Negative photoionization chloride ion attachment ion mobility spectrometry for the detection of organic acids

Abstract

A novel negative photoionization chloride ion attachment ion mobility spectrometry (NP-CA-IMS) instrument has been developed. In this technique, chloride ion attachment technology is first applied to photoionization ion mobility spectrometry in a negative detection mode. The reactant ions are chloride ions, which are generated from the reaction between carbon tetrachloride (CCl₄) and low energy electrons induced by the photoionization of acetone (CH₃COCH₃) with a commercial vacuum ultraviolet Krypton lamp. The generation efficiency of chloride ions was investigated. Subsequently, the performance of the instrument was investigated. A series of volatile straight chain organic carboxylic acids were detected in the order of magnitude of ppb, including acetic acid (CH₃COOH), propionate acid (CH₃CH₂COOH), butyric acid (CH₃(CH₂)₂COOH), valeric acid (CH₃(CH₂)₃COOH), isovaleric acid (2-CH₃(CH₂)₃COOH), hexanoic acid (CH₃(CH₂)₄COOH), heptanoic acid (CH₃(CH₂)₅COOH), and octanoic acid (CH₃(CH₂)₆COOH). Besides, the concentration of acetic acid was calibrated and a mixture of the investigated acids was also studied. Finally, the new method's capability to detect acetic acid in five different brands of edible white vinegar was evaluated. The experimental results show that negative photoionization chloride ion attachment is an excellent nonradioactive source for IMS.

---

1. Introduction

Ion mobility spectrometry (IMS) is a sensitive atmospheric pressure analysis technique for separating and detecting ions which are formed from molecules in a sample based on their mobility. The technology has been used to detect many trace compounds such as chemical warfare agents,^1,2 illegal drugs,^3,4 explosives,^5,6 and various other chemicals^7,8 due to its high sensitivity, low cost, analytical flexibility, and real time monitoring capability.

The ionization source plays a key role in any IMS instrument. Most of the ion mobility spectrometers which are used for explosive and chemical agent detection use a radioactive source such as ⁶³Ni,^9,10 which is favored due to its simplicity, stability, and convenience without the necessity for an external power supply. Unfortunately, the radioactive source has serious disadvantages including a limited linear range, inflexible selectivity, and the regulatory requirements associated with radioactive material. Recently, non-radioactive ion sources, such as ultraviolet radiation (UV) photoionization source,^11,12 corona discharge,^13,14 electrospray ionization,^15,16 surface ionization,^17,18 laser ionization,^19,20 and X-ray ionization²¹ have been developed.

UV photoionization is a primary non-radioactive ion source for ion mobility spectrometry and is attracting more and more attention. UV photoionization IMS has both a positive detection mode and a negative detection mode. In positive detection mode, the analytes are ionized either by UV lamp or by proton transfer.²² It has been successfully used for the detection of a wide variety of compounds such as terpenes,²³ alcohols,²⁴ and some other organic compounds.^25–28 In negative detection mode, the negative ions of analytes are usually formed by electron capture. The electrons can be obtained by the photoelectric effect^29,30 or with the help of suitable dopant²² with ionization energy lower than the resonance energy of UV lamp photons. However, due to the limited selectivity of electrons, its application to the detection of compounds is limited. Therefore, it is very important to develop a variety of negative reaction ions for photoionization IMS.

