Mechanisms of the electron-impact-induced glycine molecule fragmentation

Chemical Physics 404 (2012) 36–41

Contents lists available at SciVerse ScienceDirect

Chemical Physics journal homepage: www.elsevier.com/locate/chemphys

Mechanisms of the electron-impact-induced glycine molecule fragmentation J. Tamuliene a,⇑, L.G. Romanova b, V.S. Vukstich b, A.V. Snegursky b a b

Vilnius University, Institute of Theoretical Physics and Astronomy, 12 A. Goštauto str., 01108 Vilnius, Lithuania Institute of Electron Physics, Ukr. Nat. Acad. Sci., 21 Universitetska str., 88017 Uzhgorod, Ukraine

a r t i c l e

i n f o

Article history: Available online 8 February 2012

a b s t r a c t Fragmentation of the glycine (C2H5NO2) molecule by low-energy electron impact has been studied both experimentally and theoretically. The main emphasis has been given to the mechanisms of the initial molecule fragment production including formation of the doubly-charged CH2NHCO2+ fragment. A special attention has been paid to the energy characteristics of the ionic fragment yield. The geometrical parameters of the initial molecule rearrangement have also been analyzed. Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction

2. Experimental

The studies of the damages in biological systems, including amino acids, resulted from the influence of ionizing radiation, are being the hot topic of a series of investigations within last decades [1]. The majority of the above damages are not usually due to the primary high-energy radiation but results from the effect of the secondary low-energy charged particles produced in the course of ionization. Amino acids are known as the building blocks of the human body proteins responsible for forming antibodies to struggle against bacteria and viruses, constructing the hormonal system and transferring oxygen throughout the human organism. It is well-known that glycine, which is one of the non-essential amino acids, helps involving oxygen into hormone production mechanism and, thus, strengthens the immune system of live tissues [1]. The studies of amino acids in the gas phase are of great significance in biology for understanding and determining their properties. Such knowledge is fundamental to understanding the complex structure of their polymers, say, proteins and peptides. Furthermore, this information is important for a number of modern fields of biological science, e.g., for biological astrophysics and photochemistry as well as for various nano-bio-technological applications. The structures of these molecules can be accessed in different ways. Here we would like to emphasize that the complex gas-phase experimental investigations and the relevant theoretical calculations allow an unambiguous experimental data interpretation to be obtained, which, in turn, may serve to prove theoretical models. In our recent papers [2,3], we reported the experimental data on the thermal and electron-impact fragmentation of the glycine molecule. The present paper deals with the continuation of our previous studies on fragmentation of this molecule induced by the collisions with low-energy monoenergetic electrons [2,3].

The experimental apparatus and method applied are described in detail in Refs. [3,4]. The experimental technique is related to the crossed-beam method involving mass separation of the collision products by means of a modernized magnetic mass spectrometer that allows the reaction products to be selected with respect to their mass-to-charge ratio [4]. Here we would like to mention that our apparatus is capable of studying ionic fragments within the 1– 720 a.m.u. mass range with high sensitivity (1016 A) and mass resolution (m/Dm = 1100). The beam of the molecules under study was formed by an effusion source with a resistively heated oven providing the target molecule concentrations of about 1010 cm3. The operating temperature of the molecular beam source was varied up to 150 °C and controlled by a thermocouple. The experimental conditions excluded molecular cluster formation in the beam. An original three-electrode electron gun provided an electron current of 30–50 lA over the energy range under study (0–150 eV). The parent and fragment ions produced were detected after mass separation by means of an electrometer. The data acquisition and processing were controlled by a PC. The electron energy scale calibration was carried out against known ionization thresholds for an argon atom and nitrogen molecule with the accuracy not worse than ±0.1 eV [3]. The glycine molecule mass spectrum was measured at the 70 eV electron energy, the appearance energies for positively charged fragment ions were determined using technique described in Refs. [2,3] within the 5–30 eV energy range.

