Behavior and Evaluation of Tetraalkylammonium Bromides as

Article pubs.acs.org/ac

Behavior and Evaluation of Tetraalkylammonium Bromides as Instrument Test Materials in Thermal Desorption Ion Mobility Spectrometers Leonard T. Demoranville,*,†,§ Laurent Houssiau,‡ and Greg Gillen† †

Surface & Microanalysis Science Division, National Institute of Standards and Technology, MS-8371, 100 Bureau Drive, Gaithersburg, Maryland 20899, United States ‡ Research Centre in Physics of Matter and Radiation (PMR), Physics Department, University of Namur (FUNDP), 61 rue de Bruxelles, B 5000 Namur, Belgium S Supporting Information *

ABSTRACT: Ion mobility spectrometers have found widespread use for the screening of explosives, chemical warfare agents, and illicit drugs. These instruments often rely on drifttime calibrants to perform qualitative identification. Such calibrants are suitable to determine the reduced mobility of compounds, but may not necessarily provide information on instrument performance. These calibrants are often internal and behave in a fashion similar to that of analytes in terms of sensitivity to a variety of instrumental and environmental variations. Ideally, test materials used to evaluate instrument performance would be insensitive to these changes. Current calibrants are sometimes not designed to satisfy these requirements, and although several instrument test materials are also in use, a consensus has not been reached on best practices. A homologous series of tetraalkylammonium salts has been proposed as one alternative instrumental test material set and has been successfully used for electrospray ion mobility spectrometry (IMS)−MS experiments. This study extends these analyses to thermal desorption IMS instruments. The reduced mobility (K0) values for these compounds, measured on a thermal desorption IMS instrument, are reported and are similar to those reported for electrospray IMS. The variability of daily instrument response to these compounds is small, further supporting their use as test materials. The ionization behavior and thermal profile of the compounds in the thermal desorption process are discussed.

I

where K is the ion mobility, T is the drift gas temperature, P is the drift region pressure, and T0 and P0 are standard temperature and pressure, respectively . Experimentally, the reduced mobility is often calculated such that the reduced mobility of an analyte of interest is compared to that of a calibrant, giving

on mobility spectrometry (IMS) is a technique that separates gas-phase ions under the influence of an electric field at relatively high, usually atmospheric, pressure. In the simplest case the mobility of an ion is constant under uniform conditions. This mobility coefficient can be expressed as

K=

l2 tdV

K0 =

(1)

where K is mobility, l is the drift tube length, td is the drift time, and V is the potential applied across the drift tube. Variations in electric field, temperature, pressure, and humidity can cause variations in drift time for a given instrument or between instruments. Slight variations in drift tube length caused by expansion due to temperature variability can further add uncertainty to the mobility measurement. Many of these factors are not easily quantified. The preferred method of calculating mobility has become the reduced mobility, which corrects for variations in temperature and pressure such that K0 = K

T0 P T P0

(3)

where K0 and td are the reduced mobility and drift time of the cal analyte, respectively, and Kcal are the reduced mobility 0 and t and drift time of the calibrant, respectively. Ideally, a calibrant’s mobility is unaffected by the previously mentioned variations. Recently it has been suggested that IMS test materials be classed in one of two manners, either as instrument standards, which are unaffected by buffer gas contamination, or as mobility standards, which behave like the analyte in the presence of low levels of contamination.1 Herein we adopt the terminology test Received: October 10, 2012 Accepted: January 24, 2013 Published: January 24, 2013

(2) © 2013 American Chemical Society

K 0calt cal td

2652

dx.doi.org/10.1021/ac302944m | Anal. Chem. 2013, 85, 2652−2658

Analytical Chemistry

Article

published reduced mobility value in the literature. For others there are multiple published values with variability in these values. In addition, the published studies on this class of compounds are limited to electrospray ionization (ESI) ion mobility spectrometers.1,4,5,8−10 Because of the potential benefits of use of these salts as instrument standards, this study extends these evaluations to determine the behavior of the compounds in a thermal desorption IMS instrument with a 63 Ni β emitter ion source.

