Probing the microwave degradation mechanism of phenol-containing
polymeric compounds by sample pretreatment and GC–MS analysis
Yu-Cheng Chang
a,b, Fu-Hsiang Ko
a,b,∗, Chu-Jung Ko
a,b, Tieh-Chi Chu
c aNational Nano Device Laboratories, 1001-1 Ta-Hsueh Road, Hsinchu 300, TaiwanbInstitute of Nanotechnology, National Chiao Tung University, Hsinchu 300, Taiwan
cDepartment of Radiological Technology, Yuanpei University of Science and Technology, Hsinchu 300, Taiwan
Received 2 August 2004; received in revised form 29 September 2004; accepted 29 September 2004
Abstract
We have detected the volatile organic species obtained after microwave digestion of poly(hydroxystyrene) with nitric acid in closed vessels at various reaction temperatures by using headspace solid-phase microextraction (HS–SPME) and gas chromatography/mass spectrometry (GC–MS). We probed the digestion reaction by measuring the response of the detectable species – we detected a total of 20 volatile organic compounds – with respect to the temperature over the range from room temperature to 80◦C. These compounds can be classified into alkane, alkanol, and aromatic species. The alkane species decreased monotonically, whereas the alkanol and aromatic compounds first increased and then decreased, as the digestion temperature increased. We have established the degradation mechanisms, which involve bond scission, recombination, adduct formation, and ring opening, on the basis of product analyses and bond length simulations.
© 2004 Elsevier B.V. All rights reserved.
Keywords: Microwave digestion; Volatile organic species; Digestion reaction; Degradation mechanisms
1. Introduction
The semiconductor industry is an important component of the electronics industry, whose global market yield already exceeds that of the automobile industry [1]. The semicon-ductor manufacturing process, which evolves continually, involves a wide variety of distinct unit procedures[1,2]. The issue of quality control of any incoming material of interest in the manufacturing process is very crucial [3] because advancing technologies produce semiconductor devices whose physical dimensions continue to decrease [1,3]. Poly(4-hydroxystyrene) is one of the most important raw materials for preparing chemically amplified resists for the manufacture of semiconductors [4,5]. This polymer is the main constituent of resist materials and must be used at a high degree of purity (i.e., <10 ppb of metal contaminants)[6].
∗Corresponding author. Tel.: +886 35726100x7618;
fax: +886 35722715.
E-mail address: [email protected] (F.-H. Ko).
During the past few years, we had developed a series of methods [4,7–9] for analyzing lithographic materials such as resists and antireflective coatings. These methods use either closed or open-focused vessels that are sub-jected to microwave digestion and instrumental analysis. Poly(hydroxystyrene) (PHS) in resist materials is very inert and it is resistant to acid dissolution at room temperature. The material is also thermally stable; from thermogravimetric analysis, it is found to exist at ca. 24% residual mass at up to 900◦C in an ambient environment [4]. The fact that a refractory material forms indicates that the total dissolution of the polymer in an oxidant acid is not easy to achieve. Prior to analysis of the metal concentration, digestion recipes that reach at least 95% digestion efficiency are necessary to minimize spectroscopic interference from the polymer.
Kingston and Jassie evaluated the completeness of dis-solution by measuring the free amino acid content resulting from protein hydrolysis in the samples[10]. In some other studies[11,12], the total residual carbon in the digested sam-ples has been measured and used as a relative measure of the
0003-2670/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2004.09.084
efficiency of the various digestion schemes. Evaluation of the peak shape and background behavior of voltammograms also has been used to judge the quality of the digestion procedure
[13]. Yang et al. have developed a method combining radio-tracer techniques with paper electrophoresis to investigate the effectiveness of the decomposition process of65Zn-labelled liver samples[14]. In a previous report[7], we proposed that gravimetric methods can be used to evaluate the digestion ef-ficiency and kinetics of polymeric materials; we were able to evaluate digestion kinetics by weighing samples before and after digestion [15]. More conclusive and direct evidence, however, of the presence and identity of the residual mat-ter that remains afmat-ter acid dissolution is certainly desirable, especially if such matter interferes in any of the subsequent measurements.
