The technique of atomic layer deposition (ALD) has enabled the development of alternative glass microchannel plates (MCPs) with independently tunable resistive and emissive layers, resulting in excellent thickness uniformity across the large area (20 × 20 cm), high aspect ratio (60:1 L/d) glass substrates. Furthermore, the use of ALD to deposit functional layers allows the optimal substrate material to be selected, such as borosilicate glass, which has many benefits compared to the lead-oxide glass used in conventional MCPs, including increased stability and lifetime, low background noise, mechanical robustness, and larger area (at present up to 400 cm2). Resistively stable, high gain MCPs are demonstrated due to the deposition of uniform ALD resistive and emissive layers on alternative glass microcapillary substrates. The MCP performance characteristics reported include increased stability and lifetime, low background noise (0.04 events cm−2 s−1), and low gain variation (±5%).

Microchannel plates (MCPs) are used for the amplification of electrons through micron-sized channels (2–40 μm) that penetrate from the front surface to the back surface of a glass-based substrate. Incom fabricates MCPs with typical aspect ratios up to 80:1, bias angles of 5°–13°, pore diameters from 10 to 40 μm, and open-area ratio of up to ∼83%. They can be used in a range of applications for the detection of electrons, ions, neutrons, and photons.1,2 MCPs are integrated into a variety of devices such as photodetectors,1 photomultiplier tubes (PMTs),2–4 medical imagers, night vision products, cathode ray tubes, image intensifier tubes,2 and time-of-flight (ToF) mass spectrometers.5 

An MCP operates with a bias applied between the (typically) NiCr electrodes on the two planar surfaces generating an electric field of ∼106 V/m along the pores of the MCP. When an electron enters the input side of the MCP and strikes the wall of the pore, it can cause secondary electrons to be emitted, which are accelerated along the pore of the MCP, generating further secondary electrons upon impact with the pore walls. This results in the generation of an electron cascade, which is detected at the output (more positive) side of the MCP. A number of factors influence the amplification of electrons, including the pore diameter, aspect ratio of MCP, secondary electron emission (SEE) layer, and bias angle of the plate. Two MCPs can be configured as a chevron pair or three in a Z-stack for enhanced signal amplification.

Conventional MCPs are fabricated using soluble-core glass tubing with an outer lead-glass cladding layer, which is drawn into fibers, then bundled into a block, sliced, and formed into fiber optic plates. The soluble fiber cores are chemically etched out to form the pores, and the plate is hydrogen fired to impart the resistive and emissive properties on the MCP. Using this fabrication method, the MCP aspect ratio is limited to ∼100:1, as it is not possible for the chemical etchant to diffuse into longer pores without dissolving the outer glass shell. Based on the limitations of processing conventional lead-glass MCPs, it is challenging to fabricate large-area MCPs with small pores as they are delicate and can warp after hydrogen-firing. Some hydrogen tends to remain in the pores, to be released slowly during operation. The resulting hydrogen ions travel in the opposing direction to the electrons and tend to damage photocathodes in sealed photodetectors through sputtering (either directly, or by knocking other adsorbates off the pore walls). This is the so-called ion-feedback problem.6,7

As a substrate material, glass offers a broad range of properties depending on the composition, which can be selected to target MCPs with a specific chemical stability, mechanical strength, or electrical performance. The examples include soda-lime silicate glasses that are low cost and widely available; sodium borosilicate glasses that offer chemical durability, low thermal expansion, and moderate electrical resistance; alkaline earth aluminosilicate glasses for high temperature applications; and lead alkali silicate for applications similar to those used for conventional MCPs. Natural abundance levels of radioactive isotopes (such as potassium and rubidium) can cause a significant background noise in photodetector devices, but borosilicate glass mitigates this concern by having lower 40K content than lead-glass.8–10 

