A diagnostic tool for precise alignment of targets in laser-matter interactions based on confocal microscopy is presented. This device permits precision alignment of targets within the Rayleigh range of tight focusing geometries for a wide variety of target surface morphologies. This confocal high-intensity positioner achieves micron-scale target alignment by selectively accepting light reflected from a narrow range of target focal planes. Additionally, the design of the device is such that its footprint and sensitivity can be tuned for the desired chamber and experiment. The device has been demonstrated to position targets repeatably within the Rayleigh range of the Scarlet laser system at The Ohio State University, where use of the device has provided a marked increase in ion yield and maximum energy.

The push for higher intensity laser systems is driven by the prospect of unlocking new experimental capabilities and applications. Intensities in excess of 1018 W/cm2 drive relativistic electron interactions that can be used for fusion experiments,1 gamma ray production,2 high harmonic generation,3 and ion acceleration4,5 for radiography,6 among others. Intensities greater than 1021 W/cm2 will permit experiments on laboratory astrophysics and positron production,7 as well as the promise of more efficient ion acceleration mechanisms.8,9

Together with higher energy per pulse and shorter pulse lengths, the development of ultra-high intensities requires tight focusing of the laser. Faster focusing necessitates increased target alignment precision since the depth of field scales as the square of the f-number. Rayleigh ranges on the order of a few microns are common for petawatt-class Ti:Sapphire ultrashort pulse laser systems, so targets need to be repeatably aligned with this level of accuracy. Additionally, spatio-temporal coupling effects may be present in ultrashort focusing beams10,11 and may be engineered to control ion acceleration12 or attosecond pulse generation from high harmonics.13,14 These effects are strongly coupled to position within the focus, placing an additional constraint on target positioning. Finally, experiments using structured targets may have additional constraints on the target position to achieve optimal coupling of the laser into the target structures.15,16

A number of alignment techniques are available for precise target alignment, depending on the nature of the target. For example, optically smooth targets can be aligned using a specularly reflected beam,17 where the centroid location of an image of the reflected beam changes with target position along the laser axis, or in the z direction. This technique loses precision for optically rough targets as a poor reflected mode from a rough target complicates the centroid calculation. Very rough targets can be aligned by placing surface features in focus on the same image system used to view the laser focus. However, the absence of sufficiently sharp features such as scratches or pits prevents using this technique on smooth targets.

Another technique used for the thin metallic foil targets common to High Energy Density (HED) experiments relies on shadowgraphy. Here, the target is aligned by backlighting the focal plane and bringing the target edge into focus on a focal spot camera, after which the target is moved transversely to place the laser focus on the bulk of the target. Though widely used, this technique suffers from a number of systemic sources of alignment error as demonstrated in Fig. 1. Since the foils are opaque, an edge must be used for shadowgraphy. These targets are often curled or crumpled at the edges, such that the edge used for target alignment is not at the same z position as the bulk target. When the target is moved such that the laser will fire closer to the target center to prevent edge effects, this motion cannot be decoupled from the laser propagation axis entirely, and even then the surface quality of the resulting target spot is unlikely to be known. The net error from these three effects—edge curling, target motion back to the center, and center surface quality—often results in at least tens of microns inaccuracy along the direction of laser propagation, generating orders of magnitude fluctuation in intensity on subsequent shots.

FIG. 1.

Traditional alignment techniques such as shadowgraphy are limited by target morphology. For example (a) in shadowgraphy alignment, a target edge is used for alignment. (b) After the edge is aligned, the target must undergo a small translation to allow shooting on the bulk of the target. (c) The target is likely not flat on the scale of the necessary translations during target alignment, and may have curled edges from the target cutting process.

FIG. 1.

Traditional alignment techniques such as shadowgraphy are limited by target morphology. For example (a) in shadowgraphy alignment, a target edge is used for alignment. (b) After the edge is aligned, the target must undergo a small translation to allow shooting on the bulk of the target. (c) The target is likely not flat on the scale of the necessary translations during target alignment, and may have curled edges from the target cutting process.