Probing the residual organic compounds in digested so-lutions of poly(hydroxystyrene) is very difficult because of the high acidity of this matrix, the unknown polymeric dis-tribution, and the unclear set of reactions that occur during digestion. Numerous studies using many detection systems
[16–24] have focused on investigating the oxidation path-ways of phenol-related molecules subjected to wet oxygen, supercritical water, or photooxidation. Prior to instrumen-tal analysis (GC–TCD, GC–FID, HPLC–UV, HPLC–MS, GC–MS, and UV), aliquots of the samples are pretreated by liquid–liquid extraction to separate the compounds of interest. The phenol (C6H5OH) itself is oxidized into
ring-opened (e.g., CO2, CO, ethylenedicarboxylic acid, and
hexa-dienedioic acid), single-ring (e.g., catechol, benzenetriol, hydroquinone, and benzoquinone), and multiple-ring (e.g., phenoxyphenol, dibenzofuran, indanone, and dibenzofura-nol) compounds; 4-ethylphenol becomes oxidized into ring-opening products, 4-vinylphenol, phenol, etc. Despite the versatility of the methods available to determine the com-pounds derived when oxidizing phenols, to the best of our knowledge, the oxidation products and pathways followed during the microwave-assisted decomposition (digestion) of phenol-containing molecules and polymeric compounds have yet to be reported. The high nitric acid content present in the digestive solution makes the analysis of organic species more of a challenge and, hence, the reaction mechanisms and degradation kinetics remain unknown. If a database of reaction mechanisms can be established and the kinetics of these processes can be understood then selecting diges-tion recipes will become more systematic and scientifically proven.
There are the two plausible means of removing interfer-ence from the high acid concentration when studying the mi-crowave degradation of polymeric materials: liquid–liquid extraction (LLE) and solid-phase microextraction (SPME)
[25–27]. These pretreatment methods present some barriers, however, that need to be overcome. For example, nitric acid in the digestive solution can react with the extraction solvent when using the LLE method, which can limit the phase sepa-ration. The direct immersion (DI) SPME method may suffer from fouling of the fiber surface with the dissolved polymer
or corrosion of the fiber coating by the action of the strong acid. Consequently, the analytical results may not be reliable. In this study, we have evaluated the efficiency of mi-crowave digestion of PHS, in the presence of nitric acid, at various temperatures using the gravimetric method. We have studied the extent of the reaction and the ability to ef-fect phase separation using various liquid–liquid extraction solvents. We propose that using a method of HS–SPME pre-treatment and GC–MS analysis is effective for determining the volatile species in the digestion solution. Furthermore, we have characterized the degradation mechanisms of PHS under these conditions.
2. Experimental
2.1. Reagents and materials
Nitric acid (69.5%) and dichloromethane (99%) of pro-analytical grade were obtained from Merck (Darmstadt, Germany). Deionized water (18.3 M cm−1) was used throughout all the experiments. The PHS (average Mw= ca.
20,000; Tg= ca. 130–185◦C; density = ca. 1.16 g/mL) used
in this study was purchased from Aldrich (Steinhein, Germany). 4-Ethylphenol (>97%), phenol (>99.5%), and sodium hydroxide (95%) were purchased from Chem Service (West Chester, PA).
2.2. Microwave digestion procedures
Microwave digestion of PHS was accomplished by plac-ing closed vessels inside a commercial oven. The microwave system (Model MARS-5, CEM, Matthews, NC, USA) was equipped with a Teflon-coated cavity and a removable 12-position sample carousel. The oven had a variable power range (up to 1200 W) that was adjustable in 1% increments. A pressure sensor (ESP-1500 Plus) could provide “in vessel” pressure measurements of up to 1500 psi; the pressure limit was set at 350 psi. An optical fiber was used to monitor and control the digestion temperature up to 300◦C by the use of a feedback system (EST-300 Plus). The dual seal, in conjunc-tion with the frame architecture, of the XP-1500 Plus provides a completely sealed vessel that can handle temperatures of up to 300◦C and pressures up to 1500 psi.
A sample of interest (0.1 g) and HNO3(3 mL) were mixed
in the reaction vessel and digested at various temperatures. The operating parameters of the microwave system were set as follows: microwave power, 450 W; digestion time, 30 min; cooling time, 30 min. To evaluate the residual weight, the digest was transferred to a 10 mL beaker and evaporated to dryness.
2.3. Liquid–liquid extraction
The digestion solution was transferred into a glassy ex-traction tube. Dichloromethane was added into the tube up to
a total of 10 mL. The extraction tube was shaken vigorously for 10 min, left to stand for 3 h, and then a 0.3L aliquot of the solution in the organic phase was withdrawn and injected into the GC–MS system.
