Coulomb-explosion imaging using a pixel-imaging mass-spectrometry camera

Femtosecond laser-induced Coulomb-explosion imaging of 3,5-dibromo-$3{}^{\prime },5{}^{\prime }$-difluoro-$4{}^{\prime }$-cyanobiphenyl molecules prealigned in space is explored using a pixel-imaging mass-spectrometry (PImMS) camera. The fast-event-triggered camera allows the concurrent detection of the correlated two-dimensional momentum images, or covariance maps, of all the ionic fragments resulting from fragmentation of multiple molecules in each acquisition cycle. Detailed simulation of the covariance maps reveals that they provide rich information about the parent molecular structure and fragmentation dynamics. Future opportunities for imaging the real-time dynamics of intramolecular processes are considered.

Figure 1

(a) Skeleton and (b) ball and stick representations of the 3,5-dibromo-$3{}^{\prime },5{}^{\prime }$-difluoro-$4{}^{\prime }$-cyanobiphenyl (DBrDFCNBph) molecule.

Figure 2

The Aarhus VMI spectrometer used in the current experiments. The principal components are identified and labeled; see text for details.

Figure 3

Time-of-flight spectra and examples of laboratory-frame images for (a) perpendicular and (b) parallel alignment of the molecules to the detector plane. The peaks are assigned, and images corresponding to a selection of fragment ions are shown above the time-of-flight spectrum. The gray shading identifies the range of time codes contributing to each image. Where appropriate, the forward- and backward-scattered peaks are identified by arrows labeled with f and b, respectively.

Figure 4

Time-of-flight covariance map: (a) full covariance between all time codes, as calculated from the PImMS data. The autocorrelation normally present in such a measurement is not present as the correlation of each particle with itself has been ignored. (b) A magnified detail of the full covariance plot in (a) corresponding to the correlations of $\mathrm{F}{}^{+}$ ions to all others in the spectrum. The rings highlight asymmetric correlations in the data.

Figure 5

Selected traces from the time-of-flight 2D covariance map (Fig. 4). The ion species to which each trace corresponds is labeled along with the time code. The addition of a (b) indicates that a backward-scattered peak has been selected. The top trace is the integrated time-of-flight spectrum, which is included for comparison. The vertical scales have been normalized such that the most intense peak in each trace has a maximum value of 1. The up (down) arrows indicate whether peaks are enhanced (suppressed) relative to the integrated time-of-flight spectrum, and the circles represent the near absence of ion signal.

Figure 6

Illustration of the various stages involved in the generation of a recoil frame covariance image. First, two time windows corresponding to two ion species, A and B, are identified. One of the ions is designated the “reference” ion, in this case ion A. The covariance maps plotting the correlation of each pixel in the reference frame (pixels a–d) to each pixel in the partner frame are generated. Each of these covariance maps is rotated so that the velocity of the reference ion is confined to a single vector, and they are summed together to produce a recoil-frame covariance image.

Figure 7

Covariance images obtained in (a) perpendicular and (b) parallel alignment for each ion species (labeled at the top) relative to various reference ions (labeled at the left). In each case the reference ion is confined to a vertical vector from the center of the covariance image to the top edge. The covariance images are compared to the laboratory-frame ion images (shown at the bottom of each panel).

Figure 8

(a) Laboratory-frame image of ${\mathrm{H}}^{+}$ and covariance image of ${\mathrm{H}}^{+}$ with respect to ${\mathrm{N}}^{+}$ in parallel alignment. (b) ${\mathrm{F}}^{+}$ with respect to ${\mathrm{F}}^{+}$ and ${\mathrm{Br}}^{+}$ with respect to ${\mathrm{Br}}^{+}$ in perpendicular (left) and parallel alignment (right). (c) ${\mathrm{F}}^{+}$ with respect to ${\mathrm{Br}}^{2+}$ compared to ${\mathrm{F}}^{+}$ with respect to ${\mathrm{Br}}^{+}$ in perpendicular alignment (left). Also shown are ${\mathrm{F}}^{+}$ with respect to ${\mathrm{Br}}^{+}$ and ${\mathrm{Br}}^{+}$ with respect to ${\mathrm{F}}^{+}$ in parallel alignment (right).

Figure 9

Plots of the trajectories of the fragment ions following Coulomb explosion of DBrDFCNBph carrying a single positive charge on each atom (a) in three dimensions, with axis units in angstroms, and (b) as projected onto the $xz$ plane, where the $y$ axis is coincident with the $C_2$ symmetry axis of the molecule. The trajectories are plotted up to 70 fs after the start of the simulation.

Figure 10

(left) Experimental and (right) simulated recoil-frame covariance images recorded in (a) perpendicular and (b) parallel alignment. In each case the ion species plotted is labeled at the top of the column, with the reference ion labeled to the left of the rows. The reference ion is confined to a vertical vector from the center of the covariance image to the top edge. Only the covariances involving singly charged ions are shown. Covariance maps involving ${\mathrm{N}}^{+}$ in the perpendicular geometry are not very informative due to the central location of the N atom in the molecule when viewed in this alignment, and they are therefore not shown in (a).

Figure 11

Relative average scaling factors used for the simulated recoil-frame covariance images of each ion species. See text for details.

Figure 12

Comparison of selected experimental and simulated covariance images in (a) perpendicular and (b) parallel alignment. Presented in (a) are covariance images of ${\mathrm{F}}^{+}$ relative to ${\mathrm{Br}}^{+}$ and ${\mathrm{Br}}^{+}$ relative to ${\mathrm{F}}^{+}$ as indicated. The panel below the images compares the simulated (dashed line) and experimental (solid line) covariance as a function of the relative recoil angle of the partner fragment ions. Presented in (b) are the experimental and simulated covariance images of ${\mathrm{F}}^{+}$ relative to ${\mathrm{N}}^{+}$ and ${\mathrm{Br}}^{+}$ relative to ${\mathrm{N}}^{+}$. The panel below the images compares the simulated (dashed line) and experimental covariance (solid line) as a function of the relative recoil angle of the partner fragment ions, as determined from pBASEX inverted data presented previously [9].