Anal. Chem. 1996, 68, 2610-2614
Atmospheric Pressure Microwave Sample Preparation Procedure for the Combined Analysis of Total Phosphorus and Kjeldahl Nitrogen Leo W. Collins,† Stuart J. Chalk, and H. M. “Skip” Kingston*
Department of Chemistry and Biochemistry, Duquesne University, Pittsburgh, Pennsylvania 15282-1503
An atmospheric pressure microwave digestion method has been developed for the combined analysis of total phosphorus and Kjeldahl nitrogen in complex matrices. In comparison to the digestion steps in EPA Methods 365.4 (total phosphorus) and 351.x (Kjeldahl nitrogen), this method requires less time, eliminates the need for a catalyst, and reduces the toxicity of the waste significantly. It employs a microwave-assisted digestion step, using refluxing borosilicate glass vessels at atmospheric pressure. Traditionally, this method has a time-consuming sample preparation step and generates toxic waste through the use of heavy metal catalysts. These advantages are gained by the combination of a high boiling point acid (sulfuric acid) and the application of focused microwave irradiation, which enhances the digestion process by direct energy coupling. NIST standard reference materials 1572 (citrus leaves), 1577a (bovine liver), and 1566 (oyster tissue) and tryptophan were analyzed to validate the method. Phosphorus concentrations were determined by the colorimetric ascorbic acid method outlined in EPA Method 365.3. Kjeldahl nitrogen concentrations were determined using EPA Method 351.1. The results of the analyses showed good precision and are in excellent agreement with the NIST published values for both elements. There is much evidence showing that phosphorus and nitrogen are two major pollutants in the environment.1 This has recently been reinforced in a report on the fouling of the Chesapeake Bay.2 Phosphorus and nitrogen are released into environmental water by many different sources, e.g., animal and chemical fertilizer runoff and sewage. Excessive amounts of these nutrients activate eutrophication, the process of rapid growth of plankton, algae, and plants. Subsequent decay of these materials causes dissolved oxygen to be removed from a water body and with it the ability to sustain life. The analysis of these elements has thus become increasingly important. To date, the U.S. Environmental Protection Agency (EPA) has only promulgated methods for the analysis of these elements in waters and wastes. The traditional EPA Method † Present address: O. I. Analytical, P.O. Box 9010, College Station, TX 778429010. (1) U.S. Environmental Protection Agency. Report to Congress: Water Quality of the Nation’s Lakes; EPA: Washington, DC, 1989. (2) Carpenter, B.; Watson, T. U.S. News World Rep. 1994, Sept 12, 63-65.
2610 Analytical Chemistry, Vol. 68, No. 15, August 1, 1996
365.4,3 the determination of total phosphorus, and EPA Method 351.x,4 Kjeldahl nitrogen determination, are time-consuming procedures. This is primarily due to the sample preparation step rather than the elemental analysis. In addition, the sample preparation methodologies involve the use of metal catalysts, which create a toxic waste cleanup problem. Concurrent with the research on phosphorus analysis, nearly identical sample preparation steps have been developed in Methods 351.x for the analysis of Kjeldahl nitrogen (bioavailable nitrogen). These methods also incorporate metal catalysts. With emphasis now being placed on the identification of other sources of these nutrients, attention has turned to the analysis of solids, e.g., plants, fish, etc. In the area of microwave sample preparation, there have been many attempts at the determination of phosphorus in biological samples and at the determination of nitrogen. No EPA Office of Solid Waste (SW-846) methods equivalent to Method 3xx currently exist. This work describes the application of a catalyst-free, atmospheric pressure microwave