Direct evidence of low work function on SrVO₃ cathode using thermionic electron emission microscopy and high-field ultraviolet photoemission spectroscopy Open Access

Electron emission cathodes are key components of various high frequency, high power vacuum electronic devices such as traveling wave tubes and klystrons, and lower power device applications such as electron microscopes, ion thrusters for satellites, and thermionic energy converters for concentrated sunlight and industrial waste heat.^1–4 The electron emission from the cathode is largely determined by which surface terminations and orientations are present on the emitter, and their respective work function values. A stable, low work function surface is desired to provide efficient electron emission current during cathode operation, where emission may be controlled by incident photons (photoemission), high temperature (thermionic emission), electric field (field emission), or some combination.^5–8 Over the past century, significant research efforts have been devoted to discovering and engineering new low work function cathode materials and reducing the work function of existing cathode materials.^9–17 

The perovskite oxide SrVO₃ is a promising next-generation electron emission cathode due to its high electrical conductivity, ease of bulk synthesis, and low predicted work function.^18–21 The cubic SrVO₃ perovskite structure (space group Pm-3m, No. 221) consists of alternating polar SrO and VO₂ layers along the crystallographic [001] direction, and alternating SrVO and O layers along the [110] direction. According to density functional theory (DFT) calculations, the SrO-terminated (001) surface and SrVO-terminated (110), equivalent to (011), surface have low calculated work functions of just 1.9 and 2.3 eV, respectively. These low work function values are enabled by the presence of large intrinsic surface dipoles.¹⁸ Experimental studies using thermionic emission current measurements of bulk polycrystalline SrVO₃ have confirmed that SrVO₃ has a reproducible effective work function (defined and discussed more below) of 2.7–2.8 eV. In a few cases, these experiments also revealed a quite low effective work function of ∼2.3 eV from the bulk polycrystalline SrVO₃ cathode surface,¹⁹ suggesting potentially very low work functions can be obtained in this material. Furthermore, in a recent study, the bulk polycrystalline SrVO₃ cathode demonstrated stable emission during cyclic heating and cooling, and drift-free steady electron emission during 1 h of a continuous thermionic emission test,²² further demonstrating the promising potential of the SrVO₃ cathode.

It is important to note that these previous experimental measurements of thermionic emission from the heterogeneous emitting surface of polycrystalline SrVO₃ were conducted using a relatively low applied electric field. The use of a low applied electric field produces measurements of “effective” work function, that is, a complex average of many surface terminations, as opposed to directly revealing the lowest local work function. The surface electronic physics producing an effective work function under low applied fields is termed the patch field effect.¹⁴ According to the patch field effect, the higher work function patches surrounding the low work function patch pull the emitted electrons back toward the surface, resulting in a higher energy barrier for electron emission from the low work function surface patches than expected based on their local work function values. This produces an effective work function for the heterogeneous cathode surface that is higher than the lowest local work function surface present. Recent studies have unveiled the role of patch fields in physical models of cathode surfaces as well as actual surfaces from commercial cathode specimens.^23–26 Previous experiments reported the effective thermionic work function of bulk polycrystalline SrVO₃ cathode surface measured with a low applied electric field of ∼1 × 10⁶ V/m. For these surfaces, patch field analysis suggested that a field of at least several times of 10⁶ V/m would be required to cancel the patch fields and reveal the lowest local work functions of SrVO₃.¹⁹ 

In this work, we studied the thermionic emission behavior and work function of (110)-SrVO₃ cathode surfaces using the thermionic electron emission microscopy (ThEEM) imaging technique and ultraviolet photoemission spectroscopy (UPS). The experimental study revealed the presence of other crystallographic orientations along with the main (110) orientation in the crystalline specimen, prompting us to refer to this specimen as a partial-(110)-SrVO₃ cathode surface. The use of ThEEM measurements is advantageous for understanding local emission on cathode surfaces because it is performed by applying a large electric field (∼10⁷ V/m) to the cathode surface. Such large electric fields can cancel the patch field effect and enable local work function measurements of isolated SrVO₃ domains. We also studied the distribution of the local work function patches on the partial-(110)-SrVO₃ cathode surface.

The work function measurement at 950 °C revealed a value as low as 2.0 eV, likely within the uncertainty of the DFT-calculated work function of 2.3 eV for the SrVO-terminated (110) plane and 1.9 eV of SrO-terminated (100) plane of cubic SrVO₃.^15,18 This 2.0 eV work function value of (110) SrVO₃ cathode is even lower than the previously observed lowest effective work function value of SrVO₃ polycrystalline cathode (∼2.3 eV). However, this finding is consistent with the possibility that some fraction of the surface of these previously examined polycrystalline SrVO₃ samples had a work function of ∼2.0 eV. The obtained work function is significantly lower than the (001) LaB₆ single crystal cathode work function of ∼2.7 eV²⁷ and is comparable to the effective work function of a B-type dispenser cathode (∼2.1 eV),²⁸ suggesting that SrVO₃ is a promising thermionic emission material.