Close modal

To solve the problems inherent to these techniques, we have developed a target alignment device that allows micron-scale accuracy along the laser propagation direction for a wide range of target morphologies. The device is based on confocal microscopy, using alignment light normally incident onto the target surface to judge z position based on the amount of light that travels through the setup.18 In this paper, we describe the design, calibration, and practical use of this Confocal High Intensity Positioner (CHIP) for fine target alignment. In Sec. II, we present the optical configuration of the CHIP as implemented on the Scarlet laser system at The Ohio State University, including theoretical estimates for the diagnostic response as a function of z position. Section III covers the hand-off alignment procedure that ensures that the sensitive range of the CHIP is coincident with the best focal position of the laser. Finally, in Sec. IV, we present experimental evidence that use of this diagnostic leads to more reliable alignment of targets to the best laser intensity, improving the maximum ion energy and ion yield for shots on the Scarlet laser at The Ohio State University.

The CHIP operates on the concept of confocal microscopy, in which the light gathered by the detector is preferentially collected from a narrow range of focal depths.18 This is accomplished by placing a narrow collection aperture at the image plane of a microscope system, corresponding to the correct focal plane of the target. Light originating from this correct focal plane is tightly focused through the aperture and efficiently collected into a power detector such as an amplified photodiode. However, as shown in Fig. 2, light that does not originate from this optimal plane is defocused as it hits the aperture such that much of the light is rejected from the system. Thus, as the target is moved out of the correct plane, the collected beam becomes increasingly defocused on this aperture, resulting in a strong dependence of collected light power on target position.

FIG. 2.

Simplified schematic of a confocal microscope. A collimated laser is inserted via a 50% beam splitter into a microscope objective, creating a tight focus onto the target plane. The light, reflects off of the target, is re-collected through the objective and focused through a lens onto a small output aperture. If the target is not in the correct z position, the light is defocused on the collection aperture, reducing the collected power.

FIG. 2.

Simplified schematic of a confocal microscope. A collimated laser is inserted via a 50% beam splitter into a microscope objective, creating a tight focus onto the target plane. The light, reflects off of the target, is re-collected through the objective and focused through a lens onto a small output aperture. If the target is not in the correct z position, the light is defocused on the collection aperture, reducing the collected power.

Close modal

The optical configuration of the CHIP consists of a long working distance, high numerical aperture, infinity-corrected microscope objective, a 50% pick-off mirror, an achromatic doublet lens to refocus the light, and a small-core multimode fiber optic, as shown in Fig. 2. A collimated alignment laser is input from the side of the CHIP via the 50% partially reflective pick-off mirror. This laser is then focused through the infinity-corrected microscope objective, allowing collimated laser input while still obtaining a well-corrected focus onto the target plane. The diameter of the laser beam is matched to the input aperture at the back of the microscope objective such that the objective operates at its maximum numerical aperture. This reduces the illuminated spot size on the target and obtains the shortest possible depth of field. The focused laser then reflects off of the target surface at normal incidence and is recollected by the microscope objective. After returning through the microscope, the laser light passes through the beam splitter and is focused through a 200 mm focal length achromatic lens as recommended by the manufacturer of the microscope objective. The light collected by the lens is focused onto the face of a small-core (10 μm) multi-mode optical fiber, which serves as the limiting aperture for the confocal microscope. The laser power collected through the fiber is then coupled through the target chamber wall via a multimode fiber feedthrough to an amplified photodiode. The core size of the collection fiber must be chosen to be slightly smaller than the minimum diffraction-limited spot size that may be imaged to the face of the fiber such that any small defocus of the target will immediately result in decreased collection by the output fiber.

For a well-aligned target, the return path through the microscope objective is identical to the forward path, and the 200 mm achromatic lens creates a tight focus onto the collection fiber. Because the longitudinal magnification of the imaging system scales the same way as the Rayleigh range, any defocus in the target plane will produce equivalent defocus at the fiber. This allows calculation of light collection efficiency in the target plane, where the numerical aperture is well known. Additionally, if the target is shifted from the optimum focus position by some distance δz, the extra path length on reflection is 2δz. Approximating the laser as a Gaussian beam, the fraction of laser power collected into the optical fiber can be derived by integrating the Gaussian focal intensity I(r, z) produced by a focal shift of 2δz up to the radius rap of the optical fiber core and dividing by the total power incident onto the fiber Ptot. As the calculation is performed in the object plane, the fiber core size must be corrected by dividing by the magnification m, such that the collection efficiency η can be expressed as

