U.S. patent application number 14/053066 was filed with the patent office on 2014-06-19 for ultra-high speed anisotropic reactive ion etching.
This patent application is currently assigned to THE PENN STATE RESEARCH FOUNDATION. The applicant listed for this patent is Gokhan HATIPOGLU, Srinivas TADIGADAPA. Invention is credited to Gokhan HATIPOGLU, Srinivas TADIGADAPA.
Application Number | 20140166618 14/053066 |
Document ID | / |
Family ID | 50929732 |
Filed Date | 2014-06-19 |
United States Patent
Application |
20140166618 |
Kind Code |
A1 |
TADIGADAPA; Srinivas ; et
al. |
June 19, 2014 |
ULTRA-HIGH SPEED ANISOTROPIC REACTIVE ION ETCHING
Abstract
A system and method for reactive ion etching (RIE) system of a
material is provided. The system includes a plasma chamber
comprising a plasma source and a gas inlet, a diffusion chamber
comprising a substrate holder for supporting a substrate with a
surface comprising the material and a gas diffuser, and a source of
a processing gas coupled to the gas diffuser. In the system and
method, at least one radical of the processing gas is reactive with
the material to perform etching of the material, the gas diffuser
is configured to introduce the processing gas into the processing
region, and the substrate holder comprises an electrode that can be
selectively biased to draw ions generated by the plasma source into
the processing region to interact with the at least one processing
gas to generate the at least one radical at the surface.
Inventors: |
TADIGADAPA; Srinivas; (State
College, PA) ; HATIPOGLU; Gokhan; (State College,
PA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
TADIGADAPA; Srinivas
HATIPOGLU; Gokhan |
State College
State College |
PA
PA |
US
US |
|
|
Assignee: |
THE PENN STATE RESEARCH
FOUNDATION
University Park
PA
|
Family ID: |
50929732 |
Appl. No.: |
14/053066 |
Filed: |
October 14, 2013 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61737282 |
Dec 14, 2012 |
|
|
|
Current U.S.
Class: |
216/67 ;
156/345.29 |
Current CPC
Class: |
H01J 37/32357 20130101;
H01J 37/3244 20130101; H01J 37/32715 20130101; H01J 2237/3345
20130101; H01L 21/31116 20130101; B81C 99/0025 20130101; H01J
37/32449 20130101; H01L 21/3065 20130101; H01J 2237/3341 20130101;
C03C 15/00 20130101 |
Class at
Publication: |
216/67 ;
156/345.29 |
International
Class: |
B81C 1/00 20060101
B81C001/00; B81C 99/00 20060101 B81C099/00 |
Claims
1. A system for reactive ion etching (RIE) system of a material,
comprising: a plasma chamber comprising a plasma source and a gas
inlet; and a diffusion chamber comprising a substrate holder for
supporting a substrate with a surface comprising the material and a
gas diffuser; and a source of at least one processing gas coupled
to the gas diffuser, wherein at least one radical of the at least
one processing gas is reactive with the material to perform etching
of the material, wherein the substrate holder is configured to
support the substrate within a processing region of the diffusion
chamber, wherein the gas diffuser is configured to introduce the at
least one processing gas into the processing region, and wherein
the substrate holder comprises an electrode that can be selectively
biased to draw ions generated by the plasma source into the
processing region to interact with the at least one processing gas
to generate the at least one radical at the surface.
2. The RIE system of claim 1, wherein the plasma source comprises a
high density plasma source.
3. The RIE system of claim 1, wherein the high density plasma
source comprises one of an inductively coupled plasma source, a
capacitively coupled plasma source, magnetically enhanced plasma
source, or a helicon plasma source.
4. The RIE system of claim 1, wherein the gas diffuser comprises a
diffuser ring.
5. The RIE system of claim 1, wherein the diffuser ring is attached
to the substrate holder.
6. The RIE system of claim 1, wherein the gas diffuser comprises a
plurality of nozzles.
7. The RIE system of claim 1, wherein the gas inlet is coupled to a
first gas delivery system and the gas diffuser is coupled to a
second gas delivery system.
8. The RIE system of claim 7, wherein the first gas delivery system
is configured for delivering at least one gas comprising at least
one of fluorine or chlorine, Ar, O.sub.2, or any combinations
thereof and the second gas delivery system is configured for
delivering NF.sub.3.
9. The RIE system of claim 8, wherein the at least one gas
comprising at least one of fluorine or chlorine comprises at least
one of SF.sub.6, C.sub.4F.sub.8, CH.sub.4, or NF.sub.3.
10. The RIE system of claim 8, wherein the first gas delivery
system is configured for delivering Ar and NF.sub.3.
11. A method for reactive ion etching, the method comprising:
forming a plasma at a first location; introducing at least one
processing gas at a second location corresponding to a surface to
be etched, wherein at least one radical of the at least one
processing gas is reactive with the surface to perform etching of
the surface; and directing ions from the plasma to a substrate at
second location to interact with the at least one processing gas to
generate the at least one radical at the surface.
12. The method of claim 11, wherein forming the plasma comprises
generating the plasma using a high density plasma source.
13. The method of claim 12, wherein the high density plasma source
is selected from an inductively coupled plasma source, a
capacitively coupled plasma source, magnetically enhanced plasma
source, or a helicon plasma source.
14. The method of claim 11, wherein introducing the at least one
processing gas comprises providing a gas diffuser configured to
direct the at least one processing gas into the second
location.
15. The method of claim 12, wherein the gas diffuser comprises a
diffuser ring.
16. The method of claim 12, wherein the gas diffuser comprises a
plurality of nozzles.
17. The method of claim 11, wherein the forming of the plasma
comprises: directing at least one other processing gas into a
plasma chamber; generating the plasma by exciting the at least one
processing gas in the plasma chamber.
18. The method of claim 17, further comprising selecting the at
least one other processing gas to comprise at least one gas
comprising at least one of fluorine or chlorine, Ar, O.sub.2, or
any combinations thereof and selecting the at least one processing
gas introduced at the second location to comprise NF3.
19. The method of claim 18, wherein the at least one gas comprising
at least one of fluorine or chlorine comprises at least one of
SF.sub.6, C.sub.4F.sub.8, CH.sub.4, or NF.sub.3.
20. The method of claim 17, selecting the at least one other
processing gas to comprise Ar and NF.sub.3.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims the priority to and the benefit of
U.S. Provisional Patent Application No. 61/737,282, entitled
"ULTRA-HIGH SPEED ANISOTROPIC REACTIVE ION ETCHING" and filed Dec.
14, 2012, the contents of which are herein incorporated by
reference in their entirety.
