Laser-induced Electron-Transfer in the Dissociative Multiple Ionization of Argon Dimers
YanLan Wang, XuanYang Lai, ShaoGang Yu, RenPing Sun, XiaoJun Liu, Martin Dorner-Kirchner, Sonia Erattupuzha, Seyedreza Larimian, Markus Koch, Václav Hanus, Sarayoo Kangaparambil, Gerhard Paulus, Andrius Baltuška, Xinhua Xie, Markus Kitzler-Zeiler
LLaser-induced Electron-Transfer in the Dissociative Multiple Ionization of ArgonDimers
YanLan Wang, XuanYang Lai, ShaoGang Yu, RenPing Sun, XiaoJun Liu ∗ State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics,Innovation Academy for Precision Measurement Science and Technology,Chinese Academy of Sciences, Wuhan 430071, China
Martin Dorner-Kirchner a , Sonia Erattupuzha a , Seyedreza Larimian a , Markus Koch b , V´aclav Hanus a ,Sarayoo Kangaparambil a , Gerhard Paulus c , Andrius Baltuˇska a , Xinhua Xie ( 谢 新 华 ) a,d , Markus Kitzler-Zeiler a † a Photonics Institute, Technische Universit¨at Wien, A-1040 Vienna, Austria, b Institute of Experimental Physics, Graz University of Technology, A-8010 Graz, Austria, c Institute of Optics and Quantum Electronics, Friedrich Schiller University Jena, D-07743 Jena, Germany, d SwissFEL, Paul Scherrer Institute, 5232 Villigen PSI, Switzerland
We report on an experimental and theoretical study of the ionization-fragmentation dynamics ofargon dimers in intense few-cycle laser pulses with a tagged carrier-envelope phase. We find thata field-driven electron transfer process from one argon atom across the system boundary to theother argon atom triggers sub-cycle electron-electron interaction dynamics in the neighboring atom.This attosecond electron-transfer process between distant entities and its implications manifestthemselves as a distinct phase-shift between the measured asymmetry of electron emission curves ofthe Ar + + Ar and Ar + Ar fragmentation channels. Our work discloses a strong-field routeto controlling the dynamics in molecular compounds through the excitation of electronic dynamicson a distant molecule by driving inter-molecular electron-transfer processes. Photoinduced molecular charge-transfer (CT) acrosssystem boundaries is a key step in many important nat-ural or technical processes such as solar-driven energyproduction [1, 2], photocatalysis [3, 4], or photosyn-thetic activity [5, 6]. In these processes the relocationof charge, initiated by the absorption of a single pho-ton by a molecule, is determined by the energetic andspatial structure of the system. A fundamentally differ-ent mechanism for determining charge-localization pro-cesses becomes available in strong laser fields. It wasshown that the intra-molecular localization of electronsduring the dissociation of isolated, small molecules canbe determined by multi-photon processes driven by in-tense few-cycle laser pulses using their carrier-envelopephase (CEP) as the control parameter [7–11].An intriguing yet unexplored question is then, whetherstrong-field-driven multi-photon processes can influencethe localization of charge not only within one moleculebut also across system boundaries. Widely used model-systems for investigating inter-system transfer reactionsare small van der Waals (vdW) clusters and dimers.VdW dimers are used to study photoinduced biologicalprocesses [12–14], photocatalytic reactions [15, 16], andenergy or charge transfer reactions induced by soft X-ray photons [17–20] and electron impact [21]. VdW sys-tems are also studied with strong laser fields, but in thecase of dimers with a focus on the field-driven ionizationand fragmentation dynamics [22–35], or electronic energyconversion processes in the case of larger clusters [36–41].To the best of our knowledge, strong-field driven