sample preparation method for the analysis of total phosphorus and Kjeldahl nitrogen, based in part on the work of Feinberg et al.5-7 Microwave sample preparation of standard reference materials (SRMs) and tryptophan was performed at atmospheric pressure by a programmed multistep sulfuric acid-hydrogen peroxide method in a single-mode cavity. The method is faster than the sample preparation methodologies that are applied to aqueous samples and shows excellent recovery of both analytes. It may also be the basis for a sample preparation method for both solid and liquid samples for either analyte. BACKGROUND Phosphorus analysis has become a classical analytical determination since its first description in 1894.8 Current variations on spectrophotometric procedures for phosphorus are numerous.9,10 EPA Method 365, the traditional scheme for phosphorus determination in water and wastes, has several procedure varia(3) Total Phosphorus; Environmental Monitoring and Support Laboratory Report EPA-600/4-79-020; United States Environmental Protection Agency: Cincinnati, OH, March 1983. (4) Total Kjeldahl Nitrogen; Environmental Monitoring and Support Laboratory Report EPA-600/4-79-020; United States Environmental Protection Agency: Cincinnati, OH, March 1983. (5) Feinberg, M. H.; Ireland Ripert, J.; Mourel, R. M. Anal. Chim. Acta 1993, 272, 83-90. (6) Feinberg, M. H.; Suard, C.; Ireland Ripert, J. Chemom. Intell. Lab. Syst. 1994, 22, 37-47. (7) Suard, C. L.; Feinberg, M. H.; Ireland Ripert, J.; Mourel, R. M. Analusis 1993, 21, 287-291. (8) West Knights, J. Analyst 1880, 5, 195. (9) Bassett, J.; Denney, R. C.; Jeffrey, G. H. Vogel’s textbook of quantitative inorganic analysis, 4th ed.; Longman: London, UK, 1983. S0003-2700(96)00133-3 CCC: $12.00
© 1996 American Chemical Society
tions that take advantage of these chemistries.3 All are approved under the National Pollutant Discharge Elimination System (NPDES). EPA Method 365.1 (colorimetric, automated, ascorbic acid), EPA Method 365.2 (colorimetric, ascorbic acid, single reagent), and EPA Method 365.3 (colorimetric, ascorbic acid, two reagent) all require sulfuric acid hydrolysis and persulfate digestion to reduce all forms to orthophosphate. These steps can take 60-90 min to perform. Analysis of the orthophosphate ion is made by the ammonium molybdate colorimetric method, where an antimony phosphomolybdate complex is generated and reduced by ascorbic acid. The blue complex is proportional to the phosphorus concentration. EPA Method 365.4 (colorimetric, automated, block digestor AA II) requires a 150 min preparation time using sulfuric acid, K2SO4, and HgSO4 and also uses molybdenum blue reaction for detection. Microwave sample preparation for the analysis of phosphorus analysis using open (Kohlrausch) flasks in a multimode cavity has been reported.11 In this work, the digestion period was reduced from 6 to 2 h, using a multistep nitric acid-hydrogen peroxide microwave digestion procedure. Multielement determination, including that of phosphorus, in biological and environmental samples has also been reported using closed vessel systems, with subsequent analysis by atomic absorption spectroscopy (AAS) and inductively coupled plasma optical emission spectroscopy (ICP-OES).12-16 Similarly, a procedure for the determination of phosphorus in water samples, using a closed vessel microwave system, has been demonstrated.17 A closed system was used in a methodology comparison of plant tissue analysis.18 The rapid determination of phosphorus in bituminous coal, using a closed vessel microwave-assisted method, has been performed.19 Closed vessel microwave systems have also been used for drying samples in the analysis of phosphorus by the gravimetric drying technique20 and the colorimetric method.21 Kjeldahl nitrogen analysis is a classical measurement in analytical chemistry. It has been used extensively over the past 