1. The impact of the floating low-energy ion gun (FLIG) on the consumer electronics industry
Submitting Institution
University of WarwickUnit of Assessment
PhysicsSummary Impact Type
TechnologicalResearch Subject Area(s)
Physical Sciences: Atomic, Molecular, Nuclear, Particle and Plasma Physics, Other Physical Sciences
Chemical Sciences: Physical Chemistry (incl. Structural)
Summary of the impact
This case study describes the invention, development and subsequent
commercial application of the floating low-energy ion gun (FLIG), a key
enabling technology for high-resolution depth profiling, in particular of
semiconductor devices. Following its invention at the University of
Warwick, the FLIG was commercialised and now plays an important role in
the semiconductor industry as a key analytical instrument. Intel and its
competitors have used the FLIG in developing specific technologies, such
as the PentiumTM, XeonfTM and CoreTM i7
processors. Its impact extends beyond the electronics industry to
consumers worldwide since the FLIG has played a key role in the
development of multicore processors for personal computers, intense
low-energy lighting for automotive and civil engineering, mobile
telecommunications technology, and many other areas of advanced
electronic, and other material, technologies. This invention has also led
directly to an ISO standard for depth resolution.
Underpinning research
The FLIG was invented by Dowsett, (then a Principal Research Fellow and
now a Professor in the Dept. of Physics at the University of Warwick),
assisted by Noel Smith (his PhD student in Warwick between 1992 &1996)
whose thesis describes a significant fraction of the early work [1]. The
underlying science behind the FLIG concept has two major drivers:
(i) In the late 1980s, the depth resolution provided by secondary ion
mass spectrometry (SIMS) in profiling devices was falling short of that
demanded by the International Technology Roadmap for Semiconductors. SIMS
instruments were using a high energy (several keV) primary ion beam, both
to laterally focus the bombarding ions and to achieve a measureable yield
of secondary ions ejected from the device under test; however, this high
energy limited vertical resolution. In defiance of the literature of the
time, and based on extrapolation of his own laboratory results at energies
down to below 1 keV, Dowsett was convinced that lowering the SIMS beam
energy still further, to a few hundred eV would improve the depth
resolution of SIMS.
(ii) In order to do this in a way that could be developed into a routine
analytical technique, it was necessary to greatly improve ion transport in
focussed ion beam columns, from the source outwards and to reduce
chromatic and spherical aberrations within the ion column to maximize the
usefulness of the transported current. At that time a few pA on target
into a spot size of 50 µm was typical for a beam energy of 1 keV.
No real progress on reducing beam energy was made until, in 1995, Dowsett
demonstrated dramatic improvements of the ion optics in a sub-keV ion
column [2] which he created with initial support from the Royal Society's
Paul Instrument Fund [8]. The success of the prototype depended on the
superb electronics and mechanical engineering support in the Warwick Dept.
of Physics, reflected by co-authorship of several technical staff on the
early papers. Subsequently, precise details of the instrumentation could
no longer be published as they became and remain commercially
confidential.
A central part of the FLIG concept is that most of the ion column is
`floating' at a high negative potential, so that ions are transported and
conditioned at high velocities, but then slowed down to impact energies of
100's or even 10's of eV just before striking the sample. This was the key
to improving the space-charge transport, generating useful impacting
fluxes up to 50,000 times higher than would otherwise have been possible,
and improving vertical resolution because the impact velocity at the
sample is so much less. Consequently, higher-resolution data can now be
obtained in minutes or hours, rather than in impractical time periods of
months or years.
The research output comprises: (i) conference papers, related patents [9]
and PhD theses deriving directly from the original research; (ii) key
papers that establish Dowsett's contribution to the applications research,
for example, developing the concept of the SIMS response function [3, 4]
that led to an ISO standard for depth resolution (Section 4); (iii)
numerous contributions [e.g. 5,6], to the subsequent development of
ultra-low-energy (ule) SIMS as a quantitative technique that is now being
used extensively in the semiconductor industry; and (iv) papers and
advances in materials processing and device technology made by other
owners of the instrumentation.
Maximum initial impact of this research was notably achieved by
presenting the early results at the SIMS X conference in Münster
(October 1995) where both instrument manufacturers and users were present
[2]. Within a few minutes of the presentation describing the device, the
three principal SIMS manufacturers of the time all tried to close deals to
use the technology. Atomika were the successful bidders, and made an
exclusive OEM deal with Ionoptika Ltd., a UK SME based near Southampton to
whom the University of Warwick had licensed the FLIG. The Atomika (now
Cameca) 4500, 4550 and 4600 instruments were the outcome of that initial
intellectual property (IP) licensing agreement.