material, rather than standard, to more accurately describe the role of these materials in the measurement process. Mobility test materials would be used to show the presence of contamination, including humidity, and provide the opportunity either to remove contaminants or to be used as an internal test material to compensate for the influence of these factors on reduced mobility. Instrument test materials would provide confirmation of proper operational conditions for the instrument. As such, instrument test materials should provide a means of evaluating reduced mobility and resolving power over the entire instrumental operating range and the instrumental response to a known analyte concentration, while remaining insensitive to environmental conditions. Thus, having a set of instrument test materials that are applicable across a wide range of mobilities would also allow for better determinations of instrument function. In this way, instrument test materials would not be suitable for analysis in combination with analysis, as deconvolution of overlapping signals from the standard and the analyte may prove to limit the operational window of mobility. Instead a series of compounds across the mobility spectrum could be analyzed prior and subsequent to the collection of experimental data to verify instrument performance. Such performance checks are already common for verification in explosives screening, but are usually based on instrument response to a single compound. Mobility test materials would still be required for proper determination of reduced mobility values. Commercial ion mobility spectrometers have found widespread use as a screening tool for the detection of explosives, chemical warfare agents, and illicit drugs in a variety of deployment scenarios. In these situations, the instruments are generally utilized by nontechnical staff often under challenging environmental conditions. Calibrations and performance evaluation of the instruments require robust test materials to ensure proper instrumental function. In part because of the widespread use of IMS instrumentation for safety and security screening for explosives, a small but effective list of instrument and mobility test materials for the negative ion detection modeused for most explosive and chemical warfare agent detectionhas been developed.2 In the positive ion mode used for most illicit drug detectiona longer list of potential compounds exists and is in use, yet the community has not reached a consensus of a common positive ion instrument test material.1−3 While a common set of test materials is not necessarily required for effective operation, further research into the behavior of these materials and best practices for use is warranted. One proposed class of compounds for this purpose is the tetraalkylammonium salts.1,4,5 These compounds are known to produce ions when heated, both in vacuum and at atmospheric pressure.6,7 Because of this ionic character, these compounds are thought to not compete in the ionization mechanisms.1 Therefore, the use of these compounds as test materials should be less sensitive to the selection of the ionization source, i.e., radioactive sources. Additionally, they have a wide range of reduced mobilities and because of steric hindrance are thought to minimize the formation of dimers and water clusters. In one study, Fernandez-Maestre et al. found the tetraalkylammonium salts to be suitable IMS instrument test materials; that is, they are unaffected by trace drift gas contaminants. However, they expressed concerns that the reduced mobility values have yet to be completely established.1 For some salts, there is only one