The material examined in this study was a previously synthesized SrVO₃ crystalline specimen obtained from Kimball Physics Inc. The detailed synthesis process of the SrVO₃ crystalline specimen and its characterizations are reported in previous work,²⁹ and we provide a summary here. The SrVO₃ crystalline specimen was prepared from a polycrystalline mixture of SrVO₃ (85%) and Sr₂V₂O₇ (15%) using a laser floating zone growth technique. At first, the isostatically pressed SrVO₃/Sr₂V₂O₇ cylindrical rod of diameter 4 mm and length 80 mm was sintered at 1200 °C for 8 h under vacuum (∼10⁻⁵ Torr). The crystalline specimen was prepared at 1 atm pressure under Ar:H₂ (95%:5%) gas flow in the TiltLDFZ furnace (Crystal Systems Inc.) equipped with 5 × 200 W GaAs lasers. During the synthesis, the top part of the seed rod placed in the molten zone was melted, and then the feed rod was brought into contact with the seed rod in the molten zone. During the growth, both the rods were counter-rotated, and the crystalline specimen was prepared by moving both the rods downward while keeping the molten zone fixed at the same position. Laue diffraction revealed a [110] direction along the length of the cylindrical crystal, as reported before.²⁹ However, using electron backscatter diffraction (EBSD) at two separate positions on the crystal surface, we observed different Kikuchi patterns, suggesting the presence of other orientations along with the [110] orientation, as discussed later. Based on the characterizations performed, including thermionic electron emission maps and EBSD, the evidence supports the conclusion that the specimen surface consists of an array of micro-patches with varying orientations, some of which are low work function facets, like (100) or (110) orientations. Henceforth, in the remainder of this paper, we refer to such a surface as a partial-(110)-SrVO₃ surface. To prepare the cathode, a part of the crystalline specimen was cut along the circular cross-section of the crystalline specimen to get the (110) surface and polished for the electron emission test.

EBSD measurements were performed using a Tescan Vega3 scanning electron microscope at Kimball Physics Inc. Field emission electron microscopy (FESEM) images were collected using a Zeiss 1530 scanning electron microscope (SEM). Surface elemental stoichiometry was studied using an energy dispersive spectroscopy (EDS) function integrated with the Zeiss 1530 FESEM instrument.

The ThEEM image and the thermionic emission measurements were performed using an Elmitec LEEM III microscope at the XPEEM/LEEM end station of the ESM beamline (21-ID) of the National Synchrotron Light Source II (NSLS-II) at Brookhaven National Laboratory (BNL). A simplified schematic of the ThEEM and work function measurement set-up is shown in Fig. 1. Before the electron emission test, the cathode was heated at 1300 °C for 1 h in ultra-high vacuum (base pressure of ∼5 × 10⁻¹⁰ Torr) to remove any surface contamination, which may have occurred during air exposure, and recover the surface from any damage caused by polishing. After heating, the cathode was cooled slowly to room temperature, and then it was reheated to a high temperature for the ThEEM measurements at 750, 800, 850, 900, and 950 °C. The work function of the cathode surface was measured at high temperatures of 900 and 950 °C by UPS using a Hg lamp (4.88 eV) as the photon source. The temperature of the cathode was measured using a thermocouple integrated into the sample holder. During ThEEM and the UPS measurements, an applied electric bias voltage of −20 kV was applied at the cathode, and a cathode-surface-to-anode gap (AK gap) of ∼4 mm was maintained. This applied electric field was nearly five times stronger than the one used in previously reported work function measurements of polycrystalline SrVO₃ cathodes.¹⁹ The intensity map of the spatially nonuniform emission current in the ThEEM images was recorded in arbitrary units using the ImageJ software. The average intensity at different temperatures was measured from five different emission patches’ intensity on the ThEEM images.