Atomic layer deposition (ALD) is routinely used to coat a variety of substrates with uniform, conformal thin films due to the surface-saturating, self-limiting characteristics of this layered deposition process.11 Unlike other thin film deposition techniques that rely on line-of-sight between the source and the substrate, ALD growth occurs by reacting alternating doses of gaseous precursors (separated by inert-gas purges) at the substrate surface. This makes ALD the thin-film deposition method of choice for processing complex substrate geometries, high aspect ratio or large surface area substrates,12 and large batches of substrates.13 The optimized processing parameters are crucial to ensure the uniform ALD coating of complex, high area substrates and batches of substrates. To uniformly ALD-coat high surface area substrates such as MCPs, the precursor dose times and purges times have to be carefully studied to ensure there is sufficient precursor delivered to saturate the substrate, and that chemical vapor deposition is avoided which may result in nonuniform deposition. ALD has been used to functionalize commercially available conventional lead-oxide MCPs.3,4,14,15 It has been shown that deposition of an ALD SEE layer on conventional lead-oxide glass MCP-PMTs resulted in an improved gain performance and an increased device life time due to a higher secondary electron yield and the ALD layers effectively sealing the MCP surface, preventing desorption of gas-phase contaminants in the MCP glass.3,4

ALD was used to functionalize nonconventional MCPs that are fabricated by Incom, Inc., from lower cost, readily available glasses such as borosilicate (Pyrex®-like) glass. ALD offers the benefit of uniformly coating both the MCP pores and the MCP surfaces, with layers of materials that independently control both the resistance and gain of the resulting MCP. The MCPs are produced using glass tubes that are drawn into hollow fibers, thus eliminating the etching step. They are thus not limited by the aspect ratio constraints introduced by the etching step required for the production of conventional MCPs, and the elimination of the hydrogen firing step mitigates the problem of ion feedback. A range of different glasses has been explored using the hollow-core method, including MCPs with open area ratios up to 83%, low cost borosilicate glass MCPs, and alternative glasses that have low alkali content. The broad range of glasses accessible means that MCPs can be fabricated and tailored according to the final application. The use of ALD layers in the manufacture of MCPs provides independent control of the conductivity and electron emissivity of the resulting MCP.3,4,14–18 Using the hollow-core fabrication methodology, MCPs can be produced in larger formats (currently up to 20 × 20 cm), with higher aspect ratios, and open area ratios up to 65% for the largest MCP. The standard pore diameters for the MCPs manufactured using this process are 20 and 10 μm with improved temporal and spatial resolutions achieved using smaller diameter MCPs.

ALD can be used to functionalize MCPs with two layers: (1) a nanocomposite resistive layer19–22 and (2) an electron emissive layer that generates secondary electrons within the channels of the MCP (SEE layer). Previous works16,18,21,22 described the use of W:Al2O3 and Mo:Al2O3 as resistive layers to functionalize Incom's alternative glass MCPs, with the resistance of the resulting MCPs determined by the ratio of components within the nanocomposite layers. In this work, the nanocomposite resistive layer was composed of conducting metallic W nanoparticles within an insulating Al2O3 matrix, forming a W:Al2O3 layer.16,18,21,22 W:Al2O3 was selected rather than Mo:Al2O3 as the WF6 precursor was significantly less expensive than MoF6, which was an important consideration for the processing costs associated with the commercialization of large area MCPs. The challenge with ALD processing large area MCPs is the deposition of uniform, conformal resistive and emissive layers on a high surface area substrate (400 cm2, 20 μm pores, and 65% open area ratio). The resistivity of the W:Al2O3 layer was targeted by varying the ratio of Al2O3 to W ALD cycles. The SEE layer of MgO or Al2O3 was deposited on top of the resistive layer. Conventional hydrogen-fired MCPs have a secondary electron yield of 2.5–3.5,23,24 whereas ALD Al2O3 or MgO have a significantly higher secondary electron emissivity of 1–6,25 and 3–7, respectively, depending on the film thickness, incident electron angle, incident electron energy, and escape path length from the medium.25 

MCPs fabricated using borosilicate-type glasses, and functionalized with ALD, offer a robust, larger format solution with targeted, independently tunable resistive and emissive properties. Coupled with the dimensional and handling advantages of borosilicate glass MCPs, the deposition of an MgO SEE layer offers the promise of larger MCPs with higher secondary electron yield than conventional MCPs.2,26