⇑ Corresponding author. E-mail address: [email protected] (J. Tamuliene). 0301-0104/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.chemphys.2012.01.019

3. Theoretical The structure of the molecule and its fragments has been studied by using the generalized gradient approximation for the exchange-correlation potential in the density functional theory (DFT) as it is described by the Becke’s three-parameter hybrid functional, using the non-local correlation provided by Lee, Yang, and

J. Tamuliene et al. / Chemical Physics 404 (2012) 36–41

Parr. The DFT method is commonly referred to as B3LYP [5], – a representative standard DFT method. The cc-pVTZ basis set has been used as well [6]. The structures of the molecule and its fragments under study have been optimized globally without any symmetry constraint. It is necessary to mention, that the glycine molecule has several isomers [7–9]. Unfortunately, it is impossible to determine experimentally the contribution of certain isomers to the total pattern of the parent molecule fragmentation, therefore we have theoretically investigated one of the most stable conformers that is more probable to be obtained. The bond order and bond length of the most stable isomer of the glycine molecule have been investigated to find the weaker bonds that are possibly to be destructed. Additionally, the vibration spectrum has been evaluated to predict the possible elongation of bonds and the angle change aiming to analyze the most probable fragments produced due to electron impact. In order to model the fragmentation processes, the possible fragment anions, cations and fragments with a zero charge both with and without geometry optimization have been evaluated. The dissociation energies have been calculated as the difference between the total energy of the glycine molecule and the sum of the energies of the fragments predicted. We have predicted that under our experimental conditions the structure of the fragment formed could be changed thus influencing the dissociation energy. To evaluate the above influence, the dissociation energy has been calculated for the following two cases: (i) the single point energy calculation of the fragments, taking into account the geometry of a certain part of the glycine molecule, was performed (in the cases the energy of fragments formation is not the lowest one); (ii) the structure of the glycine fragments has been optimized, i.e. the fragments were allowed to reach their equilibrium geometries and the obtained energy (the lowest energy of the fragments) has been used to calculate the dissociation energy. The GAMESS and Gaussian program packages have been applied here [10,11].

4. Results and discussion The mass-spectrum of the glycine molecule measured in our experiment is shown in Fig. 1. As it is seen, the fragments observed are due to the process of electron-impact dissociative ionization of the molecule under study. It is a very peculiar feature of the above mass-spectrum that the parent ion peak (see inset in Fig. 1) is

37

rather weak allowing one to conclude that its formation is less probable because of its decay within a short period of the collision time. This pattern is typical for a number of complex molecules not comprising conjugated bonds [12]. Nevertheless, we may state, that, despite the weak intensity of the parent molecular ion peak, the process of the dissociative ionization of the initial molecule provides a rich spectrum of the ionized fragments. The most prominent peak in the mass-spectrum (Fig. 1) is due to the formation of the parent molecule fragment with the m = 30 a.m.u. mass. Probable assignments of this peak are the CH4N+ and CH2O+ ions. As we have found earlier [3], the first fragment may involve two isomers: NH2CH2+ and CH3NH+. Formation of the CH3NH+ fragment is less probable because of its thermodynamical instability [3,13]. Regarding the CH2O+ fragment, its formation is related to the substantial energy consumption for the rearrangement of the atoms and bonds in the parent molecular ion. Thus, this process is less probable as compared to that for the above CH4N+ fragment. Formation of the most intense ion fragment with the m = 30 a.m.u. mass, i.e. the CH4N+ ion, should result from the ‘amine type’ ionization of the parent C2H5NO2 molecule and the bond rupture at the adjacent carbon atom with the subsequent loss of the neutral CHO2 radical. The next (by its intensity) peak in the mass spectrum (at m = 28 a.m.u.) is related to the formation of the CH2N+ ion fragment. As it is obvious from [3], the main component of this peak is due to the HCNH+ isomer formation. The next fragment with the m = 29 a.m.u. mass should be assigned to the CH3N+ ion. The most interesting, in our opinion, is a weak peak located at m = 28.5 a.m.u (Fig. 2). To our knowledge, we were the first to observe it experimentally. Below we shall proceed with the analysis of the mechanisms of the above fragments production. Note that in Fig. 1 it is not revealed so clearly as in Fig. 2 because of the semi-logarithmic scale used. Firstly, the bond length and the bond order of the neutral glycine molecule were analyzed (Table 1). It is seen that the O3–H9 bond is the weakest one. So, the C2 H4 NO 2 fragment could be formed due to deprotonation of the glycine molecule and this process is described in detail in Ref. [9]. We had no possibility to investigate anions experimentally, thus the main task of this study was to describe the fragmentation process that leads to the appearance of positively charged fragments, therefore formation of the C2 H4 NO 2 fragment is not described here in detail.