2.4. Solid-phase microextraction
We used manual SPME holders having a 65m poly-dimethylsiloxane/divinylbenzene (PDMS/DVB) fiber assem-bly (Supelco, Bellefonte, PA). The fibers were conditioned as recommended by the manufacturer.
The acidic solution obtained after digestion was neutral-ized with 5 M NaOH solution to a value of pH 6–8 and then an aliquot (2 mL) of this solution was transferred into a headspace vial. The vial was sealed with a headspace alu-minum cap furnished with a Teflon-faced septum. Prior to HS–SPME, the vial was immersed in a water bath maintained at 50◦C and equilibrated for 5 min. The fiber was exposed to the headspace over the solution for 1 h. The fiber was imme-diately inserted into the GC injector and the chromatographic analysis was performed. The extracted compounds were des-orbed by inserting the fiber into the GC injector for 30 min at 240◦C.
2.5. Chromatographic conditions
The desorption species from the fiber were separated us-ing a GC–MS analyzer (GC: PE AutoSystem XL Model; MS: PE TurboMass, Perkin-Elmer, Norwalk, CT, USA). These species were separated on an RTX-5 column (Supelco, Belle-fonte, PA, USA) that had a length of 30 m and an i.d. of 0.25 mm and was coated with a crosslink of 5% diphenyl and 95% dimethylpolysiloxane (minimal bleed at 330◦C). The GC oven temperature program for liquid–liquid extraction was the following: 50◦C, hold 1 min, ramp rate of 10◦C/min to a final temperature of 320◦C, and then hold for 3 min. The GC oven temperature program for HS–SPME of poly(4-hydroxystyrene) was as follows: 40◦C, hold 1 min, ramp rate 2◦C/min to a final temperature of 260◦C, and then hold for 3 min. The GC oven temperature program for the HS–SPME of phenol and 4-ethylphenol was the following: 40◦C, hold 1 min, ramp rate 2◦C/min to 120◦C, hold for 3 min, ramp rate 10◦C/min to a final temperature of 260◦C, hold for 5 min.
3. Results and discussion
3.1. Evaluation of digestion efficiency with gravimetric method
In general, a basic microwave digestion process aims to achieve a high dissolution efficiency by means of adding a digestion acid (e.g., HNO3) into the closed vessel and
react-ing the mixture under a high temperature.Table 1illustrates the digestion efficiency measured by the gravimetric method
Table 1
Digestion efficiency (%) and pressure accumulated during the digestion of PHS at various temperatures Temperature (◦C) Digestion efficiency (%) Accumulated pressure (psi) 25 5.00 1 50 10.23 5 80 21.13 14 100 30.74 42 120 47.72 85 140 84.80 120 160 95.20 172 180 98.81 221
and the accumulated pressure of PHS under various digestion temperatures (25, 50, 80, 100, 120, 140, 160, and 180◦C). We found that the digestion efficiency basically increases as the reaction temperature increases and that it has a positive re-lationship with the accumulating pressure. This observation suggests that the polymer readily decomposes under high ox-idation temperatures and that the oxox-idation by-products cause the rise in pressure. In addition, PHS can react with nitric acid at room temperature, but the digestion efficiency is very low (5%). It is interesting that the digestion efficiency at 140◦C is about twice that at 120◦C. We attribute this finding to the fact that the boiling point of 69.5% nitric acid is 122◦C. The nitric acid vaporizes and refluxes in the closed vessel when the digestion temperature is elevated above the boiling point. Therefore, the digestion efficiency is significantly enhanced at 140◦C; in contrast, most of the nitric acid exists in the liquid state below 120◦C and so digestion is limited at these temperatures.
3.2. Evaluation of extraction solvent for matrix separation
The species resulting from PHS decomposition in solu-tion after microwave degradasolu-tion are very complicated, but current analytical instrumentation, such as GC–MS, does not allow direct analysis. Conventionally, solvent extraction is used to separate the matrix, but extraction is limited to mild systems. For a degradation system involving strong acid, we need to search for a suitable solvent to separate the prod-ucts from the matrix.Table 2lists the reactions of nitric acid with a series of solvents, such as ethanol, isopropanol, oc-tanol, propylene glycol monomethyl ether acetate (PGMEA),
N-methylpyrrolidone (NMP), tetrahydrofuran (THF),
cyclo-hexane, benzene, methylbenzene, and dichloromethane. The reactions inTable 2can be divided to three groups: (1) the extraction solvent has a very violent reaction with nitric acid, (2) the extraction solvent is miscible with the nitric acid phase and has small exothermic reaction, and (3) the extraction sol-vent is phase separated from the nitric acid phase and has no exothermic reaction.
