Cold Target Recoil Ion Momentum Spectroscopy: a ‘momentum microscope’ to view atomic collision dynamics

https://doi.org/10.1016/S0370-1573(99)00109-XGet rights and content

Abstract

Cold Target Recoil Ion Momentum Spectroscopy (COLTRIMS) is a novel momentum space imaging technique for the investigation of the dynamics of ionizing ion, electron or photon impact reactions with atoms or molecules. It allows the measurement of the previously undetectable small three dimensional momentum vector of the recoiling target ion created in those reactions with high resolution and 4π solid angle. Combined with novel 4π electron momentum analysers it is a momentum microscope for kinematically complete scattering experiments. We review the technical development, outline the kinematics of atomic reactions from the perspective of the recoil ion, and give an overview of the studies performed with this technique. These studies yield kinematically complete pictures of the correlated motion of the fragments of atomic and molecular breakup processes, unprecedented in resolution, detail and completeness. The multiple-dimensional momentum-space images often directly unveil the physical mechanism underlying the many-particle transitions investigated. The experiments reviewed here include reactions of single and multiple capture and ionization induced by keV proton to GeV/u U92+ impact, electron and antiproton impact ionization as well as single and double ionization by photoabsorbtion and Compton scattering from threshold to 100 keV. We give an outlook on the exciting future prospects of the method for atomic physics and other fields of science.

Introduction

Stationary systems in which the interaction potential is exactly known, can be described by quantum mechanics with an extremely high precision. For the energy eigenvalues, the central stationary observable, nearly perfect agreement between experiment and theory has been obtained (see, e.g. [1], [2], [3], [4], [5], [6], [7], [8]). Dynamical many-body systems on the other hand are still a major challenge for theory and experiment. Today they are the basic issue of many fields of physics such as solid state, nuclear and atomic physics. The last is in the fortunate situation that in atomic systems the interaction potential is exactly known. Thus all disagreement between theory and experiment in this field can be attributed to the many-body aspects and the dynamics of the problem. This makes it even more surprising that such apparently simple problems as the emission of a single electron from an atom by slow charged-particle impact or the emission of two electrons by absorption or Compton scattering of a single photon still challenge theory and experiment. This puzzling lack of understanding of the dynamics of many-body systems is in troubling contrast to the importance of such systems for our everyday world. Structure and evolution of our macroscopic world is to a large extent determined by the dynamics of many-electron processes. They are responsible for many solid state effects such as superconductivity but most prominently they govern and fuel chemical reactions and all biological systems.

Atomic and molecular many-particle reactions are characterized by fully differential cross-sections (FDCS), i.e. cross-sections differential in all observables of the final-state. In an ionization process this typically corresponds to the vector momenta, spins and internal excitation of all reaction products. Such FDCS provide the most stringent test for theory. Any integration over observables often masks important characteristics of the process. In turn, experimental FDCS in the best case directly unveil mechanisms of the many-particle transition. The lack of experimental data on such important details of simple and hence fundamental atomic reactions has for long delayed the development of many-particle collision theory. The goal of Cold Target Recoil Ion Momentum Spectroscopy COLTRIMS is to provide a tool for kinematically complete studies of three- and more-particle atomic collision systems. By kinematically complete we mean that the momenta (and thus angle and energy) of all involved particles are observed in coincidence but the spin is not determined.1

For electron impact target single ionization (so called (e,2e)-experiments), systematic experimental and theoretical studies of FDCS exist (see [9], [10], [11] for reviews). For target double ionization induced by electron impact so far only very few complete differential cross sections (see [11] for a review and [12], [13], [14], [15], [16], [17], [18]) have been published. These data typically view a very small fraction of the final state momentum space. In the field of photoionization tremendous progress has been made in the recent years. Following the pioneering work of Schwarzkopf et al. [19], experimental data for FDCS for double ionization of He have been obtained by several groups [19], [20], [21], [22], [23], [24], [25], [26], [27].