Fig. 1. Mass spectrum of the glycine molecule. Inset – that in the vicinity of the parent molecule peak.

38

J. Tamuliene et al. / Chemical Physics 404 (2012) 36–41

Fig. 2. The area of the mass-spectrum of the glycine molecule showing the doublycharged CH2NHCO2+ fragment production.

Table 1 The glycine molecule bond lengths and orders. The molecule image and the atom numbers are presented in Fig. 3. Bonds

Bond length, (Å)

Bond order

C1AC2 C1AN4 C1AH8 C1AH10 C2AO3 C2AO5 O3AH9 N4AH7

1.52 1.46 1.09 1.09 1.38 1.23 0.98 1.01

0.85 0.98 0.94 0.94 1.01 1.98 0.80 0.87

The second weakest bond is that between the carbon atoms. Based on these results and using the Mulliken and Lowdin population analysis, one may predict that the CH4N and CHO2 fragments should be produced via the glycine molecule dissociation. According to the early (that has become classical) work [14], this fragmentation process is related to the removal of one electron from the nitrogen lone pair resulting in the charge localization on the nitrogen atom and on the adjacent a-carbon atom. Such ‘amine type’ ionization seems to dominate over other possible ionization channels, hence the simplest analysis of the bond order and the frequencies allows us to predict the most probable fragmentation processes occurring at the CAC bond break. The processes could be as follows:

C2 H5 NO2 þ e !

8 CHO2 þ CH4 Nþ þ e; ðaÞ > > > > 0 þ > > < CHO2 þ CH4 N þ 2e; ðbÞ

CHOþ2 þ CH4 N þ e; ðcÞ > > > > CHOþ2 þ CH4 N0 þ 2e; ðdÞ > > : CHOþ2 þ CH4 Nþ þ 3e ðeÞ

ð1Þ

The calculated appearance energies for the above fragments are listed in Table 2. The appearance energy Eap was calculated as: jEap j ¼ jEglycine  ðECHO2 þ ECH4 N Þj. Here Eglycine is the total energy of the neutral glycine molecule, while ECHO2 and ECH4 N are the total energies of the CHO2 and CH4N fragments, respectively. This calculation does not take into account the activation energy of the molecular ion fragmentation. One may notice that the lowest energy is

