1 Introduction

Electron transfer from neutral potassium (K) atoms in collisions with neutral molecules (M) yielding ion pair formation generates a cation (\(\hbox {K}^{+})\) and a metastable temporary negative ion (TNI) with an excess of internal energy (\(\hbox {M}^{-})^{\# }\) that may result in further fragmentation. In such collision-induced dissociation processes, since the energy can be chosen larger than the threshold of electron transfer, i.e. \(\Delta \)E \(=\) IE(K) – EA(M), with IE(K) the ionization energy of the potassium atom and EA(M) the electron affinity of the molecule, a fragment anion and a neutral radical may be formed. Yet if the lifetime of the TNI is long enough, intramolecular energy redistribution may occur competing with dissociation, the latter being classified either statistical or direct. Such processes have been reported in our laboratory in different molecular targets investigated in a wide collision energy range, typically from a few eV up to 1 keV. Just to point out some examples, we note particular attention to cyclic and non-cyclic polyatomic molecules. In case of cyclic chemical compounds, an efficient redistribution of the excess energy through the different internal degrees of freedom was observed in radiosensitizers, e.g. nitroimidazoles [1, 2] and pyrimidine [3]. Regarding non-cyclic molecules, these have shown to serve as efficient doorways for enhanced bond excision due to the mainly repulsive nature of their lowest-lying anionic potential energy surfaces, as are the cases of nitromethane [4] and tetrachloromethane [5].

From the TOF analysis of fragment anions formed in electron transfer experiments, the shape and width of detected features contain relevant information on the collision dynamics. In particular, the kinetic energy gained by a given negative ion resulting from bond breaking of the TNI can be derived. As a result of a methodology used to explore the underlying molecular mechanisms resulting in dissociation of chlorobenzene and chlorinated related molecules, we present for the first time a comprehensive investigation of \(\hbox {Cl}^{-}\) kinetic-energy release distributions (KERDs) from \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) in electron transfer experiments in a wide collision energy range.

The formation of \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) negative ions has been reported in several occasions either in dissociative electron attachment (DEA) [6,7,8,9,10,11,12,13] or electron transmission spectroscopy experiments [14,15,16,17]. This molecular anion has been considered as a prototype for radiosensitization due to its closely related resemblance to halopyrimidine molecules [18,19,20]. Other studies on \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) relevant to the present work are electron impact spectroscopic experiments [21,22,23], theoretical investigations on potential energy surfaces [24] and electronic structure [25], as well as references therein. Regarding \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\), we are only aware of a DEA study yielding \(\hbox {Cl}^{-}\), where a close comparison with its hydrogenated analogue (\(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\)) revealed that deuteration has no appreciable effect in the magnitude of the cross-section [26].

As far as \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) negative ions formation is concerned, a literature survey revealed no information for the former, whereas for the latter just \(\hbox {C}_{{6}}\hbox {Cl}_{6}^{-}\) has been reported [27, 28]. Nonetheless, we have recently performed in our laboratory a thorough investigation on the formation of anionic states of \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) probed in electron transfer experiments for the energy range 10–150 eV (laboratory frame), with such study being supported by state-of-the-art quantum chemical calculations [29].

In Sect. 2, we present a brief summary of the experimental setup, and in Sect. 3, the methodology used to obtain KERDs from the experimental data. Section 4 is dedicated to results and discussion, which includes a complete comparison among the different molecular targets investigated. Section 5 deals with a summary and the conclusions that can be drawn from this study.