Chloride ion attachment is a technology which uses chloride ions as reactant ions to ionize sample molecules to get characteristic peaks. Initially, the technology was successfully used in mass spectrometry³¹ and has the advantages of good selectivity and sensitivity for certain substances. For example, it is very hard to ionize polyethylene by ordinary soft ionization. However, low molecular weight polyethylene can successfully be ionized and detected by mass spectrometry combined with the chloride ion attachment technique.³² IMS was combined with chloride attachment technique firstly in 1984 and the combination enhanced greatly the IMS detection capability. Proctor et al. investigated the ion/molecule reactions between nitrobenzene, ethylene glycol, and halide ions using an ion mobility spectrometer with ⁶³Ni radioactive source. They found that halide ions have greater selectivity of ionization and this could lead to a better sensitivity of particular compounds.³³ Similarly, Lawrence et al. detected ethylene glycol dinitrate (EGDN) by ⁶³Ni radioactive source ion mobility spectrometry using chloride reagent ions and reached the detection limit of 30 ppt for EGDN.³⁴ In addition, chloride ions can also be used to detect straight chain organic carboxylic acids through the formation of strong hydrogen bonds.³¹ Therefore, it is necessary to apply chloride ions as reactant ions in photoionization IMS.

The work reported here is the first time chloride ion attachment has been applied to photoionization ion mobility spectrometry. The reactant chloride ions were generated from the reaction between carbon tetrachloride and low energy electrons produced by the photoionization of acetone with a commercial vacuum ultraviolet Krypton lamp. To confirm the feasibility and validity of the instrument, eight volatile straight chain organic carboxylic acids and the acetic acid in vinegar were detected. Results indicated that negative photoionization chloride ion attachment is a very promising nonradioactive source for IMS.

2. Experimental

2.1. Equipment

The negative photoionization chloride ion attachment ion mobility spectrometer (NP-CA-IMS) used in this work was constructed in our laboratory at the Hefei Institute of Physical Science Chinese Academy of Sciences. A schematic diagram of the spectrometer is shown in Fig. 1. The apparatus mainly consists of two parts: the IMS cell and the detection system. The whole IMS cell consists of twelve 8 mm thick metal guards which are insulated from each other using 2 mm Teflon rings. The inner diameter of the metal guards and Teflon rings is 2.5 cm. The whole IMS cell is divided into four parts: ionization region, curtain region, reactant region, and drift region. The UV Kr lamp with 10.6 eV is installed coaxially with the IMS cell and is located at the top of the ionization region. The dopant vapor which is introduced by carrier gas 1 is ionized by UV Kr lamp in the ionization region and produces many electrons. Downstream of ionization region, there is a curtain region with 8 mm space. The curtain region consists of two electrodes, both with a 1.5 mm diameter hole, which are mounted perpendicularly to the UV Kr lamp. The curtain gas is introduced into the curtain region to prevent the sample gas from diffusing into the ionization region. The ion shutter is a Bradbury–Nielson type with a 150 μs pulse width. The electric fields used in the reaction and drift regions were 550 V cm⁻¹.

Fig. 1 Schematic diagram of the negative photoionization chloride ion attachment ion mobility spectrometer.

The current signal which is generated by ions and electrons is collected by a Faraday plate located at the end of the drift tube and amplified and converted to voltage signal by a gain of 10⁸ V A⁻¹, then fed into the computer data processing system.

2.2. The experimental procedure

Acetone was selected as dopant gas in this work. In the ionization region, acetone vapor gas was taken into ionization region by carrier gas 1, and then was ionized by a UV Kr lamp. The ionization energy of acetone is 9.70 eV. Therefore, the electrons produced during the photoionization of acetone had low initial energies around 0.9 eV and were extracted into reaction region by electric fields U₁ (800 V) and U₂ (1000 V). In the reaction region, organic acid vapors and carbon tetrachloride were taken into this region by carrier gas 2. Carbon tetrachloride easily captures these low energy electrons with its high electron affinity of 0.80 ± 0.34 eV.³⁵ The chloride ions resulting from the dissociative electron attachment reaction were the reactant ions that were used to detect organic acids in the vapor sample. The detailed mechanism for the formation of acid adducts containing chloride ions in the presence of acetone is shown in Scheme 1.

Scheme 1 Mechanism for the formation of acid adducts containing chloride ions in the presence of acetone.

