Given the rapidly increasing data volumes that are downloaded onto mobile devices such as cell phones and PDAs, there is a growing demand for suitable storage media with more and more capacity. At CeBIT, IBM for the first time shows the prototype of the MEMS- Micro Electrical Mechanical System- assembly of a nanomechanical storage system known internally as the "millipede" project. Using revolutionary nanotechnology, scientists at the IBM Zurich Research Laboratory, Switzerland, have made it to the millionths of a millimetre range, achieving data storage densities of more than one terabit (1000 gigabit) per square inch, equivalent to storing the content of 25 DVDs on an area the size of a postage stamp. With this new technique, 3040-nm-sized bit indentations of similar pitch size have been made by a single cantileverltip in a thin (50-nm) polymethylmethacrylate (PMMA) layer, resulting in a data storage density of 400500 ~ b l i n .T*h e Millipede project could bring tremendous data capacity to mobile devices such as personal digital assistants, cellular phones, and multifunctional watches, can be used to explore a variety of other applications, such as large-area microscopic imaging, nanoscale lithography or atomic and molecular manipulation.
When we think of data storage, it's common to imagine hard drive platters or solid-state memory chips. But beyond magnetic fields or electrical charges, a surprising amount of digital information is also stored in a physical form; punched paper tape and punched cards are very early examples, but our very latest CD and DVD media represent data as a series of "pits and lands" delivered to a physical surface. When IBM launched the Millipede project in 1996, it heralded another data storage effort designed to record data through microscopic physical techniques, promising very high storage densities in a In actual practice the Millipede's thousands of microscopic tips write tiny pits to a thin film of special polymer. The sequence of pits corresponds to bits. Unlike punched cards or tape, however, the data can be erased and rewritten. The high storage density of more than a terabit was achieved by using individual silicon tips to create pits approximately 10 nanometers in diameter, i.e. 50,000 times smaller than the period at the end of this sentence. Experimental chips have been designed comprising more than 4,000 of these tips arrayed in a small 6.4 mm x 6.4 mm2. These dimensions make it possible to pack an entire high-capacity storage system into the SD flash memory format package. The project is still in an advanced research state. After a decision has been made, it will take another two to three years of development until the product would be available on the market. Moreover, the nanomechanical data medium has been optimized to use a minimum amount of energy. Thus, it is ideally suited for use in mobile devices such as digital cameras, cell phones and USB sticks. However, it is likely that IBM's these criteria, mobile storage (for example, for cell phones, USB sticks, and digital cameras) is ideally suited for the Millipede probe storage technology. In this segment Millipede is able to compete against flash, which is very costly at capacities between 5GB and 40GB. Millipede's inherent shock resistance and low power requirements also bolster these features.
MOTIVATION & OBJECTIVES
In the 21st century, the nanometer will very likely play a role similar to the one played by the micrometer in the 20th century. The nanometer scale will presumably pervade the field of data storage. In magnetic storage today, there is no clear-cut way to achieve the nanometer scale in all three dimensions. Within a few years, however, magnetic storage technology will arrive at a stage of its exciting and successful evolution at which fundamental changes are likely to occur when current storage technology hits the well-known super paramagnetic limit. Several ideas have been proposed on how to overcome this limit. One such proposal involves the use of patterned magnetic media, for which the ideal writelread concept must still be demonstrated, but the biggest challenge remains the patterning of the magnetic disk in a cost-effective