2 Experimental methods

TOF mass spectra of product anions formed in electron transfer experiments in collisions of neutral potassium (K) atoms with target molecules (\(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {Cl}_{6})\), were recorded at different laboratory frame energies. The spectra have been obtained in a crossed molecular beam setup which was previously described elsewhere [30]. Briefly, a projectile beam of neutral hyperthermal K atoms crosses at right angles an effusive target molecular beam and the product anions formed in the electron transfer process are TOF mass analysed. The hyperthermal K beam is produced in a charge exchange chamber from the interaction of accelerated \(\hbox {K}^{+}\) ions, emitted from a commercial ion source (HeatWave, US) in the range of 10–150 eV in the laboratory frame, with gas-phase neutral potassium atoms from an oven source. The intensity of the neutral K beam prior to collisions with the target molecules is monitored using a Langmuir–Taylor-type ionization detector, before and after collecting the mass spectra. The TOF mass spectrometer used is a dual-stage Wiley–McLaren type, where the extraction region is formed by two parallel plates (one of it is a tungsten mesh with high transparency) placed 12 mm apart and operated at an extraction voltage of 292 \(\hbox {Vcm}^{-1}\) with a time duration of 1   \(\upmu \hbox {s}\) in an 80 \(\upmu \hbox {s}\) duty cycle. Anions are detected by a channel electron multiplier (CEM) with an effective area of 1 cm diameter. The TOF anion yield is normalized considering the primary beam current, pressure and acquisition time. The typical base pressure in the collision chamber was 4\(\times \) 10\(^{-5}\) Pa and the working pressure was 1\(\times \) 10\(^{-3}\) Pa. TOF mass calibration was performed on the basis of the well-known anionic species formed after potassium collisions with \(\hbox {CH}_{{3}}\hbox {NO}_{{2}}\) [31] and \(\hbox {CCl}_{{4}}\) [5].

\(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) samples were supplied by Sigma-Aldrich with a stated purity of 99.9%, 99%, 99% and \(\ge \) 98%, respectively. \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) were degassed through repeated freeze-pump-thaw cycles. \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) solid sample was used as delivered and gently heated up to 340 K through a temperature PID (proportional-integral-derivate controller) unit. In order to test for any thermal decomposition products from \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) molecular beam, TOF mass spectra were recorded at different temperatures, and no differences in the relative peak intensities as a function of temperature were observed.

3 Kinetic energy release distribution (KERD)

In this section, we present a concise description of the methodology used to extract the kinetic energy release from the width and shape of the TOF mass spectra of chlorine anions formed in electron transfer experiments. This approach has been used before where a comprehensive description can be found in Limão-Vieira et al. [32] and Rebelo et al. [33] Briefly, in the unimolecular dissociation mechanism of a temporary negative ion (TNI) yielding two fragments with masses \(m_{1}\) and \(m_{2}\), the kinetic energy release \(\varepsilon _{\mathrm{d}}\) of a fragment anion formed with an isotropic velocity distribution is given by:

$$\begin{aligned} \varepsilon _{d} =\frac{1}{8\mu }\left( {F.{\Delta }t} \right) ^{2} \end{aligned}$$
(1)

with \(\mu \) being the reduced mass of the system, F being the electrostatic extraction field and \(\Delta t\) the difference in extraction time between two fragments emitted in forward and backward directions.

Fig. 1
figure 1

Molecular structure schematics of \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\), and \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\)

Full size image

The KERD, \(D(\varepsilon _{\mathrm{d}})\), is closely related to the intensity and velocity of the primary K beam, the density of the target molecular beam, the dimensions of the interaction volume and Q(v) the velocity dependent emission of the anionic fragments per solid angle. The maximum velocity component (v) of a fragment anion impinging on the detector is reached by those ions which are well-aligned with the extraction field, i.e. \(v \, \approx \, v_{\mathrm{o}}\) (\(v_{\mathrm{o}}\) the initial velocity). The maximum velocity component normal to the TOF axis (\(v_{\mathrm{m}})\) is given by \(v_{m} =\frac{r}{t}\), with r the detector radius and t the total flight time. The flight time is proportional to \(\sqrt{m} \) so the maximum \(\frac{1}{2}mv_{m}^{2} \) is constant and independent of m or \(v_{\mathrm{o}}\).