The carrier, curtain, and drift gases used in this work were highly pure nitrogen N₂ gas with a purity of 99.9995%. In order to remove the impurities such as water vapor and other organic pollutants in N₂, an activated carbon trap and 13× molecular sieves were utilized downstream of the N₂ gas cylinder. CCl₄, CH₃COCH₃, CH₃COOH, CH₃CH₂COOH, CH₃(CH₂)₂COOH, CH₃(CH₂)₃COOH, 2-CH₃(CH₂)₃COOH, CH₃(CH₂)₄COOH, CH₃(CH₂)₅COOH, and CH₃(CH₂)₆COOH samples (Aladdin Co., Shanghai) were used as chemical reagents. The purity of most reagents was higher than 99% except for CH₃(CH₂)₅COOH with a purity of >98%. All chemical reagents were directly used without further purification. Their saturation vapors were prepared by 10 ml gas tight syringes and diluted by the buffer nitrogen gas flow. The gas concentrations of these organic vapors were determined and controlled by the syringe pump. The flow rates of carrier gas 1, curtain gas, carrier gas 2, and drift gases were about 100, 150, 100, and 668 ml min⁻¹, respectively. All these experiments were performed under the condition of 295 K and ambient pressure.

3. Results and discussion

3.1. Generation efficiency of the chloride ions

The detection sensitivity of the method mostly depended on the intensity of the chloride ions. In order to get sufficient chloride ions, it is necessary to make sure the dopant acetone vapors are ionized with high efficiency. Considering that the photon flux of the UV lamp is constant, the number of the electrons which is generated from acetone will have a linear relationship with the concentration of acetone according to formula (1), and then reach saturation when the photons of the UV lamp are exhausted. The intensity of electrons generated from acetone was recorded when the different concentration of acetone vapors were injected into ionization region and ionized by UV Kr lamp. Fig. 2 illustrates the relationship between the electron intensity and acetone concentration. When the acetone concentration began to increase, the electron intensity increased linearly because the number of photons provided by the UV lamp was more than enough for the photoionization. When the concentration of the acetone was around 12 ppm, the electron intensity reached saturation. Saturation means that photon supplied from the UV lamp was almost exhausted and the number of the electrons reached the maximum. In the following experiments, the concentration of the acetone was kept at 12 ppm. Under this condition, the photoionization reaction between photons of the UV lamp and organic acids in the reaction region would not occur because all photons have been exhausted in the ionization region.

Fig. 2 The relationship between electron intensity and acetone concentration.

The low energy electrons generated from the photoionization of acetone in the ionization region were extracted into the reaction region by an electric field. These electrons were captured by carbon tetrachloride molecules. The chloride ions were formed by the dissociation reaction according to formula (2). The reasons for choosing carbon tetrachloride were as follows: (1) the carbon tetrachloride molecule contains four target chlorine atoms, and more chlorine atoms can increase the reaction probability with electrons. This means that carbon tetrachloride has a relatively large electron attachment rate constant and can react with electrons more easily. (2) Chloride polymerization is difficult, which ensures the stability of the intensity of the chloride ions. These two aspects determine the efficiency of the dissociative electron attachment process at atmospheric pressure.³⁶

The intensity of electrons and chloride ions changed with the increase of the concentration of carbon tetrachloride as shown in Fig. 3. The initial intensity of the electrons was about 45 nA without carbon tetrachloride being injected into the reaction region and then decreased sharply to ∼11 nA due to formula (2) while carbon tetrachloride was injected into the reaction region. However, the signal of chloride ions reached maximum very quickly and then saturation occurred when the intensity of the electrons reached to ∼0.3 nA. The phenomenon is very similar to the results reported by Daum et al.³⁶ Those authors concluded that the carbon tetrachloride was very efficient at the formation of Cl⁻ ions and was near saturation with a low concentration of carbon tetrachloride because it has a very high electron attachment rate constant of 3.7 × 10⁻⁷ cm³ per molecule per s. After the saturation of Cl⁻ ions, the excess carbon tetrachloride did not form additional reaction ions, and only served to increase the vapor concentration of carbon tetrachloride. Those excess carbon tetrachloride vapors could diffuse into the ionization region and the drift region after the chloride ions reached saturation. The reaction between the molecules of carbon tetrachloride and electrons could occur in the ionization region and lead to the decrease of the number of electrons in the ionization region. This means that the number of chloride ions which were generated in the reaction region would decrease. Moreover, the chloride generated in the ionization region was difficult to traverse the curtain region due to the small hole on the electrodes. That might be the reason why the intensity of chloride ions slightly decreased after its saturation with the decrease of the electron intensity. The intensity of chloride ions was almost constant when all electrons were depleted under the concentration of carbon tetrachloride around 800 ppb. As shown in Fig. 3, there is a very slight change in the range of the intensity of chloride ions with the variation of carbon tetrachloride's concentration. The 200 ppb was selected as the optimal concentration of carbon tetrachloride in the following experiments because the intensity of chloride ions is the maximum under the situation.