required to divide the glycine molecule into the CHO 2 anion and the CH4N+ cation (see scheme (1)a). The positive CH4N+ ion is produced according to scheme (1)b with higher energy consumption if it is predicted that only the CAC bond of the glycine molecule is ruptured, i.e. the bond length and angles in the parent molecule and their fragments are the same. However, the appearance energies according to schemes (1)a and (1)b coincide when changeability of the geometrical structure during the fragment formation according to scheme (1)b only is taken into account. Additionally, we have observed a slight shoulder in the ionization function for this fragment, which, in our opinion, results from the contribution of a new channel of the parent molecule dissociation. This should be related both to the second ionization energy of the parent molecule and to the change of the complementary fragment charge. According to our measurements, the appearance energy for the CH4N+ fragment is 10.1 ± 0.1 eV, which is close to the calculated value (9.99 eV) for the molecule conformer under study in case of dissociation of the glycine molecule geometry with fragment change according to scheme (1)b. Taking into account the different glycine molecule conformers in the ion source of the mass-spectrometer are in the dynamical equilibrium and that the difference of their total energies is about 0.1 eV [15], one may conclude that the main dissociative channel of the parent ion with the CH4N+ ion production proceeds according to scheme (1)b. It should be noted that, according to the photoionization data [16], the appearance energy for this ion is 9.38 ± 0.05 eV. Higher ionization energy and appearance energy values for electron impact as compared to photoionization are explained usually by the influence of spin polarization and Coulomb interaction both before and after collision. For example, spin polarization reduces the ionization energy of glycine by about 0.24 eV [17]. Therefore, in the case of glycine photoionization, one has to take into account process (1)a as well, leading to the ion pair formation, which is more probable as compared to other possible processes described above. The main fragment peak CH4N+ (m = 30 a.m.u.) in the glycine molecule mass spectrum is accompanied by the satellite peaks with m = 28, 29 and 31 a.m.u., which correspond to the ionic fragments produced due to either the hydrogen atom migration/ detachment or to the main fragment dehydration. The relative intensity of the peak at m = 31 a.m.u. in our mass-spectrum is about 1.8–5.5% of the CH4N+ ion peak depending on the ionization conditions. The calculated height of the first isotope peak for this ion as the sum of the increments of the relative intensities for the atomic isotopes is 1.55% and this testifies to the fact that the observed peak is a compound one and comprises the 13CH4N+ and CH5N+ ions. Presumably, the CH5N+ ion contribution to the total peak intensity is dependent of the migration rate of the H atom from the hydroxyl group, and in our experiment it is considerably higher than that in the relevant NIST database [18]. Most probably, the structure of the CH5N+ ion is related to that of methylamine, i.e. the CH3 NHþ 2 ion. For the CH3 NHþ 2 ion, the complementary fragment corresponds to the neutral CO2 particle. In the glycine mass spectrum, the relative intensity of the peak at m = 44 a.m.u. (i.e. the COþ 2 ion) is 5.8% [18]. This pair of fragments is produced due to the hydrogen atom migration from the hydroxyl group to the carbon atom via the 4term transient state. Two alternative decay channels are possible here differing by both the reaction rate and the final charge localization:

+ O H 2N

C CH2

OH

CH3NH2+ + COOo (a) CH3NH2 o + COO+ (b)

39

J. Tamuliene et al. / Chemical Physics 404 (2012) 36–41 Table 2 Calculated appearance energies (in eV) for the CHO2 and CH4N fragments. CHO2 (m = 45 a.m.u.) charge

CH4N (m = 30 a.m.u.) charge

Glycine molecule geometry not changed*

Glycine molecule geometry changed**

1 0 1 1 1

1 1 1 0 1

10.6 12.03 16.62 15.26 23.17

8.57 9.99 16.11 14.99 21.14

*

‘Glycine molecule geometry not changed’ means that the single point energy calculation of the fragments taking into account the geometry of the certain part of the glycine molecule was performed. ** ‘Glycine molecule geometry changed’ indicates that the equilibrium geometry structure of the glycine molecule fragments is investigated.

The intensity of the corresponding peaks in the mass-spectrum may characterize the efficiency of the above reaction channels. According to our data, the intensity of the peak with m = 44 a.m.u. is higher than that of the peak at m = 31 a.m.u., thus, in case of the CAC bond dissociation accompanied with the hydroxyl group H atom migration; the cation center is mainly displaced to the CO2 fragment and the neutral CH3NH2 fragment is eliminated. The ion with the m = 28 a.m.u. mass is the second intensity-related peak in the parent molecule spectrum and may have the following gross formula: CH2N or CO (or CH2N + CO). In Ref. [19], it was suggested that this peak is due to the CH2N+and/or CO+ ions. However, the mass-spectra of the deprotonated glycine (with the HNCD2CO2H and H15NCH2CO2H composition) show that this fragment contains a nitrogen atom, while the experiment for the deuterated d5 and d3-glycine [13] has unambiguously shown that this peak belongs to the CH2N+ (CD2N+) ion. The comparison of the stability of the positively charged ions CH2N+ and CO+ (i.e. the binding energy per atom in eV), calculated by us, also allows one to conclude that formation of the CH2N+ cation is more probable than that of the CO+ cation. The structure of the CH2N+ ion depends on the parent or intermediate ion bonds being broken. Four possible relevant isomers are shown in Fig. 4. It is important that in cases when different bonds of the CH4N+ or CH3N+cation are broken the fragments become planar after the geometry optimization. Calculations [13] have shown that the HCNH structure is the most stable of those

