up to 5% without sensitivity losses. This direct technique is fast and sensitive and helps to reduce contamination connected with the sample preparation procedure. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Atomic emission spectrometry; Microwave induced plasma; Water soluble pharmaceuticals; Sodium
1. Introduction There is a limited number of analytical methods, including potentiometry with ISE, spectrophotometry, atomic absorption and atomic emission spectrometry, sufficiently sensitive and selective to determine traces of sodium in pharmaceutical [1–3] and clinical [4 – 6] materials. The utilization of * Fax: + 48-2-6283339. E-mail address: [email protected] (K. Jankowski).
sample decomposition techniques before sodium determination in organic material is a common practice due to severe matrix interferences. However, in the case of determining sodium traces the problem of sample contamination during its preparation is very serious, since this element is present practically in every chemical reactant. Regardless of matrix effects the determination of sodium is often plagued by interferences from other metals, mainly alkali metals, especially in spectrophotometric and potentiometric methods.
0039-9140/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 3 9 - 9 1 4 0 ( 0 1 ) 0 0 3 3 3 - 2
856
K. Jankowski / Talanta 54 (2001) 855–862
Microwave induced plasma atomic emission spectrometry (MIP-AES) is a promising analytical method for the determination of alkali metals. Among plasma sources widely used in various spectroscopic techniques, MIP is much sensitive toward sodium determination [7]. When compared with flame atomic emission as well as absorption spectroscopy, which are the most sensitive methods for sodium, MIP-AES provides somewhat poorer detectability (LOD about 1 ng g − 1). However, a small number of reports can be found regarding the analysis of sodium by MIPAES in real samples. In a low-power MIP Long and Perkins achieved at 589 nm LODs for sodium of 1 and 2 ng ml − 1 for helium [8] and argon plasma [9], respectively. In the air MIP at operating power of 500 W the limit of detection of 82 ng ml − 1 was obtained by Urh and Carnahan [10]. When microsamples of aqueous solutions were nebulized and excited in He-MIP by Zander and Hieftje [11], the absolute detection limit for sodium of 0.12 pg for a 1 ml sample aliquot was determined. The sample introduction to Ar-MIP after vaporization by means of a microarc permits to achieve an absolute detection limit of 0.01 pg for a 10 ml sample aliquot [12]. Hiddemann et al. [13] determined sodium in high purity quartz by laser ablation-MIP-AES and obtained LOD of 35 ng g − 1. Haas and Jamerson [14] studied the possibility of alkali metals determination by argon MIP-AES and ICP-AES using the same spectrometer. With MIP the LOD for sodium of 1.3 ng ml − 1 at 589 nm and 23 ng ml − 1 at 330 nm were obtained, while these values for ICP with radial viewing were 6.4 and 110 times higher, respectively. Jin et al. [15] reported for Ar-MPT with radial viewing the LOD for sodium of 23 ng ml − 1 at 589 nm and 1.4 ng ml − 1 if oxygen sheating is utilized. High sodium contents were determined in our laboratory in the CRMs natural waters (30– 115 mg kg − 1) [16] and in pine needles (ca. 40 mg g − 1) [17]; however, the LOD calculated was as low as 1 ng ml − 1. Layman and Hieftje [12] determined ca. 300 mg ml − 1 of sodium in blood plasma after 1000-fold sample dilution. The low tolerance of MIP toward the introduction of even a small amount of sample is often argued as a main drawback of this type of
plasma. The plasma discharge may be easily extinguished especially in the presence of organics. However, Ng and Culp [18] introduced methanolic-aqueous solutions (1:1 vol) to MIP using a direct injection nebulizer. Long and Perkins [8] studied the nebulization conditions of organic solvents. Applying xylene flow of 0.46 ml min − 1 they introduced the aerosol formed to MIP without plasma extinction. Madrid et al. [19] observed that their MPT system can handle large amounts of ethanol. Matusiewicz and Sturgeon [20], when studying slurry nebulization for MIP-AES used an organic material dispersed in nitric acid solution. The determination of trace element impurities in pharmaceutical materials is of great importance because of the high product quality requirements [21–24]. Some elements concentrations provide fingerprinting of pharmaceutical products originating from different batches or manufacturers [25]. The presence of sodium may affect the basic properties of a product, i.e. solubility or pH of its aqueous solution. Flame emission or atomic absorption spectrometry are commonly used for sodium determination in pharmaceuticals. The analysis of water-soluble organic samples as their aqueous solutions is of interest in order to simplify the analytical procedure [26]. Arpadjan and Alexandrova [27] utilized ETAAS for the analysis of water-soluble pharmaceuticals. Niebergall and Wennrich determined metal impurities in the penicillin G aqueous solution by the ICP-AES method [28]. In this work the method of direct determination of sodium in water-soluble pharmaceutical materials by MIP-AES is employed. Samples are dissolved in high purity water, nebulized with an ultrasonic nebulizer and introduced to the plasma as a wet aerosol without desolvation. The speed of this method and limitation of the possibility of contamination due to not using chemical reactants for sample preparation are its advantage. The validity of the method was evaluated by analyzing a certified material, NIST SRM 1568A Rice Flour. Microwave digestion was utilized for SRM and sample pretreatment and the results obtained were compared with those obtained for the direct determination.