electrontransfer-reactions across the system boundary from oneentity to another have thus far not been investigated. In this Letter, we show experimentally and by simula-tions, using the argon dimer, Ar , as an example, thatelectron transfer-reactions from one argon atom to theother can be driven by a strong laser field and, further-more, that they are decisive for the ionization and frag-mentation behavior of the dimer. Specifically, we demon-strate that an electron liberated at one of the two Aratoms can be captured by the neighboring atom. Thisprocess, which we refer to as the laser-induced trans-fer of electron (LITE) process, determines the emissiontiming of the electrons via electron-electron interactionand thus, depending on the CEP, influences the mo-menta of the emitted electrons. As a result, the effect ofLITE can be observed in our experiments and simulationswhen comparing the asymmetry of electron emission asa function of CEP for the two ionization-fragmentationchannels Ar(1,2) and Ar(2,2), where Ar( n, m ) denotesAr −−−→ Ar ( n + m )+2 → Ar n + + Ar m + .In our experiments, argon dimers created by super-sonic expansion of a few bars of argon gas were ionizedby intense laser pulses, linearly polarized along z , witha full width at half maximum (FWHM) duration in in-tensity of 4.5 fs and a peak intensity, calibrated in in situ [42], of 5 × W cm − , inside the ultra-high vacuumchamber of a reaction microscope [43]. Details on thereaction microscope can be found in Refs. [44–46]. Thelaser center wavelength was λ = 750 nm. The dura-tion of the pulses and their CEP were measured witha stereo electron spectrometer in phase-tagging mode[47]. Upon laser ionization of the argon dimers, the twovdW-bound argon atoms, separated by their equilibriuminternuclear distance ( R eq ) undergo fragmentation via a r X i v : . [ phy s i c s . a t m - c l u s ] J un Ar(2,2)Ar(1,2)Ar(1,1) c o un t s ( x ) KER (eV)
FIG. 1: Kinetic energy release (KER) distributionsof fragmentation channels Ar( n, m ) with ( n, m ) = { (1 , , (1 , , (2 , } . Arrows mark peaks due to electronrecapture, shaded areas highlight the peaks resulting fromCoulomb explosion at R eq . Coulomb explosion. We detected the two emerging ar-gon ions, Ar n + and Ar m + , in coincidence and from theirtime of flight and impact position on our detector cal-culated their three-dimensional momenta p n Ar and p m Ar .By imposing momentum conservation conditions onto theions detected in coincidence, the two-body fragmentationchannels of interest, Ar(1,2) and Ar(2,2), as well as thechannel Ar(1,1), were selected for further analysis. Dueto momentum conservation, the sum momentum of the( n + m ) emitted electrons, p ( n,m ) e = (cid:80) n + mi =1 p ei , with p ei the momentum of the i th electron, can be determinedfrom the center of mass recoil momentum of the ions, p ( n,m )R = p n Ar + p m Ar , using the relation p ( n,m ) e = − p ( n,m )R .Fig. 1(a) displays the measured distributions of thekinetic energy released (KER) during fragmentation,KER = (cid:2) ( p m Ar ) + ( p n Ar ) (cid:3) / (2 M ) with M the atomicmass of argon, for the Ar(1,1), Ar(1,2) and Ar(2,2) chan-nels. For each channel, at least two characteristic mainpeaks can be identified. The smaller peaks at higherKER-values (marked by arrows) were attributed to theprocess of frustrated tunnel ionization [22–24]. The dom-inant peaks at lower KER-values, highlighted by coloredareas in Fig. 1(a), originate from Coulomb explosions ofthe argon dimers at R eq and are the focus of this work.To obtain insight into the multiple ionization dynam-ics underlying the colored lower-KER peaks in Fig. 1(a),we introduce an asymmetry-parameter A ( n,m ) z = ( n up − n dn ) / ( n up + n dn ), where n up ( n dn ) denote for the chan-nel Ar(n,m) the number of events with a positive (neg-ative) electron sum momentum along z . Alternativelyto A ( n,m ) z , one could also analyze the mean electron summomentum ¯ p ( n,m ) e,z . But as we show in Suppl. Mat. [48],the two quantities feature an almost identical dependenceon the CEP. In the following we will use A ( n,m ) z , as it hasthe advantage that it can be visually connected to elec-tron yields discussed below. The measured dependenceof A z on CEP for the Ar(1,2) and Ar(2,2) channels isdepicted in Fig. 2(a). The key feature in Fig. 2(a) is thatthe A z -curve for the Ar(2,2) channel exhibits a clear left phase shift of about 0 . π to that of the Ar(1,2) channel.To understand this experimentally observed CEP-shiftbetween the two channels, we traced the correlated elec-trons and the motion of the nuclei in the combined laserand Coulomb fields by performing a 3D classical ensemblemodel calculation [34, 49], described in Suppl. Mat. [48].As the laser intensity is well above the over-the-barrierthreshold [50] the two outermost electrons are rapidlystripped from each argon atom [49]. We therefore didnot model these two initial ionization events and insteadstarted from a dimer consisting of two singly charged ar-gon ions (Ar + - Ar + ), with one active electron situatedaround the position of each ion.The CEP-dependence of A z predicted by the simula-tions for the Ar(1,2) and Ar(2,2) channels is shown inFig. 2(b). The simulated curves agree very well withthe measured ones, in particular the CEP left-shift ofthe Ar(2,2) channel is very well reproduced. The ori-gin of this phase-shift can be extracted from the simu-lations by analyzing the distributions of ionization times t st and t nd > t st of the laser-driven electron trajec-tories that lead to the channels Ar(1,2) and Ar(2,2), re-spectively. The ionization time t st marks the instantat which the single-particle energy of the first emittedelectron becomes positive for the first time. Likewise, t nd marks this instant for the second emitted electronin the Ar(2,2) channel. The distributions of t st and t nd are plotted in Figs. 2(c)-(h) for the Ar(1,2) and Ar(2,2)channels and three selected values of the CEP. For con-venience of the following discussion, the ionization time-distributions were separated depending on whether the(sum) momentum of the electron (pair) reaches positive(upper halves) or negative momentum (lower halves).To explain the CEP left-shift between Ar(1,2) andAr(2,2), we start with the CEP-dependence of A (1 , z . Asshown in Figs. 2(c)-(e), the distributions of the ioniza-tion times ( t st ) in this channel feature two maxima perpeak of the laser field. The reason underlying these twomaxima will be explained below. For ϕ CEP = 0, thetwo maxima corresponding to the field peak at t A aremarked by cyan and yellow boxes [Fig. 2(c)]. The max-ima corresponding to the field peaks at t B and t C aremuch smaller for ϕ CEP = 0. The emission directionsof electrons set free during these maxima (up or down,indicated by positive or negative time-distributions) arelargely determined by the laser vector potential accord-ing to the relation p e = − A ( t i ) = (cid:82) t i −∞ E ( t (cid:48) ) dt (cid:48) [51, 52]with t i the ionization time. Positive values of A ( t i ) areindicated by gray shading in Figs. 2(c)-(h). The smalldeviations from p e = − A ( t i ) are due to the Coulombforces of the argon ions.For ϕ CEP = 0, most of the trajectories are emittedwith p e <
0. Therefore, A (1 , z has a large negativevalue, cf. Figs. 2(a,b). For increasing CEP, the laser field-maximum at t B shifts closer to the pulse peak and be- (a) c fFig. 2( , )Fig. 2( , )d gFig. 2( , )e h (b) Measurement Ar + Ar(2,2)Ar(1,2)
