Jour

nal
Name

State-selective Photo-fragmentation Dynamics and Photoelectron

Angular Distributions of Acetylene Cations Probed by EUV Syn-
chrotron Radiation
S. Mandala , R. Gopalb , A. Sena , S Senc , A. D’Eliad , H. Srinivase , R. Richter f , M. Corenog, f ,
B. Bapata , M. Mudrichh,i , V. Sharmac† and S. R. Krishnani∗

The acetylene-vinelydene systems serves as a benchmark for investigations of ultrafast dynamical

processes where an interplay of the electronic and nuclear degrees of freedom provides a fertile
playground to explore in the femto- and sub-femto-second physics with coherent extreme-ultraviolet
(EUV) photon sources both on the table-top as well as synchrotons and free-electron lasers. We
focus on detailed investigations of this molecular system in the photon energy range 19...28 eV where
EUV and optical pulses can probe the dynamics effectively. We employ photoelectron-photoion
coincidence (PEPICO) spectroscopy to uncover hitherto unrevealed aspects of this system. In this
work, the role of excited cationic states of C2 H2+ , the primary photoion, is specifically addressed:
From photoelectron spectra and angular distributions the nature of the dissociation and isomerization
channels is discerned. Exploiting the 4π-collection geometry of velocity map imaging, we not only
probe pathways where the efficiency of photoionization is inherently high but also perform PEPICO
spectroscopy on relatively weak channels.

1 Introduction
One of the outstanding problems of interest in time-resolved spec-
troscopy and quantum dynamics of molecular systems is phenom-
ena involving the interplay between nuclear motion and electron
dynamics 1,2 . In femto- and sub-femto-second timescales a deep
understanding of these scenarios is intimately related to realiz-
ing the grand challenge of making molecular movies; "watching"
chemical reactions take place 3 . Among important aspects of the
physics of systems beyond the prominent Born-Oppenheimer ap-
proximation 4 , decoupling nuclear and electronic dynamics, the
role of conical intersections 5 , shape resonances 6 , and fast rear-
rangements within molecules 7–9 are of particular interest. Proton
migration ensuing in the rearrangement of photoexcited molecu-
lar systems has a prominent place not only owing to the intriguing
a
Indian Institute of Science Education and Research, Pune 411008, India
b
Tata Institute of Fundamental Research, Hyderabad 500107, India
c
Indian Institute of Technology Hyderabad, Kandi 502285, India; E-mail:
vsharma@iith.ac.in
d
Department of Physics, University of Trieste, Via A. Valerio 2, 34127 Trieste, Italy
e
Max-Planck-Institute für Kernphysik, 69117 Heidelberg, Germany
f
Elettra-Sincrotrone Trieste, 34149 Basovizza, Italy
g
Consiglio Nazionale delle Ricerche – Istituto di Struttura della Materia, 34149 Trieste,
Italy
h
Aarhus University, 8000 Aarhus C, Denmark
i
Indian Institute of Technology Madras, Chennai 600036, India; E-mail: srkrish-
nan@iitm.ac.in
physics, but also due to its importance in biological systems; this
plays a key role in processes underlying human vision 10 , photo-
synthesis 11 , proton tunneling in DNA 12 and radiation damage 13 ,
to name a few.
The acetylene-vinelydine system has long served as the bench-
mark for investigations of isomerization especially on ultrafast
timescales 7,14,15 as well as in static spectroscopy and theoreti-
cal investigations 16–19 . Both the photoexcitation of outer-valence
electrons 7 in the extreme-ultraviolet as well as core-shell elec-
trons in the hard-xray can effect isomerization 8 . Studying this
system paves way for investigating the dynamics of proton migra-
tion in larger systems such as benzene 9 , and proton conduction
in covalently bonded molecules 20 and weakly bound aggregates
such as bio-interfaces 21 . In order to perform time-resolved spec-
troscopy of the acetylene-vinelydene system, an intimate knowl-
edge of not only the neutral molecule but also the residual ion
and more importantly, the details of photoelectron energies and
angular distributions is essential: For example, transient absorp-
tion, laser-induced fluorescence or resonant multiphoton ioniza-
tion methods which are popular in this context are effective when
the system is spectroscopically well-characterized. The dynamics
of wavepacket resulting from the finite bandwidth of interrogat-
ing pulses can be traced if the states involved in the coherent
interplay are known apriori.
In this article, we use photoelectron imaging in coincidence
with photoion spectrometry to uncover the details of this bench-