K. Jankowski / Talanta 54 (2001) 855–862
2. Experimental
2.1. Apparatus and measurement conditions The sodium measurements were made by a MIP-AES spectrometer MIP 750MV (Analab Ltd., Warsaw, Poland) with a TE101 rectangular cavity (Plazmatronika-Service, Wrocław, Poland) and ultrasonic nebulizer (laboratory-made). The vertically positioned aerosol cooled plasma system was described earlier [17]. The monochromator (0.75 m Czerny-Turner, grating 1800 lines per mm) with two Hamamatsu photomultipliers (UV and VIS) was used for spectroscopic measurements. The sodium emission measurements were performed at a wavelength of 588 995 nm, which is characterized by the largest sensitivity of measurement and stable and low background. The plasma torch was made of quartz (Herasil®, Kleinostheim, Germany). The plasma conditions and measurement parameters used are listed in Table 1. A microwave digestion unit UniClever™ BM1z (Plazmatronika-Service, Wrocław, Poland) opTable 1 Instrumental parameters for MIP-AES system Microwave frequency/MHz Microwave power/W Plasma torch Plasma viewing mode Plasma gas Gas flow rate (ml min−1) Sample flow rate (ml min−1) Integration time (s) Analytical wavelength (nm)
2450 150 Quartz, 2.5 mm i.d. Axial Argon, 99.998% 150 30 0.5 NaI 588 995
Table 2 Operating conditions for digestion system Pressure level (atm) Maximum power (W) Pressure level (atm) PTFE vessel volume (ml) Nitric acid (ml) Program
100% power (15 min) 300 42–44 110 3 30% power (5 min), 100% power (15 min)
857
erating in the pressure-control mode was utilized for the decomposition of the reference material NIST SRM 1568A Rice Flour and selected samples.
2.2. Chemicals and standards All chemicals used were manufactured by Merck (Darmstadt, Germany). Standard and sample solutions were prepared with chromatographic grade water Ultrapur® containing maximum 0.5 ng ml − 1 of sodium. Standard sodium solution 1 mg ml − 1 was prepared from analyticalgrade sodium chloride Suprapur® dried at 400°C. For microwave digestion 60% nitric acid Ultrapur® was used.
2.3. Sample preparation for direct determination A 0.200 g mass of sample was dissolved in 10.0 ml of water in a volumetric flask. In standard addition measurements an aliquot of 10 mg ml − 1 sodium solution (containing 3–10 mg of Na) was added.
2.4. Microwa6e digestion A dry sample (containing above 90% of organic matter) or the certified reference material NIST SRM 1568A Rice Flour (0.2–0.5 g depending on sodium content) was weighed into a 110 ml PTFE vessel and 3 ml of HNO3 was added. The vessel was covered, and the sample was heated according to the program presented in Table 2. After cooling, the vessel was opened and excess of nitric acid was evaporated to about 0.5 ml. The solution was transferred to a 10.0 ml volumetric flask, an aliquot of the sodium standard solution was added if necessary, and diluted to the mark with water.
3. Results and discussion
3.1. Optimization of analytical conditions Studies on the effect of the microwave power on the intensity of sodium emission at 588 995 nm
858
K. Jankowski / Talanta 54 (2001) 855–862
Fig. 1. The effect of microwave power on signal-to-background ratio (SBR) for sodium solution at 1 mg ml − 1.
Fig. 2. The effect of argon flow rate on signal-to-background ratio (SBR) for sodium solution at 1 mg ml − 1.
were carried out for a solution of 1 mg ml − 1 concentration over the 70– 200 W range (Fig. 1). The signal and background levels increase slowly with an increase of microwave power, however, the best signal stability was obtained for about 150 W. The influence of the argon flow rate on emission intensity is shown in Fig. 2. The highest signal-tobackground ratio was obtained at about 150 ml min − 1 of argon.
Under selected optimum operating conditions the emission intensity measurements for sodium chloride as well as sodium versenate solutions (sodium concentration range, 1–30 mg ml − 1) were performed due to select a suitable substance for standard addition procedure. As expected, no difference in the measurement sensitivity for both substances was observed and NaCl was applied in further studies, since it is easier to prepare and use.