Fig. 4. The CH2N+ cation isomers before (top) and after (bottom) geometry optimization.

presented in Fig. 4. According to our calculations, after the geometry optimization with the B3LYP cc-pVTZ approach application, the trans- and cis-isomers (III and IV in Fig. 3) transit to the most stable structure with the linear configuration (the point group Cs). Note that the C and N atoms in this case undergo the sp-hybridization. Experimentally the CH3N+ and CH2N+ cations are observed in the mass-spectrum (Fig. 1). These positively charged ions could be produced at the simultaneous break of several simple bonds:

8 ðCHO2 þ HÞ þ CH3 Nþ þ e; ðaÞ > > < C2 H5 NO2 þ e ! ðCHO2 þ HÞ0 þ CH3 Nþ þ 2e; ðbÞ > > : ðCHO2 þ HÞþ þ CH3 Nþ þ 3e; ðcÞ and

8 ðCHO2 þ 2HÞ þ CH2 Nþ þ e; ðaÞ > > < C2 H5 NO2 þ e ! ðCHO2 þ 2HÞ0 þ CH2 Nþ þ 2e; ðbÞ > > : ðCHO2 þ 2HÞþ þ CH2 Nþ þ 3e: ðcÞ

ð3Þ

Calculated energies required to produce the below fragments are presented in Tables 3 and 4. It is obvious that dissociation of the CAC bond with simultaneous detachment of one/two hydrogen atom(s) from the parent ion and production of the positive CH2N+ and CH3N+ ions require more energy than those in the case corresponding to scheme (1)b. However, as shown in Tables 3 and 4, no direct correlation between the required energy and the number of detached H atoms is observed. In case of process (2)b, formation of the CH3N+ ion occurs with minimal energy consumption. As shown in [20] H atom hopping from the amino group to the carbonyl group prior to dissociation of the CAC bond is one of the preferred dissociation channels after single photon ionization of the glycine molecule. So for the case of electron impact ionization this mechanism is the most energetically favorable when the CH3N+ fragment appeared. Our analysis of the charge distribution for the [CHO2 + H] group (Table 5) shows that the minimal energies in Table 4 correspond to the production of the CH2O2 compound. Thus, the break of the CAC bond accompanied by the H atom migration from the amino group to the carbonyl group oxygen is the most probable channel of the CH3N+ fragment formation. On the other hand, the positively charged CH3N+ and CH2N+ ions could result from the secondary dissociation due to deprotonation of the CH4N+ ions, while the CH2N+ ion could also appear due to deprotonation of the CH3N+ ion. For example, the above processes could be as follows:

( CH4 Nþ !

CH3 N !

H0 þ CH3 Nþ ;

ðaÞ

ðH þ HÞ0 þ CH2 Nþ ; ðbÞ (

þ

Fig. 3. The view of the glycine molecule.

ð2Þ

H0 þ CH2 Nþ ; 0

ðaÞ þ

ðH þ HÞ þ CHN : ðbÞ

ð4Þ

ð5Þ

The glycine molecule mass spectrum (Fig. 1) reveals a diffuse peak at about m⁄  26.1 a.m.u. corresponding to the transition 30 ? 28

40

J. Tamuliene et al. / Chemical Physics 404 (2012) 36–41

Table 3 Calculated appearance energies (in eV) for the CHO2 + H and CH3N cations/anions.

* **

CH3N (m = 29 a.m.u.) charge

CHO2 + H charge

Glycine molecule geometry not changed*

Glycine molecule geometry changed**

1 1 1

1 0 1

14.89 16.51 26.76

11.55 10.52 23.61

Table 6 Calculated energies (in eV) of the CH4N + and CH3N + ion deprotonation. ACAH bond divided

Process

CH4N+ ? H0 + CH3N+ CH3N+ ? H0+CH2N+

See explanatory notes in Table 2. See explanatory notes in Table 2. * **

ANAH bond divided

Glycine molecule geometry not changed*

Glycine molecule geometry changed**

Glycine molecule geometry not changed*

Glycine molecule geometry changed**

4.84 3.71

5.36 3.27

5.92 6.72

5.58 6.04

See explanatory notes in Table 2. See explanatory notes in Table 2.

