MHS-SPME

Abstract A presentation is given of headspace (HS) extraction and headspace solid-phase microextraction (HS-SPME) techniques and their combination with multiple headspace (MHS) extraction to enable quantitative determination of volatiles in solid polymer matrixes. As an example, the development of HS, HS-SPME, and MHS-SPME methods for... 

MHS-SPME [64-67]. The additional problems caused by polar matrix are also discussed. 

A method for extraction of 2-cyclopentyl-cyclopentanone from polyamide 6.6 by MAE was developed to quantitate the amount of 2-cyclopentyl-cyclo-pentanone in the polyamide 6.6 samples [66] and to validate a MHS-SPME method for extraction of 2-cyclopentyl-cyclopentanone [67]. The method was optimized with respect to the type of solvent, extraction temperature, extraction time and sample-to-solvent ratio. Chloroform and methanol were evaluated as extracting solvents. After extraction at 90 °C for 30 min the highest recovery was achieved using methanol as a solvent. This is attributed to the better compatibility between polar polyamide 6.6 and polar methanol, which gives good swelHng of the polyamide 6.6 matrix and more effective extraction of analyte. The effect of extraction time on the recovery of 2-cyclopentyl-cyclopentanone was studied by extracting 1.0 g of... 

A prerequisite in MHS-SPME is that the area ratio of consecutive peaks is constant throughout the concentration range. In practice, the constant area ratio is checked by observing the spread of individual data points from a linear regression line in the plot of logarithms of the peak areas obtained in consecutive extractions versus the munber of extractions, hi MHE, the Hnear regression line should have a correlation coefficient of at least 0.998 to show that the matrix effects are truly absent. However, the repeatabihty of SPME is lower than the repeatability of conventional HS, and somewhat lower correlation coefficients can be accepted. For MHS-SPME the correlation coefficient should be examined together with the shape of the extraction plot and a clear linear trend should be observed in the extraction plot to ensure the absence of matrix effects. 

The effect of sample amount, extraction temperature, incubation time and addition of a displacer on the measured amount of 2-cyclopentyl-cy-clopentanone in polyamide 6.6 and the characteristics of the corresponding linear regression lines are shown in Table 1 [67]. Changing the sample amount from 50 to 75 mg did not significantly influence the measured amount of 2-cyclopentyl-cyclopentanone. The extraction temperature, however, strongly influenced the measured amount. After MHS-SPME at 50 °C, the measured concentration of 2-cyclopentyl-cyclopentanone was 0.47 xg/g. At 80 °C and 120 °C, the measured concentrations were 3.90 p.g/g and 96.07 xg/g, respectively. The very low correlation coefficient of 0.686 obtained at 120 °C immediately indicates that the measurement is invalid. At 50 °C and... 

Table 1 The amount of 2-cyclopentyl-cyclopentanone in Zytel polyamide 6.6 measured by MHS-SPME under various conditions. Reprinted from [67] with permission of Elsevier. Elsevier (2004) ...

C, the correlation coefficients were 0.936 and 0.975, respectively. However, a close examination of the MHS-SPME extraction plots, shown in Fig. 4, show that both measurements are invalid [67]. 

Figure 4a shows the extraction plot of MHS-SPME at 50 °C [67]. The extraction plot is approximately hnear from the second to the fifth extraction. However, it flattens out between the fifth and the sixth extractions. In addition, the difference in the relative peak area between the first and second extraction is larger than the difference in the relative peak areas between the following extractions. This reflects the slow migration of analyte from the sample at the low temperature used for the extraction, fii the first extraction, the readily available analyte is extracted from the sample, fii the second extraction, the analyte has had hmited time to migrate from inside of the polyamide 6.6 powder to replace the analyte removed in the first extraction. Between the fifth and sixth extraction the extraction plot flattens out, which is concluded to be due to the adsorption of the analyte to the polar polyamide 6.6 powder. In the extraction plot, the adsorption is observed only after several extractions when the amount of free analyte in relation to the hydrogen-bonded analyte has decreased. 