K. Jankowski / Talanta 54 (2001) 855–862
The effect of the matrix concentration in the sample solution on the nebulization conditions and sodium excitation has been studied. The mechanical efficiency of nebulization was measured by the weighing method as described earlier [29]. The mass of aerosol introduced to the plasma was determined by nebulizing a known mass of liquid and weighing the liquid remaining in the drain and in the spray chamber. The effect of the sample flow rate on the aerosol transport efficiency for water, 5%w/v potassium chloride and 5%w/v tartaric acid (TA) solution is presented in Fig. 3. The efficiencies obtained for solutions are slightly lower owing to the higher specific gravity. The effect of TA concentration on relative emission intensity of sodium and its signal stability is shown in Fig. 4. The presence of the matrix does not affect essentially the level of the sodium signal up to the concentration of ca. 5%w/v. However, at this concentration the signal stability worsens (4.1% R.S.D.). The effect of the argon flow rate in the range 100– 250 ml min − 1 on the intensity measured was similar in shape to those presented in Fig. 2 up to the matrix concentration of ca. 3.5%w/v.
859
Niebergall and Wennrich [28] introduced to ICP a 15% solution of penicillin G and Krasil’shchik et al. [30] 10% solutions of organic acids in order to determine metal impurities. However they utilized up to a 10-times higher power level for plasma sustaining and introduced lower aerosol concentrations since a pneumatic nebulization has been used.
3.2. Analytical performances The linear dynamic range and detection limit for sodium were determined under operating conditions listed in Table 1. Standard sodium solutions (0.005–200 mg ml − 1) were used to measure the emission intensities. The linear response was obtained in the range from 10 ng ml − 1 to 50 mg ml − 1. It could not be extended to higher concentrations owing to self-reversal. The detection limit was determined according to the 3s criterion for 11 replicate measurements. The calculated value 0.91 ng ml − 1 is lower than that obtained by other researchers [7–9,14] due to the fact that the operating conditions were established for a single element. For comparison, LOD values reported by
Fig. 3. The effect of liquid flow rate on aerosol transport efficiency of the ultrasonic nebulizer for water, 5%w/v KCl solution and 5%w/v tartaric acid (TA) solution.
860
K. Jankowski / Talanta 54 (2001) 855–862
Fig. 4. The matrix effect of TA concentration on sodium emission relative intensity.
Haas and Jamerson [14] for ICP and flame emission spectrometry were 8.3 and 0.5 ng ml − 1, respectively. The detection limits exhibit degradation in the presence of organic matrixes due to the decrease of precision of metal determination or/and increase of the background, depending on the matrix composition. Of course, organic components unfavorably affected the excitation conditions. However, the presence of alkali metals, especially potassium, as a matrix element improves the plasma stability and excitation efficiency. As a result, a slight increase of sodium emission intensity is observed (see Fig. 5) as well as a lowering of the LOD. Similar matrix effect was observed for lithium in the presence of sodium matrix, previously [31]. When ca. 2%w/v solution of the penicillin G potassium salt sample was analyzed a detection limit of 1.3 ng ml − 1 for sodium was obtained. For calcium pantothenate as a matrix the respective value was 1.6 ng ml − 1. For pure organic matrixes (tartaric acid, hydroquinone derivative) the LOD is much higher and reaches about 3 ng ml − 1. Analogously to the LOD changes, the precision of sodium determination (percent relative standard deviation) varies from 2.2 for standard solution to 4.1 for 5%w/v tartaric acid solution at 1 mg ml − 1 of sodium concentration.
3.3. Application The MIP-AES method was adapted for direct determination of sodium in pharmaceutical raw materials, process samples and commercial preparations. The water-soluble samples were dissolved with high purity water to obtain a 2%w/v solution. In order to improve the accuracy of determination a standard addition procedure was employed using a standard sodium chloride solution. The results obtained for analyzed samples are given in Table 3. Since attempts to find an appropriate organic water-soluble reference mate-
Fig. 5. The effect of potassium and calcium concentration on the emission intensity of sodium.
K. Jankowski / Talanta 54 (2001) 855–862 Table 3 Determination of sodium in samples and standard reference material NIST SRM 1568A Rice Flour by direct method and after digestion Sample
Na contenta (mg g−1) Direct method
Tartaric acid 1 Tartaric acid 2 Calcium pantothenate 1 Calcium pantothenate 2 Calcium pantothenate 3 Hydroquinone derivative Penicillin G potassium salt SRM 1568A
a
After digestion
239 0.9 65 9 2 11.990.4 148 95
1569 5
6869 19 41 9 1 35 9 1
35.19 0.9 7.1 9 0.3 (certified value) 6.69 0.8
861
4. Conclusions The MIP-AES method of sodium determination in water-soluble pharmaceuticals by direct introduction of an aqueous solution of the sample to the MIP have adequate sensitivity, accuracy and precision. Remarkable reduction of sample contamination is the more important advantage of the method. Moreover, the proposed method was found to be much simpler and more rapid than the traditional procedure involving sample digestion. A comparison of the results obtained by both procedures show a relative freedom of the direct method from matrix interferences up to 5%w/v of sample solution concentration. Recovery tests also demonstrate the suitability of the proposed direct method. This approach would minimize the necessary sample preparation in some pharmaceutical, clinical and environmental impurities determinations.
Values given are mean 9 standard deviation, n =5.
