Automated Delivery of Microfabricated Targets for Intense Laser Irradiation Experiments

Summary

A protocol is presented for automated irradiation of thin gold foils with high intensity laser pulses. The protocol includes a step-by-step description of the micromachining target fabrication process and a detailed guide for how targets are brought to the laser's focus at a rate of 0.2 Hz.

Abstract

Described is an experimental procedure that enables high-power laser irradiation of microfabricated targets. Targets are brought to the laser focus by a closed feedback loop that operates between the target manipulator and a ranging sensor. The target fabrication process is explained in detail. Representative results of MeV-level proton beams generated by irradiation of 600 nm thick gold foils at a rate of 0.2 Hz are given. The method is compared with other replenishable target systems and the prospects of increasing the shot rates to above 10 Hz are discussed.

Introduction

High-intensity laser irradiation of solid targets generates multiple forms of radiation. One of these is the emission of energetic ions with energies at the Mega electron-volt (MeV) level¹. A compact source of MeV ions has potential for many applications, such as proton fast-ignition², proton radiography³, ion radiotherapy⁴, and neutron generation⁵.

A major challenge in making laser-ion acceleration practical is the ability to position micrometer-scale targets accurately within the focus of the laser at a high rate. Few target delivery technologies were developed to answer this challenge. Most common are target systems based on micrometer-scale thick tapes. These targets are simple to replenish and may be easily positioned within the focus of the laser. Tape target has been made using VHS⁶, copper⁷, Mylar, and Kapton⁸ tapes. The tape drive system typically consists of two motorized spools for winding and unwinding and two vertical pins placed between them to keep the tape in position⁹. The accuracy in positioning the tape surface is typically less than the Rayleigh range of the focusing beam. Another type of replenishable laser target is liquid sheets¹⁰. These targets are delivered rapidly to the interaction region and introduce a very low amount of debris. This system comprises a high-pressure syringe pump continuously supplied with liquid from a reservoir. Recently, novel cryogenic hydrogen jets¹¹ were established as means to deliver ultrathin, low-debris, replenishable targets.

The main drawback of all of these replenishable target systems is the limited choice of target materials and geometries, which are dictated by mechanical requirements such as strength, viscosity, and melting temperature.

Here, a system able to bring micromachined targets to the focus of a high intensity laser at a rate of 0.2 Hz is described. Micromachining offers a wide choice of target materials in versatile geometries¹². The target positioning is performed by a closed-loop feedback between a commercial displacement sensor and a motorized manipulator.

The target delivery system was tested using a high-contrast, 20 TW laser system that delivers 25 fs-long laser pulses with 500 mJ on target. A review of the laser system’s architecture is given in Porat et al.¹³, and a technical description of the target system is given in Gershuni et al.¹⁴. This paper presents a detailed method for making and using this type of system and shows representative results of laser-ion acceleration from ultrathin gold foil targets.

The Thomson Parabola ion spectrometer (TPIS)^15,16 shown in Figure 1 was used to record the energy spectra of the emitted ions. In a TPIS, accelerated ions pass through parallel electric and magnetic fields, which places them on parabolic trajectories in the focal plane. The parabolic curvature depends on the ion’s charge-to-mass ratio, and the location along the trajectory is set by the ion’s energy.

A BAS-TR imaging plate (IP)¹⁷ positioned at the focal plane of the TPIS records the impinging ions. The IP is attached to a mechanical feedthrough to allow translation to a fresh area before each shot.

Protocol

1. Target fabrication

NOTE: Figure 2 and Figure 3 illustrate the fabrication process of freestanding gold foils.

