ENERGETIC PROCESSING OF COMPLEXMOLECULES IN THE GAS PHASE

Michael Wolf

Energetic processing of complexmolecules in the gas phase

Michael Wolf

©Michael Wolf, Stockholm University 2018 ISBN print 978-91-7797-083-5ISBN PDF 978-91-7797-084-2 Printed in Sweden by Universitetsservice US-AB, Stockholm 2017Distributor: Department of Physics, Stockholm University

For Маша

Abstract

Collisions between molecules and gas phase targets often lead to various intriguingprocesses. Such collisions may induce fragmentation of molecules that can be di-vided into different subsets depending on the projectile, target, and collision energy.One major part of the present research is the exploration of astrophysical relevantcollision mechanisms. In collisions between polycyclic aromatic hydrocarbon (PAH)molecules or fullerenes with, for example, helium, nuclear stopping can lead to theprompt knockout of a carbon atom from the molecule. Such a vacancy in the molecularcarbon backbone can be highly reactive, and lead to the formation of larger molecules.The energy dependencies of such processes are important for the understanding of as-trochemical molecular growth processes, which in turn may lead to the formationof larger and more complex molecules in space. In addition, hydrogenation of PAHschanges their structures and internal properties, including their resistance against frag-mentation. To better understand the effects of hydrogenation on the fragmentation ofPAHs, low energy photofragmentation experiments are presented along with the col-lision experiments, and a detailed comparison is made between the effects of thesedifferent types of energy transfer processes.Besides astrophysically relevant research, studies on the response of biomolecules tocollisions with gas phase targets are presented. Here, the energy dependence for for-mation of the protonated n-butyl β -ionone Schiff base through electrocyclization ofthe protonated n-butylamine Schiff base of all-trans-retinal in collisions is presented.The latter is a model compound for all-trans-retinal, the chromophore of the lightsensitive opsin proteins, and such studies are essential for the understanding of theoperation of mammal vision.While our collision studies are very successful, they are sometimes also limited bythe experimental timescale. Therefore, we have constructed an experimental setup forion storage and fragmentation analysis. The goal of this new experiment is to storeinternally hot fragments to investigate their behavior on extended timescales and asfunctions of internal excitation energies.

Sammanfattning

I denna avhandling presenteras experiment i vilka molekylära joner med ki-netiska energier av storleksordningen kiloelektronvolt kolliderar med atomer.En stor del av denna avhandling handlar om astrofysikaliskt relevanta kolli-sionsmekanismer. Två klasser av molekyler som vi har studerat är polycyklis-ka aromatiska kolväten (PAH)-molekyler och fullerener. PAH molekyler bestårav minst två sammankopplade aromatiska kolväteringar, medan fullerener harihåliga strukturer av kolatomer. Då PAH-molekyler eller fullerener kolliderarmed till exempel helium kan enskilda kolatomer knockas ut. Vakansen somdärmed skapas i den molekylära kolstrukturen är ofta mycket reaktiv och kanleda till att större molekyler på ett effektivt sätt skapas i växelverkan med andramolekyler. Energiberoendet av sådana processer kan komma att visa sig varaviktigt för förståelsen av hur komplexa molekyler skapas i rymden.

En annan del av denna avhandling handlar om studier av hydrogeneradePAH-molekyler, där ett antal extra väteatomer har bundits till PAH-molekylerna.Detta förändrar en PAH-molekyls geometri, och andra egenskaper och ävenhur stor dess motståndskraft mot fragmentering är. För att bättre förstå effek-terna av hur hydrogenering påverkar PAH-molekylernas stabilitet, presenterashär, förutom kollisionsförsök med heliumatomer, även fotofragmenteringsex-periment.

Vi har även utfört kollisionsexperiment med biomolekyler. Här presente-ras energiberoendet för bildandet av protonerade NBISB-molekyler (från eng-elska n-butyl β -ionone Schiff base) då protonerade NB-RPSB-molekyler (n-butylamine Schiff base of all-trans-retinal) aktiveras i kollisioner med atomereller molekyler. NB-RPSB används som en modell för alltrans-retinal, kromo-foren som fångar upp ljus i däggdjurs ögon, och sådana studier kan därför bidratill att öka förståelse för hur detta sker på en molekylär nivå.

Resultaten från våra kollisionsexperiment är ibland begränsade av den kor-ta experimentella tidsskalan där laddade fragment detekteras några mikrose-kunder efter att de har skapats. Därför har vi byggt upp en ny experimentupp-ställning för jonlagrings- och fragmenteringsanalyser. Målet med detta nya ex-periment är att lagra internt varma fragment för att på ett kontrollerat sätt kunnastudera deras stabilitet på längre tidsskalor och som funktion av excitationse-nergin. Detta har betydelse när man ska bedöma om denna typ av processer,som kan förekomma ibland annat i samband med supernova explosioner, kan

påverka sekundära kemiska reaktionsnätverk på astronomiska tidsskalor.

List of Papers

The following papers, referred to in the text by their Roman numerals, areincluded in this thesis.

PAPER I: Failure of hydrogenation in protecting polycyclic aromatichydrocarbons from fragmentationM. Gatchell, M. H. Stockett, N. de Ruette, T. Chen, L. Giaco-mozzi, R. F. Nascimento, M. Wolf, E. K. Anderson, R. Delau-nay, V. Vizcaino, P. Rousseau, L. Adoui, B. A. Huber, H. T.Schmidt, H. Zettergren, and H. CederquistPhysical Review A 92, 050702(R) (2015).DOI: 10.1103/PhysRevA.92.050702

PAPER II: Hydrogenated pyrene: Statistical single-carbon loss belowthe knockout thresholdM. Wolf, L. Giacomozzi, M. Gatchell, N. de Ruette, M. H. Stock-ett, H. T. Schmidt, H. Cederquist, and H. ZettergrenThe European Physical Journal D 70, 85 (2016).DOI: 10.1140/epjd/e2016-60735-3

PAPER III: Photo-stability of super-hydrogenated PAHs determined byaction spectroscopy experimentsM. Wolf, H. V. Kiefer, J. Langeland, L. H. Andersen, H. Zetter-gren, H. T. Schmidt, H. Cederquist, and M. H. StockettThe Astrophysical Journal, 832, 24 (2016).DOI: 10.3847/0004-637X/832/1/24

PAPER IV: Threshold Energies for Single-Carbon Knockout from Poly-cyclic Aromatic HydrocarbonsM. H. Stockett, M. Gatchell, T. Chen, N. de Ruette, L. Gia-comozzi, M. Wolf, H. T. Schmidt, H. Zettergren, and H. Ced-erquistThe Journal of Physical Chemistry Letters 6, 4504–4509 (2015).DOI: 10.1021/acs.jpclett.5b02080

PAPER V: The threshold displacement energy of buckminsterfullereneand formation of endohedral defect fullerenesM. H. Stockett, M. Wolf, M. Gatchell, H. T. Schmidt, H. Zetter-gen, and H. Cederquist(Manuscript)

PAPER VI: Collision Induced Dissociation of the retinal ChromophoreSchiff Base from sub-eV to keV collision energiesK. Kulyk, M. Wolf, M. Gatchell, L. Giacomozzi, A. Vegvari,O. O. Kovalenko, N. de Ruette, O. F. Wendt, R. A. Zubarev,H. T. Schmidt, H. Cederquist, and H. ZettergrenThe Journal of Physical Chemistry A, (submitted)

Reprints were made with permission from the publishers.

Author’s contribution

From the beginning of my PhD the focus of my work was in the operation ofthe EISLAB beamline in Stockholm, which is also the main experimental setupused in the present work. As time progressed, I took over more responsibilitiesin operating and expanding the experimental setup, e.g. the installation of newion optics, measures to increase the available energy range, and the design ofnew experimental features. I was involved in the discussion, analysis, and in-terpretation of the results from experiments and simulations, the preparation offigures for publications, and the writing of publications. For Paper VI I carriedout minor DFT calculations.

PAPER I: I have actively taken part in the collision induced dissociationexperiments and the discussion of the experimental and simula-tion results.

PAPER II: I have actively taken part in the collision induced dissociationexperiments and the discussion of the experimental and simu-lation results. I have analyzed the experimental data, preparedthe figures, and wrote the main part of the manuscript. I am thecorresponding author.

PAPER III: I have actively taken part in the photo induced dissociation ex-periments at the ELISA ion storage ring in Aarhus, Denmark.I was involved in the discussion of the experimental and simu-lation results. I have analyzed the experimental data, preparedthe figures, and wrote the main part of the manuscript. I am thecorresponding author.

PAPER IV: I have actively taken part in the collision induced dissociationexperiments and the discussion of the experimental and simula-tion results.

PAPER V: I have actively taken part in the collision induced dissociationexperiments and the discussion of the experimental and simula-

tion results. I did the analysis of the experimental and simulationdata. I have written parts of the manuscript.

PAPER VI: I was responsible for preparation and data taking at the EISLABbeamline. I performed the DFT calculations of dissociation en-ergies for the paper, and I had a leading role in the internal revi-sion process of the manuscript.

This thesis [PhD] contains reprocessed, and in a few cases directly adapted,material from my licentiate thesis “Carbon backbone stability of PolycyclicAromatic Hydrocarbons” [Lic]. Below I list the way the material from [Lic]has been reused in this thesis [PhD]:

• Chapter 1: The parts concerning astrophysical PAHs in [PhD] are basedon the introduction of [Lic].

• Chapter 2: The description of the EISLAB beamline in [PhD] is basedon the experimental description in [Lic]. The section “A motor driven[...] bare ions remain” on pg. 7,9 in [PhD] has been taken with onlyminor modification from pg. 8 [Lic].

• Chapter 3: Section 3.1 and 3.2 [PhD] use reformulated parts of chapter3 [Lic]. Figure 3.5 [Lic] is used as figure 3.1 [PhD]. Figure 3.2 [Lic] isused as figure 3.6 [PhD]. Table 3.2 [Lic] is used as table 3.1 [PhD].