Table 4 Calculated appearance energies (in eV) for the CHO2 + 2H and CH2N cations/anions.

* **

CH2N (m = 28 a.m.u.) charge

CHO2 + 2H charge

1 1 1

1 0 1

Glycine molecule geometry not changed*

Glycine molecule geometry changed

19.45 20.2 29.32

12.89 14.63 23.02

Table 7 Calculated energies (in eV) of the glycine molecule dissociation into the C2O2H3 and NH2 fragments.

**

See explanatory notes in Table 2. See explanatory notes in Table 2.

Table 5 Charge distribution in the [CHO2 + H] fragment group. Total charge

After optimization**

0 1 1

CH2O2 compound formed 1 CH2O2 compound formed

**

**

NH2 (m = 16 a.m.u.) charge

Glycine molecule geometry not changed*

Glycine molecule geometry changed**

1 0 1 1

1 1 1 0

15.73 16.95 14.84 14.9

14.33 16.09 12.88 12.98

See explanatory notes in Table 2. See explanatory notes in Table 2.

Before optimization* (Mulliken charge) H

CHO2

*

*

C2H3O2 (m = 59 a.m.u.) charge

0 0 1

CHO2

H

0 1 0

0

Table 8 Binding energy per atom (in eV) for the relevant doubly-charged ions.

See explanatory notes in Table 2. See explanatory notes in Table 2.

with the detachment of the neutral fragment with m = 2 a.m.u, i.e. the secondary fragmentation of the CH4N+ ion. Thus the dehydration proceeds according to scheme (4)b producing the molecular hydrogen we have observed experimentally. During the deprotonation of the CH4N+ and CH3N+ fragments, the CAH or the NAH bonds could be broken. We have checked these two cases. The obtained energy values required to deprotonate CH4N+ and CH3N+ are presented in Table 6. It should be mentioned that production of the CH3N+ ions under the CAH bond break in CH4N+ is more probable due to the latter ion stability that is higher than that of the CH3N cations with the NAH bond being broken during the deprotonation of CH4N+. Additional comparison of the appearance energy for the CH3N+ cations leads to the conclusion that the reaction with the CAH bond break during the deprotonation of CH4N+ is more relevant. However, the production of the molecular hydrogen at the CH4N+ ion dehydration occurs, evidently, with the CAH and NAH bonds break. We have calculated the energy spent for the CAN bond dissociation in the initial molecule. The possible dissociation channels are presented below:

8 C H O þ NHþ2 þ e; > > > 2 3 2 > < C H O0 þ NHþ þ 2e; 2 3 2 2 C2 H5 NO2 þ e ! > C2 H3 Oþ2 þ NH2 þ e; > > > : C2 H3 Oþ2 þ NH02 þ 2e:

ðaÞ ðbÞ ðcÞ

ð6Þ

ðdÞ

In the experimental mass spectra, the intensity of the ions produced at the above bond dissociation is 1.59% for m = 16 a.m.u. and 0% for m = 59 a.m.u. fragments. Thus, the amino group detachment from the parent molecule is less probable, despite the relatively

Ions

C 2 H3 O2þ 2 (m/ z = 29.5)

C2H3ON2+ (m/ z = 28.5)

C2H2ON2+ (m/ z = 28)

Binding energy per atom

0.50

1.63

1.015

Table 9 Calculated energies (in eV) of the neutral glycine molecule dissociation into the C2H3ON2+ and [OH + H] fragments. C2H3ON2+

[OH + H] charge

Before optimization*

After optimization**

CAH bond broken

1 0 1 1 0 1

48.4 38.87 35.33 47.96 38.44 34.9

36.8 24.55 27.0 41.13 28.89 31.34

NAH bond broken

* **

See explanatory notes in Table 2. See explanatory notes in Table 2.

