Electron Ionization (IE) is the oldest ion source developed for mass spectrometry, it has been invented by Dempster since 1918. This ionization source was the most common and widely used in coupling with gas chromatography, because this ion source allows to ionize compounds in gas phase, the molecules are ionized directly in their gaseous states. However, it does not allow to ionize directly molecules from liquid phase, moreover the later must be thermostable because the ion source operates at high temperature. The ions produced by this source are very energetic, they often present fragmentation allowing to probe the structure of the molecules on the other hand, the molecular ions are few or not present at all because of the internal energy transmitted by the kinetic energy of the electrons. Compared with other sources of ionization, the Electron Ionization is highly reproducible, the molecules are easy to identify by comparing with the mass spectra present in the libraries.
The Electron Ionization is known as electron impact Ionization however, It should be noted that there is no impact between the electron and the molecule, the electron is only cross the molecule. It is in fact the waves accompanied within the kinetic energy of the electrons that create the perturbation of the bonds of the molecule, consequently one or several electron can be detached from the molecules.
The Ion formation
When a molecule is interacted by an electron with relative high kinetic energy, a part of the energy of the electron is transferred to the molecule through a vibration process. In terms of energy transfer, if the electron interacts the molecule efficiently, the energy transferred may be greater than the ionization energy (IE) of the molecule. From the point of view in mass spectrometry it is the most desirable process: the ionization by electron ejection generating a molecular ion (positive radical ion):
M + e- → M+• + 2e– (1)
Depending on the analyte and the kinetic energy, simply, doubly and even triply charged ions can be observed.
M + e– → M2+ + 3e– (2)
M + e– → M3+• + 4e– (3)
From the equation (2 and 3) the doubly charged ion, M2+, is an ion with an even number of electrons (event electron), and the triply charged ion, M3+•, is again an odd electron. Futhermore, there are several other events from the interaction of electrons and neutral, for example a less effective interaction to put the neutral in an excited state electronically without ionizing. The electron can also be captured by the neutral to form a negative radical ion. However, electron capture (EC) is very unlikely to occur with electrons at high kinetic energy about few tens of eV. Thus, CE proceeds efficiently only with electrons having very low energy, preferably with thermal electrons. However, molecules containing highly electronegative elements can capture fast electrons to form negative ions with low efficiency. Another process that can occur during electron bombardment is Penning Ionization, this is the non-ionizing electron-neutral interaction creating the electronically excited neutrals. Ionization reactions occur when neutrons such as electronically excited noble gases interact with a molecule M in its fundamental state, this process can be divided into two classes . The first process is Penning ionization (Eq.4) , the second is associative ionization, also known as the Hornbeck-Molnar process (Eq.5) .
A* + M → A + M+• + e– (4)
A* + M → AM+• + e– (5)
It is obvious that the ionization of the neutral can only occur when the energy deposited by the electron-neutral collision is equal to or greater than the ionization energy (IE) of the corresponding neutral. In the past, ionization energy has been incorrectly called “ionization potential” (IP), from the derivative of the technique for its experimental determination.
The ionization energy (IE), is defined as the minimum amount of energy that must be absorbed by an atom or molecule in its fundamental electronic and vibrational state in order to form an ion which is also in its fundamental state by ejecting an electron. It should be noted that the ionization energy of most molecules is about 7-15 eV.