Contents

Abstract i

Sammanfattning iii

List of Papers v

Author’s contribution vii

1 Introduction 11.1 Molecules in space . . . . . . . . . . . . . . . . . . . . . . . 11.2 Biomolecules . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2 Experimental techniques 72.1 Collision induced dissociation in EISLAB . . . . . . . . . . . 72.2 Photoinduced dissociation at ELISA . . . . . . . . . . . . . . 132.3 Orbitrap mass spectrometer . . . . . . . . . . . . . . . . . . . 14

3 Results and discussion 173.1 Backbone stability of hydrogenated pyrene cations . . . . . . 173.2 Displacement energy of PAHs and C60 . . . . . . . . . . . . . 233.3 Collision induced fragmentation of the n-butylamine retinal

protonated Schiff base . . . . . . . . . . . . . . . . . . . . . 32

4 Ion storage experiments 374.1 Ion trap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

4.1.1 Trapping modes . . . . . . . . . . . . . . . . . . . . . 374.1.2 Daughter ion storage . . . . . . . . . . . . . . . . . . 39

4.2 Mass spectrometry . . . . . . . . . . . . . . . . . . . . . . . 404.3 Future development . . . . . . . . . . . . . . . . . . . . . . . 42

5 Concluding remarks 45

Acknowledgements xlvii

Additional publications xlix

References li

1. Introduction

1.1 Molecules in space

The primary focus of the present work lies in astrophysical relevant carbona-ceous molecules, more specifically Polycyclic aromatic hydrocarbons (PAHs)and fullerenes. The French name of PAHs - hydrocarbures aromatiques poly-cycliques - provides us with a very nice, systematic way to classify PAHmolecules:

• hydrocarbures - The molecules are hydrocarbons, i.e. they consist solelyof hydrogen- and carbon atoms.

• aromatiques - The molecules are aromatic: Aromatic hydrocarbons area subset of hydrocarbons, where carbon atoms form aromatic rings withdelocalized π electrons.

• polycycliques - The molecules contain multiple aromatic rings, wheremultiple usually means at least two.

The discovery of interstellar unidentified infrared radiation (UIR) in the 1970sand 1980s sparked interest in its source. A promising candidate as a carrierwas found in PAHs, as the prominent emission features of UIR at 3.3, 6.2, 7.7,8.6, and 11.2 μm are commonly ascribed to C-C and C-H bending and stretch-ing modes [1–5]. While it is widely accepted that at least some part of theUIR is emitted from PAHs, the identification of unique molecules as carriers iscomplicated by the fact that different PAH molecules often have very similaremission spectra due to their similar geometric properties. Nevertheless, theestablished consensus is that PAHs are an important constituent of interstellarmatter, with around 5–10% of the cosmic carbon bound in such molecules [6].

While PAHs might be present in space, they certainly don’t make up the bulkof interstellar matter. This honor falls to hydrogen, both in atomic and molec-ular form [5]. Despite their difference in size and complexity, interactionsbetween those molecules might play an important role for their production andsurvival in space. With PAHs most likely, and hydrogen certainly present inspace, it is not a huge leap to imagine that additional hydrogen atoms attach

�

latter reaction is the primary focus of our research, more specifically the caseof PAH-ion collisions in supernova shock waves, where typical collision en-ergies are in the range of 10–1000 eV [29]. Two types of ion-molecular in-teractions are present in such collisions: Interaction between the atom and theelectronic cloud (electronic stopping) on one hand, and Rutherford-like scat-tering processes of the projectile on a molecular atom (nuclear stopping) on theother hand. The first process leads to excited electronic states. The electronscan then relax through internal conversion, resulting in vibrationally excitedmolecules in the electronic ground state. If enough energy is deposited inthe molecule, it will then fragment predominantly through the lowest energydissociation channels, which for PAHs usually are H- and/or C2H2-loss withdissociation energies in the 5–7 eV range [24; 30; 31] and C2-loss at 10 eV forfullerenes [32; 33]. The loss of a single carbon atom on the other hand is asso-ciated with high reaction barriers and has a dissociation energy of about 15 eVboth for PAHs and fullerenes [34], making statistical fragmentation throughthis channel less likely.

The second type of interaction (nuclear stopping) can also heat up the moleculeand lead to statistical fragmentation. However, if the energy deposited is highenough it can also lead to direct atom knockout. Such knockout events onlyhappen above a certain knockout threshold energy. This has been observedfor isolated PAH molecules [34; 35], and it has been predicted by theory forfullerenes [36], although systematic studies of the latter are scarce. While for-mation of C+59 in C

−60+He collisions [37], of C

−59 from C

−60+He/Ne collisions

[38], and C+59 from C+60+Ne collisions [35] have been reported, the detection

of a fragment like C+59 is complicated by its inherent instability. C+59 readily

decays to e.g. C+58 [39], which can also be formed through statistical C2-loss.In such cases it is not possible to separate knockout fragmentation from sta-tistical fragmentation. Such knockout processes can result in highly reactivefragments, which in turn can form larger molecules as experiments on clustersof PAHs [40], neutral C60 [39; 41; 42], small hydrocarbon chains [43], andmixtures of C60 and PAHs have shown [35].

The relative strengths of the interactions depend on the center-of-mass energyand the involved interaction partners — for example in He-PAH collisions asin supernova shockwaves nuclear stopping dominates [44]. For PAHs, singlecarbon loss is an unfavorable statistical fragmentation pathway as the associ-ated dissociation energy of about 15 eV is much higher than the 5–7 eV for H-and/or C2H2-loss [24; 30; 31]. Any significant yield of CHx-loss in fragmen-tation spectra of PAHs can thus be attributed to direct single carbon knockout.By measuring the yield of the CHx-loss channel as a function of the collision

�

energy it is possible to determine the knockout threshold energy. For C60 thesame principle holds true, as it has been shown that statistical fragmentationoccurs through the loss of an even number of carbon atoms [45]. However,due to their three-dimensional, hollow structure fullerenes have an additional,fascinating property. They might form endohedral complexes, X@Cm wherean atom or molecule X is encapsulated inside the fullerene. While molecularendohedral fullerenes like H2@C60 or H2O@C60 often require the construc-tion of a carbon cage around the molecule through complex reactions [46; 47],atomic endohedral fullerenes can easily be produced in atom-fullerene colli-sions. The first atomic endohedrals from collisions, He@C+60 and He@C

+70,

were produced shortly after the discovery of fullerenes [48].

Related to the knockout threshold energy is the knockout threshold displace-ment energy, or displacement energy for short. To permanently displace anatom from its position in the molecule (i.e. for knockout to occur), a min-imum amount of kinetic energy has to be transferred to the atom. This en-ergy is the displacement energy Td [49], the energy transferred to the atomat the knockout threshold. If the molecules were rigid, Td would be the en-ergy required to break all bonds between the knock-on atom and its neigh-bors. As the molecules will bend and vibrate to preserve structural integrity,Td is well above this energy, and energy transfer below Td will not be suffi-cient to displace the atom. Instead the vibrations will heat up the molecule insuch collisions [49]. It should be noted that the strength of molecular bondsagainst breakage depends on several parameters, e.g. the direction of initialatom-displacement, leading to different displacement energies in moleculesand reported values are usually averages [49]. We have, for the first time,determined semi-empirical displacement energies for isolated molecules. Todo this purely experimentally it would be necessary to measure the energy ofall collision partners (fragments, atomic target) after the collision, somethingthat is not possible in the current experimental setup. Instead, we have useda combination of CID experiments and classical Molecular Dynamics (MD)simulations of the energy transfer to derive semi-empirical values for displace-ment energies. This method was first used in Paper IV for the PAH moleculesanthracene, pyrene and coronene, and in Paper V for C−60.

�

CH3

CH3 CH3

N+

H

CH3

C3H CH3

CH3

CH3

N+H

CH3

CH3

C3H CH3

CH3

CH3

N+

H

CH3

C3H CH3

+

CH3

Figure 1.3: Schematic of the process leading to formation of the fragment m/z= 248 (black) and a toluene molecule (red) from internally heated protonatedn-butylamine Schiff base of all-trans-retinal (NB-RPSB, top).

�

2. Experimental techniques

The main tool of choice in the present work is mass spectrometry, where, ingeneral, a mass analyzer is used to identify ions according to their mass-to-charge ratio. We use a tandem mass spectrometer (MS/MS) setup, where twostages of mass spectrometry are involved. The general principle of operationis that the first stage is used to select parent ions of the desired mass-to-chargeratio. Those ions then undergo some sort of interaction (collision, photoab-sorption) which will induce fragmentation. The resulting fragmentation prod-ucts are then analyzed by a second mass-to-charge analyzer [57]. The form ofinteraction in this work is either collision with a target gas (most commonlyHe), or photoexcitation. Collision Induced Dissociation (CID) experiments(Papers I, II, IV, V, and VI) were performed at the Electrospray IonizationSource Laboratory (EISLAB) as part of the Double ElectroStatic Ion RingExpEriment (DESIREE) facility [58; 59] at Stockholm University, Sweden.Photoinduced Dissociation (PID) experiments (Paper III) were performed atthe ELectrostatIc Storage ring for ions, Aarhus (ELISA) [60; 61] in Aarhus,Denmark. Additional low energy CID experiments of the protonated all-transn-butyl retinal Schiff base (NB-RPSB) (Paper VI) were done with an OrbitrapElite mass spectrometer at Karolinska Institute in Stockholm.

2.1 Collision induced dissociation in EISLAB

The experimental setup in EISLAB can be separated into two stages. The firststage, used for ion production and situated on a high voltage platform, is shownin figure 2.1, and has been extensively discussed in the PhD thesis of NicoleHaag [62]. To shortly summarize, gas phase ions are produced from a suitablesolution (e.g. for PAHs usually silver nitrate, dichloromethane, and methanol)by ElectroSpray Ionization (ESI). ESI is a gentle method of ion production,where the molecules receive small amounts of internal energy, making it idealto produce fragile molecules like biomolecules [50; 51] or hydrogenated poly-cyclic aromatic hydrocarbons (HPAHs) [63]. A motor driven syringe underatmospheric pressure is used to provide a constant flow to a needle, which iskept at a high potential of usually some kV. The needle is directed towardsa heated capillary at a lower potential of the order of a hundred volts. The

�

Cryostatic RingElectrode Trap

Capillary

Ion funnel

Octupole trap

Octupole guide

Quadrupole mass lter

Quadrupole de ector

Electron multiplier

Figure 2.1: Drawing of EISLAB ion platform. Adapted from the PhD thesis ofNicole Haag [62].

�

strong electric field arising from the potential difference leads to dispersion ofthe solution into a fine mist of highly charged droplets. Evaporation of thesolvent molecules, accelerated through heating of the capillary, reduces thesize of the droplets, while their charge remains constant. There are two com-peting models which explain how ions are produced. In the ion evaporationmodel (IEM) the surface field strength on the droplet becomes large enoughto produce ions by desorption [64]. In the charge residue model (CRM) thedroplet radius becomes too small for the surface tension to resist the chargerepulsion, resulting in a Coulomb explosion of the droplet into even smallercharged droplets. These smaller droplets continue to evaporate solvent and thecycle is repeated until all solvent is evaporated and only bare ions remain [65].For more information on ESI operation, we refer to reference [66]. The cap-illary acts as an atmospheric interface to the vacuum region of the experimentthrough which the ions enter a low-vacuum region with pressure of the orderof 3 ·10−1 mbar. In this first chamber an ion funnel is used to improve collec-tion and transmission of the ions [67; 68]. The ion funnel is followed by twodifferential pumping stages, the first containing an octupole trap, and the sec-ond containing an octupole guide. The octupole trap can be used to produceion bunches, though the CID experiments in the present work were all donewith a continuous ion beam. After the two octupoles a quadrupole mass filteris used to select ions of the desired mass-to-charge ratio. After mass selectiona quadrupole bend can be used to either deflect the molecules into a cryogenicRing Electrode Trap (RET), to an electron multiplier detector, or towards theexperimental section. The ring electrode trap can be used to cool and shortenthe bunches produced in the octupole trap, but it is not used in the present ex-periments. As the first available detector in the system, the electron multiplieris used for optimization and monitoring of ion production in the source.

