>Date: Tue, 11 Apr 1995 13:48:56 -0500 (EST)
>From:OXFORDCOMPUT@delphi.com
>To: MEMS@ISI.EDU
>Subject: Cornell's Supposed Nanotechnology Breakthrough
>
>April 11, 1995
>
>Can someone tell me what the breakthrough is? I do not see in the
>press release the news of any new work. The use of STM's to store data
>at very high densities is well known, see, for example, "Technology of
>Proximal Probe Lithography" by Christie Marrian (ed.), SPIE IS10,
>1993.
>
>Also, since they say they can scan an area 1 x 1 um, but it takes a micro-
>actuator 100 x 100 um to do this, the utilization of the storage medium is
>only 1/10,000 -- very, very inefficient. Also, the data rate, as the press
>release noted is very (very) slow, so it will take thousands of such probes
>running in parallel to give the data rate of current disk drives.
>
>-- Steve Morton, Oxford Computer, Inc., Oxford, CT
>
Dear Steve,

In answer to your question: Can someone tell me what the breakthrough is?

Yes I can, but the press release you read does not convey the information. The
breakthrough is that we have fabricated an integrated, micro-STM with an xyz
actuator and integrated tip and we demonstrated that it can function as a
scanning tunneling microscope (STM) (Yang Xu, Scott A. Miller, and Noel C.
MacDonald, "Microelectromechanical Scanning Tunneling Microscope," American
Physical Society March Meeting, San Jose, CA, Mar. 20-24, 1995. Bulletion of
the American Physical Society , vol 40, No. 1, p. 63). Further we have reduced
the size of the STM to about 200 micrometers on-a-side with xyz actuators and
tip (s); we now have a fully integrated, very high tip density silicon process
(K.A. Shaw, et al., Sensors and Actuators A 40, 63 (1994))and(Z. L. Zhang and
N. C. MacDonald, J. Micromech. Microeng. 2, 31 (1992)) to make arrays of
MICROSTMs and MICROAFMs. Most important we have a process technology that now
works for this application. Also, the technology can scale to 100,000's of tips
and some architectures provide very fast read out of the data, but this
information is not part of the statement you read. Again our contribution is
the development of integrated xyz microactuators with integrated tips in
silicon technology. Furthermore the technology can be scaled to massively
parallel probe architectures.

Our STM is small enough to be an array element for a massively parallel STM/AFM
architecture (again that's the point) -including an 'on-chip', integrated z
position microsensor for AFM (no external light source or mirror is required)-
and is compatible with silicon processing techniques. Furthermore, each xyz
actuator can support many tips (maybe up to 100) - not just one tip per
actuator - see for example a 25 tip array on one XYZ actuator ([1] Z. Lisa
Zhang and N. C. MacDonald, "Integrated Silicon Process for Micro-Dynamic Vacuum
Field Emission Cathodes," J. Vac. Sci. Technol.,Vol. B 11(6), pp. 2538-2543,
(Nov/Dec. 1993). [2] Z. L. Zhang and N. C. MacDonald, J. Micromech. Microeng.
2, 31 (1992)). As is pointed out in "Technology of Proximal Probe Lithography"
by Christie Marrian (ed.), SPIE IS10,1993, page 7 and page 30, the only
practical way to use proximal probes in lithography and data storage is to
operate arrays of tips in parallel. The point is we have realized a single
crystal silicon process with integrated tips, xyz actuators and electrical
contacts on moving structures. This is the paradigm we have been working on for
8 or 9 years. Fast, highly integrated xyz microactuators with integrated tips
is the game -no assembly and integrated transistors on the moving structure (J.
J. Yao et al., Sensors and Actuators A 40, 77 (1994)) for addressing! Silicon
is a great choice for such a device or array of devices for massively parallel
information storage or massively parallel lithography etc.

We have demonstrated tip-arrays with 5 micrometer spacing using the same
process used to make the working microSTM ([1] J. P. Spallas and N. C.
MacDonald, "Self-aligned Silicon Field Emission Cathode Arrays formed by
selective, lateral thermal oxidation of silicon," J. Vac. Sci. Technol., Vol. B
11 (2), pp. 437-440, (Mar./Apr. 1993). [2] S. Arney and N. C. MacDonald,
"Formation of Submicron Silicon on Insulator Structures by Lateral Oxidation of
Substrate-Silicon Islands", J. Vac. Sci. Technol. B:6 (1), 341-345, (Jan/Feb
1988).). Furthermore, with the same technology tips or small arrays of tips can
be spaced 25-50 micrometers apart and integrated with individual z
micro-actuators; so one xy manipulator can support many tips but each tip would
have a seperate z actuator. Contacts, isolation, interconnects, and transistors
(for amplifiers) have been fabricated on moving single crystal silicon beams
with our silicon MEMS technology (Z. L. Zhang and N. C. MacDonald, J. Vac. Sci.
Technol. B 11, 2538 (1993)) Yes, we need all the pieces to produce massively
parallel tip-array architecture's and now we have them - refer to the
publications etc.. Here we finally demonstrated an image with our microSTM to
show that all the integrated pieces work.

It's an all or nothing technology for massively parallel scan-probe array
architecture's! You must have all the pieces. In addition, we have demonstrated
a very sensitive' integrated Single Crystal Silicon torsional cantilever
technology (both STMs incorporate this z actuator) which is useful for array
architecture's. So the 'big deal' is the integrated micro STM works; it is made
using a robust single crystal silicon MEMS process; one xy(z) actuactor can
support many tips using our process depending on the architecture; it is a
highly integrated, batch fabricated technology; and all the pieces have now
been demonstrated. Other questions (you could ask) are can we make these
suspended silicon structures over an area a few mms on-a-side, and can we make
actuators that move 50 or so micrometers per tip? Fortunately the answer is now
yes to both questions. We have recently demonstrated a suspended large area
actuator (4mm x 5mm) that produces 10 milliNewtons of force using the same
technology used to make the microSTM (M. T. A. Saif and N. C. MacDonald, [1]
"Design Considerations for Large MEMS," Proceedings paper 2448-10; 1995 North
American Conference on Smart Structures and Materials, San Diego, Ca., 26
February - 3 March 1995; M. T. A Saif and N. C. MacDonald. [2] `A Milli-Newton
Micro-Loading Device" accepted for Transducers '95, June 25-29, Stockholm,
Sweden, June 25 - 29, 1995.); so things are finally looking positive for making
large arrays of tips that move in xyz over a 50 micrometer x 50 micrometer area
per tip and cover a total area (with all the tips) of at least 5 mm x 5mm - all
the tips and microactuators are mounted on a large, suspended and planar single
crystal silicon structure.

Steve, Give me a call if you need additional information or discussion.

Regards, Noel

Prof. Noel C. MacDonald, Cornell University
(607) 255-3388.
or contact
Scott A. Miller; Cornell University, School of Electrical Engineering; 409
Phillips Hall,Ithaca, NY 14853;Phone: (607) 255-7377; FAX: (607) 254-4565;
E-mail: sam8@cornell.edu