THE JOURNAL OF CHEMICAL PHYSICS 128, 154309 共2008兲

Functional group dependent dissociative electron attachment to simple organic molecules Vaibhav S. Prabhudesai,a兲 Dhananjay Nandi, Aditya H. Kelkar, and E. Krishnakumarb兲 Tata Institute of Fundamental Research, Colaba, Mumbai 400005, India

共Received 4 January 2008; accepted 26 February 2008; published online 16 April 2008兲 Dissociative electron attachment 共DEA兲 cross sections for simple organic molecules, namely, acetic acid, propanoic acid, methanol, ethanol, and n-propyl amine are measured in a crossed beam experiment. We find that the H− ion formation is the dominant channel of DEA for these molecules and takes place at relatively higher energies 共⬎4 eV兲 through the core excited resonances. Comparison of the cross sections of the H− channel from these molecules with those from NH3, H2O, and CH4 shows the presence of functional group dependence in the DEA process. We analyze this new phenomenon in the context of the results reported on other organic molecules. This discovery of functional group dependence has important implications such as control in electron induced chemistry and understanding radiation induced damage in biological systems. © 2008 American Institute of Physics. 关DOI: 10.1063/1.2899330兴 I. INTRODUCTION

Dissociative electron attachment 共DEA兲 to molecules that are biologically important has drawn a lot of attention in the recent time.1–8 The main aim behind these experiments is to understand the dynamics that leads to the single and double strand breaks in DNA in radiation damage of the tissues. DEA of the low energy secondary electrons has been found to be the prominent mechanism toward the radiation damage of the biological tissues through DNA strand breaks.9 In order to understand the complex molecular dynamics which leads to the DNA strand breaks, the DEA studies on simpler molecular systems such as DNA constituents and sugar molecules have been carried out in the recent past.10–17 The DEA study on even simpler systems such as simple carboxylic acids, alcohols, and simple aromatic compounds are taken as the starting points for these understandings. Apart from this, the presence of these simple compounds in the interstellar medium and large molecular clouds in deep space18,19 imply that these studies are important from the astrobiology point of view as well. The basic features that have been discovered in the DEA to these simple molecules are of two classes. In the very low energy regime 共below 4 eV兲, the prominent dissociation channel is found to be the hydrogen abstraction.20 Mostly, the formation of single particle shape resonance is known to play a role in DEA at these energies.1 Another class of resonances that are present at relatively higher energies, i.e., above 4 eV, is the core excited resonances. These are also known as two particle resonances, where the incoming electron excites one of the bound electrons while getting captured in the potential of the electronically excited neutral molecule.21 As the core excited resonances are associated with the electronic excitation of the neutral molecule, these a兲

Present address: Department of Physics of Complex Systems, Weizmann Institute of Science, Rehovot 76100, Israel. b兲 Electronic mail: [email protected]. 0021-9606/2008/128共15兲/154309/7/$23.00

resonances have direct resemblance with the excited state spectrum of the corresponding neutral molecule. Many dissociation channels are observed in this energy regime as the complex potential energy surfaces of the excited states of the polyatomic molecules provide large number of possibilities in the dissociation channel. It is seen that the presence of specific functional groups in neutral molecules appears as the characteristic feature in their UV absorption spectra.22 These features have been found to resemble the characteristic pattern of the absorption band in the precursor molecule that leads to the specific functional group. For example, in the case of amines, the optical absorption spectra are found to show a similar pattern that resembles the absorption spectrum associated with the ammonia. This pattern is observed along with the other bands that are associated with the remaining part of the molecules. The excitation of the molecules by optical absorption is guided by the dipole selection rule. The excitation by electron impact provides a way of accessing the excited states that are not correlated with the ground state with the dipole transition, although in the electron impact excitation, the dipole transition may play the most dominant role. Thus, as in the case of optical absorption spectra, we may have similar excitations in electron impact process in molecules containing the same functional groups. However, very few systematic studies have been reported so far23,24 in this direction in the DEA process. Our earlier effort is the only attempt toward looking at the patterns in the H− ion formation.25 The DEA cross section is a product of the electron capture cross section to form the resonance and the survival probability of the resonance against autodetachment enabling it to dissociate. The survival probability is dependent on the nuclear coordinates and thus changes along different dissociation pathways of the resonance. Hence, the absolute cross sections for different channels of dissociation of the resonance are an input to understand the dynamics of the DEA.

