Anal. Chem. 2005, 77, 3018-3020
Correspondence
Chemically Modified Glass Fiber as a Matrix-Free Reference Material for Volatile Compounds Anna S Ä witaj-Zawadka,† Piotr Konieczka,† Jan F. Biernat,‡ Jan Wo´jcik,§ and Jacek Namies´nik*,†
Department of Analytical Chemistry and Department of Chemical Technology, Chemical Faculty, Gdan´ sk University of Technology, 11/12 Narutowicza Street, 80-952 Gdan´ sk, Poland, and Laboratory of Optical Fiber Technology, Faculty of Chemistry, Maria Curie Skłodowska University, 3 M.C. Skłodowska Square, 20-031 Lublin, Poland
Difficulties with obtaining suitable reference materials, which are applied for validation of analytical procedures and calibration of measured devices used in gaseous examinations (atmospheric air, indoor air, workplace atmosphere), are one of the most important challenges that stand in front of analytics. Known gaseous standard mixtures preparation techniques, both static and dynamic, are burdened with lots of inconveniences and disadvantages, which can be responsible for false analytical results. It appears that a very important step in the way of obtaining suitable reference materials useful in gaseous analysis is the technique of thermal decomposition of surface compounds. The paper describes a new type of matrix-free reference materialschemically modified glass fiber, which significantly makes it easier and simplifies the usage of reference materials produced using this technique. The role of reference materials in analytical practice is very important. They are mostly used for the verification of measuring instruments reliability, calibration of analytical equipment, and applicability of procedures used in various areas of analytical chemistry. They are responsible for the quality of results. Reference materials should be characterized with respect to their homogeneity, stability, and the certified property values. In the case of gaseous chemicals, gaseous standard mixtures are used as analytical standards. The problems with finding suitable reference materials are more serious challenges. The preparation of proper reference materials used in the process of gaseous sample analysis is a difficult task. A variety of known techniques of gaseous standard mixtures preparation, both static and dynamic,1 have numerous disadvantages that may result in potential errors affecting analytical results. We are proposing a new technique that seems to be an important step in the development of standards suitable for * To whom correspondence should be addressed. E-mail: chemanal@ pg.gda.pl. Fax: 00-48-59-347-26-94. † Department of Analytical Chemistry, Gdan ´ sk University of Technology. ‡ Department of Chemical Technology, Gdan ´ sk University of Technology. § Maria Curie Skłodowska University. (1) Namies´nik, J.; Przyk, E. Chem. Inz˘ . Ekol. 1999, 6, 289-300.
3018 Analytical Chemistry, Vol. 77, No. 9, May 1, 2005
gaseous analysis. The proposed technique can be classified as a dynamic technique of generation of gaseous standard mixtures. This technique is based on thermal decomposition of immobilized compounds in the stream of a carrier gas. Suitable immobilized compounds are obtained by chemical modification of the solid support surface. During chemical reaction at high temperature, a surface compound decomposes or undergoes a transformation. It is followed by liberating one or more volatile chemicals of a precisely known identity. In this technique, a significant role is played by the type of support material for the surface compounds. This is because the amount of released analyte depends on the selected support. The technique is convenient due to elimination of the steps connected with analyte dilution or enrichment in a gaseous sample. We can distinguish three supports that are best suited for surface compounds that can be a source of volatile analytes. Silica gel is used in chemical synthesis2,3 most often. It is popular mostly because of its availability and the ease of preparation of many varieties. The second one is porous glass, which is characterized by high chemical inertness over a wide pH range toward both organic and aqueous solvents. The most suitable support material for surface compounds, which can be a source of trace amounts of analytes, is glass fiber. Its relatively small specific surface area allows us to obtain a gaseous standard mixture in which the concentration of the analyte is lower by 1 or 3 orders of magnitude as compared to porous glass and silica gel, respectively. Depending on the conditions under which pyrolysis process takes place, we can obtain