All these data have been measured using traditional electron-spectroscopy techniques. To yield sufficiently high resolution for the momentum of the ejected electron, the solid angle Ω of the electron spectrometer is strongly limited, typically to less than 10−3 of 4π. In a two- or even three-electron coincidence experiment the total coincidence efficiency is thus extremely small (10−8). This explains why, so far, systematic investigations of FDCS could be performed only for a few many-particle reactions. In addition, the low counting rate and the geometry of the experimental setup mostly restricted those studies to specific kinematical conditions (e.g. all emerging particles in one plane). Obtaining a desirable overview over the complete final state with this techniques would in general require hundreds of separate experiments (see [28] for such an attempt in (e,2e)-experiments).

In this paper we review a rapid development of a new experimental approach, the COLTRIMS-technique, which overcame many of these problems on a basic level (for earlier reviews on this field see [29], [30], [31], [32], [33]). The experimental solution is an imaging technique using a well localized reaction zone and electromagnetic fields to guide all charged fragments towards large-area position- and time-sensitive detectors. From the measurement of the time-of-flight and the position of impact for each particle its momentum vector can be determined. Such an imaging technique has first been used for measuring the ion momenta from atomic reactions and has soon been generalized to electron detection as well.

For an atomic reaction it is for at least three reasons particularly desirable to determine the vector momentum of the recoiling ion. First, the charge state of the ion gives the multiplicity of the process. Second, there are many reactions of charged particles or photons with atoms where the momentum imparted to the ion gives unique information on the dynamical mechanism of the reaction. The measurement of the ion momentum alone allows, e.g. already the separation of Compton scattering from photo absorption, the identification of the electron–electron interaction between two centers or higher-order processes in transfer ionization or the determination of the final state of a capture reaction. These and many more examples of the richness of information inherent in the ionic momentum will be discussed in detail in 4 Experimental results for charged particle impact, 5 Experimental results for photon impact. Third, for a kinematically complete experiment with n particles in the final state, it is necessary to measure 3n−3 momentum components (3n−4 if the Q-value is known). The remaining 3(4) momenta can be inferred from the others by exploiting the momentum- and energy-conservation laws. In fast ion impact atomic reactions the projectile suffers typically a very small relative momentum change. Only for very selective collisions systems this momentum change could be measured [34], [35], [36]. For reactions like 3.6 MeV Ni24+ on He (see Section 4.2.6) for example the energy loss of the projectile (ΔE/E) is in the range of 10−7 and the typical scattering angle leads to a deflection in the range of 1 mm over 1 km (μrad). Thus for such fast heavy particle collisions the coincident detection of ion and electron momenta gives, via momentum conservation, the only practical access to the projectile momentum change and thus to a kinematically complete experiment. Historically this problem of the unmeasurably small projectile momentum change motivated the development of the first spectrometer for measuring transverse momenta of recoil ions [37].

The idea of such a recoil ion momentum measurement has certainly been considered by many experimental atomic physicists. One immediately recognizes, however, that the target atom at room temperature has already such a large initial momentum spread that typical momenta of the recoil ions gained in the collision are largely covered by the target thermal motion at room temperature (He at room temperature has about 4.6 a.u. mean momentum). This was the reason why the measurement of the momentum of the recoil ion was for a long time not seriously exploited as an alternative high-resolution spectroscopy technique in ion-atom collisions [38]. Using existing techniques for gas target cooling, however, the initial target momentum spread can be reduced dramatically. As will be shown below, the target can be cooled to temperatures far below 1 K. With modern laser-cooling techniques, in particular successful for cooling of the alkali atoms, temperatures in the few μ K range, i.e. a few neV kinetic energy of the target atoms, are feasible. The use of such cold trapped atoms for ion momentum measurements is currently explored in several laboratories. Using supersonic gas target devices for He a few mK has already been obtained, which is equivalent to a He kinetic energy of below 1μeV. In present COLTRIMS devices based on supersonic gas jet targets a momentum resolution of 0.05–0.2 a.u. for the recoil ion and a detection efficiency of about 60% of all ions from a reaction (4π solid angle but 60% detection efficiency) is routinely achieved.