References
rial of low sodium content failed, the accuracy of determination was evaluated by analyzing a certified NIST SRM 1568A Rice Flour material after microwave digestion. Additionally, some of the analyzed samples were decomposed under the same digestion conditions and sodium content was determined to compare with the direct method results. The result obtained for SRM compares favorably with the certified value. The results of sodium content determination for samples of penicillin G and calcium pantothenate after microwave digestion are also in good agreement with those obtained by the direct method, when taking into account the contamination during sample pretreatment. In spike recovery tests, yields of 98.59 1.1% (n=5) for the direct determination method and 99.99 2.5% for the method with sample digestion were obtained.
[1] N. Lewen, M. Schenkenberger, T. Raglione, S. Mathews, Spectroscopy (Eugene, Oreg.) 12 (1997) 14. [2] T.H. Risby, Chem. Anal. NY 85 (1986) 179. [3] S. Ahuja, Chem. Anal. NY 85 (1986) 217. [4] G.E. Pacey, Y.P. Wu, Talanta 31 (1984) 165. [5] C.W. McLeod, P.J. Worsfold, A.G. Cox, Analyst 109 (1984) 327. [6] J. Ruzˇ icˇ ka, E.H. Hansen, E.A. Zagatto, Anal. Chim. Acta 88 (1977) 1. [7] A.T. Zander, Anal. Chem. 58 (1986) 1139A. [8] L.D. Perkins, G.L. Long, Appl. Spectrosc. 43 (1989) 499. [9] G.L. Long, L.D. Perkins, Appl. Spectrosc. 41 (1987) 980. [10] J.J. Urh, J.W. Carnahan, Anal. Chem. 57 (1985) 1253. [11] A.T. Zander, G.M. Hieftje, Anal. Chem. 50 (1978) 1257. [12] L.R. Layman, G.M. Hieftje, Anal. Chem. 47 (1975) 194. [13] L. Hiddemann, J. Uebbing, A. Ciocan, O. Dessenne, K. Niemax, Anal. Chim. Acta 283 (1993) 152. [14] D.L. Haas, J.D. Jamerson, Spectrochim. Acta 42 (1987) 299 Part B. [15] Q. Jin, W. Yang, F. Liang, H. Zhang, A. Yu, Y. Cao, J. Zhou, B. Xu, J. Anal. At. Spectrom. 13 (1998) 377. [16] K. Jankowski, J. Anal. At. Spectrom. 14 (1999) 1419. [17] K. Jankowski, R. Parosa, A. Ramsza, E. Reszke, Spectrochim. Acta 54 (1999) 515 Part B. [18] K.C. Ng, R.C. Culp, Appl. Spectrosc. 51 (1997) 1447. [19] Y. Madrid, M. Wu, Q. Jin, G.M. Hieftje, Anal. Chim. Acta 277 (1993) 1.
862
K. Jankowski / Talanta 54 (2001) 855–862
[20] H. Matusiewicz, R.E. Sturgeon, Spectrochim. Acta 48 (1993) 723 Part B. [21] L.G. Danielsson, Z. Huazhang, J. Pharm. Biomed. Anal. 7 (1989) 937. [22] P.D. Krampitz, K.W. Barnes, At. Spectrosc. 19 (1998) 43. [23] N.V. Petrov, S.A. Listov, A.P. Arzamastsev, A.V. Chuppin, Farmatsija (Moscow) 39 (1990) 51. [24] A.A. Almeida, X. Jun, J.L.F.C. Lima, Analyst 123 (1998) 1283. [25] M. Wagner, P. Rostam-Khani, A. Wittershagen, C. Rittmeyer, B.O. Kolbesen, H. Hoffmann, Spectrochim. Acta 52 (1997) 961 Part B.
.
[26] S.B. Knight, M.H. Peterson, Anal. Chem. 24 (1952) 1514. [27] S. Arpadjan, A. Alexandrova, J. Anal. At. Spectrom. 10 (1995) 799. [28] K. Niebergall, R. Wennrich, Spectrochim. Acta 47 (1992) 173 Part B. [29] K. Jankowski, D. Karmasz, A. Ramsza, E. Reszke, Spectrochim. Acta 42 (1997) 1813 Part B. [30] V.Z. Krasil’shchik, G.A. Voronina, M.S. Chupakhin, Vysokochist. Veshchest. 5 (1989) 205. [31] K. Jankowski, M. Dreger, J. Anal. At. Spectrom. 15 (2000) 269.