different amounts of analyte. The measurand concentration in the gaseous mixture depends on the following: (i) mass or length of the modified support material used; (ii) temperature at which decomposition takes place; (iii) time of gaseous standard mixtures generation; (iv) type and geometry of the support material. Three criteria should be taken into account during the evaluation of the suitability of chemically modified support materials. High purity is a primary criterion and is related to the necessity of generation of a pure, strictly defined substance in the gaseous standard mixture. When a material of required purity (2) Buszewski, B.; Leboda, R. Chem. Stosowana 1990, 34, 175-196. (3) Os´cik, J. Adsorpcja; PWN: Warszawa, Poland, 1973. 10.1021/ac048565r CCC: $30.25
© 2005 American Chemical Society Published on Web 03/12/2005
is prepared, then the optimum temperature should be determined. This temperature is characterized as a temperature at which the largest and the most repeatable amount of analyte is obtained. The minimum temperature at which the thermal decomposition process starts should be much higher than ambient temperature. Otherwise a standard in the form of chemically modified support material cannot be stored for a long time without analyte losses. High temperature of the decomposition also prevents an analyte from adsorption on the device walls. Once optimum decomposition temperature is found, the kinetics of the process should be investigated. The final steps of the study include the determination of characteristics of chemically modified support materials, namely, homogeneity of coverage by surface compound and degree of saturation of functional groups on the surface by the compound used during chemical modification process. The investigation of a degree of homogeneity of coverage of the support surface by the immobilized compound involves statistical evaluation of the relationship between the amount of analyte liberated per unit length of the fiber and the length of the fiber taken for analysis. If the support surface is covered homogeneously with the immobilized compound, then the amount of analyte generated per unit length of the fiber should not depend on the fiber length. These parameters can be evaluated on the basis of determination of repeatability of the process of thermal decomposition. The proposed technique can be used in the preparation of onecomponent or multi-component mixtures with a very wide and controlled range of concentrations of analytes. This technique has been already used for the generation of the following analytes: carbon dioxide and/or monoxide, hydrocarbons, amines, aldehydes, isothiocyanates, thiols, ethene, acetone, and methyl chloride.4-10 Silica gel has been mostly used as a support material. Others supports were porous glass and glass fiber. Glass fiber is produced from fused silica of type III (brand name SQ1) by flame hydrolysis of silica tetrachloride in an oxygen-hydrogen blowpipe in the Vernouille configuration. The content of transition elements and other metals is below 1 ppb. Water content is in the range of 800-1000 ppm. The glass is optically homogeneous; therefore, it can be used for the preparation of fiber optic of damping below 10 dB/km. Type III glass fibers were drawn from a rod (10 mm in diameter) using a fiber feeder for fiber optic telecommunication at the Laboratory of Optical Fibers Technology. The outer fiber diameter is 100 ( 2 µm. Fibers were drawn at a temperature of 2000 °C in the argon atmosphere (O2 < 1 ppm; H2O < 1 ppm). At such a high temperature and dry atmosphere, glass surface is completely dehydroxylated. Therefore, after cutting in a laminar chamber, fibers surface was subjected to hydroxylation by immersing in (4) Konieczka, P. Fresenius Anal. Chem. 2000, 367, 132-140. (5) Przyk, E.; Konieczka, P.; Szczygelska-Tao, J.; Biernat, J. F.; Namies´nik, J. J. Sep. Sci. 2001, 24, 226-229. (6) Przyk, E.; Konieczka, P.; Szczygelska-Tao, J.; Teschner, R.; Biernat, J. F.; Namies´nik, J. J. Chromatogr. A 2001, 928, 99-108. (7) SÄ witaj, A.; Przyk, E.; Szczygelska-Tao, J.; Wo´jcik, J.; Biernat, J. F.; Namies´nik, J. J. Sep. Sci. 2003, 26, 1057-1062. (8) Przyk, E.; SÄ witaj-Zawadka, A.; Konieczka, P.; Szczygelska-Tao, J.; Biernat, J. F.; Namies´nik, J. Anal. Chim. Acta 2003, 488, 89-96. (9) Przyk, E.; SÄ witaj-Zawadka, A.; Szczygelska-Tao, J.; Przyjazny, A.; Biernat, J. F.; Namies´nik, J. Crit. Rev. Anal. Chem. 2003, 33, 249-267. (10) SÄ witaj-Zawadka, A.; Konieczka, P.; Szczygelska-Tao, J.; Biernat, J. F.; Namies´nik, J. J. Chromatogr. A 2004, 1033, 145-151.