way. Other proposals call for totally different media and techniques such as local probes or holographic methods. In general, if an existing technology reaches its limits in the course of its evolution and new alternatives are emerging in parallel, two things usually happen: First, the existing and well-established technology will-be explored further and everything possible done to push its limits to take maximum advantage of the considerable investments made. Then, when the possibilities for improvements have been exhausted, the technology may still survive for certain niche applications, but the emerging technology will take over, opening up new perspectives and new directions.Today we are witnessing in many fields the transition from structures of the micrometer scale to those of the nanometer scale, a dimension at which nature has long been building the finest devices with a high degree of local functionality. Many of the I It techniques we use today are not suitable for the coming nanometer age. In any case, an A emerging technology being considered as a serious candidate to replace an existing but limited technology must offer long-term perspectives. The only available tool known today that is simple and yet provides these very longterm perspectives is a nanometer sharp tip. Such tips are now used in every atomic force microscope (AFM) and scanning tunneling microscope (STM) for imaging and structuring down to the atomic scale. In the early 1990s, Mamin and Rugar at the IBM Almaden Research Center pioneered the possibility of using an AFM tip for readback and writing of topographic features for the purposes of data storage. In one scheme developed by them, reading and writing were demonstrated with a single AFM tip in contact with a rotating polycarbonate substrate. The data were written thermomechanically via heating of the tip. In this way, densities of up to 30 ~ b / i nw.e~re achieved. The objectives of the research activities within the Micro- and Nanomechanics Project at the IBM Zurich Research Laboratory are to explore highly parallel AFM data storage with areal storage densities far beyond the expected super paramagnetic limit (60100 ~ b l i n .a~n)d data rates comparable to those of today's magnetic recording. The "Millipede" concept is a new approach for storing data at high speed and with an ultrahigh density. It is not a modification of an existing storage technology, although the use of magnetic materials as storage media is not excluded. The ultimate locality is given by a tip, and high data rates are a result of massive parallel operation of such tips. The current effort is focused on demonstrating the Millipede concept with areal densities up to 500 Gblin.* and parallel operation of very large 2D (32 x 32) AFM cantilever arrays with integrated tips and writelread storage functionality. The AFM-based data storage concept, Millipede has a potentially ultrahigh density, terabit capacity, small form factor, and high data rate. Its potential for ultrahigh storage density has been demonstrated by a new thermo mechanical local-probe technique to store and read back data in very thin polymer films. With this new technique, cantileverltip in a thin (50-nm) polymethylmethacrylate (PMMA) layer, resulting in a data storage density of 400500 ~ b l i nH. i~gh data rates are achieved by parallel operation of large two-dimensional (2D) AFM arrays that have been batch-fabricated by silicon surface-micromachining techniques. The very large scale integration (VLSI) of microlnanomechanical devices (cantileversltips) on a single chip leads to the largest and densest 2D array of 32 x 32 (1024) AFM cantilevers with integrated writelread storage functionality ever built. Time-multiplexed electronics control the writelread storage cycles for parallel operation of the Millipede array chip. Initial areal densities of 100200 ~ b l i n . ~ have been achieved with the 32 x 32 array chip, which has potential for further improvements. In addition to data storage in polymers or other media, and not excluding magnetics, we envision areas in nanoscale science and technology such as lithography, high-speedllarge-scale imaging, molecular and atomic manipulation.