Fig. 3 The intensity of electrons and chloride ions change in the relationship with the concentration of CCl₄.

The clustering between neutral reagent molecules and chloride ions will affect the stability of the chloride ion intensity and reduce the effectiveness of the detection. It was reported that carbon tetrachloride has a low exothermic clustering reaction.³⁷ The higher and lower concentrations of carbon tetrachloride were used to verify the clustering possibility of the excess neutral regent molecules. Fig. 4 shows the ion mobility spectra using lower (130 ppb) and higher (1300 ppb) carbon tetrachloride concentrations. Only one product was observed under two different concentrations. The results show that the clustering between neutral carbon tetrachloride molecules and chloride ions was lower in this work. The one product might be hydrated chloride ions, as is explained in detail below.

Fig. 4 Ion mobility spectra of CCl₄ with concentration of 130 ppb and 1300 ppb.

The reduced mobility of the product ion in this work was calculated with 2.22 cm² V⁻¹ s⁻¹ according to the drift time of 5.98 ms. This value agrees with our previous work,^13,29 whereas it is far smaller than the reported reduced mobility of chloride ion as shown in Table 1. Lawrence et al.³⁸ studied the reduced mobility of chloride ion at different temperatures and reported the reduced mobility of chloride ion was 2.8 cm² V⁻¹ s⁻¹ and 2.93 cm² V⁻¹ s⁻¹ at 398 K and 443 K, respectively. Tabrizchi et al.³⁹ reported the reduced mobility of chloride ion was 2.99 cm² V⁻¹ s⁻¹ at 433 K.

Table 1 Reduced mobility K₀ of chloride ions

T (K)	373 (ref. 38)	398 (ref. 38)	423 (ref. 38)	433 (ref. 39)	443 (ref. 38)	463 (ref. 38)	295 (this work)
K0 (cm^2 V^−1 s^−1)	2.80	2.89	2.90	2.99	2.93	2.98	2.22

---

Generally, the hydrated chloride ions Cl⁻(H₂O)_n (where n is the number of water molecules) formed through adduct formation with water clusters and the chloride ions are common in ion mobility spectrometry instruments. The number of water molecules n depends on the temperature and water content.⁴⁰ Although an activated carbon trap and 13× molecular sieves were utilized to try to remove the water vapor and organic impurities in N₂ gas in the experiment, the water vapor cannot be eliminated completely. Additionally, there is also a small amount of water vapor existing in carbon tetrachloride sample. These water molecules clustered to the chloride ions and formed hydrated chloride ions Cl⁻(H₂O)_n at room temperature. However, the temperature in the IMS tube was higher than 398 K in the experiments of Lawrence et al. and Tabrizchi et al. The water molecules were vaporized under the high temperature. Therefore the chloride ions in Lawrence et al.'s and Tabrizchi et al.'s reports should be “bare” and have higher mobility in comparison with our data due to the lower mass weight and smaller collision cross section.

