Portions: Copyright 2001 The Rowland Institute for Science.

Electron Diffraction of Trapped Cluster Ions

Mathias Maier-Borst, Douglas Cameron, Mordechai Rokni and Joel H. Parks

Introduction

DiffractFirst results of electron diffraction from C60+ ions stored in an rf Paul trap are reported. Gas-phase electron diffraction (GED) has been applied1 over the past several decades to the study of cluster structures by scattering electrons from cluster beams formed within a supersonic expansion. In this case the relatively broad distributions of cluster size and temperature prevents detailed measurements of structural properties as a function of cluster size under well determined thermodynamic conditions. The application of rf Paul trap techniques both to store mass selected cluster ions and to control their internal energy2, provides the possibility to apply GED methods to study the size dependence of cluster structure and structural dynamics. This paper introduces the first measurements of trapped ion electron diffraction (TIED).

The experimental apparatus is composed of an electron gun, an ion trap, a multichannel plate detector and a CCD camera assembled on a common axis to maintain alignment. In GED, the orientationally averaged scattering from the molecular structure produces a ring pattern superimposed on a smooth background resulting from the atomic scattering. GED from a neutral C60 beam was used to calibrate and develope this technique with a well defined species whose GED spectrum has been studied previously3 and to evaluate the constraints imposed by background scattering, detector saturation and rf field perturbations.

Primary effort during the development of the TIED technique was directed towards controlling the background electron scattering, optimizing trap conditions for elastic scattering and evaluating the inelastic electron scattering. We have successfully obtained diffraction patterns from >104 trapped C60+ ions. The dominant inelastic channels have been identified as primarily due to the formation of multicharged C60z+ ions. No evidence of C60+ ion fragmentation has been observed.

Diffraction from Trapped Ions Electron Scattering in Molecules Trap rf Field Effects C60+ Trapped Ion Diffraction
Experimental Setup Beam Data Sequence Trap Data Sequence Inelastic Electron Scattering
Electron Background with Faraday Cup Electron Trap C60 Neutral Beam Diffraction Mass Selection and Detection Summary and Discussion

 

Addendum: Exposure to Ionizing Radiation

Click on images to enlarge.


Diffraction from Trapped Ions

Diffraction from trapped ions.

Fig. 1. The overall experimental concept for obtaining diffraction patterns from trapped cluster ions is shown. Cluster ions can be stored in an rf Paul trap by in-situ ionization of an effusive beam emitted by oven. The trapped ions are then exposed to a high energy electron beam to obtain a diffraction pattern with adequate S/N to derive structural information. As a result of the orientational and spatial disorder of the trapped cluster ions, the diffraction pattern will be in the form of Debye-Scherrer rings similar to powder diffraction. The primary issue is the small number (102–105) and density (106-108 cm–3) of trapped cluster ions which leads to a low rate of elastic scattering relative to the incident electron beam. In this situation, the challenge is to minimize the electron background scattering within the confined trap volume and the backscattering of high energy electrons from the Faraday cup.

To table of contents


Experimental Setup

Experimental setup.

Fig. 2. The experimental setup includes an alignment structure on which the individual components are mounted. The structure maintains a cylindrical symmetry around the electron beam axis and aligns the rf trap, Faraday cup and microchannel plate (MCP) detector along this axis. The CCD camera is mounted external to the UHV chamber and images the phosphor screen within the MCP assembly which displays the diffraction pattern. The vacuum chamber achieves a base pressure of ~10–9 Torr and during data taking the pressure is ~10–8 Torr. Diffraction data is obtained for e-beam energy of 40 keV and beam current of ~0.4 mA.

To table of contents

Addendum: Exposure to Ionizing Radiation

Electron Background with Faraday Cup Electron Trap

Electron background with Faraday cup electron trap.

Fig. 3. The background electron scattering was measured by counting the rate of single electron events detected by the MCP at low incident electron beam current (~2nA) in the absence of cluster ions and at the chamber base pressure. The low background level shown above was achieved for a Faraday cup designed as an electron trap which provided asymmetric entry and escape solid angles. In addition the cup was constructed to maximize the conversion of high energy electrons to x-rays which were then absorbed in the walls.