η = P collect P tot = 1 P tot 0 2 π 0 r ap m I ( r , 2 δ z ) r d r d ϕ = 1 exp 8 ( r ap / m ) 2 ( f / # ) 2 π 2 16 λ 2 ( f / # ) 4 + ( 2 δ z ) 2 π 2 ,
(1)

where λ is the wavelength of the laser in the confocal positioner and f/# is the f-number of the focus onto the target plane. For short wavelengths and tight focus, the collected power falls off very quickly with defocus δz such that maximizing the collected power optimizes the z position of the target within a very narrow range. For a confocal system operating at 532 nm with an f-number of 2, a magnification of 10 × and a collection aperture of rap = 5 μm as implemented at OSU, the theoretical sensitivity of the confocal position sensor is well below the Rayleigh range of an 800 nm laser at an f/2.2 focus (typical focus at OSU’s Scarlet laser), as shown in Fig. 3.

FIG. 3.

CHIP sensitivity curves: The ideal theoretical signal output for a Gaussian beam to the CHIP (as implemented on the Scarlet laser at OSU) is plotted in blue (solid line). The theoretical signal for a flat-top input beam with moderate amounts of astigmatism (0.25 waves P-V) and spherical aberration (0.3 waves P-V) is plotted in red (dotted-dashed line) and agrees well with the measured sensitivity curve with a flat-top diode laser input, as plotted in black dots. For comparison, the Rayleigh range zR for a short pulse beam at a wavelength of 800 nm focusing at an f-number of 2.2 (typical of the Scarlet laser at OSU) is denoted on the x-axis. Even with an imperfect input laser, optimization of the CHIP output by translating the target results in a target position well within a Rayleigh range of best focus.

FIG. 3.

CHIP sensitivity curves: The ideal theoretical signal output for a Gaussian beam to the CHIP (as implemented on the Scarlet laser at OSU) is plotted in blue (solid line). The theoretical signal for a flat-top input beam with moderate amounts of astigmatism (0.25 waves P-V) and spherical aberration (0.3 waves P-V) is plotted in red (dotted-dashed line) and agrees well with the measured sensitivity curve with a flat-top diode laser input, as plotted in black dots. For comparison, the Rayleigh range zR for a short pulse beam at a wavelength of 800 nm focusing at an f-number of 2.2 (typical of the Scarlet laser at OSU) is denoted on the x-axis. Even with an imperfect input laser, optimization of the CHIP output by translating the target results in a target position well within a Rayleigh range of best focus.

Close modal

To ensure that the measured sensitivity curve of the CHIP is as close as possible to the theoretical sensitivity for a Gaussian input beam as described in Eq. (1), it is critical to use an input laser with a very high quality mode (M2 beam propagation parameter as close as possible to 1). For example, the use of an alignment laser with a flat-top intensity distribution in testing of the CHIP at OSU produces a significantly wider sensitivity curve compared to theory, as indicated in Fig. 3.

For such a non-Gaussian input mode, the sensitivity curve can be numerically modeled by producing a 2D amplitude and phase mask of the input electric field, accounting for aberrations both in the input beam, and any aberrations inherent to the focusing optics in the CHIP itself. The resulting mode in the focal plane can then be calculated using Fourier optics by taking the two dimensional Fourier transform of the input amplitude and phase functions. To study the mode as a function of defocus, the input beam mode can be modulated by an additional parabolic phase before performing the Fourier transform, providing the defocused mode near the focal plane. As with the analytical result in Eq. (1), the resulting intensity is integrated within the radius of the collection fiber optic to determine the collected power at the output of the CHIP as a function of defocus. As an example, we approximate the test beam as implemented at OSU as a flat-top mode possessing small amounts of astigmatism (inherent to most diode laser outputs) and spherical aberration (resulting from residual uncorrected aberration in the CHIP itself). The predicted response curve in this case is calculated as described above, and plotted in Fig. 3, in very good agreement with the measured sensitivity curve. It should be noted that the majority of the broadening of the sensitivity curve comes from the flat-top intensity profile of the input beam, which increases the M2 of the beam, and increases the effective Rayleigh range of the focus. The small level of spherical aberration and astigmatism is responsible for the slight asymmetry in the measured response curve.