Direct determination of trace amounts of sodium in water-soluble organic pharmaceuticals by microwave induced plasma atomic emission spectrometry Krzysztof Jankowski * Department of Analytical Chemistry, Faculty of Chemistry, Warsaw Uni6ersity of Technology, ul. Naokowskiego 3, 00 -664 Warsaw, Poland Received 19 October 2000; received in revised form 10 January 2001; accepted 24 January 2001
Abstract The direct determination of trace sodium by microwave induced plasma atomic emission spectrometry (MIP-AES) in water-soluble organic substances utilized in pharmaceutical preparations was developed. No decomposition of the organic constituents was required. Samples were dissolved with water and introduced to the plasma after ultrasonic nebulization without desolvation. A limit of detection (3s) of 0.91– 3.0 ng ml − 1 was obtained under experimental conditions. The quantitative MIP-AES procedure involved the standard addition method. The sodium content determined in reference material NIST SRM 1568A Rice Flour agreed with the certified value (6.6 90.8 mg g − 1). Physical and chemical interferences were investigated. It was found for the microwave plasma that it is possible to introduce organic substances solutions of concentration up to 5% without sensitivity losses. This direct technique is fast and sensitive and helps to reduce contamination connected with the sample preparation procedure. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Atomic emission spectrometry; Microwave induced plasma; Water soluble pharmaceuticals; Sodium
1. Introduction There is a limited number of analytical methods, including potentiometry with ISE, spectrophotometry, atomic absorption and atomic emission spectrometry, sufficiently sensitive and selective to determine traces of sodium in pharmaceutical [1–3] and clinical [4 – 6] materials. The utilization of * Fax: + 48-2-6283339. E-mail address: [email protected] (K. Jankowski).
sample decomposition techniques before sodium determination in organic material is a common practice due to severe matrix interferences. However, in the case of determining sodium traces the problem of sample contamination during its preparation is very serious, since this element is present practically in every chemical reactant. Regardless of matrix effects the determination of sodium is often plagued by interferences from other metals, mainly alkali metals, especially in spectrophotometric and potentiometric methods.
0039-9140/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 3 9 - 9 1 4 0 ( 0 1 ) 0 0 3 3 3 - 2
856
K. Jankowski / Talanta 54 (2001) 855–862
Microwave induced plasma atomic emission spectrometry (MIP-AES) is a promising analytical method for the determination of alkali metals. Among plasma sources widely used in various spectroscopic techniques, MIP is much sensitive toward sodium determination [7]. When compared with flame atomic emission as well as absorption spectroscopy, which are the most sensitive methods for sodium, MIP-AES provides somewhat poorer detectability (LOD about 1 ng g − 1). However, a small number of reports can be found regarding the analysis of sodium by MIPAES in real samples. In a low-power MIP Long and Perkins achieved at 589 nm LODs for sodium of 1 and 2 ng ml − 1 for helium [8] and argon plasma [9], respectively. In the air MIP at operating power of 500 W the limit of detection of 82 ng ml − 1 was obtained by Urh and Carnahan [10]. When microsamples of aqueous solutions were nebulized and excited in He-MIP by Zander and Hieftje [11], the absolute detection limit for sodium of 0.12 pg for a 1 ml sample aliquot was determined. The sample introduction to Ar-MIP after vaporization by means of a microarc permits to achieve an absolute detection limit of 0.01 pg for a 10 ml sample aliquot [12]. Hiddemann et al. [13] determined sodium in high purity quartz by laser ablation-MIP-AES and obtained LOD of 35 ng g − 1. Haas and Jamerson [14] studied the possibility of alkali metals determination by argon MIP-AES and ICP-AES using the same spectrometer. With MIP the LOD for sodium of 1.3 ng ml − 1 at 589 nm and 23 ng ml − 1 at 330 nm were obtained, while these values for ICP with radial viewing were 6.4 and 110 times higher, respectively. Jin et al. [15] reported for Ar-MPT with radial viewing the LOD for sodium of 23 ng ml − 1 at 589 nm and 1.4 ng ml − 1 if oxygen sheating is utilized. High sodium contents were determined in our laboratory in the CRMs natural waters (30– 115 mg kg − 1) [16] and in pine needles (ca. 40 mg g − 1) [17]; however, the LOD calculated was as low as 1 ng ml − 1. Layman and Hieftje [12] determined ca. 300 mg ml − 1 of sodium in blood plasma after 1000-fold sample dilution. The low tolerance of MIP toward the introduction of even a small amount of sample is often argued as a main drawback of this type of
plasma. The plasma discharge may be easily extinguished especially in the presence of organics. However, Ng and Culp [18] introduced methanolic-aqueous solutions (1:1 vol) to MIP using a direct injection nebulizer. Long and Perkins [8] studied the nebulization conditions of organic solvents. Applying xylene flow of 0.46 ml min − 1 they introduced the aerosol formed to MIP without plasma extinction. Madrid et al. [19] observed that their MPT system can handle large amounts of ethanol. Matusiewicz and Sturgeon [20], when studying slurry nebulization for MIP-AES used an organic material dispersed in nitric acid solution. The determination of trace element impurities in pharmaceutical materials is of great importance because of the high product quality requirements [21–24]. Some elements concentrations provide fingerprinting of pharmaceutical products originating from different batches or manufacturers [25]. The presence of sodium may affect the basic properties of a product, i.e. solubility or pH of its aqueous solution. Flame emission or atomic absorption spectrometry are commonly used for sodium determination in pharmaceuticals. The analysis of water-soluble organic samples as their aqueous solutions is of interest in order to simplify the analytical procedure [26]. Arpadjan and Alexandrova [27] utilized ETAAS for the analysis of water-soluble pharmaceuticals. Niebergall and Wennrich determined metal impurities in the penicillin G aqueous solution by the ICP-AES method [28]. In this work the method of direct determination of sodium in water-soluble pharmaceutical materials by MIP-AES is employed. Samples are dissolved in high purity water, nebulized with an ultrasonic nebulizer and introduced to the plasma as a wet aerosol without desolvation. The speed of this method and limitation of the possibility of contamination due to not using chemical reactants for sample preparation are its advantage. The validity of the method was evaluated by analyzing a certified material, NIST SRM 1568A Rice Flour. Microwave digestion was utilized for SRM and sample pretreatment and the results obtained were compared with those obtained for the direct determination.