115 years and has recently been excellently reviewed.22 The majority of the modifications to the basic method have focused on improving the sample preparation steps for quantitative recovery of nitrogen. In EPA Methods 351.1-351.4, sulfuric acid, K2SO4, and HgSO4 are again used for digestion, with times ranging from 3 min to 2.5 h. Although Method 351.1 uses autoanalyzer technology (air-segmented continuous flow) to process samples very quickly, it generates a significant amount of toxic waste due to the high flow rates used. In the more recent attempts at developing sample preparation procedures for the subsequent determination of Kjeldahl nitrogen, (10) Robards, K.; McKelvie, I. D.; Benson, R. L.; Worsfold, P. J.; Blundell, N. J.; Casey, H. Anal. Chim. Acta 1994, 287, 147-190. (11) White, R. T.; Douthit, G. E. J. Assoc. Off. Anal. Chem. 1985, 68, 766-769. (12) Mingorance, M. D.; Perez Vazquez, M. L.; Lachica, M. J. Anal. At. Spectrom. 1993, 8, 853-858. (13) Xu, L.; Shen, W. Fenxi Shiyanshi 1989, 8, 33-34. (14) Xu, L.; Shen, W.; Zhu, J. Fenxi-Huaxue 1990, 18, 597-601. (15) Millward, C. G.; Kluckner, P. D. J. Anal. At. Spectrom. 1989, 4, 709-713. (16) Wandt, M. A. E.; Pougnet, M. A. B. Analyst 1986, 111, 1249-1253. (17) Johnes, P. J.; Heathwaite, A. L. Water Res. 1992, 26, 1281-1287. (18) Rechcigl, J. E.; Payne, G. G. Commun. Soil Sci. Plant Anal. 1990, 21, 22092218. (19) Riley, K. W.; Schafer, H. N. S.; Orban, H. Analyst 1990, 115, 1405-1406. (20) Melton, J. R.; Hoover, W. L. J. Assoc. Off. Anal. Chem. 1981, 64, 13191321. (21) Smith, M. W.; Gaines, T. P. Hortic. Sci. 1980, 15, 614. (22) McKenzie, H. A. Trends Anal. Chem. 1994, 13, 138.
Figure 1. Schematic of the Prolabo Microdigest 401.
a significant amount of effort has been directed toward the use of microwave irradiation.5-7 Optimization,5 validation,7 and automation6 of this atmospheric pressure single-mode microwave method has led to its incorporation in the French AFNOR method23 as an alternative sample preparation procedure. EXPERIMENTAL SECTION Apparatus. The atmospheric pressure microwave unit used in this work was a Microdigest 401 (Prolabo, Paris, France). Samples were placed in an open borosilicate glass vessel, customdesigned by Prolabo for use in the Microdigest 401. The vessel is “open” only with respect to pressure equilibration, since a refluxing head allows the flask to retain all solvent and volatile elements. This particular model can accompany sample sizes up to 10 g. Unique features of the atmospheric pressure unit include the ability for sequential addition of reagents during digestion and precise control of the energy, due to the generation of a focused, single-mode microwave coupling mechanism. A schematic of the Prolabo Microdigest is shown in Figure 1. Analysis of the digestates for phosphorus was performed on a Shimadzu UV-visible spectrophotometer, Model UV-1201 (Shimadzu Scientific Instruments, Columbia, MD). Total Kjeldahl nitrogen analysis was performed by GeoChem Labs (Somerset, PA) using EPA Method 351. Reagents. Analytical standard reference materials (SRMs) used in this study were obtained from the National Institute of Standards and Technology (NIST, Gaithersburg, MD). The materials used in this study were SRM 1572 (citrus leaves), SRM 1577a (bovine liver), SRM 1566 (oyster tissue), and tryptophan (ACS certified grade, Fisher Scientific, Pittsburgh, PA). ACS certified grade sulfuric acid and hydrogen peroxide were obtained from Fisher. ACS certified grade potassium hydrogen phosphate (Fisher) was used as the phosphorus standard. Procedures. Sample preparation was accomplished by a tworeagent (sulfuric acid and hydrogen peroxide) digestion and oxidation procedure. An accurately weighed 0.5-1.0 g portion (23) Kjeldahl Nitrogen Determination with Microwave Sample Preparation; Report NF V 03-100; Association Francaise de Normalisation: Paris, France, Sept 6, 1992.