Incom's ALD-functionalized borosilicate MCPs have been extensively tested and shown to have low intrinsic background (∼0.04 events cm−2 s−1)8 and stable gain during >7 C/cm2 charge extraction after preconditioning (vacuum bake and burn-in).27 The overall aim of producing these high gain, robust, stable, low cost, large area (20 × 20 cm) MCPs is to integrate them into a sealed photodetector package as part of the Large Area Picosecond Photodetector (LAPPD™) program27–31 funded by the U.S. Department of Energy. The large area photodetectors are designed for use in high energy physics applications such as water Cherenkov counters, large scintillation detectors, vertex separation and particle identification in time-of-flight measurements, and accelerator beam diagnostics.28,29 Other uses include neutron and neutrino detection for defense and Homeland Security applications, medical imaging including positron emission tomography (PET) scanners, and UV detection for space applications.28,29

ALD layers of W, Al2O3, and MgO were deposited at 200 °C using a Beneq TFS 500 system. The W component of the nanocomposite layer was deposited by alternating between pulses of WF6 (99.8%, Sigma-Aldrich) and Si2H6 (99.998%, Sigma-Aldrich).32 Al2O3 was deposited with trimethylaluminum (TMA, 97%, Sigma-Aldrich) and deionized H2O. MgO was deposited with biscylopentadientylmagneisum (MgCp2, 99.9%, Strem) and H2O. TMA and H2O were regulated to 20 °C to achieve consistent vapor pressures. WF6 and Si2H6 were maintained at room temperature. MgCp2 was heated to 80 °C. A constant flow of 300 sccm ultrahigh purity (99.999%) nitrogen was used as a carrier gas throughout the ALD process.

Using the Beneq TFS 500 ALD system with a specially modified reaction chamber, recipes for targeted resistances were developed for different MCP geometries (Table I). For all ALD processes, the precursor dosing and purge times were optimized to ensure sufficient delivery of the precursors to saturate the high surface area MCP substrates resulting in uniform deposition on both the front and back surfaces of the MCP and throughout the substrate pores. All MCPs were functionalized with resistive and emissive layers. To target the required resistive range for a certain MCP geometry, the W:Al2O3 recipes were optimized (precursor dose time, purge time, and ratio of W:Al2O3 cycles), resulting in typical resistance values from 10 to 120 MΩ. The thickness of the W:Al2O3 and SEE layers was measured on 300 mm strips of Si(100) wafer using a M-2000 ellipsometer from J. A. Woollam.

Table I.

Dimensions of MCPs fabricated by Incom, Inc., referenced in this work.

Pore size (μm)Bias angleOpen area ratio (%)Aspect ratio (L/d)MCP thickness (mm)MCP dimensions (mm)
20 8° 70–75 60:1 1.2 33 (diameter) 
20 8° 65 60:1 1.2 33 (diameter) 
10 8° 65 60:1 0.6 33 (diameter) 
20 8° 65 60:1 1.2 200 × 200 
Pore size (μm)Bias angleOpen area ratio (%)Aspect ratio (L/d)MCP thickness (mm)MCP dimensions (mm)
20 8° 70–75 60:1 1.2 33 (diameter) 
20 8° 65 60:1 1.2 33 (diameter) 
10 8° 65 60:1 0.6 33 (diameter) 
20 8° 65 60:1 1.2 200 × 200 

To create an electrical contact, a 200 nm nickel-chromium (NiCr) layer was deposited by evaporation of 80/20 NiCr on the input and output surfaces of the MCP using a Sharon Vacuum thermal evaporator. The MCPs are loaded in a rotating planetary fixture with the axis of the pores at 45° relative to the evaporation source. The 8° bias angle of the MCPs is factored into the angle of deposition of the planetary with the NiCr depositing 1.0–1.7 pore diameters into the MCP,8 known as end-spoiling.