The second, experimental stage, shown in figure 2.2, is on ground potentialand consists of a collision gas cell, ion optics, a pair of electrostatic horizontaldeflectors for kinetic-energy-per-charge selection after interaction in the gascell, and a position sensitive MCP detector (not shown in figure 2.2), which issituated approximately 1.3 m after the second horizontal deflector. The colli-sion energy is controlled by the potential difference between the two stages,as ions will gain kinetic energy according to ΔE = q ·ΔV , with the ion chargeq and the potential difference ΔV . This kinetic energy in the laboratory frametranslates to collision energies in the center-of-mass frame, ECM, of

ECM = EkinLAB

(mtarget

mtarget +mprojectile

), (2.1)

�

Gas cell

To detector

Einzel lens

Horizontal and vertical de ectors

Horizontal Slits

Einzel lens

From source

Horizontal de ectors

Figure 2.2: Drawing of the EISLAB experimental section. The horizontal de-flectors are followed by a 1.3 m long straight, empty section with a positionsensitive MCP detector at the end, which is not shown.

where mtarget and mprojectile are the masses of the neutral target atom (or molecule)and the projectile ion, respectively. The available potential on the high voltageplatform is ΔV = 0–14 kV. By scanning the voltage on the horizontal deflectors,we can scan the fragmentation products according to their energy-to-charge ra-tio, leading to energy spectra. Assuming that the velocity is the same before(for the projectile ion) and after (for charged fragments) the collision, such en-ergy spectra can be converted into mass spectra by simple scaling.

Absolute destruction cross sections for the projectile ion are measured withthe beam attenuation method [69] where the primary beam intensity I is mea-sured as a function of the gas cell pressure p. I decreases exponentially as

I(p) = I0e−p σLkbT (2.2)

with the intensity of the unperturbed beam I0, Boltzmann constant kB, temper-ature T , length of the gas cell L = 4 cm, and total destruction cross section σ .An example of such a measurement is shown in black in figure 2.3.Cross sections for specific channels σi can be obtained by measuring the in-crease of daughter ions i as a function of pressure. For low pressures a firstorder Taylor expansion of equation 2.2 gives

I(p)� I0 − I0 σLkbT p (2.3)

��

ion peak is not required for this method. As all fragments have to be recordedin the mass spectrum, it is unsuitable for anions which can lose electrons andform neutral and/or positive fragments.

2.2 Photoinduced dissociation at ELISA

Action spectroscopy experiments of hydrogenated pyrene cations were per-formed at the ELISA storage ring in Aarhus, Denmark, shown in figure 2.5[60; 61]. Like EISLAB in Stockholm ELISA utilizes an ESI source on a highvoltage platform for ion production. To provide bunches which can be injectedinto the storage ring, a 22-pole ion trap filled with He buffer gas is used tocollect, cool, and bunch the ions before acceleration. After acceleration, ionbunches are injected into the storage ring through a 90-degree bending magnetfor mass separation.

Figure 2.5: Schematic of ELISA, adapted from Paper III

After half a revolution in the ring, the ions are overlapped with a laser pulseand might absorb one or several photons sequentially. The absorbed electronicexcitation energy from those photoabsorption processes is converted into inter-nal vibrational energy, and the ions may then relax through dissociation. Anyneutral fragment formed while the ions are still in the first section (few mi-crosecond timescale – a “prompt” action signal [70]) will continue on a straightpath and can be detected with the Secondary Emission Detector (SED) [71].Excited ions that survive long enough to reach the second straight section andfragment there are detected on the MCP detector and give a “delayed” actionsignal.

ELISA can be operated as a MS-MS spectrometer similar to the EISLABbeamline. After photoexcitation and prompt fragmentation in the first straight

��

emanating from the HCD cell is used to cool the ions down and bunch them inthe center of the C-trap [72]. Ion bunches are accelerated into the HCD cell,a straight multipole inside a metal tube, by a potential difference of up to 5 Vbetween the C-trap and the encasing metal tube. Collision induced fragmenta-tion products are then returned to the C-trap and from there sideways extractedinto the Orbitrap for mass analysis [72]. The Oribtrap [75] is a modified King-don trap [76], consisting of three electrodes: A central spindle-like electrode isenclosed inside two outer cusp shaped ones (see figure 2.6). A DC voltage be-tween the central and outer electrodes generates a radial logarithmic potential,resulting in stable orbits around the central electrode [75] where the frequencyis dependent on the mass-to-charge ratio of the ions. The outer electrode isthen used to measure the image charge of the oscillating ions. Fast FourierTransformation of the image charge gives the frequency of the oscillations,and thereby the mass-to-charge ratio of stored ions.

��

��

3. Results and discussion

The experimental results presented in this thesis are grouped into three sec-tions. In section 3.1, I will discuss our studies of the effect of hydrogenationon the carbon backbone stability of pyrene molecules (Paper I–III). In section3.2, I will discuss our studies of the threshold displacement energies of PAHand C60 molecules (Paper IV, V). In section 3.3, I will discuss our studies ofthe energy dependence of the electrocyclization process in collision inducedfragmentation of the retinal protonated Schiff base.

3.1 Backbone stability of hydrogenated pyrene cations

The study of hydrogenation effects on the stability of the PAH carbon back-bone focuses on two aspects. The first is the total carbon backbone fragmenta-tion cross section, i.e. the sum of all cross sections for all fragmentation path-ways where at least one carbon atom has been lost (C16H+10+m −CxHy), x > 0from the pyrene cation in hydrogenation stage m. The dependency of this frag-mentation cross section on the degree of hydrogenation is a direct measure-ment of the backbone stability of the molecules in the collisions. The secondaspect is a measurement of the energy required to induce photodissociation.Here the molecules are irradiated with low energy (less than 3 eV) photons,and the number of photons (and therefore the total energy) absorbed beforefragmentation is derived from the experiment.

The total carbon backbone fragmentation cross sections for native and hy-drogenated pyrene cations, C16H+10+m (m = 0,6,16), in collisions with He aremeasured in attenuation measurements. In figure 3.1, those cross sections areshown as a function of the center-of-mass collision energy. At every collisionenergy, the absolute value of the total carbon backbone fragmentation crosssection is larger for hydrogenated than for native pyrene, and here it increaseswith the degree of hydrogenation. This shows that for those particular hy-drogenated molecules the possibility to lose additional, more weakly boundH-atoms does not fully compensate for the carbon backbone weakening due tostructural changes.

��

Figure 3.1: Absolute carbon-backbone fragmentation cross sections in collisionsbetween He and C16H+10+m with, from top to bottom, m = 16,6, and 0, as a func-tion of the collision energy. The statistical errors are one standard deviation. Thelines are to guide the eye. Adapted from Paper II.

Additional action spectroscopy experiments of the same molecules at theELISA ion storage ring were carried out to gain further insights into the del-icate balance between carbon backbone fragmentation and cooling throughhydrogen emission. There, the molecules were irradiated with laser pulses(λ = 400–650 nm, ∼ 2–3 eV/photon). In figure 3.2, prompt (fragmentationwithin a few microseconds) and delayed (fragmentation after ∼ 30 microsec-onds) action spectra are shown. The intensities in these spectra reflect thenumber of all neutral fragments from backbone fragmentation and hydrogenloss as a function of laser wavelength. This information is used to determinethe absorption maxima for fragmentation. We find maxima in the range of2.7–3 eV for all three ions, and an additional maximum for C16H+16 at 2.2 eV.

For the photon number dependency measurements, the laser wavelength isset to the measured absorption maximum. A continuously variable attenuatoris placed inside the laser beam path. The ion bunch is overlapped with a laserpulse to induce photofragmentation and the number of neutrals is measured.Then the attenuator is used to decrease the laser pulse energy by decreasingthe number of photons available, leading to fewer photoabsorption events. Theyield of neutral fragments, I, as a function of the laser pulse energy, P, followsa power law,

I(P) ∝ Pn (3.1)

��

Figure 3.3: Power dependence of total prompt fragmentation (including H-loss)for C16H+10+m,m = 16,6, or 0. The excitation energies are given in the bracket.The data follows power laws, with the exponents (see insets) giving the numberof photons absorbed before fragmentation. The deviation of the power law at lowenergies for pyrene (black dots) is due to hydrogen loss. The statistical errors areone standard deviation. Adapted from Paper III.

with n, the number of photons absorbed before fragmentation, being the expo-nent [77]. Figure 3.3 shows the total neutral yield on the SED as a functionof laser pulse energy. It should be noted that this includes not only backbonefragmentation, but also hydrogen losses.

Individual fragmentation channels can be measured by changing the electrodevoltages of the storage ring to accommodate fragments formed after photoex-citation instead of the intact molecule. After half a revolution in the ring thevoltages are turned off and the fragments hit the MCP detector. The resultingpower dependencies for individual fragmentation channels are shown in fig-ure 3.4. The photodependence results show that native pyrene (C16H+10) hasto absorb three photons, hexahydropyrene (C16H+10+6) has to absorb two, andhexadecahydropyrene (C16H+10+16) only needs to absorb one photon to frag-ment (the photons have similar, but not precisely the same energies for thethree ions). With the given photon energies, the total energies absorbed beforebackbone fragmentation are 8.16 eV (3× 2.72 eV) for C16H+10, 5.76 eV and4.36 eV (2× 2.88 eV and 2× 2.18 eV, respectively, for the two observed ab-sorption maxima) for C16H+16, and 2.95 eV for C16H

+26.