The first group, comprising alkanols, such as ethanol, isopropanol, and octanol, react vigorously with the extraction solvent and lead to a serious degree of spraying. This group
Table 2
Effects observed when mixing the extraction solvents with nitric acid Extraction solvent Rigorous
reaction
Exothermic reaction
Phase separation
Ethanol Yes Yes No
Isopropyl alcohol Yes Yes No
1-Octanol Yes Yes No
PGMEA No Yes No
NMP No Yes No
THF No Yes No
Cyclohexane No No Yes
Benzene No No Yes
Methyl benzene No No Yes
Dichloromethane No No Yes
of solvents clearly cannot be used to extract the digestion species. The second group solvents (PGMEA, NMP and THF) are readily miscible with nitric acid and, hence, they are also unfavorable for matrix separation. The most suitable solvents for separating the nitric acid matrix are cyclohex-ane, benzene, methylbenzene, and dichloromethane; these organic solvents demonstrate good phase separation with nitric acid solution and, furthermore, no exothermic reaction occurs upon shaking the mixture. Although these solvents are candidates for matrix separation after PHS polymer digestion, we need to consider the effect of interference by the solvent in the GC–MS detection. Because benzene and methylbenzene moieties are subunits of PHS and the digestion solution may contain these compounds, we would be unable to differentiate these signals in the chromatogram as arising from either the extraction solvent or the digestion species. In addition, benzene or methylbenzene might react with the degradation species of the PHS polymer. For these reasons, we do not believe that these solvents are suitable extractants. Dichloromethane seems more appropriate than cyclohexane because the presence of its Cl atoms. By GC–MS detection, we can infer the stability of dichloromethane with the nitric acid. If the extraction species from the digested PHS polymer can be identified as Cl-containing molecules, we would know that the solvent reacts with the digestion products. As a result, dichloromethane is a candidate solvent for both dissolving the digestion species and separating them from the nitric acid matrix.Fig. 1 illus-trates the GC–MS chromatograms for PHS after microwave digestion at various digestion temperatures and extraction with dichloromethane. It is interesting that the species signals are very weak for the nitric acid digestion at 25, 80, and 140◦C. As indicated inTable 1, the digestion efficiencies are 5, 21, and 85%, respectively. The degradable species after microwave digestion do exist in the nitric acid solution, but the process of solvent extraction and GC–MS analysis does not indicate their presence. We attribute this contradictory result to the low extraction efficiency, the low concentration of the products, or their non-volatile nature. Clearly, to probe the degradation products of PHS after digestion with nitric acid, we must adopt a different analytical method.
Fig. 1. Chromatograms of the digestion species of PHS upon liquid extrac-tion and GC–MS detecextrac-tion. Digesextrac-tion temperatures: (A) 25◦C, (B) 80◦C, (C) 140◦C; total ion counts: 5.64E+5.
3.3. Headspace solid-phase microextraction
Solid-phase microextraction (SPME), which was devel-oped by Pawliszyn et al. in 1990, is an attractive and widely used method for sample concentration and matrix separation
[25,26]. This technique has the advantages of a high recovery, a more-efficient matrix separation, a low detection limit, and the ability to use rapid and solvent-free extraction methods. Thus, we adapted SPME and GC–MS methods to probe the digestion species.