■

EXPERIMENTAL SECTION Solutions of tetrapropylammonium bromide, tetrabutylammonium bromide, tetrapentylammonium bromide, tetrahexylammonium bromide, tetraheptylammonium bromide, tetraoctylammonium bromide, and tetradodecylammonium bromide (research grade, Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and tetradecylammonium bromide (TCI America, Portland, OR) were prepared in isobutyl alcohol (99%, SigmaAldrich, St. Louis, MO) at concentrations of 3 × 10−4 and 7.5 × 10−4 M. An additional solution was prepared that contained a mixture of all the salts, each at a concentration of 3 × 10−4 M. A Smiths Detection 500DT thermal desorption ion mobility spectrometer (Danbury, CT) was used for all analyses. Generally, samples were pipetted as 2.5 ± 0.1 μL aliquots onto Nomex swabs (500DTNE, Smiths Detection). In the case of one study comparing the effects of different loading levels on resolving power, appropriate volumes ranging from 0.5 ± 0.1 to 2.5 ± 0.1 μL were deposited to attain the desired deposited mass. After sufficient time for solvent evaporation, the samples were analyzed using IMS. The instrumental conditions were thermal desorber temperature of 250 °C, inlet temperature of 265 °C, drift temperature of 237 °C, and drift flow of 300 cm3/min. During desorption temperature studies, the desorption temperature was varied from 200 to 300 °C. These instrumental conditions were set using the instrument software. Experimental verification of these values was not readily achieved. While there may be slight variability in the internal temperatures contributing to uncertainty in the data, the reproducibility of the data remained satisfactory. The instrument was set to collect 35 spectra (labeled segments by the instrument) per sample collected over 15 s. Each spectrum was a composite of 20 scans with an analysis time of 30 ms per scan. Each scan contained 1157 data points. Laboratory air was used as the drift gas. The Smiths 500DT utilizes an internal regenerative air purification system which passes the environmental air through molecular sieves to remove moisture and hydrocarbons. Additionally, the laboratory is temperature and relative humidity (RH) controlled to 20 ± 0.5 °C and 45 ± 5%, respectively. For this paper, instrument response is defined as the peak amplitude, determined by a Gaussian fit of the peak applied by the instrumental software, from the segment with the maximum response at a given drift time. The reduced mobility values are reported for the segment with the maximum response and are calculated by the instrument, against a nicotinamide internal calibrant. In all figures, error bars represent ±1 standard deviation. For instances where no error bars are visible, the error is below the size of the data point symbol. Standard deviations for instrumental response and resolving power were calculated using three to five replicate analyses. The standard deviations for reduced mobility were determined from measurements at a 250 °C desorber temperature using 40 replicates for the propyl, pentyl, hexyl, heptyl, and decyl 2653

dx.doi.org/10.1021/ac302944m | Anal. Chem. 2013, 85, 2652−2658

Analytical Chemistry

Article

compounds, 102 replicates for the octyl compound, and 150 replicates for the butyl and dodecyl compounds.

■

RESULTS AND DISCUSSION Samples of the mixed salt solution were analyzed using the 500DT to determine the reduced mobility and drift time. Figure 1 is a two-dimensional plasmagram of the homologous

Figure 2. Linearity of drift time and K0−1 for the tetraalkylammonium homologous series.

number of carbons and inverse reduced mobility.14,15 When the data presented in Figure 2 are plotted against another metric of the molecular size, namely, the number of carbon atoms in the compound, linearity is retained. These results provide evidence that the homologous series is well suited to evaluate the linearity of instrument performance over the complete range of mobilities. This could offer advantages over the typical onepoint calibration for ensuring linear performance with respect to mobility. A suitable instrument test material should provide a predictable instrument response with low day-to-day and within-day variations. Aliquots of the 3 × 10−4 M butyl, octyl, and dodecyl salt solutions were analyzed on 20 different days over two months to determine the repeatability of the daily instrument response. These data are presented in Figure 3. The relative standard deviations of the average daily

Figure 1. Average ion mobility spectrum for an equimolar mixture of the homologous series. The first peak on the left is the nicotinamide calibrant. Sequential peaks arise from each of the studied compounds from the homologous series and are marked with the number of carbons in each side chain. The inset is the generalized molecular structure of the tetraalkylammonium salts.

series of ammonium salts averaged over the complete desorption time, while Table 1 presents the reduced mobilities Table 1. Reduced Mobility Values for Tetraalkylammonium Salts ammonium salt

molecular weight

nicotinamide tetrapropyl tetrabutyl

122.12 186.36 242.47

tetrapentyl tetrahexyl tetraheptyl tetraoctyl tetradecyl tetradodecyl

298.57 354.68 410.79 466.89 579.11 691.32

lit. K0a (cm2/(V s)) 5

11

1.56, 1.47 1.33,5 1.26,11 1.31,9 1.19,10 1.26,4 1.40,12 1.15,5 1.10,11 1.12,9 1.0310 1.02,5 0.97,11 0.999 0.92,5 0.88,11 0.89,9 0.8110 0.84,5 0.80,11 0.809 0.73,5 0.7011 0.67,5 0.6411