FIELD OF THE INVENTION
[0002] The present invention relates to reactive ion etching (RIE),
and more specifically to apparatus and methods for ultra-high speed
anisotropic RIE.
BACKGROUND
[0003] Silicon dioxide in its crystalline form (quartz) as well as
its amorphous form (glass) is finding increasing applications in
microsystems, as active resonator structure as well as passive
support and packaging components. Recently borosilicate glass and
quartz substrates have been etched with high aspect ratios and high
surface smoothness using SF6 and Ar/Xe gases [1-3]. The main focus
of these etch process development has been the achievement of high
etch rates and high aspect ratio etching of silica. In this context
the processes developed thus far rely upon ion bombardment to
accelerate the etching process while fluorine based gases are used
to provide the reactive component for etching. The use of heavier
Xe helps reduce the re-deposition and more effectively removes any
non-volatile residues resulting in smoother surfaces with an
average surface roughness of .about.2 nm.
[0004] In spite of more than an order of magnitude increase in the
etch rate of silicon dioxide compared to comparable processes used
in CMOS industry, the currently obtained etch rates are not
attractive enough for MEMS through wafer and high aspect ratio
etching of glass substrates. Considering that typical glass
substrates are 100-500 .mu.m, these etches can take between 200
(.about.3 hours) and 1000 minutes (.about.16 hours) at a rate of
.about.0.5 .mu.m/min making such processes impractical for glass
based device development and their commercialization. Etch
processes that can potentially break through this etch rate
limitation for glass can dramatically affect several MEMS
devices--including inertial and microfluidic devices. Thus, the
question to ask is what is limiting the etch rate of silica at just
under the 1 .mu.m/min.
[0005] To date, efforts in silicon dioxide (glass) etching, have
been primarily directed towards realizing features for
microelectronics applications such as interconnect vias [5],
waveguides [6], phase shift masks [7], etc. Hence, process
optimization has traditionally aimed at increasing the selectivity
of silicon dioxide over silicon substrate [8], reducing gate oxide
damage [9], decreasing sidewall roughness [10], and increasing
sidewall angle of the etched features [11]. With the advent of
microelectromechanical systems (MEMS) and microsystems in the last
decade, focus has shifted to high aspect ratio etching of silicon
dioxide for applications in microfluidics [12], microsensors [13],
and lab-on-a-chip applications. Many of these applications require
greater than 100 .mu.m of silicon dioxide (glass) etching while
maintaining the surface finish, with RMS surface roughness of less
than 5 nm [14, 15]. Hence, these applications impose additional new
requirements on borosilicate glass etching processes such as high
etch rate, high selectivity to masking material, high anisotropy,
low surface roughness for mirror polish, uniformity of etch across
the wafer and within a pattern [16], etc.
[0006] Traditional RIE processes are limited by the fact that the
substrate power and RF plasma power are coupled to each other often
resulting in etch non-uniformity across the wafer, low density of
plasmas, and limited control over the processing conditions.
However, in an inductively coupled plasma (ICP) RIE system, the
substrate power and the coil (source) power are independent of each
other thus providing excellent control over plasma density
(controlled by ICP power) and energy of etchant ions (controlled by
substrate power) [17]. As a result, plasma can be generated even at
relatively low pressures in the range of 10.sup.-4 Ton to 10.sup.-3
Torr. However, at such low pressures, the plasma in traditional RIE
systems is not stable.
[0007] Nonetheless, low processing pressure is advantageous for
rapid removal of etching products from the surface and also for the
removal of stray particles generated from the masking material,
substrate holder and walls of the reaction chamber. The presence of
stray particles results in micro-masking wherein the
micro-particles or reaction products on the substrate shield the
surface from the etchant species resulting in surface roughness,
micro-trenching and formation of plateau-like structures.
Additionally, the increased mean free path at low pressures
improves the anisotropy of the etch by minimizing the randomizing
collisions between the radicals, ions and other plasma species.
Borosilicate glass substrates are known to have a typical
composition of SiO.sub.2 (79.6%), B.sub.2O.sub.3 (12.5%), Na.sub.2O
(3.72%), Al.sub.2O.sub.3 (2.4%), and K.sub.2O (0.02%).
[0008] In the case of deep reactive ion etching of silicon dioxide
(quartz or borosilicate glass) a high Ar to SF.sub.6 ratio is
required to maintain low RMS surface roughness. FIG. 1 shows the
dependence of the etch rate and RMS surface roughness as a function
of substrate RF power, chamber pressure, Ar, and SF.sub.6 flow
rate. In all cases the pressure in the chamber was maintained at
0.26 Pa throughout the flow ranges. The ICP source power was 2000 W
and a substrate bias power of 475 W (Bias Voltage of 80 V) was used
in generating these results. From the graphs it can be seen that
the best surface roughness of .about.2 nm is obtained at high Ar
flow rates, low chamber pressure, and high substrate
power--corresponding to conditions dominated by physical sputtering
of the material. The etch rate can be increased by increasing the
SF.sub.6 flow rate from 5 sccm to 50 sccm from 0.54 .mu.m/min to
0.74 .mu.m/min however the surface roughness was found to degrade
under these conditions to >100 nm. Pulse electroplated nickel is
typically used as the etch mask layer and a selectivity of
.about.25:1 can be obtained for SiO.sub.2 etching under these
conditions. FIG. 1(e) shows an SEM of a high aspect ratio feature
etched in quartz using these conditions. Similar results were
obtained by Li et al. while etching SiO.sub.2 using Xe instead of
Ar. The higher sputter yield of Xe gave a lower RMS surface
roughness value as compared to Ar for the same mole fraction of the
inert gas in SF.sub.6. Although silicon grease or a small drop of a
fluoropolymer based oil, such as FOMBLIN manufactured by Solvay S.
A. of Belgium, Brussels, can be used for mounting the quartz/glass
substrates onto a 4'' silicon carrier wafer, these materials cannot
withstand the long process times and can leave the backside of the
sample with hard to remove, stubborn residues. Furthermore, these
mounting materials do not provide a reliable and uniform thermal
contact, between the carrier wafer and the sample, throughout the
entire etch process. In order to avoid these problems, indium
solder can be used for mounting the sample directly onto a silicon
wafer. However, the mounting side of the SiO.sub.2 sample needs to
be coated with 20/80 nm of Cr/Au to provide a surface to which the
solder can adhere. Of course if the sample is large enough it can
be directly mechanically clamped or an electrostatic chuck can be
used for the mounting of the sample. In all cases the backside of
the chuck/substrate is cooled using helium gas maintained at the
desired temperature.