To table of contents


Electron Scattering in Molecules

Electron scattering in molecules.

Fig. 4. Gas phase electron scattering in molecules is composed of (a) elastic scattering from individual atoms, (b) the interference of waves scattered from atoms separated by distances characteristic of the molecular structure, and (c) inelastic scattering characteristic of the molecular electron energy states. The orientationally averaged elastic contributions are expressed above in terms of the atomic scattering amplitudes fij , the atomic separations rij and the vibrational mean amplitudes lij . The total elastic scattering intensity distribution is shown on the left as a function of s and the separate contributions of the atomic and molecular scattering are displayed on the right.

To table of contents


Beam Data Sequence

Beam data sequence.

 

Fig. 5. Diffraction data was obtained first from a neutral beam of C60 emitted by the oven source shown in Fig. 2 as it passed through the trap in the absence of rf voltage. In the beam data sequence, the diffraction signal is obtained for a period of time determined by the CCD pixel saturation and then an electron background signal is obtained under identical conditions but with the C60 oven shutter closed. This data sequence is repeated over a total experimental runtime of ~2.5 h at an e-beam current of 50 nA and an oven temperature of ~800 K. Using the C60 vapor pressure, the number of molecules in the volume defined by the intersection of the molecular and electron beams is estimated to be ~5x105.

To table of contents


Signal (Shutter Open)

Signal (Shutter open).

Fig. 6. The pseudocolor image displays the CCD data obtained with the oven shutter open for a total exposure time of 70 min. The MCP diameter of 25.4 mm is indicated. The width of the Faraday cup mount of ~ 7 mm shadows the pattern and the central spot is produced by X-ray emission from an aperture in the cup as shown in Fig. 3. The plot of average signal intensity vs s (-1) is obtained by averaging the CCD pixels forming a circle around the electron beam axis neglecting pixels shadowed by the Faraday cup. The radius of the circle is related to s through the scattering angle as defined in Fig. 4.

To table of contents


Background (Shutter Closed)

Background (Shutter closed).

Fig. 7. The pseudocolor image displays the background CCD data obtained with the oven shutter closed for total exposure time of 70 min. The plot of average background intensity vs s (-1) is obtained by averaging the CCD pixels forming a circle similar to Fig. 6.

To table of contents


Difference = Signal - Background

Difference = signal - background.

Fig. 8. The pseudocolor image displays the CCD diffraction data obtained by subtracting each background pixel intensity from the corresponding signal pixel intensity and averaging this difference over all sets taken during the 70 min exposure. The plot of average difference intensity vs s (-1) is obtained by averaging the CCD pixels forming a circle around the electron beam axis similar to Fig. 6. Note the emerging ring pattern obtained after the background is removed. Also note that except for the region near s=3 –1 the background intensity exceeds the corresponding diffraction intensity.

To table of contents


C60 Neutral Beam Diffraction

C60 neutral beam diffraction.

 

Fig. 9. The pseudocolor image displays the CCD diffraction data obtained after background subtraction. The average diffraction intensity vs s (–1) is obtained by averaging over 40 data sequences each with an exposure time of 45 s. The C60 molecular diffraction data is compared with theoretical calculations using lij vibrational amplitude values derived from GED C60 data of Hedberg3 . The theoretical curve has been smoothed to account for the electron beam diameter of ~0.05 cm. The experimental curve is obtained after subtracting the background scattering signal produced by atomic contributions from C60 and residual gases as well as that due to inelastic scattering. This background signal is derived from the total scattering data using standard methods which rely upon the calculated zero crossings of the C60 molecular diffraction. The fit varied 2 parameters: a smoothing parameter and a scaling parameter for s (–1). The green curve superimposed on the data represents the standard deviation at each data point. A total exposure time of 30 min at e-beam current of 410 nA was required to obtain data with high S/N over the range 3<s(–1)<13.

To table of contents


Trap rf Field Effects

Trap rf field effects.