J
our
nal
Name,
[yea
r][
,vol
.
], 1–10 | 1
mark system in the spirit of preparing the ground for further in-
vestigations of this system using table-top as well as free-electron
pulsed laser sources. While reports on the transient dynamics of
this system have been published, our recent investigations of this
molecular system embedded in an environment behoove further
time-resolved studies 22 . One of the key advantages of photoelec-
tron spectroscopy is that it can be readily applied in time-resolved
studies bringing with it the advantage that it can access the en-
tire reaction coordinate even when the electronic and vibrational
states evolve in time. Thus, photon energy dependent study of
partial cross sections and photoelectron angular distributions has
proved to be a useful tool to probe different resonant autoion-
ization processes and shape resonance phenomena in molecular
species 23–26 .
In these investigations of C2 H2 photoionization by the
photoelectron-photoion coincidence technique, we report the
photoelectron energy spectra (PES) corresponding to different
ionization channels of C2 H2+ along with their accompanying
photoelectron angular distributions (PADs). This allows us to
discern PES and PADs distinct to each resulting ionic state of
C2 H2+ as a function of photon energy, some of which undergo
further dissociation and isomerization. The choice of the velocity-
Fig. 1 Schematic of the experimental setup. Acetylene gas is effused
map-imaging scheme for photoelectrons is deliberate. This tech-
through a leak valve in the source chamber adjacent to the spectrom-
nique is a preferred method for studying molecules and clus- eter chamber which is connected by a conical skimmer. Subsequently,
ters with extreme-ultraviolet pulses and high-harmonic genera- through the skimmer, C2 H2 is flooded into the spectrometer chamber
tion methodology 27 owing to the elegance of the method; when where the electron-VMI and the ion-ToF spectrometers are situated. In
combined with the inherently high-collection efficiency over the the spectrometer chamber, C2 H2 is ionized by linearly polarized EUV
synchrotron radiation.
entire solid angle this is a very potent tool 28–30 . Thus, our results
can be immediately carried forward and applied to these scenar-
ios. Moreover, to the best of our knowledge, this is the first such
study of the coincidence spectroscopy undertaken to reveal key of the monochromator upstream, ∆E/E ≤ 10−4 ; using a set of
aspects of this paradigmatic molecular system. gratings high-quality photon beams in the energy range 10 and
900 eV are accessible here. The synchrotron ring delivers the pho-
2 Experimental Methods ton beam in this case in the form of ∼ 150 ps pulses with typical
The experiments reported here were carried out at the Gasphase peak intensity of ∼ 15 W/m2 and repetition rate of 500 MHz.
beamline of the Elettra Sincrotrone, Trieste. Fig.1 shows the In the spectrometer chamber, randomly oriented C2 H2
schematic diagram of the experimental setup, whose details have molecules are photoionized by the EUV light and the resultant
been published earlier 31 . Here, high-purity acetylene (C2 H2 ) gas photoelectrons and ions are detected in coincidence with the VMI
was effused into the source chamber through a dosing valve. The and ToF spectrometer, respectively. The charge particle count rate
C2 H2 gas was distilled before entering this valve to remove ace- were maintained at ∼ 18 kHz by adjusting two slits on the photon
tone contamination. In the distillation process, the gas mixture beam path. This synchronous detection scheme of photoelctrons
was passed through a slurry of ethanol and liquid N2 maintained and ions enables us to measure photoelectron velocity distribu-
at −100 ◦C. The source chamber is connected to spectrometer tions correlated to different ions formed due to C2 H2 photoion-
chamber through a conical skimmer which maintains a differen- ization. Therefore, unlike previous studies 23,32,33 , not only we
tial pressure; C2 H2 gas effuses into spectrometer chamber which get the photoelectron velocity distribution of C2 H2 photoioniza-
is maintained at ∼ 10−8 mbar, while the source chamber remains tion, but also we have the information about the same specific to
at ∼ 10−5 . different fragmentation channels of C2 H2+ ion.
The spectrometer chamber holds two co-axial spectrometers - We implemented Abel inversion using the well-established pro-
a Velocity Map Imaging (VMI) spectrometer and a Time of Flight gram, MEVELER 34 , to obtain the full 3D velocity distribution of
(ToF) spectrometer (cf. fig.1). A focused beam of linearly polar- photoelectrons from 2D projection images captured by the VMI
ized extreme ultraviolet (EUV) photons passes through the geo- spectrometer. We used known photoelectron velocity distribution
metric center of the two spectrometers at right angle to the spec- of He at different photon energies above the atomic He ionization
trometer axis which is also perpendicular to its electric polariza- threshold (24.58 eV) to calibrate VMI spectrometer. The average
tion vector (ε). Photon energies in the range 19 and 28 eV were energy resolution (∆E/E) achieved by the spectrometer is about
used in our study. We exploited the excellent photon energy defi- 7%. For single-photon ionization by linearly polarized light, un-
nition possible at this beamline quantified by the resolving power der dipole approximation, the differential cross section can be

2| J
our
nal
Name,
[yea
r][
,vol
.
], 1–10
 (a) (b) (c) 0
PAD
330 30
44.00

VMI Image Abel Inverted High

Fit
image
+
300 60 C H
2 2 19.0 eV
1.0
21.6 eV
10
Y axis

23.9 eV
270 90 26.0 eV
0.8

Ion Yield (arb. u.)

28.0 eV

240 120
0.6
+
Low
1.000
CH
210 150
X axis
180
0.4 +
C H
+ 2
C
Fig. 2 (a) VMI distribution and (b) Abel inverted distribution of the
+
photoelectrons due to photoionization of effusive He at 28 eV and (c) the 0.2 CH
2
+ +
photoectron angular distribution (PAD) obtained from the Abel inverted C
2
N
2

image. The red line shows the fitting of the PAD for β = 2.01 ± 0.08.
0.0
The Z axis in panel (a,b) is in logarithmic scale. 11 12 13 14 23 24 25 26 27 28 29 30

M/q

expressed as:
dσ σ Fig. 3 The ion-ToF mass spectra at different photon energies due to
= total [1 + β P2 (cos θ )]. (1)
dΩ 4π
Since, the photoelectron velocity (~v) has the cylindrical symmetry
along the polarization axis (ε), it has no azimuthal (φ ) depen-
dence. P2 (cos θ ) is the second order Legendre polynomial and θ
is the angle between ~v and ε. The photoelectron angular distribu-
tion (PAD) is characterized by the asymmetry parameter, β .
In the current study, to obtain the value of β specific to different
ionization levels, we used the following scheme: Multiple Gaus-
sian functions are fitted on the PES to determine different ioniza-
tion states and their full-width-at-half-maximum (FWHM). Then,
we obtain the PAD for each state by integrating the angular pho-
toelectron counts over the FWHM limit of each state. Finally, we
fitted eq.(1) on the PAD to get the asymmetry parameter β . For
example, fig.2(a,b) show the experimental VMI distribution and
the Abel inverted distribution of the photoelectron emitted due
to photoionization of effusive He at 28 eV, respectively. Fig.3(c)
shows the PAD of observed the He 1s ionization, where the value
of β obtained from fitting eq.(1) is 2.01 ± 0.08 which correlates
to the PAD of p- partial wave resulting from one-photon ioniza-
tion 35 .
3 Results and Discussion
Acetylene in its neutral ground state (1 Σ+
g ) has the following elec-
tronic configuration:
2 2 2
(1σg ) (1σu ) (2σg ) (2σu ) (3σg ) (1πu ) 2
, with 1πg , 3σu , 4σg and 4σu being the lowest lying unoccupied
orbitals. In the spectral range (19−28 eV) we carried out these in-
vestigations, electrons are predominantly excited or ionized from
the valance orbitals, 1πu , 3σu , 2σu and 2σg . Considering the inde-
pendent particle model, ionization from the 1πu , 3σu , 2σu and 2σg
orbitals leads to X 2 Πu , A2 Σ+ 2 + 2 + + C2 H2+ and C2 H+ ions are the most abundant ionic species, con-
g , B Σu , and C Σg states in C2 H2 ,
respectively. Along with these direct ionization channels, there
exists several indirect autoionizing resonances in C2 H2 , where
electrons are excited from the valance orbitals to the virtual or-
bitals upon photo absorption. As these excitations converge to
the ionic states X, A and B, the corresponding kinetic energies of
C2 H2 photoionization. The x-axis represents mass to charge ratio (M/q)
of the ions. The shaded portion from M/q = 11 to 15 is magnified 10
times for better visibility of the low-yield ions.
photoelectrons remain same irrespective of the ionization mech-
anism. However, the ionization cross sections of these states and
the associated photoelectron angular distributions are greatly in-
fluenced by the involved ionization processes 23,33,36–38 . Here, we
will discuss the photon energy dependent photoionization cross
section and the photoelectron angular distributions associated
with different cationic states (X, A, B,C and D) of C2 H2 both for
the photoionization and for different photodissociation channels.
The remainder of this paper is organized as follows: First we dis-
cern report and discuss photoion mass spectra which enables us
to identify distinct ionization channels characterized by the dis-
sociation pathways of the the C2 H2+ ion. The mainstay of this
paper, PES specific to these ionization channels, as well as an-
gular distributions and the β -parameter particular to each ionic
dissociation channel and state are presented. We compare our
work with existing studies wherever it is relevant to underscore
new findings.
3.1 Photoion Mass Spectra and dissociation channels
2 4 To identify C2 H2 photodissociation channels, we recorded the
ion-ToF mass spectra, presented in fig.3, at different photon ener-
gies. We observe several fragmented ions, C2 H+ , C2+ , CH2+ ,
CH+ and C+ as well as unfragmented parent molecular ion
(C2 H2+ ). Each of these fragment ions represents a distinct pho-
todissociation channel, where the respective ionic fragment is ac-
companied by undetected neutrals. Among these ionic products,
stituting ∼ 95% of total ion yield, while the other fragments com-
prise the remaining fraction. Notably, photoion yields in fig.3 vary
with photon energy, evidencing the corresponding dependence
the ionization efficiency of the channels involved on the same pa-
rameter.