Simulation Ar + Ar(2,2)Ar(1,2)-0.50.5000.1-0.1 A s y mm e t r y CEP (rad)0 0.5 1.5 2 A r ( , ) A r ( , ) c o un t s ( x ) e l e c t r i c f i e l d ( a . u . ) Σ p e,z >0 z ( n , m ) (c) Σ p e,z <0 Σ p e,z >0 Σ p e,z <0 FIG. 2: (a) Measured asymmetry A ( n,m ) z of electron emission along z for Ar(n,m), ( n, m ) = { (1 , , (2 , } over CEP. (b) Sameas (a) but simulated. The curves of the Ar + monomer (gray) serve as a reference in (a) and (b). (c-e) Simulated distributionsof ionization times of the first electron, t st , for trajectories leading to Ar(1,2) for three values of the CEP. (f-h) Same as (c-e)but for trajectory pairs leading to Ar(2,2) with the distributions of the ionization times of the second electron, t nd , shown inblue. The laser electric field E z ( t ) (gray dashed) is also shown for reference. The time-distributions are separated dependingon whether the (sum) momentum of the electron (pair) reaches positive (upper halves) or negative momentum (lower halves). comes stronger. Accordingly, the positive valued double-peak structure corresponding to the field maximum at t B becomes gradually larger; at ϕ CEP = 0 . π the negativeand positive double-peak structures are roughly equal inarea. As a consequence, A (1 , z varies from a large neg-ative value at ϕ CEP = 0 to roughly 0 at ϕ CEP = 0 . π .Thus, the CEP-dependence of A (1 , z in Figs. 2(a,b) canto a good degree be explained straightforwardly usingstandard strong-field arguments based on the relation p e = − A ( t i ) and the sub-cycle dependence of the ion-ization rate on CEP.To explain the CEP-dependence of A (2 , z for Ar(2,2),also the distribution of t nd , blue-colored in Fig. 2, mustbe considered. The distributions of t st in the Ar(2,2)channel, although different in amplitude from those ofchannel Ar(1,2), are also dominated by two peaks perlaser cycle. In contrast, the distribution of t nd for ϕ CEP = 0 in Fig. 2(f) is dominated by only one peak.It is delayed by a laser-half-cycle to the strongest fieldmaximum at t A and points into the negative direction.Again, the reasons for the delay and the single peak-structure will be discussed below. Together with the t st peaks that also point into the negative direction, thissingle t nd peak leads to A (2 , z < ϕ CEP = 0, inagreement with Figs. 2(a,b). As the CEP increases, thehalf-cycle-delayed negative t nd peak due to the decreas-ing field-maximum at t A becomes weaker, and the posi-tive t nd peak due to the increasing field maximum at t B becomes stronger. Together with the t st distributionsthat behave similarly as in the Ar(1,2) case, this causesthat A (2 , z moves towards positive values, reaches ≈ ϕ CEP = 0 . π and a large positive value for ϕ CEP = 0 . π [see Figs. 2(a,b)].We now turn to discussing the origin of the t st double- peak and the half-cycle delayed single-peak structure of t nd . As we will see, this will also explain the CEPleft-shift of A (2 , z relative to A (1 , z . To this end, wetraced the classical trajectories leading to the Ar(2,2)channel. For simplicity, but without loss of generality, weselect for this in-depth analysis the electron pairs emittedwithin [ − . T, . T ] and with negative sum momen-tum for the case of ϕ CEP = 0 [indicated by a green boxin Fig. 2(f)]. The resulting time-distributions, displayedin Fig. 3(a), show that the emissions can be classified intotwo types according to the relative emission time of thefirst and second electrons: One, where the two emissionshappen isolated of each other ( t st ∈ [ − . T, . T ]),and a second one, where the two emission steps hap-pen in a concerted manner within the same half-cycle( t st ∈ [0 . T, . T ]).Typical trajectories for both the isolated and concerted case are displayed in Fig. 3. We show in the Suppl. Mat.