Laser-Driven Electron Lensing in Silicon Microstructures

We demonstrate a laser-driven, tunable electron lens fabricated in monolithic silicon. The lens consists of an array of silicon pillars pumped symmetrically by two 300 fs, $1.95\mu \mathrm{m}$ wavelength, nJ-class laser pulses from an optical parametric amplifier. The optical near field of the pillar structure focuses electrons in the plane perpendicular to the pillar axes. With $100\pm 10\mathrm{MV}$/m incident laser fields, the lens focal length is measured to be $50\pm 4\mu \mathrm{m}$, which corresponds to an equivalent quadrupole focusing gradient $B^{\prime }$ of $1.4\pm 0.1\mathrm{MT}/\mathrm{m}$. By varying the incident laser field strength, the lens can be tuned from a $21\pm 2\mu \mathrm{m}$ focal length ($B^{\prime }>3.3\mathrm{MT}/\mathrm{m}$) to focal lengths on the centimeter scale.

Figure 1

(a) An electron beam passes through a DLA lens with two identical laser pulses normally incident upon it. The beam is focused, travels approximately one focal length, and is filtered by an aperture of two silicon blocks with a small gap between them. (b) An SEM of the lens and aperture, viewed from above. The lens is composed of two rows of 15 pillars each. The drift length is $39.6\mu \mathrm{m}$. (c) An SEM showing the aperture structure.

Figure 2

(a) Simulated contrast plotted against relative drive laser phase $\theta$ and electric field $E_{\text{inc}}$. The simulation is a symplectic 2D particle-tracking scheme based on Ref. [21] which applies a momentum kick once per lens period equal to the time integral of Eq. (2) over one structure period. Expected contrast is calculated by a Monte Carlo approach. (b) Contrast is measured as a function of $\theta$ and $E_{\text{inc}}$.

Figure 3

(a) The measured contrast as a function of $\theta$ for $E_{\text{inc}}=137\pm 13\mathrm{MV}$/m. The blue line is measured data, the black line and gray shaded area are the moving average and standard deviation, respectively. (b) The electron energy spectrum is measured simultaneously with the phase sweep in (a). Electron counts are normalized to the maximum value.

Figure 4

This figure plots the contrast and standard deviation as a function of $E_{\text{inc}}$ for $\theta =0$. The simulation curve is a cross section of Fig. 2 at $\theta =0$. Duplicate runs have been included.