Following deposition of the ALD and NiCr layers, the resistance of the MCPs was determined by measuring the current–voltage characteristics in high vacuum using a Keithley 6487 picoammeter/voltage source. Additional characterization of the MCPs has been previously outlined.8 

Optimized ALD recipes for the deposition of W:Al2O3, Al2O3, and MgO were developed for processing single 20 × 20 cm MCPs, and batches of 33 mm MCPs. An excellent thickness uniformity was achieved across the 300 mm reaction chamber with a thickness standard deviation of 3.9%, 2.1%, 5.1% for W:Al2O3, Al2O3, and MgO, respectively, measured on Si(100) witness coupons positioned beneath the MCP(s) in the reaction chamber (Fig. 1). A growth per cycle of 1.1 Å/cycle for W:Al2O3 (500 cycles), 1.3 Å/cycle (100 cycles) for Al2O3, and 1.7 Å/cycle for MgO (50 cycles) was determined.

Fig. 1.

(Color online) Film thickness measured on Si in 300 mm reaction chamber for deposition of W:Al2O3, Al2O3, and MgO.

Fig. 1.

(Color online) Film thickness measured on Si in 300 mm reaction chamber for deposition of W:Al2O3, Al2O3, and MgO.

Close modal

A batch of 33 mm diameter MCPs was ALD functionalized with W:Al2O3 as outlined in Table II, with resistively matched MCPs (within 12% for each MCP pore geometry) produced independent of the glass type. The deposition of optimized ALD films on MCPs was dependent not only on the process parameters used during deposition, but also the geometry and dimensions of the MCPs, and the use of appropriate substrate holders. The total surface area of a 33 mm MCP with 10 μm pores and 60:1 L/d is the same as the surface area of a 33 mm MCP with 20 μm pores and 60:1 L/d. The resistance of single 10 μm 60:1 L/d pore is the same as the resistance of a 20 μm 60:1 L/d pore. However, there are four times the number of 10 μm pores in a 33 mm MCP with 60:1 L/d, compared to a 33 mm MCP with 20 μm pores and 60:1 L/d. It was expected that the resistance of a 10 μm pore MCP processed with the same W:Al2O3 film would be ¼ the resistance of the 20 μm MCP. A reduced resistance was observed for 10 μm MCPs (Table II). However, the measured resistance of the 10 μm MCPs was ½ the resistance of the 20 μm MCPs. There was no additional NiCr electrode deposited on top of the W:Al2O3 layer, and the resistance was measured with a rim contact. Therefore, in addition to the resistance of the MCP pores, the resistive path of the top and bottom surfaces of the MCP, which may be different for the 10 and 20 μm MCPs, must be accounted for. The observed decrease in resistance is currently under investigation, including depositing an additional NiCr layer to eliminate the resistive pathways on the top and bottom surfaces of the MCP.

Table II.

MCPs of 33 mm with 20 or 10 μm pores processed with W:Al2O3.

MCP diameter (mm), thickness (mm), pore size (μm)Resistance (MΩ)
33, 0.6, 10 44 
33, 0.6, 10 42 
33, 0.6, 10 43 
33, 0.6, 10 43 
33, 0.6, 10 48 
33, 1.2, 20 109 
33, 1.2, 20 106 
33, 1.2, 20 103 
33, 1.2, 20 123 
33, 1.2, 20 107 
33, 1.2, 20 107 
33, 1.2, 20 118 
MCP diameter (mm), thickness (mm), pore size (μm)Resistance (MΩ)
33, 0.6, 10 44 
33, 0.6, 10 42 
33, 0.6, 10 43 
33, 0.6, 10 43 
33, 0.6, 10 48 
33, 1.2, 20 109 
33, 1.2, 20 106 
33, 1.2, 20 103 
33, 1.2, 20 123 
33, 1.2, 20 107 
33, 1.2, 20 107 
33, 1.2, 20 118 