Prior to SPME, the digestion solution had to be neutral-ized; we used NaOH solution. The experimental conditions we used for the SPME and GC–MS analysis basically followed those described in the literature [25,28–32] Fig. 2 displays HS–SPME–GC–MS chromatograms of the products of the microwave digestion of PHS con-ducted under different reaction temperatures. Fig. 2A illustrates that 20 molecules are extracted from the so-lution at room temperature. From database searches, we identified these molecules as (1) 2,2,6-trimethyloctane, (2) 2,5,6-trimethyloctane, (3) 2,2,6-trimethyldecane, (4) 3-methyl-5-propylnonane, (5) 2,6,6-trimethyldecane, (6) 2,6,9-trimethyldecane, (7) 2,6,10-trimethyldodecane, (8) 2-ethyl-1-hexanol, (9) 4-ethyl-2,2,6,6,-tetramethylheptane, (10) 2,7,10-trimethyldodecane, (11) 1-octadecane, (12) 2,2,3, 4,6,6-hexamethylheptane, (13) 3-methylheptane, (14) non-adecane, (15) 2,6,10,14-tetramethylpentadecane, (16) 2,2,11,11-tetramethyldodecane, (17) 2,6,10,15-tetramethyl-heptadecane, (18) 2,2,4,10,12,12-hexamethyl-7-(3,5,5-tri-methylhexyl)tridecane, (19) 2,6-di-tert-butyl-4-methyl-phenol, and (20) 2,4-di-tert-butyl2,6-di-tert-butyl-4-methyl-phenol, respectively. When the digestion temperature was elevated to 80◦C, the inten-sity of various molecules was reduced significantly in the resulting chromatogram. This observation suggests that PHS and its digestion products were decomposed by the action of nitric acid. Interestingly, only three of these molecules (2-ethyl-1-hexanol, 2,6-di-tert-butyl-4-methylphenol and 2,4-di-tert-butylphenol) were present in the chromatogram after digestion at 140◦C. The digestion efficiency of PHS is
Fig. 2. HS–SPME-GC–MS chromatograms of the products of mi-crowave digestion of poly(4-hydroxystyrene) at different reaction temperatures: (A) 25◦C, (B) 80◦C, (C) 140◦C. Peak identification: (1) 2,2,6-trimethyloctane, (2) 2,5,6-trimethyloctane, (3) 2,2,6-trimethyldecane, (4) 3-methyl-5-propylnonane, (5) 2,6,6-trimethyldecane, (6) 2,6,9-trimethyldecane, (7) 2,6,10-trimethyldodecane, (8) 2-ethyl-1-hexanol, (9) 4-ethyl-2,2,6,6-tetramethylheptane, (10) 2,7,10-trimethyldodecane, (11) 1-octadecane, (12) 2,2,3,4,6,6-hexamethylheptane, (13) 3-methylheptane, (14) nonadecane, (15) 2,6,10,14-tetramethylpentadecane, (16) 2,2,11,11-tetramethyldodecane, (17) 2,6,10,15-tetramethylheptadecane, (18) 2,2,4,10,12,12-hexamethyl-7-(3,5,5-trimethylhexyl)tridecane, (19) 2,6-di-tert-butyl-4-methylphenol, (20) 2,4-di-tert-butylphenol; total ion counts: 1.02E+6.
>80% (see below).Table 3summarizes the retention time, name, formula, and relative area for each peak at the various temperatures. The peak area is obtained from three repli-cates. Despite the fact that PHS contains a benzene moiety in each unit, it is interesting that only two compounds (2,6-di-tert-butyl-4-methylphenol and 2,4-di-tert-butylphenol) contain benzene rings after microwave digestion; the other detectable species are aliphatic compounds. Most of these aliphatic compounds are alkanes (C11–C28), and only one species (2-ethyl-1-hexanol; C8) is an alkanol. We note that other non-volatile species exist in the digestion solution, but these species are not easily detected.
3.4. Classifying the degradation species
To better understand the digestion reactions that lead to the detectable species, we classified the compounds inTable 3
into three categories: alkanes (Type A), alkanols (Type B), and aromatics (Type C).Fig. 3illustrates the effect that the digestion temperature has on the response of peak area for the detectable species. The abundances of the alkane species de-crease as the digestion temperature inde-creases. This observa-tion implies that PHS is degraded into alkanes, and that these alkanes are then oxidized by nitric acid. The alkane species are totally decomposed at reaction temperatures >100◦C. The alkanol and aromatic species inFig. 3 initially increase in abundance as the reaction temperature increases, but then they decrease upon further increase in temperature. The tem-perature at which the alkanols’ abundance begins to decrease is higher than that for the aromatic species. The question
Fig. 3. Peak areas, obtained by solid-phase microextraction and GC–MS, of the three classes of products from the digestion of poly(4-hydroxystyrene) at different temperatures. The dashed line is the digestion efficiency for poly(4-hydroxystyrene).
arises: Why do these latter two species behave differently with respect to the Type A species? In essence, we observe in
Fig. 3that the digestion efficiency of PHS increases gradually as the reaction temperature increases. The observation, that a relationship exists between the digestion efficiency of PHS and the abundances of the detectable volatile species, seems to suggest that the digestion process is very complicated and involves degradation, digestion, oxidation, and recombina-tion mechanisms.