K0 (cm2/(V s)) 1.8600b 1.627 ± 0.002c 1.377 ± 0.007 1.195 1.057 0.950 0.864 0.736 0.646

± ± ± ± ± ±

0.007 0.006 0.005 0.005 0.004 0.003

a

Absolute (calculated) reduced mobility measured in nitrogen by ESIIMS. bNicotinamide value used by instrument for calibration purposes. c Uncertainty represents 1 standard deviation.

measured in this study compared with those found in the literature for ESI-IMS. For all the salts, the reduced mobility is in reasonable agreement with the published data, although there is some variability in the published values for the tetrabutyl-, tetrapentyl-, and tetraheptylammonium salts. Additionally, the salts show a strong linear trend versus cation mass in both drift time and K0−1 over the complete series as presented in Figure 2. Here the data are plotted versus cation mass, although mobility is determined not by mass but by collisional cross section. Previously, the molecular shapes for these compounds have been modeled as spherical,13 and thus, there is a direct relation between mobility and mass. Previous work has shown that, for primary and secondary alcohols, aldehydes, and ketones, there is a linear relationship between

Figure 3. Interday comparison of instrument response for three tetraalkylammonium salts.

instrument response measurements were 3.7%, 3.7%, and 2.6% for the tetrapropyl-, tetraoctyl-, and tetradodecylammonium salts, suggesting a low day-to-day variability. Additionally, reasonable within-day precision is achieved, with relative standard deviations within a single day and for single compounds ranging from 0.38% to 11.3%, with half of the daily measurements producing relative standard deviations below 3%. 2654

dx.doi.org/10.1021/ac302944m | Anal. Chem. 2013, 85, 2652−2658

Analytical Chemistry

Article

higher masses are not generating ions as efficiently, the generated ions are not as effectively introduced to the drift tube, or losses may be occurring during the analysis process. For the equimolar mixture, the pattern of instrumental response exhibited by the equimolar mixture and presented in Figure 4B is similar to that seen in earlier studies of the salts by ESI-IMS.5 In that study, the pattern for an equimolar mixture shows a high instrumental response for the propyl, butyl, and pentyl compounds, with a large reduction in response for the hexyl compound, after which the response gradually diminished.5 Although in this case the peak response occurs at the butyl compound, the overall pattern is similar. The difference in response profiles between the responses of the instrument to individual compounds and an equimolar mixture of the homologous series may partially be explained by matrix effects. In the report by Chen et. al, it was suggested that favorable matrix effects may exist in atmospheric pressure thermal desorption ionization.6 They suggest the matrix may limit thermal decomposition. It is possible that some complex interaction of the salts may be modifying the desorption behavior in this case as well. Additionally, as shown in Figure 5, there is a strong temporal variation in signal during desorption. Differences between the temporal desorption profile of individual compounds and the equimolar mixture analysis are again seen. For the individual compounds, at all temperatures, the propyl and butyl

To determine the instrumental response to the homologous series, aliquots of individual salts and the equimolar mixture of the salts were analyzed. For the equimolar mixture, the individual amount of a given salt was 0.75 nmol; for the individual compounds, 0.60 nmol of individual salt was analyzed. As presented in Figure 4, there is a difference not only in the intensity but also in the profile of the instrument response between the analyses of the equimolar mixture and individual compounds.

Figure 4. Instrument response for the homologous series at three desorber temperatures. (A) is the instrument response to 0.60 nmol of an individual compound, while (B) is the instrument response to an equimolar mixture of the series. The aliquot contains 0.75 nmol of each salt.