For targets with roughened surfaces, such as the sub-micron thickness foils often used on high intensity laser experiments, the sensitivity curve will depart from the ideal cases as described above. In the case that the target is very close to the correct focal plane, the portion of the target illuminated by the CHIP is very small (on the scale of a few microns), and the target will likely be locally flat within the illumination spot size. The main effect at small defocus is that the local target normal may not be collinear with the axis of the CHIP. However, the microscope objective implemented on the CHIP has a large numerical aperture such that most of the light is still collected from the target, and the sensitivity curve will resemble that from a flat reflective target. As the roughened target is defocused, the illumination spot size on the target increases, and the target ceases to be flat over the illuminated spot size. This results in an aberrated beam being reflected from the target, producing laser “speckle” which may provide constructive or destructive interference within the aperture of the collection fiber. This laser speckle will result in local maxima in the sensitivity curve, corresponding to when a bright point in the laser speckle is collected by the fiber. However, these local maxima are still much lower in intensity than the peak signal observed at zero defocus as the power is still spread over a defocused beam. Thus, the maximum signal observed on the CHIP will still correspond to the zero-defocus position.

The microscope objective, beam splitter, lens, and fiber that make up the CHIP are assembled into an optical cage system to rigidly combine the elements into a single unit, as shown in Fig. 4. Due to the close proximity of the microscope objective to the target, the CHIP is likely to be blocking some of the solid angle of the focusing short pulse beam, requiring the CHIP to be moved out of the target plane before a laser shot. Additionally, the exploding target is likely to coat the microscope objective with target debris or damage the front of the microscope unless the CHIP is moved before the laser shot. To provide the necessary motion away from the target, the optical cage system is supported on a two-inch travel motorized vertical translation stage. Additionally, a two-axis motorized horizontal stage allows for fine positioning under vacuum to ensure that the CHIP focus and short-pulse focus are coincident. The entire structure allows for the CHIP to be easily repositioned in the target chamber with minimal realignment.

FIG. 4.

Mechanical drawing of the CHIP. (a) The optical components are assembled onto a 30 mm cage system and (b) supported on a vertical lift stage, allowing the CHIP to be lowered out of the target plane before a shot. (c) The assembly is mounted on a 2-axis motorized translation stage to allow fine position control under vacuum.

FIG. 4.

Mechanical drawing of the CHIP. (a) The optical components are assembled onto a 30 mm cage system and (b) supported on a vertical lift stage, allowing the CHIP to be lowered out of the target plane before a shot. (c) The assembly is mounted on a 2-axis motorized translation stage to allow fine position control under vacuum.

Close modal

The requirement to move the CHIP out of the focal plane to clear the incoming short-pulse beam and prevent damage makes this alignment technique most advantageous for experiments in the single shot regime. However, for solid targets’ shot at sufficient angle of incidence that the CHIP will clear the incoming short pulse beam, the device can be adapted to operate at high repetition rate. The addition of a thin cover glass to protect the microscope objective from target debris and a dichroic filter to prevent short pulse light from reaching the collection aperture will allow the CHIP to remain in position during shots, further enabling high repetition rates.

The design is scalable, allowing for an inexpensive and compact system that may be customized to meet the needs of the laser system or target chamber on which it is used. For example, if a large standoff distance is not required, the long-working distance objective can be replaced with a more compact one, allowing both for beam size reduction and a shorter tube lens to focus onto the collection fiber. Alternately, higher sensitivity can be achieved by replacing the 10 × objective with a higher numerical aperture objective, further decreasing the depth of field of the system.

Additionally, the signal can be split via an additional 50% beam splitter into two collection fibers, where one of the fibers is intentionally defocused from the optimal focal position. Thus, when the signal decreases away from the maximum on the primary fiber, the signal on the other fiber can be referenced to provide direction information about the target misalignment.

As the CHIP gives very precise alignment via a rapid drop in collected light as the target is shifted out of the focal plane of the device, it is critical that the CHIP focal point is co-focal with the short-pulse laser focal point. To accomplish this, a hand-off procedure is implemented to bring the CHIP into proper alignment before aligning targets.

