The spatial and energy resolutions of state-of-the-art transmission electron microscopes (TEMs) have surpassed 50 pm and 5 meV. However, with respect to the time domain, even the fastest detectors combined with the brightest sources may only be able to reach the microsecond timescale. Thus, conventional methods are incapable of resolving the myriad fundamental ultrafast (i.e., attosecond to picosecond) atomic-scale dynamics. The successful demonstration of femtosecond (fs) laser-based (LB) ultrafast electron microscopy (UEM) nearly 20 years ago provided a means to span this nearly 10-order-of-magnitude temporal gap. While nanometer-picosecond UEM studies of dynamics are now well established, ultrafast Å-scale imaging has gone largely unrealized. Further, while instrument development has rightly been an emphasis, and while new modalities and uses of pulsed-beam TEM continue to emerge, the overall chemical and materials application space has been only modestly explored to date. In this Perspective, we argue that these apparent shortfalls can be attributed to a simple lack of data and detail. We speculate that present work and continued growth of the field will ultimately lead to the realization that Å-scale fs dynamics can indeed be imaged with minimally modified UEM instrumentation and with repetition rates (frep) belowand perhaps even well below1 MHz. We further argue that the use of low frep, whether for LB UEM or for chopped/bunched beams, significantly expands the accessible application space. This calls for systematically establishing modality-specific limits so that especially promising technologies can be pursued, thus, ultimately facilitating broader adoption as individual instrument capabilities expand.

In 2005, Lobastov et al. at Caltech published a paper demonstrating the feasibility of coupling a femtosecond (fs) pulsed laser to a commercial transmission electron microscope (TEM). This first system consisted of an FEI Tecnai T12 120 kV TEM with a standard Wehnelt electrode, a custom 0.3 mm diameter LaB6 source, and a 3 W, 80 MHz Spectra-Physics Ti:sapphire fs laser oscillator.1 The goal was to reach sub-picosecond scales while also preserving the base instrument capabilities and resolutions, and especially to enable coherent, in situ fs optical specimen triggering for conducting ultrafast laser-pump/electron-probe measurements. They called this approach four-dimensional ultrafast electron microscopy (4D UEM), which aligned with their ultrafast gas/vapor-phase electron diffraction (UED) and ultrafast electron crystallography (UEC) work.2–4 This first-generation 4D UEM instrument was dubbed UEM-1. Roughly two years later, the same group developed a second-generation instrument dubbed UEM-2. This instrument was based on an FEI Tecnai TF20 200 kV TEM equipped with a hybrid LaB6 field-emission gun (FEG) coupled to a Clark-MXR Yb-doped fiber oscillator/amplifier.5 Of particular note is that the development of UEM-2 was motivated in large part by some of what was learned from working with UEM-1: the need to access lower repetition rates (frep) and higher pump pulse energies in order to expand the application space. See Baskin and Zewail for an historical account of the development of fs laser-based (LB) UEM at Caltech.6 

Prior to the developments at Caltech, efforts to improve TEM temporal resolution beyond detector limits by using a pulsed beam had occurred to some degree for decades, with significant contributions coming from Bostanjoglo et al. beginning in the 1970s, though Spivak, Petrov, and Pavlyuchenko appear to be the pioneers, reporting stroboscopic TEM experiments on magnetic domain wall motion at least as early as 1966.7–11 These early efforts focused mainly on reaching nanosecond timescales via beam blanking or, later, with nanosecond pulsed lasers and photoemission.12–14 Around the time fs LB UEM was being developed at Caltech, nanosecond LB single-shot TEMdubbed dynamic TEM (DTEM)was being further advanced at Lawrence Livermore National Lab (LLNL).15,16 Members of the LLNL team later formed the company Integrated Dynamic Electron Solutions (IDES), which is now a wholly-owned subsidiary of JEOL. The technology developed at LLNL for single-shot DTEM is now employed by several groups for fs LB UEM.17–22 

Assessment of the overall stroboscopic electron microscopy literature [including scanning electron microscopy (SEM)], which again dates back decades, makes it apparent that the first use of a fs pulsed laser by Lobastov et al. to conduct ultrafast pump-probe experiments in the TEM (and later by Yang et al. in the SEM in 2010),23 was a watershed moment that catalyzed growth of the field and greatly expanded the materials, chemical, and solid-state physics application space. (It should be noted that Merano et al. first used a fs pulsed laser to generate ∼10 ps photoelectron packets in a JEOL 6360 SEM in 2005, but they did not trigger the specimen with fs laser pulses.24) Indeed, since the seminal demonstration by Zewail et al., fs LB UEM has grown in use and in scope and is now being further developed, applied, and expanded by groups around the world.17–21,25–38 Excitingly, pulsed-beam TEM is beginning to be used for new applications extending beyond time-resolved measurements, such as for the study of electron-beam radiation damage.39–43 Other variants continue to emerge and be developed as well, including new takes on photon-induced near-field electron microscopy (PINEM) and new (or re-visited) forms of specimen triggering.21,38,44–50

Being a chemist, Zewail’s ultimate vision for fs LB UEM was to be able to directly image (in real space) molecular bond dynamics.4,25,51–54 He also had an intense interest in imaging biomolecular dynamics with UEM, especially protein folding.55 Realizing such a grand vision requires, at a minimum, combined Å-fs spatiotemporal resolutions and, thus, identification and quantification of the factors that impact resolution and define the limits. Beyond this, we argue that another important factor is having access to large ranges of instrument frepespecially below 1 MHz. This allows one to probe dynamics stemming from robust ground-state excitations, including relaxation dynamics, in the largest possible number of chemical and materials systems. Note that we are not suggesting that 1 MHz is a hard cut-off and that nothing useful can be done above this frep (quite the contrary). Instead, we chose this value based on previous experience with high laser frep in UEM-1 and also based on simulations of photothermal heating and dissipation in thin, electron-transparent specimens.56–58 

Despite having been a significant motivation for the development of fs LB UEM, imaging dynamics at combined Å-fs scales has yet to be demonstrated with any UEM variant at any frep. In fact, only a few examples of sub-nm fs LB UEM static images have been reported, and none have been obtained with in situ fs specimen photoexcitation. Details of reported UEM lattice fringe images are compiled in Table I. While some examples are of lattice spacings larger than 1 nm, these are included so that a broader comparison can be made. Note that while fs LB UEM imaging has been shown to be sensitive to sub-nm dynamic changes in, for example, nanoparticle dimensions,59 it has not been used to resolve sub-nm features, such as lattice spacings, during fs photoexcitation.

TABLE I.

Compilation of fs LB HR-UEM imaging results. NR = not reported. SFEG = Schottky FEG. frep = laser repetition rate (pulses s−1).

PI/countryfrepSpacing, specimenAcquisition timeMag.e/packetτep (fwhm)Source, gun, TEMPumped?YearReference
Zewail/USA NR 1.46 nm,  NR NR NR NR 16 μm LaB6, hybrid FEG,  No 2007 5  
  C32Cl16CuN8     front illumination, FEI    
Zewail/USA NR 3.4 Å, graphite NR NR NR NR 16 μm LaB6, hybrid FEG,  No 2008 60  
       front illumination, FEI    
Li/China 80 MHz 3.4 Å, CNTs NR NR NR NR LaB6, Wehnelt,  No 2015 61  
       front illumination, JEOL    
Banhart/France 2 MHz 2.3 Å, Au 10 s, binning NR NR NR NR 0.9 mm Ta disk,  No 2016 18  
       biased Wehnelt + C0,     
       front illumination, JEOL    
Houdellier/France NR 9 Åa, crocidolite 150 s, binning NR NR NR NR W needle, cold FEG,  No 2018 29  
       side illumination, Hitachi    
Houdellier/France 2 MHz 8.65 nm, catalase 150 s, binning 4 NR 11 NR W needle, cold FEG,  No 2018 29  
       side illumination, Hitachi    
Li/Chinab 1 MHz 3.4 Å, CNTs 20 s, binning 1 NR 33, 18 <1 ps ZrO/W needle, SFEG,  No 2020 34  
       front illumination, JEOL    
Flannigan/USA 200 kHz 1.1 nm,  10 s, binning 1 150kx 811 3.6 psc 100 μm LaB6, unbiased No ⋯ This paper 
  C32Cl16CuN8     Wehnelt, front    
       illumination, FEI    
PI/countryfrepSpacing, specimenAcquisition timeMag.e/packetτep (fwhm)Source, gun, TEMPumped?YearReference
Zewail/USA NR 1.46 nm,  NR NR NR NR 16 μm LaB6, hybrid FEG,  No 2007 5  
  C32Cl16CuN8     front illumination, FEI    
Zewail/USA NR 3.4 Å, graphite NR NR NR NR 16 μm LaB6, hybrid FEG,  No 2008 60  
       front illumination, FEI    
Li/China 80 MHz 3.4 Å, CNTs NR NR NR NR LaB6, Wehnelt,  No 2015 61  
       front illumination, JEOL    
Banhart/France 2 MHz 2.3 Å, Au 10 s, binning NR NR NR NR 0.9 mm Ta disk,  No 2016 18  
       biased Wehnelt + C0,     
       front illumination, JEOL    
Houdellier/France NR 9 Åa, crocidolite 150 s, binning NR NR NR NR W needle, cold FEG,  No 2018 29  
       side illumination, Hitachi    
Houdellier/France 2 MHz 8.65 nm, catalase 150 s, binning 4 NR 11 NR W needle, cold FEG,  No 2018 29  
       side illumination, Hitachi    
Li/Chinab 1 MHz 3.4 Å, CNTs 20 s, binning 1 NR 33, 18 <1 ps ZrO/W needle, SFEG,  No 2020 34  
       front illumination, JEOL    
Flannigan/USA 200 kHz 1.1 nm,  10 s, binning 1 150kx 811 3.6 psc 100 μm LaB6, unbiased No ⋯ This paper 
  C32Cl16CuN8     Wehnelt, front    
       illumination, FEI    
a

Image not shown.

b

Claimed to have resolved CNT interlayer lattice spacings with frep = 200 kHz using LaB6, but the image was not shown.

c

Calculated using σep400ne0.2 from Ref. 131, where σep is the electron-packet temporal standard deviation, and ne is the number of electrons per packet. Gaussian fwhm τep is reported in the table.