SUMMARY
[0009] Embodiments of the invention concern systems and methods for
RIE, and more specifically to apparatus and methods for ultra-high
speed and ultra-high smooth anisotropic RIE of quartz and other
glasses. In some embodiments, the apparatus and methods can be
selected to provide etched glass surface with excellent
smoothness.
[0010] In a first embodiment of the invention, a system for
reactive ion etching (RIE) system of a material is provided. The
system includes a plasma chamber comprising a plasma source and a
gas inlet. The system also includes a diffusion chamber comprising
a substrate holder for supporting a substrate with a surface
comprising the material and a gas diffuser. The system further
includes a source of at least one processing gas coupled to the gas
diffuser. In the system, at least one radical of the at least one
processing gas is reactive with the material to perform etching of
the material. Further, the substrate holder is configured to
support the substrate within a processing region of the diffusion
chamber. Additionally, the gas diffuser is configured to introduce
the at least one processing gas into the processing region and the
substrate holder comprises an electrode that can be selectively
biased to draw ions generated by the plasma source into the
processing region to interact with the at least one processing gas
to generate the at least one radical at the surface.
[0011] In a second embodiment of the invention, a method for
reactive ion etching is provided. The method can include forming
plasma at a first location and introducing at least one processing
gas at a second location corresponding to a surface to be etched,
wherein at least one radical of the at least one processing gas is
reactive or enables reactivity with the surface to perform etching
of the surface. The method can further include directing ions from
the plasma to a substrate at second location to interact with the
at least one processing gas to generate the at least one radical at
the surface.
[0012] In the various embodiments, the forming of the plasma can
include generating the plasma using an inductively coupled plasma
source. Further, the introducing of the at least one processing gas
can include configuring the gas diffuser to direct the at least one
processing gas into the second location. In some configurations,
the gas diffuser can be a diffuser ring. Alternatively, the gas
diffuser can be a plurality of nozzles.
[0013] In the various embodiments, the forming of the upstream
plasma can include generating the plasma using a magnetically
enhanced plasma source or a helicon source, a capacitively coupled
plasma source, or any other source of high density plasma.
Alternatively, the forming of the upstream plasma can include
generating the plasma using a non-high density source, such as a
parallel plate capacitor plasma source, an inductively coupled
plasma source, a microwave plasma source, an ionization chamber, or
any other source of low to normal density plasma. Further, the
introducing of the at least one processing gas can include
configuring the gas diffuser to direct the at least one processing
gas into the second location. In some configurations, the gas
diffuser can be at least one diffuser ring. Alternatively, the gas
diffuser can be a plurality of nozzles.
BRIEF DESCRIPTION OF THE DRAWINGS
[0014] FIG. 1 is a schematic of a system in accordance with the
various embodiments;
[0015] FIG. 2 is a schematic of the system of FIG. 1 utilizing a
diffuser ring;
[0016] FIG. 3 shows a modified etching system in accordance with
the various embodiments;
[0017] FIG. 4 shows a ring diffuser with 1 mm holes in accordance
with the various embodiments;
[0018] FIG. 5A is a plot of NF3 flow rate through the ring diffuser
versus etch rate for a system in accordance with the various
embodiments;
[0019] FIG. 5B is a plot of NF3 flow rate through the ring diffuser
versus smoothness for a system in accordance with the various
embodiments;
[0020] FIG. 6A is an AFM image of the smoothest substrate obtained
with an ring active etch in accordance with the various
embodiments;
[0021] FIG. 6B is an SEM image of the etch region cross section for
the substrate of FIG. 6A;
[0022] FIG. 7A shows residual gas analyzer data for two methods,
including a method in accordance with the various embodiments;
and
[0023] FIG. 7B shows a detailed view of a portion of the data of
FIG. 7A.
DETAILED DESCRIPTION
[0024] The present invention is described with reference to the
attached figures, wherein like reference numerals are used
throughout the figures to designate similar or equivalent elements.
The figures are not drawn to scale and they are provided merely to
illustrate the instant invention. Several aspects of the invention
are described below with reference to example applications for
illustration. It should be understood that numerous specific
details, relationships, and methods are set forth to provide a full
understanding of the invention. One having ordinary skill in the
relevant art, however, will readily recognize that the invention
can be practiced without one or more of the specific details or
with other methods. In other instances, well-known structures or
operations are not shown in detail to avoid obscuring the
invention. The present invention is not limited by the illustrated
ordering of acts or events, as some acts may occur in different
orders and/or concurrently with other acts or events. Furthermore,
not all illustrated acts or events are required to implement a
methodology in accordance with the present invention.
[0025] The various embodiments are directed to etch apparatus and
methods that achieve glass high etch rates, at least in the 1-100
.mu.m/min range, for glass substrates resulting in another order of
magnitude improvement in the etch rate of glass.
[0026] In 2004, Yamakawa et al. [19] demonstrated a very high speed
etching of silicon dioxide (BPSG) film using microwave excited
non-equilibrium atmospheric pressure plasma. The authors were able
to demonstrate an ultra-high etch rate of SiO.sub.2 (14 .mu.m/min)
and an unprecedented selectivity of 200 with respect to silicon
using NF.sub.3/He along with addition of H.sub.2O as the etching
gas. In this configuration, the inventors note that the fast etch
rate can be attributed to the presence of NF.sub.x radicals arising
from the breakup of NF.sub.3 into NF.sub.x*+F*--radicals.
NF.sub.x--radicals (with x>0). These radicals are extremely
aggressive towards SiO.sub.2 and other dielectrics (and polymers)
and can etch them at extremely high speed. The addition of water
vapor (H.sub.2O) consumes the fluorine radicals by HF formation
(2F*+H.sub.2O.fwdarw.2HF+<0>) whereas NF.sub.x radicals are
much less scavenged. Accordingly, water vapor therefore acts as a
selective fluorine radical scavenger and consequently reduces any
undesirable attack of silicon. However, if NF.sub.3 is completely
broken down, e.g., as in high-density, high power plasma, then the
plasma primarily consists only of fluorine and nitrogen radicals.
Under these conditions, a complete quenching of the fast etch rate
of SiO.sub.2 and other dielectrics has been observed. The complete
breakup of NF.sub.3 molecules under appropriate plasma conditions
can also confirmed by the plasma color which changes from a deep
red to a bluish color, indicating a change in preponderance from
NF.sub.x to fluorine radicals in the plasma.