128, 154309-1

© 2008 American Institute of Physics

Downloaded 17 Apr 2008 to 147.47.241.116. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/jcp/copyright.jsp

154309-2

Prabhudesai et al.

FIG. 1. Schematic of the experimental setup.

In this context, the measurement of absolute cross sections for DEA is important, over and above the importance of these cross sections for modeling various chemical processes. One of the approaches toward modeling and understanding of the radiation damage resulting from DEA to complex molecules such as DNA is through understanding the DEA to the constituent molecules.26 We have tried to simplify this approach and attempted to make it understandable at a more fundamental level by looking at much simpler molecules. Our results show that the DEA patterns observed in the precursor molecules are replicated in the bigger molecules due to functional group dependence in DEA.

II. EXPERIMENTAL

The experimental setup has been described earlier in detail elsewhere.27 It consists of a magnetically collimated and pulsed electron beam crossing an effusive molecular beam at right angle. The electron beam is derived from a hairpin filament. The typical energy resolution of the electron beam is about 0.5 eV. The electron current is measured using a Faraday cup placed at the opposite end along the axis of the electron gun, as shown in Fig. 1. The negative ions produced by the interaction in the crossed beam volume are pushed into the time of flight 共TOF兲 mass spectrometer by a pulsed electric field. The ions are then detected by a channel electron multiplier operated in pulse counting mode. The axis of the TOF spectrometer coincides with the axis of the molecular beam. The ion extraction field pulse follows the electron pulse with a delay of few tens of nanoseconds, and hence, does not affect the latter. The flight tube of the TOF spectrometer is made up of four separate sections, which are designed to act like an electrostatic lens assembly to transport the divergent beam of ions extracted from the interaction region to the detector situated at its exit. The high field pulsed extraction and the segmented TOF mass spectrometer are significant features of this experiment, as they facilitate the complete collection and analysis of all ions produced in the interaction region without any discrimination due to their mass, initial kinetic energy, and angular distribution.28

J. Chem. Phys. 128, 154309 共2008兲

The background pressure in the chamber during the experiment was always kept less than 5 ⫻ 10−6 torr. The needle valve and the gas line introducing the gaseous targets into the vacuum chamber were heated to a temperature of about 90 ° C. The interaction region including the capillary array was at a temperature of 70 ° C due to the heating caused by the magnetic coils used for collimating the primary electron beam. The organic molecular samples in the glass bulbs were sufficiently pumped by a rotary vacuum pump to remove the volatile impurities in them. The measurements were carried out in two steps. In the first step, the mass spectrometer was tuned on a specific mass and the ion yield measured as a function of electron energy. In the second step, the cross sections at the resonance peaks in the yield functions were determined using the relative flow techniques28 using O− from O2 共Ref. 29兲 as the standard. The cross sections at the peaks were then used to normalize the respective ion yield functions. The errors in the measurement are mostly statistical in nature, apart from the uncertainty in the cross section for O− from O2, which is used as the standard for normalization of the cross sections. The uncertainty in these data is ⫾10%.29 The possible systematic errors arising from the collection and detection efficiency of the ions have been minimized, as discussed elsewhere.30 The systematic error due to the limited electron energy resolution could not be estimated, but is assumed to be small since the resonances in these molecules are found to be relatively broad. The statistical errors in our measurements are mainly arising from the ion count rate. This has been minimized while carrying out normalization of the cross sections to absolute values using the relative flow technique by collecting data for sufficiently long time. In all the above measurements, the estimated error inclusive of the random error is at most 15%. III. RESULTS AND DISCUSSIONS A. Acetic acid „CH3COOH… and propanoic acid „C2H5COOH…

DEA to simple carboxylic acids, namely, acetic acid and propanoic acid has been reported earlier.31,32 In these measurements, various negative ion fragments are observed except the H− ions. These measurements are carried out at high electron energy resolution and very good mass resolution. However, in these experiments, relatively weak electric fields are used to extract the ions along with a quadrupole mass spectrometer for mass analysis. Hence, these experiments strongly discriminate against ions with large translational energies.33 Since H− ions carry almost all the excess energy of the DEA process as kinetic energy, it becomes difficult to detect them using such an experimental setup. The DEA measurements carried out on formic acid in our laboratory using the same setup that we use for the present measurements has shown that the H− ion formation is one of the prominent DEA channel present, which was not observed earlier.34 The previous measurements in acetic acid found resonances at 1.5 eV for CH3COO−, 0.75 eV for CH2O2−, and about 10 eV for CH2COO−, CHCO−, CCO−, OH−, O−, and

