MM-wave cavity/klystron developments using deep x-ray lithography at the Advanced Photon Source. Page: 2 of 3
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to UNT Digital Library by the UNT Libraries Government Documents Department.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
(1-GeV Aladdin). For the DXRL mask, 45-pm-thick layer
of Au was plated over a 300-pm-thick Si wafer where the
x-ray was exposed and the resist removed. To observe the
high depth-to-width aspect ratio in the final product, mi-
cron-range structures were patterned on the DXRL mask.
To avoid alignment problems and x-ray diffraction, these
two steps were done on the same sample substrate without
a physical gap.
Poly-methylmethacrylate (PMMA) up to 1 mm thick
was used as a positive resist. The copper substrate was
diamond-finished to have a flatness of 1 pm over 4 inches.
Then either an oxide film was grown to one micron thick
or an equally thick Ti coating was deposited in order to
promote better adhesion to the copper substrate. Through
these processes, the flatness of the copper surface is main-
tained, but is still rough enough to give good adhesion to
the PMMA sheet, which has a roughness of less than 0.1
pm [6].
When the PMMA film was cast onto the copper sub-
strate, it was annealed at various temperatures from 110 to
1700 C for one to three hours [7]. The National Synchro-
tron Light Source (NSLS) X-26C beamline and the APS
2-BM-A beamline were used to expose the sample. The
transmitted x-ray intensity was calculated based on the
APS bending magnet parameters and is plotted in Fig. 2.
The ratio of the top dose to the bottom dose for the 1-
mm-thick PMMA is about 1.1. During the exposure, the
sample was enclosed in a He-purged housing with a Kap-
ton window, and the sample holder baseplate was water
cooled. The first Platinum (Pt) mirror with a grazing an-
gle of 0.150 was used to cut off all the high-energy x-rays
above 40 keV as shown in Figure 2. More information on
the APS 2-BM beamline for the DXRL can be found in
Ref. 8.
Transmitted Intensity during PMMA exposures for LIGA
In nsay (wavs/honzanmal_cm) u
1e-R
2 5
Figure 2: Transmitted x-ray intensity during PMMA ex-
posures for DXRL at APS, energy vs. intensity.
Two different developers were used in the developing
process. The first developer, so-called GG, was a mixture
of 60% vol 2-(2-butoxy-ethoxy) ethanol, 20% tetrahydro-
1, 4-oxazine, 5% 2-aminoethanol-1, and 15% deionized
water. The allowed dose range was from 3 to about 10kJ/cm3. Below that threshold the crosslinked resist could
not be dissolved, and above that range damage to the resist
can occur from production of gases in the PMMA. The
second developer was methyl-iso-butyl ketone (MIBK)
diluted with 2-propanol. After developing the microstruc-
ture, copper can be electroplated to the positive resist and
the surface can be diamond-finished.
3 MM-WAVE STRUCTURES
Due to DXRL's ability to maintain precise tolerances, it
is ideally suited for the manufacture of rf components op-
erating at frequencies above 30 GHz. The first two
structures fabricated were a 32-cell, 108-GHz constant-
impedance cavity and a 66-cell, 94-GHz constant-gradient
cavity. A 32-cell, 108-GHz constant-impedance cavity is a
planar accelerating structure with parameters as shown in
Table 1. The change from a typical cylindrical symmetri-
cal disk-like structure to a planar accelerating structure
results in a loss in shunt impedance and Q value of less
than 5%. An accelerating gradient of 50 MV/m is chosen
for a practical 50-MeV microlinac application, but it is
not limited to 50 MV/m when the rf system is operational
in the pulse mode with less repetition rate.
Table 1: The rf Parameters of a 32-Cell, 108-GHz
Constant-Impedance CavityFrequency
Shunt impedance
Quality factor
Operating mode
Group velocity
Attenuation factor
Accelerating gradient
Peak powerf
R
Q
TW
vg
a
E
P108 GHz
312 MQ/m
2160
27c/3
0.043C
13.5 m-
50 MV/m
30 kWFor a constant-impedance planar structure, the double-
periodic structures with confluence in the it-mode design
were considered. The 2/3-mode operation in these struc-
tures can give high shunt impedance, group velocity, and
low sensitivity on dimensional errors. More detailed de-
scriptions, such as rf simulation using the MAFIA
computer code, and the thermal analysis related to this
structure can be found in Ref. 9.
Constant-gradient accelerating structures are used in
many present-day accelerators because of their higher en-
ergy gain and better frequency characteristics, such as
higher shunt impedance, more uniform power dissipation,
and lower sensitive to frequency deviations, when com-
pared to the constant-impedance structure. Tapering the
cells along the structure while keeping the gap and cell
depth constant is difficult. Since the structure needs to be
manufactured on a planar wafer, adjusting the cell width
and length while maintaining a constant depth within the
structure is necessary. Figure 3 shows the constant-
gradient structure with cuts in the irises; its rf parameters
can be found in Ref. 10.
Upcoming Pages
Here’s what’s next.
Search Inside
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Song, J.J.; Decarlo, F.; Kang, Y.W.; Kustom, R.L.; Mancini, D.c.; Nassiri, A. et al. MM-wave cavity/klystron developments using deep x-ray lithography at the Advanced Photon Source., article, March 31, 1998; Argonne, Illinois. (https://digital.library.unt.edu/ark:/67531/metadc623000/m1/2/: accessed April 25, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.