Before high-resolution COLTRIMS was developed recoil ion momenta in atomic physics have been measured already in the sixties for slow or small impact parameter collisions, which lead to ion energies of 101–103 eV. The momenta of those ions, whose energies in the present context were extremely high, have been measured for example by Everhard and Kessel [39], [40] and Federenko [41]. For fast charged-particle collisions and photoionization, the recoil ion energies are in the meV regime or even below. The charge state distribution of such slow ions has been measured in numerous experiments using a time-of-flight technique or magnetic deflection (see [42] for a review). First attempts on measuring angles and momenta of slow recoil ions were reported already in the seventies [43], [44], [45], [46], [47], [48]. Ullrich and Schmidt-Böcking [37] succeeded 1987 in the first quantitative measurement of transverse recoil ion momenta in 340 MeV U32+ on Ne collisions [37], [49], [50]. They used a static room temperature gas target, time-of-flight measurement in a field-free drift tube and magnetic ion charge state selection. The technical development from this first recoil ion momentum spectrometer to todays most advanced momentum microscopes will be discussed in detail in Section 3. The momentum resolution of this gas cell spectrometer was considerably improved using a cooled gas cell (30 K) [51], [52], and first Multiple-DCS have been measured by coincident detection of recoil ion and projectile transverse momenta [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61]. Parallel to this development at University Frankfurt Germany, Cocke and coworkers at Kansas State University have used warm effusive gas-jet targets and an electric projection field [62], [63]. They measured the first recoil ion longitudinal momentum distributions, deducing the Q-value of the collision. Since 1991 at the University Frankfurt (Germany) [64] and at CIRIL/GANIL (Caen, France) [65] recoil ion momentum spectrometers based on supersonic gas-jet targets have been developed. The extremely low internal temperature of such gas-jets (typically below 1 K) yielded a momentum resolution of 0.2 a.u. and below, more than a factor of 10 better than effusive or static target devices. An even colder gas target has been built using precooling of the target gas to 15 K [66], [67]. The ionic momentum resolution in these devices was in most cases not limited by the internal target temperature but by the extension of the target in the ion momentum spectrometer. Spectrometers with electrostatic lenses today eliminate this problem by focussing in all three spatial dimensions and yield an ion-momentum resolution of 0.05 a.u. [31], [68], [69], [70]. Presently at least in seven laboratories such COLTRIMS devices are in operation (Argonne, GANIL, GSI, Frankfurt, KSU, LBNL, RIKEN) and many more are in preparation.

The latest step in this rapid development was the combination of such COLTRIMS spectrometers with novel electron projection spectrometers. These electron imaging systems apply the basic principle of COLTRIMS to electron detection. For very low electron energies (typically <5 eV) the same electric field which projects the positive ions onto one detector guides the negative electrons towards another detector [71], [72], [73], [74], [75]. Moshammer and coworkers have developed an electron spectrometer which superimposes a solenoidal magnetic field parallel to the electrostatic field. This novel electron analyzer extends the projection technique for electrons towards much higher electron energies [76], [77]. Such electron projection spectrometers achieve high resolution and 2π–4π solid angle even for zero energy electrons. A resolution of ±5 meV at zero energy and full solid angle up to 300 eV has been reported [77]. In particular the detection of such very low-energy electrons is extremely difficult by conventional spectrometers. These projection spectrometers can be equipped with fast delayline detectors, capable of handling multiple hits [78]. Thus, all electrons from multi-electron reactions can be analyzed simultaneously [79]. Merging a high resolution recoil ion with such a multi-electron projection spectrometer creates an extremely versatile and powerful tool for atomic and molecular collision physics. Kinematically complete experiments for single ionization in slow, adiabatic collisions, for collisions of fast protons and antiprotons, for highly charged projectile impact, for relativistic collisions, for electron-impact and photon-induced double ionization by linearly and circularly polarized light have been performed already. In pioneering experiments for multiple ionization by fast ion impact, more than two electrons have been detected together with the recoil ion momentum [80].