The first sub-nm, fs LB high-resolution (HR)-UEM image was published in 2008 by Barwick et al.60 The team used UEM-2 to acquire an image of graphitized carbon with the 3.4 Å interlayer spacing resolved. Prior to this, they used UEM-2 to acquire lattice fringe images of chlorinated copper phthalocyanine crystals (1.46 nm spacing with a fringe fwhm of 7 Å).5 Unfortunately, no specific details were reported for either of these images [e.g., frep, laser pulse duration (τlp), electrons per packet, or image acquisition time].

In 2016, Bücker et al. published a fs LB HR-UEM image of the (111) lattice spacing of Au (2.3 Å), and they also saw hints of the (002) spacing (2.0 Å) in the FFT.18 This was done using an frep of 2 MHz and a remarkable 10 s acquisition time with an independently-biased, Wehnelt-based thermionic electron gun (TEG) and an additional condenser lens (C0) incorporated into the column. Again, however, some key details were not reported. Furthermore, given that the resolution is roughly the same as that specified for the base TEM, additional details about the repeatability and robustness of the measurement would have been especially informative. In 2018, another group in France, Houdellier et al., stated that they resolved the 9 Å lattice spacing of a crocidolite crystal using a side-illuminated cold-FEG tungsten needle.29 Unfortunately, no image was shown, and no other specific experimental details were given. They did, however, show an image of a catalase crystal with the 8.65 nm spacing resolved.

Finally, in 2015, Cao et al. reported an image of resolved 3.4 Å interlayer spacings in multi-walled carbon nanotubes (CNTs).61 This was reportedly accomplished with an frep of 80 MHz and a front-illuminated LaB6 source, but no additional information connected to the image was provided. In 2020, the same group claimed to have resolved the 3.4 Å interlayer lattice spacing of CNTs with a LaB6 source, but this time with a much lower frep of 200 kHz, a τlp of 300 fs, and a remarkable ∼5000 electrons per packet.34 Unfortunately, no image was shown. They did, however, report an image claiming to show the 3.4 Å interlayer spacing resolved using a ZrO/W Schottky FEG (SFEG; frep = 1 MHz, τlp = 190 fs, 33 and 18 electrons per packet, and a 20 s acquisition timethis is the entry in Table I). However, it is difficult to determine from the image and the FFT if the spacings are clearly resolved. Furthermore, none of the other several claimed sub-nm images were shown.

An oft-stated hypothesis is that imaging sub-nm, fs dynamics with HR-UEM will not be possible until the deleterious effects of photothermal accumulation are minimized, and the photomechanical stability is robust, such that pulse-to-pulse whole-specimen spatial variation is smaller than the feature of interest. This is intuitive and was indeed commented on by Zewail and his team at least as early as 2007 and was also a factor in the design and development of UEM-2 at Caltech and of the UEM at Minnesota.5,26,60 Unfortunately, little in the way of systematic studies aimed at establishing spatiotemporal image resolution limits and trends has been reported for any UEM modality. Indeed, even basic specifications remain ill-defined, such as how static HR-UEM image resolution varies with frep (all else being the same, lower frep would require longer acquisition times to reach the same signal level). We argue that such work is critical to the advancement and growth of the field, as the results could be used to guide data-driven optimization efforts and would inform future designs and configurations. The dearth of data is itself a compelling reason to systematically explore the limits of HR-UEM.

Because of the lack of data, proposed limitations to imaging sub-nm dynamics with UEMno matter how intuitiveare supported by one-off measurements and shared anecdotes; trends are often absent, and measurements are often not repeated (or are perhaps not easily repeatable). That is, statements about what presently limits the realization of HR-UEM imaging dynamics are speculative and have not yet been rigorously tested. Thus, the degree to which each (speculated) factor is limiting is unknown. To illustrate this, the lattice spacings that have been resolved with HR-UEM imaging, for which (1) an image was shown and (2) an associated frep was reported, are compiled in Fig. 1. As can be seen, the data are sparse, there are large areas of the resolution-frep space that are empty, and no trends are obvious. Indeed, one could argue that these measurements cannot be directly compared because it is not clear if they represent the instrument/lab limits. It is also entirely unknown how fs photoexcitation would quantitatively impact these limits. We argue, in particular, that it is critically important to establish the resolution-frep limit-trend below ∼1 MHz, for which there is presently only one data point (Fig. 2). (As noted in Table I caption, Li et al. claimed to have resolved 3.4 Å lattice fringes with frep = 200 kHz and ∼5000 electrons per packet, but no image was shown.) There are many open questions, and, as we argue here, the answers can be found by simply conducting the measurements and providing all the critical details so that comparisons across instruments and labs can be made.

FIG. 1.

Illustration of some of the static fs LB HR-UEM measurements compiled in Table I that show an image and that report an associated frep. The 200 kHz, 1.1 nm image obtained in this lab is shown in Fig. 2. The 6 GHz* image was obtained with a radio frequency (RF) chopped beam and is included for comparison. The red dashed line represents an apparent limit at those frep values, while the resolution limits below ∼1 MHz are unknown.

FIG. 1.

Illustration of some of the static fs LB HR-UEM measurements compiled in Table I that show an image and that report an associated frep. The 200 kHz, 1.1 nm image obtained in this lab is shown in Fig. 2. The 6 GHz* image was obtained with a radio frequency (RF) chopped beam and is included for comparison. The red dashed line represents an apparent limit at those frep values, while the resolution limits below ∼1 MHz are unknown.

Close modal
FIG. 2.

(a) Contrast-enhanced UEM bright-field image of a C32Cl16CuN8 crystal with 1.1 nm lattice spacings resolved. Details: frep = 200 kHz, τlp = 240 fs (measured), τep = 3.6 ps (calculated), 811 electrons per packet (measured at the detector), 150kx magnification, 10 s acquisition time, binning 1, specimen unpumped, TEM column not floated. (b) Contrast-enhanced FFT of (a) showing the 0.91 nm−1 spatial frequency, with the lower-right spot magnified.

FIG. 2.

(a) Contrast-enhanced UEM bright-field image of a C32Cl16CuN8 crystal with 1.1 nm lattice spacings resolved. Details: frep = 200 kHz, τlp = 240 fs (measured), τep = 3.6 ps (calculated), 811 electrons per packet (measured at the detector), 150kx magnification, 10 s acquisition time, binning 1, specimen unpumped, TEM column not floated. (b) Contrast-enhanced FFT of (a) showing the 0.91 nm−1 spatial frequency, with the lower-right spot magnified.

Close modal

Adding to the difficulties described above is the general trend illustrated in Table I of not clearly and thoroughly reporting critical details or performing direct comparison experiments to conventional TEM (e.g., resolution at the same dose rate and total accumulated dose). Furthermore, as more results are ultimately generated, it will be critical to know what the lab conditions and the baseline instrument performance were during data acquisition. This will be an additional means with which to glean what factors impact performance. For example, the UEM Lab at Minnesota has a typical steady-state temperature control of ±0.2 °C. Indeed, without key details, evaluation and benchmarking within and across different approaches, instruments, and labs is impossible. This is also potentially problematic to growth of the field, as a lack of clearly defined performance metrics based on validated and reproducible studies could be perceived as high-risk with respect to investment by industry, funding agencies, and universities.