[0027] Thus, the inventors note that a main requirement for
speeding up glass etch rate requires the creation of NF.sub.x
radicals. However, to date no reports of such high etch rates in
high density plasma sources such as ICP-RIE systems exist. This is
mainly due to the fact that these conditions cannot be readily
reproduced in an ICP-RIE system. In an ICP-RIE process, the
pressure is typically too low (.about.1.times.10.sup.-3 mbar) and
the coupling of high RF power results in a complete breakdown of
the NF.sub.3 into nitrogen and fluorine radicals. The high density
of plasma achieved in ICP chambers results in a very efficient
transfer of energy between the RF source and the gas in the chamber
and essentially breaks up the molecules into individual atomic
radicals and ions. To overcome this limitation and to achieve the
incomplete breakup of the NF.sub.3 molecules, high chamber pressure
is required. However, even if the pressure is increased a 100
times, conventional ICP-RIE processes still cannot achieve the
results reported in microwave plasma. Furthermore, at these
pressures a high mismatch between the source and chamber load
occurs and the typical tuning circuits available on the etching
equipment are unable to tune. As a result, any resulting plasma is
typically extinguished.
[0028] Although the various embodiments will be described primarily
with respect to an ICP system, this is solely for illustrative
purposes. That is, the various embodiments are not limited to any
particular high density plasma source. For example, the high
density plasma can be generated using a magnetically enhanced
plasma source, a helicon plasma source, or a capacitively coupled
plasma source, to name a few. However, other methods for generating
high density or any other source of plasma can be considered as
suitable for this application. Thus, an ICP source described below
with respect to the various figures can be substituted with any
other type of high density or ordinary plasma source.
[0029] Based on the foregoing, the inventors have determined that
in order to exploit the most of the ICP-RIE etchers for glass
etching the following conditions are required: [0030] a. Operation
of the ICP-RIE etcher under its normal operating pressure and power
range to obtain efficient plasma conditions. [0031] b. Providing
for conditions within the chamber to create incomplete breakup of
the NF.sub.3 molecules to achieve the high etch rate. [0032] c.
Achieving control over the etch rate, etch smoothness, and
anisotropy of the structure independently of the plasma source
conditions. [0033] d. Introduction of water molecules into the
chamber either through the nozzle or any other means in the
proximity of the substrate.
[0034] In view of the foregoing, the various embodiments of the
invention are directed to new systems and methods for etching of
glass using NF.sub.3 gas using ICP-RIE etching. In particular, new
systems and method that meet the goals listed above during etching
of glass. In the various embodiments, these goals are achieved by
incorporating a gas diffuser in the vicinity of the wafer or other
material to be etched to provide a localized source of NF.sub.3 gas
in the ICP-RIE reactor. For example, in one embodiment, a stainless
steel gas diffuser ring can be provided on the metal plate on the
mechanical clamping plate of the etcher. In particular, the gas
diffuser is positioned in the ICP-RIE reactor so that the gas
diffuser dispenses NF.sub.3 and/or H.sub.2O and/or any other gas of
interest right above the wafer.
[0035] In operation, the etch process is imitated in a conventional
manner. That is, conventional etch process gases are introduced
into the ICP-RIE reactor (e.g., Ar, SF.sub.6) and a plasma is
generated, remote from the wafer, using normal operating
conditions. Concurrently, NF.sub.3 gas is introduced in the
vicinity of the wafer using the gas diffuser. A substrate bias
power is applied during this process, which drives the high energy
ions generated in the plasma towards the NF.sub.3 gas in the
vicinity of the wafer. These ions then interact with the NF.sub.3
gas to break it down into NF.sub.x radicals. Effectively, this
creates a high pressure plasma in the vicinity of the wafer inside
a low pressure ICP-RIE chamber. This effective high density
NF.sub.3 plasma with the incomplete breakdown of the gas molecules
is then used to achieve a high etch rate of glasses. In particular,
etch rates of 1 .mu.m/min range or higher for glass substrates. In
other words, etch rates that are at least an order of magnitude
than conventional methods.
[0036] In another variation, H.sub.2O is introduced in the
proximity of the substrates/wafer to enhance the reactivity of the
NF.sub.3 radicals with the glass substrate. Water molecules can
react with the NF.sub.x and F radicals and scavenge them to create
the right etching mixture. Furthermore, water can also help in
wetting the glass surface and initiate and mediate reactions on the
surface through the interactions with various radicals and ions
thus formed in the chamber.
[0037] Any sequence of gas introduction through the diffuser ring
can be considered. For example, alternate introduction of NF.sub.3
gas followed by H.sub.2O gas, or simultaneous introduction of the
two gases, or concurrent introduction using two independent nozzles
or diffuser rings. Neither gas is limited to NF.sub.3 or H.sub.2O
only and any gases that contain active species relevant for glass
etching can be considered in this arrangement.
[0038] In the various embodiments, the energy of the bombarding
ions from the plasma can be effectively controlled by the substrate
bias which provides an independent control to change the etching
conditions for tuning the composition of the NF.sub.x radicals as
well as to control the anisotropy. Further, by controlling the
plasma gas composition from the gas diffuser, in-situ reactions and
gas chemistries in the vicinity of the wafer can be increased or
decreased as needed.
[0039] FIG. 1 shows a schematic illustration of an ICP-RIE system
100 in accordance with the various embodiments. The system 100
includes many components of a conventional ICP-RIE system. For
example, the system 100 can include a plasma chamber 102, serving
as a source of inductively coupled plasma or ICP source. The plasma
chamber 102 is coupled to a diffusion chamber 104 lined with
magnetic portions 106, in which the wafer to be etched is disposed
and into which ions from the ICP source are directed towards the
wafer to perform the etching. The plasma chamber 102 can include a
gas inlet 108 for introducing gases into the plasma chamber 102.
This gas inlet can be coupled to a gas delivery system (not shown)
for delivering one or more gases. For example, the gas delivery
system can be configured to delivery of SF.sub.6, C.sub.4F.sub.8,
Ar, O.sub.2, CH.sub.4, CHF.sub.3, CF.sub.4, Cl.sub.2, Xe, Ne,
N.sub.2, or NF.sub.3, or any combinations thereof. The plasma
chamber 102 can also be associated with an antenna or coil 110, an
RF matching or tuning network 112 and an RF power supply 113 to
provide the energy for generating the plasma in the plasma chamber
102.
[0040] The diffusion chamber 104 can include a substrate holder and
clamp 114 for supporting and securing a wafer. As shown in FIG. 2,
the substrate holder 114 can be configured such that the position
(i.e., the height) of the wafer in the diffusion chamber can be
changed as required to reduce or increase the distance between the
plasma chamber 102 and the wafer. During etching, the temperature
of the wafer can be controlled using back side Helium cooling lines
116 coupling the substrate holder 108 to a chiller (not shown). The
He cooling lines can be configured to allow temperatures between
-20-50 .degree. C. The substrate holder and clamp 114 can be
further configured to serve as an electrode coupled to a second RF
generator or power supply 116 via a second RF matching/tuning
network 115 to provide substrate bias. This second RF generator 116
and second RF matching/tuning network 115 can be configured to be
decoupled from and operate independently of the RF generator 113
and RF matching tuning network 112 associated with the plasma
chamber 102.
