Effects of nonlinear photoemission on mean transverse energy from metal photocathodes

Reducing the mean transverse energy (MTE) of electrons emitted from photocathodes will improve the performance of linear accelerator applications like x-ray free electrons lasers (XFELs) and ultrafast electron diffraction (UED) and microscopy (UEM) experiments. Metallic photocathodes are popular for these applications due to their convenience of use. However, they typically have a very low quantum efficiency and require a large laser fluence in order to extract the desired charge density for applications like XFELs and single-shot UED/UEM experiments. Recent theoretical investigations have shown that nonlinear photoemission effects of multiphoton emission and electron heating increase the MTE dramatically at these large laser fluences. In this paper, we report on measurements of nonlinear near-threshold photoemission from a graphene-coated Cu(110) photocathode at laser pulse lengths of 130 fs, 1 ps, and 10 ps. We extrapolate our measured data to find the ideal irradiating photon energy that results in a minimum MTE from such metallic photocathodes for charge densities relevant to photoinjectors and specify quantum efficiency requirements to obtain the thermally limited MTE at these charge densities.

Figure 1

The schematic of an in-house pulse stretcher that was built using a pair of transmission diffraction gratings. The schematic shows the pulse stretcher in relation to the optical parametric amplifier (OPA) output. By fixing one diffraction grating and moving the other, we can set the correct distance according to Eq. (4) and stretch the pulse to lengths ranging from 1 and 10 ps for photon energies between 3.70 and 4.77 eV.

Figure 2

Measured MTE as a function of excess energy for a 130-fs pulse length. The line joining the points is a guide for the eye. The laser fluence was kept small enough to ensure linear photoemission at 0 and positive excess energies. At large excess energies, the emission process is dominated by single-photon emission and the MTE is roughly equal to $E_{\text{excess}}/3$. At negative excess energies, the MTE increases significantly due to nonlinear photoemission effects. The error in the measurement was estimated to be 10%.

Figure 3

Measured data points at $-0.06\mathrm{eV}$ excess energy for 130 fs, 1 ps, and 10 ps pulse lengths as a function of laser fluence. At this excess energy, a significant increase in MTE is observed for the shorter pulse lengths due to nonlinear effects as the laser fluence increases. At the longer 10 ps pulse length, the increase in MTE is only a few meV for the highest laser fluence. The solid lines are obtained from the extrapolation technique described in Sec. 4. From this plot, we see that our extrapolation technique provides a lower limit on the MTE.

Figure 4

Linear quantum efficiency as a function of excess energy for 130 fs. The current was measured from electron counts on the detector and therefore represents only a fraction of the actual current emitted. Hence the measured QE is some unknown factor on the order of unity lower than the actual QE.

Figure 5

Total energy distribution at 325 nm ($-0.06\mathrm{eV}$ excess energy) for 130 fs, 1 ps, and 10 ps. We see that multiphoton emission is significantly reduced and the number of electrons emitted from a single photon is increased for the 10-ps data. This shows that multiphoton can be significantly mitigated by operating at 10 ps pulse lengths.

Figure 6

The $k_x=0$ slice of the energy integrated transverse momentum distribution on a log scale for 130 fs, 1 ps, and 10 ps under the same photoemission conditions as shown in Fig. 5. The small wiggles are an artifact of a mesh grid placed at the entrance of the detector.

Figure 7

MTE vs fluence curves for 130 fs, 1 ps, and 10 ps pulse lengths calculated from Eq. (6). Although the curves plateau at higher fluences, this is an artifact of not including beyond second order photoemission in the extrapolation technique. In practice, we would expect the curves to continue rising with fluence due to third and higher order photoemission effects. The curves are identical to those presented in Fig. 3, except at higher laser fluences.

Figure 8

Minimum contribution to the MTE from nonlinear effects as calculated from Eq. (8) for charge densities ranging from ${10}^{-1}$ and ${10}^4\mathrm{pC}/{\mathrm{mm}}^2$ for (a) 130 fs, (b) 1 ps, and (c) 10 ps pulse widths. The calculation was performed for the measured QE data as well as for higher QEs. It shows that a nearly full mitigation of nonlinear effects can be achieved by using photocathodes with QEs on the order of ${10}^{-4}$ at the threshold.