3.1. Technological Background
The 2D AFM cantilever array storage technique called "Millipede" is illustrated in figure. It is based on a mechanical parallel x/y scanning of either the entire cantilever array chip or the storage medium. In addition, a feedback-controlled z-approaching and - leveling scheme brings the entire cantilever array chip into contact with the storage medium. This tipmedium contact is maintained and controlled while x/y scanning is performed for writelread. It is important to note that the Millipede approach is not based on individual z-feedback for each cantilever; rather, it uses a feedback control for the entire chip, which greatly simplifies the system. However, this requires stringent control and uniformity of tip height and cantilever bending. Chip approach and leveling make use of four integrated approaching cantilever sensors in the corners of the array chip to control the approach of the chip to the storage medium. Signals from three sensors provide feedback signals to adjust three magnetic z-actuators until the three approaching sensors are in contact with the medium. The three sensors with the individual feedback loop maintain the chip leveled and in contact with the surface while x/y scanning is performed for writelread operations. This basic concept of the entire chip approachlleveling has been tested and demonstrated for the first time by parallel imaging with a 5 x 5 array chip. These parallel imaging results have shown that all 25 cantilever tips have approached the substrate within less than 1 pm of z-activation. This promising result has led us to believe that chips with a tip-apex height control of less than 500 nm are feasible. This stringent requirement for tip-apex uniformity over the entire chip is a consequence of the uniform force needed During the storage operation, the chip is raster-scanned over an area called the storage field by a magnetic x/y scanner. The scanning distance is equivalent to the cantilever x/y pitch, which is currently 92 pm. Each cantileverltip of the array writes and reads data only in its own storage field. This eliminates the need for lateral positioning adjustments of the tip to offset lateral position tolerances in tip fabrication. Consequently, a 32 x 32 array chip will generate 32 x 32 (1 024) storage fields on an area of less than 3 mm x 3 mm. Assuming an areal density of 500 ~ b l i n .o~n,e storage field of 92 pm x 92 pm has a capacity of about 10 Mb, and the entire 32 x 32 array with 1024 storage fields has a capacity of about 10 Gb on 3 mm x 3 mm. As the storage capacity scales with the number of elements in the array, cantilever pitch (storage-field size) and areal density, and depends on the application requirements. Lateral tracking will also be performed for the entire chip, with integrated tracking sensors at the chip periphery. This assumes and requires very good temperature control of the array chip and the medium substrate between write and read cycles. For this reason the array chip and medium substrate should be held within about 1 "C operating temperature for bit sizes of 30 to 40 nm and array chip sizes of a few millimeters. This will be achieved by using the same material (silicon) for both the array chip and the medium substrate in conjunction with four integrated heat sensors that control four heaters on the chip to maintain a constant array-chip temperature during operation. True parallel operation of large 2D arrays results in very large chip sizes because of the space required for the individual writelread wiring to each cantilever and the many 110 pads. The row and column time-multiplexing addressing scheme implemented successfully in every DRAM is a very elegant solution to this issue. In the case of Millipede, the time-multiplexed addressing scheme is used to address the array row by row with full parallel writelread operation within one row. Temperature plays a critical role in every part of the device's operation once the tips contact the polymer surface. Bits are written by heating the tip to a temperature above the glass transition temperature of the polymer by means of the heating resistor integrated in the cantilever. The polymer in close proximity to the tip is heated and becomes softer allowing the tip to indent a few nanometers into the film, mechanically stressing the material. For reading the cantilever's reading sensor, which is separate from the tip, is heated slightly. As the polymer film is scanned under the tip, the tip moves in and out of the written indentations. When the tip moves into an indent, it cools down because of the reduced distance to the substrate. This cooling results in a measurable change in electrical conductivity of the sensor. To ovetwrite data, thermo-mechanical effects are used. They cause the stressed polymer material closely around a newly created bit to relax. The current Millipede storage approach is based on a new thermomechanical writehead process in nanometer-thick polymer films. Thermomechanical writing in polycarbonate films and optical readback were first investigated and demonstrated with a single cantilever by Mamin and Rugar. Although the storage density of 30 ~ b l i n . ~ obtained original& was not overwhelming, the results encouraged to use polymer films as well to achieve density improvements.