The current TOF configuration yields \(\varepsilon _{m} =\frac{1}{2}mv_{m}^{2} \approx 0.03\, eV\). We can now express \(Q(v_{\mathrm{o}})\) as a function of the experimental \(\Delta t\) and \(I(\Delta t)\) functions, the latter obtained by measuring the intensity of the TOF mass feature, according to ref. [32] as:

$$\begin{aligned} Q\left( {v_{{o}} } \right) =\frac{F}{2A}\left\{ {1+\frac{3}{4}\frac{v_{m}^{2} }{v_{{o}}^{2} }+\cdots } \right\} {\Delta }t^{2}.I\left( {{\Delta }t} \right) \end{aligned}$$
(2)

where the constant A is a function of \(\varepsilon _{\mathrm{m}}\) [32], and so the KERD can now be given as:

$$\begin{aligned} D\left( {\varepsilon _{d} } \right) =\frac{1}{A}\left\{ {1+\frac{3}{4}\frac{v_{m}^{2} }{v_{{o}}^{2} }+\cdots } \right\} {\Delta }t.I\left( {{\Delta }t} \right) \end{aligned}$$
(3)

this provided that \(v_{m}^{2} /v_{o}^{2} \ll 1\). The quadratic term in Eq. (3) can be ignored for \(\varepsilon _{\mathrm{d}}\) > 0.1 eV, thus \(D(\varepsilon _{\mathrm{d}})\) is given as:

$$\begin{aligned} D\left( {\varepsilon _{d} } \right) =\frac{1}{A}.{\Delta }t.I\left( {{\Delta }t} \right) \end{aligned}$$
(4)

4 Results and discussion

Negative ion formation in potassium collisions with \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) yields only chlorine anions, whereas in the case of \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\), the main anion signal intensities have been assigned to the undissociated parent anion (\(\hbox {C}_{{6}}\hbox {Cl}_{6}^{-})\) and the formation of \(\hbox {C}_{{6}}\hbox {Cl}_{5}^{-}\) and \(\hbox {Cl}^{-}\), with other minor contributions from ions resulting from the loss of Cl units and ring breaking due complex internal reactions within the TNI [29].

Fig. 2
figure 2

Kinetic energy release distributions \(D(\upvarepsilon _{\mathrm{d}})\) in collisions of K \(+\) \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) obtained from the TOF spectra of \(\hbox {Cl}^{-}\) at 12, 43 and 117 eV collision energy in the centre-of-mass frame. Statistical fitting procedure for 12 and 117 eV (see text for details). The shape of the distribution does not change appreciably with the 5% error bars, so these have not been included to avoid congestion of the figure. The anion states are marked as arrows (see text for details)

Full size image

KERDs for the dominant fragment anion \(\hbox {Cl}^{-}\) have been obtained from the TOF mass spectra induced by potassium electron transfer to \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) (Fig. 1) at collision energies, in the centre-of-mass frame, of about 12, 40 and 118 eV. Note that in the present arrangement, the associated collision energy uncertainties are 0.34 eV for \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\), while for \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) is 0.4 eV. The shape of an asymmetric TOF mass peak is related to the fragment anion velocity in the extraction region, where the maximum intensity corresponds to ions formed with no initial velocity, while the left and right flanks of such peaks are related to the extraction of ions with a velocity component in the forward and backward directions, respectively. From the height of \(\hbox {Cl}^{-}\) TOF mass peaks at different widths, one obtains the KERD. Figures 2, 3, 4 and 5 depict the KERDs, \(D(\varepsilon _{\mathrm{d}})\), obtained from Eq. (4) for \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\), respectively. For values of \(\varepsilon _{\mathrm{d}}\) < 0.1 eV, the distributions are plotted with dashed lines, meaning the less accuracy of such data due to neglecting, as already noted, the quadratic term contribution in Eq. (3). The uncertainty limits associated to each experimental data point do not change the shape of the distributions, within ± 5%, so these have not been included in the figures to avoid congestion. Figure 6 shows \(\hbox {Cl}^{-}\) kinetic energy release maximum from \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) at 12, 40 and 118 eV collision energy in the centre-of-mass frame. The KERDs of \(\hbox {Cl}^{-}\) are limited to \(\sim \) 14 eV for K \(+\) \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), K \(+\) \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\) and K \(+\) \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) reaching this value at 118 eV collision energy (Figs. 2, 3 and 4), whereas those for K \(+\) \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) are within \(\sim \) 8.0 eV with the maximum value at 40 eV (Fig. 5).