[0041] Although FIG. 1 and other descriptions refer to discrete
plasma and diffusion chambers, the various embodiments are not
limited in this regard. That is, in some configurations, a single
chamber can be provided with different regions to serve as the
plasma chamber 102 and the diffusion chamber 104.
[0042] In addition to the foregoing, the system 100 can include a
gas diffuser 118 coupled to a gas delivery system (not shown) via
feedthroughs 120 for introducing NF.sub.3 gas and other gases,
separately or in combination, into the diffusion chamber 104 in the
vicinity of substrate holder 114 when in the elevated position.
Such other gases can include, but are not limited to H.sub.2O. In
the various embodiments, the gas delivery system for the gas
diffuser 118 can be the same or different than the gas delivery
system for inlet 108. However, providing a separate gas delivery
systems for inlet 108 and gas diffuser 118 can be advantageous.
That is, separate systems provide greater flexibility and control.
For example, the mixture of gases introduced via inlet 108 and gas
diffuser 118 can be different.
[0043] In the various embodiments, the gas diffuser 118 can be
implemented in a variety of ways. In some embodiments, as
illustrated in FIG. 2, the gas diffuser 118 can be implemented as
diffuser ring 218, flexible feedthroughs 220 coupling the diffuser
ring 218 to a gas panel 222, and an NF3 gas source 224 coupled to
gas panel 222. In one particular embodiment, the diffuser ring 214
can be constructed using a 1/8 inch electro-polished stainless
steel tube drilled with 0.5 mm holes at a distance of 10 mm bent
into a 4'' circular tube. Thereafter this tube can be welded onto
the substrate holder and clamp 108. The tube can be coupled via
flexible gas tubing, serving as flexible feedthroughs 220, to the
gas panel 222 to permit gas delivery and to account for the
different positions of the substrate holder and clamp 108. For
example, stainless steel flexible gas tubing using VCR fittings can
be used to bring the process gas to the diffuser ring 218.
Additionally one or more the flexible feedthroughs 220 can include
non-flexible portions or segments 221, as shown in FIG. 3. Further,
additional tubing 226 can be provided to connect the various
components described above.
[0044] Although the discussion above describes that the gas
diffuser 114 is implemented as a diffuser ring, the various
embodiments are not limited in this regard. Rather, any other
structure can be used in the various embodiments. For example, the
gas diffuser 118 can be implemented using individual nozzles
disposed in the vicinity of the substrate. In another example, gas
diffuser 118 can be implemented using multiple sections of tubing
with holes or openings. Further, although the diffuser ring 214 is
described as containing holes or openings disposed substantially
along a same plane parallel to that of the wafer, having a
particular size, and having a particular spacing, the various
embodiments are not limited in this regard. Rather, a gas diffuser
in accordance with the various embodiments can provide any
placement, spacing, size, or other arrangement of holes or openings
for introducing gases into diffusion chamber 104. Similarly, other
configurations of gas diffuser 118 can be configured with a wide
array of geometries.
[0045] The gas panel 222 can also be configured in a variety of
ways. In one particular embodiment, a precision needle valve can be
used to control the gas flow rate of NF.sub.3 (or other gases) into
the diffusion chamber 104. In one particular embodiment, a NF.sub.3
flow rate of 300 sccm or less can be used. However, any other
methods for controlling gas flow rate can be used in the various
embodiments.
[0046] In one particular embodiment, a precision needle valve can
be used to control the gas flow rate of H.sub.2O (or other gases)
into the diffusion chamber 104. In one particular embodiment, a
H.sub.2O flow rate of 300 sccm or less can be used. However, the
flow rates and methods for controlling gas flow rate can vary in
the various embodiments. In addition to the foregoing, the system
100 can include additional modifications compared to conventional
ICP-RIE systems. For example, plasma shielding plates may need to
be modified to accommodate the additional components for the system
100. In some embodiments, the system 100 can also be coupled to a
mass-spectrometer system (e.g., a residual gas analyzer) to measure
the gas species created within the chamber to monitor and adjust
the etching process. Additionally, optical windows can be provided
to monitor the visible/UV spectra of the plasma and evaluate the
plasma glow composition during etching.
[0047] The modifications described above allow an ICP-RIE reactor
in accordance with the various embodiments to be used in a variety
of modes. A first mode of operation is simply to utilize the
ICP-RIE reactor to perform etching under normal operating
conditions. That is, to perform conventional (i.e., low etch rate)
etch processes. For such normal operations, the plasma etching
occurs by generating a plasma using an etching gas chemistry (e.g.,
NF.sub.3 and Ar) without introducing gases through the gas
diffuser. That is, referring back to FIG. 1, gases are introduced
view inlet 108 and a plasma 130 is generated in plasma chamber 102.
Concurrently, a substrate bias is provided at substrate holder and
clamp 114, which causes ions generated in the plasma 130 to be
directed towards substrate holder and clamp 114 and any wafer
disposed thereon. The ions are then used to perform conventional
etching. Such a mode can be utilized when low etch rates are
desirable to accurately.
[0048] A second mode is to utilize the gas diffuser to generate
enhanced etching conditions. In such a mode, the system can be
initially setup in the first mode, as described above. Thereafter,
the etch gas (NF.sub.3) can be introduced into the ICP-RIE reactor
via the gas diffuser and the enhanced etching (high etch rate) will
be performed. That is, as described above, the plasma 130 is
generated and the substrate bias is used to direct ions from the
plasma 130 towards the wafer. However, via gas diffuser 118, a
region 132 with an increased density of NF.sub.3 gas, localized
around the wafer, is provided. The interaction of the ions from the
plasma 130 with the NF.sub.3 132 in region 132 causes the
generation of NF.sub.x radicals. That is, the ions are utilized to
break up the NF.sub.3 in region 132 to provide the incomplete
breakup of NF.sub.3. This results in effectively forming a high
density plasma of NF.sub.x radicals in the vicinity of the wafer.
Consequently, the presence of a high concentration of these
NF.sub.x radicals enhances the etch rate.
[0049] Although the various embodiments are described with respect
to the enhanced etching of glasses using a NF.sub.3-based chemistry
in an ICP-RIE reactor modified to include a gas diffuser, the
various embodiments are not limited in this regard. Rather, the
systems and methods described herein can be applied to the etching
of any other types of materials and using other types of
chemistries.