low values of calculated energies required to break the CAN bond (see schemes (6)c and (6)d and Table 7). As mentioned above, the weak peak at m = 28.5 a.m.u. (Fig. 2) is of particular interest. Obviously, it could be assigned to the doublycharged C2H3ON2+ ion. The single-charged C2H3ON+ ion with m = 57 a.m.u. in the glycine mass-spectrum [18] has the intensity of 0.4%, however, in the mass spectra of other a-amino acids, the fragment with this mass is very typical together with the fragments having masses m = 30, 44 and 75 a.m.u. [14]. Moreover, the tandem mass spectrometry spectra of the deprotonated glycine marked by the D and 15N isotopes reveal the most pronounced peaks at m = 57–59 a.m.u.[11]. The replacement of two hydrogen atoms at the positions H8, H10 (Fig. 1) by deuterium indicates that the formation of a water molecule takes place at the hydrogen atom detachment from the amine group. In Ref. [21], the 7-stage fragmentation trajectory of one of the glycine conformers was calculated. It has been shown that a water molecule is formed

J. Tamuliene et al. / Chemical Physics 404 (2012) 36–41

and released within the time of 3 ps due to the transition of the hydrogen atom from the amine group to the oxygen atom of the hydroxyl group. According to the authors of this paper, the optimization of the fragments favors the formation of the neutral water molecule and the positive glycine fragment over the ionic water and the neutral glycine fragment. In accordance with Ref. [14], the loss of a water molecule by the CH2NH2COOH+ (m/z = 75) fragment is accompanied by the metastable ion production with m⁄ = 43.3 a.m.u. In our case, the diffuse peak with this mass is not observed, while the small intensity of the peak at m = 57 a.m.u. indicates the lack of fragmentation channel leading to the single-charged CH2NHCO+ ion formation. Thus, formation of the doubly-charged CH2NHCO2+ ion occurs, probably, simultaneously (during 105–107 s) with water molecule elimination, which, in turn, becomes possible after the two-electron loss by the parent molecule from the first two highest occupied molecular orbitals (i.e. non-bonding n-orbitals of the nitrogen atom and the oxygen atom of the hydroxyl group). Since we seem to be the first to observe the above doubly-charged ion in the amine acid mass spectra, we have calculated the total energy for some possible 2þ 2þ 2þ relevant ions. Thus, the CHO2þ 2 ; C2 H3 O2 ; C2 H3 ON ; C2 H2 ON ; 2þ 2þ ions were investigated. It should be C2 H3 ON ; C2 H2 ON 2þ and C2H2ON2+ ions mentioned that only the C2 H3 O2þ 2 ; C2 H3 ON are stable, although their stability is not very high (Table 8). Hence, analyzing the data of Table 8, one may conclude that in our experimental conditions the most stable ion is produced. The calculation of the energy required to produce the doubly-charged C2H3ON2+ ion from the neutral molecule with multiplicity 1 shows that the hydrogen atom detachment from the carbon atom proceeds with a higher probability than that from the nitrogen atom, and in this case the minimal energy corresponds to the neutral [OH + H] fragment group (Table 9). The results of optimization indicate that in the [OH + H]0 and [OH + H] case a water molecule should be produced. Unfortunately, weak intensity of the C2H3ON2+ ion peak did not allow us to measure the dissociative ionization function for this ion and to determine its appearance energy experimentally. 5. Conclusions We have studied the possible mechanisms of the glycine molecule fragmentation by slow monoenergetic electrons. The experimental observations of the ionized fragment formation were supported by relevant theoretical calculations, predicting possible fragmentation paths. The analysis of the theoretically calculated channels of the main ion production from the glycine molecule (CH4N+) and the experimental dissociative ionization function for this ion, measured earlier, show that the stepwise structure of this curve may be due not only to the ionization of the parent molecule molecular orbitals, but also due to the charge change of the complementary particle produced.