To validate the developed MHS-SPME method, the concentration of 2-cyclopentyl-cyclopentanone in five different polyamide 6.6 samples was measured by MHS-SPME and the results were compared to the concentrations determined by the previously developed MAE method [67]. In general, the amounts determined by MHS-SPME and MAE agreed well. However, the concentrations determined by MHS-SPME were up to 30% higher than the concentrations measured after MAE. As it was shown that the developed MHS-SPME method eliminated the matrix effects, it is likely that the con-... 

HS-SPME is a very useful tool in polymer analysis and can be applied for absolute and semi-quantitative determination of the volatile content in polymers, for degradation studies, in the assessment of polymer durabihty, for screening tests and for quality control of recycled materials. For quantitative determination of volatiles in polymers, SPME can be combined with multiple headspace extraction to remove the matrix effects. If the hnearity of the MHS-SPME plot has been verified, the number of extractions can be reduced to two, which considerably reduces the total analysis time. Advantages of MHS-SPME compared to MAE are its higher sensitivity, the small sample amount required, solvent free nature and if an autosampler is used a low demand of labor time. In addition, if the matrix effects are absent, the recovery will always be 100%. This is valuable compared to other techniques for extracting volatiles in polymers in which the recovery should be calculated from the extraction of spiked samples, which are very difficult to produce in the case of polymeric materials. 

Strong interactions between the polar matrix and polar analytes may lead to extremely long equilibrium times and errors in quantitation even when the MHS technique is used. In these cases, a displacer may be added to break the interactions between the matrix and analyte. Polar 2-cyclopentyl-cyclopentanone could be quantitatively determined in polar polyamide 6.6 by MHS-SPME if water was added as a displacer to break the hydrogen bonding between 2-cyclopentyl-cyclopentanone and polyamide. The addition of water also significantly reduced the equilibrium time. A correlation was found between the amount of 2-cyclopentyl-cyclopentanone emitted from polyamide 6.6 and the total amount of 2-cyclopentyl-cyclopentanone in the material. This correlation enables rapid assessment of the 2-cyclopentyl-cy-clopentanone content using headspace techniques under non-equilibrium conditions. The analysis time is significantly reduced if the polymer samples are milled to a powder prior to extraction. 

A method based on the use of multiple-headspace (MHS) SPME using a carbowax-divinylbenzene (CW/DVB) fiber (three consecutive extractions of the same sample to minimize the possible matrix effects), showed LODs (S/N = 3) of 0.06qg/L for 4-EG and 4-EP, 0.20qg/L for 4-VG, and 0.12 qg/L for 4-VP, below the sensory thresholds of these compounds in wines (Pizarro et al.,2007). 

The experimental conditions for MHS-SPME-GC/MS/MS of volatile phenols in wines are reported in Table 5.14. The MS/MS parameters and method performances are reported in Table 5.15. 

TABLE 5.14. Multiple Headspace (MHS) SPME (n = 3) and GC/MS/MS Conditions for Analysis of Volatile Phenols in Wine2... 

The most sensitive method for CVAA has recently been reported by Wooten et al. (39) using solid-phase microextraction to concentrate the derivatized analyte. Urine, with added ammonium acetate buffer and PhAsO as an internal standard, was derivatized directly with 1,3-propanedithiol and the derivative concentrated on a poly(dimethylsiloxane) (PDMS) solid-phase microextraction (SPME) fiber. Analysis was by automated GC/MS using SIM of the isotopic MH+ ions. An impressive detection limit of 7.4pg/ml was reported, using a benchtop GC/MS system. The method was validated using spiked human urine. 

We focused our attention on performing a simple, less time-consuming method with minimum risk to human health. A HS-SPME/GC-MS approach without on-fiber derivatisation was developed, which allows quantitative analysis for quantification of about a dozen ng/L of each analyte (values close to the sensory thresholds of both 3-MH and... 

Figure 5.9 Comparison between GC-MS analysis of 3-mercaptohexan-l-ol (3-MH) and 3-mercaptohexanol acetate (3-MHA) in wine by HS-SPME without on-fiber derivatisation and by SPE using ENV+-type resin enrichment. (From Gas Chromatography-Mass Spectrometry determination of 3-mercaptohexan-l-ol and 3-mercaptohexyl acetate in wine. A comparison between Solid Phase Extraction and Headspace Solid Phase Microextraction methods, Fedrizzi et al., 2007b, Anal. Chim. Acta, 596(2), 291-297)...