J
our
nal
Name,
[yea
r][
,vol
.
], 1–10 | 3
Table 1 Photoionization efficiencies as a function of photon energy, compared with Hayaishi et al. 36

Photoionization Efficiency
hν (eV)
C2 H2+ C2 H+ C2+ CH2+ CH+

Current Previous 36 Current Previous 36 Current Previous 36 Current Previous 36 Current Previous 36
.
19.0 0.782 0.777 0.106 0.112 0.006 0.004 0.005 0.002 0.007 0.000
21.6 0.737 0.734 0.182 0.191 0.014 0.008 0.008 0.011 0.029 0.004
23.9 0.639 0.628 0.146 0.149 0.016 0.024 0.008 0.026 0.027 0.019
26.0 0.518 0.511 0.120 0.119 0.017 0.025 0.006 0.016 0.021 0.025
28.0 0.407 0.412 0.097 0.087 0.016 0.021 0.005 0.012 0.018 0.023

Hayaishi et al. 36 extensively studied the photoionization dy- , and

namics of C2 H2 by measuring the photoion yield of C2 H2+ , C2 H+ ,
C2+ , CH2+ and CH+ ions as a function of photon energy. They
discussed the appearance energies of these ions as well as as-
signed electronic excitations that result in these ions from ab ini-
tio theoretical calculations. Table.1 shows the comparison of the
observed ionization efficiencies of C2 H2+ , C2 H+ , C2+ , CH2+ and
CH+ ions with the result obtained by Hayaishi et al. 36 . We ob-
serve good agreement between our result and the previous one
for C2 H2+ and C2 H+ . The reported photoionization threshold of
C2 H2 is 11.4 eV and appearance energy of C2 H+ is 16.8 eV 36,39
which can be associated with the cationic states X 2 Πu and A2 Σ+ g,
respectively. Between the X and A states, two major autoion-
izing resonances at 13.3 eV and 15.3 eV were observed; several
electronic excitations were assigned as the origins of these reso-
nances. Therein, the resonance at 13.3 eV was assigned to the
3σg → 3σu excitation, the 15.3 eV resonance to the transitions
2σu → 1πg and 3σg → npσu in the Rydberg series converging to A
state 23,33,36,40 . The high oscillator strength corresponding to this
15.3 eV resonance is also indicative of shape resonance enhanced
final states associated with the centrifugal barrier corresponding
to the l = 2 component 23 . For photon energies above the ioniza-
tion potential of the A state, several resonances were observed;
these converge to the A and B states, e.g., the resonance around
17.68 eV is due to the 2σu → 4σg transition 41 . Further, around
20.0 eV, another shape resonance associated with the 3σg → 3σu
transition may be identified 38 . Thus, a series of dominant res-
onances for photon energies hν < 16 eV stronger than the set of
resonances at higher photon energies hν ≥ 16 eV, characterize the
electronic excitations of acetylene in the extreme ultraviolet. For
ions whose appearance energies are higher than that of C2 H2+
and C2 H+ ions, > 16eV, the yield and photoionization efficiencies
are also relatively lower. We identify C2+ ion originating from the
following two dissociation channels:
C2 H2 + hν → C2+ + H2 + e−
, and
C2 H2 + hν → C2+ + 2H + e−
with appearance energies of 18.1 eV and 22.7 eV, respectively 36 .
The difference between these two C2+ appearance energies
matches the dissociation energy of the H2 molecule 36 . Similarly,
for CH+ ion, there are two distinct appearance energies at 20.7
eV and 24.1 eV arising due to the following channels:
C2 H2 + hν → CH+ + CH + e−
C2 H2 + hν → CH+ + C + H + e− (5)
, respectively 36 . Whereas the appearance energies of the CH2+
+ C and C+ + CH2 photodissociation channels are 19.4 eV 36 and
24.0 eV 42 , respectively. At hν = 19.0 eV, the observed ionization
efficiencies of C2+ , CH2+ , CH+ and C+ ions are even lower. Ta-
ble.1 is a concise summary of the measured photoionization ef-
ficiencies of these fragments, corresponding to aforementioned
channels; these compare well with numbers reported in literature
earlier 36 , affirming the reliability of our measurements. We are
now in a position to obtain insights into these processes taking ad-
vantage of the photoelectron imaging correlated to each of these
photoions. This enables us to derive insights into state-selective
photofragmentation dynamics.
3.2 Photoelectron Energy Spectra and Angular Distribution
To investigate into the details of the mechanisms underlying the
photoionization dynamics of C2 H2 , we recorded the photoelec-
tron energy spectra (PES) and photoelectron angular distribution
(PAD). These performed in two modes: Firstly, these distributions
are measured in the total cumulative electron mode and, secondly,
we record these in the coicindence mode where photoelectron VMI
images are tagged to particular photoions. The first, cumulative
mode, corresponds to PES and PAD measurements summed over
all photoions; these are not tagged to any fragment in particular.
We will first discuss the C2 H2 photoionization by considering the
total photoelectron and the photoionization channels that pro-
duce most abundant ions, C2 H2+ and C2 H+ . In the latter part,
we will discuss rest of the photodissociation channels. This al-
lows us to compare our work with earlier reports, whereas the
coincidence mode is a particular specialty of this work.
Fig.4 a) shows the cumulative PES of all photoelectrons pro-
duced in C2 H2 photoionization. There are four distinct peak
structures in the PES centered at 11.8, 17.0, 18.8 and 24 eV rep-
(2) resenting different ionized states of C2 H2 . The vertical green
dashed lines in fig.4 a) present the known ionization energies
of first five cationic states X 2 Πu , A2 Σ+ 2 + 2 + 2 +
g , B Σu , C Σg and D Σu at
(3) 38
11.4, 16.71, 18.64, 23.33 and 23.53 eV, respectively . Therefore,
in this photon energy range (19 − 28 eV), these five states are
mainly populated upon photoionization of C2 H2 . Interestingly,
PES correlated to the unfragmented C2 H2+ ion (see fig.4 b)) do
not have the fourth peak corresponding to the the C, D states, evi-
dencing that the unfragmented C2 H2+ ion is only produced from
first three states. In contrast, PES correlated to C2 H+ ions (see
(4) fig.4 c)) indicate that only the higher excited states, excluding
X, lead to the C2 H+ fragment, which in addition release a neu-