[48] that they are representative for all emitted trajecto-ries. The trajectories for the isolated case of double ion-ization (DI) [Figs. 3(b,c)] show that the first electron isimmediately flying away from its own parent nucleus (redcurve). The second electron, in contrast, is transferredto the other nucleus, where it is subsequently temporallycaptured by the Coulomb potential of the neighboringAr ion. It becomes ionized only during the next laser-half-cycle around the peak of the field. We refer to thiselectron transfer process across the system boundariesand the subsequent capture process that results in theionization delay as LITE.LITE also plays a significant role in the concerted typeof DI. Two cases can be distinguished: Representativeelectron trajectories [Figs. 3(d,e)] show that for case 1 one of the electrons is emitted at one site and trans-ferred to the other site by LITE. There it is captured by c t E t c isolated concerted i s o l a t e d c o n c e r t e d -5 0 5 0 1 2 c o un t s ( x ) z ( a . u . ) x (a.u.) En e r g y ( a . u . ) time (optical cycles)100-10 e l . - f i e l d ( a . u . ) time (optical cycles) z ( a . u . ) z ( a . u . ) En e r g y ( a . u . ) case 1case 2 el.-field t t /t t t t e ion e ion t t t c -3 0 3 6 FIG. 3: Classical trajectory analysis for channel Ar(2,2). (a)Ionization time distributions of first (red), t st , and second(blue), t nd , electrons for electron pairs emitted within [-0.25T, 0.75T] and with negative sum momentum for ϕ CEP =0. The left (b,d,f) and right columns (c,e,g) show typical elec-tron trajectories in space and over time, respectively. Thetrajectories are classified in isolated ( t st ∈ [ − . T, . T ],(b,c)) and concerted ( t st ∈ [0 . T, . T ], (d-g)). t C , t E de-note the times of collision and excitation. For better visibilty,the orbits in (b,d,f) are shown for t > − . T . the Coulomb potential, collides with the second electroninitially on this site, and produces a doubly excited neu-tral atom, i.e., an Ar -Ar ∗∗ dimer. The highly excitedAr ∗∗ atom is then doubly ionized before the next peakof the laser field, resulting in an Ar(2,2) dimer. Case 2 [Figs. 3(f,g)] starts similarly: An electron is emitted atone site and is transferred to the other by LITE. How-ever, in this case the energy exchange by collisions withthe second electron is larger, so that one of the electronsgains enough energy to ionize soon. The other electronloses some of its energy and is trapped by the Coulombpotential, forming a transient Ar -Ar + ∗ complex. Thecaptured electron finally ionizes at or after the next peakof the laser field and produces an Ar(2,2) dimer. -0.8-0.40.00.40.8 0.5 1.5 20 Ar(1,2)Ar(2,2) isolatedAr(2,2) concerted
CEP (rad) A s y mm e t r y z ( n , m ) FIG. 4: Simulated CEP-dependence of the asymmetry A z for channels Ar(1,2) and Ar(2,2). The latter is separatedinto isolated and concerted two-electron emissions based on∆ t = t st − t A,B,C , where t A,B,C (indicated in Fig. 2) arethe instants of the laser field-maxima right before a given t st peak that initiates the electron emission at t st : Isolated for0 ≤ ∆ t ≤ . T , concerted for 0 . T ≤ ∆ t ≤ . T . This second case is reminiscent of the recollision-induced excitation with subsequent field ionization(RESI) process well-known for monomers [53–55]. Here,however, the collision-excitation step takes place on a dis-tant entity and is enabled only by a preceding LITE pro-cess. Further explanations and a visualization of the roleof LITE in the three different DI