After the resistance of the MCPs has been measured, the MCP performance is characterized in detail. To determine the gain–voltage characteristics of a single MCP, two MCPs are assembled into a stack. The first MCP is illuminated with UV light, to serve as a source of electrons that are directed to the second MCP. The gain of the second MCP is determined by measuring the ratio of its input current (typically 0.1–1.0 nA) to its output current, at different voltages applied across the second MCP (Fig. 2). Figure 2 shows the gain–voltage relationship for a series of 33 mm diameter 20 μm pore MCPs with an Al2O3 SEE layer with measured gain of 103 for a single MCP at 800 V.8 In previous work, it has been demonstrated that the gain achievable from borosilicate MCPs with an ALD SEE layer of Al2O3 is comparable to or slightly higher than the gain measured with lead-oxide MCPs (Refs. 2 and 26) with the potential for higher gains when using an SEE layer with a higher electron yield such as MgO and with an optimized layer thickness.25 The gain was measured as a function of charge extracted from the MCP under UV illumination.

Fig. 2.

(Color online) Gain–voltage plot for five different Al2O3-coated 33 mm 20 μm pore MCPs with gains of ∼103 at 800 V (gain has dimensionless unit).

Fig. 2.

(Color online) Gain–voltage plot for five different Al2O3-coated 33 mm 20 μm pore MCPs with gains of ∼103 at 800 V (gain has dimensionless unit).

Close modal

To determine the gain-voltage characteristics of a pair of MCPs, the first MCP is illuminated with UV light, which ionizes the input side of the first MCP creating a photoelectron. This phototelectron causes charge multiplication in both MCPs. The gain is the output charge from the second MCP divided by the charge of an electron. In this case, enough charge is generated to form a pulse height distribution of their charge output, as shown in Fig. 3.

Fig. 3.

(Color online) Pulse height distributions of pair of 33 mm MCPs (20 μm, 1.2 mm thick, 60:1 L/d) with Al2O3 SEE layer (counts and gain have dimensionless units).

Fig. 3.

(Color online) Pulse height distributions of pair of 33 mm MCPs (20 μm, 1.2 mm thick, 60:1 L/d) with Al2O3 SEE layer (counts and gain have dimensionless units).

Close modal

Longer term testing was completed to determine the characteristics, stability, and performance of the MCPs. The MCPs were preconditioned by illumination with a uniform source of electrons from a standard MCP and setting the output current to ∼1 μA to “burn in” the test MCP (Refs. 8 and 33) with the gain measured under increasing charge extraction (lifetime testing). It was demonstrated that MgO-coated MCPs initially increase in gain (up to 0.03 C/cm2 extracted charge) but then stabilize to their initial gain value with 1 C/cm2 extracted from the MCP [Fig. 4(a)]. For Al2O3-coated MCPs, the gain decreases with the initial extracted charge (a factor of 10 gain decrease with 0.05 C/cm2) and then is observed to slightly increase and stabilize [Fig. 4(b)].8 This reduction in the initial gain measured may be attributed to gradual desorption of adsorbed species such as water from the pore surfaces of the MCP.

Fig. 4.

(Color online) (a) Gain (dimensionless unit) with extracted charge during burn-in for an MgO-coated 33 mm 20 μm pore MCP (1.2 mm thick, 60:1 L/d) up to 1 C/cm2 extracted charge. (b) Gain (dimensionless unit) with extracted charge during burn-in for an Al2O3-coated 33 mm 20 μm pore MCP (1.2 mm thick, 60:1 L/d) up to 0.25 C/cm2 extracted charge.

Fig. 4.

(Color online) (a) Gain (dimensionless unit) with extracted charge during burn-in for an MgO-coated 33 mm 20 μm pore MCP (1.2 mm thick, 60:1 L/d) up to 1 C/cm2 extracted charge. (b) Gain (dimensionless unit) with extracted charge during burn-in for an Al2O3-coated 33 mm 20 μm pore MCP (1.2 mm thick, 60:1 L/d) up to 0.25 C/cm2 extracted charge.