3.2. Determination of organic acids

The reaction between low energy electrons and organic acids, such as formic acid and acetic acid has been investigated intensively.^41–47 The low energy electron attachment to formic and acetic acid were studied for the first time by mass spectrometry in 2003.⁴⁶ Those early results showed that the nine negative product ions were formed in the energy range of 0–13 eV in the attachment reaction between acetic acid and low energy electrons. The first product ion (HCOOH⁻) was found when the electron energy is 0.65 eV and the pressure in the apparatus was 10⁻⁶ mbar. In this study, the initial energy of the electrons produced from the photoionization of acetone is 0.9 eV. Considering that experiment was executed under atmospheric pressure, the mean free path of the electrons is shorter and the electron energy can decay quickly due to collisions between the electron and molecules in the ambient atmosphere. The energy of most of the electrons was too low to attach to organic acid molecules at atmospheric pressure. In this work, the electron intensity with and without organic acids was recorded as shown in Fig. 5. It can be seen in Fig. 5 that the absolute electron intensity was unchanged even if a mixture of eight organic acids was injected into the reaction region, the mixture including acetic acid, propionate acid, butyric acid valeric acid, isovaleric acid, hexanoic acid, heptanoic acid, and octanoic acid. Also no other peaks appeared. We believe that energy of most of the electrons was lower than 0.65 eV. Electrons under those conditions were unable to react with the acid sample. The photoionization IMS with dopant, such as acetone in the negative mode, is unable to detect organic acid samples. However, chloride ions proved to be a good reaction ion to react with these organic acids.³¹ As for the organic acid M–COOH, the main product [M–COOH + Cl(H₂O)_n]⁻ was formed due to the reaction between the hydrated chloride ions and the organic acid in the reaction region.

Fig. 5 The ion mobility spectra of electrons (the solid line) and the ion mobility spectra of the mixture of 8 kinds of organic acids when electrons are used as reactant ions (the dotted line). To avoid the overlap of these two lines, the dotted line was deliberately raised to above zero point coordinates.

In the experiment, a gas acid sample with a concentration of approximately 50 ppb was injected into the reaction region. The mobility spectra of CH₃COOH, CH₃CH₂COOH, CH₃(CH₂)₂COOH, CH₃(CH₂)₃COOH, 2-CH₃(CH₂)₃COOH, CH₃(CH₂)₄COOH, CH₃(CH₂)₅COOH, and CH₃(CH₂)₆COOH are illustrated in Fig. 6. It can be seen that, the position of the ion peak gradually moves to longer drift time with the increase of the carbon chain length. All these acids were identified clearly. The drift times of the two isomers CH₃(CH₂)₃COOH and 2-CH₃(CH₂)₃COOH were 8.1 ms and 8.06 ms, respectively. These two isomers cannot be sufficiently distinguished at present due to this insufficient resolution. The reduced mobility K₀ values of the eight acids are listed in Table 2, which is the first report of their reduced mobility.

Fig. 6 The ion mobility spectra of 50 ppb CH₃COOH, CH₃CH₂COOH, CH₃(CH₂)₂COOH, CH₃(CH₂)₃COOH, 2-CH₃(CH₂)₃COOH, CH₃(CH₂)₄COOH, CH₃(CH₂)₅COOH, and CH₃(CH₂)₆COOH.

Table 2 Drift time and reduced mobility K₀ of ions formed in eight acids

Samples	Acetic acid	Propionate acid	Butyric acid	Valeric acid	Isovaleric acid	Hexanoic acid	Heptanoic acid	Octanoic acid
Drift time (ms)	6.72	7.14	7.6	8.1	8.06	8.54	8.96	9.3
K0 (cm^2 V^−1 s^−1)	2.18	1.86	1.75	1.64	1.65	1.56	1.48	1.43

---

3.3. Limit of the detection and acids mixture determination

The response of NP-CA-IMS against the amount of acetic acid was evaluated and the result is shown in Fig. 7. The working range was 1.7–834 ppb and the corresponding linear calibration range, which is illustrated in the inset plot, was 1.7–26 ppb. The further development of the instrument will be performed in order to broaden its linear calibration range.