In table 3.1, density functional theory calculations for lowest dissociation en-ergy pathways of the three pyrene species are shown. Focusing only on back-

��

s������yH�s.�so

)0dy

�,�y���:y5mmy�y

�y�,xhmHs,0

��������������y

H�s.�s.)0d

�,�y���:54oy�

��y���:y54oy�

�,�y���:ym9oy�

��y���:ym9oy�

�y�,xooH90

������������������

H�s.�,.)0d

�4�y���y5,oy�

��y���:y5,oy�

�y�oxh9H.0

Figure 3.4: Power dependencies for individual fragmentation channels. The ex-ponents shown in the power laws on the right are weighted averages over allfragmentation channels for a given molecule. The statistical errors are one stan-dard deviation. Adapted from Paper III.

bone fragmentation channels, we find that C2H2 loss is the lowest energy dis-sociation channel for C16H+10. For the hydrogenated molecules on the otherhand, loss of CH3 is energetically more favorable with dissociation energiesof 2.26 eV and 1.60 eV for C16H+16 and C16H

+26, respectively. When compar-

ing the experimental results to theoretical values one has to keep in mind twothings. First, it is likely that transition states are involved in the fragmenta-tion. For example, loss of CH3 from the hydrogenated molecules requires anH migration before fragmentation. Thus, the energy needed for fragmentationmight be higher than the dissociation energy, and even if the barrier is lowerthan the dissociation energy these transition states will slow down the reac-tion. Second, if the excitation energy is only slightly above the dissociationenergy, fragmentation might not occur on the experimental timescale. Mea-surement on doubly charged anthracene (C14H2+10 ) have shown that excitationenergies above 10 eV are required to induce fragmentation on microsecondtimescales [24]. In case of the fully hydrogenated hexadecahydropyrene, thesingle 2.95 eV photon can activate all fragmentation pathways (neglecting anyadditional barriers) shown in table 3.1. This is also reflected in the fragmen-tation mass spectra, which display a wide range of smaller fragments, both inCID and PID, compared to the other molecules, as shown in figure 3.5.

Both the CID and PID measurements indicate an overall weakening effect ofhydrogenation on the carbon backbone of pyrene molecules. If this is the case

��

Table 3.1: Lowest dissociation energies for H-loss and carbon-backbone frag-mentation pathways (in eV) calculated with density functional theory (B3LYP/6-31G(d)). Adapted from Paper I.

DissociationChannel C16H+10 C16H

+16 C16H

+26

H 5.16 2.56 2.02H + H 8.67 5.19 5.13CHx 7.10 (x = 1) 2.26 (x = 3) 1.60 (x = 3)C2Hx 6.30 (x = 2) 3.88 (x = 4) 2.40 (x = 4)C3Hx 10.67 (x = 3) 6.24 (x = 6) 2.19 (x = 6)

for all degrees of hydrogenation, it is unlikely that hydrogenation protects thecarbon backbone of PAH molecules against fragmentation in the interstellarmedium. However, experiments by Reitsma et al. [13] have shown a protec-tive effect for coronene, albeit at lower hydrogenation states (up to 7 out of24 possible additional H atoms). In the present study on hydrogenated pyrenethe number was 6 and 16 out of 16 possible additional H atoms. This mightbe an indication that a low number of additional H atoms has a net protectiveeffect on the molecule, while for large numbers the weakening effect prevails.Additional studies are needed to clarify these aspects.

3.2 Displacement energy of PAHs and C60

The collisions in our experiments are nearly elastic, therefore the He target willalways have some kinetic energy after collision. To determine the displace-ment energy purely experimentally would require knowledge of this kineticenergy. In the current setup, this energy cannot be measured directly. Instead,we perform classical Molecular Dynamics (MD) simulations of the collisionsto determine theoretical values for the knockout cross sections and the en-ergy transfer to the molecular system as a whole in such knockout collisions.Through a combination of the experimental results for knockout threshold en-ergies with the MD energy transfers we obtain semi-empirical displacementenergies.

The knockout threshold displacement energy Td is the energy transferred tothe atom at the knockout threshold energy Eth. Eth is obtained from a measure-ment of the cross section for knockout of a single carbon atom as a function ofcollision energy. The knockout cross section follows a modified law for C-He

��

collisions [40] as

σKO =A/ECM

π2 arccos−2(√

Eth/ECM)−4

(3.2)

where A is a constant, and ECM is the center-of-mass collision energy forPAH+He collisions.

Experimental single carbon loss cross sections are derived from mass spectra.The blue line in figure 3.6 shows a fit to the single carbon loss peak for massspectra in He-Pyrene collisions with a range of 25–125 eV center-of-mass col-lision energy. The relative area of the peak is proportional to the cross sectionof the related process, and one can clearly see that the threshold for singlecarbon loss is between 25 and 30 eV collision energy. The single carbon losscross section as a function of collision energy is shown in the top part of fig-ure 3.7 (filled symbols) together with MD (open symbols) results for the PAHmolecules anthracene (C14H+10), pyrene (C16H

+16), and coronene (C24H

+12). The

lines are fits to equation 3.2. While the qualitative agreement between exper-iment and simulation is good, there is a noticeable shift δ = 8.5 ± 0.5 eVtowards higher threshold energies in the simulations. The origin of this shiftis still discussed, and it might be due to limitations in the simulations, e.g. anoverestimation of the C-C bond strength or the lack of treating electronic stop-ping in the simulations.

The lower part of figure 3.7 shows the mean energy lost by the He atom inknockout events, 〈ΔEHe〉, as a function of the center-of-mass energy. It is ob-tained from molecular dynamics simulations, where a He projectile with ran-dom start parameters is launched at a randomly oriented PAH molecule. Aver-aging over all impact parameters and orientations shows that 〈ΔEHe〉 follows apower law of

〈ΔEHe〉= (2.55±0.14) E0.636±0.013CM (3.3)in the experimentally relevant center-of-mass collision energy region of 30–150 eV, independent of the actual PAH molecule. Threshold energies fromequation 3.2 and equation 3.3 are used to derive Td . The experimental and the-oretical results are shown in table 3.2 and indicated in figure 3.7.

The present average semi-empirical displacement energies for three PAHmolecules (T SEd = 23.3± 0.3 eV) compare favorably with those for graphene(Td = 23.6±0.3 eV [78; 79]). As PAHs are basically small pieces of graphenewith H atoms terminating the edges, this comes not as a big surprise. Es-pecially for larger PAHs the central carbon atoms are indistinguishable from

��

Figure 3.7: Top panel: Normalized experimental (filled symbols) and MD sim-ulation (open symbols) single carbon knockout cross sections for anthracene(C14H+10), pyrene (C16H

+10), and coronene (C24H

+12) cations. The lines are fits

to the combined experimental (dashed line) and simulated (solid line) results us-ing equation 3.2. The statistical errors are one standard deviation.Bottom panel: The mean energy lost by the He atom in collisions that lead toknockout, 〈ΔEHe〉, is obtained from molecular dynamics simulations. The valuesof 〈ΔEHe〉 at the knockout thresholds, Eth, are the displacement energies, Td , andindicated by rings. The solid line is a fit to a power and given in equation 3.3.The statistical errors are smaller than the data points.Adapted from Paper IV.

��

as a function of the center-of-mass collision energy are shown in red alongwith the cross section for knockout directly extracted from MD simulations(green points in figure 3.8). In addition, we have estimated the cross sectionsfor statistical electron loss and added them to the knockout results (black). Wehave used a simple Arrhenius expression for the electron emission rate [81]

k = νe−Eb/kBTe f f . (3.4)

Here, the pre-exponential factor is set to ν = 1013 s−1 [82], the electron bind-ing energy of C−60 is Eb =2.664 eV [80], kB is Boltzmann’s constant, andthe effective temperature is Te f f = T −Eb/C, where C = 0.0138 K/eV is theheat capacity of C60 [83]. The internal temperature is approximately T [K] =1000+ (Eint [eV ])/C [83]. The contribution to the internal energy Eint fromthe energy transfer from He to C−60 is directly taken from our MD simulations.The electronic stopping contribution was estimated by scaling the results ofSchlathölter et al. [84], and a small internal energy correction of 0.44 eV ac-counting for the internal energy of room-temperature C−60 before the collision(the C−60 ions are assumed to be thermalized at 300 K during their transportfrom the source to the gas cell) [85]. We find that this simple model repro-duces our experimental results for total loss of C−60 ions from the primary beam(figure 3.8).

In figure 3.9, negative product mass spectra of C−60+He collisions in a center-of-mass energy range of ECoM = 22–77 eV are shown. For each spectrum, asum of Gaussian curves was fitted to the data (solid red line), which can besplit up into unique contributions of each peaks (dashed blue lines). The mainfragmentation products in the present energy range are found to be single anddouble carbon loss, and the corresponding endohedrals He@C−59 and He@C

−58.

It is known that C−59 forms through non-statistical single-carbon knockout inC−60+H collisions [38]. C

−58 is in the present experiment most likely formed

through secondary statistical C-loss following an initial knockout event. Wehave calculated the electron affinity of C−59 to be 3.56 eV, and the dissociationenergy for C-loss from C−59 is 4.54 eV according to the same calculations. Withsuch similar energies competition between C-loss and electron loss is to be ex-pected, leading to the formation of C−58 in some cases. The dissociation energyfor C2-loss (10 eV [32; 33]) from C−60 on the other hand is much larger than theelectron affinity (2.664 eV [80]), and hence statistical fragmentation throughthis channel is highly unlikely. This is supported by measurements of C2-lossfrom C2+60 , which has a dissociation energy of 10.19 eV for C2-loss [86], andrequires a minimal excitation energy of 40 eV for C2-loss to occur within 3.5μs [87]. The endohedrals He@C−59 and He@C

−58 are then formed in the same

way, except that in addition to fragmentation the He atom is also captured.

��

The experimental absolute single carbon knockout cross sections in figure3.10 are obtained by scaling the total area of negative fragment peaks pro-duced by an initial knockout event leading to any of the products (C−59, C

−58,

He@C−59,He@C−58) by the area of the peak corresponding to the transmitted

parent ion beam following equation 2.6. The MD simulations are handled inthe same way as for the total destruction cross section, but with the heat ca-pacity scaled for C59, and with an electron affinity of 3.56 eV, as calculatedat the B3LYP/6-31+G(d) level using Gaussian 09[88]. In figure 3.10, crosssections (top part) and mean energy loss of the He atom (bottom part) as afunction of center-of-mass energy are shown, following the same approach asfor PAHs in figure 3.7. The lower part of figure 3.10 shows the mean energylost by the He atom in such collisions that result in knockout, 〈ΔEHe〉, as afunction of the center-of-mass energy. It is obtained from molecular dynamicssimulations, where a He projectile with random start parameters is launched ata randomly oriented C−60 molecule [43]. Averaging over all impact parametersand orientations shows that 〈ΔEHe〉 follows a power law of

〈ΔEHe〉= (2.31±0.17) E0.655±0.015CM (3.5)

in the experimentally relevant center-of-mass collision energy region. The ex-perimental and theoretical threshold energies and displacement energies, alsoindicated in figure 3.10, are then

Eexpth = 35.8±0.5 eV T SEd = 24.1±0.5 eVEMDth = 41.8±1.5 eV T MDd = 26.5±0.8 eV

which is similar to the values found for PAHs and graphene[78; 79].