scenarios, as well as ad-ditional data and discussion on the role of the alignmentof the argon dimer with respect to the laser polariza-tion direction, the correlation between the two emittedelectrons due to the collisions induced by LITE, and aspatio-temporal analysis of the electron transfer is pro-vided in the Suppl. Mat. [48].The finding that the DI dynamics to Ar(2,2) is dom-inated by an electron transfer process (LITE), explainswhy the second electron emission is delayed by a laserhalf-cycle to its initiating laser field-peak [cf. the t nd distributions in Figs. 2(f)-(h)]. Likewise, also the double-peak structure of t st can be explained by LITE: In theconcerted cases of DI the first electron is transferred andtherefore is emitted with delay, giving rise to the secondpeak. The first, undelayed peak arises during the isolatedcases of DI and during single ionization (SI) to Ar(1,2).The delayed peak in SI corresponds to cases where thefirst electron becomes transferred but the second electronstays bound, see Suppl. Mat. [48] for further details.Finally, based on the fact that the first ionization stepproceeds similarly for the Ar(1,2) and Ar(2,2) channels[cf. Figs. 2(c)-(h)], we can now investigate which of thetwo DI cases, the isolated or the concerted one, is re-sponsible for the distinct CEP-shift observed betweenthe A (1 , z and A (2 , z curves in Figs. 2(a,b). To see this,we plot in Fig. 4 A (2 , z separately for the isolated andconcerted contributions to Ar(2,2), in comparison with A (1 , z taken from Fig. 2(b). The separated curves re-veal that the uncorrelated two-electron emission of theisolated case introduces a notable shift, but the mainshift is introduced by the concerted pathway. The rea-son is that for this case the electron-electron interac-tion dynamics triggered in the excited argon atom uponelectron-transfer by LITE leads to electron emission overa much broader range of time within the laser cycle ascompared to a purely field-driven ionization dynamicsconfined to around the crests of the laser cycle.In conclusion, we have experimentally and theoreti-cally studied the ionization-fragmentation dynamics ofargon dimers in intense few-cycle laser pulses with aknown CEP. We observe a distinct CEP-shift of the elec-tron emission asymmetry between the Ar + + Ar andAr + Ar fragmentation channels. Using a classicalensemble model we find that this CEP-shift is due toelectron-electron interaction mediated by a field-drivenelectron transfer process (LITE) from one argon atomto the other. Our work, thus, heralds the possibilityto use strong laser fields for controlling sub-cycle inter-molecular electron-transfer processes where the trans-ferred electron can excite electronic dynamics on a dis-tant molecule. This finding opens up a new route forcontrolling molecular processes with intense laser pulsesbeyond mere bond-breaking reactions.We thank Prof. Difa Ye and Jing Chen, and Dr.Zongqiang Yuan for stimulating discussion. This workwas supported by the Austrian Science Fund (FWF),Grants No. P28475-N27, and P30465-N27, the NationalKey Research and Development Program of China (No.2019YFA0307702), the National Natural Science Foun-dation of China (Nos. 11834015, 11847243, 11804374,11922413 and 11874392), and the Strategic Priority Re-search Program of the Chinese Academy of Sciences (No.XDB21010400). ∗ [email protected] † [email protected][1] A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran,M. K. Nazeeruddin, E. W.-G. Diau, C.-Y. Yeh, S. M.Zakeeruddin, and M. Gratzel, Science , 629 (2011).[2] J. Lu, S. Liu, and M. Wang, Front. Chem. (2018),10.3389/fchem.2018.00541.[3] O. Morawski, K. Izdebska, E. Karpiuk, J. Nowacki,A. Suchocki, and A. L. Sobolewski, Phys. Chem. Chem.Phys. , 15256 (2014).