ThEEM is an important research tool to study the local distribution of electron emission from a heterogeneous thermionic cathode surface. It also allows us to observe the emission at high temperatures under a high applied electric field, which cancels out the patch-field effect. Figures 2(a)–2(e) show the ThEEM images of a 50 × 50 μm² region on the cathode surface collected at 750, 800, 850, 900, and 950 °C, respectively. The ThEEM images show that the measurable emission starts at around 750 °C, suggesting the presence of low work function domains on the cathode surface. At this low temperature, we observe that the electron emission is limited to a few small, low work function patches. However, as the temperature increased, more patches of progressively higher work function became activated and contributed to the overall emission. The patchy thermionic emission suggests the presence of non-uniform crystallographic grain orientations and/or terminations on the cathode surface. The dimension of the emission patches ranges from a few 100 nm to 2 μm, suggesting that the emission is not from any sharp asperity-like structure. Hence, we rule out the possibility of any field emission from the surface, as the field emission from the surface would require a very sharp, high asperity structure to have field emission at 5 kV/mm electric field. Besides, the cathode was sintered at 1300 °C before emission test, which would make it very unlikely that any structural asperities of sufficient sharpness to cause non-negligible field emission to remain on the surface as high temperature sintering will anneal away such asperities. The emission current density trend, qualitatively indicated by the brightness of the ThEEM images, was observed to increase exponentially with temperature. Figure 2(f) shows the temperature-dependent trend of the ThEEM-image-averaged intensity of the electron emitting patches. This image-averaged intensity trend is qualitatively consistent with the Richardson–Dushman equation of thermionic emission current. Only a small fraction of the cathode surface is showing emission with low work function, indicating this SrVO₃ cathode surface is a very patchy emitter, similar to observations made on previous polycrystalline specimens.^19,22

We also used the ThEEM to study the emission from a polycrystalline SrVO₃ cathode surface at a higher temperature of 1100 °C at an electric field of 2.5 kV/mm as shown in Fig. S2 of the supplementary material. ThEEM measurements of the polycrystalline sample reveal the presence of bright emission spots on the cathode surface, very similar to those observed in Fig. 2. The bright emission spots, which are the low work function regions, could originate from the low work function (110)-oriented SrVO-termination and (100)-oriented SrO-termination surfaces discovered using the DFT calculations. However, from this study, we cannot definitively identify the orientation or termination of the observed low work function regions, and doing so would require more detailed fundamental characterizations of the material.

The kinetic energy distributions of the thermally and photothermally emitted electrons were measured using a LEEM III energy analysis system at BNL. The LEEM microscope was operated at an accelerating voltage of 20 kV. The emitted electrons from the cathode passed through the hemispherical electron analyzer system as shown in Fig. 1, and generated energy-filtered images at varying kinetic energies.³⁰  Figures 3(a) and 3(b) show the kinetic energy distributions of the combined thermal- and photo-generated electrons from the cathode surface at 900 and 950 °C, respectively. The UPS valence band spectra do not show any shift in the emitted electron energy maximum position (A) as observed in Figs. 3(a) and 3(b). Meanwhile, the shifting of the secondary electron cutoff energy position (B) toward lower kinetic energy at 950 °C as compared to 900 °C suggests a reduction of the effective work function at 950 °C.^14,31

The high applied electric field in the ThEEM is expected to cancel the patch field effect from the high work function region surrounding the low work function patches (observed in Fig. 2), enabling us to measure the local work function of the low work function patches using thermionic emission.¹⁴ Using two complementary methods (photothermal emission and pure ThEEM), the measurements consistently show a decreased work function at 950 °C as compared to the work function measured at 900 °C. The fitting of the pure thermionic emission data shows a 0.15 eV work function reduction as the temperature increased from 900 °C (2.15 ± 0.05 eV) to 950 °C (2.00 ± 0.05 eV). This 0.15 eV work function reduction includes a strong cancellation of the measurement errors, so we expect it to have a much smaller uncertainty. A temperature-dependent shifting of the work function has also been observed in other patchy emitters, including Sc₂O₃-BaO-coated-W and polycrystalline SrVO₃ cathodes.^19,30 The origin of the observed temperature trend is not clear but the change of 0.15 eV over just 50 °C seems unexpectedly large to be part of a smooth equilibrium trend with heating on a fixed surface, as simple linear extrapolation would lead to very large 0.6 eV changes over just 200 °C. We speculate instead that the surface is evolving. Specifically, the surface domains that have bulk orientations consistent with the lowest work function are different at 900 °C vs 950 °C, and reach the required surface chemistry and structure for the lowest work function values during the 950 °C testing. This could be an equilibration during the experiment, which first heats to 900 °C and then to 950 °C, or different equilibrium surface morphologies or terminations that exist at equilibrium at the two temperatures. However, understanding the origin of this work function change requires more in situ research on the surface structure–work function correlation of the SrVO₃ cathode. The measured 2.0 eV work function is comparable with the DFT-calculated 2.3 eV work function of the SrVO-terminated [110] surface and 1.9 eV work function of the SrO-terminated (100) SrVO₃ surface.¹⁸ To the best of our knowledge, this 2.0 eV effective work function at 950 °C is the lowest experimentally measured local work function reported from any bulk conductive oxide material not relying on volatile surface species or doping to produce the low work function.