This review paper is organized as follows: In Section 2 we present the kinematics of atomic reactions induced by heavy projectiles, electrons and photons from the perspective of the momentum transfer to the recoiling ion. The goal is to give guidance for the interpretation of recoil ion momentum spectra and to illustrate the large amount of information on the reaction which can be obtained from the momentum of the recoiling ion. This is followed (Section 3) by a review of the technical development from the first recoil ion spectrometers to the most advanced ion-electron momentum imaging systems, termed reaction microscopes. We complete this experimental section by a brief discussion of supersonic gas-jet targets and position sensitive detectors, the two most important ingredients of COLTRIMS. In Section 4 we give an overview on the experimental results obtained by this technique so far for charged particle impact and in Section 5 for photon impact. The overview covers most fields of atomic collision physics, including single capture and single ionization by ion impact, multiple-electron processes like double capture, transfer ionization and multiple ionization, electron impact ionization, photon-induced double ionization by linearly and circularly polarized light, Compton scattering and electron emission from aligned molecules. The energies of the projectiles range, for charged particles, from a hundred eV electrons to GeV/u bare uranium and, for photons, from threshold to 100 keV. This broad range of topics illustrates the fruitful and wide impact which the still young technique of COLTRIMS has already had. We conclude this review with an outline of some of the future perspectives of this technique and by at least naming some of the possible applications to other fields of physics.

Section snippets

Kinematics of recoil ion production

Independent of the dynamics of the ionization process, the observable momenta of collision fragments (recoiling ion, electrons and scattered projectile) are interrelated by the conservation laws of momentum and energy. The final state of a reaction with n fragments in the exit channel is described by 3n momentum components (neglecting the spin) plus internal excitation energies (Q-value). Due to momentum and energy conservation, however, only 3n−3 of these momenta are independent. Thus the

Experimental technique

Today's recoil ion momentum spectroscopy is the result of 15 yr of experimental development. Historically, the first recoil ion momentum spectrometers used extended gas targets and a field free drift path for the ions. We will first briefly review these devices (Section 3.1), which were used for some of the pioneering experiments in recoil ion momentum spectroscopy. They allowed a measurement of the transverse recoil ion momentum with a solid angle of a few percent. The resolution was restricted

One-electron processes

Two types of one-electron processes have been investigated with COLTRIMS: First the transfer of one electron from a bound state of the target to a bound state of the projectile and second the emission of one target electron to the continuum. A third possible one-electron process, target excitation, does not result in a charged recoil ion and is thus in most cases impossible to detect by COLTRIMS. The same is true for projectile ionization or excitation in a one-electron process, since these

Experimental results for photon impact

The latest field to which COLTRIMS has been applied are studies on photon-induced ionization. So far experiments on He and D2 targets in the energy range from the double ionization threshold (79 eV for Helium) to 90 keV have been reported. The work covers measurements of the ratio of double to single ionization by photoabsorption R=σ2+/σ+ from threshold to 400 eV [92], the separation of Compton scattering and photoabsorption between 9 and 100 keV [93], [104], [261], the determination of fully

Outlook

The investigation of many-particle atomic collision systems, where small momentum transfers between the collision partners dominate might be the most important application of COLTRIMS. To study the correlated motion of few electron systems in momentum and spin space, wherever the latter is possible too, is of fundamental interest. Here COLTRIMS can provide a momentum resolution which is about a factor hundred better than the mean momentum of the most weakly bound electron in any stable atom.

Acknowledgements

Many people have been involved in the development of COLTRIMS reviewed here. Most of the experimental work and technical development has been done, like mostly in todays experimental physics, by collaborators who left academic physics after they received their Diploma or Ph.D. They are to numerous to mention all of them. The work was driven by a longstanding close collaboration between several experimental and theoretical groups. Here we are particularly grateful to our friends and colleagues

References (365)

  • W.R. Johnson et al.

    At. Data Nucl. Data Tables

    (1985)
  • R. Dörner et al.

    Nucl. Instr. and Meth.

    (1997)
  • J. Ullrich et al.

    Nucl. Instr. and Meth.

    (1988)
  • J. Ullrich et al.

    Nucl. Instr. and Meth.

    (1991)
  • R. Moshammer et al.

    Nucl. Instr. and Meth.

    (1996)
  • G.W. Fraser

    Nucl. Instr. and Meth.

    (1984)
  • G.W. Fraser et al.

    Nucl. Instr. and Meth. in Phys. Res.

    (1988)
  • G.W. Fraser et al.

    Nucl. Instr. and Meth. in Phys. Res.

    (1988)
  • H. Dehmelt

    Rev. Mod. Phys.