For the individual compounds, the general trend of decreasing response as a function of mass may be due to a number of considerations. The results may suggest a decreasing thermal desorption yield with increasing mass. As can be seen in Figure 4A, the low-mass compounds experience the largest increases in instrument response at higher temperatures. While still significant, the high-mass compounds, however, display a lower change in instrument response with temperature. Losses due to fragmentation should also be considered. However, with the exception of a peak occasionally present in the tetrabutyl and tetrapentyl samples, the mobility spectra contain no evidence of fragmentation. Fragmentation may be occurring, but does not produce species that are detected by the ion mobility spectrometer. Finally, ion losses may be significant. Mass-selective losses to the transfer lines due to competitive absorption may cause a reduction in signal. Also, at the charge density expected (see the Supporting Information) for the instrument response levels presented in Figure 4A, spacecharge effects can be an important consideration. Because of the repulsion due to space charge and the comparatively longer time higher mass species spend in the drift tube, losses due to recombination or to the drift tube walls may reduce the overall signal intensity. These experiments provide confirmatory evidence that, when compared to the low-mass salts, salts of

Figure 5. Temporal profile of thermal desorption for the homologous series. The color chart represents the number of carbons in each side chain. (a), (b), and (c) are the profiles for compounds analyzed individually at 200, 250, and 300 °C, respectively. (d), (e), and (f) are the profiles for compounds analyzed as part of the equimolar mixture at 200, 250, and 300 °C, respectively. 2655

dx.doi.org/10.1021/ac302944m | Anal. Chem. 2013, 85, 2652−2658

Analytical Chemistry

Article

compounds remain the first compounds to desorb. In the individual compounds at low temperature there is an initial sharp peak, followed by a broader peak. As the temperature is increased, this broad peak shifts earlier in the profile, reducing the first peak to a slight shoulder. This shift may be due to a more rapid increase of desorber temperature at higher temperature values, causing the higher mass compounds to enter the gas phase more quickly than at lower temperatures. In the equimolar mixture the sharp initial peak is also present, but the second, broad peak is only seen at 300 °C. The larger compounds exhibit a slower, broader desorption and thus lower overall intensities at peak thermal desorption. Integrating the peaks under the curve over the total time of desorption does not provide a more accurate representation, because the desorption profile has not been completed in the 14 s desorption time. Due to a software limitation of acquisition time, obtaining the complete desorption profile was not possible. This temporal variation makes interpretation of instrument response complex, and a fundamental understanding of the behavior of these compounds requires further experimentation and is beyond the scope of this study. Such studies would need to consider the impact of temperature ramping of the sample desorber on the profile and would elucidate the differences between the individual compounds and equimolar mixture. Understanding the role of these differences is an important component of developing a daily use test material that would be analyzed to determine proper instrumental response. As explained in the introduction such a test material would likely be comprised of a mixture of the compounds in the homologous series to provide validation for the complete range of drift times accessible to the instrument prior to analysis of samples. While a fundamental explanation of the behavior is not available at this time, the behavior is reproducible and suggests that the homologous series can be used as a test material. An additional consideration in the difference between the profiles of the individual compounds and the equimolar mixture is the total amount of material analyzed. Figure 6 presents the instrumental response for the comparison of individual salts and the equimolar mixture at approximately equimolar amounts of each salt deposited as well as approximately equal moles of

total deposited salt. The pattern of instrumental response suggests that, regardless of the amount of salt deposited, the profile of the individual compounds is different from that of the equimolar mixtures. This is suggestive that the profile of instrument response is determined by the matrix rather than by the total amount of salt analyzed. Resolving power is another important parameter in IMS instrumentation performance, as it determines the ability of the instrument to resolve interferences. Therefore, the resolving power of the instrument for each of the tetraalkylammonium salts was also determined and was calculated as the ratio of the drift time to the peak width (full width at half-maximum, fwhm). The resolving power for three desorption temperatures is presented in Figure 7. A trend toward higher resolving power at high mass, particularly at elevated desorption temperatures, is evident.

Figure 7. Resolving power as a function of mass at three desorption temperatures for both individual compounds and the equimolar mixture at the peak instrumental response. All compounds are present at 0.75 nmol.