EXAMPLES
[0050] The following examples and results are presented solely for
illustrating the various embodiments and are not intended to limit
the various embodiments in any way.
[0051] Smooth dry etching of some types of borosilicate glass
wafers has been a challenge due to nonvolatile or less volatile
metal-halogen compounds forming on the surface of the etched area.
Thus, the methods of the various embodiments have been evaluated.
In particular, to evaluate the methods of the various embodiments,
two alternative etching method were investigated in regards of
smooth and fast etching; i) Conventional Ar/NF.sub.3 downstream
plasma etching ii) Ar/O2/NF.sub.3 plasma etching where an
inductively coupled plasma reactor is modified such that a ring
shape diffuser system is adapted to the substrate holder in order
to supply high local etchant flux, NF.sub.3, over the substrate
surface. The latter approach, in accordance with the various
embodiments, reduces the full dissociation of NF.sub.3 in
NF.sub.3/AR and O2 based plasma and resulted in forming more
reactive species in the plasma RGA (Residual gas analysis) show
that the diffuser system allows to form larger amounts of NF.sub.x
radicals in the plasma. Further, ex-situ X-ray Photoelectron
Studies (XPS) showed non-volatile metal alkali compounds are formed
on the surface on rough surfaces, with NaF being the prominent
compound for the Schott Borofloat glass. For smooth surfaces, less
alkali compounds are observed. The results demonstrated that
ultra-smooth etch surfaces can be obtained within a window of
parameters, where the arithmetic mean smoothness (Ra) varies from
3.4 .ANG. to 8 .ANG. at considerable etch rates of 0.343 .mu.m/min
to 0.55 .mu.m/min.
[0052] I. Introduction
[0053] One of the most important and challenging parameters in
micromachining of Borofloat glass is obtaining smooth surfaces at
considerable etch rates [13]. Despite the fact that, controlled
surface roughness is shown to be useful for some specific
applications [14], surface roughness in general is highly unwanted
in most of the applications such as optical applications and
micro-electro-mechanical systems (MEMS) [15]. If the surface is
rough and wavy, the light is scattered in optical applications.
Therefore, high precision optical applications require close to
atomically flat surface roughness (Ra.about.0.5-1 nm). For micro
machined resonators and sensors, roughness negatively affects the
quality factor and thus decreases the fabricated device
performance. The causes of increased roughness are due to the
impurities exist in glass substrates and due to the less percent
metal containing compounds other than SiO.sub.2, where these metal
based compounds form non-volatile or less volatile metal halogens
in fluorine containing plasmas (i.e., Argon/SF.sub.6) during dry
etching. For instance, Borosilicate glass, is mainly composed of
81% of SiO.sub.2, 13% of B.sub.2O.sub.3, 4% of Na.sub.2O/K.sub.2O
and 2% of Al.sub.2O.sub.3. The non-volatile metal/alkali metal
halogens (i.e., NaF and AlF.sub.3) forming on the surface cause
micro-masking in the etch regions resulting in the formation of
needle-like structures and therefore destructing the smoothness. As
a solution, highly energetic argon ions are proposed as the
physical etch mechanism to remove these compounds from the surface
to achieve better roughness, but as the argon percent ratio in the
gas mixture is increased, the etch rate decreases significantly
[16]. Furthermore, the arithmetic mean roughness (Ra) is still
reported on the range of 2 nm to 100 nm in these cases
[16]-[19].
[0054] II. Etching mechanism
[0055] In this study, an etching system in accordance with the
various embodiments was utilized in order to achieve smoothness at
considerable etch rates. A radio frequency (RF) inductively coupled
plasma etching system (Alcatel AMS 100, manufactured by
Alcatel-Lucent of Paris, France) is modified, where a ring diffuser
mechanism is attached to the substrate holder. The whole system and
the ring are shown in FIGS. 3 and 4. FIG. 3 shows a modified
etching system in accordance with the various embodiments and FIG.
4 shows a ring diffuser with 1 mm holes in accordance with the
various embodiments. The ring has a diameter of 9.6 cm and has
diffuser holes that are placed within 1 mm spacing.
[0056] In most of the ICP applications and in the case of
unmodified etching system chemical and physical etchants are
introduced via the gas inlets located at the top of the ICP chamber
[20]. The plasma is formed within this chamber and directed towards
the substrate in a downstream fashion via the applied electric
field between the source and the substrate. The main purpose of the
ring diffuser proposed here is to enhance the reactive ion etching
via establishing high local flux of reactive species on etch face
of the wafer, combining the downstream flux with the local ring
flux. Before characterizing the effect of ring design on
smoothness, etch rate and hard mask selectivity over substrate,
classical downstream etches using Ar and NF.sub.3, where the ring
is inactive, are performed in sake of determining the etch
performance. A set of experiments utilizing design of experiments
involving orthogonal arrays are carried out to obtain optimum
conventional ring-inactive downstream etching parameters using
Ar/NF.sub.3 plasma (Please refer to the supplemental information
for the detailed experimentation). Double-side polished, 4 inches
Borosilicate wafers are used as the substrates during all
experiments performed. The wafers are patterned with 100 um, 250
um, 500 um and 1 mm rectangular openings. 2-3 um thick nickel is
plated on the wafers, acting as hard mask during the etch. Each
sample is etched for 1 hour at 20.degree. C. substrate temperature.
The results for fastest and smoothest Ar/NF.sub.3 etches are
demonstrated in Table 1.
TABLE-US-00001 TABLE 1 Baseline results for AR/NF.sub.3
conventional ICP plasma. ICP Substrate Stage Ar flow NF3 flow
Smoothness power Power position Pressure rate rate Etch rate (Ra)
Fastest 2000 W 400 W 120 mm 6.5 .times. 10.sup.-6 bar 10 sccm 30
sccm 0.41 .mu.m/min 79.6 nm Smoothest 2000 W 400 W 120 mm 12
.times. 10.sup.-6 bar 30 sccm 12.5 sccm 0.31 .mu.m/min 28.1 nm
Summarized results from Table 1 are taken into consideration to
start experimental design of the ring active experiments. For the
ring-active experiments, 20 sccm of Ar and 10 sccm of O.sub.2 are
introduced from top into the ICP chamber. The ICP power, substrate
power and the stage position are all held constant with the ones
stated in Table 1. The pressure is not regulated. Various flow
rates of NF.sub.3 are feed through the ring diffuser. FIGS. 5A and
5B shows the etch rate and recorded operating pressure along with
the smoothness in the chamber when the ring is active. The etch
depths are measured via a Tencor P16+ profilometer, manufactured by
KLA Tencor of Milpitas, Calif. and the roughness is measured via a
PSIA XE-100 Scanning Probe Microscope, manufactured by Park Systems
Corporation of Santa Clara, Calif.