Downloaded 17 Apr 2008 to 147.47.241.116. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/jcp/copyright.jsp

154309-3

J. Chem. Phys. 128, 154309 共2008兲

Dissociative electron attachment

TABLE I. Absolute DEA cross sections for various negative ion fragments observed from acetic acid and propanoic acid in comparison with the earlier reported data 关for acetic acid 共Ref. 31兲 and propanoic acid 共Ref. 32兲兴. Our results Fragment negative ion

Resonance position 共eV兲

Earlier reported data Cross section 共cm2兲

Fragment negative ion

Acetic acid 共CH3COOH兲 2.0⫻ 10−19 H− −19 CH2− 1.8⫻ 10 O− 1.3⫻ 10−19

H−

06.7 07.7 09.1

CH2− O− + OH− CCO− + CHCO− HCOO− 共CH2O2兲− CH2COO− CH3COO−

10.5 10.1 10.0

H−

06.7 07.7 09.1

O− + OH−

05.1 9.3

4.9⫻ 10−21 2.2⫻ 10−20

OH−

C 2H 2− + C 2H 3−

03.9 08.0

1.8⫻ 10−21 5.6⫻ 10−21

C 3H 5− + C 2H 2O −+ CHO2− + CH2O2−

00.0 03.4 09.3

C 2H 3O 2− C2H5COO−

1.9⫻ 10−20 2.1⫻ 10−20 4.0⫻ 10−20 Not observed Not observed 10.2 2.2⫻ 10−20 01.5 1.3⫻ 10−19

Resonance position 共eV兲

Cross section 共cm2兲

Not Observed ⬃10.2 Not estimated ⬃05.5 Not estimated

OH− CCO− CHCO− HCOO− 共CH2O2兲− CH2COO− CH3COO−

Propanoic Acid 共C2H5COOH兲 8.8⫻ 10−20 H− −19 O− 1.4⫻ 10 −20 9.0⫻ 10

⬃09.5 ⬃10.0 ⬃10.8 ⬃10.0 01.3 0.75 ⬃10.0 01.5

Not estimated Not estimated Not estimated Not estimated Not estimated Not estimated Not estimated 6.0⫻ 10−19

Not Observed ⬃04.8 1.0⫻ 10−19 ⬃07.0 1.0⫻ 10−19 00.3 ⬃09.0

5.0⫻ 10−19 5.0⫻ 10−19

C 2H 2− C 2H 3−

01.9 01.9

5.0⫻ 10−19 1.1⫻ 10−18

2.0⫻ 10−20 2.2⫻ 10−20 1.0⫻ 10−20

C 3H 5− C 2H 2O − CHO2−

04.0 04.0 01.3

4.1⫻ 10−19 8.2⫻ 10−19 2.0⫻ 10−17

01.5 09.2

5.4⫻ 10−21 9.0⫻ 10−21

CH2O2−

00.0 01.4

1.6⫻ 10−16 0.5⫻ 10−16

01.5

1.3⫻ 10−19

03.9 01.5 01.5

0.3⫻ 10−16 5.3⫻ 10−18 1.7⫻ 10−16

−

C 2H 3O 2 C2H5COO−

CH2− fragment ion channels, respectively. Of these the one at 1.5 eV in the CH3COO− was found to be the dominant one with a cross section of 6 ⫻ 10−19 cm2 with an uncertainty of about an order of magnitude.31 Because of the limited mass resolution in our experiment, we are unable to separate all the fragment ions as observed in the previous experiment. However, due to the characteristic resonance structure, we are able to identify the contribution from various fragments by comparing with the previous results. The peak positions and the absolute cross sections that we obtained are given in Table I along with the earlier reported values for various fragment ions. We find the cross section for the formation of CH3COO− ion peaking at 1.5 eV with a magnitude of 1.3 ⫻ 10−19 cm2, while the cross sections for other fragment ions are relatively small with the notable exception of H−. We do see a resonance at about 10 eV in the CHCO−, CCO−, OH−, O−, and CH2− fragment ion channels, although we are unable to separate out CHCO− from CCO− and OH− from O− due to poor mass resolution. As compared to the previous measurements, we did not observe CH2O2− and HCOO− fragments.