Analytical Chemistry, Vol. 68, No. 15, August 1, 1996
2611
Table 1. Results of the Atmospheric Pressure Microwave Digestion and Analysis for Total Kjeldahl Nitrogen and Total Phosphorus nitrogen (% w/w) ( 95% CI sample tryptophan bovine liver oyster tissue citrus leaves
certified 13.72a 10.6 ( 0.6
exptl 13 ( 0.62 10 ( 0.91
phosphorus (% w/w) ( 95% CI certified
exptl
1.11 ( 0.04 (0.81) 0.13 ( 0.02
1.1 ( 0.12 0.807 ( 0.0048 0.133 ( 0.0016
a Calculated from stoichiometry; NIST noncertified value in parentheses.
Figure 2. Generalized atmospheric pressure microwave sample preparation digestion program.
of sample is transferred into the borosilicate vessel. After placement of the vessel into the microwave unit, the reflux/ reagent addition head is connected to the top of the vessel. The digestion program used is shown in Figure 2. Twenty milliliters of sulfuric acid was introduced into the digestion vessel for an 11 min digestion step. Over the first 4 min, the microwave power was linearly ramped to 210 W. For the next 6 min, the power remained at 210 W, generating a solution temperature of 275-310 °C. Following a short cooldown period, 6-12 mL of 30% hydrogen peroxide was added until complete oxidation was obtained. Oxidation took an additional 10 min to complete at 285 W. This procedure was performed with the microwave unit controlled by a TX32 (Prolabo) computer module. The module was programmed to provide both the power/time curve and the reagent timing and volume addition. Note: Safe laboratory procedures should always be used when performing the microwave sample preparation methods described. The operator should use an appropriate vapor exhaust system and wear protective clothing and eyeware. Oil- or grease-based lubricants should not be used with any of the glassware. Caution should always be exercised when working with hydrogen peroxide. All samples, particularly those having high organic content, must be sufficiently digested prior to any treatment with hydrogen peroxide. It is recommended that at least 10 mL of acid/g of sample be used for the digestion, and the resultant solution should appear as a homogeneous brown or black solution. The solution should also be allowed to cool down below ∼100 °C prior to the addition of hydrogen peroxide for the final oxidation step. Accurate temperature measurements were made at various points in the digestion program using fiber-optic technology. A Luxtron fluoroptic thermometry system, Model 750 (Luxtron Corp., Santa Clara, CA), was used. It was equipped with a Luxtron fiber-optic probe and high-temperature tip, necessary to measure temperatures approaching the boiling point of sulfuric acid (340 °C). Temperatures at various stages of the digestion program are shown in Figure 2. The analysis was performed on quadruplicate samples by diluting the acidic digestates and complexing the phosphorus as mentioned previously and then analyzing in triplicate with the UV-visible. The spectrophotometer was first calibrated with the phosphorus standard, and measurements were made at 650 nm as recommended in EPA Method 365. 2612 Analytical Chemistry, Vol. 68, No. 15, August 1, 1996
RESULTS The procedure in Figure 2 represents a generalized digestion program for the analysis of phosphorus. Minor modifications in the program, such as slower power ramping during the initial digestion step to 210 W, were made, depending on the sample matrix being tested. This was necessary to minimize foaming during the early minutes of the digestion of, for instance, the citrus leaves SRM. The more complex matrices, such as the bovine liver SRM, required a slightly longer oxidation period. When a modification was necessary, the total program time increased by only 3-6 min, and thus the total time necessary for complete digestion was under 30 min in every case. The atmospheric pressure microwave procedure developed here significantly reduces the digestion period for all the matrices studied without the need for a catalyst. The rapid digestion is made possible by an efficient energy coupling using a microwave digestion system and is aided by using a high-temperature boiling