    (1990)
  • N.F. Ramsey

    Rev. Mod. Phys.

    (1990)
  • P.J. Mohr et al.

    Phys. Rev. Lett.

    (1993)
  • Th. Stöhlker et al.

    Phys. Rev. Lett.

    (1993)
  • G. Werth

    Phys. Scripta

    (1987)
  • M. Weitz et al.

    Phys. Rev. Lett.

    (1992)
  • Th. Udem et al.

    Phys. Rev. Lett.

    (1997)
  • M.A. Coplan et al., Rev. Mod. Phys. 66 (1994) 985 and References...
  • I.E. McCarthy et al.

    Rep. Prog. Phys.

    (1991)
  • A. Lahmam-Bennani

    J. Phys. B

    (1991)
  • B. El Marji et al.

    J. Phys. B

    (1996)
  • B. El Marji et al.

    J. Phys. B

    (1997)
  • C. Schröter et al.

    J. Phys. B

    (1998)
  • M.J. Ford et al.

    Phys. Rev. A

    (1995)
  • A. Lahmam-Bennani et al.

    Phys. Rev. Lett.

    (1582)
  • A. Dorn, R. Moshammer, C.D. Schröter, T. Zouros, W. Schmitt, H. Kollmus, R. Mann, J. Ullrich, Phys. Rev. Lett. 82...
  • I. Taouil, A. Lahmam-Bennani, A. Duguet, A. Duguet, L. Avaldi, Phys. Rev. Lett. 81 (1998)...
  • O. Schwarzkopf, B. Krässig, J. Elmiger, V. Schmidt, Phys. Rev. Lett. 70 (1993)...
  • O. Schwarzkopf, B. Krässig, V. Schmidt, F. Maulbetsch, J. Briggs, J. Phys. B 27 (1994)...
  • O. Schwarzkopf et al.

    J. Phys. B

    (1995)
  • A. Huetz et al.

    J. Phys. B

    (1994)
  • G. Dawber et al.

    J. Phys. B

    (1995)
  • P. Lablanquie et al.

    Phys. Rev. Lett.

    (1995)
  • J. Viefhaus et al.

    J. Phys. B

    (1996)
  • J. Viefhaus et al.

    Phys. Rev. Lett.

    (1996)
  • J.P. Wightman et al.

    J. Phys. B

    (1998)
  • A.J. Murray et al.

    Phys. Rev. A

    (1993)
  • J. Ullrich et al.

    Comm. At. Mol. Phys.

    (1994)
  • R. Dörner, V. Mergel, L. Spielberger, O. Jagutzki, S. Nüttgens, M. Unverzagt, H. Schmidt-Böcking, J. Ullrich, R.E....
  • J. Ullrich, W. Schmitt, R. Dörner, O. Jagutzki, V. Mergel, R. Moshammer, H. Schmidt-Böcking, L. Spielberger, M....
  • J. Ullrich, R. Moshammer, R. Dörner, O. Jagutzki, V. Mergel, H. Schmidt-Böcking, L. Spielberger, J. Phys. B 30 (1997)...
  • J.T. Park, in: R.G. Cook (Ed.), Collision Spectroscopy, Plenum Press, New York, 1978, p....
  • W. Wu et al.

    Phys. Rev. Lett.

    (1996)
  • R. Schuch et al.

    Phys. Rev. Lett.

    (1988)
  • J. Ullrich, H. Schmidt-Böcking, Phys. Lett. A 125 (1987)...
  • B. Fastrup, in: Methods of experimental Physics, Vol. 17, Academic Press, New York, 1980, p....
  • E. Everhart et al.

    Phys. Rev. Lett.

    (1995)
  • Q.C. Kessel et al.

    Phys. Rev.

    (1996)
  • N.V. Federenko et al.

    Sov. Phys.-Tech. Phys.

    (1956)
  • C.L. Cocke et al.

    Phys. Rep.

    (1991)
  • J.W. McConkey et al.

    Phys. Rev. Lett.

    (1972)
  • L.J. Puckett et al.

    Phys. Rev.

    (1976)
  • Cited by (1232)

    View all citing articles on Scopus
    View full text