The measurements of displacement energies presented in this thesis are thefirst such results on isolated molecules. Astrophysical models concerning, forexample, the interaction between supernova shockwaves and PAH moleculesneed accurate values of the threshold displacement energy [29]. Despite theirimportance, experimental results for such displacement energies have not beenavailable earlier and instead theoretical values that are far too low have beenused. The first semi-empirical measurements of the displacement energies ofPAH and C−60 molecules reported here may therefore be useful as inputs inastrophysical models, for our understanding of atom-molecule interactions inspace.

��

3.3 Collision induced fragmentation of the n-butylamineretinal protonated Schiff base

Fragmentation mass spectra for collision between the n-butylamine retinal pro-tonated Schiff base (NB-RPSB) and He at center-of-mass collision energies inthe range ECM = 10–100 eV are shown in the top panel of figure 3.11. In thisenergy region, no fragment at m/z = 248 could be observed. Instead, all frag-mentation products can be explained by simple bond cleavages of the molecu-lar backbone (see figure 3.12). Results of low energy collisions between NB-RPSB and N2 are shown in the bottom part of figure 3.11. Here the protonatedn-butyl β -ionone Schiff base (NB-BISB) fragment at m/z = 248 is formed asa result of the collisions although this process only occurs in a narrow energywindow. The abundance of smaller fragments in these spectra indicates that itis possible for several collisions to take place, thereby increasing the energytransferred.

In figure 3.13 the ratio of abundances of NB-BISB to NB-RPSB ions as afunction of center-of-mass collision energy is shown. Formation of NB-BISBstarts around ECM = 0.35 eV, reaches maximal relative abundance around ECM= 2 eV, and disappears around ECM = 4 eV as the energy is increased further.The theoretical activation energy for this process as calculated by Coughlanet al. is 1.55 eV [55]. In addition, the simple bond cleavage CH3-loss frag-ment is shown, which starts around ECM = 1.8 eV, peaks at ECM = 2.9 eV, andvanishes at ECM = 3.8 eV. We have carried out DFT calculations for the disso-ciation energy for different CH3-loss pathways as shown in figure 3.14 usingGaussian09 [88] with the M06-2X functional and the cc-pVDZ basis set, andfind a lowest dissociation energy of 3.19 eV. This is the same method whichhas been used previously by Coughlan et al. [55] to calculate the energy bar-rier for electrocyclization. Both the energy barrier for electrocyclization andthe lowest dissociation energy for CH3-loss are indicated by red dotted linesin figure 3.13, and for both a shift towards lower energies can be seen for theappearance of the corresponding fragment in the experiment. Here one has tokeep in mind that the energy transfer to the molecule is smaller than the totalcenter-of-mass collision energy, but multiple collisions might easily heat upthe molecules to induce statistical fragmentation. Such a view is supported bythe relative constant shifts of 1.2 eV and 1.4 eV for NB-BISB formation andCH3-loss, respectively.

Interestingly, in collisions at much higher energies than the present ones (50keV NB-RPSB colliding with air) formation of NB-BISB has been observed[53]. On the other hand, the present experimental results show that NB-RPSB

��

Figure 3.11: Top: CID mass spectra of NB-RPSB in collision with He as afunction of center-of-mass energy in the range 10–100 eV, where no fragment atm/z = 248 is observed.Bottom: HCD mass spectra of NB-RPSB in collision with N2 as a function ofcenter-of-mass energy in the range 0–4.64 eV. The large abundance of smallerfragments indicates that multiple collisions take place.Adapted from Paper VI.

��

CH3

CH3 CH3

N+

H

CH3

C3H CH3

3.19 eV

4.18 eV

4.32 eV 4.00 eV

3.95 eV

Figure 3.14: Dissociation energies for CH3-loss calculated with the M06-2Xfunctional and the cc-pVDZ basis set in Gaussian 09 [88], and corrected for zeropoint energies. The lowest dissociation energy for CH3-loss is found to be 3.19eV.

will either break before rearrangement, or the NB-BISB will break apart afterformation if more energy than a few eV is put into the system, although this isstill an open question. A possible reason could be the difference in interactionat the different energies. At sub-keV energies nuclear interaction, which canlead to prompt knockout, will dominate. At the high collision energy used inthe experiment by Toker et al. [53] on the other hand, energy transfer throughinteraction with the electrons of the molecule is likely dominant.

��

��

4. Ion storage experiments

The knockout fragmentation products discussed in chapter 3 are often highlyreactive and have been shown to be involved in growth processes both for PAHs[89] and fullerenes [41; 42]. However, it should be noted that molecular growthstudies do not use isolated molecules, but molecular clusters, as targets. Thiscan limit the applicability of these results on isolated molecules in two ways.First, the cluster environment provides a way to release excess energy. Second,the reactive fragments are formed in close vicinity to partners with which theycan form bonds on very short timescales (typically within picoseconds). Inspace on the other hand, any excess energy remaining in an isolated molecularsystem after the initial fragmentation step might well lead to further, secondaryfragmentation before a reaction partner is encountered. The present EISLABexperiments can only give a lower limit for the fragments lifetime on the orderof some tens of microseconds (the flight time after excitation). To investigatewhether the molecules relax through non-destructive channels (i.e. throughphoton emission), we have performed simulations with SIMION 8 to design areflectron based time-of-flight mass spectrometer in combination with a smallelectrostatic Conetrap ion trap [90] between the horizontal energy analyzer andthe MCP detector. In this chapter I will discuss the development and imple-mentation of this novel setup, which as shown schematically in figure 4.1.

4.1 Ion trap

4.1.1 Trapping modes

The Conetrap ion trap used in our experiment consist of two cone-shaped endelectrodes, witch act as ion mirrors, and a central cylindrical ground electrodein between. Such ion traps are often operated in a symmetric storage mode, i.e.the cones are set to the same potential [90; 91], although other modes of oper-ation with non-symmetric potentials are possible [91; 92]. In the present work,we plan on using one of the two symmetric storage modes, of which SIMION8 simulations are shown in figure 4.2. The first trapping mode will here bereferred to as ‘soft’ storage mode due to the smooth (i.e. parallel and centered)

��

Position sensitiveMCP detector

Position sensitiveMCP detector

Re ectron

Einzel lens

Conetrap

Horizontal de ectors

Figure 4.1: Schematic of the new ion trapping and analysis setup.

reflection is shown at the top of figure 4.2. The second, ‘hard’ storage modeopens up at larger trapping potentials. As shown at the bottom of figure 4.2,ion trajectories in the hard storage mode take up a shorter, but broader volumein the ion trap. This broader volume arises from a larger angular acceptancefor stable trajectories.

In figure 4.3 we show simulations for the trapping efficiency for a beam of300 amu singly charged ions, with a kinetic energy of 3 keV. The trapping ef-ficiency (the number of ions trapped) is measured as a function of the trappingpotential relative to the ion kinetic energy and the trap closure time. In thosesimulations, the entrance cone is on ground potential at first while the exit coneis on the trapping potential. At the trap activation time, the entrance cone volt-age is switched to the trapping potential. The time axis in figure 4.3 is thereforea measurement of the ion position in the trap, while it is also dependent on thedirection in which the ion is moving. A longer time window corresponds toa longer path that ions can travel in the trap, something that can also easilybe seen from figure 4.2. The important information of those simulations isthe energy acceptance of the trapping modes, which is independent of the ionkinetic energy. Trapping for the soft mode is only viable in a narrow energywindow close to trapping potentials of 104% of the ion kinetic energy, whilethe hard mode has a broad energy acceptance in the range 120 – 145% of theion kinetic energy. In addition to the larger energy acceptance, the hard trap-ping mode has higher probabilities to trap ions with larger divergences fromthe ideal beam path, something that is crucial for efficient ion storage in our

��

neutrals leaving the trap will mostly hit the walls of the vacuum chamber.

Beam from gas cell Conetrap

Einzel lens

Position sensitiveMCP detector 2

Position sensitiveMCP detector 1

Re ectron

Figure 4.6: Schematic of the current time-of-flight setup (top) and the proposedturned configuration, which allows for additional measurement of neutrals leav-ing the trap (bottom).

��

��

5. Concluding remarks

Summary

In this thesis theoretical and experimental work on the fragmentation of com-plex molecules in collisions with gas phase targets was presented. For poly-cyclic aromatic hydrocarbons and fullerenes collisions with He targets are di-rectly applicable for astrophysical research. Such collisions are believed tobe important in certain astrophysical environments as e.g. in supernova shockwaves [29] and could also be of interest for processes in planetary atmospheres[95]. From measurements of the single carbon knockout cross section, and re-sults from MD simulations on typical energy transfers in such collisions, wehave for the first time determined semi-empirical displacement energies forisolated PAH and fullerene molecules — and in fact these are the first empiri-cally based determinations of displacement energies for any isolated molecule.We have further shown that certain hydrogenation states of the PAH moleculepyrene (C16H10) do not protect the carbon backbone against fragmentation fortwo different excitation methods (collisions and photo-absorptions). Resultson photo-induced dissociation of the same molecules, performed at the ionstorage ring ELISA (ELectrostatic Ion Storage ring, Aarhus) [61] at AarhusUniversity, suggests that superhydrogenation leads to a weakening of the car-bon backbone for low-energy photo excitation. This may suggest that theweakening effect is independent of the excitation method. However, it hasto be noted that no systematic study of all hydrogenation states was carriedout, and the effect of hydrogenation might well depend on the degree of hy-drogenation, and size and shape of the underlying (native) PAH molecule.We have used collision induced fragmentation of the n-butylamine retinal pro-tonated Schiff base (NB-RPSB) by He as means of energy deposition in themolecule. We have shown that the two-step electrocyclization process of theNB-RPSB only occurs in a narrow energy window, leading to more insight inthe complex mechanics of biomolecules, as for example their role for mammalvision.

��

Outlook

In chapter 4 I have presented a new experimental setup intended to extend theexperimental timescale of the EISLAB CID experiments. Due to the designions that fragment in the electrostatic ion-beam trap, Conetrap, can be ana-lyzed with a time-of-flight setup. Important applications of this experimentare studies of the survival of highly reactive PAH knockout fragments, whichmay relax through non-destructive channels, e.g. through electron- or photonemission processes. If this is the case these highly reactive fragments mightbe very important for astrophysical reaction networks. If the fragments sur-vive on extended timescales, reactions involving such highly reactive speciesand oppositely charged ions (atoms or molecules) will be studied in DESIREE[58; 59] under interstellar conditions. DESIREE consists of a pair of electro-static cryogenic storage rings with a common straight section, ideally suited tostudy such uni-molecular reactions on timescales ranging up to hours.

��

Acknowledgements

In the last four years, when I was working on my PhD, I was not alone. I wantto thank all those great people who I have met and worked with.

First, I thank my three supervisors Henning, Henning, and Henrik for giv-ing me the chance to do a PhD, and always supporting me along the way.