Close modal

The flat field and gain map images of a 33 mm MCP with an MgO SEE layer are shown in Figs. 5(a) and 5(b), respectively. The uniformity in the gain of this MCP is attributed to the conformal deposition of the MgO SEE layer with gain variation of ±5% in the x- and y-directions [Fig. 5(c)]. The production of MCPs with a uniform film thickness for both the resistive and SEE layers (Fig. 1), matched, stable resistance (Table II) and high, uniform gain (Figs. 2, 3, and 4) demonstrates the merits of using ALD as the deposition method for the functionalization of MCPs.

Fig. 5.

(Color online) (a) Flat field image of MgO-coated 33 mm 20 μm pore (1.2 mm thick, 60:1 L/d) MCP. (b) Gain map of MgO-coated 33 mm 20 μm pore (1.2 mm thick, 60:1 L/d) MCP under initial UV illumination. (c) Gain distribution across 33 mm 20 μm MCP in x-direction, and y-direction, with variation of ±5% (dimensionless units of relative gain, and position pixel).

Fig. 5.

(Color online) (a) Flat field image of MgO-coated 33 mm 20 μm pore (1.2 mm thick, 60:1 L/d) MCP. (b) Gain map of MgO-coated 33 mm 20 μm pore (1.2 mm thick, 60:1 L/d) MCP under initial UV illumination. (c) Gain distribution across 33 mm 20 μm MCP in x-direction, and y-direction, with variation of ±5% (dimensionless units of relative gain, and position pixel).

Close modal

To demonstrate the scalability of the ALD processes, Incom's large area 20 × 20 cm MCPs with 20 μm pores (1.2 mm thick, 60:1 L/d) were coated with resistive and emissive layers and characterized to ensure uniform, stable resistance and high gain. In order to determine the uniformity of the ALD layers deposited on the large area MCP, an imaging detector was used with a cross delay line readout,8,34 which allows the MCP to be imaged, the gain uniformity to be mapped, and the pulse amplitude distributions measured.8 We can measure the spatial variation of the gain, but the resistance measured is a bulk value. To determine the uniformity of the ALD response, the gain is measured on both sides of the MCP (Fig. 6) with a gain of 103 measured at 900 V for a single 20 × 20 cm MCP with an Al2O3 SEE layer. This gain value is comparable to the performance achieved on 33 mm MCPs with an Al2O3 SEE layer [Figs. 2, 3, and 4(b)], demonstrating the scalability of the ALD resistive and emissive processes. Efforts are on-going to functionalize and characterize large area MCPs with an ALD MgO SEE layer.

Fig. 6.

(Color online) Gain–voltage plot of 20 × 20 cm MCP (20 μm pores, 1.2 mm thick, 60:1 L/d) with an Al2O3 SEE layer, measured on both sides of MCP.

Fig. 6.

(Color online) Gain–voltage plot of 20 × 20 cm MCP (20 μm pores, 1.2 mm thick, 60:1 L/d) with an Al2O3 SEE layer, measured on both sides of MCP.

Close modal

To determine the uniformity of the Al2O3 ALD SEE layer on a 20 × 20 cm MCP, a gain map of the illuminated MCP was taken [Fig. 7(a)], revealing excellent gain uniformity across the substrate surface (typically ∼15% variation). The background of the MCP [Fig. 7(b)] shows uniform event rates of 0.17 events cm−2 s−1 with even lower background rates of 0.04 events cm−2 s−1 measured when “hot-spots” caused by debris on the MCP surface are excluded.8 This background rate of 0.04 events cm−2 s−1 is >5× lower than conventional glass MCPs.35 

Fig. 7.

(Color online) (a) Gain map of Al2O3-coated 20 × 20 cm 20 μm pore (1.2 mm thick, 60:1 L/d) MCP showing uniform gain across entire surface. (b) Background image of pair of 20 × 20 cm 20 μm MCPs with count rates of 0.04 events cm−2 s−1 when hot-spots on MCPs excluded (0.17 events cm−2 s−1 otherwise).

Fig. 7.

(Color online) (a) Gain map of Al2O3-coated 20 × 20 cm 20 μm pore (1.2 mm thick, 60:1 L/d) MCP showing uniform gain across entire surface. (b) Background image of pair of 20 × 20 cm 20 μm MCPs with count rates of 0.04 events cm−2 s−1 when hot-spots on MCPs excluded (0.17 events cm−2 s−1 otherwise).