The absolute cross sections for various channels other than H− are given in Fig. 2. In contrast to the previous measurements, we find the formation of H− ions as three resonances peaking at 6.7, 7.7, and 9.1 eV, respectively. We also note that H− formation is the most dominant channel of DEA to acetic acid. The absolute cross sections for the formation of H− from acetic acid along with that from propanoic acid are given in Fig. 3. Pelc et al.32 have reported ten different fragment ion channels in the case of DEA to propanoic acid. In this case, also, the H− channel was not observed. However, like in the case of acetic acid, we find H− ion channel to be the dominant one, with almost identical resonances 共Fig. 3兲. The cross sections for H− channel along with other channels in propanoic acid are given in Table I. Although lacking in mass resolution to separate all the fragment ion channels, we are able to identify various resonance positions for most of the channels similar to what has been reported earlier.32 Apart from observing H−, the present results differ from the previous measurement in the following. We find a resonance at

Downloaded 17 Apr 2008 to 147.47.241.116. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/jcp/copyright.jsp

154309-4

J. Chem. Phys. 128, 154309 共2008兲

Prabhudesai et al.

FIG. 4. DEA cross section for the H− channel from 共a兲 methanol and 共b兲 ethanol.

FIG. 2. DEA cross section for 共a兲 CH2−, 共b兲 O− + OH−, 共c兲 CHCO− and CCO−, and 共d兲 CH2COO− and CH3COO− from acetic acid.

9.2 eV in addition to that observed earlier at 1.5 eV in the C2H3O2− channel. Most importantly, we observe substantially low cross section for the hydrogen abstraction channel leading to the formation of C2H5COO− on comparison with the previous reported value. The previous measurements32 had estimated the cross section at 1.5 eV as 1.7 ⫻ 10−16 cm2 with an order of magnitude uncertainty. The absolute cross section that we measured for this channel is 1.3⫻ 10−19 cm2, which is about three orders of magnitude smaller. From Fig. 3, we note that the H− ion yield functions from both the acids show striking similarity in the resonance positions. These resonances could be correlated with the resonances seen in H2O and CH4, as discussed in Sec. IV below,25,35 due to functional group dependence.

et al.36 estimated the cross section for the O− channel at the 10.5 eV resonance to be of the order of 10−19 cm2. We observe only H− and O− ions from methanol. CH3O− is found to be absent. The resonance patterns in the H− channels are given in Fig. 4 along with that for ethanol. The absolute cross sections at various observed resonances are listed in Table II. We find three resonances in the H− channel at 6.4, 7.9, and 10.2 eV, respectively, which are in agreement with the previously reported studies. The O− channel displays only one resonance centered at 10.3 eV. In the case of ethanol too, we observe three resonances in the H− channel in the 6 to 10 eV range, as given in Fig. 4. However, we do not see the formation of C2H5O− as observed in a recent experiment.38 This experiment, which used a partially deuterated sample, showed three resonances at 6.34, 7.85, and 9.18 eV, respectively, in the H− channel. The absolute cross sections for the three resonances that we have measured are given in Table II. Similar to the two carboxylic acids, the three resonances seen in both the alcohols, in almost identical energy range, could be correlated to the resonances seen in H2O and CH4, as discussed later. C. n-propyl amine „C3H7NH2…

B. Methanol „CH3OH… and ethanol „C2H5OH…

There have been reports about the DEA to methanol in the past.36–38 The main fragments reported from DEA to methanol are, namely, O−, OH−, and CH3O−36 with the H− being reported by Curtis and Walker37 and Ibanescu et al.38 The O− channel is found to show the major resonance at 10.5 eV with a minor peak at around 8 eV. The OH− channel is also observed showing similar pattern.36 The CH3O− and H− channels have been reported with three resonances at 6.4, 7.9, and 10.5 eV.37,38 The cross section measurements for these channels have not been carried out, although Kuhn

There has been no report in the literature about the DEA to n-propyl amine. The aim behind studying the DEA to this molecule was to extend the observed functional group TABLE II. Absolute DEA cross sections for various negative ion fragments observed from methanol, ethanol, and n-propyl amine. Fragment negative ion

H−

O− + OH−

H−

H− FIG. 3. DEA cross section for the H− channel from 共a兲 acetic acid and 共b兲 propanoic acid.