[0057] As the ring is utilized and the NF.sub.3 flow is kept
between 20-50 sccm, high etch rates and ultrasmooth surfaces are
obtained. The maximum etch rate, 0.55 .mu.m/min, is achieved at 20
sccm Ar/10 sccm O2/48 sccm NF.sub.3 flows. The best smoothness,
Ra=3.4 .ANG., is achieved at 20 sccm flow, where the wall slope was
82.4.degree.. An Atomic Force microscope (AFM) image and an
scanning electron microscope (SEM) image of the substrate are shown
in FIGS. 6A and 6B, respectively. The obtained smooth substrates
can be readily used for optical applications involving high
precision (Ra<5 .ANG.) and super precision (Ra<3 .ANG.). Even
at the highest etch rate, Ra is measured as 8.4 .ANG.. However, the
smoothness is heavily degraded once the ring flow rate is above 50
sccm. In this case, Ra is at least 2 orders of magnitude larger in
high flow regions where the chamber pressure is >6.3
.mu.Bar.
[0058] When compared to conventional downstream etching with
Ar/NF.sub.3 , where the results are tabulated in Table 1, the etch
rate with the ring is improved 34%. The nickel selectivity at
downstream Ar/NF.sub.3 plasma is measured as 7.35:1 over
Borosilicate glass. In ring active experiments of FIGS. 4A and 4B,
the selectivity is measured as 9.32:1 for the fastest etch rate,
indicating a 27% improvement.
[0059] The superior performance of one etching method among two
etching methods is mainly due to the diffuser mechanism. The aim to
have diffuser mechanism in place is to decrease the dissociation
rate of NF.sub.3 and obtain more NF.sub.x radicals in tha plasma.
It is previously stated that the low percent dissociation of
NF.sub.3 improves the etching speed of borosilicate glass [21]. In
order to determine the gas species forming in the plasma, a
Residual Gas Analyzer (Exton XT 100, manufactured by Extorr, Inc.
of New Kensington, Pa.) is used. FIGS. 7A and 7B show two RGA data:
for the ring active etch at 48 sccm NF.sub.3 flow rate and for
Ar/NF.sub.3 plasma with fastest etch rate shown in Table 1. From
FIGS. 7A and 7B, it is clearly observed that the plasma creates
considerable partial pressure of NF.sub.x (NF.sup.+ and
NF.sub.2.sup.+ at the peaks 33 and 52 respectively) radicals in the
plasma. In addition, NF.sub.3 at peak 71 has considerable partial
pressure. However, in conventional Ar/NF.sub.3 plasma, there is no
peak at NF.sub.3, proving a full dissociation in the gas. There are
less NF.sub.2 and NF radicals, indicating NF.sub.3 dissociated
mostly to N and F. The highly energetic Argon ions, which is
accelerating towards the wafer surface first interacts with a local
flux of NF.sub.3 gas that is just over the substrate. The
collisions between Argon and NF.sub.3 create high concentrations of
NF.sub.x. NF.sub.x radicals may act more aggressive to glass and
thus improve the etch rate. In addition, heavier gasses are formed
when the ring is active. The peak 84 may highly correspond to
NF.sub.2O.sub.2 gas. The authors articulate there may be
NF.sub.xO.sub.x based gas formation when NF.sub.3 is diffused
through the ring, which may be playing a critical role in
smoothness and fast etching. This is further discussed in
below.
[0060] As it is pointed out above, smoother surfaces are obtained
via the diffuser system. In order to understand the smoothness
achievement further and to determine the surface chemistry by
estimating the atomic composition of the surfaces, X-ray
Photoelectron Spectroscopy (XPS) is used. Three etching results are
compared in XPS study: i) the fastest etch obtained in Ar/NF.sub.3
etching system shown in table 1 ii) Ring active etch with 20 sccm
NF.sub.3 flow iii) Ring active etch with 80 sccm flow. All these
have 79.6 nm, 0.34 nm and 350 nm of Ra respectively. These 3 runs
are selected to compare two different etching methods and determine
the effect of ring diffuser system on roughness. As previous
studies in literature indicate, the non-volatile species negatively
affect the smoothness. Not only the smoothness is lost by these
species, but the etch rate also decreases (the degree of decrease
depending on the composition of the glass type) as the non-volatile
species concentration/accumulation on the substrate surface
increases during the etching [22]. Therefore, high resolution XPS
scans have been performed for Sodium (Na), Boron (B), Aluminum (Al)
and Potassium (K). Table 2 shows the atomic percent calculations of
3 etching results emphasized.
TABLE-US-00002 TABLE 2 Approximate atomic percent calculations from
XPS surveys for 3 different etch cases comparing two etching
approach. All values are in percent. Si 2s O 1s F 1s Na 1s B 1s Al
2s K 2s Ni 2p Ar/NF.sub.3 fastest etch 4.01 8.979 57.76 18.56 ND*
7.646 1.279 1.768 Ring active etch- 20 sccm 20.7 51.696 15.751
7.526 2.105 2.22 ND* ND* NF.sub.3 Ring active etch- 80 sccm 4.483
6.487 56.448 20.946 4.144 5.102 1.887 ND* NF.sub.3 Unprocessed
wafer 27.91 66.92 ND* 0.46 4.69 ND* ND* ND* *ND: Not Detectable
amount.
[0061] From approximate atomic percent calculations, it is observed
that the rough samples have high Al, K and Na percentage. However,
it is interesting to note that Na is the most prominent amongst
them. For the conventional Ar/NF.sub.3 etching and high flow ring
active etching have very high Na concentrations (>15%) at the
surface, whereas smooth ring active etch has 7.5% Na concentration.
When Ra roughness is considered, this much of difference in Na
percentages causes 2 orders of magnitude worse roughness. These
results infer that the smoothness is enhanced via ion enhanced
chemical reaction introduced by the ring diffuser system.
[0062] In conclusion, a system configured in accordance with the
various embodiments can be used to obtain smooth surfaces on
borosilicate glass and is integrated with a commercial ICP etching
system, where NF.sub.3 is supplied to the diffuser. The effects of
the mechanism are compared with conventional Ar/NF.sub.3 plasma
etching. Ultra-smooth etching is achieved with Ra<5 .ANG. at
considerable etch rates. The smoothness is degraded at high flows
of NF.sub.3 (>50 sccm). The main reason of this degradation is
due to the high percentages of Na (in NaF form) on the surface. At
smooth etches, the Na concentration is found out to be less in
atomic percent. The authors articulate that higher amount of
NF.sub.x radicals and possible existence NF.sub.xO.sub.x gasses
effectively remove Na from the surface in the light of RGA
readouts.