As mentioned above, another pulsed-beam approachlaser-free stroboscopic UEM based on beam blankinghas been developed and applied since at least the 1960s and is now making an exciting re-emergence in the configuration of a radio-frequency (RF) module paired with an aperture for beam chopping and bunching.8,9,11,62–87 Perhaps owing to very high frep (e.g., 100s of MHz to 10s of GHz) and laser-free probe operation, such pulsed-beam TEMs have recently been shown to resolve deep sub-nm features.88 Thus, background is provided here for qualitative comparison to fs LB UEM. (Note that several laser-free variants have been developed and explored, including line sampling, combination chopped beam and signal gating, and gating of the secondary electron signal.89–94)

As early as 1978, Hosokawa et al. demonstrated electron packet durations as short as 200 fs in an SEM using a combination chopper/buncher, though they resolved relatively slow dynamics (100s of picoseconds).72,95,96 More recently, Kruit et al. proposed beam deflection for electron microscopes based on photoconductive switches predicted to reach 100 fs and better than 10 nm resolution. Note that SEM beam blankers are currently being explored for a number of applications, including time-resolved cathodoluminescence.97–102 Early work to develop stroboscopic electron microscopes was motivated by the desire to probe cyclic integrated circuit operation or cyclic magnetic responses using oscillatory, phase-shifted specimen driving signals.63,65–68,78 This specific specimen driving mechanism (oscillatory electric or magnetic waveforms) was also employed for chopped-beam TEMs by Bostanjoglo et al. but at much lower frequencies, and recent work has improved the spatial resolution of such an approach to slightly better than 1.2 Å (no dynamics) by using phase locking (10 kHz but with long 12.5 µs electron pulses).11,47 Early chopped-beam SEMs could be operated with an frep of 100s of kHz up to 64 GHz with custom aperture wheels; 100s of nanometer spatial resolutions and picosecond time resolutions were routinely reached.66,67,103 (Note that while SEM beam blankers are still under active study and much more could be described, TEMs are the primary focus of this Perspective.99,100)

As previously noted, recent work on laser-free pulsed-beam TEM has focused on using GHz RF modules paired with apertures to reach 10s of picoseconds to 100s of fs. The instruments now in service have shown promise for reaching nm-ps resolutions, with potential for going below this barrier. For example, a team from Lawrence Berkeley National Lab and University of California, Berkeley, TU Eindhoven, Dow Chemical, and Thermo Fisher Scientific reported packet durations in an RF stroboscopic TEM as short as 2 ps, though they did not explicitly report a pulsed-beam real-space resolution, and no dynamics were studied.39 Members of this team have worked on RF cavity pulsers for UEM and UED for years and have suggested that this approach may be able to reach 100 fs temporal resolution at frep ≫ 1 MHz.104–109 

Another team comprised of researchers from Brookhaven National Lab, NIST, and Euclid Labs recently reported resolving 2 Å lattice fringes (Au, no dynamics) using a frequency-tunable RF strip-line pulser with an effective frep of 6 GHz (tunable from 100 MHz to 12 GHz) and with τep = 60 ps (acquisition time was not reported for the lattice fringe image).88,110,111 They also reported imaging dynamics (RF-driven mechanical oscillations) at 1200× magnification with ∼10 ps temporal resolution and a 1.5 s acquisition time.50 This again is conceptually similar to time-resolved SEM sampling of voltage- or RF-driven microcantilever beams, wherein nm-ns displacements can be resolved, or to cyclic ultrasonic excitation and sampling with TEM.11,93,94,112,113 For example, Gopinath and Hill reported 300 nm and 10 ps resolutions of potential-field dynamics (∼1 μm domains propagating at ∼100 nm ps−1) in a Gunn device with a stroboscopic SEM operated at 9.1 GHz.66,67 There are many other examples.62,63,65,68,70–75,78 Interestingly, this particular application (carrier contrast dynamics) has also been explored with LB scanning UEM (S-UEM) for some time, beginning at IBM in the 1980s with picosecond lasers (again for testing integrated circuits) and currently continued by several groups using fs lasers, beginning with a group at EPFL in 2005 for probing carrier dynamics (τlp = 200 fs, frep = 80.7 MHz, as low as 1 electron per packet, 50 nm and 10 ps spatiotemporal resolution).23,24,114–127 Note that the early LB S-UEM at IBM demonstrated resolutions of (up to) 4 ps and 100 nm.114 

As a point of reference, fs LB UEM (TEM-based) has demonstrated τep = 200 fs (τlp = 50 fs) and better than 100 fs sensitivity.27,44,128 We have conducted simulations of electron trajectories in the FEI Tecnai Femto UEM that indicate τep may become laser limited (τlp = 300 fs), which is interesting considering the instrument has a standard Wehnelt TEG.129,130 Experiments on this system also indicate the temporal resolution can be laser limited, at least for laser pulse durations of 100s of fs.131 A near laser-limited τep has been reported for a side-illuminated cold-FEG source by Houdellier et al. (τlp = 350 fs, τep = 360 fs, 20 electrons per packet, Hitachi HF2000).29 Other claims of τep becoming laser limited at low packet density have been made,132 though additional systematic studies based on various combinations of laser and instrument configurations still need to be performed in order to establish robust trends. Regardless of modality, current thinking based on the relationship between laser parameters and beam current, and on fundamental electron-beam physics (e.g., electron–electron repulsion), is that a balance must be struck between coherence, beam current, and acquisition time for achieving robust and routine Å-fs imaging of the broadest-possible range of phenomena and materials. This balance can be appreciated by inspecting Eq. (1) for the LB UEM photoelectron beam current (Ipe) reaching the detector,

Ipe=EphvηCEefrep.
(1)

Here, Ep is the pulse energy of the probe laser striking the electron source, hv is the photon energy, η is the photoelectron quantum efficiency, CE is the collection efficiency (i.e., the ratio of the number of photoelectrons reaching the detector to the number generated at the source), and e is the fundamental charge.

The challenges associated with using high frep LB UEM for studying fs–ps dynamics have been known since at least the development of UEM-1 at Caltech.5,56,133,134 Myriad nanomechanical, energy, and quantum materials phenomena are either not accessible or are obscured when using high frep owing to relatively long-lived excited states or high quality-factor oscillatory relaxations. For example, while carrier recombination times in thin films of pristine ferroelectric materials span nanoseconds to microseconds, the creation of gap states by defects can lead to trapping and long lifetimes, thus producing long-lived incoherent and coherent lattice responses and protracted relaxation times.135–137 Furthermore, using high frep can hinder full dissipation of photothermal energy prior to the next pump event, thus activating or enhancing thermally driven mechanisms that produce atomic displacements and plastic (inelastic) morphological changes (e.g., mass diffusion, melting, ablation, dislocation motion, fracture, and solid–solid structural transformations). A schematic illustrating the timescales of select chemical and materials dynamics is shown in Fig. 3 and is meant to convey how the use of low frep UEM can generally expand the application space.

FIG. 3.

Select zoology of ultrafast photoinduced dynamics and their approximate timescales. Boundary conditions become increasingly important to energy dissipation with increasing time owing to dynamics encompassing the entire specimen (i.e., initiating with charge-carrier dynamics and increasing in scale to whole-crystal/specimen motion and thermal diffusion). Relatively long-lived states and dissipation channels will limit the UEM operational frep if full recovery is desired, which can take microseconds or longer. For comparison, the frep (and the associated time between packets) needed to capture full recovery of the discrete examples is shown in relation to the overall chain of events, further illustrating the need for frep < MHz for full recovery of the broadest range of chemical and materials phenomena.

FIG. 3.

Select zoology of ultrafast photoinduced dynamics and their approximate timescales. Boundary conditions become increasingly important to energy dissipation with increasing time owing to dynamics encompassing the entire specimen (i.e., initiating with charge-carrier dynamics and increasing in scale to whole-crystal/specimen motion and thermal diffusion). Relatively long-lived states and dissipation channels will limit the UEM operational frep if full recovery is desired, which can take microseconds or longer. For comparison, the frep (and the associated time between packets) needed to capture full recovery of the discrete examples is shown in relation to the overall chain of events, further illustrating the need for frep < MHz for full recovery of the broadest range of chemical and materials phenomena.

Close modal

Though the relatively high beam currents afforded by using high frep ease difficulties associated with obtaining HR-UEM images, the trade-off is that the accessible application space becomes limited. This is, of course, not to say that high frep UEM should not be pursued (see the next paragraph). Rather, our perspective is that the reach and impact of a characterization tool is dependent not only upon the insights that can be gathered but also upon how broadly it can be applied. Thus, we argue that the growth of UEM requires broader adoption, and that broader adoption will come about by expanding the application space. Additionally, considering the high barrier to entry, we have long advocated for the development of user-friendly instruments that are fully operational as both a pulsed-beam UEM and a conventional, HR analytical TEM. This would increase the attractiveness of installing such instruments in a university user facility, owing to the typical reliance on hourly fees.

In light of the points raised thus far, having access to a large range of frepfrom 1 kHz to 1 GHz, for exampletogether with fs specimen photoexcitation in one instrument would be extremely enabling. Indeed, the relatively recent emergence of new applications for pulsed-beam TEM has shown that there are numerous compelling reasons to pursue high frep UEM. Some of these include the study of cyclic waveform excitations with high spatial resolution, electron-photon quantum effects (PINEM), and damage-reduced (or even damage-free) probing, as being pursued, for example, in optical near-field electron microscopy (ONEM).44,50,138 Ultimately, large-scale development will be driven by the emergence of critical need areas, as has occurred with quantum and energy materials research. For these specific applications, a range of frep is needed, though the issue of beam coherence is also critical.

As noted above, nature often requires a balance to be struck and compromises to be made. For example, side-illuminated cold-FEG and SFEG sources have high coherence and high brightness at the expense of electrons per packet (as well as long-term stability), which therefore also requires high frep in order to increase beam current and reduce acquisition time.27,29,128,139 As such, the reasons put forth for using sharp-tip emitters and (S)FEGs for fs LB UEM are essentially the same as those for conventional TEM—higher brightness and better spatial coherence. However, one might glean from the discussion thus far that much is still unknown about the effects of laser and electron–source parameters for front-illuminated flat photocathodes, especially for conventional Wehnelt-based TEGs. For example, as noted above, Zhu et al. claim to have resolved 3.4 Å interlayer spacings in CNTs using a LaB6 emitter and an astounding ∼5000 electrons per packet and τep < 1 ps.34 If correct, this seems to clearly indicate that the electron trajectories are behaving in a way that is yet to be fully understood—it is remarkable that packets of that size would retain coherence such that τep < 1 ps and 3.4 Å spacings could be resolved.