A hand-off target was created using a 100 nm thick silicon nitride transmission electron microscope (TEM) support window. This hand-off target is optically smooth, which ensures that light from the CHIP maintains a clean mode upon reflection and permits easier optimization of the output signal. This target is primarily transmissive, permitting simultaneous viewing of the short-pulse focal spot using the focal spot optimization camera and the reflected beam measured by the CHIP. To ensure that the TEM window target is in the correct focal plane, one side is sparsely coated with copper nanoparticles. These nanoparticles provide surface features that may be sharply imaged using a focal spot camera to bring the hand-off target into optimum focus. The hand-off target is prepared by first mixing a few mg of Cu nanoparticles with 100 ml of water and agitating these in an ultrasonic bath. A small droplet of the resulting nanoparticle suspension is placed onto the TEM window using an eyedropper and allowed to evaporate, leaving few-micron clusters of nanoparticles dispersed across the surface. One such cluster of these nanoparticles is brought into focus on the focal-spot camera by backlighting the focal plane with an alignment laser and sharply focusing the shadow produced by the nanoparticles as shown in Fig. 5. This cluster of nanoparticles is chosen to be on the scale of the imaging resolution of the focal spot camera to allow the clearest indication of best focal position. Once the nanoparticles are in optimal focus, the reflective surface of the hand-off target is known to be at the best z position within the depth of field of the focal spot imaging system. The CHIP is then raised into position via its motorized lift stage and is moved transversely to be co-focal with the short pulse on the focal spot camera using the lift stage motor and a motorized horizontal translation stage. The CHIP signal is then maximized by fine-tuning its z position and the input k-vector of light into the chip. This ensures that the sensitivity range of the CHIP is centered on the correct focal depth, and that the light is coupled into the output fiber as effectively as possible.

FIG. 5.

The CHIP is aligned by use of a thin handoff target coated in copper nanoparticles. (a) The nanoparticles are brought into focus on a focal spot imaging camera to ensure that the handoff target is in the same plane as the best short-pulse focus. The surface of the handoff target can then be used to optimize the CHIP position. (b) The handoff target must be very thin on the scale of the desired accuracy, as refraction through the handoff target creates a shift in the observed position of the nanoparticle-coated surface.

FIG. 5.

The CHIP is aligned by use of a thin handoff target coated in copper nanoparticles. (a) The nanoparticles are brought into focus on a focal spot imaging camera to ensure that the handoff target is in the same plane as the best short-pulse focus. The surface of the handoff target can then be used to optimize the CHIP position. (b) The handoff target must be very thin on the scale of the desired accuracy, as refraction through the handoff target creates a shift in the observed position of the nanoparticle-coated surface.

Close modal

Due to the necessity of performing this procedure to bring the CHIP to the focal spot of the short-pulse laser, the accuracy of target alignment is determined by the depth of field of the imaging system used to view the short-pulse focal spot and align the microparticle surface features to the focal plane. However, after the hand-off procedure has been completed, the CHIP provides precise alignment, giving sub-Rayleigh range repeatability to the same position as previous target alignments.

The CHIP was built for and tested on the Scarlet laser facility at The Ohio State University. Scarlet is a 400 TW, 30 fs, one shot per minute Ti:Sapphire laser system operating at mid 1021 W/cm2. The final focusing optic is an f/2.2 off-axis parabola, which generates a <5 μm FWHM focal spot on target.

To test the CHIP, two sets of shots were taken on 2 μm Cu foils placed at 22.5° angle of incidence. For the first series of shots, these targets were aligned with the shadowgraphy method described in Sec. I, where the back-lit target shadow and filtered laser focus were both placed exactly at the focal plane of the focal spot alignment camera situated directly behind target chamber center. As previously discussed, this alignment is limited by the foil morphology and target positioner translation axis coupling, so that the discrepancy between best laser focus and target position was on the order of tens of microns. Subsequent shots were taken on the same type of targets aligned using the CHIP via the hand-off procedure described in Sec. III. The primary diagnostic for these shots was a Thomson parabola spectrometer19 located in the rear target normal direction. Figure 6 shows the ion traces from two typical shots, where the color scales have been normalized to the same values. Here, more energetic protons will lie closer to the upper left of the trace. In Fig. 6(a), the Cu target was aligned without the CHIP, resulting in 17 MeV maximum energy protons from 10.3 J of laser energy. For the second shot, this time aligned with the CHIP, a significantly lower laser energy of 5.1 J on target produced protons with more than 30 MeV. Additionally, the yield of protons is higher in the CHIP-aligned target, indicating a hotter target from more intense laser interaction due to superior focal positioning. This trend of improved experimental results has been demonstrated in other types of targets as well, including flat and rough foils, structured targets, reduced mass foils with low surface area for reflection, and transparent liquid film targets.20 

FIG. 6.