I also want to thank the whole atomic physics group for the great social atmo-sphere at work, which made the days spent in the office much more enjoyable.A special thank goes to my main collaborators in the EISLAB experiments,Linda, Nathalie, Mark, John, and Michael.

During my time at work I also made some good friends. Thanks Moa andJesper for everything from after work pubs and Song Contest parties to climb-ing and Boda Borg.

Finally, I want to thank Mariia for supporting and enduring me during thestressful time when I was writing this thesis.

Additional publications

i DESIREE electrospray ion source test bench and setup for CollisionInduced Dissociation experimentsN. de Ruette, M. Wolf, L. Giacomozzi, J. D. Alexander, M. Gatchell,M. H. Stockett, N. Haag, H. Zettergren, H. T. Schmidt, and H. CederquistManuscript (2017)

ii Knockout driven fragmentation of porphyrinsL. Giacomozzi, M. Gatchell, N. de Ruette, M. Wolf, G. D’Angelo, H. T.Schmidt, H. Cederquist, and H. ZettergrenPhysical Chemistry Chemical Physics, 19, 19750–19755, 2017.DOI: 10.1039/C7CP01583F

iii Non-statistical fragmentation of large molecules in collisions with atomsM. H. Stockett, L. Adoui, E. K. Anderson, T. Chen, J.-Y. Chesnel, N. deRuette, M. Gatchell, L. Giacomozzi, B. A. Huber, K. Kulyk, S. Maclot, P.Rousseau, M. Wolf, H. Zettergen, H. T. Schmidt, and H. CederquistJournal of Physics: Conference Series, 635, 012036, 2015.DOI: 10.1088/1742-6596/635/1/012036

References

[1] A. LEGER AND J. L. PUGET. Identification of the’unidentified’IR emission features of interstel-lar dust? Astronomy & Astrophysics, 137:L5–L8, 1984.

[2] L. J. ALLAMANDOLA, A. G. G. M. TIELENS, AND J. R. BARKER. Polycyclic aromatic hydro-carbons and the unidentified infrared emission bands-Auto exhaust along the Milky Way. TheAstrophysical Journal, 290:L25–L28, 1985.

[3] L. J. ALLAMANDOLA, A. G. G. M. TIELENS, AND J. R. BARKER. Interstellar polycyclic aro-matic hydrocarbons-The infrared emission bands, the excitation/emission mechanism, and theastrophysical implications. The Astrophysical Journal Supplement Series, 71:733–775, 1989.

[4] A. G. G. M. TIELENS. Interstellar Polycyclic Aromatic Hydrocarbon Molecules. Annual Reviewof Astronomy and Astrophysics, 46:289–337, 2008.

[5] A. G. G. M. TIELENS. The molecular universe. Reviews of Modern Physics, 85:1021–1081, 2013.

[6] A. G. G. M. TIELENS. 25 years of PAH hypothesis. EAS Publications Series, 46:3–10, 2011.

[7] M. P. BERNSTEIN, S. A. SANDFORD, AND L. J. ALLAMANDOLA. Hydrogenated polycyclicaromatic hydrocarbons and the 2940 and 2850 wavenumber (3.40 and 3.51 micron) infraredemission features. The Astrophysical Journal Letters, 472(2):L127, 1996.

[8] E. RAULS AND L. HORNEKÆR. Catalyzed routes to molecular hydrogen formation and hydro-gen addition reactions on neutral polycyclic aromatic hydrocarbons under interstellar condi-tions. The Astrophysical Journal, 679(1):531, 2008.

[9] V. MENNELLA, L. HORNEKÆR, J. THROWER, AND M. ACCOLLA. The catalytic role of coronenefor molecular hydrogen formation. The Astrophysical Journal Letters, 745(1):L2, 2011.

[10] L. BOSCHMAN, G. REITSMA, S. CAZAUX, T. SCHLATHÖLTER, R. HOEKSTRA, M. SPAANS, ANDO. GONZALEZ-MAGANA. Hydrogenation of PAH Cations: A First Step toward H2 Formation.The Astrophysical Journal Letters, 761(2):L33, 2012.

[11] L. BOSCHMAN, S. CAZAUX, M. SPAANS, R. HOEKSTRA, AND T. SCHLATHÖLTER. H2 formationon PAHs in photodissociation regions: a high-temperature pathway to molecular hydrogen.Astronomy & Astrophysics, 579:A72, 2015.

[12] M. HIRAMA, T. TOKOSUMI, T. ISHIDA, AND J. AIHARA. Possible molecular hydrogen forma-tion mediated by the inner and outer carbon atoms of typical PAH cations. Chemical Physics,305(1):307–316, 2004.

[13] G. REITSMA, L. BOSCHMAN, M. J. DEUZEMAN, O. GONZÁLEZ-MAGAÑA, S. HOEKSTRA,S. CAZAUX, R. HOEKSTRA, AND T. SCHLATHÖLTER. Deexcitation Dynamics of Superhydro-genated Polycyclic Aromatic Hydrocarbon Cations after Soft-x-Ray Absorption. Physical Re-view Letters, 113:053002, 2014.

[14] P. J. SARRE. The diffuse interstellar bands: A major problem in astronomical spectroscopy.Journal of Molecular Spectroscopy, 238(1):1 – 10, 2006.

[15] T. P. SNOW, V. LE PAGE, Y. KEHEYAN, AND V. M. BIERBAUM. The interstellar chemistry ofPAH cations. Nature, 391(6664):259–260, 1998.

[16] E. K. CAMPBELL, M. HOLZ, D. GERLICH, AND J. P. MAIER. Laboratory confirmation of C+60as the carrier of two diffuse interstellar bands. Nature, 523(7560):322–323, 2015.

[17] G. A. H. WALKER, D. A. BOHLENDER, J. P. MAIER, AND E. K. CAMPBELL. Identification ofMore Interstellar C+60 Bands. The Astrophysical Journal Letters, 812(1):L8, 2015.

[18] T. ALLAIN, S. LEACH, AND E. SEDLMAYR. Photodestruction of PAHs in the interstellarmedium. I. Photodissociation rates for the loss of an acetylenic group. Astronomy & Astrophysics,305:602, 1996.

[19] T. ALLAIN, S. LEACH, AND E. SEDLMAYR. Photodestruction of PAHs in the interstellarmedium. II. Influence of the states of ionization and hydrogenation. Astronomy & Astrophysics,305:616, 1996.

[20] B. WEST, F. USELI-BACCHITTA, H. SABBAH, V. BLANCHET, A. BODI, P. M. MAYER, ANDC. JOBLIN. Photodissociation of Pyrene Cations: Structure and Energetics from C16H+10 to C

+14

and Almost Everything in Between. The Journal of Physical Chemistry A, 118(36):7824–7831,2014.

[21] E. R. MICELOTTA, A. P. JONES, AND A. G. G. M. TIELENS. Polycyclic aromatic hydrocarbonprocessing by cosmic rays. Astronomy & Astrophysics, 52(arXiv: 1012.1599):A52, 2010.

[22] E. R. MICELOTTA, A. P. JONES, AND A. G. G. M. TIELENS. Polycyclic aromatic hydrocarbonprocessing in a hot gas. Astronomy & Astrophysics, 510:A37, 2010.

[23] J. POSTMA, S. BARI, R. HOEKSTRA, A. G. G. M. TIELENS, AND T. SCHLATHÖLTER. Ionizationand Fragmentation of Anthracene upon Interaction with keV Protons and α Particles. TheAstrophysical Journal, 708(1):435, 2010.

[24] S. MARTIN, L. CHEN, R. BRÉDY, G. MONTAGNE, C. ORTEGA, T. SCHLATHÖLTER, G. REITSMA,AND J. BERNARD. Statistical fragmentation of doubly charged anthracene induced by fluorine-beam impact at 3 keV. Physical Review A, 85(5):052715, 2012.

[25] G. REITSMA, H. ZETTERGREN, L. BOSCHMAN, E. BODEWITS, R. HOEKSTRA, ANDT. SCHLATHÖLTER. Ion–polycyclic aromatic hydrocarbon collisions: kinetic energy releasesfor specific fragmentation channels. Journal of Physics B: Atomic, Molecular and Optical Physics,46(24):245201, 2013.

[26] J. POSTMA, R. HOEKSTRA, A. G. G. M. TIELENS, AND T. SCHLATHÖLTER. A Molecular Dy-namics Study on Slow Ion Interactions with the Polycyclic Aromatic Hydrocarbon MoleculeAnthracene. The Astrophysical Journal, 783(1):61, 2014.

[27] P. M. MISHRA, J. RAJPUT, C. P. SAFVAN, S. VIG, AND U. KADHANE. Velocity dependence offragmentation yields in protonâĂŞnaphthalene collision and comparison with electronic energyloss calculation. Journal of Physics B: Atomic, Molecular and Optical Physics, 47(8):085202, 2014.

[28] M. H. STOCKETT, M. GATCHELL, J. D. ALEXANDER, U. BERZINS, T. CHEN, K. FARID, A. JO-HANSSON, K. KULYK, P. ROUSSEAU, K. STOCHKEL, L. ADOUI, P. HVELPLUND, B. A. HUBER,H. T. SCHMIDT, H. ZETTERGREN, AND H. CEDERQUIST. Fragmentation of anthracene C14H10,acridine C13H9N and phenazine C12H8N2 ions in collisions with atoms. Physical ChemistryChemical Physics, 16:21980, 2014.

[29] E. R. MICELOTTA, A. P. JONES, AND A. G. G. M. TIELENS. Polycyclic aromatic hydrocarbonprocessing in interstellar shocks. Astronomy & Astrophysics, 510:A36, 2010.

[30] A. I. S. HOLM, H. A. B. JOHANSSON, H. CEDERQUIST, AND H. ZETTERGREN. Dissociation andmultiple ionization energies for five polycyclic aromatic hydrocarbon molecules. The Journal ofChemical Physics, 134(4):044301, 2011.

[31] C. PARIS, M. ALCAMÍ, F. MARTÍN, AND S. DÍAZ-TENDERO. Multiple ionization and hy-drogen loss from neutral and positively-charged coronene. The Journal of Chemical Physics,140(20):204307, 2014.

[32] S. TOMITA, J. U. ANDERSEN, C. GOTTRUP, P. HVELPLUND, AND U. V. PEDERSEN. DissociationEnergy for C2 Loss from Fullerene Cations in a Storage Ring. Physical Review Letters, 87:073401,2001.

[33] S. TOMITA, J. U. ANDERSEN, K. HANSEN, AND P. HVELPLUND. Stability of Buckminster-fullerene, C60. Chemical Physics Letters, 382(1):120 – 125, 2003.