 

Fig. 10. The possible effects of the trap rf field on the scattered electrons was studied by comparing diffraction data produced by e-beam scattering from a neutral C60 beam passing through trap center with the rf field on, with data obtained with the rf field off. The diffraction data shown on the left indicates that Vrf=800 V0-p produces no appreciable effect on the diffraction pattern. This is consistent with estimates of the field perturbations of the scattering angle and the electron energy. Plots of the signal and background electron scattering obtained for the C60 beam are shown on the right. These indicate that the electron scattering background is slightly increased by the prescence of the rf fields. Apparently, electrons derived from ionization events and beam scattering from apertures are swept out of the trap by the rf fields and increase the measured CCD signals almost uniformly across the detector. The degree to which this might constrain the rf voltage amplitude will depend on the diffracting species.

To table of contents


Trap Data Sequence

Trap data sequence.

 

Fig. 11. Diffraction data was obtained from trapped C60+ by multiple trap loadings. The total experimental runtime is composed of repetitions of the data sequence shown above. After the trap is loaded with ions by in-situ ionization of the C60 beam, the C60+ ion species is isolated by resonantly ejecting all other m/z ions. The C60+ internal energy is then relaxed using a He gas pulse. The UHV chamber is then evacuated to ~10-8 Torr and the ions are exposed to the 40 keV e-beam for 45 s to obtain their diffraction signal. After the exposure, the ions are resonantly ejected into an ion detector to observe the mass spectrum. This provides a continuous observation of the inelastic scattering channels as well as a measure of the initial ion number. The background signal is obtained after the ejection of the ions from the trap.

To table of contents


Mass Selection and Detection

Mass selection and detection.

Fig. 12. The trap stability diagram on the left defines the parameter space (az, qz) associated with stable ion trajectories. In these experiments, operating points in this space lie along the line az=0 as shown. Mass spectra of ions stored in the trap after in-situ ionization of C60 include C60+ and fragmented fullerenes C60–2n+, as displayed in the plot at the top right. This spectrum was obtained by varying the rf drive voltage Vrf to sweep each specific ion of mass m and charge z sequentially through resonance with an external rf voltage Vexc. These excited ions are ejected through the trap endcap aperture and impinge upon the external dynode of an electron multiplier. This resonant excitation is used to isolate C60+ in these diffraction experiments as shown in the bottom right mass spectrum. This resonant ejection process is also applied after the ions are exposed to the 40 keV beam to detect the products of inelastic scattering.

To table of contents


C60+ Trapped Ion Diffraction

C60+ trapped ion diffraction.

 

Fig. 13. The pseudocolor image displays the CCD diffraction data obtained after background subtraction. The average diffraction intensity vs s (–1) is obtained by averaging over 360 data sequences each with an exposure time of 45 s. The trapped C60+ ion diffraction data is compared with theoretical calculations using the vibrational amplitudes of Hedberg3 and smoothing to account for the electron beam diameter of ~0.05 cm. The experimental curve is obtained by the procedure described in Fig. 9 and it includes 5 point smoothing. The fit varied 2 parameters similiar to analysis of the neutral beam data. The green curve superimposed on the data represents the standard deviation at each data point. The deviation between data and theory as s increases is not only a result of weaker scattering at larger angles but also because of less reliable fitting of the background scattering. A total exposure time of 4.5 h was required to obtain adequate S/N over the range 3<s(-1)<12. The total corresponding experimental runtime of 12 h indicates that the experiment is capable of providing reproducible results over such long durations.

To table of contents


Inelastic Electron Scattering

Inelastic electron scattering.

Fig. 14. Inelastic scattering data is shown which compares the mass spectra resulting from exposure of C60+ ions to low and high energy electrons. The spectrum on the left was obtained by resonant ejection of the trapped ions after ~1 s exposure to the beam of the ionization electron gun having an energy of ~100 eV at a current of ~10 mA. In addition to fragmentation products C60–2n+, a significant number of multiply charged ions C60z+ and C60–2nz+ are also produced. The mass spectrum on the right (red curve) was obtained by exposing the trapped ions for 5 min to the 0.1 mA beam of 40 keV electrons used for the diffraction experiments. The blue curve shows the mass spectrum obtained after 5 min without exposure. Note that exposure to the 40 keV beam results in only a slight decrease in the number of C60+ ions. The dominant inelastic scattering channel observed at these high energies is the production of multiply charged ions; however, in sharp contrast, to low energy electrons, no fragmentation is observed. This is consistent with previous measurements4 of excitation of gas phase C60 by high energy electrons and photons which identified the surface plasmon excitation near 20 eV as the autoionization channel. Since the ionization of C60+ requires ~12 eV, such an ionization process leaves ~8 eV in internal energy which is not expected to produce significant fragmentation. The trap operating point was chosen so that ions with z >1 were unstable, so that the trapped species was predominantly C60+ during the 45 s exposure of the diffraction measurements.