(a) Ion traces from a Thomson parabola Spectrometer for a 10 J Scarlet shot, resulting in a maximum of 17 MeV protons. The target here was 2 μm Cu foil, aligned with in situ shadowgraphy. (b) Another 2 μm Cu foil shot, this time with only 5.1 J of energy aligned with the CHIP, resulting in a maximum of 30 MeV protons.

FIG. 6.

(a) Ion traces from a Thomson parabola Spectrometer for a 10 J Scarlet shot, resulting in a maximum of 17 MeV protons. The target here was 2 μm Cu foil, aligned with in situ shadowgraphy. (b) Another 2 μm Cu foil shot, this time with only 5.1 J of energy aligned with the CHIP, resulting in a maximum of 30 MeV protons.

Close modal

The confocal high intensity positioner allows for repeatable, micron-scale target placement for ultra-high intensity laser experiments by leveraging the high axial resolution of a confocal microscope to provide a simple reference of target position. Any defocus of the target results in a shifted object plane of the imaging system such that much of the imaged light is rejected by a small-core multimode optical fiber at the output of the microscope. The CHIP relies on light collected from a small area coincident with the short-pulse laser focus such that even very rough targets are effectively flat on this scale. This allows for precise and repeatable alignment on a broad variety of targets without complications inherent to traditional alignment procedures. Since the critical elements of this confocal setup are relatively simple and inexpensive, the device can be redesigned to have increased sensitivity or to fit in a confined target chamber space. As demonstrated on the Scarlet laser at The Ohio State University, targets aligned using this device will have interactions more repeatably near maximum laser intensity, thus improving accelerated ion yield and energy with less overall laser energy on shot. Additionally, the sub-Rayleigh range accuracy of the CHIP promises highly comparable shot conditions across the entirety of an experimental run.

This work was supported by the Air Force Office of Scientific Research (AFOSR) and Defense University Research Instrumentation Program (DURIP) under Contract No. 60037764, by the National Nuclear Security Administration (NNSA) under Grant No. DE-NA0001976, and by the DARPA PULSE program through a grant from AMRDEC.