Some of our own recent work on large, front-illuminated flat cathodes and unbiased Wehnelt electrodes suggests that such UEM beam properties may, in fact, not be so farfetched. The image shown in Fig. 2 is one example, which was obtained with over 800 electrons per packet. Clearly, if the measurements are correct, there are differences that must be reconciled between this and the measurement by Li et al. (e.g., our calculated τep is 3.6 ps for 811 electrons per packet). This aside, we feel there is enough evidence to suggest that there are new electron-packet behaviors waiting to be discovered and new, clever instrument configurations that can be designed that will allow for the HR-UEM frep limit to be pushed well below the MHz regime. In this regard, some of the new insights we have recently gathered through simulations have indeed been surprising. For example, we found that improvements in coherence and temporal resolution are seen when photoelectron emission occurs from a region centered on the optic axis, the size of which is dictated by the (unbiased) Wehnelt aperture.129,130 We note that this is strictly for single-electron packets; in addition to experimental testing, the simulations need to be extended to multi-electron packets incorporating particle–particle interactions.

Finally, we again note that fs specimen photoexcitation may indeed pose a challenge to reaching Å-fs imaging dynamics, though systematic studies are still needed. It is thus worth considering other potentially less spatially disruptive forms of excitation. However, in our view, there are strong and compelling reasons to use fs photoexcitation for HR-UEM and thus to establish the limits through systematic experimentation. First, the use of a fs pulsed laser to coherently trigger dynamics in situ provides significant versatility and flexibility—for example, it enables polarization-dependent near-UV to near-IR excitation, above-gap carrier excitation, resonant and non-resonant excitation of electron density and vibrational/phonon modes, and both strong and weak excitations of large-gap semiconductors (e.g., the oxide perovskites). Ultrafast photoexcitation is also coherent, in that it is a temporally well-defined trigger, occurring on timescales commensurate with charge-carrier responses, as explained by Zewail for molecular ensembles.140 Importantly, the ability to directly probe these processes with high spatiotemporal resolutions (and with small probe sizes on select, well-characterized specimen regions), as well as coupling them to ultrafast x-ray spectroscopies (for example),51,141,142 will enable development of a comprehensive understanding of ultrafast dynamics spanning charge carriers to nanomechanics and without temporal gaps. This encompasses generation, coupling, conversion, transport, and decay and the associated influences of structure, morphology, geometry, and boundary conditions.

Second, the practical use of mechanical delay stages to select the time delay for fs LB UEM means the fundamental laser pulse properties may dictate the ultimate resolution limits if the microscope pinch points can be identified and optimized. Indeed, mechanical delay stages have resolutions better than 1 fs, calibrated accuracies better than ±5 fs, and bidirectional repeatabilities of better than ±2 fs. Furthermore, the full range of typical delay stages spans nanoseconds, thus providing access to initial relaxation dynamics. This again provides significant flexibility, versatility, and tunability. Third, owing to the thin specimens and large accelerating voltages, arguments pertaining to confounding factors arising from differences in electron and photon penetration depths in fs LB UEM are unfounded—the large body of quantifiable and readily-modeled published fs LB UEM data attests to the accessibility of the approach. Furthermore, claimed limitations due to laser-beam pointing stabilities are also unfounded and have not been quantifiably demonstrated to limit performance at any magnification when applied in a stable lab environment. Fourth and finally, using a fs pulsed laser is often simply cited as “challenging,” but we have operated a UEM lab for nearly 8 years primarily with a group of 5–10 graduate and undergraduate students and without technical staff. Growing the field requires moving beyond anecdotes and addressing dogma with data and detail, so it is paramount that the limits of fs LB UEM be determined and that the entire experimental parameter space be systematically and quantitatively mapped so that a clear delineation of the limits can be made and so that the chemical and materials application space can be expanded.

The sub-nm HR-UEM images that have been published to date have typically been obtained with frep ≥ 1 MHz and never with photoexcitation. However, the number of experimentally accessible ultrafast chemical and materials phenomena drops as the instrument lower-bound frep goes up owing to insufficient energy dissipation between excitation pulses. Speculative predictions are that the requirement of high spatial repeatability and efficient thermal dissipation between pulses for stroboscopic fs LB UEM will limit the resolution to ∼1 nm for coherently triggered specimens, regardless of frep or the base instrument. However, such conceptions are not based on the results of systematic studiesthe spatial resolution and frep-limits of fs LB UEM have yet to be quantitatively, rigorously, and systematically established for any UEM modality, and no information on achievable resolutions with in situ fs photoexcitation has been published.

Though fs LB UEM has been touted as a viable path to reaching Å-fs real-space imaging, as well as being uniquely positioned to probe dark states (i.e., sensitive to all elements and structures), to probe individual defect dynamics, and to probe complex heterogeneous behaviors, only a tiny fraction of the expansive parameter space has been explored and characterized. This is especially true for a minimally modified, TEG-based UEM, which may in fact be well-suited for low frep HR-UEM studies of the broadest range of chemical and materials phenomena. Further, there is a need to determine the resolution limits for such instruments so that comparisons across different configurations and modalities can be made. As can be gleaned from Fig. 1, the dearth of information on fs UEM operation below 1 MHz—regardless of instrument configuration—illustrates a significant opportunity to establish new development directions and to greatly expand the application space.

This work was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award No. DE-SC0018204, the National Science Foundation under Grant No. DMR-1654318, the National Science Foundation through the University of Minnesota MRSEC under Award No. DMR-2011401, and the National Science Foundation Graduate Research Fellowship Program under Grant No. DGE-1839286. Acknowledgment is made to the donors of the American Chemical Society Petroleum Research Fund for partial support of this research under Award No. 60584-ND10.

The authors have no conflicts to disclose.