Close modal

Optimized ALD processes were developed for the deposition of uniform layers of W:Al2O3, Al2O3, and MgO on batches of 33 mm MCPs, and single 20 × 20 cm MCPs. The uniformity of the ALD layers was verified by measurement of film thickness, resistance characterization, and gain mapping of MCPs, revealing excellent thickness uniformity for all ALD processes investigated, independent of the substrate size or glass type used. Resistively matched MCPs were fabricated with high, stable, uniform gain (103–104) tested under extended charge extraction, and low background rates (0.04 events cm−2 s−1) due to the use of alternative glass substrates. The deposition of an ALD W:Al2O3 layer that can be tuned to target specified resistance ranges, in combination with an ALD SEE layer of Al2O3 or MgO with higher electron emissivity, offers the opportunity to manufacture application-targeted and customer-specified MCPs. The performance and fabrication advantages of ALD functionalized MCPs, coupled with the benefits of using alternative glass substrates, provides many opportunities for commercialization of these MCP products in addition to exploration of alternate ALD chemistries and glass substrates.

The authors acknowledge Matthew J. Wetstein, Andrey Elagin, and Henry J. Frisch for their contributions to this work. This material is based upon work supported by the U.S. Department of Energy SBIR/STTR Programs, under Award Nos. DE-SC0009717 and DE-SC0011262.