Resonance position 共eV兲

Cross section 共cm2兲

Methanol 共CH3OH兲 06.4 07.9 10.2 10.3

7.6⫻ 10−20 3.6⫻ 10−20 4.3⫻ 10−20 4.5⫻ 10−20

Ethanol 共C2H5OH兲 06.4 07.9 09.3 n-propyl amine 共C3H7NH2兲 05.2 08.8

7.5⫻ 10−20 2.5⫻ 10−20 2.8⫻ 10−20 5.2⫻ 10−20 1.7⫻ 10−20

Downloaded 17 Apr 2008 to 147.47.241.116. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/jcp/copyright.jsp

154309-5

Dissociative electron attachment

FIG. 5. DEA cross section for the H− channel from n-propyl amine.

dependence on the DEA pattern in the H− ion channel to the functional group other than the hydroxyl and alkyl groups.25 We observed that the H− ion channel is the most dominant channel of dissociation from this molecule with the two resonances at 5.2 and 8.8 eV. The peak cross sections at these resonances, as shown in Fig. 5, were found to be 5.2 ⫻ 10−20 and 1.7⫻ 10−20 cm2, respectively. In comparison with the resonance pattern observed in ammonia, a distinct similarity can be noted.35 We observed extremely weak signal in the NH2− channel and no other fragments were observed against the background. IV. FUNCTIONAL GROUP DEPENDENT DEA

As can be seen from the observed H− ion channel from simple carboxylic acids and simple alcohols, there is a similarity in the resonance patterns. In all these molecules, there are two possible sites that can contribute to the H− signal, namely, the alkyl site that corresponds to the C–H bond breakage and the hydroxyl site 共carboxyl in the case of acids兲 that corresponds to the O–H bond breakage and alkyl 共C–H bond兲 site and amine site 共corresponding to N–H bond breakage兲 for the n-propyl amine. Interestingly, all these compounds show more than one resonance in the H− channel and in the energy range that is indicative of the role of core excited resonances. Also, the electron energies associated with these resonances are well above the thresholds for obtaining H− ions from any of the sites of a given molecule from these groups. For example, the appearance energy for H− from the O–H part in acetic acid is 4.02 eV, where as that from the C–H part is below 4 eV. From this, a simple question arises whether we can associate a specific resonance with a specific site. We addressed this question by making measurements on H− and D− channels from partially deuterated acetic acid. The results are given in Fig. 6, which show that the first two resonances are located at the O–H site. The data available on partially deuterated methanol and ethanol also show that the first two resonances at 6.5 and 8 eV seen in the hydride ion channel arise from the O site.37,38 These resonance positions also resemble the resonance pattern observed for water in the H− channel.35 This similarity seen in the electron attachment process can be traced to the intrinsic similarity in the optical absorption spectra of acetic acid34 and alcohols to that of water22 arising from a lone pair excitation at the O site to the unoccupied Rydberg orbitals.38 The lone pair excitation is a

J. Chem. Phys. 128, 154309 共2008兲

FIG. 6. 共Color online兲 Yield functions for H− 共triangle兲 and D− 共square兲 ions from partially deuterated acetic acid 共CH3COOD兲. Also shown is the H− 共cross兲 yield function from pure acetic acid.

characteristic excitation of oxygen site as these electrons do not participate in the bond formation in the molecule. This implies that there should be a characteristic resonance in all the compounds that contain the O–H group, where this valance transition is available as manifested in the case of alcohols and carboxylic acids. In the case of the third resonance from alcohols and acids, the resonance position is the characteristic one observed in simple alkanes. For example, methane shows a characteristic resonance at 9.8 eV in the H− channel.35 It is also observed that for higher alkanes, this resonance shifts to the lower electron energies.39 We may assign it as a Feshbach resonance corresponding to the electronic excitation observed in the optical absorption spectra in these molecules.40 Our results on partially deuterated acetic acid and the results on deuterted methanol by Kuhn et al.36 and Curtis and Walker37 show that the resonance seen in alkanes is manifesting as a resonance localized at the C site in these molecules. Thus, one may generalize that the resonances seen in alkanes have a common origin and they are carried over to even bigger molecules with a C–H bond. The present results on acetic acid as well as those from methanol36,37 indicate a contribution from the O–H bond also at the third resonance. This was interpreted as indicative of complex dynamics, such as scrambling, which takes place in the dissociation of this resonance.36 On the other hand, this contribution to the third resonance from the O–H site may also be ascribed to the third resonance seen in water.35 The results from n-propyl amine help in further generalizing the functional group dependence on amines. The resonance pattern observed in the n-propyl amine shows a strong similarity with the resonance pattern observed in ammonia. In the optical absorption spectra, simple amines and ammonia molecule show distinct similarity.22 If the functional group dependence exists in DEA, this similarity is expected to manifest in the resonant attachment through core excited resonances. Thus, the observed similarity in the resonance patterns in the H− channel in the amine and ammonia strengthens the argument for the presence of functional group dependence in DEA. It may be noted that the amine group has a lone pair nonbonding electrons on the N site. We may argue that like in the case of the O site, the N site should also have the characteristic electronic excitations that will manifest in the resonance pattern. This has been observed in