4| J
our
nal
Name,
[yea
r][
,vol
.
], 1–10
 2 2
X A
Electron Signal (arb. u.) Electron Signal (arb. u.) Electron Signal (arb. u.)
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
8 10
Fig. 4 The photoelectron energy spectra (PES) correlated to (a) total
electron, (b) C2 H2+ ion and (c) C2 H+ ion, respectively, at different
photon energies. The vertical green dashed lines denote the binding
energy of C2 H2+ states.
tral H. This observation reveals the mechanism underlying the
previously reported appearance energy of C2 H+ at 16.8 eV 36,39 .
Coincident photoelectron imaging in forthcoming discussions will
reveal further details of the dynamics in this channel and others.
In order to determine the binding energies, multiple Gaussian
functions are fitted to the PES; peak positions and relative inten-
sities associated with different maxima are determined. Owing
to the finite energy resolution of the VMI spectrometer, we are
not able to distinguish between the closely spaced C and D states.
Therefore, we will address the properties of this peak by labelling
it as (C, D). Table.2 presents the binding energies corresponding
to different ionic states obtained from cumulative total electron
PES as well as spectra correlated with specific photoions, C2 H2+
and C2 H+ , respectively.
In fig.4, we note systematic upward shifts in PES peaks with
increasing photon energies. Assigning the dashed lines obtained
from literature to the respective ionization energies of cationic
ground vibrational states, we draw the following conclusion.
At photon energies significantly higher than the photoionization
threshold of C2 H2 (11.4 eV) higher vibrational levels are increas-
ingly populated. For example, C2 H2+ is likely to be populated
to the ν = 2 and 3 vibrational level in its ground electronic state
+ 2 + 2 + 2 +
(X) whose energies are 11.85 and 12.1 eV, respectively 40 , cf. ta-
B C D
u g u g u ble.2. This observation is important in the context of studying
time-resolved under coherent EUV pulse excitation. From table.2,
(a) Total 19.0 eV
21.6 eV
we see that the PES peaks corresponding to cumulative total elec-
Electron
23.9 eV trons, and those correlated to C2 H2+ are nearly at the same po-
26.0 eV
sitions for X and A states, whereas maxima in spectra correlated
28.0 eV
to A and B states in C2 H+ are significantly shifted towards higher
binding energies by ∼ 0.66 eV and ∼ 0.23 eV, respectively, as com-
pared to the same in C2 H2+ , also evident panel c) of fig. 4. Thus,
additional binding energy is expended in climbing up the vibra-
+ 19.0 eV
tional manifold of the A and B states of the C2 H2+ ion leading
(b) C H
2 2
21.6 eV up to the dissociation into the C2 H+ ion and the neutral H 43 .
23.9 eV
For this dissociative ionization channel, the contributions of the
26.0 eV
28.0 eV
higher excited states (C, D) are also significantly enhanced.
Furthermore, even though we have been able to decipher the
electronic states resulting from C2 H2 photoionization, particu-
larly in the case of C2 H2+ and C2 H+ product ions, it is difficult
to discern the exact ionization process for the following reasons
(c) C H
+
19.0 eV without additional knowledge: Both the direct photoionization
2
21.6 eV
and indirect autoionization processes lead to the same final elec-
23.9 eV
26.0 eV
tronic state; photoelectron kinetic energies emerging from the fi-
28.0 eV nal state remain identical irrespective of the ionization mecha-
nisms. However, the ionization cross sections of the final states
and the associated photoelectron angular distributions will de-
pend on the details of the ionization processes. At photon en-
ergies near the autoionization resonances, we may expect to see
12 14 16 18 20 22 24 26 28
the impact of resonances both in the partial photoionization cross
Binding Energy (eV)
section of these states and in the associated photoelectron angular
distributions (PADs). Therefore, to discern the C2 H2 photoioniza-
tion processes, we will discuss photon energy dependent partial
ionization cross sections and photoelectron asymmetry parame-
ter of these states in detail. To the best of our knowledge, frag-
mentation channel specific ionization cross section and the corre-
sponding photoelectron asymmetry for different cationic states as
a function of photon energy are reported for the first time.
The relative intensities of the cationic states as a function of
photon energy for the total electron are presented in fig.5 a). As
a trend, with increasing photon energy, the populations of X and
A states decrease while that of higher excited B and C, D states
increase. This decreasing nature of ionization cross section of X
and A states can be attributed to the involved 2p atomic orbital.
On the other hand, increasing ionization cross sections of B and
C, D states are due to the involved 2s atomic orbital 33 . In addi-
tion, a weak local maximum is observed in the A state around 21.6
eV. A comparison between the relative ionization cross section of
different cationic states obtained from our study with the same
from theoretical time dependent density functional (TDDFT) cal-
culation 37 is reported in table.3 where good agreement for the
relative cross sections of X and A states are observed with the
theoretical calculation. However, for the higher excited state (B),
the trends observed in our experimental are opposite to that re-
sulting from this TDDFT calculation.
To our advantage, the PEPICO technique enables the measure-
ment of relative ionization cross-sections associated with the frag-
mentation channels producing C2 H2+ and C2 H+ ions as a func-
tion of photon energy, see fig.5 b) and c). This leads to a compre-
J
our
nal
Name,
[yea
r][
,vol
.
], 1–10 | 5
Table 2 State-specific binding energies from cumulative and coincidence PES, at different photon energies