[34] M. H. STOCKETT, H. ZETTERGREN, L. ADOUI, J. D. ALEXANDER, U. BĒRZIŅŠ, T. CHEN,M. GATCHELL, N. HAAG, B. A. HUBER, P. HVELPLUND, A. JOHANSSON, H. A. B. JOHANSSON,K. KULYK, S. ROSÉN, P. ROUSSEAU, K. STØCHKEL, H. T. SCHMIDT, AND H. CEDERQUIST.Nonstatistical fragmentation of large molecules. Physical Review A, 89:032701, 2014.

[35] M. GATCHELL, M. H. STOCKETT, P. ROUSSEAU, T. CHEN, K. KULYK, H. T. SCHMIDT, J.Y.CHESNEL, A. DOMARACKA, A. MÉRY, S. MACLOT, L. ADOUI, K. STØCHKEL, P. HVELPLUND,Y. WANG, M. ALCAMÍ, B. A. HUBER, F. MARTÍN, H. ZETTERGREN, AND H. CEDERQUIST. Non-statistical fragmentation of PAHs and fullerenes in collisions with atoms. International Journalof Mass Spectrometry, 365–366(0):260 – 265, 2014. Special issue: Tilmann mark.

[36] F. Z. CUI, H. D. LI, AND X. Y. HUANG. Atomistic simulation of radiation damage to C60.Physical Review B, 49(14):9962, 1994.

[37] M. C. LARSEN, P. HVELPLUND, M. O. LARSSON, AND H. SHEN. Fragmentation of fast posi-tive and negative C60 ions in collisions with rare gas atoms. The European Physical Journal D,5(2):283–289, 1999.

[38] S. TOMITA, P. HVELPLUND, S. B. NIELSEN, AND T. MURAMOTO. C59-ion formation in high-energy collisions between cold C−60 and noble gases. Physical Review A, 65:043201, 2002.

[39] Y. WANG, H. ZETTERGREN, P. ROUSSEAU, T. CHEN, M. GATCHELL, M. H. STOCKETT, A. DO-MARACKA, L. ADOUI, B. A. HUBER, H. CEDERQUIST, M. ALCAMÍ, AND F. MARTÍN. Formationdynamics of fullerene dimers C+118, C

+119, and C

+120. Physical Review A, 89:062708, 2014.

[40] T. CHEN, M. GATCHELL, M. H. STOCKETT, J. D. ALEXANDER, Y. ZHANG, P. ROUSSEAU,A. DOMARACKA, S. MACLOT, R. DELAUNAY, L. ADOUI, B. A. HUBER, T. SCHLATHÖLTER,H. T. SCHMIDT, H. CEDERQUIST, AND H. ZETTERGREN. Absolute fragmentation cross sec-tions in atom-molecule collisions: Scaling laws for non-statistical fragmentation of polycyclicaromatic hydrocarbon molecules. The Journal of Chemical Physics, 140(22):224306, 2014.

[41] H. ZETTERGREN, P. ROUSSEAU, Y. WANG, F. SEITZ, T. CHEN, M. GATCHELL, J. D. ALEXAN-DER, M. H. STOCKETT, J. RANGAMA, J. Y. CHESNEL, M. CAPRON, J. C. POULLY, A. DO-MARACKA, A. MÉRY, S. MACLOT, H. T. SCHMIDT, L. ADOUI, M. ALCAMÍ, A. G. G. M. TIE-LENS, F. MARTÍN, B. A. HUBER, AND H. CEDERQUIST. Formations of Dumbbell C118 andC119 inside Clusters of C60 Molecules by Collision with α Particles. Physical Review Letters,110:185501, 2013.

[42] F. SEITZ, H. ZETTERGREN, P. ROUSSEAU, Y. WANG, T. CHEN, M. GATCHELL, J. D. ALEXAN-DER, M. H. STOCKETT, J. RANGAMA, J. Y. CHESNEL, M. CAPRON, J. C. POULLY, A. DO-MARACKA, A. MÉRY, S. MACLOT, V. VIZCAINO, H. T. SCHMIDT, L. ADOUI, M. ALCAMÍ, A. G.G. M. TIELENS, F. MARTÍN, B. A. HUBER, AND H. CEDERQUIST. Ions colliding with clustersof fullerenes–Decay pathways and covalent bond formations. The Journal of Chemical Physics,139(3):034309, 2013.

[43] M. GATCHELL, R. DELAUNAY, G. D’ANGELO, A. MIKA, K. KULYK, A. DOMARACKA,P. ROUSSEAU, H. ZETTERGREN, B. A. HUBER, AND H. CEDERQUIST. Ion-induced moleculargrowth in clusters of small hydrocarbon chains. Physical Chemistry Chemical Physics, 19:19665–19672, 2017.

[44] M. GATCHELL AND H. ZETTERGREN. Knockout driven reactions in complex molecules andtheir clusters. Journal of Physics B: Atomic, Molecular and Optical Physics, 49(16):162001, 2016.

[45] A. B. YOUNG, L. M. COUSINS, AND A. G. HARRISON. A collisional study of some C60 and C70fullerene ions. Rapid Communications in Mass Spectrometry, 5(5):226–229, 1991.

[46] K. KOMATSU, M. MURATA, AND Y. MURATA. Encapsulation of Molecular Hydrogen inFullerene C60 by Organic Synthesis. Science, 307(5707):238–240, 2005.

[47] K. KUROTOBI AND Y. MURATA. A Single Molecule of Water Encapsulated in Fullerene C60.Science, 333(6042):613–616, 2011.

[48] T. WEISKE, D. K. BÖHME, J. HRUŠÁK, W. KRÄTSCHMER, AND H. SCHWARZ. EndohedralCluster Compounds: Inclusion of Helium within C·⊕60 and C

·⊕70 through Collision Experiments.

Angewandte Chemie International Edition in English, 30(7):884–886, 1991.

[49] The Displacement of Atoms, pages 73–124. Springer Berlin Heidelberg, Berlin, Heidelberg, 2007.

[50] J. B. FENN, M. MANN, C. K. MENG, S. F. WONG, AND C. M. WHITEHOUSE. Electrosprayionization for mass spectrometry of large biomolecules. Science, 246(4926):64–71, 1989.

[51] M. MANN. Electrospray: Its potential and limitations as an ionization method for biomolecules.Organic Mass Spectrometry, 25(11):575–587, 1990.

[52] Y. TOKER, A. SVENDSEN, A. V. BOCHENKOVA, AND L. H. ANDERSEN. Probing the Barrierfor Internal Rotation of the Retinal Chromophore. Angewandte Chemie International Edition,51(35):8757–8761, 2012.

[53] Y. TOKER, D. B. RAHBEK, H. V. KIEFER, J. RAJPUT, R. ANTOINE, P. DUGOURD, S. BROND-STED NIELSEN, A. V. BOCHENKOVA, AND L. H. ANDERSEN. Photoresponse of the proto-nated Schiff-base retinal chromophore in the gas phase. Physical Chemistry Chemical Physics,15:19566–19569, 2013.

[54] L. MUSBAT, M. NIHAMKIN, S. YTZHAK, A. HIRSHFELD, N. FRIEDMAN, J. M. DILGER,M. SHEVES, AND Y. TOKER. Isotope Labeling Study of Retinal Chromophore Fragmentation.The Journal of Physical Chemistry A, 120(16):2547–2549, 2016. PMID: 27046667.

[55] N. J. A. COUGHLAN, B. D. ADAMSON, K. J. CATANI, U. WILLE, AND E. J. BIESKE. Ion MobilityUnlocks the Photofragmentation Mechanism of Retinal Protonated Schiff Base. The Journal ofPhysical Chemistry Letters, 5(18):3195–3199, 2014. PMID: 26276331.

[56] N. J. A. COUGHLAN, B. D. ADAMSON, L. GAMON, K. CATANI, AND E. J. BIESKE. Retinalshows its true colours: photoisomerization action spectra of mobility-selected isomers of theretinal protonated Schiff base. Physical Chemistry Chemical Physics, 17:22623–22631, 2015.

[57] F. W. MCLAFFERTY. Tandem mass spectrometry. Science, 214(4518):280–287, 1981.

[58] R. D. THOMAS, H. T. SCHMIDT, G. ANDLER, M. BJÖRKHAGE, M. BLOM, L. BRÄNNHOLM,E. BÄCKSTRÖM, H. DANARED, S. DAS, N. HAAG, P. HALLDÉN, F. HELLBERG, A. I. S. HOLM,H. A. B. JOHANSSON, A. KÄLLBERG, G. KÄLLERSJÖ, M. LARSSON, S. LEONTEIN, L. LIL-JEBY, P. LÖFGREN, B. MALM, S. MANNERVIK, M. MASUDA, D. MISRA, A. ORBÃĄN, A. PAÁL,P. REINHED, K.-G. RENSFELT, S. ROSÉN, K. SCHMIDT, F. SEITZ, A. SIMONSSON, J. WEIMER,H. ZETTERGREN, AND H. CEDERQUIST. The double electrostatic ion ring experiment: A uniquecryogenic electrostatic storage ring for merged ion-beams studies. Review of Scientific Instru-ments, 82(6), 2011.

[59] H. T. SCHMIDT, R. D. THOMAS, M. GATCHELL, S. ROSÉN, P. REINHED, P. LÖFGREN,L. BRÄNNHOLM, M. BLOM, M. BJÖRKHAGE, E. BÄCKSTRÖM, J. D. ALEXANDER, S. LEON-TEIN, D. HANSTORP, H. ZETTERGREN, L. LILJEBY, A. KÄLLBERG, A. SIMONSSON, F. HELL-BERG, S. MANNERVIK, M. LARSSON, W. D. GEPPERT, K. G. RENSFELT, H. DANARED,A. PAÁL, M. MASUDA, P. HALLDÉN, G. ANDLER, M. H. STOCKETT, T. CHEN, G. KÄLLER-SJÖ, J. WEIMER, K. HANSEN, H. HARTMAN, AND H. CEDERQUIST. First storage of ion beamsin the Double Electrostatic Ion-Ring Experiment: DESIREE. Review of Scientific Instruments,84(5), 2013.

[60] S. P. MOLLER. ELISA, an electrostatic storage ring for atomic physics. Nuclear Instrumentsand Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and AssociatedEquipment, 394(3):281 – 286, 1997.

[61] J. U. ANDERSEN, P. HVELPLUND, S. BRØNDSTED NIELSEN, S. TOMITA, H. WAHLGREEN, S. P.MØLLER, U. V. PEDERSEN, J. S. FORSTER, AND T. J. D. JØRGENSEN. The combination of anelectrospray ion source and an electrostatic storage ring for lifetime and spectroscopy experi-ments on biomolecules. Review of Scientific Instruments, 73(3):1284–1287, 2002.

[62] N. HAAG. Probing biomolecular fragmentation. Phd thesis, Stockholm University, 2011.

[63] B. KLÆRKE, Y. TOKER, D. B. RAHBEK, L. HORNEKÆR, AND L. H. ANDERSEN. Formation andstability of hydrogenated PAHs in the gas phase. Astronomy & Astrophysics, 549:A84, 2013.