D.J.F.: Conceptualization, data curation, funding acquisition, methodology, project administration, resources, supervision, visualization, writing – original draft, writing – review and editing. W.A.C.: Data curation, investigation, methodology, and validation. E.J.V.: Conceptualization, writing – original draft. Y.Z.: Conceptualization, writing – original draft. See the NISO CRediT taxonomy for definitions of contributing roles (credit.niso.org).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
V. A.
Lobastov
,
R.
Srinivasan
, and
A. H.
Zewail
, “
Four-dimensional ultrafast electron microscopy
,”
Proc. Natl. Acad. Sci. U. S. A.
102
,
7069
7073
(
2005
).
2.
J. C.
Williamson
and
A. H.
Zewail
, “
Structural femtochemistry: Experimental methodology
,”
Proc. Natl. Acad. Sci. U. S. A.
88
,
5021
5025
(
1991
).
3.
C.-Y.
Ruan
,
F.
Vigliotti
,
V. A.
Lobastov
,
S.
Chen
, and
A. H.
Zewail
, “
Ultrafast electron crystallography: Transient structures of molecules, surfaces, and phase transitions
,”
Proc. Natl. Acad. Sci. U. S. A.
101
,
1123
1128
(
2004
).
4.
A. H.
Zewail
, “
4D ultrafast electron diffraction, crystallography, and microscopy
,”
Annu. Rev. Phys. Chem.
57
,
65
103
(
2006
).
5.
H. S.
Park
,
J. S.
Baskin
,
O.-H.
Kwon
, and
A. H.
Zewail
, “
Atomic-scale imaging in real and energy space developed in ultrafast electron microscopy
,”
Nano Lett.
7
,
2545
2551
(
2007
).
6.
J. S.
Baskin
and
A. H.
Zewail
, “
Seeing in 4D with electrons: Development of ultrafast electron microscopy at Caltech
,”
C. R. Phys.
15
,
176
189
(
2014
).
7.
V. I.
Petrov
,
G. V.
Spivak
, and
O. P.
Pavlyuchenko
, “
Transmission electron microscope observation of the domain pattern of speedily remagnetized thin ferromagnetic films
,” in
Sixth International Congress for Electron Microscopy, Kyoto, Japan
(
Maruzen Company
,
1966
), pp.
615
616
.
8.
G. V.
Spivak
,
O. P.
Pavlyuchenko
, and
V. I.
Petrov
, “
Electron microscopic observation of alterations of the domain structure of magnetic films
,”
Bull. Acad. Sci. USSR, Phys. Ser.
30
,
822
826
(
1966
).
9.
V. I.
Petrov
,
G. V.
Spivak
, and
O. P.
Pavlyuchenko
, “
Observation of switching of thin films by means of a stroboscopic transmission electron microscope
,”
Izv. Akad. Nauk SSSR, Ser. Fiz.
31
,
362
369
(
1967
).
10.
O.
Bostanjoglo
and
T.
Rosin
, “
Stroboscopic study on ultrasonic activity in an electron microscope
,”
Mikroskopie
32
,
190
(
1976
).
11.
O.
Bostanjoglo
and
T.
Rosin
, “
Ultrasonically induced magnetic reversals observed by stroboscopic electron microscopy
,”
Opt. Acta
24
,
657
664
(
1977
).
12.
O.
Bostanjoglo
and
F.
Heinricht
, “
A reflection electron microscope for imaging of fast phase transitions on surfaces
,”
Rev. Sci. Instrum.
61
,
1223
1229
(
1990
).
13.
O.
Bostanjoglo
and
D.
Otte
, “
High speed transmission electron microscopy of laser quenching
,”
Mater. Sci. Eng.: A
173
,
407
411
(
1993
).
14.
O.
Bostanjoglo
,
R.
Elschner
,
Z.
Mao
,
T.
Nink
, and
M.
Weingärtner
, “
Nanosecond electron microscopes
,”
Ultramicroscopy
81
,
141
147
(
2000
).
15.
J.
Kang
,
A.
Ziegler
,
N.
Browning
, and
G.
Campbell
, “
Development of a dynamic transmission electron microscope (DTEM) for the study of self-propagating reactions in multilayer foils
,”
Microsc. Microanal.
11
,
484
485
(
2005
).
16.
W. E.
King
,
G. H.
Campbell
,
A.
Frank
,
B.
Reed
,
J. F.
Schmerge
,
B. J.
Siwick
,
B. C.
Stuart
, and
P. M.
Weber
, “
Ultrafast electron microscopy in materials science, biology, and chemistry
,”
J. Appl. Phys.
97
,
111101
(
2005
).
17.
L.
Piazza
,
D. J.
Masiel
,
T.
LaGrange
,
B. W.
Reed
,
B.
Barwick
, and
F.
Carbone
, “
Design and implementation of a fs-resolved transmission electron microscope based on thermionic gun technology
,”
Chem. Phys.
423
,
79
84
(
2013
).
18.
K.
Bücker
,
M.
Picher
,
O.
Crégut
,
T.
LaGrange
,
B. W.
Reed
,
S. T.
Park
,
D. J.
Masiel
, and
F.
Banhart
, “
Electron beam dynamics in an ultrafast transmission electron microscope with Wehnelt electrode
,”
Ultramicroscopy
171
,
8
18
(
2016
).
19.
S.
Ji
,
L.
Piazza
,
G.
Cao
,
S. T.
Park
,
B. W.
Reed
,
D. J.
Masiel
, and
J.
Weissenrieder
, “
Influence of cathode geometry on electron dynamics in an ultrafast electron microscope
,”
Struct. Dyn.
4
,
054303
(
2017
).
20.
Y. M.
Lee
,
Y. J.
Kim
,
Y.-J.
Kim
, and
O.-H.
Kwon
, “
Ultrafast electron microscopy integrated with a direct electron detection camera
,”
Struct. Dyn.
4
,
044023
(
2017
).
21.
H.
Liu
,
T. E.
Gage
,
P.
Singh
,
A.
Jaiswal
,
R. D.
Schaller
,
J.
Tang
,
S. T.
Park
,
S. K.
Gray
, and
I.
Arslan
, “
Visualization of plasmonic couplings using ultrafast electron microscopy
,”
Nano Lett.
21
,
5842
5849
(
2021
).
22.
B. W.
Reed
,
T.
LaGrange
,
R. M.
Shuttlesworth
,
D. J.
Gibson
,
G. H.
Campbell
, and
N. D.
Browning
, “
Solving the accelerator-condenser coupling problem in a nanosecond dynamic transmission electron microscope
,”
Rev. Sci. Instrum.
81
,
053706
(
2010
).
23.
D.-S.
Yang
,
O. F.
Mohammed
, and
A. H.
Zewail
, “
Scanning ultrafast electron microscopy
,”
Proc. Natl. Acad. Sci. U. S. A.
107
,
14993
14998
(
2010
).
24.
M.
Merano
,
S.
Sonderegger
,
A.
Crottini
,
S.
Collin
,
P.
Renucci
,
E.
Pelucchi
,
A.
Malko
,
M. H.
Baier
,
E.
Kapon
,
B.
Deveaud
, and
J.-D.
Ganière
, “
Probing carrier dynamics in nanostructures by picosecond cathodoluminescence
,”
Nature
438
,
479
482
(
2005
).
25.
D. J.
Flannigan
and
A. H.
Zewail
, “
4D electron microscopy: Principles and applications
,”
Acc. Chem. Res.
45
,
1828
1839
(
2012
).
26.
D. A.
Plemmons
,
P. K.
Suri
, and
D. J.
Flannigan
, “
Probing structural and electronic dynamics with ultrafast electron microscopy
,”
Chem. Mater.
27
,
3178
3192
(
2015
).
27.
A.
Feist
,
N.
Bach
,
N.
Rubiano da Silva
,
T.
Danz
,
M.
Möller
,
K. E.
Priebe
,
T.
Domröse
,
J. G.
Gatzmann
,
S.
Rost
,
J.
Schauss
,
S.
Strauch
,
R.
Bormann
,
M.
Sivis
,
S.
Schäfer
, and
C.
Ropers
, “
Ultrafast transmission electron microscopy using a laser-driven field emitter: Femtosecond resolution with a high coherence electron beam
,”
Ultramicroscopy
176
,
63
73
(
2017
).
28.
J.
Williams
,
F.
Zhou
,
T.
Sun
,
Z.
Tao
,
K.
Chang
,
K.
Makino
,
M.
Berz
,
P. M.
Duxbury
, and
C.-Y.
Ruan
, “
Active control of bright electron beams with RF optics for femtosecond microscopy
,”
Struct. Dyn.
4
,
044035
(
2017
).
29.
F.
Houdellier
,
G. M.
Caruso
,
S.
Weber
,
M.
Kociak
, and
A.
Arbouet
, “
Development of a high brightness ultrafast transmission electron microscope based on a laser-driven cold field emission source
,”
Ultramicroscopy
186
,
128
138
(
2018
).
30.
D.
Li
,
Z. L.
Wang
, and
Z.
Wang
, “
Revealing electron–phonon interactions and lattice dynamics in nanocrystal films by combining in situ thermal heating and femtosecond laser excitations in 4D transmission electron microscopy
,”
J. Phys. Chem. Lett.
9
,
6795
6800
(
2018
).
31.
C. W.
Barlow Myers
,
N. J.
Pine
, and
W. A.
Bryan
, “
Femtosecond transmission electron microscopy for nanoscale photonics: A numerical study
,”
Nanoscale
10
,
20628
20639
(
2018
).
32.
J.
Yang
,
Y.
Yoshida
, and
H.
Yasuda
, “
Ultrafast electron microscopy with relativistic femtosecond electron pulses
,”
Microscopy
67
,
291
295
(
2018
).
33.
M. T.
Hassan
, “
Attomicroscopy: From femtosecond to attosecond electron microscopy
,”
J. Phys. B: At., Mol. Opt. Phys.
51
,
032005
(
2018
).
34.
C.
Zhu
,
D.
Zheng
,
H.
Wang
,
M.
Zhang
,
Z.
Li
,
S.
Sun
,
P.
Xu
,
H.
Tian
,
Z.
Li
,
H.
Yang
, and
J.