5% nitric acid for 2 h at 90 °C. Then, the fibers were rinsed several times with distilled water, dried at 105 °C in a laboratory oven, cooled and placed into plastic bags in a laminar chamber. A gaseous standard mixture can be generated in two ways, using a desorber or using an SPME device.11 In the former case, samples of chemically modified glass fiber were placed in glass tubes between silanized glass wool plugs. The glass tubes were silanized prior to use to minimize interactions between the tube wall and the analyte. The tube with a modified glass fiber was placed in a thermal desorber, which served as a sampling device and was connected to the GC-FID system via a four-port valve. Depending upon its position, the valve directed a stream of carrier gas either through the thermal desorber or directly onto the front of a GC column. The desorber was heated electrically using a temperature controller. The time of decomposition process was measured since the start of heating of the thermal desorber. After a specified period of heating, the four-port valve was switched for 1 min to the position in which the carrier gas purged the sampling loop including the desorber. The gaseous standard mixture generated in the desorber outlet was then directed to the front of a GC column. During the dosing of the standard mixture the desorber was heated continuously. Depending on the type of investigation, the sequence was repeated or a new sample of the modified glass fiber was loaded into the cool (room temperature) oven of the desorber. We have already used chemically modified glass fibers as a source of ethene and methyl chloride in gaseous standard mixtures. In case of using an SPME device, a chemically modified glass fiber was placed in the device needle. After that, the SPME device was introduced into the hot injector and the fiber exposed. After a period of time, the fiber was withdrawn into the needle, and the stream of a gaseous standard mixture was directed to the front of the column. In this case, we simplify the system by omitting the desorber. The amount of a measured component introduced depends, among others, on fiber length placed in a desorber or an SPME device. This new approach has been already applied to the generation of a gaseous standard mixture containing methyl chloride as the analyte. The investigation involved thermal decomposition of an immobilized compound from lengths of glass fiber whose surface was chemically modified using two procedures in order to generate methyl chloride (I using trimethylamine or II fibers treated with a silylating agent and N-methylmorpholine). Lengths of chemically modified glass fibers were heated to 280 °C (modification procedure I) or 270 °C (modification procedure II) for 11 min. Figure 1 shows the dependence between the amount of methyl chloride liberated and the length of glass fiber modified by the two procedures. The results of statistical evaluation of the results of determination of methyl chloride generated through thermal decomposition of the immobilized compounds obtained via the two modification procedures are compiled in Table 1. On the basis of the results shown in Table 1, it can be concluded that the fiber surface is homogeneously covered with the immobilized compound for both modification procedures. The high optimum temperature (270 or 280 °C) provides stability of a chemically modified glass fiber at room temperature (11) Ligor, M.; SÄ ciborek, M.; Buszewski, B. J. Microcolumn Sep. 1999, 11, 377383.
Analytical Chemistry, Vol. 77, No. 9, May 1, 2005
3019
Table 1. Statistical Evaluation of the Results of Measurement of the Amount of Methyl Chloride Obtained from Unit Length of the Glass Fiber type of modification
Figure 1. Dependence of the amount of methyl chloride on the length of glass fiber.
without any analyte losses. This is the evidence that a chemically modified glass fiber can be treated as a matrix-free reference material in some special cases. The proposed technique has a number of advantages. It is very easy to determine the amount of a particular gaseous analyte on the basis of a known amount of support with chemically modified surface. The possibility of generation of the analyte during the calibration process reduces memory effect associated with adsorption of compounds on the walls of apparatus and tubing. The temperature at which the process of thermal decomposition occurs is significantly higher than room temperature. That is why chemically modified glass fibers can be safely stored for a long time with no risk of liberating the analyte to the environment. This technique allows generating a wide range of the analyte concentrations in a diluent gas by varying parameters of thermal decomposition process (e.g. temperature, mass, or length of chemically modified support material). We can obtain the analyte in a gaseous standard mixture at the concentration level close to the detection limit of the calibrated instrument. A very important feature is that this technique is suitable for the generation of gaseous standard mixtures of toxic, reactive, labile, and malodorous components.
3020 Analytical Chemistry, Vol. 77, No. 9, May 1, 2005
statistical parameters
I
II
average value, x ( U (k ) 2) [ng/g] standard deviation, Sn-1 slope a standard deviation, Sa calcd Student’s t value, tcalc ) |a|/Sa critical t value, tcrit (0.05; 22) intercept b standard deviation, Sb calcd Student’s t value, tcalc.) |b - x|/Sb critical t value, tcrit (0.05; 22) correlation coefficient r calcd value, |rcalc| critical value, rcrit (0.05; 22)
0.911 ( 0.041 0.050 -0.0027 0.0014 1.91
2.62 ( 0.11 0.075 -0.0000022 0.0022 0.00099
2.07 0.944 0.020 1.64
2.07 2.626 0.032 0.031
2.07
2.07
0.377 0.392
0.00021 0.392
In summary, chemically modified glass fibers can be treated as a matrix-free reference material. Using the technique of thermal decomposition of immobilized compounds, on the basis of the length of the fiber we can obtain suitable (required) amounts of the analyte. This is a very easy, time- and cost-saving solution. ACKNOWLEDGMENT The new approach described in this paper is financed by the Committee of Scientific Research (KBN), EU grants (VI-RM and QUA-NAS), and Center of Excellence in Environmental Analytics and MonitoringsCEEAM. Note Added after ASAP Publication. This paper was posted on March 12, 2005, before all modifications of data in Table 1 had been made. The version posted on March 25, 2005, is correct.
Received for review September 27, 2004. Accepted February 24, 2005. AC048565R