A complete discussion of the factors involved in IMS resolution is not the focus of this work; a more detailed discussion can be found in refs 13−20. In brief, the peak width in IMS has been established as being derived from four components, namely, peak broadening via diffusion, the gate pulse width, Coulombic expansion due to space charge, and reactions in the drift tube causing conversion of ions into species of different mobilities.16−19 Under normal conditions, the first two factors are the significant drivers of peak width,16−22 while space-charge effects may play a small but measurable role.21−23 Xu et al. have shown that the resolution due to the initial gate width is correlated indirectly with mobility, while the resolution due to diffusion is unaffected by the analyte.16 Additionally, it can be determined that, as the analyte molecule increases in size, and therefore has a lower mobility, the resolving power will increase. The data presented in Figure 7 follow this general trend for the individual compounds. In contrast, no significant differences are seen in the resolution of the equimolar mixtures, in relation to either molecular size or temperature. The differences in resolving power for the individual compounds seen with variations in desorption temperature suggest it is possible that there are space-charge effects introduced (see the Supporting Information for a more

Figure 6. Comparison of instrument response to individual compounds and the equimolar mixture at equal aliquots for each salt and for the total analysis. 2656

dx.doi.org/10.1021/ac302944m | Anal. Chem. 2013, 85, 2652−2658

Analytical Chemistry

Article

complete description). At elevated temperatures, the lower mass species have an increased instrumental response for the individual compounds. This suggests an increased number of ions are introduced to the instrument. For the equimolar mixture, however, the instrumental response remains low regardless of the temperature or molecular size. This lower response suggests that space charge would have a smaller effect on these species, explaining the uniform resolution for the equimolar mixture. The individual compounds see a lower resolution for higher temperature and for smaller compounds because of the increased response, leading to space-charge effects. As further evidence for the potential effects of space charge, the impact of increased instrumental response is illustrated in Figure 8, from which it can be determined that at low sample

Figure 9. Resolving power by desorption time for selected compounds at three desorption temperatures. (a) presents the resolving power for individual compounds, and (b) presents the resolving power for the equimolar mixture.

Figure 8. Resolving power for three compounds at different sample loading masses.

understood that this level of charge density is reached (see the Supporting Information).

■

loading the resolving power increases for a given compound at a single temperature, while the instrumental response increases. This again can be related to the space-charge effect since at low loading levels fewer ions will enter the drift region, resulting in lower space-charge effects and therefore a higher resolving power. In addition, differences in resolving power are seen over the course of instrument desorption time. Figure 9 presents the resolving power over the course of the entire desorption time. Comparing these results with the desorption profiles presented in Figure 5, it is possible to determine that there is a correlation of resolving power with instrument response when the instrument response is high. For the individual compounds at 200 °C and the equimolar mixture, there is generally little change in resolution and instrument response across the entire range of desorption time. For the individual compounds at higher temperatures, the resolution has an initial drop in resolving power, corresponding to the initial peak in instrument response, followed by a gradual increase in resolving power, corresponding to a gradually diminishing instrument response. This relationship between instrument response and resolving power again suggests a possible contribution of space charge to the resolving power in these cases. In cases of high charge density, it has been shown that charge repulsion causes the ion clouds to expand, resulting in ion losses and lower resolving power.24 In the case of the individual compounds, it is

CONCLUSIONS Tetraalkylammonium salts have been tested in a thermal desorption ion mobility spectrometer as a wide-mobility-range calibrant and for the determination of instrument performance characterization. The homologous series has reduced mobility values similar to those found in ESI-IMS systems. Because of the previously published literature that shows little sensitivity to drift gas contamination and because the introduction of the thermal desorption does not introduce variability in reduced mobility or signal intensity, these salts have the potential to be used as instrument test materials for thermal desorption IMS units, particularly those deployed for the detection of illicit drugs.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: (859) 238-6066. E-mail: leonard.demoranville@ centre.edu. Present Address §

Centre College, 600 W. Walnut St., Danville, KY 40422.