Li
, “
Development of analytical ultrafast transmission electron microscopy based on laser-driven Schottky field emission
,”
Ultramicroscopy
209
,
112887
(
2020
).
35.
O.
Zandi
,
A. E.
Sykes
,
R. D.
Cornelius
,
F. M.
Alcorn
,
B. S.
Zerbe
,
P. M.
Duxbury
,
B. W.
Reed
, and
R. M.
van der Veen
, “
Transient lensing from a photoemitted electron gas imaged by ultrafast electron microscopy
,”
Nat. Commun.
11
,
3001
(
2020
).
36.
P. K.
Olshin
,
M.
Drabbels
, and
U. J.
Lorenz
, “
Characterization of a time-resolved electron microscope with a Schottky field emission gun
,”
Struct. Dyn.
7
,
054304
(
2020
).
37.
Y.
Kurman
,
R.
Dahan
,
H. H.
Sheinfux
,
K.
Wang
,
M.
Yannai
,
Y.
Adiv
,
O.
Reinhardt
,
L. H. G.
Tizei
,
S. Y.
Woo
,
J.
Li
,
J. H.
Edgar
,
M.
Kociak
,
F. H. L.
Koppens
, and
I.
Kaminer
, “
Spatiotemporal imaging of 2D polariton wave packet dynamics using free electrons
,”
Science
372
,
1181
1186
(
2021
).
38.
X.
Fu
,
Z.
Sun
,
S.
Ji
,
F.
Liu
,
M.
Feng
,
B.-K.
Yoo
, and
Y.
Zhu
, “
Nanoscale-femtosecond imaging of evanescent surface plasmons on silver film by photon-induced near-field electron microscopy
,”
Nano Lett.
22
,
2009
2015
(
2022
).
39.
C.
Kisielowski
,
P.
Specht
,
B.
Freitag
,
E. R.
Kieft
,
W.
Verhoeven
,
J. F. M.
van Rens
,
P.
Mutsaers
,
J.
Luiten
,
S.
Rozeveld
,
J.
Kang
,
A. J.
McKenna
,
P.
Nickias
, and
D. F.
Yancey
, “
Discovering hidden material properties of MgCl2 at atomic resolution with structured temporal electron illumination of picosecond time resolution
,”
Adv. Funct. Mater.
29
,
1807818
(
2019
).
40.
E. J.
VandenBussche
and
D. J.
Flannigan
, “
Reducing radiation damage in soft matter with femtosecond-timed single-electron packets
,”
Nano Lett.
19
,
6687
6694
(
2019
).
41.
H.
Choe
,
I.
Ponomarev
,
E.
Montgomery
,
J. W.
Lau
,
Y.
Zhu
,
Y.
Zhao
,
A.
Liu
,
A.
Kanareykin
, and
C.
Jing
, “
Mitigation of radiation damage in biological macromolecules via tunable picosecond pulsed transmission electron microscopy
,” bioRxiv:2020.05.15.099036 (
2020
).
42.
E. J.
VandenBussche
,
C. P.
Clark
,
R. J.
Holmes
, and
D. J.
Flannigan
, “
Mitigating damage to hybrid perovskites using pulsed-beam TEM
,”
ACS Omega
5
,
31867
31871
(
2020
).
43.
C.
Kisielowski
,
P.
Specht
,
S.
Rozeveld
,
B.
Freitag
,
E. R.
Kieft
,
J.
Kang
,
A. J.
Fielitz
,
T. R.
Fielitz
,
D.
van Dyck
, and
D. F.
Yancey
, “
Exploring functional materials by understanding beam-sample interactions
,”
Adv. Funct. Mater.
32
,
2201112
(
2022
).
44.
B.
Barwick
,
D. J.
Flannigan
, and
A. H.
Zewail
, “
Photon-induced near-field electron microscopy
,”
Nature
462
,
902
906
(
2009
).
45.
A.
Feist
,
K. E.
Echternkamp
,
J.
Schauss
,
S. V.
Yalunin
,
S.
Schäfer
, and
C.
Ropers
, “
Quantum coherent optical phase modulation in an ultrafast transmission electron microscope
,”
Nature
521
,
200
203
(
2015
).
46.
L.
Piazza
,
T. T. A.
Lummen
,
E.
Quiñonez
,
Y.
Murooka
,
B. W.
Reed
,
B.
Barwick
, and
F.
Carbone
, “
Simultaneous observation of the quantization and the interference pattern of a plasmonic near-field
,”
Nat. Commun.
6
,
6407
(
2015
).
47.
K.
Soma
,
S.
Konings
,
R.
Aso
,
N.
Kamiuchi
,
G.
Kobayashi
,
H.
Yoshida
, and
S.
Takeda
, “
Detecting dynamic responses of materials and devices under an alternating electric potential by phase-locked transmission electron microscopy
,”
Ultramicroscopy
181
,
27
41
(
2017
).
48.
M. T.
Hassan
,
J. S.
Baskin
,
B.
Liao
, and
A. H.
Zewail
, “
High-temporal-resolution electron microscopy for imaging ultrafast electron dynamics
,”
Nat. Photonics
11
,
425
430
(
2017
).
49.
R.
Dahan
,
S.
Nehemia
,
M.
Shentcis
,
O.
Reinhardt
,
Y.
Adiv
,
X.
Shi
,
O.
Be’er
,
M. H.
Lynch
,
Y.
Kurman
,
K.
Wang
, and
I.
Kaminer
, “
Resonant phase-matching between a light wave and a free-electron wavefunction
,”
Nat. Phys.
16
,
1123
1131
(
2020
).
50.
X.
Fu
,
E.
Wang
,
Y.
Zhao
,
A.
Liu
,
E.
Montgomery
,
V. J.
Gokhale
,
J. J.
Gorman
,
C.
Jing
,
J. W.
Lau
, and
Y.
Zhu
, “
Direct visualization of electromagnetic wave dynamics by laser-free ultrafast electron microscopy
,”
Sci. Adv.
6
,
eabc3456
(
2020
).
51.
M.
Chergui
and
A. H.
Zewail
, “
Electron and X-ray methods of ultrafast structural dynamics: Advances and applications
,”
ChemPhysChem
10
,
28
43
(
2009
).
52.
A. H.
Zewail
, “
Micrographia of the twenty-first century: From camera obscura to 4D microscopy
,”
Philos. Trans. R. Soc., A
368
,
1191
1204
(
2010
).
53.
A. H.
Zewail
, “
Four-dimensional electron microscopy
,”
Science
328
,
187
193
(
2010
).
54.
G. M.
Vanacore
,
A. W. P.
Fitzpatrick
, and
A. H.
Zewail
, “
Four-dimensional electron microscopy: Ultrafast imaging, diffraction and spectroscopy in materials science and biology
,”
Nano Today
11
,
228
249
(
2016
).
55.
A. H.
Zewail
, “
Filming the invisible in 4-D
,”
Sci. Am.
303
,
74
81
(
2010
).
56.
D. J.
Flannigan
and
A. H.
Zewail
, “
Optomechanical and crystallization phenomena visualized with 4D electron microscopy: Interfacial carbon nanotubes on silicon nitride
,”
Nano Lett.
10
,
1892
1899
(
2010
).
57.
S. A.
Reisbick
,
Y.
Zhang
,
J.
Chen
,
P. E.
Engen
, and
D. J.
Flannigan
, “
Coherent phonon disruption and lock-in during a photoinduced charge-density-wave phase transition
,”
J. Phys. Chem. Lett.
12
,
6439
6447
(
2021
).
58.
R. A.
Gnabasik
,
P. K.
Suri
,
J.
Chen
, and
D. J.
Flannigan
, “
Imaging coherent phonons and precursor dynamics in LaFeAsO with 4D ultrafast electron microscopy
,”
Phys. Rev. Mater.
6
,
024802
(
2022
).
59.
Y.-J.
Kim
,
H.
Jung
,
S. W.
Han
, and
O.-H.
Kwon
, “
Ultrafast electron microscopy visualizes acoustic vibrations of plasmonic nanorods at the interfaces
,”
Matter
1
,
481
495
(
2019
).
60.
B.
Barwick
,
H. S.
Park
,
O.-H.
Kwon
,
J. S.
Baskin
, and
A. H.
Zewail
, “
4D imaging of transient structures and morphologies in ultrafast electron microscopy
,”
Science
322
,
1227
1231
(
2008
).
61.
G.
Cao
,
S.
Sun
,
Z.
Li
,
H.
Tian
,
H.
Yang
, and
J.
Li
, “
Clocking the anisotropic lattice dynamics of multi-walled carbon nanotubes by four-dimensional ultrafast transmission electron microscopy
,”
Sci. Rep.
5
,
8404
(
2015
).
62.
G. S.
Plows
and
W. C.
Nixon
, “
Stroboscopic scanning electron microscopy
,”
J. Phys. E: Sci. Instrum.
1
,
595
600
(
1968
).
63.
G. Y.
Robinson
,
R. M.
White
, and
N. C.
MacDonald
, “
Probing of Gunn effect domains with a scanning electron microscope
,”
Appl. Phys. Lett.
13
,
407
408
(
1968
).
64.
G. Y.
Robinson
, “
Stroboscopic scanning electron microscopy at gigahertz frequencies
,”
Rev. Sci. Instrum.
42
,
251
255
(
1971
).
65.
G. S.
Plows
and
W. C.
Nixon
, “
Operational testing of L.S.I. arrays by stroboscopic scanning electron microscopy
,”
Microelectron. Reliab.
10
,
317
318
(
1971
).
66.
A.
Gopinath
and
M. S.
Hill
, “
A technique for the study of Gunn devices at 9.1 GHz using a scanning electron microscope
,”
IEEE Trans. Electron Devices
20
,
610
612
(
1973
).
67.
M. S.
Hill
and
A.
Gopinath
, “
Probing Gunn domains at X-band microwave frequencies using a scanning microscope
,”
J. Phys. D: Appl. Phys.
7
,
69
77
(
1974
).
68.
P. J.
Fentem
and
A.
Gopinath
, “
Voltage distributions in X-band n+-n-n+ Gunn devices using a SEM
,”
IEEE Trans. Electron Devices
23
,
1159
1165
(
1976
).
69.
A.
Gopinath
and
M. S.
Hill
, “
Deflection beam-chopping in the SEM
,”
J. Phys. E: Sci. Instrum.
10
,
229
236
(
1977
).
70.
T.
Hosokawa
,
H.
Fujioka
, and
K.
Ura
, “
Two‐dimensional observation of Gunn domains at 1 GHz by picosecond pulse stroboscopic SEM