K. Jankowski / Talanta 54 (2001) 855–862
2. Experimental
2.1. Apparatus and measurement conditions The sodium measurements were made by a MIP-AES spectrometer MIP 750MV (Analab Ltd., Warsaw, Poland) with a TE101 rectangular cavity (Plazmatronika-Service, Wrocław, Poland) and ultrasonic nebulizer (laboratory-made). The vertically positioned aerosol cooled plasma system was described earlier [17]. The monochromator (0.75 m Czerny-Turner, grating 1800 lines per mm) with two Hamamatsu photomultipliers (UV and VIS) was used for spectroscopic measurements. The sodium emission measurements were performed at a wavelength of 588 995 nm, which is characterized by the largest sensitivity of measurement and stable and low background. The plasma torch was made of quartz (Herasil®, Kleinostheim, Germany). The plasma conditions and measurement parameters used are listed in Table 1. A microwave digestion unit UniClever™ BM1z (Plazmatronika-Service, Wrocław, Poland) opTable 1 Instrumental parameters for MIP-AES system Microwave frequency/MHz Microwave power/W Plasma torch Plasma viewing mode Plasma gas Gas flow rate (ml min−1) Sample flow rate (ml min−1) Integration time (s) Analytical wavelength (nm)
2450 150 Quartz, 2.5 mm i.d. Axial Argon, 99.998% 150 30 0.5 NaI 588 995
Table 2 Operating conditions for digestion system Pressure level (atm) Maximum power (W) Pressure level (atm) PTFE vessel volume (ml) Nitric acid (ml) Program
100% power (15 min) 300 42–44 110 3 30% power (5 min), 100% power (15 min)
857
erating in the pressure-control mode was utilized for the decomposition of the reference material NIST SRM 1568A Rice Flour and selected samples.
2.2. Chemicals and standards All chemicals used were manufactured by Merck (Darmstadt, Germany). Standard and sample solutions were prepared with chromatographic grade water Ultrapur® containing maximum 0.5 ng ml − 1 of sodium. Standard sodium solution 1 mg ml − 1 was prepared from analyticalgrade sodium chloride Suprapur® dried at 400°C. For microwave digestion 60% nitric acid Ultrapur® was used.
2.3. Sample preparation for direct determination A 0.200 g mass of sample was dissolved in 10.0 ml of water in a volumetric flask. In standard addition measurements an aliquot of 10 mg ml − 1 sodium solution (containing 3–10 mg of Na) was added.
2.4. Microwa6e digestion A dry sample (containing above 90% of organic matter) or the certified reference material NIST SRM 1568A Rice Flour (0.2–0.5 g depending on sodium content) was weighed into a 110 ml PTFE vessel and 3 ml of HNO3 was added. The vessel was covered, and the sample was heated according to the program presented in Table 2. After cooling, the vessel was opened and excess of nitric acid was evaporated to about 0.5 ml. The solution was transferred to a 10.0 ml volumetric flask, an aliquot of the sodium standard solution was added if necessary, and diluted to the mark with water.
3. Results and discussion
3.1. Optimization of analytical conditions Studies on the effect of the microwave power on the intensity of sodium emission at 588 995 nm
858
K. Jankowski / Talanta 54 (2001) 855–862
Fig. 1. The effect of microwave power on signal-to-background ratio (SBR) for sodium solution at 1 mg ml − 1.
Fig. 2. The effect of argon flow rate on signal-to-background ratio (SBR) for sodium solution at 1 mg ml − 1.
were carried out for a solution of 1 mg ml − 1 concentration over the 70– 200 W range (Fig. 1). The signal and background levels increase slowly with an increase of microwave power, however, the best signal stability was obtained for about 150 W. The influence of the argon flow rate on emission intensity is shown in Fig. 2. The highest signal-tobackground ratio was obtained at about 150 ml min − 1 of argon.
Under selected optimum operating conditions the emission intensity measurements for sodium chloride as well as sodium versenate solutions (sodium concentration range, 1–30 mg ml − 1) were performed due to select a suitable substance for standard addition procedure. As expected, no difference in the measurement sensitivity for both substances was observed and NaCl was applied in further studies, since it is easier to prepare and use.