The ionization process by electron bombardment occurs extremely fast. The time required for an electron of 70 eV to cross a distance corresponding to more or less half a dozen bond lengths (about 1nm), is only about 2×10-16s, and even the largest molecules can be traversed in a few femtoseconds. The molecule being interacted by the electron can be considered as immobilize, because its thermal velocity of some 100 ms-1 is negligible compared to the speed of the electron passing through it. The vibrational motions are slower by at least two orders of magnitude, for example, the stretching vibration cycle of CH takes 1,1×10-14s which can be calculated from its IR absorbance at about 3000 cm-1. According to the Born-Oppenheimer approximation, electronic movements and nuclear movements can be separated due to the large difference in mass between nuclei and electrons   . Meaning the positions of the atoms and the bond lengths remain unchanged during the ionization. The transitions are represented as vertical transitions (fig 2). The probability of a neutral transition in its fundamental state to a certain vibrational level of the ion is expressed by its Franck-Condon factor. The Franck-Condon factors comes from the fact that the most likely transition is where the two fundamental state and excited electronic wave functions overlap. Whereas for the fundamental state this is at equilibrium position, the wave functions of the vibrational states reach their maxima upon return to equilibrium. Regardless of where the electron is taken, ionization tends to cause a weakening of the bond within the precursor ion relative to the neutral. On average, the weak bond is longer and has a strong tendency towards bond dissociation. In terms of potential energy surfaces, the situation can be visualized easier by focusing on a single bond within the molecule or simply by discussing a diatomic molecule. A diatomic molecule has only one vibratory motion, so its energy potential can be represented by potential energy curves rather than the potential energy surface. The minimum of the potential energy curve of the neutral, is supposed to be in its fundamental vibrational state, it is located at the length of the bond which is shorter than that of the radical ion in its ground state, r1 (Fig. 1). Therefore, the ionization is accompanied by vibrational excitation, the transitions are vertical, ie, the positions of the atoms are fixed during this short period.
Fig. 1: The vertical transition of excited molecules by the kinetic energy of the ions.
The factor distribution of Franck-Condon, FFC, describes the distribution of vibrational states for an excited ion . Compare to r0, the larger r1 is the more the generation of excited ions above the dissociation energy is likely. Photoelectron spectroscopy allows both the determination of adiabatic ionization energy and Franck-Condon factors. The counterpart of vertical ionization is a process where the ionization of the neutral in its vibrational ground state would also give the radical ion in its fundamental vibrational state, namely the (0 ← 0) transition. This is called adiabatic ionization and should be represented by a diagonal line in the diagram. The IEvert – IEad difference can lead to ionization energy errors is in the orders of 0.1-0.7 eV . The other fate of the ion depends on the shape of its potential energy surface. In the case where there is a minimum and the excitation level is lower than the energy barrier for dissociation D, the ion can exist for a very long time. Ions with an internal energy above the dissociation energy level will dissociate at some point, causing fragment ions. In some unfavorable cases, some ions carry no minimum on their energy surface. These will suffer from spontaneous dissociation and there is no chance of observing the molecular ion.
Ionization Efficiency and Cross Section
The ionization energy represents the minimum energy required for ionization of the neutral, for an effective ionization, electrons which interact must carry at least that amount of energy. If the energy is then quantitatively transferred during the collision, the ionization would take place. Obviously, such an event is of rather low probability and, as a result, the ionization efficiency is close to zero with electrons carrying only the energy corresponding to IE of the neutral. However, a slight increase in electron energy results in a steady increase in ionization efficiency. Strictly speaking, all molecular species has an ionization efficiency curve and its own function of the ionization cross section of the specific molecule. In the case of methane, this question has been studied several times (Fig. 2) . The ionization section describes an area through which the electron must move to effectively interact with the neutral, and therefore, the ionization cross section is given in units of square meters.
Fortunately, ionization cross section curves versus electron energy are all of the same type, with a maximum at electron energies of about 70 eV (Fig. 2). That is why the EI spectra are almost always acquired at 70 eV.
Fig. 2 The abundance of molecular ions for CH4 as a function of the kinetic energy of the electrons.
The plateau of the ionization efficiency curve at ~ 70 eV implies that the electron energy variations are negligible around 70eV. Therefore, the electronic impact offers a better mass spectra reproducibility, allowing the comparison of the spectra obtained from different mass spectrometers or from mass spectrometry databases.
Covalent bond dissociation
Great efforts have been made to generate accurate and reliable thermodynamic data for ions. Once, these data are available, they can be used to elucidate fragmentation, mechanisms. In addition, it is useful for general information on energy dimensions in mass spectrometry.
The heat from the formation of neutral molecules, ΔHf (RH), can be obtained from the combustion data with great precision. The covalent bond dissociation energy may be either derived from homolytic dissociation.