PHS contains benzene groups, but after microwave di-gestion, only two of the detectable compounds contain ben-zene units. We have used chemical bond simulation[33]to determine the most likely bonds will cleave and to under-stand the formation of alkanes (Type A) during microwave digestion.Table 4 lists the bond lengths within the repeat-ing units of PHS. Bond “c” is longer than the others in the phenyl group, irrespective of the number of repeating units. This observation suggests the bond “c” is more scissile than bond “d–j”. When comparing bonds “a–c”, we find that bond “c” is the shortest, especially at higher repeating unit. This is because bond “c” links sp2- and sp3-hybridized C atoms, whereas bond “a” and “b” both link two sp3-hybridized C atoms. This phenomenon suggests that the bonds tethering adjacent monomer units are the easiest to cleave. Hence, the Type A compounds originate from the scission of bonds “c” and “a” (or “b”), which explains why 17 alkanes are ob-served by HS–SPME and GC–MS analysis. In addition, it still has the other possibility that the benzene ring might be fragmented and undergone complicated rearrangements to form the alkanes.
The origin of the Type B compound, 2-ethyl-1-hexanol, is difficult to identify at first glance of the structure of PHS. Hence, we subjected two simple model compounds, phenol and 4-ethylphenol, to the same nitric acid digestion conditions. It is interesting to note from their GC–MS chromatograms that 2-ethyl-1-hexanol is a product of the digestion of both of these molecules.Table 5reveals that the amounts of 2-ethyl-1-hexanol products obtained from the
Y .-C. Chang et al. / Analytica Chimica Acta 526 (2004) 121–129 Table 3
Digestion products of poly(4-hydroxystyrene) and their peak areas (n = 3) at different temperatures as measured by solid-phase microextraction Peak Retention
time (min)
Compound name Formula 25◦C 50◦C 80◦C 100◦C 120◦C 140◦C
1 12.55 2,2,6-Trimethyloctane C11H24 142762± 6424 13645± 5943 9377± 973 – – – 2 13.29 2,5,6-Trimethyloctane C11H24 72182± 4645 7566± 721 4683± 698 – – – 3 13.82 2,2,6-Trimethyldecane C13H28 67739± 4598 9021± 1045 6520± 978 – – – 4 14.88 3-Methyl-5-propylnonane C13H28 52951± 6765 8068± 1765 3975± 632 – – – 5 15.53 2,6,6-Trimethyldecane C13H28 36556± 5764 8575± 1397 6999± 935 – – – 6 16.47 2,6,9-Trimethyldecane C13H28 676115± 39764 74395± 7173 67656± 7308 – – – 7 17.03 2,6,10-Trimethyldodecane C15H32 490169± 87134 48533± 5673 34773± 6512 – – – 8 17.35 2-Ethyl-1-hexanol C8H18O 191565± 29831 231250± 24981 785202± 43761 515646± 93429 92675± 6276 34449± 2945 9 18.30 4-Ethyl-2,2,6,6-tetramethylheptane C13H28 723143± 12974 58317± 9141 44263± 9962 – – – 10 19.02 2,7,10-Trimethyldodecane C15H32 568792± 24781 55621± 6217 47591± 8925 – – – 11 19.75 1-Octadecane C18H38 244377± 35872 23827± 1365 19608± 1743 – – – 12 22.79 2,2,3,4,6,6-Hexamethylheptane C13H28 145260± 9897 24170± 4745 22187± 1254 – – – 13 23.11 3-Methylheptane C18H38 359762± 34212 54615± 4196 49600± 3923 – – – 14 23.96 1-Nonadecane C19H40 67275± 8731 12986± 2769 12964± 3461 – – – 15 25.82 2,6,10,14-Tetramethylpentadecane C19H40 341631± 24921 67138± 21981 58536± 8182 – – – 16 26.11 2,2,11,11-Tetramethyldodecane C16H34 253908± 34976 48872± 6548 39150± 2176 – – – 17 26.80 2,6,10,15-Tetramethylheptadecane C21H44 64827± 9824 16548± 2874 11679± 1973 – – – 18 29.69 2,2,4,10,12,12-Hexamethyl-7-(3,5,5-trimethylhexyl)tridecane C28H58 35359± 4931 11296± 3721 7400± 1032 – – – 19 45.77 2,6-Di-tert-butyl-4-methylphenol C15H24O 114680± 7631 820175± 32974 344671± 29761 75067± 9987 43215± 3218 26441± 8132 20 48.22 2,4-Di-tert-butylphenol C14H22O 11965± 1934 76419± 5987 77738± 9273 35137± 2873 28763± 3454 23471± 4263
Table 4