K. Jankowski / Talanta 54 (2001) 855–862
The effect of the matrix concentration in the sample solution on the nebulization conditions and sodium excitation has been studied. The mechanical efficiency of nebulization was measured by the weighing method as described earlier [29]. The mass of aerosol introduced to the plasma was determined by nebulizing a known mass of liquid and weighing the liquid remaining in the drain and in the spray chamber. The effect of the sample flow rate on the aerosol transport efficiency for water, 5%w/v potassium chloride and 5%w/v tartaric acid (TA) solution is presented in Fig. 3. The efficiencies obtained for solutions are slightly lower owing to the higher specific gravity. The effect of TA concentration on relative emission intensity of sodium and its signal stability is shown in Fig. 4. The presence of the matrix does not affect essentially the level of the sodium signal up to the concentration of ca. 5%w/v. However, at this concentration the signal stability worsens (4.1% R.S.D.). The effect of the argon flow rate in the range 100– 250 ml min − 1 on the intensity measured was similar in shape to those presented in Fig. 2 up to the matrix concentration of ca. 3.5%w/v.
859
Niebergall and Wennrich [28] introduced to ICP a 15% solution of penicillin G and Krasil’shchik et al. [30] 10% solutions of organic acids in order to determine metal impurities. However they utilized up to a 10-times higher power level for plasma sustaining and introduced lower aerosol concentrations since a pneumatic nebulization has been used.
3.2. Analytical performances The linear dynamic range and detection limit for sodium were determined under operating conditions listed in Table 1. Standard sodium solutions (0.005–200 mg ml − 1) were used to measure the emission intensities. The linear response was obtained in the range from 10 ng ml − 1 to 50 mg ml − 1. It could not be extended to higher concentrations owing to self-reversal. The detection limit was determined according to the 3s criterion for 11 replicate measurements. The calculated value 0.91 ng ml − 1 is lower than that obtained by other researchers [7–9,14] due to the fact that the operating conditions were established for a single element. For comparison, LOD values reported by
Fig. 3. The effect of liquid flow rate on aerosol transport efficiency of the ultrasonic nebulizer for water, 5%w/v KCl solution and 5%w/v tartaric acid (TA) solution.
860
K. Jankowski / Talanta 54 (2001) 855–862
Fig. 4. The matrix effect of TA concentration on sodium emission relative intensity.
Haas and Jamerson [14] for ICP and flame emission spectrometry were 8.3 and 0.5 ng ml − 1, respectively. The detection limits exhibit degradation in the presence of organic matrixes due to the decrease of precision of metal determination or/and increase of the background, depending on the matrix composition. Of course, organic components unfavorably affected the excitation conditions. However, the presence of alkali metals, especially potassium, as a matrix element improves the plasma stability and excitation efficiency. As a result, a slight increase of sodium emission intensity is observed (see Fig. 5) as well as a lowering of the LOD. Similar matrix effect was observed for lithium in the presence of sodium matrix, previously [31]. When ca. 2%w/v solution of the penicillin G potassium salt sample was analyzed a detection limit of 1.3 ng ml − 1 for sodium was obtained. For calcium pantothenate as a matrix the respective value was 1.6 ng ml − 1. For pure organic matrixes (tartaric acid, hydroquinone derivative) the LOD is much higher and reaches about 3 ng ml − 1. Analogously to the LOD changes, the precision of sodium determination (percent relative standard deviation) varies from 2.2 for standard solution to 4.1 for 5%w/v tartaric acid solution at 1 mg ml − 1 of sodium concentration.
3.3. Application The MIP-AES method was adapted for direct determination of sodium in pharmaceutical raw materials, process samples and commercial preparations. The water-soluble samples were dissolved with high purity water to obtain a 2%w/v solution. In order to improve the accuracy of determination a standard addition procedure was employed using a standard sodium chloride solution. The results obtained for analyzed samples are given in Table 3. Since attempts to find an appropriate organic water-soluble reference mate-
Fig. 5. The effect of potassium and calcium concentration on the emission intensity of sodium.
K. Jankowski / Talanta 54 (2001) 855–862 Table 3 Determination of sodium in samples and standard reference material NIST SRM 1568A Rice Flour by direct method and after digestion Sample
Na contenta (mg g−1) Direct method
Tartaric acid 1 Tartaric acid 2 Calcium pantothenate 1 Calcium pantothenate 2 Calcium pantothenate 3 Hydroquinone derivative Penicillin G potassium salt SRM 1568A
a
After digestion
239 0.9 65 9 2 11.990.4 148 95
1569 5
6869 19 41 9 1 35 9 1
35.19 0.9 7.1 9 0.3 (certified value) 6.69 0.8
861
4. Conclusions The MIP-AES method of sodium determination in water-soluble pharmaceuticals by direct introduction of an aqueous solution of the sample to the MIP have adequate sensitivity, accuracy and precision. Remarkable reduction of sample contamination is the more important advantage of the method. Moreover, the proposed method was found to be much simpler and more rapid than the traditional procedure involving sample digestion. A comparison of the results obtained by both procedures show a relative freedom of the direct method from matrix interferences up to 5%w/v of sample solution concentration. Recovery tests also demonstrate the suitability of the proposed direct method. This approach would minimize the necessary sample preparation in some pharmaceutical, clinical and environmental impurities determinations.