2657

dx.doi.org/10.1021/ac302944m | Anal. Chem. 2013, 85, 2652−2658

Analytical Chemistry

Article

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Drs. R. DeBono and H. Zaleski for helpful conversations and information. L.T.D. recognizes the National Research Council for funding his postdoctoral fellowship. Certain commercial equipment, instruments, or materials are identified in this paper to adequately specify the experimental procedure. Such identification does not imply recommendation nor endorsement of the National Institute of Standards and Technology, nor does it imply that the materials or equipment identified is necessarily the best available for the purpose.

■

REFERENCES

(1) Fernández-Maestre, R.; Harden, C. S.; Ewing, R. G.; Crawford, C. L.; Hill, H. H. Analyst 2010, 135, 1433−1442. (2) Kaur-Atwal, G.; O’Connor, G.; Aksenov, A. A.; Bocos-Bintintan, V.; Paul Thomas, C.; Creaser, C. S. Int. J. Ion Mobility Spectrom. 2009, 12, 1−14. (3) Eiceman, G. A.; Nazarov, E. G.; Stone, J. A. Anal. Chim. Acta 2003, 493, 185−194. (4) Jafari, M. T. Talanta 2009, 77, 1632−1639. (5) Viidanoja, J.; Sysoev, A.; Adamov, A.; Kotiaho, T. Rapid Commun. Mass Spectrom. 2005, 19, 3051−3055. (6) Chen, H.; Ouyang, Z.; Cooks, R. G. Angew. Chem., Int. Ed. 2006, 45, 3656−3660. (7) Stoll, R.; Röllgen, F. W. J. Chem. Soc., Chem. Commun. 1980, 789−789. (8) Jafari, M. T.; Azimi, M.; Borhani, K. Proc. Fourth Int. Iran Russia Conf. 2004, 832−838. (9) Guevremont, R.; Siu, K. W. M.; Wang, J.; Ding, L. Anal. Chem. 1997, 69, 3959−3965. (10) Wittmer, D.; Chen, Y. H.; Luckenbill, B. K.; Hill, H. H., Jr. Anal. Chem. 1994, 66, 2348−2355. (11) Adamov, A.; Mauriala, T.; Teplov, V.; Laakia, J.; Pedersen, C. S.; Kotiaho, T.; Sysoev, A. A. Int. J. Mass Spectrom. 2010, 298, 24−29. (12) Hallen, R.; Shumate, C.; Siems, W.; Tsuda, T.; Hill, H., Jr. J. Chromatogr. 1989, 480, 233−245. (13) Bennett, J.; Gillen, G. J. Am. Soc. Mass Spectrom. 1993, 4, 930− 937. (14) Hariharan, C. B.; Baumbach, J. I.; Vautz, W. Anal. Chem. 2010, 82, 427−431. (15) Hariharan, C.; Ingo Baumbach, J.; Vautz, W. Int. J. Ion Mobility Spectrom. 2009, 12, 59−63. (16) Xu, J.; Whitten, W. B.; Ramsey, J. M. Anal. Chem. 2000, 72, 5787−5791. (17) Watts, P.; Wilders, A. Int. J. Mass Spectrom. Ion Processes 1992, 112, 179−190. (18) Rokushika, S.; Hatano, H.; Baim, M. A.; Hill, H. H. Anal. Chem. 1985, 57, 1902−1907. (19) Kanu, A. B.; Gribb, M. M.; Hill, H. H. Anal. Chem. 2008, 80, 6610−6619. (20) Spangler, G. E. Int. J. Mass Spectrom. 2002, 220, 399−418. (21) Spangler, G. E. Anal. Chem. 1992, 64, 1312. (22) Spangler, G. E.; Collins, C. I. Anal. Chem. 1975, 47, 403−407. (23) Siems, W. F.; Wu, C.; Tarver, E. E.; Hill, H. H. J.; Larsen, P. R.; McMinn, D. G. Anal. Chem. 1994, 66, 4195−4201. (24) Mariano, A. V.; Su, W.; Guharay, S. K. Anal. Chem. 2009, 81, 3385−3391.

2658

dx.doi.org/10.1021/ac302944m | Anal. Chem. 2013, 85, 2652−2658