1.
O. H. W.
Siegmund
,
J.
Vallerga
, and
A.
Tremsin
,
Proc. SPIE
5898
,
58980H
(
2005
).
2.
J. L.
Wiza
,
Nucl. Instrum. Methods
162
,
587
(
1979
).
3.
T. M.
Conneely
,
J. S.
Milnes
, and
J.
Howorth
,
Nucl. Instrum. Methods A
732
,
391
(
2013
).
4.
A.
Britting
,
W.
Eyrich
,
A.
Lehmann
, and
F.
Uhlig
,
Acta Phys. Pol. B Proc. Suppl.
7
,
701
(
2014
).
5.
R.
Liu
,
Q.
Li
, and
L. M.
Smith
,
J. Am. Soc. Mass Spectrom.
25
,
1374
(
2014
).
6.
N.
Kishimoto
,
M.
Nagamine
,
K.
Inami
,
Y.
Enari
, and
T.
Ohshima
,
Nucl. Instrum. Methods A
564
,
204
(
2006
).
7.
K.
Inami
 et al,
Nucl. Instrum. Methods A
592
,
247
(
2008
).
8.
C.
Ertley
,
O. H. W.
Siegmund
,
J.
Schwarz
,
A. U.
Mane
,
M. J.
Minot
,
A.
O'Mahony
,
C. A.
Craven
, and
M.
Popecki
,
Proc. SPIE
9601
,
96010S
(
2015
).
9.
O. H. W.
Siegmund
 et al,
IEEE Trans. Nucl. Sci.
60
,
923
(
2013
).
10.
O. H. W.
Siegmund
,
M. A.
Gummin
,
J.
Stock
, and
D.
Marsh
, “
Microchannel plate imaging detectors for the ultraviolet
,”
Proceedings of an ESA Symposium on Photon Detectors for Space Instrumentation SP-356
(
1992
).
11.
S. M.
George
,
Chem. Rev.
110
,
111
(
2010
).
12.
J. W.
Elam
, “
Coatings on high aspect ratio structures
,” in
Atomic Layer Deposition of Nanostructured Materials
, edited by
N.
Pinna
and
M.
Knez
(
Wiley-VCH
,
Weinheim
,
2012
), pp.
227
.
13.
E.
Granneman
,
P.
Fischer
,
D.
Pierreux
,
H.
Terhorst
, and
P.
Zagwijn
,
Surf. Coat. Technol.
201
,
8899
(
2007
).
14.
N. T.
Sullivan
,
A.
Tremsin
,
P.
de Rouffignac
,
D.
Beaulieu
,
K.
Saadatmand
,
S.
Bachman
,
K.
Stenton
, and
D.
Gorelikov
, U.S. patent 8,237,129 (7 August
2012
).
15.
N. T.
Sullivan
,
S.
Bachman
,
P.
de Rouffignac
,
A.
Tremsin
,
D.
Beaulieu
, and
D.
Gorelikov
, U.S. patent 8,227,965 (24 July
2012
).
16.
A. U.
Mane
 et al,
Phys. Proc.
37
,
722
(
2012
).
17.
D. R.
Beaulieu
,
D.
Gorelikov
,
H.
Klotzsch
,
P.
de Rouffignac
,
K.
Saadatmand
,
K.
Stenton
,
N.
Sullivan
, and
A. S.
Tremsin
,
Nucl. Instrum. Methods Phys. Res. Sect. A
633
,
S59
(
2011
).
18.
J. W.
Elam
 et al,
ECS Trans.
58
,
249
(
2013
).
19.
J. W.
Elam
and
S. M.
George
,
Chem. Mater.
15
,
1020
(
2003
).
20.
J. W.
Elam
,
D.
Routkevitch
, and
S. M.
George
,
J. Electrochem. Soc.
150
,
G339
(
2003
).
21.
A. U.
Mane
and
J. W.
Elam
,
Chem. Vap. Deposition
19
,
186
(
2013
).
22.
J. W.
Elam
,
A. U.
Mane
, and
Q.
Peng
, U.S. patent 8,969,823 B2 (3 March
2015
).
23.
J. J.
Fijol
,
A. M.
Then
,
G. W.
Tasker
, and
R. J.
Soave
,
Appl. Surf. Sci.
48/49
,
464
(
1991
).
24.
A. M.
Then
and
C. G.
Pantano
,
J. Non-Cryst. Solids
120
,
178
(
1990
).
25.
S. J.
Jokela
,
I. V.
Veryovkin
,
A. V.
Zinovev
,
J. W.
Elam
,
A. U.
Mane
,
Q.
Peng
, and
Z.
Insepov
,
Phys Proc.
37
,
740
(
2012
).
26.
O. H. W.
Siegmund
, “
Amplifying and position sensitive detectors
,” in
Methods of Vacuum Ultraviolet Physics
, 2nd ed. edited by
J. A. R.
Samson
and
D. L.
Ederer
(
Academic
,
New York
,
1998
), Chap. III.
27.
O. H. W.
Siegmund
,
N.
Richner
,
G.
Gunjala
,
J. B.
McPhate
,
A.
Tremsin
,
H. J.
Frisch
,
J. W.
Elam
,
A. U.
Mane
, and
R. G.
Wagner
,
Proc. SPIE
8859
,
88590Y
(
2013
).
28.
M. J.
Minot
 et al,
Nucl. Instrum. Methods A
787
,
78
(
2015
).
29.
C. A.
Craven
 et al,
Proceedings of Technology and Instrumentation in Particle Physics
(
2014
).
30.
H. J.
Frisch
, “
The development of large-area detectors with space and time resolution
,”
Advances in Neutrino Technology Workshop
(
2009
).
31.
B. W.
Adams
,
A.
Elagin
,
H. J.
Frisch
,
R.
Obaid
,
E.
Oberla
,
A.
Vostrikov
,
R. G.
Wagner
,
J.
Wang
, and
M.
Wetstein
,
Nucl. Instrum. Methods A
795
,
1
(
2015
).
32.
J. W.
Klaus
,
S. J.
Ferro
, and
S. M.
George
,
Thin Solid Films
360
,
145
(
2000
).
33.
O. H. W.
Siegmund
,
Proc. SPIE
1072
,
111
(
1989
).
34.
O. H. W.
Siegmund
 et al,
Proceedings of the Advanced Maui Optical and Space Surveillance Technologies Conference
(
2012
).
35.
O. H. W.
Siegmund
and
C.
Ertley
, “
Very large area 20 cm × 20 cm flat panel phototubes using ALD microchannel plates
,”
IEEE Nuclear Science Symposium
(unpublished).