41

The above calculations allowed us to suggest the most energetically beneficial dissociation channels for the glycine molecule with the charge of all produced particles to be taken into account. This extends the possibilities of the commonly used qualitative approach in the electron-impact mass spectrometry that takes into account the positive ions only, while the charge of the complementary fragmentation products is not detected and is taken to be zero. Production of the doubly-charged CH2NHCO2+ fragment was observed for the first time. The assumption about the mechanism of production of this fragment by a simultaneous elimination of two electrons and a water molecule is confirmed by the calculation results according to which the least energy consumption corresponds to the C-H bond break being accompanied with the neutral [OH + H] fragment yield. Acknowledgments The authors would like to thank Prof. J. Bojarski and Prof. M. Cegla from Jagellonian University (Krakow, Poland), Prof. P.D. Burrow from Lincoln University (Nebraska, USA) and our colleagues Prof. A. Imre and V. Patasiene for their assistance and fruitful discussions. Special thanks to InSpire and NGI.LT projects for the resources and technical support provided as well as COST MP0802 activity. References [1] S. Cristoni, L.R. Bernardi, Mass. Spectr. Rev. 22 (2003) 369. [2] V.S. Vukstich, A.I. Imre, A.V. Snegursky, Tech. Phys. Lett. 15 (2009) 1071. [3] V.S. Vukstich, A.I. Imre, L.G. Romanova, A.V. Snegursky, J. Phys. B At. Mol. Phys. 43 (2010) 185208. [4] V.S. Vukstich, A.I. Imre, A.V. Snegursky, Instr. Exp. Technol. 54 (2011) 66. [5] A.D. Becke, J. Chem. Phys. 98 (1993) 5648. [6] R.A. Kendall, T.H. Dunning Jr., R.J. Harrison, J. Chem. Phys. 96 (1992) 6796. [7] S. Simon, A. Gil, M. Sodupe, J. Bertran, J. Mol. Struct. (Teochem). 727 (2005) 191. [8] B. Herrera, O. Dolgounitcheva, V.G. Zakrzewski, A. Toro-Labbe, J.V. Ortiz, J. Phys. Chem. A. 108 (2004) 11703. [9] R.A.J. O’Hair, S. Blanksby, M. Styles, J.H. Bowie, Int. J. Mass Spectr. 182/183 (1999) 203. [10] M.W. Schmidt, K.K. Baldrige, J.A. Boatz, S.T. Elbert, M.S. Gordon, J.H. Jensen, S. Koseki, N. Matsunaga, K.A. Nguyen, S.J. Su, T.L. Windus, M. Dupuis, J.A. Montgomery, J. Comput. Chem. 14 (1993) 1347. [11] Gaussian 03, Revision C.02, Gaussian, Inc., Wallingford CT, 2004. [12] A.V. Snegursky, F.F. Chipev, A.N. Zavilopulo, O.B. Shpenik, Radiat. Phys. Chem. 76 (2007) 604. [13] H.-W. Jochims, M. Schwell, J.-L. Chotin, Chem. Phys. 298 (2004) 279. [14] G.A. Junk, H.J. Svec, J. Am. Chem. Soc. 85 (1963) 839. [15] L.F. Pacios, P.C. Go9 mez, J. Mol. Struct. (Teochem). 544 (2001) 237. [16] M. Schwell, H.-W. Jochims, H. Baumgartel, F. Dulieu, S. Leach, Planet. Space Sci. 54 (2006) 1073. [17] R. Maul, M. Preuss, F. Ortmann, K. Hannewald, F. Bechstedt, J. Phys. Chem. A 111 (2007) 4370. [18] National Institute of Standards, Standard Reference Database: Chemistry Webbook http://webbook.nist gov. [19] A.F. Lago, L.H. Coutinho, R.R.T. Marinho, A. Naves de Brito, G.G.B. de Souza, Chem. Phys. 307 (2004) 9. [20] D. Shemesh, G.M. Chaban, R.B. Gerber, J. Phys. Chem. A. 108 (2004) 11477. [21] D. Shemesh, R.B. Gerber, J. Chem. Phys. 122 (2005) 241104.