acid (sulfuric acid, bp 340 °C), which allows for elevated temperatures to be used in the procedure. As a result, all sample digestates were clear solutions and precipitate-free on visual inspection. Therefore, filtering was unnecessary before analysis. However, it should be noted that, if this procedure were extended to other analytes, the residual sulfuric acid content of the digests may not be compatible with all detection systems. Results of the total phosphorus analysis for the three SRMs and the Kjeldahl nitrogen analysis of bovine liver and tryptophan are shown in Table 1. Excellent agreement was obtained between the SRM values and the experimental results. The precision and accuracy for the phosphorus results of SRMs 1572 and 1566 are superb. The precision for both the phosphorus and nitrogen results of SRM 1577a are very good, particularly in light of the sample being one of the more difficult matrices to digest. Precision is better than 3% RSD for the nitrogen analysis of tryptophan, with good accuracy. Kjeldahl nitrogen analysis of the bovine liver SRM has not been reported in the microwave sample preparation literature. However, significant work has been done on the analysis of phosphorus in each of the three SRMs, and this is summarized in Tables 2-4. For the bovine liver, the accuracy of the result is better than anything previously reported, but the precision is much poorer. (24) Que Hee, S. S.; Boyle, J. R. Anal. Chem. 1988, 60, 1033-1042. (25) White, R. T. In Introduction to Microwave Sample Preparation: Theory and Practice; Kingston, H. M., Jassie, L. B., Eds.; American Chemical Society: Washington, DC, 1988; pp 53-78. (26) Andrasi, E.; Dozsa, A.; Bezur, L.; Ernyei, L.; Molnar, Z. Fresenius’ J. Anal. Chem. 1993, 345, 340-342. (27) Nadkarni, R. A. Anal. Chem. 1984, 56, 2233-2237.
Table 2. Comparison of Results for Phosphorus Analysis in NIST SRM 1577 Bovine Liver
a
resultsa
n
reagents
type of digestion
ref
1.11 ( 0.04 1.1 ( 0.12 1.24 ( 0.02 1.21 ( 0.01 1.32 ( 0.03 1.01 ( 0.01 1.07 ( 0.014 0.93 1.07 ( 0.05
4 10 3 11 3 10 1 nrb
H2SO4, H2O2 HNO3, H2O2 HNO3, HClO4 HNO3, H2O2 HNO3, H2O2, HF HNO3 aqua regia, HF HNO3 or HNO3, HCl or NH4EDTA
focused microwave, open vessel cavity microwave, open vessel cavity microwave, closed vessel cavity microwave, open vessel cavity microwave, closed vessel cavity microwave, closed vessel cavity microwave, closed vessel focused microwave, open vessel
certified value this work 11 24 25 25 26 27 28
Mean ( 95% CI. b Not reported
Table 3. Comparison of Results for Phosphorus Analysis in NIST SRM 1566 Oyster Tissue
a
resultsa
n
reagents
type of digestion
ref
0.81 0.807 ( 0.005 0.77 ( 0.01 0.87 ( 0.03 0.68 0.56 ( 0.02 0.72 ( 0.02
4 10 1 1 4 nrb
H2SO4, H2O2 HNO3, H2O2 HNO3, H2O2 HNO3, H2O2, HF aqua regia, HF HNO3 or HNO3, HCl or NH4EDTA
focused microwave, open vessel cavity microwave, open vessel cavity microwave, open vessel cavity microwave, closed vessel cavity microwave, closed vessel focused microwave, open vessel
reference value this work 11 25 25 27 28
Mean ( 95% CI. b Not reported.
Table 4. Comparison of Results for Phosphorus Analysis in NIST SRM 1577 Citrus leaves
a
resultsa
n
reagents
type of digestion
ref
0.13 ( 0.02 0.133 ( 0.0016 0.13 ( 0.00 0.133 ( 0.003 0.13 ( 0.00 0.13 ( 0.00 0.154 ( 0.004 0.13 ( 0.0016 0.12 0.11 ( 0.0005 0.13 ( 0.004 0.14 ( 0.002 0.134 ( 0.0044 0.133 ( 0.001 0.11 ( 0.03
4 10 3 11 3 6 nrb 1 nr nr nr 49 7 42
H2SO4, H2O2 HNO3, H2O2 HNO3, HClO4 HNO3, H2O2 HNO3, H2O2, HF HNO3, H2O2 HNO3, HCl HNO3, HClO4 HNO3, HCl, HClO4 HNO3, HCl, HF HNO3, HCl, H2O2 HNO3, HCl, H2O2 HNO3, HClO4 HNO3, HF, H2O2
focused microwave, open vessel cavity microwave, open vessel cavity microwave, closed vessel cavity microwave, open vessel cavity microwave, closed vessel cavity microwave, closed vessel cavity microwave, closed vessel cavity microwave, closed vessel cavity microwave, closed vessel cavity microwave, closed vessel cavity microwave, closed vessel cavity microwave, open vessel cavity microwave, open vessel cavity microwave, closed vessel
certified value this work 11 24 14 14 29 30 30 30 30 30 31 32 33
Mean ( 95% CI. b Not reported.