1.
J. J.
Honrubia
and
J. M.
ter Vehn
,
Plasma Phys. Controlled Fusion
51
,
014008
(
2009
).
2.
K. W. D.
Ledingham
,
I.
Spencer
,
T.
McCanny
,
R. P.
Singhal
,
M. I. K.
Santala
,
E.
Clark
,
I.
Watts
,
F. N.
Beg
,
M.
Zepf
,
K.
Krushelnick
,
M.
Tatarakis
,
A. E.
Dangor
,
P. A.
Norreys
,
R.
Allott
,
D.
Neely
,
R. J.
Clark
,
A. C.
Machacek
,
J. S.
Wark
,
A. J.
Cresswell
,
D. C. W.
Sanderson
, and
J.
Magill
,
Phys. Rev. Lett.
84
,
899
(
2000
).
3.
H.
Vincenti
,
S.
Monchocé
,
S.
Kahaly
,
G.
Bonnaud
,
P.
Martin
, and
F.
Quéré
,
Nat. Commun.
5
,
3403
(
2014
).
4.
R. A.
Snavely
,
M. H.
Key
,
S. P.
Hatchett
,
T. E.
Cowan
,
M.
Roth
,
T. W.
Phillips
,
M. A.
Stoyer
,
E. A.
Henry
,
T. C.
Sangster
,
M. S.
Singh
,
S. C.
Wilks
,
A.
MacKinnon
,
A.
Offenberger
,
D. M.
Pennington
,
K.
Yasuike
,
A. B.
Langdon
,
B. F.
Lasinski
,
J.
Johnson
,
M. D.
Perry
, and
E. M.
Campbell
,
Phys. Rev. Lett.
85
,
2945
(
2000
).
5.
S. P.
Hatchett
,
C. G.
Brown
,
T. E.
Cowan
,
E. A.
Henry
,
J. S.
Johnson
,
M. H.
Key
,
J. A.
Koch
,
A. B.
Langdon
,
B. F.
Lasinski
,
R. W.
Lee
,
A. J.
Mackinnon
,
D. M.
Pennington
,
M. D.
Perry
,
T. W.
Phillips
,
M.
Roth
,
T. C.
Sangster
,
M. S.
Singh
,
R. A.
Snavely
,
M. A.
Stoyer
,
S. C.
Wilks
, and
K.
Yasuike
,
Phys. Plasmas
7
,
2076
(
2000
).
6.
M.
Roth
,
D.
Jung
,
K.
Falk
,
N.
Guler
,
O.
Deppert
,
M.
Devlin
,
A.
Favalli
,
J.
Fernandez
,
D.
Gautier
,
M.
Geissel
,
R.
Haight
,
C. E.
Hamilton
,
B. M.
Hegelich
,
R. P.
Johnson
,
F.
Merrill
,
G.
Schaumann
,
K.
Schoenberg
,
M.
Schollmeier
,
T.
Shimada
,
T.
Taddeucci
,
J. L.
Tybo
,
F.
Wagner
,
S. A.
Wender
,
C. H.
Wilde
, and
G. A.
Wurden
,
Phys. Rev. Lett.
110
,
044802
(
2013
).
7.
H.
Chen
,
S. C.
Wilks
,
J. D.
Bonlie
,
E. P.
Liang
,
J.
Myatt
,
D. F.
Price
,
D. D.
Meyerhofer
, and
P.
Beiersdorfer
,
Phys. Rev. Lett.
102
,
105001
(
2009
).
8.
T.
Esirkepov
,
M.
Borghesi
,
S. V.
Bulanov
,
G.
Mourou
, and
T.
Tajima
,
Phys. Rev. Lett.
92
,
175003
(
2004
).
9.
L.
Yin
,
B. J.
Albright
,
B. M.
Hegelich
, and
J. C.
Fernandez
,
Laser Part. Beams
24
,
291
(
2006
).
10.
S.
Kahaly
,
S.
Monchocé
,
V.
Gallet
,
O.
Gobert
,
F.
Réau
,
O.
Tcherbakoff
,
P. D.
Oliveira
,
P.
Martin
, and
F.
Quéré
,
Appl. Phys. Lett.
104
,
054103
(
2014
).
11.
V.
Gallet
,
S.
Kahaly
,
O.
Gobert
, and
F.
Quéré
,
Opt. Lett.
39
,
4687
(
2014
).
12.
K.
Zeil
,
J.
Metzkes
,
M.
Bussmann
,
T.
Cowan
,
S.
Kraft
,
R.
Sauerbrey
, and
U.
Schramm
,
Nat. Commun.
3
,
874
(
2012
).
13.
J. A.
Wheeler
,
A.
Borot
,
S.
Monchocé
,
H.
Vincenti
,
A.
Ricci
,
A.
Malvache
,
R.
Lopez-Martens
, and
Q.
Fabien
,
Nat. Photonics
6
,
829
(
2012
).
14.
K. T.
Kim
,
C.
Zhang
,
T.
Ruchon
,
J.-F.
Hergott
,
T.
Auguste
,
D.
Villeneuve
,
P.
Corkum
, and
F.
Quéré
,
Nat. Photonics
7
,
651
(
2013
).
15.
S.
Monchocé
,
S.
Kahaly
,
A.
Leblanc
,
L.
Videau
,
P.
Combis
,
F.
Réau
,
D.
Garzella
,
P.
D’Oliveira
,
P.
Martin
, and
F.
Quéré
,
Phys. Rev. Lett.
112
,
145008
(
2014
).
16.
M. A.
Purvis
,
V. N.
Shlyaptsev
,
R.
Hollinger
,
C.
Bargsten
,
A.
Pukhov
,
A.
Prieto
,
Y.
Wang
,
B. M.
Luther
,
L.
Yin
,
S.
Wang
, and
J. J.
Rocca
,
Nat. Photonics
7
,
796
(
2013
).
17.
J. T.
Morrison
, “
Selective deuteron acceleration using target normal sheath acceleration
,” Ph.D. dissertation (
The Ohio State University
,
2013
).
18.
Handbook of Biological Confocal Microscopy
, 3rd ed., edited by
J.
Pawley
(
Springer Science + Business Media, LLC
,
2006
).
19.
J. T.
Morrison
,
C.
Willis
,
R. R.
Freeman
, and
L.
Van Woerkom
,
Rev. Sci. Instrum.
82
,
033506
(
2011
).
20.
P. L.
Poole
,
C. D.
Andereck
,
D. W.
Schumacher
,
R. L.
Daskalova
,
S.
Feister
,
K. M.
George
,
C.
Willis
,
K. U.
Akli
, and
E. A.
Chowdhury
,
Phys. Plasmas
21
,
063109
(
2014
).