Regarding DEA experiments in \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), Khatymov et al. [10] report anion formation with masses of 111, 75, 25 and 35 u, being \(\hbox {Cl}^{-}\) (35 u) the most abundant. From the resonance profiles obtained through the different DEA [7,8,9,10] and ETS [14,15,16,17, 34] data, a general consensus has been reached to assign \(\hbox {Cl}^{-}\) formation to the prominent feature found at \(\sim \) 0.8 eV electron impact energy, as a shape resonance of \(\uppi \)* character (Table 1). The grand total cross section for electron scattering measurements reports the main resonance peak at 8.5 eV [22] in good agreement with the electron transmission value of 8.22 eV. [15] The features above 5.0 eV are related to core-excited resonances, such assignment consistent with the lowest-lying excited states of \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) which are discernible at photon energies > 5.5 eV. [35] In the case of \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), the main anion states’ positions from DEA [10] and/or electron impact spectroscopy [21] experiments are marked in Fig. 2 with arrows and the corresponding values are shown in Table 1. The lowest resonance \(\pi _{1}^{*} \) at \(\sim \) 0.8 eV, and presumably \(\pi _{2}^{*} \) at 1.2 eV, give rise to \(\hbox {Cl}^{-}\) formation provided a \(C_{\mathrm{2v}}\) symmetry lowering induced by vibronic coupling of the \(\pi _{1}^{*} \left( {b_{1} } \right) \) and \(\pi _{2}^{*} \left( {a_{2} } \right) \) with the \(\sigma ^{*}\left( {a_{1} } \right) \) anion states prevails [21], though strong excitation of out-of-plane (\(a_{2})\) and in-plane non-totally symmetric vibrations (\(b_{2})\) are the effective path for C–Cl bond excision [21]. Moreover, the broad resonance feature at 2.6 eV has been assigned to a temporary electron capture into the \(\sigma _{C-Cl}^{*} \left( {a_{1} } \right) \) molecular orbital, the band at 4.50 eV to a \(\pi _{3}^{*} \left( {b_{1} } \right) \) of the anion [21], while the features at 5.8 and 6.4 eV to inter-shell resonances of core-excited character [10]. Note that Palmer et al. [35] report vertical excitation energies for neutral \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) at 5.807 (\(^{1}\hbox {A}_{1})\) and 6.724 eV (\(^{1}\hbox {A}_{1})\) yielding C–Cl bond excision, in good agreement with the values of 5.91 and 6.77 eV from Ari et al. [36]

From the KERD profile in Fig. 2, the first and second resonances at 0.8 and 1.2 eV are the dominant reaction paths for \(\hbox {Cl}^{-}\) formation. The distributions maxima peak at 0.47 eV and 0.55 eV for 12 (and 43 eV) and 117 eV collision experiments (Fig. 6). Note that at 12 and 43 eV, the downward slope shows a tendency to change at \( > rsim \)2.6 eV which may be reminiscent of the excess energy in the TNI character changing relative to \(\hbox {C}_{{6}}\hbox {H}_{{5}}\)–\(\hbox {Cl}^{-}\) dissociation level, i.e. at lower energies is due to a statistical process (see below) and at higher energies to direct dissociation. Actually, at 2.6 eV, \(\hbox {Cl}^{-}\) is formed through a \(\sigma _{C-Cl}^{*} \left( {a_{1} } \right) \) resonance, so the broad contribution at higher kinetic energies, \(\varepsilon _{\mathrm{d}}\), is due to direct dissociation with the excess energy converted into translational energy of the fragment. A close inspection of Fig. 2 also reveals that the downward slope of the distribution at 117 eV seems to level off at \(\sim \) 8 eV with character changing \( > rsim \) 6.0 eV. This is due to the contribution of two additional resonances at 5.80 and 6.40 eV (Table 1) that are associated to core-excited resonances. We now detain ourselves to the lower kinetic energy release region where the distribution profile has been fitted with a statistical fitting in Figs. 2, 3, 4 and 5 (dashed lines), as thoroughly described before [37]:

$$\begin{aligned} D\left( {\varepsilon _{d} } \right) =C\left( {1-\frac{\varepsilon _{d} }{E_{e} }} \right) ^{s-2} \end{aligned}$$
(5)

with C a constant independent of the energy, s the adapted degree of freedom of the anion and \(E_{\mathrm{e}}\) the available access energy. In the specific case of \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), for 12 eV and 117 eV collision energy, \(E_{\mathrm{e}} \, =\) 6 eV and \(E_{\mathrm{e}} \, =\) 7.5 eV. The reasonably good fitting accord with the distributions yield C and s values of 2.65\(\times \) 10\(^{-6}\) and 3.0, and 2.86\(\times \) 10\(^{-6}\) and 3.0, respectively. Given the arbitrary units of the KERD intensities, constant C has no physical meaning. At this stage is relevant to comment on the s values. Actually, three degrees of freedom rather than 30 internal modes in \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), is rather small, this meaning that only part of the reaction pathway leading to dissociation is channelled through a particular isolated state of the TNI, whereas the remaining is certainly going through direct bond breaking.

Fig. 3
figure 3

Kinetic energy release distributions \(D(\upvarepsilon _{\mathrm{d}})\) in collisions of K \(+\) \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\) obtained from the TOF spectra of \(\hbox {Cl}^{-}\) at 12, 40 and 118 eV collision energy in the centre-of-mass frame. Statistical fitting procedure for 12 and 40/118 eV (see text for details). The shape of the distribution does not change appreciably with the 5% error bars, so these have not been included to avoid congestion of the figure

Full size image
Fig. 4
figure 4

Kinetic energy release distributions \(D(\upvarepsilon _{\mathrm{d}})\) in collisions of K \(+\) \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) obtained from the TOF spectra of \(\hbox {Cl}^{-}\) at 12, 40 and 1118 eV collision energy in the centre-of-mass frame. Statistical fitting procedure for 12/40 and 118 eV (see text for details). The shape of the distribution does not change appreciably with the 5% error bars, so these have not been included to avoid congestion of the figure

Full size image
Fig. 5
figure 5

Kinetic energy release distributions \(D(\upvarepsilon _{\mathrm{d}})\) in collisions of K \(+\) \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) obtained from the TOF spectra of \(\hbox {Cl}^{-}\) at 12, 40 and 118 eV collision energy in the centre-of-mass frame. Statistical fitting procedure for 12 and 40 eV (see text for details). The shape of the distribution does not change appreciably with the 5% error bars, so these have not been included to avoid congestion of the figure. The anion states are marked as arrows (see text for details)