Binding energy (eV) of the PES peak correlated to

hν (eV)
Cumulative total electron C2 H2+ ion C2 H+ ion

X 2 Πu A2 Σ+
g B2 Σ+
u C 2 Σ+ 2 +
g , D Σu X 2 Πu A2 Σ+
g B2 Σ+
u A2 Σ+
g B2 Σ+
u C2 Σ+ 2 +
g , D Σu

19.0 11.79 16.95 17.49 11.79 16.92 17.62 18.54

21.6 11.72 17.09 18.76 11.72 17.01 18.46 17.67 18.86
23.9 11.78 17.14 19.01 23.29 11.78 17.09 18.96 17.72 19.00 23.08
26.0 11.83 17.16 19.00 23.92 11.83 17.12 18.87 17.78 19.10 23.44
28.0 12.01 17.27 19.13 24.24 12.01 17.22 18.98 17.88 19.23 23.60

Table 3 Comparison of relative intensity of different cationic sates of C2H2 photoionization with TDDFT calculation 37

Relative peak intensities (eV)

hν (eV)
X 2 Πu A 2 Σ+
g B 2 Σ+
u C 2 Σ+ 2 +
g , D Σu

Current TDDFT 37 Current TDDFT 37 Current TDDFT 37 Current TDDFT 37

19.0 40.91 40.80 28.77 47.25 17.32 13.36
21.6 39.49 36.75 37.04 29.91 16.51 16.69
23.9 37.01 26.86 28.89 24.97 15.61 13.46 1.57 2.69
26.0 28.90 23.30 20.09 21.30 19.29 11.36 2.91 3.49
28.0 16.64 20.83 15.00 18.32 19.90 10.26 4.32 3.59

hensive picture of ionization and dissociation in the photofrag-

mentation process examining the PES correlated to C2 H2+ and
C2 H+ ions. Whereas the higher excited ionic states, (C, D), do
not leave behind unfragmented C2 H2+ ions, the lowest ionized
state, X, does not participate in the dissociation process to pro-
45 duce C2 H+ . For the C2 H2+ ion, the relative cross section of X
Relative Intensity (%)

2 2 + 2 + 2 + 2 +
X A B C , D
u g u g u

states dominates over the same of B and C, D states. The rela-

(a) Total

30 Electron
tive cross sections of the X and A states decrease with increas-
ing photon energy and are significantly larger than that of the B
state which remains largely independent of photon energy. On
15
the other hand for the dissociated C2 H+ ion, the contributions of
B states are dominant the A and C, D states. Beyond photon en-
0 ergies of ∼ 21 eV, the ionization cross sections of A and B states
Relative Intensity (%)

+ 2 2 + 2 +

60 (b) C H
2 2
X
u
A
g
B
u decreases with increasing photon energy.

The most significant aspect of correlated photoelectron imag-

40 ing is the opportunity it provides to examine photoion- and state-
specific PADs. This immediately reveals the variation and system-
20 atics of the asymmetry parameter (β ) of the photoelectron angu-
lar distribution correlated to different electron states. Fig.6 a),
0 b) and c) depict the photoelectron asymmetry parameter (β ) as
a function of photon energy; these are determined for the cases
Relative Intensity (%)

+ 2 + 2 + 2 + 2 +
(c) C H A B C , D
15 2 g u g u

of cumulative total photoelectrons as well as PADs correlated to

C2 H2+ and C2 H+ ions. Since, the π electron ejection usually
10 leads to higher degree of asymmetry than that of σ electron ejec-
tion, the observed higher value of β for X states than that of the
5 other states is justified. Considering the low relative efficiencies
photoionization of some of the channels, the particularly high col-
0 lection of photoelectrons enabled by the VMI technique in which
18 19 20 21 22 23 24 25 26 27 28 29 photoelectrons are collected over the entire solid angle plays an
Photon Energy (eV)
important role in the measurement of asymmetry parameters. In
this case, the measured β (∼ 1.5) for X states is higher than the
Fig. 5 Relative intensities of different states correlated to (a) total elec- corresponding values obtained from the previous experimental
tron (b) C2 H2+ ion and (c) C2 H+ ion, respectively, at different photon studies where angle-resolved spectra were recorded using hemi-
energies. spherical electron analysers 32,33 .

The behaviour of the asymmetry parameter for the lower ex-

cited, X and A, as well as the higher excited, B and C, D, states is
revealing. The asymmetry parameter increases steadily with pho-

6| J
our
nal
Name,
[yea
r][
,vol
.
], 1–10
 ton energy for X and A state (black and red lines in fig.6). As the
autoionizing resonances are known to influence the photoelec-
tron angular distribution, the absence of strong autoionization
channels converging to the lower ionized states X (1πu−1 ) and A
(3σg−1 ) states for hν > 19 eV can be attributed to this steady in-
crease in the β parameter for these state 44 . Earlier results 33,45
show a sharp minima in β at 19 eV due to shape resonance in σu
continuum, and the value of β increases steadily above 19 eV for A
state,. This is in accordance with our current result. However, the
higher ionized B (2σu−1 ) state shows considerable variations in the
β parameter with photon energy in the PADs of the cumulative to-
tal electron distribution and those correlated with the C2 H2+ ion
(blue line in fig.6 a), b)). We observe a local maxima around 21.6
eV for total electron and a local minima around 26.0 eV for C2 H2+
X
2
A
2 +
B
2 +
C
2 +
, D
2 +
ion. Previous experimental and theoretical studies 33,37,45 also re-
2.0 u g u g u

ported such minima around 25 eV for the B state occurring due to

1.5 interplay between 2σu → kσg and 2σu → kπg transitions. But, for
C2 H+ ion, there is no such structure in the β parameter, only a
1.0
steady increase is seen with increasing photon energy (blue line in
0.5 fig.6 c)). This is indicative of autoionizing resonances in this pho-
0.0
(a) Total electron ton energy range (19 eV ≤ hν ≤ 28.0) that leads to unfragmented
C2 H2+ ion, whereas these resonances do not lead to photodisso-
+

2.0
(b) C H
2 2 ciation into C2 H+ ion, only direct photoionization of C2 H2 pro-
ceeds to C2 H+ + H fragmentation. This result calls for detailed
1.5
theoretical calculations of ionization channel specific asymmetry
1.0 parameter considering multichannel interaction to include possi-
ble autoionization channels combining both nuclear and electron
0.5
dynamics. Finally, for C, D state, the value of β increases steadily
with increasing photon energy for the total electrons and C2 H+
2 2 + 2 +

