Image of protons ejected from H2 molecules after multiphoton excitation with intense UV (201 nm) and visible light (532 nm). The different rings indicate different channels of dissociation.

Technical Details

Ion imaging was first demonstrated at Sandia National Labs in 1987 by a collaborative research effort between Dr. David Chandler (Sandia) and Dr. Paul Houston (Cornell University). Since then, more than 100 other laboratories have adopted this technique for the study of photochemistry, bimolecular reactivity, energy transfer, dissociative electron attachment and electron scattering.

At Sandia, we have utilized this technique to study low energy secondary electrons emitted from an electron microscope. We have facilities to study unimolecular and bimolecular chemistry as well as electron driven processes (either electron scattering, dissociative electron attachment or laser driven plasmas).

We have also developed an electron imaging apparatus that allows one to study electron scattering under zero potential field and then quickly sweep the electron out of the region and image them to reveal inelastic electron scattering as well as study dissociative electron attachment processes. Once again, we have the capability to pre-excite the molecules with a laser before proceeding with the electron driven chemical process. This is a unique capability made possible by our use of laser produced electrons. The work is being supervised by David Chandler and Jonathan Frank.

Some of the “firsts” demonstrated in our laboratory include the first to demonstrate ion imaging (CH3I in 1987, collaboration with Paul Houston); the utilization polarized light to use imaging to measure the alignment of a product molecule as a function of angle for reaction products (CH3 from CH3I); the imaging of products of a bimolecular reaction (H + HI and H + D2); the measurement of the alignment and orientation of bimolecular products; and the measurement of the differential cross sections, alignment, and orientation of collision products of electronically excited state molecules.

Electron and Ion Imaging for Study of Chemical Processes

The imaging of the charged products of unimolecular or bimolecular chemical processes (either collision or photon induced) provides deep understanding into the chemical dynamics of the process. Products can be charged as a natural consequence of the chemical process or laser ionized. The image below provides a measure of the velocity (speed and angle of motion) of the ion or electron and from this measurement the energetics, and often the mechanism, of the process can be determined.

A simple example that demonstrates this power to determine the mechanism and energy of the process is shown in this image of electrons ejected by Kr atoms after multiphoton excitation. The Kr is excited with UV (214 nm) light and ultraviolet (266 nm) light and the two different rings are observed indicating two different processes that are shown next to the image.

Images of specific quantum states of NO(jf) product after scattering NO(j=.5) from Argon and collecting the images with circularly polarized light. The patterns indicate that the product molecules in specific quantum states scattering at specific angles have a propensity to rotate either clockwise or counterclockwise and this propensity alternates more quickly with low rotational state products than with high rotational state products.

The measurement of the velocity of a particular quantum state (rotational, vibrational and electronic state specified) of a chemical process gives unprecedented information about the process that produced that product. These studies, both at Sandia in the other laboratories) have provided information that has impacted combustion mechanisms, atmospheric photochemistry, the ozone hole formation, plasma processes, catalytic behavior at surfaces and the bonding of water and solvation mechanisms. These detailed studies provide data to rigorously test the best theories of chemical reactivity.


Image of NO(A, j=7) scattering r NO(A, j=0) is scattered from Ar atoms. In these experiments the NO is excited to the A electronic state that lives for 200 ns and during that time the NO(A) has a collision and a single quantum state of the product of that collisional excitation is ionized and detected, in this case the j=7 rotational state of the A electronic state.



A partial list of recent collaborators includes Dr. Richard Zare, Stanford; Dr. Peter Rakitzis, University of Crete; Drs. Mathew Costen and Kenneth McKendrick, Harriot Watt University; and Dr. Javier Aoiz, Madrid.

PIs: David W. Chandler, Jonathan Frank, Lenny Sheps, Krupa Ramasesha

Many reacting flows are inherently three dimensional and temporally evolving. High-speed 3D imaging measurements are required to capture the dynamics of key processes such as localized extinction and re-ignition in turbulent combustion. To meet that need, we have combined multi-kHz rate tomographic particle image velocimetry (tomo-PIV) and laser-induced fluorescence (LIF) imaging. Tomo-PIV provides a 3D measurement of all three velocity components, enabling us to determine key fluid dynamic quantities, such as strain rate and vorticity. Laser-induced fluorescence provides 2D and 3D measurements of select species. In one approach to 3D LIF measurements, we have combined a Sandia-built 100 kHz burst-mode laser and an acousto-optic modulator to rapidly raster scan the laser sheet across a probe volume and record LIF images on a high-speed CMOS camera. Simultaneous high-speed 3D velocity and species measurements provide a more complete picture of the dynamics of turbulence-chemistry interactions than was previously available. (See Figures 1 and 2, below.)



Fig. 1: Time sequence of simultaneous tomographic particle image velocimetry (PIV) and planar laser-induced fluorescence (PLIF) measurements at 10 kHz in a turbulent partially premixed DME/air jet flame. Blue surfaces are isosurfaces of the compressive strain rate for Uo = -15 x 103 s-1. Velocity vectors are shown in the same plane as the OH PLIF images (1 out of 16 in-plane vectors displayed). Source: Coriton, B.; Steinberg, A. M.; Frank, J. H., High-speed tomographic PIV and OH PLIF measurements in turbulent reactive flows. Exp. Fluids 2014, 55, 1743.