Downloaded 17 Apr 2008 to 147.47.241.116. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/jcp/copyright.jsp

154309-6

J. Chem. Phys. 128, 154309 共2008兲

Prabhudesai et al.

the hydrogen abstraction channel in the free electron attachment experiments.24 In the case of electron induced desorption of H− from the condensed film of aniline, this resonance is identified.41 This implies that the characteristic resonance corresponding to N site is observed in both saturated and unsaturated amines at around 5 eV. The site selectivity observed in simple acids and alcohols as well as in amine has also been observed in more complicated molecule such as thymine.42 In this case, the hydride ions coming from the N site contribute to the lower resonances which are located at 5.5 and 6.8 eV. The resonances observed in the H− channel around 10 eV are found to be coming from the C sites. It has also been shown that one can identify the contribution from the methyl group and the aromatic ring in the molecules using the methylated compounds.42 Our preliminary studies on benzene and toluene are also indicative of such differentiation, although due to poor electron energy resolution, we could not resolve the two peaks very effectively. Interestingly, one may cite the presence of the resonance at 6.5 eV attributed to the N site in thymine42 as failure of the argument for functional group dependence since we do not expect to observe a resonance from the N site at around 6.5 eV, and the resonance near 6.5 eV is a characteristic of the O site. However, a closer look at the thymine structure shows that this resonance indeed arises from the presence of the O atom near the N sites. We believe that the 6.8 eV resonance in thymine is arising from the tautomeric structures of thymine in which the hydrogen atom from the N site is found to be hopping around the adjacent O site.43 We have discussed above the qualitative similarity in the resonance positions leading to the identification of the functional group dependence in DEA. The question is whether it manifests in the magnitude of the cross sections as well, despite DEA being a very complex process and the absolute cross sections are very sensitive to initial conditions even in a given molecule. In DEA, the absolute cross section depends on the capture cross section for the electrons at a given energy and the survival probability of the resonance against autodetachment before the dissociation. The capture cross section is expected to be a characteristic of the individual molecule, and hence, is expected to vary from molecule to molecule. The survival probability of the resonance depends on the lifetime of the resonance against autodetachment and the relative motion of the dissociating fragments. For a diatomic molecule, the survival probability is the characteristic of the potential energy curve of the negative ion resonance. In the case of polyatomic molecules, this scenario becomes even more complicated as the potential energy surface for the resonance is multidimensional. The multidimensional nature of the surface makes the dynamics of the process very complicated apart from allowing redistribution of the energy if the resonance is relatively longer lived. In the case of a core excited resonance, many dissociation channels can become active as the energy provided to the system by the captured electron is much more than the threshold energies for each of the dissociation channels. All these will affect the cross section of a particular channel and the effect will be difficult to quantify. In spite of all these, we notice a certain

pattern in the magnitude of the cross sections as well. To begin with, in all these molecules that we have studied, the cross section for the H− channel is dominant as compared to other channels. This is only to be expected since the breakup leading to the formation of H− is expected to have the fastest kinematics in the two body fragmentation process due to low reduced mass. We also note that within a class of molecules, the magnitude of cross section for the H− channel is similar. For example, in the case of alcohols, the cross sections at the first two resonances are within a factor of 2 from one alcohol to another. However, this is about two orders of magnitude smaller than that observed for water at those resonances. Within carboxylic acids as well, the peak cross section at the two lower resonances are within a factor of 2. Here, the cross section is one order of magnitude lower than that observed in water. Interestingly, the optical absorption cross sections for the electronic transition, corresponding to the lone pair of electrons from oxygen atom in all these molecules such as water and simple alcohols as well as simple carboxylic acids, have very similar values.22 The order of magnitude difference between the DEA cross sections may be depicting the role of the survival probability of the negative ion resonance.