0.0 X A B
u g u

+
ions (green line in fig.6 a), c)). This indicates absence of autoion-
2 + 2 + 2 + 2 +
(c) C H A B C , D
2.0
2 g u g u izing channels leading to higher ionized C, D states. Although a
maxima in β at 26 eV was reported due to shape resonance to kσu
1.5
continuum 33,37,45 .
1.0 Several theoretical studies performed hitherto 37,38 addressed
0.5
the variation of electronic state specific asymmetry parameter by
considering multichannel interaction of electronic excitations in
0.0
this photon energy range. Wells and Lucchese 38 implemented
18 19 20 21 22 23 24 25 26 27 28
multichannel scattering methodology (MCSCF) where the par-
tial cross sections and corresponding β parameters of different
Photon Energy (eV)
ionic final states were calculated for different autoionization reso-
nances. Fronzoni et al. 37 used time dependent density functional
Fig. 6 Photoelectron asymmetry parameter (β ) of different states corre- method on a fixed nuclei geometry to calculated the autoioniza-
lated to (a) total electron (b) C2 H2+ ion and (c) C2 H+ ion, respectively,
tion channels and the asymmtery parameters. Table.4 shows the
as a function of photon energy.
comparison of the β parameter obtained in our experiment with
the previous theoretical studies 33,38 . Only for the B state, reason-
able agreement between our experimental result and the theoret-
ical calculation is observed. While for other ionized states our β
values are quite different from the same calculated from theory.
It should be noted that both the theoretical studies estimate dif-
ferent values of β for the X, A and C, D states, with reasonable
agreement only for B state.
In the rest of this article, we will discuss the photoionization
channels that produce C2+ , CH2+ , CH+ and C+ ions. Fig.7 a),
b), c) and d) show the PES correlated to C2+ , CH2+ , CH+ and C+
ions, respectively, at different photon energies. Among these ions,
the PES and PAD associated with the CH2+ ion are of particular

J
our
nal
Name,
[yea
r][
,vol
.
], 1–10 | 7
Table 4 Comparison of asymmetry parameter correlated to the total electron with previous theoretical studies

Asymmetry parameter (β ) correlated to

hν (eV)
X 2 Πu A 2 Σ+
g B2 Σ+
u C 2 Σ+ 2 +
g , D Σu

Current TDDFT 37 MCSCF 38 Current TDDFT 37 MCSCF 38 Current TDDFT 37 MCSCF 38 Current TDDFT 37 MCSCF 38
19.0 1.54 0.88 0.74 0.60 0.31 0.02 0.59 0.67 0.59
21.6 1.74 1.03 0.79 0.61 0.32 0.15 1.30 1.16 0.82
23.9 1.72 1.19 1.06 0.67 0.36 0.22 1.04 1.17 1.22 0.10 1.06 0.01
26.0 1.80 1.26 0.83 0.41 0.30 0.86 1.13 1.25 0.32 1.47 -0.31
28.0 1.74 1.32 1.10 0.46 0.41 0.86 1.08 1.23 0.67 1.43 -0.13

interest as this ion result from the photodissociation of C−CH2+

ion, which is produced due to isomerization in HC−CH+ ion. In
this isomerization process, a H atom attached one of the C’s mi-
grates from its native location to the adjacent C atom upon single
ionization of C2 H2 7,16 . The PES correlated to these ions can be
divided into two sections: a. Intense peak structures for binding
energies > 22 eV and b. low intensity features below 22 eV. The
highly intense peak structures above 22 eV indicate that these ions 2 2 + 2 +
C
2 +
D
2 +

2 2
g u
X A B 3 3
are predominantly produced from the higher excited ionic states u g u g u

of C2 H2+ , specifically from C state and beyond. On the other (a) C

+

2
19.0 eV 26.0 eV
21.6 eV 28.0 eV
hand, the low intensity features below ∼ 22 eV indicate that the 1.0
23.9 eV
lower ionized states below C state can also produce these ions,
although previously reported appearance energies of these ions 0.5

were above 18 eV 36,42 . Owing to the fact that these low energy
features are complex, it is difficult to decompose these into mul- 0.0
+
tiple Gaussian peak function associated with X, A and B states as (b) CH
2
19.0 eV 26.0 eV
21.6 eV 28.0 eV
done in case of total electron, C2 H2+ and CH2+ ions. Therefore, 1.0
Electron Signal (arb. u.)

23.9 eV

we will only concentrate on the high binding energy part of the

spectra for these low-yield ions. 0.5

0.0
+
(c) CH 19.0 eV 26.0 eV
21.6 eV 28.0 eV
1.0
23.9 eV
At hν = 23.9 eV, PES correlated to all these photoions have
(blue line in fig.7) similar maxima around a binding energy of 0.5

23.3 eV corresponding to the C, D states. However, the associ-

ated photoelectron asymmetry parameter (β ) for these ions are 0.0
+
(d) C 19.0 eV 26.0 eV
quite different (cf. table.5). The observed β decreases from
21.6 eV 28.0 eV
0.25 ± 0.14 for C2+ ion to −0.03 ± 0.13 for the C+ ion. A broad 1.0
23.9 eV

peak is observed (green line in fig.7) for a photoexcitation of 26.0

eV which can be decomposed into two closely separated peaks at 0.5

24.1 and 24.9 eV. The corresponding asymmetry parameters for

these peaks are ∼ 0.3 and ∼ 0, respectively. However, at a pho- 0.0

8 10 12 14 16 18 20 22 24 26 28
ton energy of 28.0 eV, the high energy PES mainly arise from three
Binding Energy (eV)
peaks at 24.5, 26.4 and 27.3 eV (orange line in fig.7). Among these
peaks, the 26.4 and 27.3 eV features can be assigned to the 3 Πg
Fig. 7 The photoelectron energy spectra (PES)correlated to (a) C2+ ,
and 3 Πu states, respectively 46 , whereas, the peak around 24.5
(b) CH2+ , (c) CH+ and (d) C+ ion, respectively, at different photon
eV (vertical blue dashed line in fig.7) has not been reported ear- energies. The vertical green dashed lines are the ionization energies of
lier. For C2+ and CH2+ ions, the new peak dominates over the different C2 H2+ states 46 . The blue vertical dashed lines show the new
states (3 Πg,u ), while for CH+ and C+ ion all the three peaks are peak around 24.5 eV.
almost at equal intensity. For 3 Πu,g states, the observed photo-
electron angular distributions are isotropic, which resulted in β
values close to zero. This is in contrast to the observed asymmetry
(β ) for X 2 Πu state where we observe higher degree of asymmetry
(β ∼ 1.5). However, the new peak around 24.5 eV shows higher
degree of asymmetry which increases from 0.40 ± 0.07 for C2+ to
0.77 ± 0.11 for C+ .