,”
Appl. Phys. Lett.
31
,
340
(
1977
).
71.
E.
Hieke
and
G.
Meusburger
, “
SEM stroboscopy for the evaluation of IC operating functions in the subnanosecond range
,”
Rev. Sci. Instrum.
49
,
802
805
(
1978
).
72.
K.
Ura
,
H.
Fujioka
, and
T.
Hosokawa
, “
Picosecond pulse stroboscopic scanning electron microscope
,”
J. Electron Microsc.
27
,
247
252
(
1978
).
73.
M.
Masuda
,
T.
Ogura
, and
J.
Koyama
, “
Study of two‐dimensional Gunn domain dynamics using a stroboscopic SEM
,”
Appl. Phys. Lett.
33
,
888
889
(
1978
).
74.
M.
Masuda
,
T.
Ogura
,
J.
Koyama
,
H.
Fujioka
,
T.
Hosokawa
, and
K.
Ura
, “
Analysis of the highfield domain dynamics in a planar Gunn diode by using a stroboscopic SEM
,”
J. Appl. Phys.
50
,
530
536
(
1979
).
75.
H.
Fujioka
,
K.
Nakamae
, and
K.
Ura
, “
Function testing of bipolar ICs and LSIs with the stroboscopic scanning electron microscope
,”
IEEE J. Solid-State Circuits
15
,
177
183
(
1980
).
76.
O.
Bostanjoglo
and
R.
Liedtke
, “
Tracing fast phase transitions by electron microscopy
,”
Phys. Status Solidi A
60
,
451
455
(
1980
).
77.
O.
Bostanjoglo
and
T.
Rosin
, “
Mass and relaxation time of domain walls in thin NiFe films from forced oscillations
,”
J. Magn. Magn. Mater.
15-18
,
1529
1530
(
1980
).
78.
H.
Fujioka
and
K.
Ura
, “
Waveform measurements on gigahertz semiconductor devices by scanning electron microscope stroboscopy
,”
Appl. Phys. Lett.
39
,
81
82
(
1981
).
79.
A.
Steckenborn
,
H.
Münzel
, and
D.
Bimberg
, “
Cathodoluminescence lifetime pattern of GaAs surfaces around dislocations
,”
J. Lumin.
24-25
,
351
354
(
1981
).
80.
K.
Ura
,
H.
Fujioka
,
K.
Nakamae
, and
M.
Ishisaka
, “
Stroboscopic observation of passivated microprocessor chips by scanning electron microscopy
,”
Scanning Electron Microsc.
3
,
1061
1068
(
1982
).
81.
O.
Bostanjoglo
,
G.
Schlotzhauer
, and
S.
Schade
, “
Shaping trigger pulses from noisy signals and time-resolved TEM of fast phase transitions
,”
Optik
61
,
91
98
(
1982
).
82.
B.
Lischke
,
E.
Plies
, and
R.
Schmitt
, “
Resolution limits in stroboscopic electron-beam instruments
,”
Scanning Electron Microsc.
3
,
1177
1185
(
1983
).
83.
H.
Todokoro
,
S.
Fukuhara
, and
T.
Komoda
, “
Stroboscopic scanning electron microscope with 1 keV electrons
,”
Scanning Electron Microsc.
2
,
561
568
(
1983
).
84.
S. K.
Obyden
,
G. V.
Saparin
, and
S. I.
Popov
, “
Principle and possibilities of differential (poliphase) stroboscopy in the SEM
,”
Scanning
6
,
55
57
(
1984
).
85.
O.
Bostanjoglo
and
E.
Endruschat
, “
Time-resolved TEM of laser-annealing
,”
Phys. Status Solidi A
82
,
K1
K3
(
1984
).
86.
J. T. L.
Thong
, “
Picosecond electron pulse generation via beam deflection-chopping in the SEM
,”
Meas. Sci. Technol.
2
,
207
216
(
1991
).
87.
L.
Zhang
,
J. P.
Hoogenboom
,
B.
Cook
, and
P.
Kruit
, “
Photoemission sources and beam blankers for ultrafast electron microscopy
,”
Struct. Dyn.
6
,
051501
(
2019
).
88.
S. A.
Reisbick
,
M.-G.
Han
,
C.
Liu
,
Y.
Zhao
,
E.
Montgomery
,
C.
Jing
,
V. J.
Gokhale
,
J. J.
Gorman
,
J. W.
Lau
, and
Y.
Zhu
, “
Stroboscopic ultrafast imaging using RF strip-lines in a commercial transmission electron microscope
,”
Ultramicroscopy
235
,
113497
(
2022
).
89.
T.
Ikuta
and
R.
Shimizu
, “
A simple technique for the stroboscopic observation of magnetic domain response using scanning electron microscopy
,”
J. Phys. E: Sci. Instrum.
9
,
721
724
(
1976
).
90.
R.
Shimizu
and
T.
Ikuta
, “
Line‐sampling multiframe stroboscopic technique for dynamic observation of periodic phenomena at low frequencies
,”
Appl. Phys. Lett.
44
,
811
812
(
1984
).
91.
O.
Bostanjoglo
,
R. P.
Tornow
, and
W.
Tornow
, “
Nanosecond transmission electron microscopy and diffraction
,”
J. Phys. E: Sci. Instrum.
20
,
556
557
(
1987
).
92.
O.
Bostanjoglo
,
R. P.
Tornow
, and
W.
Tornow
, “
Nanosecond-exposure electron microscopy of laser-induced phase transformations
,”
Ultramicroscopy
21
,
367
372
(
1987
).
93.
H.
Ishikawa
,
H.
Dobashi
,
T.
Kodama
,
T.
Furuhashi
, and
Y.
Uchikawa
, “
Investigation of micro mechanical vibration of piezoelectric actuators using a stroboscopic SEM
,”
J. Electron Microsc.
42
,
35
40
(
1993
).
94.
C.-L.
Wong
and
W.-K.
Wong
, “
In-plane motion characterization of MEMS resonators using stroboscopic scanning electron microscopy
,”
Sens. Actuators, A
138
,
167
178
(
2007
).
95.
T.
Hosokawa
,
H.
Fujioka
, and
K.
Ura
, “
Generation and measurement of subpicosecond electron beam pulses
,”
Rev. Sci. Instrum.
49
,
624
628
(
1978
).
96.
T.
Hosokawa
,
H.
Fujioka
, and
K.
Ura
, “
Gigahertz stroboscopy with the scanning electron microscope
,”
Rev. Sci. Instrum.
49
,
1293
1299
(
1978
).
97.
A.
Bell
,
J.
Christen
,
F.
Bertram
,
F. A.
Ponce
,
H.
Marui
, and
S.
Tanaka
, “
Localization versus field effects in single InGaN quantum wells
,”
Appl. Phys. Lett.
84
,
58
60
(
2003
).
98.
S.
Sonderegger
,
E.
Feltin
,
M.
Merano
,
A.
Crottini
,
J. F.
Carlin
,
R.
Sachot
,
B.
Deveaud
,
N.
Grandjean
, and
J. D.
Ganière
, “
High spatial resolution picosecond cathodoluminescence of InGaN quantum wells
,”
Appl. Phys. Lett.
89
,
232109
(
2006
).
99.
I. G. C.
Weppelman
,
R. J.
Moerland
,
J. P.
Hoogenboom
, and
P.
Kruit
, “
Concept and design of a beam blanker with integrated photoconductive switch for ultrafast electron microscopy
,”
Ultramicroscopy
184
,
8
17
(
2018
).
100.
S.
Meuret
,
M.
Solà Garcia
,
T.
Coenen
,
E.
Kieft
,
H.
Zeijlemaker
,
M.
Lätzel
,
S.
Christiansen
,
S. Y.
Woo
,
Y. H.
Ra
,
Z.
Mi
, and
A.
Polman
, “
Complementary cathodoluminescence lifetime imaging configurations in a scanning electron microscope
,”
Ultramicroscopy
197
,
28
38
(
2019
).
101.
I. G. C.
Weppelman
,
R. J.
Moerland
,
L.
Zhang
,
E.
Kieft
,
P.
Kruit
, and
J. P.
Hoogenboom
, “
Pulse length, energy spread, and temporal evolution of electron pulses generated with an ultrafast beam blanker
,”
Struct. Dyn.
6
,
024102
(
2019
).
102.
L.
Zhang
,
M. W. H.
Garming
,
J. P.
Hoogenboom
, and
P.
Kruit
, “
Beam displacement and blur caused by fast electron beam deflection
,”
Ultramicroscopy
211
,
112925
(
2020
).
103.
J. T. L.
Thong
,
B. C.
Breton
, and
W. C.
Nixon
, “
Electron pulse generation at 64 GHz in scanning electron microscope
,”
Electron. Lett.
25
,
978
979
(
1989
).
104.
A.
Lassise
,
P. H. A.
Mutsaers
, and
O. J.
Luiten
, “
Compact, low power radio frequency cavity for femtosecond electron microscopy
,”
Rev. Sci. Instrum.
83
,
043705
(
2012
).
105.
J. F. M.
van Rens
,
W.
Verhoeven
,
J. G. H.
Franssen
,
A. C.
Lassise
,
X. F. D.
Stragier
,
E. R.
Kieft
,
P. H. A.
Mutsaers
, and
O. J.
Luiten
, “
Theory and particle tracking simulations of a resonant radiofrequency deflection cavity in TM110 mode for ultrafast electron microscopy
,”
Ultramicroscopy
184
,
77
89
(
2018
).
106.
W.
Verhoeven
,
J. F. M.
van Rens
,
E. R.
Kieft
,
P. H. A.
Mutsaers
, and
O. J.
Luiten
, “
High quality ultrafast transmission electron microscopy using resonant microwave cavities
,”
Ultramicroscopy
188
,
85
89
(
2018
).
107.
J. F. M.
van Rens
,
W.
Verhoeven
,
E. R.
Kieft
,
P. H. A.
Mutsaers
, and
O. J.
Luiten
, “
Dual mode microwave deflection cavities for ultrafast electron microscopy
,”
Appl. Phys. Lett.
113
,
163104
(
2018
).
108.
W. F.
Toonen
,
X. F. D.
Stragier
,
P. H. A.
Mutsaers
, and
O. J.
Luiten
, “
Gigahertz repetition rate thermionic electron gun concept
,”
Phys. Rev. Accel. Beams
22
,
123401
(
2019
).
109.
W.
Verhoeven
,
J. F. M.
van Rens
,
A. H.