Poly(4-hydroxystyrene) bond lengths ( ˚A), as a function of the number of monomer units, calculated using ACD Labs Freeware 5.0 software
Position n 3 6 9 12 a 1.50 1.52 1.53 1.56 b 1.50 1.52 1.53 1.56 c 1.49 1.48 1.47 1.47 d 1.36 1.35 1.35 1.34 e 1.36 1.35 1.35 1.34 f 1.36 1.35 1.35 1.34 g 1.33 1.33 1.33 1.32 h 1.36 1.35 1.35 1.34 i 1.36 1.35 1.35 1.34 j 1.36 1.35 1.35 1.34
digestions of both phenol and 4-ethylphenol increase as the reaction temperature increases up to 80◦C, but then they de-crease as the temperature rises further. The peak area is ob-tained from three replicates. These observations indicate that 2-ethyl-1-hexanol is the product of digestion of the phenol units of PHS. The digestion of PHS must break bonds “a” to “c” to release an aryl group that then decomposes by the ac-tion of the acid to give fragments that recombine to become, ultimately, 2-ethyl-1-hexanol. Although studying digestion products under high temperature and oxidizing conditions is very difficult, the use of simple model compounds can be an effective means of clarifying the complex processes that occur.
Finally, let us consider the Type C digest products,
2,6-di-tert-butyl-4-methylphenol and 2,4-di-tert-butylphenol. The
digestion of PHS fragments bonds “a”, “b”, and “c” to release an aryl unit, which may react with other digestion products to form the Type C compounds. We observe inTable 3that the peak area of 2,6-di-tert-butyl-4-methylphenol is at least 10-fold higher than that of 2,4-di-tert-butylphenol when diges-tion occurs at either 25 or 50◦C, but this peak ratio decreases gradually as the temperature is raised from 50 to 140◦C. These results imply that 2,6-di-tert-butyl-4-methylphenol is more stable than 2,4-di-tert-butylphenol during microwave digestion.
Table 5
Peak areas (n = 3) of 2-ethyl-1-hexanol, recorded using headspace solid-phase microextraction, after the digestion of phenol and 4-ethylphenol at different temperatures
Temperature (◦C) Phenol 4-Ethylphenol
25 174267± 9265 165841± 9787 50 421929± 12974 525085± 18854 80 1292764± 21926 1107981± 16984 100 631040± 10487 467407± 12984 120 265437± 7653 137547± 9219 140 87867± 3479 21881± 4569
3.5. The mechanism of the microwave degradation of PHS
Taking the findings above into account, inFig. 4we pro-pose degradation mechanisms for PHS subjected to nitric acid digestion. The reaction of PHS with nitric acid in a closed vessel leads to the detachment of phenol and alka-nes. The detached alkanes further digest with nitric acid to create a number of C11–C28 compounds, i.e., Type A com-pounds. Some phenols and alkanes (I) will react together with the nitric acid to form other intermediates (II), which, af-ter bond scission and recombination, yield 2-ethyl-1-hexanol that can be detected by HS–SPME and GC–MS. Reactions of the bond scission products of the alkanes with phenols lead to adducts 2,6-di-tert-butyl-4-methylphenol and 2,4-di-tert-butylphenol.
The species formed from the microwave-mediated degradation of PHS in the presence of nitric acid de-tected by HS–SPME and GC–MS analysis, and we have proposed degradation mechanisms on the basis of the bond lengths in the polymer. Microwave digestion in this high-temperature and -pressure system is a very complicated process. The chemical species are not easy to detect and analyze in the nitric acid environment, especially the non-volatile species. This study proposes a method to analyze the volatile species present in the digestion solution at various reaction temperatures. The possible reaction pathways can be understood by measuring the abundances of the digestion species as functions of temperature.
Acknowledgment
We thank the National Science Council, Taiwan, for sup-porting this research financially through contracts NSC 92-2113-M-492-002.