8| J
our
nal
Name,
[yea
r][
,vol
.
], 1–10
Table 5 Photoelectron asymmetry parameters for C2+ , CH2+ , CH+ and C+ ions at different photon energies

Fragment Asymmetry Parameter (β ) for different photon energies

hν =23.9 eV hν =26.0 eV hν =28.0 eV
C 2 Σ+ 2 +
g , D Σu 24.1 eV peak 24.9 eV peak 24.5 eV peak 3Π
u
3Π
g

C2 + 0.25±0.14 0.28±0.08 0.00±0.09 0.40±0.07 0.07±0.09 0.15±0.13

CH2+ 0.19±0.14 0.34±0.07 -0.04±0.09 0.57±0.09 -0.01±0.11 -0.02±0.11
CH+ 0.02±0.12 0.23±0.07 -0.02±0.10 0.59±0.07 0.14±0.09 -0.23±0.15
C+ -0.03±0.13 0.33±0.08 -0.01±0.10 0.66±0.11 0.22±0.11 0.23±0.14

4 Conclusion 6 S. Nandi, E. Plésiat, S. Zhong, A. Palacios, D. Busto, M. Isinger,

We have uncovered several intriguing details of the dynamics of
the proton acetylene-vinelydene system which plays a central role
in our understanding of proton migration and isomerization in
the extreme ultraviolet, 19...28 eV. In the photofragmentation dy-
namics, we observe that the unfragemented C2 H2+ ion remains
intact in the lower lying X, A and B states. Photodissociation
producing C2 H+ and other fragments are mainly produced from
even higher excited states, C and beyond. Upon photoexciation
at 28.0 eV, C2+ , CH2+ , CH+ and C+ originate from a newly re-
vealed state centered around 24.5 eV binding energy and from
3Π
g,u states. PES and PAD measurements evidence the important
role of autoionizing resonances at higher photon energies. Par-
tial photoionization cross sections and photoelectron asymmetry
correlated to different cationic states are reported for the pho-
toelectron in coincidence with these photoions. State-selective
photoelectron asymmetry parameter (β ) as a function of photon
energy shows distinct behaviour for the photoionization channels
leading to C2 H2+ and C2 H+ ions. This finding suggests further
theoretical investigation of photoelectron asymmetry by combin-
ing both the nuclear and the electron dynamics.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
VS, RG and SRK are grateful to DST, India and ICTP, Trieste, for
support (proposal # 20165468) to carry out this campaign at the
Elettra Synchrotron facility. VS and SRK acknowledge financial
support from the IMPRINT and DAE-BRNS scheme. SRK thanks
the Max Planck Society for supporting this research via the Part-
ner group. MM acknowledges support from the Carlsberg Foun-
dation, and with SRK and VS for the funding from the SPARC
programme, MHRD, India.
Notes and references
1 A. Stolow, Annual Review of Physical Chemistry, 2003, 54, 89–
119.
2 A. Stolow, A. E. Bragg and D. M. Neumark, Chemical Reviews,
2004, 104, 1719–1758.
3 A. H. Zewail, The Journal of Physical Chemistry A, 2000, 104,
5660–5694.
4 L. S. Cederbaum, The Journal of Chemical Physics, 2008, 128,
124101.
5 G. A. Worth and L. S. Cederbaum, Annual Review of Physical
Chemistry, 2004, 55, 127–158.
L. Neoričić, C. L. Arnold, R. J. Squibb, R. Feifel, P. Decleva,
A. L’Huillier, F. Martín and M. Gisselbrecht, Science Advances,
2020, 6,.
7 Y. H. Jiang, A. Rudenko, O. Herrwerth, L. Foucar, M. Kurka,
K. U. Kühnel, M. Lezius, M. F. Kling, J. van Tilborg, A. Belka-
cem, K. Ueda, S. Düsterer, R. Treusch, C. D. Schröter,
R. Moshammer and J. Ullrich, Physical Review Letters, 2010,
105, 263002.
8 C. E. Liekhus-Schmaltz, I. Tenney, T. Osipov, A. Sanchez-
Gonzalez, N. Berrah, R. Boll, C. Bomme, C. Bostedt, J. D.
Bozek, S. Carron, R. Coffee, J. Devin, B. Erk, K. R. Fergu-
son, R. W. Field, L. Foucar, L. J. Frasinski, J. M. Glownia,
M. Gühr, A. Kamalov, J. Krzywinski, H. Li, J. P. Marangos, T. J.
Martinez, B. K. McFarland, S. Miyabe, B. Murphy, A. Natan,
D. Rolles, A. Rudenko, M. Siano, E. R. Simpson, L. Spector,
M. Swiggers, D. Walke, S. Wang, T. Weber, P. H. Bucksbaum
and V. S. Petrovic, Nature Communications, 2015, 6, 8199.
9 M. C. E. Galbraith, C. T. L. Smeenk, G. Reitsma, A. Marciniak,
V. Despré, J. Mikosch, N. Zhavoronkov, M. J. J. Vrakking,
O. Kornilov and F. Lépine, Physical Chemistry Chemical
Physics, 2017, 19, 19822–19828.
10 R. Schoenlein, L. Peteanu, R. Mathies and C. Shank, Science,
1991, 254, 412–415.
11 W. Kreutz, Zeitschrift für Naturforschung B, 01 Jan. 1970, 25,
88 – 94.
12 P.-O. Löwdin, Reviews of Modern Physics, 1963, 35, 724–732.
13 R. Rein and F. E. Harris, Science, 1964, 146, 649–650.
14 C. Burger, A. Atia-Tul-Noor, T. Schnappinger, H. Xu, P. Rosen-
berger, N. Haram, S. Beaulieu, F. Légaré, A. S. Alnaser,
R. Moshammer, R. T. Sang, B. Bergues, M. S. Schuurman,
R. de Vivie-Riedle, I. V. Litvinyuk and M. F. Kling, Structural
Dynamics, 2018, 5, 044302.
15 Z. Li, L. Inhester, C. Liekhus-Schmaltz, B. F. E. Curchod, J. W.
Snyder, N. Medvedev, J. Cryan, T. Osipov, S. Pabst, O. Ven-
drell, P. Bucksbaum and T. J. Martinez, Nature Communica-
tions, 2017, 8, 453.
16 S. Boyé-Péronne, D. Gauyacq and J. Liévin, The Journal of
Chemical Physics, 2006, 124, 214305.
17 T. S. Zyubina, Y. A. Dyakov, S. H. Lin, A. D. Bandrauk and
A. M. Mebel, The Journal of Chemical Physics, 2005, 123,
134320.
18 M. Alagia, C. Callegari, P. Candori, S. Falcinelli, F. Pirani,
R. Richter, S. Stranges and F. Vecchiocattivi, The Journal of
Chemical Physics, 2012, 136, 204302.
19 R. Krishnan, M. J. Frisch, J. A. Pople and P. von R. Schleyer,