Kemper
,
E. H.
Rietman
,
H. A.
van Doorn
,
I.
Koole
,
E. R.
Kieft
,
P. H. A.
Mutsaers
, and
O. J.
Luiten
, “
Design and characterization of dielectric filled TM110 microwave cavities for ultrafast electron microscopy
,”
Rev. Sci. Instrum.
90
,
083703
(
2019
).
110.
C.
Jing
,
Y.
Zhu
,
A.
Liu
,
K.
Schliep
,
X.
Fu
,
Y.
Zhao
,
E.
Montgomery
,
W.
Rush
,
A.
Kanareykin
,
M.
Katz
, and
J.
Lau
, “
Tunable electron beam pulser for picoseconds stroboscopic microscopy in transmission electron microscopes
,”
Ultramicroscopy
207
,
112829
(
2019
).
111.
J. W.
Lau
,
K. B.
Schliep
,
M. B.
Katz
,
V. J.
Gokhale
,
J. J.
Gorman
,
C.
Jing
,
A.
Liu
,
Y.
Zhao
,
E.
Montgomery
,
H.
Choe
,
W.
Rush
,
A.
Kanareykin
,
X.
Fu
, and
Y.
Zhu
, “
Laser-free GHz stroboscopic transmission electron microscope: Components, system integration, and practical considerations for pump–probe measurements
,”
Rev. Sci. Instrum.
91
,
021301
(
2020
).
112.
J. J.
Yao
and
N. C.
Macdonald
, “
Time-resolved scanning electron-microscopy analysis of nanodynamical structures
,”
Scanning Microsc.
6
,
937
942
(
1992
).
113.
I.
Ogo
and
N. C.
MacDonald
, “
Application of time‐resolved scanning electron microscopy to the analysis of the motion of micromechanical structures
,”
J. Vac. Sci. Technol. B
14
,
1630
1634
(
1996
).
114.
P.
May
,
J. M.
Halbout
, and
G.
Chiu
, “
Picosecond photoelectron scanning electron microscope for noncontact testing of integrated circuits
,”
Appl. Phys. Lett.
51
,
145
147
(
1987
).
115.
P. G.
May
,
J.-M.
Halbout
, and
G. L.-T.
Chiu
, “
Noncontact high-speed waveform measurements with the picosecond photoelectron scanning electron microscope
,”
IEEE J. Quantum Electron.
24
,
234
239
(
1988
).
116.
G.
Chiu
,
J. M.
Halbout
, and
P.
May
, “
High‐speed electron beam testing
,”
J. Vac. Sci. Technol. B
6
,
1814
1819
(
1988
).
117.
J.-M.
Halbout
,
P.
May
, and
G.
Chiu
, “
Ultrashort electron-pulse probing of integrated circuits
,”
J. Mod. Opt.
35
,
1995
2005
(
1988
).
118.
P.
May
,
J.-M.
Halbout
,
C. T.
Chuang
, and
G. P.
Li
, “
Waveform measurements in high-speed silicon bipolar circuits using a picosecond photoelectron scanning electron microscope
,”
IEEE Trans. Electron Devices
35
,
1126
1129
(
1988
).
119.
Y.
Pastol
,
J.-M.
Halbout
,
P.
May
, and
G.
Chiu
, “
Electron beam testing of integrated circuits using a picosecond photoelectron scanning electron microscope
,”
Scanning Microsc.
3
,
443
447
(
1989
).
120.
P.
May
,
Y.
Pastol
,
J.-M.
Halbout
, and
G.
Chiu
, “
Picosecond photoelectron microscope for high-speed testing of integrated circuits
,”
IBM J. Res. Dev.
34
,
204
214
(
1990
).
121.
A.
Adhikari
,
J. K.
Eliason
,
J.
Sun
,
R.
Bose
,
D. J.
Flannigan
, and
O. F.
Mohammed
, “
Four-dimensional ultrafast electron microscopy: Insights into an emerging technique
,”
ACS Appl. Mater. Interfaces
9
,
3
16
(
2017
).
122.
E.
Najafi
,
B.
Liao
,
T.
Scarborough
, and
A.
Zewail
, “
Imaging surface acoustic wave dynamics in semiconducting polymers by scanning ultrafast electron microscopy
,”
Ultramicroscopy
184
,
46
50
(
2018
).
123.
M.
Zani
,
V.
Sala
,
G.
Irde
,
S. M.
Pietralunga
,
C.
Manzoni
,
G.
Cerullo
,
G.
Lanzani
, and
A.
Tagliaferri
, “
Charge dynamics in aluminum oxide thin film studied by ultrafast scanning electron microscopy
,”
Ultramicroscopy
187
,
93
97
(
2018
).
124.
D.-S.
Yang
,
B.
Liao
, and
O. F.
Mohammed
, “
Scanning ultrafast electron microscopy: Four-dimensional imaging of materials dynamics in space and time
,”
MRS Bull.
43
,
491
496
(
2018
).
125.
M. W. H.
Garming
,
M.
Bolhuis
,
S.
Conesa-Boj
,
P.
Kruit
, and
J. P.
Hoogenboom
, “
Lock-in ultrafast electron microscopy simultaneously visualizes carrier recombination and interface-mediated trapping
,”
J. Phys. Chem. Lett.
11
,
8880
8886
(
2020
).
126.
J.
Zhao
,
R.
Nughays
,
O. M.
Bakr
, and
O. F.
Mohammed
, “
Access to ultrafast surface and interface carrier dynamics simultaneously in space and time
,”
J. Phys. Chem. C
125
,
14495
14516
(
2021
).
127.
S. R.
Ellis
,
N. C.
Bartelt
,
F.
Léonard
,
K. C.
Celio
,
E. J.
Fuller
,
D. R.
Hughart
,
D.
Garland
,
M. J.
Marinella
,
J. R.
Michael
,
D. W.
Chandler
,
B.
Liao
, and
A. A.
Talin
, “
Scanning ultrafast electron microscopy reveals photovoltage dynamics at a deeply buried p-Si/SiO2 interface
,”
Phys. Rev. B
104
,
L161303
(
2021
).
128.
R.
Bormann
,
S.
Strauch
,
S.
Schäfer
, and
C.
Ropers
, “
An ultrafast electron microscope gun driven by two-photon photoemission from a nanotip cathode
,”
J. Appl. Phys.
118
,
173105
(
2015
).
129.
W. A.
Curtis
and
D. J.
Flannigan
, “
Toward Å–fs–meV resolution in electron microscopy: Systematic simulation of the temporal spread of single-electron packets
,”
Phys. Chem. Chem. Phys.
23
,
23544
23553
(
2021
).
130.
W. A.
Curtis
,
S. A.
Willis
, and
D. J.
Flannigan
, “
Single-photoelectron collection efficiency in 4D ultrafast electron microscopy
,”
Phys. Chem. Chem. Phys.
24
,
14044
14054
(
2022
).
131.
D. A.
Plemmons
and
D. J.
Flannigan
, “
Ultrafast electron microscopy: Instrument response from the single-electron to high bunch-charge regimes
,”
Chem. Phys. Lett.
683
,
186
192
(
2017
).
132.
A.
Gahlmann
,
S.
Tae Park
, and
A. H.
Zewail
, “
Ultrashort electron pulses for diffraction, crystallography and microscopy: Theoretical and experimental resolutions
,”
Phys. Chem. Chem. Phys.
10
,
2894
2909
(
2008
).
133.
M. S.
Grinolds
,
V. A.
Lobastov
,
J.
Weissenrieder
, and
A. H.
Zewail
, “
Four-dimensional ultrafast electron microscopy of phase transitions
,”
Proc. Natl. Acad. Sci. U. S. A.
103
,
18427
18431
(
2006
).
134.
V. A.
Lobastov
,
J.
Weissenrieder
,
J.
Tang
, and
A. H.
Zewail
, “
Ultrafast electron microscopy (UEM): Four-dimensional imaging and diffraction of nanostructures during phase transitions
,”
Nano Lett.
7
,
2552
2558
(
2007
).
135.
S. J.
Clark
and
J.
Robertson
, “
Energy levels of oxygen vacancies in BiFeO3 by screened exchange
,”
Appl. Phys. Lett.
94
,
022902
(
2009
).
136.
Y.
Yuan
,
Z.
Xiao
,
B.
Yang
, and
J.
Huang
, “
Arising applications of ferroelectric materials in photovoltaic devices
,”
J. Mater. Chem. A
2
,
6027
6041
(
2014
).
137.
M. R.
Morris
,
S. R.
Pendlebury
,
J.
Hong
,
S.
Dunn
, and
J. R.
Durrant
, “
Effect of internal electric fields on charge carrier dynamics in a ferroelectric material for solar energy conversion
,”
Adv. Mater.
28
,
7123
7128
(
2016
).
138.
R.
Marchand
,
R.
Šachl
,
M.
Kalbáč
,
M.
Hof
,
R.
Tromp
,
M.
Amaro
,
S. J.
van der Molen
, and
T.
Juffmann
, “
Optical near-field electron microscopy
,”
Phys. Rev. Appl.
16
,
014008
(
2021
).
139.
G. M.
Caruso
,
F.
Houdellier
,
P.
Abeilhou
, and
A.
Arbouet
, “
Development of an ultrafast electron source based on a cold-field emission gun for ultrafast coherent TEM
,”
Appl. Phys. Lett.
111
,
023101
(
2017
).
140.
A. H.
Zewail
, “
The fog that was not
,”
Nature
412
,
279
(
2001
).
141.
H.
Liu
,
I. M.
Klein
,
J. M.
Michelsen
, and
S. K.
Cushing
, “
Element-specific electronic and structural dynamics using transient XUV and soft X-ray spectroscopy
,”
Chem
7
,
2569
2584
(
2021
).
142.
B.
Buades
,
A.
Picón
,
E.
Berger
,
I.
León
,
N.
Di Palo
,
S. L.
Cousin
,
C.
Cocchi
,
E.
Pellegrin
,
J. H.
Martin
,
S.
Mañas-Valero
,
E.
Coronado
,
T.
Danz
,
C.
Draxl
,
M.
Uemoto
,
K.
Yabana
,
M.
Schultze
,
S.
Wall
,
M.
Zürch
, and
J.
Biegert
, “
Attosecond state-resolved carrier motion in quantum materials probed by soft x-ray XANES
,”
Appl. Phys. Rev.
8
,
011408
(
2021
).