R-H → R• H+•, ΔHDhom (7)
Or from the heterolytic dissociation
R-H → R + H–, ΔHDhet (8)
The homolytic enthalpies of covalent bond dissociation give the energy necessary to cleave a bond of the neutral molecule that is in its fundamental electronic and vibrational state in order to obtain a pair of radicals that are not in excited states either. The homolytic bond dissociation enthalpies are about 3-5 eV. Heterolytic bond dissociation energies are valid for the case of the creation of a cation and an anion in their fundamental state from the neutral precursor, these values include the amount of energy required for the separation of charge; they are in the orders of 10-13 eV. Due to the significant bond change, bond dissociation for ion molecular requires significantly less energy than those for neutral molecules.
Direct cleavage and rearrangement
The reactions of excited ions are not always as simple as expected. Of course, the existence of multiple fragmentation pathways for an ion which contains few tens of atoms, brings about the different types of reactions that all certainly will not lead to the same function k(E) (Reaction rate constant) .
The K(E) functions of two reactions of the same type appear to be “parallel” in the comparison, starting at different energies of activation (Fig. 3), while different types of reactions will show a crossover of their functions K(E) to an excess of intermediate energy (Fig. 3b).
For example, case (a) represents two competition cleavages homolytic covalent bond. Homolytic cleavage is a direct cleavage in the covalent bond of molecular ions creating an event-electron ion and a radical. A direct cleavage may require more energy activation than the other (E02> E01) because of the difference between the bonds to be cleaved, and their excessively high energy rates will increase significantly. The more the ion acquires energy, the more the cleavage of the bond can occur easily. A further increase in excess energy will be ineffective only if the velocity approaches the upper limit as defined by the vibration frequency of the bond to be cleaved.
Case (b) compares a rearrangement (reaction 1) to a direct cleavage (reaction 2). During a rearrangement process the precursor ion ejects a neutral fragment which is an intact molecule after rearranging in an energetically favorable way. The rearrangement starts at low energy excess, but the rate gets closer quickly, while the cleavage starts later on, then becomes dominant when the excess energy is higher. The differences can be explained by the different transition states of the two types of reactions. The cleavage has a loose transition state , ie, it does not require that certain parts of the molecule adopt a specific position at the time of cleavage. The dissociation simply requires enough energy in the bond to overcome the covalent bond strength, once the bond is too tight, the fragments separate. The rearrangement requires less energy because the energy for breaking the bond on one side is compensated by the energy received from the other.
Fig. 3. The comparison of the different types of reactions, direct cleavage and rearrangement.
Structure des ions moléculaires en phase gazeuse
La perte d’un électron à partir de la molécule méthane neutre.
 Jones, E.G.; Harrison, A.G. Study of Penning Ionization Reactions Using a Single-Source Mass spectrometer. Int. J. Mass Spectrom. Ion Phys. 1970, 5, 137- 156.
 Penning, F.M. Ionization by Metastable Atoms. Naturwissenschaften 1927, 15,818.
 Hornbeck, J.A.; Molnar, J.P. Mass- Spectrometric Studies of Molecular Ions in the Noble Gases. Phys. Rev. 1951, 84,621-625.
 Born, M.; Oppenheimer, J.R. Zur Quantentheorie Der Molekeln. Annalen der Physik 1927, 84, 457-484.
 Lipson, R.H. Ultraviolet and Visible Absorption Spectroscopy, in Encyclopedia of Applied Spectroscopy, Andrews, D.L. (ed.); Wiley-VCH: Berlin, 2009; Chap. 11, pp. 353-380.
 Seiler, R. Born-Oppenheimer Approximation. International Journal of Quantum Chemistry 1969, 3, 25-32.
 Dunn, G.H. Franck-Condon Factors for the Ionization of H2 and D2. J. Chem. Phys. 1966, 44, 2592-2594
 Märk, T.D. Fundamental Aspects of Electron Impact Ionization. Int. J. Mass Spectrom. Ion Phys. 1982, 45, 125-145.
 McAdoo, D.J.; Bente, P.F.I.; Gross, M.L.; McLafferty, F.W. Metastable Ion Characteristics. XXIII. Internal Energy of Product Ions Formed in Massspectral Reactions. Org. Mass Spectrom. 1974, 9, 525-535.
 McLafferty, F.W.; Wachs, T.; Lifshitz, C.; Innorta, G.; Irving, P. Substituent Effects in Unimolecular Ion Decompositions. XV. Mechanistic Interpretations and the Quasi-Equilibrium Theory. J. Am. Chem. Soc. 1970, 92, 6867-6880.