Fig. 2: (a) Temporal evolution of the 3D formaldehyde distribution at the base of a lifted jet flame recorded using laser-induced fluorescence excited by a 100 kHz burst-mode laser with the laser beam raster scanned using an acousto-optic modulator. (b) Temporal evolution of 3D formaldehyde and velocity field. Source: Frank, J.H., Advances in imaging of chemically reacting flows. J. Chem. Phys., 2021, 154 (4), 040901. Results based on experiments from: Zhou, B.;  Li, T.;  Frank, J. H.;  Dreizler, A.; Bohm, B., Simultaneous 10 kHz three-dimensional CH2O and tomographic PIV measurements in a lifted partially-premixed jet flame. Proc. Combust. Inst. 2021, 38, in press.


Li, T.;  Zhou, B.;  Frank, J. H.;  Dreizler, A.; Böhm, B., High-speed volumetric imaging of formaldehyde in a lifted turbulent jet flame using an acousto-optic deflector. Exp. Fluids 2020, 61, 112.


  • Bruno Coriton, former Sandia postdoc now at General Atomics, San Diego, CA
  • Bo Zhou, former Sandia postdoc now at Southern Univ. of Science and Tech., Shenzhen, China
  • Adam M. Steinberg, Georgia Institute of Technology, Atlanta, GA
  • Tao Li, Technical University of Darmstadt, Darmstadt, Germany
  • Andreas Dreizler, Technical University of Darmstadt, Darmstadt, Germany
  • Benjamin Böhm, Technical University of Darmstadt, Darmstadt, Germany

PI: Jonathan H. Frank

Figure: A 1-dimensional femtosecond/picosecond CARS image during a flame-wall quenching event. Panels (a) and (b) provide images at differing stages of quenching. Panel (c) shows the bandwidth of the ~7 fs laser source is demonstrated by recording a CARS spectrum in a nonresonant gas, Argon. A higher-resolution spectrum of the pure-rotational portion of N2 and O2 spectrum is presented in panel (e) and the associated spatially resolved temperature fit is presented in (d).

Gas-surface interfaces are present in many important chemical systems.  From the valorization of methane at catalytically active surfaces, to the impinging of combustion processes on chemically inert interfaces, the complex heterogeneous chemistry and molecular energy transfer processes occurring in the gas phase very near to interfaces must be elucidated to describe overall chemical conversions. In any application of combustion technology, for example, ubiquitous surfaces play a major role in combustion efficiency, as energy transfer into the surface and radical-radical recombination reactions quench the local chemistry. This quenching, in turn, reduces efficiency and increases harmful pollutant production. Over catalytic surfaces, the near surface gas-phase not only acts as a local reporter of catalytic activity but may also be an active participant in the overall chemical mechanism. To study this environment, we have developed a suite of multiplexed ultrafast nonlinear optical approaches capable of imaging the near-surface gas phase with ~10 µm spatial resolution. These approaches are simultaneously capable of measuring chemical speciation as well as instantaneous molecular energy distributions.  The figure above displays a one-dimensional ultrabroadband femtosecond/picosecond coherent anti-Stokes Raman spectroscopy image during a flame-wall quenching event. The approach enables direct measurement of the molecular temperature and relative concentration for all Raman-active molecules in the probed region.

Key Contributions

  • Multiplexed probing of time-resolved chemical reactions with high spatial resolution near surfaces
  • Simultaneous probing of all Raman-active species


  • Alexis Bohlin, TU Delft
  • Andreas Dreizler, TU Darmstadt

PI: Christopher J. Kliewer

Nonlinear coherent Raman spectroscopy has long been the gold standard for the nonintrusive determination of molecular temperature and speciation during chemical reactions. The coherent, laser-like, signal beam allows for the remote probing of even optically hostile environments, while blue-shifted signal of anti-Stokes Raman enables separation from the optical interference of fluorescence. However, because the third-order nonlinear optical approach has traditionally required not only the phase-matching of three input beams, but also very high laser intensities, the approach has been limited to the detection of a single point in space.

Showing both (a) 2D-CARS single-shot spectra recorded in pure N2. In the top row, a mask has been imaged within the probe laser beam to the interrogation plane and is mapped onto each rotational state of the N2 pure-rotational spectrum. On the bottom row, the mask has been removed; and (b) A higher dispersion spectrometer was assembled. The top row shows the 2D-CARS spectrum of pure N2 while the bottom spectrum is taken in air. Over 15,000 spatially correlated CARS spectra are thus acquired across a planar spatial field within a single pulse of the laser system.

At the CRF, we have developed a new two-beam “imperfect” phase-matching approach that has enabled the extension of coherent Raman spectroscopy to both 1D and 2D imaging within a single laser pulse. In the figure below, we demonstrate the simultaneous acquisition of over 15,000 spatially correlated CARS spectra across a plane of space within a single laser pulse by combining this approach with a hyperspectral imaging spectrometer, also developed in our labs. This capability for imaging the state-specific population of molecules across space is now being used in spatially, temporally, and chemically resolved studies of combustion, plasma chemistry, and catalytic reactions.

Key Contributions

  • Invention of the 2D-CARS chemical imaging approach
  • Extension of this approach to ultrabroadband pulses has enabled multiplex detection of many species

Partner: Prof. Alexis Bohlin, TU Delft

PI: Christopher J. Kliewer

Understanding complex chemical physics systems often involves correlations among properties, making multidimensional diagnostics a natural tool. The CRF invents and employs imaging methods to probe such correlations: from ion imaging, which directly measures correlated angular and velocity distributions in single molecular collisions or photodissociation to four-dimensional imaging of chemical processes in turbulent flames.