[4] M. D. K¨ark¨as, O. Verho, E. V. Johnston, andB. ˚Akermark, Chem. Rev. , 11863 (2014).[5] J. Barber, Chem. Soc. Rev. , 185 (2009).[6] J.-d. Rochaix, Biochim. Biophys. Acta - Bioenerg. ,375 (2011).[7] M. F. Kling, C. Siedschlag, A. J. Verhoef, J. I.Khan, M. Schultze, T. Uphues, Y. Ni, M. Uiberacker,M. Drescher, F. Krausz, and M. J. J. Vrakking, Science , 246 (2006).[8] M. F. Kling, P. von den Hoff, I. Znakovskaya, andR. de Vivie-Riedle, Phys. Chem. Chem. Phys. , 9448(2013).[9] M. Kremer, B. Fischer, B. Feuerstein, V. L. B. de Jesus, V. Sharma, C. Hofrichter, A. Rudenko, U. Thumm, C. D.Schr¨oter, R. Moshammer, and J. Ullrich, Phys. Rev.Lett. , 213003 (2009).[10] B. Fischer, M. Kremer, T. Pfeifer, B. Feuerstein,V. Sharma, U. Thumm, C. Schr¨oter, R. Moshammer,and J. Ullrich, Phys. Rev. Lett. , 223001 (2010).[11] I. Znakovskaya, P. von den Hoff, G. Marcus,S. Zherebtsov, B. Bergues, X. Gu, Y. Deng, M. Vrakking,R. Kienberger, F. Krausz, R. de Vivie-Riedle, andM. Kling, Phys. Rev. Lett. , 063002 (2012).[12] H. Lippert, V. Stert, C. P. Schulz, I. V. Hertel, andW. Radloff, Phys. Chem. Chem. Phys. , 2718 (2004).[13] T. Schultz, E. Samoylova, W. Radloff, I. V. Hertel, A. L.Sobolewski, and W. Domcke, Science , 1765 (2004).[14] D. A. Horke, H. M. Watts, A. D. Smith, E. Jager,E. Springate, O. Alexander, C. Cacho, R. T. Chapman,and R. S. Minns, Phys. Rev. Lett. , 163002 (2016).[15] J. Ehrmaier, T. N. Karsili, A. L. Sobolewski, andW. Domcke, J. Phys. Chem. A , 4754 (2017).[16] E. J. Rabe, K. L. Corp, A. L. Sobolewski, W. Dom-cke, and C. W. Schlenker, J. Phys. Chem. Lett. , 6257(2018).[17] L. Cederbaum, J. Zobeley, and F. Tarantelli, Phys. Rev.Lett. , 4778 (1997).[18] T. Jahnke, A. Czasch, M. S. Sch¨offler, S. Sch¨ossler,A. Knapp, M. K¨asz, J. Titze, C. Wimmer, K. Kreidi,R. E. Grisenti, A. Staudte, O. Jagutzki, U. Hergenhahn,H. Schmidt-B¨ocking, and R. D¨orner, Phys. Rev. Lett. , 163401 (2004).[19] N. Sisourat, N. V. Kryzhevoi, P. Kolorenˇc, S. Scheit,T. Jahnke, and L. S. Cederbaum, Nat. Phys. , 508(2010).[20] T. Mizuno, P. C¨orlin, T. Miteva, K. Gokhberg, A. Kuleff,L. S. Cederbaum, T. Pfeifer, A. Fischer, and R. Mosham-mer, J. Chem. Phys. , 104305 (2017).[21] X. Ren, E. Jabbour Al Maalouf, A. Dorn, and S. Denifl,Nat. Commun. , 11093 (2016).[22] B. Manschwetus, H. Rottke, G. Steinmeyer, L. Foucar,A. Czasch, H. Schmidt-B¨ocking, and W. Sandner, Phys.Rev. A , 013413 (2010).[23] B. Ulrich, A. Vredenborg, A. Malakzadeh, M. Meckel,K. Cole, M. Smolarski, Z. Chang, T. Jahnke, andR. D¨orner, Phys. Rev. A - At. Mol. Opt. Phys. , 013412(2010).[24] J. Wu, A. Vredenborg, B. Ulrich, L. Schmidt, M. Meckel,S. Voss, H. Sann, H. Kim, T. Jahnke, and R. D¨orner,Phys. Rev. Lett. , 043003 (2011).[25] B. Ulrich, A. Vredenborg, A. Malakzadeh, L. P. H.Schmidt, T. Havermeier, M. Meckel, K. Cole, M. Smo-larski, Z. Chang, T. Jahnke, and R. D¨orner, J. Phys.Chem. A , 6936 (2011).[26] J. Wu, A. Vredenborg, B. Ulrich, L. P. H. Schmidt,M. Meckel, S. Voss, H. Sann, H. Kim, T. Jahnke, andR. D¨orner, Phys. Rev. A - At. Mol. Opt. Phys. , 061403(2011).[27] J. Wu, M. Kunitski, L. P. H. Schmidt, T. Jahnke, andR. D¨orner, J. Chem. Phys. , 104308 (2012).[28] K. Hoshina, H. Hagihara, and M. Tsuge, J. Phys. Chem.A , 826 (2012).[29] J. Wu, M. Meckel, L. Schmidt, M. Kunitski, S. Voss,H. Sann, H. Kim, T. Jahnke, A. Czasch, and R. D¨orner,Nat. Commun. , 1113 (2012).[30] A. von Veltheim, B. Manschwetus, W. Quan,B. Borchers, G. Steinmeyer, H. Rottke, and W. Sandner, Phys. Rev. Lett. , 023001 (2013).[31] C. Wu, C. Wu, D. Song, H. Su, X. Xie, M. Li, Y. Deng,Y. Liu, and Q. Gong, J. Chem. Phys. , 141101(2014).