A previous study confirmed the phase purity of the SrVO₃ crystal and stoichiometric elemental ratios.²⁹ Hence, the heterogeneity in the emission behavior of the SrVO₃ cathode is hypothesized to arise due to the presence of different crystallographic orientations and terminations on the emission tested cathode surface. To check the crystallinity of the SrVO₃ cathode, EBSD was performed on the polished cathode surface. The EBSD study, as shown in Figs. 4(a) and 4(b), reveals different patterns at two different positions on the sample surface. This result suggests the emitting surface has additional crystallographic domains other than the dominant (110) orientation. This heterogeneity is consistent with the ThEEM results in Fig. 2 showing heterogeneous emission, consistent with some domains on the cathode surface having higher work function compared to the initially activated 2.0 eV low work function patches.

To better understand the origin of the emission heterogeneity, we studied the elemental distribution on the cathode surface using EDS elemental mapping, as shown in Fig. 5. The EDS mapping shows a uniform distribution of the Sr, V, and O atoms on the polished cathode surface. These data imply that the reason for the heterogeneity in the local work function on the cathode surface, as observed in Fig. 2, is not from major variations in the surface stoichiometry. However, the heterogeneity in the local work function may arise due to the presence of different crystallographic domains and terminations, consistent with the EBSD results in Fig. 4, and only a small fraction of the cathode surface may have the (110)-oriented SrVO-termination and/or (100)-oriented SrO-termination, with the remainder of the surface comprising other surface orientations or terminations.

A key challenge for future work is to identify the crystallographic orientation and termination of the surfaces that give rise to the low work function and then seek to increase their role in the emission. If the areas of the low work function patches were increased, or these areas occupied a larger fraction of the emission surface in the form of a higher density of small, copiously emitting patches, it could provide high emission current at a reduced temperature and an applied electric field (i.e., the field required to cancel the patch effects would be reduced). Such surface engineering would provide a promising path toward realizing SrVO₃ cathodes with enhanced emission, increasing the value of SrVO₃ emitters for a range of applications.

Conducting perovskite oxide SrVO₃ is a promising thermionic cathode material due to its low effective work function. In this study, we used thermionic electron emission microscopy to study the local work function distribution on the partial-(110)-SrVO₃ cathode surface. ThEEM demonstrated the presence of low work function patches on the surface, which started emitting at around 750 °C. The presence of low work function patches on the partial-(110)-SrVO₃ surface was revealed by using a high applied electric field to cancel the patch field effect, where these local work function patches had a low work function value of just 2.00 eV, much lower than (001) LaB₆ single crystal cathodes and comparable to B-type dispenser cathodes. Overall, this study provides two important results: (1) experimental verification that a high applied surface electric field enables emission-based local measurements of low work function patches on heterogeneous surfaces, and (2) verification of the existence of patches on a heterogeneous SrVO₃ surface that have work functions comparable to the DFT-predicted values of 2.3 eV for a SrVO-terminated (110) surface and 1.9 eV for a SrO-terminated (100) surface of cubic SrVO₃, further illustrating the potential of SrVO₃ as a thermionic emission material.

The synthesis procedure of the polycrystalline SrVO3 pellet, its room temperature XRD pattern, and the thermionic electron emission microscopy (ThEEM) measurement are reported in the supplementary material.

The authors would like to thank the Wisconsin Alumni Research Foundation (WARF) for providing financial support for this work. This research used resources from the Center for Functional Nanomaterials and the National Synchrotron Light Source II, which are U.S. Department of Energy (DOE) Office of Science facilities at Brookhaven National Laboratory under Contract No. DE-SC0012704. The authors gratefully acknowledge the use of facilities and instrumentation supported by NSF through the University of Wisconsin Materials Research Science and Engineering Center (Grant No. DMR-1720415).

The authors have no conflicts to disclose.

Md Sariful Sheikh: Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Lin Lin: Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Writing – review & editing (equal). Ryan Jacobs: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Project administration (equal); Resources (equal); Supervision (equal); Validation (equal); Writing – review & editing (equal). Martin E. Kordesch: Investigation (equal); Methodology (equal); Resources (equal); Validation (equal); Visualization (equal); Writing – review & editing (equal). Jerzy T. Sadowski: Investigation (equal); Methodology (equal); Resources (equal); Supervision (equal); Validation (equal); Writing – review & editing (equal). Margaret Charpentier: Investigation (equal); Methodology (equal); Resources (equal). Dane Morgan: Conceptualization (equal); Funding acquisition (equal); Project administration (equal); Resources (equal); Supervision (equal); Validation (equal); Writing – review & editing (equal). John Booske: Conceptualization (equal); Funding acquisition (equal); Project administration (equal); Resources (equal); Supervision (equal); Validation (equal); Writing – review & editing (equal).

The raw ThEEM, UPS, EBSD, FESEM, and EDS data as reported in Fig. 2 through Fig. 5 are openly available in Figshare at https://doi.org/10.6084/m9.figshare.25202624.³³