The reason for this does not seem apparent and cannot be attributed to the loss of volatiles during initial oxidation, the reaction being less violent compared to the other matrices. Both precision and accuracy are far superior than previously reported for the analysis of the oyster tissue. Finally, the analysis accuracy and precision for citrus leaves is equal to that reported by other researchers and better than the majority. These tables show the great improvement this method provides over other reaction chemistries and microwave implementations. As a general sample (28) Krushevska, A.; Barnes, R. M.; Amarasiriwaradena, C. J. Analyst 1993, 118, 1175-1181. (29) Miller, R. O. Principles and Practices of Microwave Dissolution/Digestion of Plant Material. International Soil Testing and Plant Analysis Symposium, Orlando, FL, 1991. (30) Schelkoph, G. M.; Milne, D. B. Anal. Chem. 1988, 60, 2060-2062. (31) Kalra, Y. P.; Maynard, D. G.; Radford, F. G. Can. J. For. Res. 1989, 19, 981-985. (32) Mateo, M. A.; Sabate, S. Anal. Chim. Acta 1993, 279, 273-279. (33) Zunk, B. Anal. Chim. Acta 1990, 236, 337-343.
preparation methodology, it shows improved robustness for phosphorus analysis. The use of sulfuric acid seems to have a significant benefit over the nitric acid-based methods previously reported. CONCLUSIONS The results obtained for the analysis of total phosphorus on the three SRMs are in excellent agreement with the NIST values and have superb precision for oyster tissue and citrus leaves. Although the phosphorus analysis of bovine liver is not as good, the joint analysis for both phosphorus and nitrogen is accomplished with this single procedure with good accuracy and precision for both elements. As demonstrated, the sample preparation time using the microwave method was significantly reduced from 90-150 min (EPA Method 365 procedures) to less than 30 min. In addition, a catalyst was not necessary for the digestion and analysis for phosphorus, thus eliminating the concern of toxic waste Analytical Chemistry, Vol. 68, No. 15, August 1, 1996
2613
buildup in the laboratory. The ability to automate the digestion process provides increased safety as well as improved reproducibility. Atmospheric pressure microwave sample preparation can be as effective a method as closed vessel microwave sample preparation methods, if not better, based on the other values reported in the literature. The advantages of working at atmospheric pressure, including the ability to add reagents sequentially, and the safety improvement when avoiding high pressures, make it an attractive alternative. While the technique is clearly not a panacea for the sample preparation of all matrix types, it should be extendible to other analytes and matrices. Limitations of sample throughput and the lack of reaction monitoring are being addressed at this time.
2614
Analytical Chemistry, Vol. 68, No. 15, August 1, 1996
ACKNOWLEDGMENT This research was supported by a grant from the Prolabo Corp. The authors thank Dr. Daniel Mathe´, Mr. Jean-Louis DiMartino, and Mr. Jean-Luc Lenoir at Prolabo for their useful discussions. The authors also thank Dr. Elke Lorentzen, Dr. Peter Walter, and Mr. James Ferguson at Duquesne University for their valuable suggestions and assistance. Luxtron Corp. provided the hightemperature fiber-optic tips at no cost for use in these experiments, and their support is appreciated. Received for review February 7, 1996. Accepted May 13, 1996.X AC960133W X
Abstract published in Advance ACS Abstracts, July 1, 1996.