[0063] While various embodiments of the present invention have been
described above, it should be understood that they have been
presented by way of example only, and not limitation. Numerous
changes to the disclosed embodiments can be made in accordance with
the disclosure herein without departing from the spirit or scope of
the invention. Thus, the breadth and scope of the present invention
should not be limited by any of the above described embodiments.
Rather, the scope of the invention should be defined in accordance
with the following claims and their equivalents.
[0064] Although the invention has been illustrated and described
with respect to one or more implementations, equivalent alterations
and modifications will occur to others skilled in the art upon the
reading and understanding of this specification and the annexed
drawings. In addition, while a particular feature of the invention
may have been disclosed with respect to only one of several
implementations, such feature may be combined with one or more
other features of the other implementations as may be desired and
advantageous for any given or particular application.
[0065] The terminology used herein is for the purpose of describing
particular embodiments only and is not intended to be limiting of
the invention. As used herein, the singular forms "a", "an" and
"the" are intended to include the plural forms as well, unless the
context clearly indicates otherwise. Furthermore, to the extent
that the terms "including", "includes", "having", "has", "with", or
variants thereof are used in either the detailed description and/or
the claims, such terms are intended to be inclusive in a manner
similar to the term "comprising."
[0066] Unless otherwise defined, all terms (including technical and
scientific terms) used herein have the same meaning as commonly
understood by one of ordinary skill in the art to which this
invention belongs. It will be further understood that terms, such
as those defined in commonly used dictionaries, should be
interpreted as having a meaning that is consistent with their
meaning in the context of the relevant art and will not be
interpreted in an idealized or overly formal sense unless expressly
so defined herein.
REFERENCES
[0067] The following references describe concepts associated with
various aspects of the invention and are each incorporated by
reference in their entirety: [0068] [1] L. Li, T. Abe, and M.
Esashi, Journal of Vacuum Science & Technology B
(Microelectronics and Nanometer Structures), vol. 21, p. 2545,
2003. [0069] [2] X. Li, T. Abe, and M. Esashi, Proceedings of IEEE
Thirteenth Annual International Conference on Micro Electro
Mechanical Systems, Miyazaki, Japan, 2000, p. 271. [0070] [3] A.
Goyal, V. Hood, and S. Tadigadapa, Journal of Non-Crystalline
Solids, vol. 352, p. 657, 2006. [0071] [4] S. Queste, R. Salut, S.
Clatot, J.-Y. Rauch, G. Chantal, and M. Khan, Microsystems
Technology, vol. 16, pp. 1485-1493, 2010. [0072] [5] S. Karecki, R.
Chatterjee, L. Pruette, R. Reif, T. Sparks, L. Beu, V. Vartanian,
and K. Novoselov, Journal of the Electrochemical Society, vol. 148,
pp. 141-9, 2001/03/2001. [0073] [6] S. S. Choi, D. W. Kim, and M.
J. Park, Journal of the Korean Physical Society, vol. 45, pp.
1500-1504, December 2004. [0074] [7] C.-H. Yang and C.-M. Dai,
Proceedings of the SPIE--The International Society for Optical
Engineering Optical Microlithography XI, 25-27 Feb. 1998, vol.
3334, pp. 553-8, 1998//1998. [0075] [8] F. H. Bell, O. Joubert, G.
S. Oehrlein, Y. Zhang, and D. Vender, Journal of Vacuum Science
& Technology A (Vacuum, Surfaces, and Films), vol. 12, pp.
3095-101, 1994/11/1994. [0076] [9] G. Adegboyega, I. PerezQuintana,
A. Poggi, E. Susi, and M. Merli, Journal of Vacuum Science &
Technology B, vol. 15, pp. 623-628, May-June 1997. [0077] [10] B.
A. Cruden, M. V. V. S. Rao, S. P. Sharma, and M. Meyyappan, Journal
of Vacuum Science & Technology B (Microelectronics and
Nanometer Structures), vol. 20, pp. 353-63, 2002/01/2002. [0078]
[11] A. Nagy, Optical Engineering, vol. 31, pp. 335-340, February
1992. [0079] [12] C. H. Lin, G. B. Lee, Y. H. Lin, and G. L. Chang,
Journal of Micromechanics and Microengineering, vol. 11, pp.
726-732, November 2001. [0080] [13] Xinghua Li, Takashi Abe, and
Masayoshi Esashi, Sensors and Actuators A: Physical 87 (3), 139
(2001). [0081] [14] E. Hein, D. Fox, and H. Fouckhardt, Journal of
Applied Physics 107 (3), 033301 (2010). [0082] [15] A. M. Hynes, H.
Ashraf, J. K. Bhardwaj, J. Hopkins, I. Johnston, and J. N.
Shepherd, Sensors and Actuators A: Physical 74 (1-3), 13 (1999);
Lee A. Donohue, Janet Hopkins, Richard Barnett, Andrew Newton, and
Anthony Barker, 44 (2004). [0083] [16] T. Ichiki, Y. Sugiyama, R.
Taura, T. Koidesawa, and Y. Horiike, Thin Solid Films 435 (1), 62
(2003). [0084] [17] D. A. Zeze, R. D. Forrest, J. D. Carey, D. C.
Cox, I. D. Robertson, B. L. Weiss, and S. R. P. Silva, Journal of
Applied Physics 92 (7), 3624 (2002). [0085] [18] S. Queste, R.
Salut, S. Clatot, J. Y. Rauch, and ChantalG Khan Malek, Microsystem
Technologies 16 (8-9), 1485 (2010). [0086] [19] Abhijat Goyal,
Vincent Hood, and Srinivas Tadigadapa, 61110P (2006). [0087] [20]
H. Sugai, K. Nakamura, Y. Hikosaka, and M. Nakamura, Journal of
Vacuum Science & Technology A: Vacuum, Surfaces, and Films 13
(3), 887 (1995). [0088] [21] Koji Yamakawa, Masaru Hori, Toshio
Goto, Shoji Den, Toshirou Katagiri, and Hiroyuki Kano, Applied
Physics Letters 85 (4), 549 (2004). [0089] [22] EzzEldin Metwalli
and Carlo G. Pantano, Nuclear Instruments and Methods in Physics
Research Section B: Beam Interactions with Materials and Atoms 207
(1), 21 (2003).
* * * * *