V. CONCLUSION

We present absolute cross sections for the formation of various fragment negative ions from acetic acid, propanoic acid, methanol, ethanol, and n-propyl amine through DEA. In all these molecules, the H− channel is found to be the dominant channel in DEA. A comparison of the resonances in the H− channel from these molecules with those from H2O, CH4, and NH3 shows that the electron attachment properties leading to the formation of H− are retained by the hydroxyl, alkyl, and amine groups even in bigger molecules. We identify this as due to a phenomenon which occurs at the very fundamental level due to the core excited nature of the resonances occurring at the O, C, and N sites, respectively. Since the electronically excited states of neutral molecules display functional group dependence, it is carried over to the electron attachment process in core excited resonances. The characteristic resonances associated with the O, N, and C sites are very important in the context of DEA to complex molecules such as DNA. We believe that the functional group dependence in DEA may be a very general phenomenon and thus, the characteristic resonances would be expected from molecules containing atoms such as S and P, which are also present in the big biologically important molecules. Moreover, the observation of the functional group dependence in DEA will also help in modeling electron interaction with biological molecules in the context of radiation damage. The functional group dependence present in DEA allows selective bond breaking in molecules as a function of electron energy. Due to its inherent simplicity, chemical control using DEA holds enormous potential for practical applications.

Downloaded 17 Apr 2008 to 147.47.241.116. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/jcp/copyright.jsp

154309-7

ACKNOWLEDGMENTS

We thank M. A. Rahman for his help in repeating some of the measurements and cross-checking the absolute cross sections. 1

F. Martin, P. Burrow, Z. Cai, P. Cloutier, D. Hunting, and L. Sanche, Phys. Rev. Lett. 93, 068101 共2004兲. S. Ptasinska, S. Denfil, P. Scheier, and T. D. Märk, J. Chem. Phys. 120, 8505 共2004兲. 3 H. Abdoul-Carime and E. Illenberger, Chem. Phys. Lett. 390, 481 共2004兲. 4 S. Denfil, S. Ptasinska, G. Hanel, B. Gstir, M. Probst, P. Scheier, and T. D. Märk, J. Chem. Phys. 120, 6557 共2004兲. 5 S. Ptasinska, S. Denifl, P. Candori, S. Matejcik, and T. D. Märk, Chem. Phys. Lett. 403, 107 共2005兲. 6 L. Sanche, Eur. Phys. J. D 35, 364 共2005兲. 7 P. Sulzer, S. Ptasinska, F. Zappa, B. Mielewska, A. R. Milosavljevic, P. Scheier, T. D. Märk, I. Bald, S. Gohlke, M. A. Huels, and E. Illenberger, J. Chem. Phys. 125, 044304 共2006兲. 8 P. Papp, J. Urban, S. Matejcik, M. Stano, and O. Ingolfsson, J. Chem. Phys. 125, 204301 共2006兲. 9 B. Boudaiffa, P. Cloutier, D. Hunting, M. A. Huels, and L. Sanche, Science 287, 1658 共2000兲. 10 M. Huels, I. Hahndorf, E. Illenberger, and T. D. Märk, J. Chem. Phys. 108, 1309 共1998兲. 11 S. Gohlke, A. Rosa, E. Illenberger, F. Brunnings, and M. A. Huels, J. Chem. Phys. 116, 10164 共2002兲. 12 G. Hanel, B. Gstir, P. Scheier, M. Probst, B. Farizon, M. Farizon, E. Illenberger, and T. D. Märk, Phys. Rev. Lett. 90, 188104 共2003兲. 13 H. Abdoul-Carime, S. Gohlke, and E. Illenberger, Phys. Rev. Lett. 92, 168103 共2004兲. 14 C. Konig, J. Kopyra, I. Bald, and E. Illenberger, Phys. Rev. Lett. 97, 018105 共2006兲. 15 I. Bald, J. Koprya, and E. Illenberger, Angew. Chem., Int. Ed. 45, 4851 共2006兲. 16 I. Bald, J. Koprya, I. Dabkowska, E. Antonsson, and E. Illenberger, J. Chem. Phys. 126, 074308 共2007兲. 17 I. Baccarelli, F. A. Gianturco, A. Grandi, N. Sanna, P. R. Lucchese, I. Bald, J. Koprya, and E. Illenberger, J. Am. Chem. Soc. 129, 6269 共2007兲. 18 J. M. Hollis, J. A. Pedelty, L. E. Synder, P. R. Jewell, F. J. Lovas, P. Palmer, and S.-Y. Liu, Astrophys. J. 588, 353 共2003兲. 19 Y.-J. Kaun, S. B. Charnley, H.-C. Huang, W. L. Tseng, and Z. Kisiel, Astrophys. J. 593, 848 共2003兲. 2