Dowsett has continued to develop the technology. The commercial version
of the FLIG, which can now deliver around 500 nA of usable beam current at
500 eV, has re-engineered optics and its own custom high brightness
duoplasmatron with a low energy spread (both designed by Dowsett, in 1995
and 2004, respectively, the latter being descended from an earlier design
by Drummond & Long.) The success of the commercial FLIG depended on
the development of this very high brightness, cold cathode duoplasmatron,
which delivers both oxygen and other gas species such as inert gases and
nitrogen (that normally require a hot cathode source). Further elements of
the innovative design include a matched column that transports the beam at
high energy to the vicinity of the sample, and a multi-element, low
aberration retarding lens to reduce the impact energy by a factor of 10 to
30. Dowsett also designed the secondary ion optical and detector systems
of the Atomika 4550 and the 4600, and continues to provide support and
expertise to Ionoptika and users of the technology as a consultant.
Warwick has recently signed a further licence agreement with Ionoptika,
covering new FLIG-related ion source technology developed by Dowsett and
his son David, who completed his PhD at the University of Warwick in 2007
[7].
References to the research
(Principal Warwick authors in bold, Warwick technical staff in red)
Publications:
1. N.S. Smith, Ultra low energy SIMS depth profiling, PhD
thesis, University of Warwick (1996)
2. M.G. Dowsett, N.S. Smith, R. Bridgeland, D. Richards, A.C.
Lovejoy & P. Pedrick, An ultralow energy ion column for sputter
profiling. In Secondary Ion Mass Spectrometry (SIMS X), eds. A.
Benninghoven, B. Hagenhoff & H.W. Werner, (Wiley 1996) p.367 ISBN:
0471958972; N.S. Smith, M.G. Dowsett, B. McGregor & P.J.
Phillips, Rapid low energy depth profiling using SIMS, ibid.
p.363; G.A. Cooke, N.S. Smith, M.G. Dowsett, H.R.G. Burke &
P.J. Phillips, Ultra low energy SIMS depth profiling and the dilute limit,
ibid. p.371.
3. M.G. Dowsett, G. Rowlands, P.N. Allen and R.D. Barlow, An
analytic form for the SIMS response function measured from ultra-thin
impurity layers, Surf. Interface Anal. 21, 310-315 (1994)
DOI: 10.1002/sia.740210508
(Cited 65 times)
4. D.P. Chu and M.G. Dowsett, Dopant spatial distributions:
Sample-independent response function and maximum-entropy reconstruction, Phys.
Rev. B 56, 15167-15170 (1997). DOI: 10.1103/PhysRevB.56.15167
5. M.G. Dowsett, Depth profiling using ultra-low-energy secondary
ion mass spectrometry, Appl. Surf. Sci., 203, 5 (2003)
DOI: 10.1016/S0169-4332(02)00630-X
(Cited 47 times)
6. R.J.H. Morris and M.G. Dowsett, Ion yields and erosion rates
for Si1-xGex (0 ≤ x ≤ 1) ultralow energy O2+
secondary ion mass spectrometry in the energy range of 0.25-1 keV, J.
Appl. Phys. 105, 114316 (2009) DOI: 10.1063/1.3139279
7. D.M.F. Dowsett, High brightness ion sources for surface
analysis, PhD thesis, University of Warwick (2007). Led to Patent
application GB0815077.3 Ion Sources For Secondary Ion Mass
Spectroscopy (SIMS) (2008).
Grants/awards:
8. Royal Society's Paul Instrument Fund, PI M.G. Dowsett, The
development of a sub-keV ion column, 1990-1993, £38,225
Patents:
9. J.L. Maul, M.G. Dowsett, Sample Analysis Method EP 0878827 (1998)
& US
6078045 (2000); M.G. Dowsett, J.L. Maul, Secondary ion
mass spectrometer with aperture mask, US
6080986 (2000)
Details of the impact
All modern semiconductor devices depend on highly reproducible,
nanometre-scale multilayer engineering for dielectrics, semiconductor
doping and heterostructures. The FLIG system has enabled semiconductor
companies to develop new generations of devices with ever increasing
miniaturisation and hence has been an essential technology in the
development of all modern electronics. The nanoanalysis capability that
the FLIG provides has contributed to the creation of many of today's small
computers, mobile phones, communications & entertainment technologies.