References
[1] S.M. Sze, Semiconductor Devices: Physics and Technology, 2nd ed., Wiley, New York, 2002 (Chapter 1).
[2] G.S. May, S.M. Sze, Fundamentals of Semiconductor Fabrication, Wiley, New York, 2004 (Chapters 4–8).
[3] International Technology Roadmap for Semiconductors, Semicon-ductor Industry Association, 2003.http://public.itrs.net/.
[4] F.-H. Ko, J.-K. Lu, T.-C. Chu, C.-T. Chou, L.-T. Hsiao, H.-C. Lin, J. Anal. At. Spectrom. 15 (2000) 715.
[5] M. Quirk, J. Serda, Semiconductor Manufacturing Technology, Prentice-Hall, New Jersey, 2001, p. 621.
[6] C.-C. Yang, F.-H. Ko, M.-Y. Wang, T.-K. Wang, S.-C. Wu, J. Elec-trochem. Soc. 147 (2001) 3853.
[7] F.-H. Ko, M.-Y. Wang, T.-K. Wang, Anal. Chem. 71 (1999) 5413. [8] J.-K. Lu, F.-H. Ko, T.-C. Chu, Y.-C. Sun, M.-Y. Wang, T.-K. Wang,
Anal. Chim. Acta 407 (2000) 291.
[9] P.-H. Chi, F.-H. Ko, C.-T. Hsu, H.-L. Chen, C.-K. Yang, Y.-C. Sun, M.-H. Yang, J. Anal. At. Spectrom. 17 (2002) 358.
[10] H.M. Kingston, L.B. Jassie, Anal. Chem. 58 (1986) 2534. [11] M. Wurfels, E. Jackwerth, M. Stoeppler, Anal. Chim. Acta 226
(1989) 1.
[12] S. Nakashima, R.E. Sturgeon, S.N. Willie, S.S. Berman, Analyst 113 (1988) 159.
[13] M. Wurfels, E. Jackwerth, M. Stoeppler, Anal. Chim. Acta 226 (1989) 31.
[14] J.Y. Yang, M.H. Yang, S.M. Lin, Anal. Chem. 62 (1990) 146.
[15] F.-H. Ko, H.-L. Chen, J. Anal. At. Spectrom. 16 (2001) 1337. [16] H.S. Joglekar, S.D. Samant, J.B. Joshi, Water Res. 25 (1991) 135. [17] C. Bencher, C. Ngai, B. Roman, S. Lian, T. Vuong, J. Photochem.
Photobiol. A 148 (2002) 183.
[18] C.J. Martino, P.E. Savage, Ind. Eng. Chem. Res. 38 (1999) 1775. [19] C.J. Martino, P.E. Savage, Ind. Eng. Chem. Res. 38 (1999) 1784. [20] C.J. Martino, P.E. Savage, Ind. Eng. Chem. Res. 36 (1997) 1385. [21] C.J. Martino, P.E. Savage, Ind. Eng. Chem. Res. 36 (1997) 1391. [22] S. Gopalan, P.E. Savage, J. Phys. Chem. 98 (1994) 12646. [23] T.D. Thornton, P.E. Savage, Ind. Eng. Chem. Res. 31 (1992) 2451. [24] T.D. Thornton, D.E. Ladue, P.E. Savage, Environ. Sci. Technol. 25
(1991) 1507.
[25] J. Pawliszyn, Application of Solid Phase Microextraction, Royal So-ciety of Chemistry, Cambridge, 1999 (Chapters 43–44).
[26] Z. Mester, R. Sturgeon, J. Pawliszyn, Spectrochim. Acta B 53 (1998) 239.
[27] M.-C. Monje, C. Privat, V. Gastine, F. Nepveu, Anal. Chim. Acta 458 (2002) 111.
[28] H. Kataoka, H.L. Lord, J. Pawliszyn, J. Chromatogr. A 880 (2000) 35.
[29] R.A. Doong, S.M. Chang, Anal. Chem. 72 (2000) 3647.
[30] M. Liompart, M. Pazos, P. Landin, R. Cela, Anal. Chem. 72 (2001) 5858.
[31] M.E. Miller, J.D. Stuart, Anal. Chem. 71 (1999) 23. [32] H.L. Lord, J. Pawliszyn, J. Chromatogr. A 885 (2000) 153. [33] ACD Labs Freeware 5.0, Advanced Chemistry Development Inc.,