Full size image

As noted previously, we have no information related to electron affinities of the lowest-lying electronic states of \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\). Yet, Figs. 3 and 4 show the KERDs of both molecules at 12, 40 and 118 eV collision energy in the centre-of-mass frame. The \(\hbox {Cl}^{-}\) distributions from \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\) peak at 0.33 eV and shift to slightly higher energy at 0.45 eV, while for \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\), the maximum appears at 0.47 eV and shifts to 0.87 eV as the collision energy is increased (Fig. 6). Of relevance the trending of such peak maxima as a function of the collision energy relative to \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) (see Fig. 6). We note in the case of chlorobenzenes (\(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\)) that above 40 eV, the peak maximum distribution is, roughly speaking, insensitive to the increasing collision energy. Such seems reasonable given the role of the \(\sigma _{C-Cl}^{*} \) resonances and even core-excited resonances that are mainly dissociative in character. Above 40 eV, the difference in \(\hbox {Cl}^{-}\) kinetic energy release maximum is not particularly significant for \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\) against \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), yet below 40 eV that becomes noticeable. Such is certainly related to the lifetime of the low-lying resonances, where in case of \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\), autodetachment strongly competes with dissociation and rejection of the extra electron may prevail in the deuterated compound. Moreover, the statistical fittings in Fig. 3 have been obtained by using Eq. (5) for 12 eV and 118 eV collision energy with excess energies of \(E_{\mathrm{e}} \, =\) 4.5 eV and \(E_{\mathrm{e}} \, =\) 6.0 eV, certainly lower than the related values of \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\). In case of \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\), the striking difference to \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) is noted above 40 eV (Fig. 6). Note that \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) does not have \(\uppi \) delocalized electrons over the ring, so any \(\uppi \)*/\(\upsigma \)* mixing does not hold, and the dissociation mechanism may certainly be dictated by relevant \(\upsigma \)* antibonding character along the C–Cl coordinate resulting in higher kinetic energy of the fragment anion. This is in contrast with \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), where Skalicky et al. [21] have shown that the SOMO (singly occupied molecular orbital) of the negative ion has a \(\uppi \)*(\(a_{2})\)-like molecular orbital character strongly mixing with \(\sigma _{C-Cl}^{*} \left( {a_{1} } \right) \) MO favouring \(\uppi \)*/\(\upsigma \)* mixing with relevant implications regarding the dissociation mechanism as the nuclear wave packet moves along the potential energy surface. The statistical fittings in Fig. 4 result from Eq. (5) for 12 eV and 118 eV collision energy, and with values of \(E_{\mathrm{e}} \, =\) 6.0 eV and \(E_{\mathrm{e}} \, =\) 11.5 eV, respectively.

As far as \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) is concerned, the ground-state character has been assigned to valence \(\uppi \)* in \(\hbox {C}_{{6}}\hbox {Cl}_{6}^{-}\) from generalized Kohn–Sham semicanonical projected random phase approximation, yielding an electron affinity from –0.03 up to –0.30 eV at different levels of accuracy. [38] Recently, we have calculated at the RKS/B3LYP\(+\)D3/aug-cc-pvtz level of theory a vertical electron affinity of –0.14 eV (\(\uppi \)*) for the ground-state anion and also obtained from the \(\hbox {K}^{+}\) energy loss spectrum in collisions of potassium with \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\), vertical electron affinities for the different anion states at (–1.86 ± 0.30), (–2.96 ± 0.20) and (–3.76 ± 0.20) eV [29]. These have been assigned to \(\sigma _{C-Cl}^{*} \), \(\pi _{CC}^{*} /\sigma _{CC}^{*} +\sigma _{C-Cl}^{*} \) and \(\sigma _{C-Cl}^{*} \) states, respectively, and are marked as arrows in Fig. 5. From the TOF mass spectrometry and energy loss data [29], the first resonance is not the dominant reaction path for \(\hbox {Cl}^{-}\) formation. Moreover, we have reported that for collision energies below 30 eV, the extra electron is captured by the LUMO with mostly \(\pi _{CC}^{*} \) character. Note that in \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) with relevant ring delocalized \(\uppi \) system, the six identical Cl atoms are “highly competitive” among each other for the extra charge. In such K–\(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) collisions at energies above the threshold of ion-pair formation, the excess energy may be channelled into the available degrees of freedom, resulting mainly in a stable \(\hbox {C}_{{6}}\hbox {Cl}_{6}^{-}\) ion rather than prevalent bond breaking. However, the SOMO of the negative ion has \(\pi _{CCl}^{*} \) character, while the closest MO shows a strong \(\sigma _{C-Cl}^{*} \) antibonding nature. Therefore, C–Cl bond breaking resulting in \(\hbox {Cl}^{-}\) formation may be only operative by efficient diabatic curve crossing between \(\pi _{CCl}^{*} \) and \(\sigma _{C-Cl}^{*} \), as long as the nuclear wave packet survives long enough along the C–Cl coordinate to yield \(\hbox {Cl}^{-}\). From the KERD profiles in Fig. 5, the second, third and fourth resonances are the dominant reaction paths for \(\hbox {Cl}^{-}\) formation. Note that at 12 and 118 eV the downward slope shows a tendency to change at \( > rsim \) 1.86 eV, while at 40 eV such is certainly visible at higher kinetic energies closely related to the anionic states at 2.96 and 3.76 eV. Thus, the excess energy in the TNI at lower energies is dictated by a statistical process and at higher energies to direct dissociation. Of relevance the fact that at 1.86 eV, \(\hbox {Cl}^{-}\) is formed through a \(\sigma _{C-Cl}^{*} \) resonance. The statistical fitting for 12 eV and 40 eV collision energy, results in \(E_{\mathrm{e}} \, =\) 2 eV and \(E_{\mathrm{e}} \, =\) 4.5 eV, respectively, and with \(s \, =\) 3.0 in both cases. Finally, the distributions maxima in Fig. 6 peak at 0.14 eV and 0.34 eV for 12 (and 118 eV) and 40 eV collision energies. From the chlorinated molecules investigated, \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) shows the \(\hbox {Cl}^{-}\) kinetic energy release maximum with the lowest values (Fig. 6). In this comparison, such does not seem to be unexpected given that \(\hbox {Cl}^{-}\) is the only anion formed from electron transfer to \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\). In \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\), the parent anion formation strongly competes with \(\hbox {Cl}^{-}\), the former accounting for a significant branching ratio at low collision energies, i.e. more than 70% below 10 eV [29]. At intermediate energies, 40 eV, \(\hbox {C}_{{6}}\hbox {Cl}_{6}^{-}\) amounts 30% and \(\hbox {Cl}^{-}\) 60%, while at higher energies other fragments stemming from ring breaking amount 25% of the total anion yield, although the role of the \(\sigma _{C-Cl}^{*} \) resonances and even core-excited resonances that are mainly dissociative in character [29].