Research and process development at this level requires analytical tools
with sufficient sensitivity and spatial resolution to measure prototype
and production wafer compositions accurately, and uleSIMS is a key method
in this area. Dowsett's prototype FLIG-based SIMS instrument (EVA 3000)
was the first SIMS instrument in the world capable of routine SIMS
analysis at energies down to 200 eV. Subsequently, the IP for the FLIG was
licensed to Ionoptika Ltd (UK) and the Atomika Instruments GmbH 4500 SIMS
depth profiler was specifically developed as a vehicle for the FLIG. This
was the world's first commercial instrument with ultra-low energy SIMS
capability.
Since its initial development in late 1993, its commercialization for OEM
application by Ionoptika Ltd. [10] and the first sale to Intel of a
FLIG-based instrument in 1996 by Atomika Instruments GmbH [11], the FLIG
has been in constant use in the R&D laboratories of the world's major
semiconductor and technology companies, such as AMD, Fibics, IMEC
(Belgium), Intel, MA-tek, Motorola, Nichia, Osram, Toshiba, TSMC4, and
around 20 more worldwide, as well as in universities such as Imperial
College, NUS (Singapore) & Warwick [12].
The FLIG was the founding product for Ionoptika Ltd. Their CEO says:
"I would point out the benefit to my company, a small British enterprise,
only recently started up at the time that we took on the FLIG licence from
the University of Warwick. The FLIG quickly became the mainstay of our
business and remained so for the next 6 years. In that time it earned us
close to £4M in revenue... However, the biggest impact is yet more
significant. This unique ion beam system has held its place as the state
of the art in shallow depth profiling from the day of its first
application in the Department of Physics at Warwick right up until the
present. Nothing else matches its depth resolution (measured in Angstroms)
and etch rates. The FLIG gave the semiconductor companies the ability to
develop new generations of devices with ever increasing miniaturisation.
So, the FLIG was an essential technology in the development of all modern
electronics..." [10].
Through mounting FLIG systems onto instruments by Atomika, and more
recently Cameca ISA, further sales worth tens of millions of pounds have
been generated and it remains an integral part of the Atomika (Cameca)
range of SIMS instrumentation [13]. Ionoptika continues to supply the FLIG
amongst other instrumentation that includes state-of-the-art
time-of-flight-SIMS. Substantial royalties have also been paid to the
University of Warwick, the inventor, and the Paul Instrument Fund (who
recovered their initial investment).
The impact of FLIG-based tools has been continuous since 1996 and
continues to grow alongside the expansion of microelectronic technology
into every area of modern life. They are in routine use today as process
control tools for the most demanding silicon industry, such as Intel,
where they have contributed to the development of PentiumTM,
XeonTM and CoreTM i7 (2008-date) processors:
"The development of the floating low energy ion gun (FLIG) by Professor
Mark Dowsett and his colleagues at the University of Warwick in the early
1990's was one of the most significant developments in analytical
instrumentation in the last 20 years. The refinement of the design and the
subsequent implementation on commercial SIMS instrumentation enabled
reliable, fast and accurate SIMS depth profiling of thin layers and
ultra-shallow junctions. This capability was critically important to the
semiconductor industry that was in an era of continuous junction scaling.
Intel was an early adopter of the FLIG technology in the mid 90's and the
FLIG remains a key capability in our laboratory today." [14]
FLIG-based tools are routinely employed in research on established
electronic materials and new semiconductors, such as diamond, carbides and
nitrides [15]. The high resolution depth profiling enabled by FLIG-based
tools is invaluable in developing a wide range of generic semiconductor
devices and other technologies, such as dynamic random access memory
(DRAMs) central to all computer data storage, quantum wells e.g. for GaN
light emitting diodes (LEDs), silicon germanium alloys, and high and low-k
dielectrics. Furthermore, the FLIG has had;
"... wider impact in organic electronics, pharmaceuticals, medical devices
and the characterisation of nanoparticles ... Dowsett and colleagues
pioneered the development of floating low energy ion gun technology (FLIG)
to achieve the high depth resolution necessary for inorganic
semiconductors. This continues to be state-of-the-art for the
semiconductor sector." [16]
To date, there are over 1,000 Google hits on FLIG associated items.