Table 5.14 Data of HS-SPME/GC-MS peak areas and recoveries of 3-MHA and 3-MH in wines in dependence of their concentration and the sugars and ethanol content (relevant standard deviations in brackets).

The method is more sensitive than HS-SPME, mostly in the determination of 3-MHA for which LD is about 60% lower (0.057p-g/L of HS-SPME with respect to 0.036 p-g/L for PT), in the case of 3-MH Ld it is about 43% lower (0.069 with respect to 0.048 p-g/L), and close to the sensory threshold. LD by SPE are just a little higher than by HS-SPME and the difference with the PT method is more evident in the case of 3-MHA for which LD achieved by SPE is 0.083 xg/L. In conclusion, the best improvement in sensitivity has been reached... 

By performing analysis of the same sample, PT shows slightly higher data in comparison with those from the other two methods, probably due to better extraction of the analytes. PT sampling coupled with GC-MS analysis is a simple method applicable in the analysis of both 3-MH and 3-MHA moreover, it provides lower detection limits which are close to sensory thresholds. Among the three sampling methods, PT seems to overcome the matrix effects affecting the HS-SPME and, to a lesser extent, the SPE approaches. As for SPE, we have to emphasize that, due to its exhaustive extraction for both free and bound forms of... 

Several separation and preconcentration methods for analysis of 3-MH, 3-MHA, and 4-MP, were reported (Tominaga et al., 1998 Schneider et al, 2003 Mateo-Vivaracho et al., 2007). Recently, HS-SPME-GC/MS method with limits of quantification (LOQs) of a dozen ng/L (close to the sensory thresholds of 3-MH and 3-MHA) was proposed (Fedrizzi et al., 2007b). This method does not include determination of 4-MP because the sensory threshold of this compound in wines ( 0.4 ng/L) is too low. The method has to be calibrated using a white wine pretreated with charcoal, corrected to a total S02 100 g/L, and spiked with different quantities of standard solutions. The optimized HS-SPME conditions (temperature, sampling time, solution pH) are reported in Table 5.6. 

TABLE 5.6. Optimized HS-SPME Conditions for Analysis of 3-MH and 3-MHA in Wine2... 

Recent studies showed that the wine matrix strongly influences the apparent partition between the liquid phase and the SPME fiber coatings (Fedrizzi et al., 2007a 2007b). Wine components may potentially affect the sampling remarkable sugar content may induce the signals to rise, and polyphenols and esters can participate in competitive adsorption processes (Murray, 2001 Lestremau et al.,2004) and redox reactions (Murat et al., 2003 Blanchard et al., 2004). Consequently, a reliable HS-SPME method needs the use of a suitable internal standard, such as 6-mercaptohexan-l-ol (6-MH). 

Comparison among the HS-SPME, SPE, and PT methods, coupled with GC/MS-SIM analysis, showed that PT is more sensitive than HS-SPME in both 3-MHA (LOD -0.036 and 0.057 gg/L, respectively) and 3-MH (LOD 0.048 and 0.069 gg/L, respectively) analysis, and SPE LODs are comparable with the HS-SPME ones. Purge and Trap overcomes the matrix effects observed for HS-SPME and SPE. The latter is more suitable for analysis of volatile compounds by providing more exhaustive extraction (Versini et al.,2008). 

Microwave-assisted desorption coupled to in situ headspace solid-phase microextraction (HS-SPME) was first proposed as a possible alternative pretreatment of samples collected from workplace monitoring. Therefore, pretreatment that takes a short time and uses little or no organic solvents has led to the recent development of a new extraction technique. Solid-phase micro-extraction (SPME) coupled with GC analysis has been used successfully to analyze pollutants in environmental matrices. MHS has been developed to achieve one-step, in situ headspace sampling of semivolatile organic compounds in aqueous samples, vegetables, and soil [7, 55-58]. 

---