To table of contents


Mass Scaling

Mass Scaling.

 

Fig. 15. It is useful to estimate the cluster mass range which could be stored in a given trap geometry considering the rf voltage as a limiting parameter. The mass scaling plot above has been calculated for a trap frequency of 600 KHz and endcap separation z0=0.3 cm with r0=2 z0. These parameters represent a reasonable compromise to reach heavier mass limits with minimum rf voltage. The ion mass and charge which can be trapped at a specific trap stability parameter qz is shown above. In general lower values of qz require less rf voltage for a given ion mass. Plots of M/z ion vs qz and the related trapped ion well depth are shown for several rf drive voltages.

To table of contents


Addendum: Exposure to Ionizing Radiation


Summary and Discussion

This poster has introduced a new technique to measure electron diffraction from trapped ions (TIED) and has demonstrated the technique by measuring diffraction patterns from ~104 C60+ ions stored within an rf Paul trap. These measurements required a 4.5 h exposure to the ~400 nA current of a 40 keV e-beam corresponding to a total runtime of ~12 h. In this respect, diffraction from C60+ is an exceptionally severe test of the technique. The experimental system has been designed to integrate a high energy electron gun, an ion trap, a multichannel plate detector and a CCD camera so that diffraction data can be acquired reproducibly over multiple hour exposure times involving hundreds of trap loading cycles. The Faraday cup and the alignment structure have been designed to reduce background electron scattering rates to 10–8 of the primary e-beam current. In these first measurements, the diffraction patterns exhibited S/N~10 –2 over the range of momentum transfer s ~3 – 11 –1 respectively. The dominant noise contributions were from high energy background electrons, residual neutral gas electron scattering and lower energy electrons ejected from the trap by the rf field. The saturation of the CCD detector by this electron background currently places upper limits on the e-beam current of ~400 nA and on the duration of an exposure cycle of ~45 s. The cloud density was not optimized in these measurements since elastic scattering from He at the pressures (>10–6 Torr) required to relax the ion kinetic energy introduces excessive background scattering.

The inelastic scattering channels will always be an important part of these measurements. Monitoring these channels is routinely performed by resonant ejection of the trap contents into an electron multiplier after an exposure cycle. The dominant inelastic channels have been identified as primarily due to the formation of multicharged ions C60z+. No evidence of C60+ ion fragmentation has been observed. This is consistent with past observations in high energy electron scattering experiments4 which indicated that the C60+ surface plasmon excitation with ensuing second ionization dominated the interaction. Since the trap stability of multicharged species is easily manipulated, these higher z ions do not contribute to the diffraction pattern but only lead to a decrease in the number of the parent species.

To table of contents


References

1. J.H.Parks, S. Pollack and W. Hill, J. Chem. Phys. 101, 6666 (1994); J. H. Parks and A. Szke. J. Chem. Phys. 103, 1422 (1995).

2. B. Raoult and J. Farges, Rev. Sci. Instrum. 44, 430 (1973); A. Yokozeki and G. Stein, J. Appl.Phys. 49, 2224 (1978).

3. K. Hedberg, L. Hedberg, D. Bethune, C. Brown, H. Dorn, R. Johnson, M. De Vries. Science 254, 410 (1991).

4. J. W. Keller and M. A. Coplan. Chem. Phys. Lett. 193, 89 (1992); I. V. Hertel, H. Steger, J. De Vries, B. Weisser, C, Menzel, B. Kamke and W. Kamke, Phys. Rev. Lett. 68, 784 (1992).

To table of contents