Dowsett's name for the ion gun and its acronym FLIG, although trademarked
by Ionoptika (US
Trademark No. 75047263), have become generically synonymous with
this type of technology. There are at least 100 peer reviewed papers, from
a wide range of international academic and commercial bodies, that
acknowledge use of the FLIG or Atomika / Cameca 4500, 4550 and 4600
instruments and hundreds of presentations at semiconductor workshops [e.g.
17], which are the principal means of dissemination in the industry. Use
of the FLIG has spread well beyond its original technological drivers of
silicon, silicon germanium [17] and gallium arsenide, into nanotechnology,
diamond electronics, and cultural heritage [18].
Finally, the SIMS response function developed, in 1994, by Dowsett to
relate the data obtained from a SIMS instrument to the underlying material
profile has given rise to a current ISO standard (ISO
20341:2003) [19]. The function itself is thus also a significant
outcome of the original FLIG development. It is becoming known in the
field as the `Dowsett Function' [20] and has become important both for the
semiconductor depth profiling community and outside of the original
research field in the new science arising from depth profiling of organic
materials [21].
"... the Dowsett function could be used to characterise the sputtering
process in these very different materials ... This now allows modern
organic electronic devices to be imaged in 3D with exquisite chemical
detail." [16]
Sources to corroborate the impact
- Supporting letter from CEO Ionoptika Ltd.
- Supporting statement from Sales Director Atomika GmbH/Cameca GmbH
-
www.ionoptika.com ;
www.cameca.com/instruments-for-research/sims4550.aspx
- List of further companies and institutions with FLIG-based instruments
available on request.
- Supporting letter from Senior Engineer, Intel Corporation.
- E.g. Fibics Inc. (operating from the Canadian Government's Materials
Technology Laboratory) uses an Atomika 4500 to offer SIMS services www.fibics.com/SC_SIMSOverview.html
- Supporting letter from National Physical Laboratory (NPL) Fellow.
- Publication from IMEC at recent workshop:
R. Pureti and W. Vandervorst, Quantification of Ge in Si1-x
Gex by using low-energy Cs+ and O2+
ion beams, Surf. Interface Anal. 45, 402-405 (2013) DOI:
10.1002/sia.5049
- Publications with reach into nanotechnology: Li Yang et
al. Depth Profiling and Melting of Nanoparticles in Secondary Ion
Mass Spectrometry, J. Phys. Chem. 117, 16042-16052
(2013) DOI: 10.1021/jp4048538;
R.J.H. Morris et al., The use of low-energy SIMS (LE-SIMS) for
nanoscale fuel cell material development, Surf. Interface Anal.
43, 635-638 (2010) DOI: 10.1002/sia.3526;
diamond electronics: R. Edgington et al. Growth and
electrical characterisation of d-doped boron layers on (111) diamond
surfaces, J. Appl. Phys. 111, 033710 (2012) DOI: 10.1063/1.3682760;
cultural heritage: M.G. Dowsett, et al., The use of
ultra-low-energy dynamic SIMS in the study of the tarnishing of silver,
Nucl. Instrum. Meth. Phys. Res. B239 51-64 (2005). DOI: 10.1016/j.nimb.2005.06.179;
G. Spoto & G. Grasso, Spatially resolved mass spectrometry in the
study of art and archaeological objects, TrAC — Trends in Analytical
Chemistry 30, 856-863 (2011) DOI: 10.1016/j.trac.2011.01.016
- ISO 20341:2003 (reviewed and confirmed in 2009) can be downloaded from
the Standards Centre at http://tinyurl.com/ptsde9e,
or at
www.iso.org/obp/ui/#iso:std:iso:20341:ed-1:v1:en
- Demonstrates the `Dowsett Function' as referred to by [20]: P.
Sjovall, et al., Sample cooling or rotation improves C60 organic
depth profiles of multilayered reference samples: Results from a VAMAS
interlaboratory study, J. Phys Chem. B 114 769-774
(2010)
- Examples in: Proceedings of the Seventeenth International Conference
on Secondary Ion Mass Spectrometry, SIMS XVII, Surf. Interface Anal.
43, 1-676 (2011);Cluster Secondary Ion Mass Spectrometry: Principles
and Applications Ed. C.M. Mahoney (Wiley 2013) ISBN:1118589246