Fig. 7 Response curve and linear calibration curve for CH₃COOH.

The resolution of the instrument was evaluated by the detection of a multi-component volatile organic compounds mixture. Fig. 8 shows the ion mobility spectra of a mixture of CH₃COOH, CH₃CH₂COOH, CH₃(CH₂)₂COOH, CH₃(CH₂)₃COOH, CH₃(CH₂)₄COOH, CH₃(CH₂)₅COOH, and CH₃(CH₂)₆COOH. There are eight peaks in the spectra. Peak 1 is the reactant chloride ions and the other seven peaks are assigned to the seven organic acids according to their drift times.

Fig. 8 The ion mobility spectra of a mixture of 7 kinds of organic acids: CH₃COOH, CH₃CH₂COOH, CH₃(CH₂)₂COOH, CH₃(CH₂)₃COOH, CH₃(CH₂)₄COOH, CH₃(CH₂)₅COOH, and CH₃(CH₂)₆COOH.

During the experiment, the volumes of the seven organic acids were about 1–10 ml in the mixture and were roughly selected according to their saturated vapor pressure at room temperature. Organic acids with higher saturated vapor pressure appear less in the mixture. In contrast, organic acids with lower saturated vapor pressure appear more in the mixture. The headspace of the mixture was extracted and analyzed by the NP-CA-IMS instrument. The ion intensity of the certain organic acid depends on its ratio in the mixture (i.e. its concentration), its saturated vapor pressure, and its rate constant with chloride ions. The saturated vapor pressures of the seven organic acids are all known. If carbon tetrachloride is injected into drift tube with the drift gas, the rate constant k between the certain organic acid and chloride ions can be written as where i is the ion intensity of the product ions, t_p is the drift time of the product ions from the shutter to the Faraday plate, t_d is the drift time of the product ions from a point loaded at x in the drift region to the Faraday plate, t_Cl⁻(H2O)n is the drift time of is the drift time of the hydrated chloride ions, and [M] is the concentration of the organic acids in the mixture. The rate constant k can be measured by plotting against the natural logarithm of the ion intensity i.³⁷ Detailed results of this experiment will be published elsewhere.

3.4. Detection of the practical sample

Finally, the capability of NP-CA-IMS for detecting in practical samples was also investigated. Five different brands of edible white vinegar were selected in this work. The components in the headspace of the white vinegar was extracted without further diluting and injected into the reaction region. Only one product ion (peak B) was found in the measurement for each brand sample. As shown in Fig. 9, the positions of these product ions are exactly the same and agree with the position of pure CH₃COOH.

Fig. 9 The ion mobility spectra of CH₃COOH and five different brands of edible white vinegar. Peak A: Cl⁻(H₂O)_n, peak B: [CH₃COOH + Cl(H₂O)_n]⁻.

In order to further confirm the main component in the edible white vinegar, the headspace of five brands of edible white vinegar were analyzed by Proton transfer reaction mass spectrometer (PTR-MS). Fig. 10 shows the mass spectra of chemical compound in headspace. Three peaks were found at m/z 33, 43 and 61. The m/z 61 is assigned to protonated acetic acid (CH₃COOH)H⁺, and the m/z 43 is acylium ion which is produced by dehydration reaction from protonated acetic acid. These results agree with the Haase's report.⁴⁸

Fig. 10 Proton-transfer reaction mass spectra of product ions for white vinegar samples.

The results measured by PTR-MS show that the main component in edible white vinegar is acetic acid, which means that the result measured by the NP-CA-IMS is reliable. The successful detection in the edible white vinegar tests indicates the NP-CA-IMS device has practical application value for edible product analysis.