[64] J. V. IRIBARNE AND B. A. THOMSON. On the evaporation of small ions from charged droplets.The Journal of Chemical Physics, 64(6), 1976.

[65] M. DOLE, L. L. MACK, R. L. HINES, R. C. MOBLEY, L. D. FERGUSON, AND M. B. ALICE.Molecular Beams of Macroions. The Journal of Chemical Physics, 49(5):2240–2249, 1968.

[66] P. KEBARLE AND U. H. VERKERK. Electrospray: From ions in solution to ions in the gas phase,what we know now. Mass Spectrometry Reviews, 28(6):898–917, 2009.

[67] T. KIM, A. V. TOLMACHEV, R. HARKEWICZ, D. C. PRIOR, G. ANDERSON, H. R. UDSETH, R. D.SMITH, T. H. BAILEY, S. RAKOV, AND J. H. FUTRELL. Design and Implementation of a NewElectrodynamic Ion Funnel. Analytical Chemistry, 72(10):2247–2255, 2000.

[68] R. T. KELLY, A. V. TOLMACHEV, J. S. PAGE, K. TANG, AND R. D. SMITH. The ion funnel:Theory, implementations, and applications. Mass Spectrometry Reviews, 29(2):294–312, 2010.

[69] P. B. ARMENTROUT. Fundamentals of ion–molecule chemistry. Journal of Analytical AtomicSpectrometry, 19(5):571–580, 2004.

[70] L. LAMMICH, J. RAJPUT, AND L. H. ANDERSEN. Photodissociation pathways of gas-phasephotoactive yellow protein chromophores. Physical Review E, 78:051916, 2008.

[71] D. HANSTORP. A secondary emission detector capable of preventing detection of the photoelec-tric effect induced by pulsed lasers. Measurement Science and Technology, 3(5):523, 1992.

[72] THERMO FISHER SCIENTIFIC. Orbitrap Elite Hardware Manual. Thermo Fisher Scientific Inc,2011.

[73] THERMO FISHER SCIENTIFIC. LTQ Series Hardware Manual. Thermo Fisher Scientific Inc., 2012.

[74] J. V. OLSEN, J. C. SCHWARTZ, J. GRIEP-RAMING, M. L. NIELSEN, E. DAMOC, E. DENISOV,O. LANGE, P. REMES, D. TAYLOR, M. SPLENDORE, E. R. WOUTERS, M. SENKO, A. MAKAROV,M. MANN, AND S. HORNING. A dual pressure linear ion trap Orbitrap instrument with veryhigh sequencing speed. Molecular & cellular proteomics, 8(12):2759–2769, 2009.

[75] Q. HU, R. J. NOLL, H. LI, A. MAKAROV, M. HARDMAN, AND R. G. COOKS. The Orbitrap: anew mass spectrometer. Journal of Mass Spectrometry, 40(4):430–443, 2005.

[76] K. H. KINGDON. A Method for the Neutralization of Electron Space Charge by Positive Ioniza-tion at Very Low Gas Pressures. Physical Review, 21:408–418, 1923.

[77] L. H. ANDERSEN, H. BLUHME, S. BOYE, T. J. D. JORGENSEN, H. KROGH, I. B. NIELSEN,S. BRONDSTED NIELSEN, AND A. SVENDSEN. Experimental studies of the photophysics of gas-phase fluorescent protein chromophores. Physical Chemistry Chemical Physics, 6:2617–2627,2004.

[78] J. C. MEYER, F. EDER, S. KURASCH, V. SKAKALOVA, J. KOTAKOSKI, H. J. PARK, S. ROTH,A. CHUVILIN, S. EYHUSEN, G. BENNER, A. V.. KRASHENINNIKOV, AND U. KAISER. Accu-rate Measurement of Electron Beam Induced Displacement Cross Sections for Single-LayerGraphene. Physical Review Letters, 108:196102, 2012.

[79] J. C. MEYER, F. EDER, S. KURASCH, V. SKAKALOVA, J. KOTAKOSKI, H. J. PARK, S. ROTH,A. CHUVILIN, S. EYHUSEN, G. BENNER, A. V. KRASHENINNIKOV, AND U. KAISER. Erratum:Accurate Measurement of Electron Beam Induced Displacement Cross Sections for Single-Layer Graphene [Phys. Rev. Lett. 108 , 196102 (2012)]. Physical Review Letters, 110:239902,2013.

[80] K. STØCHKEL AND J. U. ANDERSEN. Photo excitation and laser detachment of C−60 anions in astorage ring. J. Chem. Phys., 139(16):164304, 2013.

[81] S. TOMITA, P. HVELPLUND, S. B. NIELSEN, AND T. MURAMOTO. C59-ion formation in high-energy collisions between cold C−60 and noble gases. Phys. Rev. A, 65:043201, 2002.

[82] J. U. ANDERSEN AND E. BONDERUP. Classical dielectric models of fullerenes and estimation ofheat radiation. The European Physical Journal D-Atomic, Molecular, Optical and Plasma Physics,11(3):413–434, 2000.

[83] J. U. ANDERSEN, C. GOTTRUP, K. HANSEN, P. HVELPLUND, AND M. O. LARSSON. Radiativecooling of fullerene anions in a storage ring. The European Physical Journal D-Atomic, Molecular,Optical and Plasma Physics, 17(2):189–204, 2001.

[84] T. SCHLATHÖLTER, O. HADJAR, R. HOEKSTRA, AND R. MORGENSTERN. Strong Velocity Effectsin Collisions of He+ with Fullerenes. Physical Review Letters, 82:73–76, 1999.

[85] R. K. YOO, B. RUSCIC, AND J. BERKOWITZ. Vacuum ultraviolet photoionization mass spectro-metric study of C60. J Chem Phys, 96(2):911–918, 1992.

[86] S. DÍAZ-TENDERO, M. ALCAMÍ, AND F. MARTÍN. Theoretical study of ionization potentialsand dissociation energies of Cnq+ fullerenes (n=50–60, q=0, 1 and 2). The Journal of ChemicalPhysics, 119(11):5545–5557, 2003.

[87] L. CHEN, S. MARTIN, J. BERNARD, AND R. BRÉDY. Direct Measurement of Internal Energy ofFragmented C60. Phys. Rev. Lett., 98:193401, 2007.

[88] M. J. FRISCH, G. W. TRUCKS, H. B. SCHLEGEL, G. E. SCUSERIA, M. A. ROBB, J. R. CHEESE-MAN, G. SCALMANI, V. BARONE, B. MENNUCCI, G. A. PETERSSON, H. NAKATSUJI, M. CAR-ICATO, X. LI, H. P. HRATCHIAN, A. F. IZMAYLOV, J. BLOINO, G. ZHENG, J. L. SONNEN-BERG, M. HADA, M. EHARA, K. TOYOTA, R. FUKUDA, J. HASEGAWA, M. ISHIDA, T. NAKA-JIMA, Y. HONDA, O. KITAO, H. NAKAI, T. VREVEN, J. A. MONTGOMERY, J. E. PERALTA,F. OGLIARO, M. BEARPARK, J. J. HEYD, E. BROTHERS, K. N. KUDIN, V. N. STAROVEROV,R. KOBAYASHI, J. NORMAND, K. RAGHAVACHARI, A. RENDELL, J. C. BURANT, S. S. IYEN-GAR, J. TOMASI, M. COSSI, N. REGA, J. M. MILLAM, M. KLENE, J. E. KNOX, J. B. CROSS,V. BAKKEN, C. ADAMO, J. JARAMILLO, R. GOMPERTS, R. E. STRATMANN, O. YAZYEV, A. J.AUSTIN, R. CAMMI, C. POMELLI, J. W. OCHTERSKI, R. L. MARTIN, K. MOROKUMA, V. G.ZAKRZEWSKI, G. A. VOTH, P. SALVADOR, J. J. DANNENBERG, S. DAPPRICH, A. D. DANIELS,FARKAS, J. B. FORESMAN, J. V. ORTIZ, J. CIOSLOWSKI, AND D. J. FOX. Gaussian 09, RevisionD.01, 2009.

[89] R. DELAUNAY, M. GATCHELL, P. ROUSSEAU, A. DOMARACKA, S. MACLOT, Y. WANG, M. H.STOCKETT, T CHEN, L. ADOUI, M. ALCAMÍ, F. MARTÍN, H. ZETTERGREN, H. CEDERQUIST,AND B. A. HUBER. Molecular Growth Inside of Polycyclic Aromatic Hydrocarbon ClustersInduced by Ion Collisions. The Journal of Physical Chemistry Letters, 6(9):1536–1542, 2015.

[90] H. T. SCHMIDT, H. CEDERQUIST, J. JENSEN, AND A. FARDI. Conetrap: A compact electrostaticion trap. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions withMaterials and Atoms, 173(4):523 – 527, 2001.

[91] P. REINHED, A. ORBÁN, S. ROSÉN, R.D. THOMAS, I. KASHPERKA, H.A.B. JOHANSSON,D. MISRA, A. FARDI, L. BRÄNNHOLM, M. BJÖRKHAGE, H. CEDERQUIST, AND H.T. SCHMIDT.Cryogenic keV ion-beam storage in ConeTrap–A tool for ion-temperature control. Nuclear In-struments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors andAssociated Equipment, 621(1):83 – 90, 2010.

[92] S. KELLY, P. CAMPBELL, B. CHEAL, T. ERONEN, S. GELDHOF, A. JOKINEN, I. D. MOORE,H. PENTTILÄ, I. POHJALAINEN, S. RINTA-ANTILA, V. SONNENSCHEIN, AND A. VOSS. Laserspectroscopy with an electrostatic ConeTrap. Hyperfine Interactions, 238(1):42, 2017.

[93] B. A. MAMYRIN, V. I. KARATAEV, D. V. SHMIKK, AND V. A. ZAGULIN. The massreflectron, anew non-magnetic time-of-flight mass spectrometer with high resolution. Journal of Experimen-tal and Theoretical Physics, 64:82–89, 1973.

[94] X. TANG, R. BEAVIS, W. ENS, F. LAFORTUNE, B. SCHUELER, AND K. G. STANDING. A sec-ondary ion time-of-flight mass spectrometer with an ion mirror. International journal of massspectrometry and ion processes, 85(1):43–67, 1988.

[95] M. LÓPEZ-PUERTAS, B. M. DINELLI, A. ADRIANI, B. FUNKE, M. GARCÍA-COMAS, M. L.MORICONI, E. D’AVERSA, C. BOERSMA, AND L. J. ALLAMANDOLA. Large Abundances ofPolycyclic Aromatic Hydrocarbons in Titan’s Upper Atmosphere. The Astrophysical Journal,770(2):132, 2013.