Values given are mean 9 standard deviation, n =5.
References
rial of low sodium content failed, the accuracy of determination was evaluated by analyzing a certified NIST SRM 1568A Rice Flour material after microwave digestion. Additionally, some of the analyzed samples were decomposed under the same digestion conditions and sodium content was determined to compare with the direct method results. The result obtained for SRM compares favorably with the certified value. The results of sodium content determination for samples of penicillin G and calcium pantothenate after microwave digestion are also in good agreement with those obtained by the direct method, when taking into account the contamination during sample pretreatment. In spike recovery tests, yields of 98.59 1.1% (n=5) for the direct determination method and 99.99 2.5% for the method with sample digestion were obtained.
[1] N. Lewen, M. Schenkenberger, T. Raglione, S. Mathews, Spectroscopy (Eugene, Oreg.) 12 (1997) 14. [2] T.H. Risby, Chem. Anal. NY 85 (1986) 179. [3] S. Ahuja, Chem. Anal. NY 85 (1986) 217. [4] G.E. Pacey, Y.P. Wu, Talanta 31 (1984) 165. [5] C.W. McLeod, P.J. Worsfold, A.G. Cox, Analyst 109 (1984) 327. [6] J. Ruzˇ icˇ ka, E.H. Hansen, E.A. Zagatto, Anal. Chim. Acta 88 (1977) 1. [7] A.T. Zander, Anal. Chem. 58 (1986) 1139A. [8] L.D. Perkins, G.L. Long, Appl. Spectrosc. 43 (1989) 499. [9] G.L. Long, L.D. Perkins, Appl. Spectrosc. 41 (1987) 980. [10] J.J. Urh, J.W. Carnahan, Anal. Chem. 57 (1985) 1253. [11] A.T. Zander, G.M. Hieftje, Anal. Chem. 50 (1978) 1257. [12] L.R. Layman, G.M. Hieftje, Anal. Chem. 47 (1975) 194. [13] L. Hiddemann, J. Uebbing, A. Ciocan, O. Dessenne, K. Niemax, Anal. Chim. Acta 283 (1993) 152. [14] D.L. Haas, J.D. Jamerson, Spectrochim. Acta 42 (1987) 299 Part B. [15] Q. Jin, W. Yang, F. Liang, H. Zhang, A. Yu, Y. Cao, J. Zhou, B. Xu, J. Anal. At. Spectrom. 13 (1998) 377. [16] K. Jankowski, J. Anal. At. Spectrom. 14 (1999) 1419. [17] K. Jankowski, R. Parosa, A. Ramsza, E. Reszke, Spectrochim. Acta 54 (1999) 515 Part B. [18] K.C. Ng, R.C. Culp, Appl. Spectrosc. 51 (1997) 1447. [19] Y. Madrid, M. Wu, Q. Jin, G.M. Hieftje, Anal. Chim. Acta 277 (1993) 1.
862
K. Jankowski / Talanta 54 (2001) 855–862
[20] H. Matusiewicz, R.E. Sturgeon, Spectrochim. Acta 48 (1993) 723 Part B. [21] L.G. Danielsson, Z. Huazhang, J. Pharm. Biomed. Anal. 7 (1989) 937. [22] P.D. Krampitz, K.W. Barnes, At. Spectrosc. 19 (1998) 43. [23] N.V. Petrov, S.A. Listov, A.P. Arzamastsev, A.V. Chuppin, Farmatsija (Moscow) 39 (1990) 51. [24] A.A. Almeida, X. Jun, J.L.F.C. Lima, Analyst 123 (1998) 1283. [25] M. Wagner, P. Rostam-Khani, A. Wittershagen, C. Rittmeyer, B.O. Kolbesen, H. Hoffmann, Spectrochim. Acta 52 (1997) 961 Part B.
.
[26] S.B. Knight, M.H. Peterson, Anal. Chem. 24 (1952) 1514. [27] S. Arpadjan, A. Alexandrova, J. Anal. At. Spectrom. 10 (1995) 799. [28] K. Niebergall, R. Wennrich, Spectrochim. Acta 47 (1992) 173 Part B. [29] K. Jankowski, D. Karmasz, A. Ramsza, E. Reszke, Spectrochim. Acta 42 (1997) 1813 Part B. [30] V.Z. Krasil’shchik, G.A. Voronina, M.S. Chupakhin, Vysokochist. Veshchest. 5 (1989) 205. [31] K. Jankowski, M. Dreger, J. Anal. At. Spectrom. 15 (2000) 269.