1. Back Side
   1. Use a 250 μm thick, 100 mm diameter, high-stress silicon wafer in a <100> crystal formation, coated on both sides with silicon nitride.
   2. Clean the wafer using acetone followed by isopropanol and dry with nitrogen.  Then spin coat a layer of HMDS to form an adhesive layer following the steps outlined in Table 1.
   3. Spin-coat the wafer with an AZ1518 photoresist layer following the steps outlined in Table 2.
   4. Bake the wafer at 100 °C for 1 min, then let it cool.
   5. Photolithograph 1,000 μm x 1,000 μm square openings under vacuum, exposing the wafer in 1 cycle of 4 to 7 seconds to a 400 nm UV lamp. The wafer is exposed to an overall fluence of 40 J/cm². Use an AZ726K developer to expose the silicon nitride, and a bath of dehydrated water to stop the process.
   6. Use Reactive Ion Etcher (RIE) to remove the silicon nitride in the location of the squares.
   7. Use an N-methyl-2-pyrrolidone (NMP) bath for 20 min to remove the residual resist and photoresist, producing a replica of the mask on the silicon nitride layer. Wash the wafer under fresh water and dry with nitrogen.
   8. Sink the wafer in a 30%, 90 °C, potassium hydroxide solution to etch the silicon through the square openings. Sink the wafer for 40 min for every 50 μm of silicon that needs to be etched. Because the etch rate in the <100> plane is much higher than in others, the potassium hydroxide reaches the bottom silicon nitride layer through the silicon bulk before etching any significant depth in the silicon nitride mask.
2. Front Side
   1. For the front side, repeat steps 1.1.1–1.1.6 with a mask shaped as three concentric rings.
   2. Use RIE to remove the silicon nitride where the rings are located, followed by an NMP bath to remove resist and photoresist leftovers.
   3. Finally, to roughen the silicon rings, sink the wafer in nitric acid and in a solution of 0.02 M silver nitrate and 4 M hydrogen fluoride.
   4. On the etched side of the wafer, use a physical vapor deposition machine (PVD)¹⁸ to sputter a layer of a few hundred nanometers of gold on top of a ~10 nm thin film of adhesive titanium, nickel, or chrome. The sputtered gold layer will become the freestanding membrane target.

2. Alignment

NOTE: Figure 4 shows the target irradiation setup.

1. Bring a first arbitrarily chosen target into view under a 100x magnification microscope.
2. Point a triangulation ranging sensor (e.g., MTI/MicroTrak 3 LTS 120-20)¹⁹ to the roughened ring closest to the target, and record its displacement reading.
   NOTE: The ranging sensor model used is not intended for high-vacuum applications. Different models, like the MTI-2100 from the same vendor, are compatible with low-outgassing applications.
3. While leaving the microscope in place, move the wafer away a known distance to clear the beam path.
4. Using two folding mirrors and the off-axis parabolic mirror (OAP), align the beam in low power into the field of view of the microscope.
5. Adjust these three mirrors to correct astigmatisms in the beam. The result should be a nearly diffraction-limited focal spot.
6. Block the laser beam and bring the target back to the focus of the microscope. Validate its position using the microscope and the ranging sensor's reading.
7. Move the microscope to a position in which it will be kept safe from laser light and debris.

3. Irradiation sequence and automated target positioning

1. Implement a closed-loop feedback between the focal axis manipulator of the target and the displacement sensor reading using software. Use the recorded value from protocol step 2.2 as the setpoint. The main PID²⁰ control sequence, prepared with LabView, is shown in Figure 5.
2. Once the closed-loop positioning has reached a desired tolerance distance from the setpoint, irradiate the target with a single high-power laser pulse.
3. Translate the IP using the mechanical feedthrough to a new position.
4. Repeat the irradiation sequence with the next target brought to focus by the software.

Representative Results

This target delivery system was employed to accelerate ions from the back side of 600 nm thick gold foils. When irradiated with a normalized laser intensity of a₀ = 5.6, these ions were accelerated by the target normal sheath acceleration (TNSA) mechanism²¹. In TNSA, the lower-intensity light that preceded the main laser pulse ionized the front surface of the target foil. The ponderomotive force exerted by the main laser pulse drove hot electrons through the bulk matter. A charge separation on the back surface, induced by these electrons²², created an extreme electrostatic gradient that accelerated ion contaminants in the target-normal direction.

A time series of the target displacement along the focal axis is shown in Figure 6. The values are relative to the focal position setpoint. The green dots indicate when the target displacement was within a tolerance value of 1 μm from the setpoint; this is when a laser shot was taken.

Figure 7 shows TPIS traces from 14 consecutive irradiations of 600 nm thick gold foil targets. The energy spectrum derived from these traces is shown in Figure 8. The peak-to-peak stability of the maximum proton energy is within 10%.