4. Conclusion

Chloride ion attachment technology combined with photoionization ion mobility spectrometry was developed for the first time. The chloride ions were formed by the reaction between carbon tetrachloride and low energy electrons produced from an acetone photoionization process. Eight pure samples of straight chain organic carboxylic acids were detected and their reduced mobility was calculated for the first time. The main component of five brands of edible white vinegar was successfully identified. The results demonstrate that the novel NP-CA-IMS apparatus is a promising and powerful device.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (21403245 and 21107112), Anhui Provincial Program for Science and Technology Development (1301042095), and Direction Program of Hefei Center of Physical Science and Technology (2012FXCX009). Authors would like to give their thanks to Ms Lanlan Li for polishing the language.

References

1. A. B. Kanu, P. E. Haigh and H. H. Hill, Anal. Chim. Acta, 2005, 553, 148.
2. W. E. Steiner, B. H. Clowers, L. M. Matz, W. F. Siems and H. H. Hill, Anal. Chem., 2002, 74, 4343.
3. A. H. Lawrence, Anal. Chem., 1989, 61, 343.
4. J. K. Lokhnauth and N. H. Snow, J. Sep. Sci., 2005, 28, 612.
5. P. Guerra, H. Lai and J. R. Almirall, J. Sep. Sci., 2008, 31, 2891.
6. T. Khayamian, M. Tabrizchi and M. T. Jafari, Talanta, 2003, 59, 327.
7. J. D. Dunn, C. M. Gryniewicz-Ruzicka, J. F. Kauffman, B. J. Westenberger and L. F. Buhse, J. Pharm. Biomed. Anal., 2011, 54, 469.
8. Z. Karpas, A. V. Guamán, D. Calvo, A. Pardo and S. Marco, Talanta, 2012, 93, 200.
9. M. A. Mäkinen, O. A. Anttalainen and M. E. T. Sillanpää, Anal. Chem., 2010, 82, 9594.
10. R. G. Ewing, D. A. Atkinson, G. A. Eiceman and G. J. Ewing, Talanta, 2001, 54, 515.
11. S. Sielemann, J. I. Baumbach, H. Schmidt and P. Pilzecker, Field Anal. Chem. Technol., 2000, 4, 157.
12. J. Laakia, A. Adamov, M. Jussila, C. S. Pedersen, A. A. Sysoev and T. Kotiaho, J. Am. Soc. Mass Spectrom., 2010, 21, 1565.
13. H. T. Feng, W. Q. Niu, H. H. Han, C. Q. Huang, H. M. Wang, J. Matuska, M. Sabo, S. Matejcik, H. H. Jiang and Y. N. Chu, Int. J. Mass Spectrom., 2011, 305, 30.
14. M. Tabrizchi and F. Rouholahnejad, Rev. Sci. Instrum., 2004, 75, 4656.
15. G. R. Asbury, J. Klasmeier and H. H. Hill, Talanta, 2000, 50, 1291.
16. C. Wu, W. F. Siems, J. Klasmeier and H. H. Hill, Anal. Chem., 2000, 72, 391.
17. U. K. Rasulev, U. Khasanov and V. V. Palitcin, J. Chromatogr. A, 2000, 896, 3.
18. C. Wu, H. H. Hill, U. K. Rasulev and E. G. Nazarov, Anal. Chem., 1999, 71, 273.
19. J. M. Headrick, T. A. Reichardt, T. B. Settersten, R. P. Bambha and D. A. V. Kliner, Appl. Opt., 2010, 49, 2204.