J. Chem. Phys. 128, 154309 共2008兲

Dissociative electron attachment

A. Pelc, W. Sailer, P. Scheier, and T. D. Märk, Vacuum 78, 631 共2005兲. L. G. Christophorou, Electron-Molecule Interactions and Their Applications 共Academic, Orlando, FL 1984兲, Vols. 1 and 2. 22 M. B. Robin, Higher Excited States of Polyatomic Molecules 共Academic, New York, 1975兲, Vols. 1 and 2. 23 M. V. Muftakhov and A. Fokin, Rapid Commun. Mass Spectrom. 11, 1923 共1997兲. 24 T. Skalicky and M. Allan, J. Phys. B 37, 4849 共2004兲. 25 V. S. Prabhudesai, A. H. Kelkar, D. Nandi, and E. Krishnakumar, Phys. Rev. Lett. 95, 143202 共2005兲. 26 P. Swiderek, Angew. Chem., Int. Ed. 45, 4056 共2006兲. 27 D. Nandi, S. A. Rangwala, S. V. K. Kumar, and E. Krishnakumar, Int. J. Mass. Spectrom. 205, 111 共2001兲. 28 E. Krishnakumar and K. Nagesha, Rapid Commun. Mass Spectrom. 9, 336 共1995兲. 29 D. Rapp and D. D. Briglia, J. Chem. Phys. 43, 1480 共1965兲. 30 D. Nandi, E. Krishnakumar, A. Rosa, W.-F. Schmidt, and E. Illenberger, Chem. Phys. Lett. 373, 454 共2003兲. 31 W. Sailer, A. Pelc, M. Probst, J. Limtrakul, P. Scheier, E. Illenberger, and T. D. Märk, Chem. Phys. Lett. 378, 250 共2003兲. 32 A. Pelc, W. Sailer, P. Scheier, T. D. Märk, and E. Illenberger, Chem. Phys. Lett. 392, 465 共2004兲. 33 A. Pelc, W. Sailer, P. Scheier, M. Probst, N. J. Mason, E. Illenberger, and T. D. Märk, Chem. Phys. Lett. 361, 277 共2002兲. 34 V. S. Prabhudesai, D. Nandi, A. H. Kelkar, R. Parajuli, and E. Krishnakumar, Chem. Phys. Lett. 405, 172 共2005兲. 35 P. Rawat, V. S. Prabhudesai, G. Aravind, N. Bhargavaram, M. A. Rahman, and E. Krishnakumar, J. Phys.: Conf. Ser. 80, 012018 共2007兲. 36 A. Kuhn, H.-P. Fenzlaff, and E. Illenberger, J. Chem. Phys. 88, 7453 共1988兲. 37 M. G. Curtis and I. C. Walker, J. Chem. Soc., Faraday Trans. 88, 2805 共1992兲. 38 B. C. Ibanescu, O. May, A. Monney, and M. Allan, Phys. Chem. Chem. Phys. 9, 3163 共2007兲. 39 F. H. Dorman, J. Chem. Phys. 44, 3856 共1966兲. 40 M. B. Robin, Higher Excited States of Polyatomic Molecules 共Academic, New York, 1983兲, Vol. 3. 41 M. A. Huels, L. Parenteau, and L. Sanche, Chem. Phys. Lett. 279, 223 共1997兲. 42 S. Ptasinska, S. Deniff, V. Grill, T. D. Mark, E. Illenberger, and P. Scheier, Phys. Rev. Lett. 95, 093201 共2005兲. 43 J. Rejnek, M. Hanus, M. Kabelac, F. Ryjacek, and P. Hobza, Phys. Chem. Chem. Phys. 7, 2006 共2005兲. 20 21

Downloaded 17 Apr 2008 to 147.47.241.116. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/jcp/copyright.jsp