Figure 1: A technical layout of the Thomson parabola ion spectrometer. Please click here to view a larger version of this figure.

Figure 2: A schematics sketch of the target wafer.
The front side, showing 300 gold foil targets ordered in three concentric rings (left). The back, showing roughened fiducial rings positioned between the target foil locations (right). Please click here to view a larger version of this figure.

Figure 3: An illustration of the wafer fabrication process. Please click here to view a larger version of this figure.

Figure 4: A schematic layout (left) and photo (right) of the interaction chamber. Please click here to view a larger version of this figure.

Figure 5: Target positioning PID LabView code (VI). Please click here to view a larger version of this figure.

Figure 6: Target displacement during a shot sequence of 20 targets. Please click here to view a larger version of this figure.

Figure 7: TPIS traces from 14 consecutive shots. The trajectories of ions and X-rays passing through the TPIS are illustrated. Please click here to view a larger version of this figure.

Figure 8: Ion energy spectra derived from the 14 traces shown in Figure 7. Please click here to view a larger version of this figure.

Figure 9: A TPIS trace recorded using a low dynamic range CCD imaging of a CsI(TI) scintillator. Please click here to view a larger version of this figure.

Step	ν [rps]	ramp [rps2]	Duration [s]
1	500	500	10
2	4000	1000	45
3	0	1000	0

Table 1: Resist spin coat steps.

Step	ν [rps]	ramp [rps2]	Duration [s]
1	500	500	10
2	4000	1000	45
3	0	1000	0

Table 2: Photoresist spin coat steps.

Discussion

With some variations, the target fabrication process described in this protocol is common (e.g., Zaffino et al.²³). Here, one unique step that is critical to the operation of automatic positioning is the addition of nanometer-scale roughening in ring-shaped areas on the back of the wafer (step 1.2.3). The purpose of this step is to increase the diffused scattering of light incident on the wafer in those areas. The ranging sensor shines a low-power laser beam on the wafer, collects the scattered light, and determines its displacement by triangulation.

The data shown above were taken at a rate of one shot per 5 s, with the rate-limiting factor being the translation time of the IP. Shown here is a preliminary result of a simple, inexpensive, online readout method that will increase the shot duty cycle. Online readouts have been traditionally made using either microchannel plates²⁴ or plastic scintillators^25,26. In the latter case, an expensive, image-intensified gated CCD was required to record the relatively low amount of scintillation light. The current system uses a simpler readout system based on a different scintillator material, Csl(Tl), which is bright enough to be recorded with an inexpensive, low dynamic range CCD. This choice of scintillator has been suggested and discussed by Pappalardo et al.²⁷.

Figure 9 shows a sample image of a TPIS trace taken with a low dynamic range CCD image of a Csl(Tl) scintillating screen. These traces were taken with a relatively large aperture, to produce a high quantity of scintillation light. Further study is required to identify the optimal settings in terms of signal-to-noise ratio and energy resolution.

The image shown in Figure 8 was acquired using a 1.6 megapixel camera. At a 10 Hz rate and 8-bit pixel depth, the data stream would amount to about 130 Mbps. This data rate is supported by either a USB3 or GigE communication interface.

The mechanical stability of any replenishable laser target delivery system may be compromised by a higher delivery rate or by the higher impact induced by higher energy laser pulses. Table 3 presents a comparison between this work and various other target delivery technologies. The performance of this system at higher shot rates and higher energy pulses will be investigated in the near future.

Reference	Target type	Materials	Thickness	Repetition Rate	Laser Energy
[6]	Tape	Mylar	15 µm	0.2 Hz	5 J
[10]	Liquid Sheet	Ethelyne Glycol	0.4 µm	1 kHz	0.011 J
[11]	Hydrogen Jet	H2	20 µm	1 Hz	600 J
This work	Micro-machined Au foil	Au	0.6 µm	0.2 Hz	0.5 J

Table 3: Comparison of different target types.

Materials

Name	Company	Catalog Number	Comments
76.2 x 127mm EFL 90° Protected Gold 100Å Off-Axis Parabolic Mirror	Edmund optics	35-535
MicroTrak 3 LTS 120-20	MTI Instruments
Ultrafast high power dielectric mirrors for 800 nm	Thorlabs