20. A. Tempez, M. Ugarov, T. Egan, J. A. Schultz, A. Novikov, S. Della-Negra, Y. Lebeyec, M. Pautrat, M. Caroff, V. S. Smentkowski, H. Y. J. Wang, S. N. Jackson and A. S. Woods, J. Proteome Res., 2005, 4, 540.
21. V. S. Pershenkov, A. D. Tremasov, V. V. Belyakov, A. U. Razvalyaev and V. S. Mochkin, Microelectron. Reliab., 2006, 46, 641.
22. E. G. Nazarov, R. A. Miller, G. A. Eiceman and J. A. Stone, Anal. Chem., 2006, 78, 4553.
23. W. Vautz, S. Sielemann and J. I. Baumbach, Anal. Chim. Acta, 2004, 513, 393.
24. S. Sielemann, J. I. Baumbach, H. Schmidt and P. Pilzecker, Anal. Chim. Acta, 2001, 431, 293.
25. H. Bahrami and M. Tabrizchi, Talanta, 2012, 97, 400.
26. J. I. Baumbach, S. Sielemann, Z. Y. Xie and H. Schmidt, Anal. Chem., 2003, 75, 1483.
27. H. Li, W. Q. Niu, H. M. Wang, C. Q. Huang, H. H. Jiang and Y. N. Chu, Spectrosc. Spectral Anal., 2012, 32, 29.
28. Z. Xie, S. Sielemann, H. Schmidt, F. Li and J. I. Baumbach, Anal. Bioanal. Chem., 2002, 372, 606.
29. D. S. Su, W. Q. Niu, S. Liu, C. Y. Shen, C. Q. Huang, H. M. Wang, H. H. Jiang and Y. N. Chu, Radiat. Phys. Chem., 2012, 81, 1869.
30. P. Begley, R. Corbin, B. E. Foulger and P. G. Simmonds, J. Chromatogr., 1991, 588, 239.
31. H. P. Tannenbaum, J. D. Roberts and R. C. Dougherty, Anal. Chem., 1975, 47, 49.
32. S. Kéki, L. Nagy, A. Kuki and M. Zsuga, Macromolecules, 2008, 41, 3772.
33. C. J. Proctor and J. F. J. Todd, Anal. Chem., 1984, 56, 1794.
34. A. H. Lawrence and P. Neudorfl, Anal. Chem., 1988, 60, 104.
35. P. O. Staneke, G. Groothuis, S. Ingemann and N. M. M. Nibbering, Int. J. Mass Spectrom. Ion Processes, 1995, 142, 83.
36. K. A. Daum, D. A. Atkinson and R. G. Ewing, Talanta, 2001, 55, 491.
37. M. Tabrizchi and A. Abedi, J. Phys. Chem. A, 2004, 108, 6319.
38. A. H. Lawrence, P. Neudorfl and J. A. Stone, Int. J. Mass Spectrom., 2001, 209, 185.
39. M. Tabrizchi and A. Abedi, Int. J. Mass Spectrom., 2002, 218, 75.
40. H. Borsdorf and T. Mayer, Talanta, 2012, 101, 17.
41. V. S. Prabhudesai, D. Nandi, A. H. Kelkar, R. Parajuli and E. Krishnakumar, Chem. Phys. Lett., 2005, 405, 172.
42. T. Sedlacko, R. Balog, A. Lafosse, M. Stano, S. Matejcik, R. Azria and E. Illenberger, Phys. Chem. Chem. Phys., 2005, 7, 1277.
43. A. Pelc, W. Sailer, P. Scheier and T. D. Märk, Vacuum, 2005, 78, 631.
44. A. Pelc, W. Sailer, P. Scheier, N. J. Mason and T. D. Märk, Eur. Phys. J. D, 2002, 20, 441.
45. A. Pelc, W. Sailer, P. Scheier, M. Probst, N. J. Mason, E. Illenberger and T. D. Märk, Chem. Phys. Lett., 2002, 361, 277.
46. W. Sailer, A. Pelc, M. Probst, J. Limtrakul, P. Scheier, E. Illenberger and T. D. Märk, Chem. Phys. Lett., 2003, 378, 250.
47. A. Pelc, W. Sailer, P. Scheier, N. J. Mason, E. Illenberger and T. D. Märk, Vacuum, 2003, 70, 429.
48. K. B. Haase, W. C. Keene, A. A. P. Pszenny, H. R. Mayne, R. W. Talbot and B. C. Sive, Atmos. Meas. Tech., 2012, 5, 2739.

---