Table 1 Resonance positions (in eV) and their assignments from dissociative electron attachment, electron transmission spectroscopy and electron scattering from \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\)
Full size table
Fig. 6
figure 6

\(\hbox {Cl}^{-}\) kinetic energy release maximum in collisions of K atoms with \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\), and \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) at 12, 40 and 118 eV in the centre-of-mass frame. Error bars have been added and account for ± 5%

Full size image

5 Conclusions

We have reported a comprehensive investigation on the KERDs of \(\hbox {C}_{{6}}\hbox {H}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {D}_{{5}}\hbox {Cl}\), \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) and \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) at collision energies in the centre-of-mass frame of about 12, 40 and 118 eV. Such distributions have been obtained from the shape and width of the TOF mass spectra of \(\hbox {Cl}^{-}\) ions. Collision induced dissociation results in an energy redistribution either through the different internal degrees of freedom (via vibrational excitation) or direct transformation of the excess energy via translational motion of the fragment anions involved. The statistical fittings of \(\hbox {Cl}^{-}\) distributions at low kinetic energy release, \(\varepsilon _{\mathrm{d}}\), reveal at least three effective degrees of freedom, while the role of statistical and direct dissociation has been thoroughly discussed according to the nature of the different resonances obtained by dissociative electron attachment and/or electron transmission spectroscopy, where applicable. Additionally, we have made some considerations on the collision dynamics for the set of chosen molecules, and the conclusions drawn lend support to previous investigations of \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) negative ion formation in electron transfer experiments, while for the remaining molecular systems the present results will serve to help interpreting the underlying molecular mechanisms within electron transfer induced fragmentation. Finally, the \(\hbox {Cl}^{-}\) kinetic-energy release maxima as a function of the collision energy, show that in case of \(\hbox {C}_{{6}}\hbox {Cl}_{{6}}\) the lowest values correspond to the strong competition of \(\hbox {Cl}^{-}\) ion with \(\hbox {C}_{{6}}\hbox {Cl}_{6}^{-}\) formation. Moreover, due to absence of \(\uppi \) delocalized electrons over the \(\hbox {C}_{{6}}\hbox {H}_{{11}}\hbox {Cl}\) ring, the electronic state spectroscopy is dictated by \(\upsigma \)* antibonding character along the C–Cl bond.