[32] M. Amada, Y. Sato, M. Tsuge, and K. Hoshina, Chem.Phys. Lett. , 24 (2015).[33] X. Ding, M. Haertelt, S. Schlauderer, M. S. Schuurman,A. Y. Naumov, D. M. Villeneuve, A. R. W. McKellar,P. B. Corkum, and A. Staudte, Phys. Rev. Lett. ,153001 (2017).[34] Q. Cheng, X. Xie, Z. Yuan, X. Zhong, Y. Liu, Q. Gong,and C. Wu, J. Phys. Chem. A , 3891 (2017).[35] A. S. Bogomolov, V. G. Goldort, S. A. Kochubei, andA. V. Baklanov, J. Chem. Phys. , 234304 (2017).[36] T. Ditmire, J. W. G. Tisch, E. Springate, M. B. Ma-son, N. Hay, R. A. Smith, J. Marangos, and M. H. R.Hutchinson, Nature , 54 (1997).[37] V. Kumarappan, M. Krishnamurthy, and D. Mathur,Phys. Rev. Lett. , 085005 (2001).[38] C. Jungreuthmayer, M. Geissler, J. Zanghellini, andT. Brabec, Phys. Rev. Lett. , 133401 (2004).[39] T. Fennel, K. H. Meiwes-Broer, J. Tiggesb¨aumker, P. G.Reinhard, P. M. Dinh, and E. Suraud, Rev. Mod. Phys. , 1793 (2010) .[40] B. Sch¨utte, M. Arbeiter, T. Fennel, G. Jabbari, A. Kuleff,M. Vrakking, and A. Rouz´ee, Nat. Commun. , 8596(2015).[41] B. Sch¨utte, C. Peltz, D. R. Austin, C. Str¨uber, P. Ye,A. Rouz´ee, M. J. Vrakking, N. Golubev, A. I. Kuleff,T. Fennel, and J. P. Marangos, Phys. Rev. Lett. ,063202 (2018).[42] C. Smeenk, J. Z. Salvail, L. Arissian, P. B. Corkum, C. T.Hebeisen, and A. Staudte, Opt. Express , 9336 (2011).[43] R. D¨orner, V. Mergel, O. Jagutzki, L. Spielberger, J. Ull-rich, R. Moshammer, and H. Schmidt-B¨ocking, Phys.Rep. , 95 (2000).[44] X. Xie, K. Doblhoff-Dier, S. Roither, M. S. Sch¨offler,D. Kartashov, H. Xu, T. Rathje, G. G. Paulus,A. Baltuˇska, S. Gr¨afe, and M. Kitzler, Phys. Rev. Lett. , 243001 (2012).[45] L. Zhang, X. Xie, S. Roither, D. Kartashov, Y. Wang,C. Wang, M. Sch¨offler, D. Shafir, P. B. Corkum,A. Baltuˇska, I. Ivanov, A. Kheifets, X. Liu, A. Staudte,and M. Kitzler, Phys. Rev. A , 061401 (2014). [46] X. Xie, T. Wang, S. Yu, X. Lai, S. Roither, D. Kartashov,A. Baltuˇska, X. Liu, A. Staudte, and M. Kitzler, Phys.Rev. Lett. , 243201 (2017).[47] T. Rathje, N. G. Johnson, M. M¨oller, F. S¨ußmann,D. Adolph, M. K¨ubel, R. Kienberger, M. F. Kling, G. G.Paulus, and A. M. Sayler, J. Phys. B At. Mol. Opt. Phys. , 074003 (2012).[48] See Supplemental Material at [URL] for additional ex-planations and supporting data on the electron emissionanisotropy, the numerical model and the LITE mecha-nism, for which we have made use of Refs. [56–58].[49] X. Xie, C. Wu, Z. Yuan, D. Ye, P. Wang, Y. Deng, L. Fu,J. Liu, Y. Liu, and Q. Gong, Phys. Rev. A , 023417(2015).[50] A. Scrinzi, M. Geissler, and T. Brabec, Phys. Rev. Lett. , 706 (1999).[51] F. Faisal, J. Phys. B At. Mol. , L89 (1973).[52] H. Reiss, Phys. Rev. A , 1786 (1980).[53] B. Feuerstein, R. Moshammer, D. Fischer, A. Dorn,C. Schr¨oter, J. Deipenwisch, J. Crespo Lopez-Urrutia,C. H¨ohr, P. Neumayer, J. Ullrich, H. Rottke, C. Trump,M. Wittmann, G. Korn, and W. Sandner, Phys. Rev.Lett. , 043003 (2001).[54] A. Rudenko, K. Zrost, B. Feuerstein, V. de Jesus,C. Schr¨oter, R. Moshammer, and J. Ullrich, Phys. Rev.Lett. , 253001 (2004).[55] Y. Wang, S. Xu, W. Quan, C. Gong, X. Lai, S. Hu,M. Liu, J. Chen, and X. Liu, Phys. Rev. A , 053412(2016).[56] L. Zhang, X. Xie, S. Roither, Y. Zhou, P. Lu,D. Kartashov, M. Sch¨offler, D. Shafir, P. B. Corkum,A. Baltuˇska, A. Staudte, and M. Kitzler, Phys. Rev.Lett. , 193002 (2014).[57] N. Camus, B. Fischer, M. Kremer, V. Sharma,A. Rudenko, B. Bergues, M. K¨ubel, N. G. Johnson, M. F.Kling, T. Pfeifer, J. Ullrich, and R. Moshammer, Phys.Rev. Lett. , 073003 (2012).[58] B. Bergues, M. K¨ubel, N. G. Johnson, B. Fischer, N. Ca-mus, K. J. Betsch, O. Herrwerth, A. Senftleben, A. M.Sayler, T. Rathje, T. Pfeifer, I. Ben-Itzhak, R. R. Jones,G. G. Paulus, F. Krausz, R. Moshammer, J. Ullrich, andM. F. Kling, Nat. Commun.3