J
our
nal
Name,
[yea
r][
,vol
.
], 1–10 | 9
 Chemical Physics Letters, 1981, 79, 408 – 411. Y. Iida and M. Nakamura, Journal of Physics B: Atomic and
20 L. Vilčiauskas, M. E. Tuckerman, G. Bester, S. J. Paddison and Molecular Physics, 1982, 15, 79–92.
K.-D. Kreuer, Nature Chemistry, 2012, 4, 461–466. 37 G. Fronzoni, M. Stener and P. Decleva, Chemical Physics,
21 M. Brändén, T. Sandén, P. Brzezinski and J. Widengren, 2004, 298, 141 – 153.
Proceedings of the National Academy of Sciences, 2006, 103, 38 M. C. Wells and R. R. Lucchese, The Journal of Chemical
19766–19770. Physics, 1999, 111, 6290–6299.
22 S. Mandal, R. Gopal, M. Shcherbinin, A. D’Elia, H. Srinivas, 39 Y. Ono and C. Y. Ng, The Journal of Chemical Physics, 1981,
R. Richter, M. Coreno, B. Bapat, M. Mudrich, S. R. Krishnan 74, 6985–6986.
and V. Sharma, Physical Chemistry Chemical Physics, 2020, 22, 40 L. Avaldi, G. Dawber, R. Hall, G. King, A. McConkey, M. Mac-
10149–10157. Donald and G. Stefani, Journal of Electron Spectroscopy and
23 A. C. Parr, D. L. Ederer, J. B. West, D. M. P. Holland and J. L. Related Phenomena, 1995, 71, 93 – 105.
Dehmer, The Journal of Chemical Physics, 1982, 76, 4349– 41 P. Langhoff, B. McKoy, R. Unwin and A. Bradshaw, Chemical
4355. Physics Letters, 1981, 83, 270 – 275.
24 A. C. Parr, D. L. Ederer, B. E. Cole, J. B. West, R. Stockbauer, 42 G. Cooper, T. Ibuki, Y. Iida and C. Brion, Chemical Physics,
K. Codling and J. L. Dehmer, Physical Review Letters, 1981, 46, 1988, 125, 307 – 320.
22–25. 43 C. Servais and R. Locht, Chemical Physics Letters, 1995, 236,
25 K. Codling, A. C. Parr, D. L. Ederer, R. Stockbauer, J. B. West, 96 – 102.
B. E. Cole and J. L. Dehmer, Journal of Physics B: Atomic and 44 Z. H. Levine and P. Soven, Physical Review Letters, 1983, 50,
Molecular Physics, 1981, 14, 657–666. 2074–2077.
26 J. B. West, K. Codling, A. C. Parr, D. L. Ederer, B. E. Cole, 45 D. Lynch, M. Lee, R. R. Lucchese and V. McKoy, The Journal of
R. Stockbauer and J. L. Dehmer, Journal of Physics B: Atomic Chemical Physics, 1984, 80, 1907–1916.
and Molecular Physics, 1981, 14, 1791–1801. 46 M. Wells and R. R. Lucchese, The Journal of Chemical Physics,
27 T. Kreibich, M. Lein, V. Engel and E. K. U. Gross, Physical 1999, 110, 6365–6380.
Review Letters, 2001, 87, 103901.
28 J. Mauritsson, T. Remetter, M. Swoboda, K. Klünder,
A. L’Huillier, K. J. Schafer, O. Ghafur, F. Kelkensberg, W. Siu,
P. Johnsson, M. J. J. Vrakking, I. Znakovskaya, T. Uphues,
S. Zherebtsov, M. F. Kling, F. Lépine, E. Benedetti, F. Ferrari,
G. Sansone and M. Nisoli, Physical Review Letters, 2010, 105,
053001.
29 F. Kelkensberg, W. Siu, J. F. Pérez-Torres, F. Morales, G. Gade-
mann, A. Rouzée, P. Johnsson, M. Lucchini, F. Calegari, J. L.
Sanz-Vicario, F. Martín and M. J. J. Vrakking, Physical Review
Letters, 2011, 107, 043002.
30 G. Sansone, F. Kelkensberg, J. F. Pérez-Torres, F. Morales,
M. F. Kling, W. Siu, O. Ghafur, P. Johnsson, M. Swo-
boda, E. Benedetti, F. Ferrari, F. Lépine, J. L. Sanz-Vicario,
S. Zherebtsov, I. Znakovskaya, A. L’Huillier, M. Y. Ivanov,
M. Nisoli, F. Martín and M. J. J. Vrakking, Nature, 2010, 465,
763–766.
31 P. O’Keeffe, P. Bolognesi, M. Coreno, A. Moise, R. Richter,
G. Cautero, L. Stebel, R. Sergo, L. Pravica, Y. Ovcharenko and
L. Avaldi, Review of Scientific Instruments, 2011, 82, 033109.
32 J. Kreile and A. Schweig, Chemical Physics Letters, 1980, 69,
71 – 74.
33 D. Holland, M. MacDonald, M. Hayes, L. Karlsson and
B. Wannberg, Journal of Electron Spectroscopy and Related Phe-
nomena, 1998, 97, 253 – 263.
34 B. Dick, Physical Chemistry Chemical Physics, 2014, 16, 570–
580.
35 P. O'Keeffe, P. Bolognesi, R. Richter, A. Moise, E. Ovcharenko,
L. Pravica, R. Sergo, L. Stebel, G. Cautero and L. Avaldi, Jour-
nal of Physics: Conference Series, 2010, 235, 012006.
36 T. Hayaishi, S. Iwata, M. Sasanuma, E. Ishiguro, Y. Morioka,

10 | J
our
nal
Name,
[yea
r][
,vol
.
], 1–10