Thermomechanical AFM data storage
Thermomechanical writing is a combination of applying a local force by the cantileverltip to the polymer layer and softening it by local heating. Initially, the heat transfer from the tip to the polymer through the small contact area is very poor, improving as the contact area increases. This means that the tip must be heated to a relatively high temperature (about 400Ã‚Â°C) to initiate the melting process. Once melting has commenced, the tip is pressed into the polymer, which increases the heat transfer to the polymer, increases the volume of melted polymer, and hence increases the bit size. It is estimated that at the beginning of the writing process only about 0.2% of the heating power is used in the very small contact zone (1040 nm2) to melt the polymer locally, whereas about 80% is lost through the cantilever legs to the chip body and about 20% is radiated from the heater platform through the air gap to the medium/substrate. After melting has started and the contact area has increased, the heating power available for generating the indentations increases by at least ten times to become 2% or more of the total heating power. With this highly nonlinear heat-transfer mechanism, it is very difficult to achieve small tip penetration and thus small bit sizes, as well as to control and reproduce the thermomechanical writing process. This situation can be improved if the thermal conductivity of the substrate is increased, and if the depth of tip penetration is limited. They have explored the use of very thin polymer layers deposited on Si substrates to improve these characteristics. The hard Si substrate prevents the tip from penetrating farther than the film thickness allows, and it enables more rapid transport of heat away from the heated region because Si is a much better conductor of heat than the polymer. Si substrates are coated with a 40-nm film of polymethylmethacrylate (PMMA) to achieve bit sizes ranging between 10 and 50 nm. However there is increased tip wear, probably caused by the contact between Si tip and Si substrate during writing. So introduced a 70-nm layer ofcross-linked photoresist (SU-8) between the Si substrate and the PMMA film to act as a softer penetration stop that avoids tip wear but remains thermally stable. Using this layered storage medium, data bits 40 nm in diameter have been written. These results were obtained using a I-pm-thick, 70-pm-long, two-legged Si cantilever. The cantilever legs are made highly conducting by high-dose ion implantation, whereas the heater region remains low-doped. Electrical pulses 2 ps in duration were applied to the cantilever with a period of 50 ps Imaging and reading are done using a new thermomechanical sensing concept. The heater cantilever originally used only for writing was given the additional function of a thermal readback sensor by exploiting its temperature-dependent resistance. The resistance ® increases nonlinearly with heating powerltemperature from room temperature to a peak value of 500700Ã‚Â°C.For sensing, the resistor is operated at about 350Ã‚Â°C, a temperature that is not high enough to soften the polymer, as is necessary for writing.. When the distance between heater and sample is reduced as the tip moves into a bit indentation, the heat transport through air will be more efficient, and the heater's temperature and hence its resistance will decrease. Thus, changes in temperature of the continuously heated resistor are monitored while the cantilever is scanned over data bits, providing a means of detecting the bits.
3.3. Array design, technology, and fabrication
As a first step, a 5 x 5 array chip was designed and fabricated to test the basic Millipede concept. All 25 cantilevers had integrated tip heating for thermomechanical writing and piezoresistive deflection sensing for read-back. No timemultiplexing addressing scheme was used for this test vehicle; rather, each cantilever was individually addressable for both thermomechanical writing and piezoresistive deflection sensing. A complete resistive bridge for integrated detection has also been incorporated for each cantilever. The array of tiny levers at the heart of the M~ll~pedsyes tem The chip has been used to demonstrate x/y/z scanning and approaching of the entire array, as well as parallel operation for imaging. This was the first parallel imaging by a 2D AFM array chip with integrated piezoresistive deflection sensing. The imaging results also confirmed the global chip-approaching and -leveling scheme, since all 25 tips approached the medium within less than 1 pm of z-actuation. Unfortunately, the chip was not able to demonstrate parallel writing because of electro migration problems due to temperature and current density in the Al wiring of the heater. However, the results got from 5 x 5 test vehicle are I)gl obal chip approaching and leveling is possible and promising, and 2) metal (Al) wiring on the cantilevers should be avoided to eliminate electromigration and cantilever deflection due to bimorph effects while heating. With the findings from the fabrication and operation of the 5 x 5 array and the very dense thermomechanical writinglreading in thin polymers with single cantilevers, they made some important changes in the chip functionality and fabrication processes. The major differences are 1) surface micromachining to form cantilevers at the wafer surface, 2) all-silicon cantilevers, 3) thermal instead of piezoresistive sensing, and 4) first- and second-level wiring with an insulating layer for a multiplexed row/columnaddressing scheme. Since the heater platform functions as a writelread element and no individual cantilever actuation is required, the basic array cantilever cell becomes a simple twoterminal device addressed by multiplexed x/y wiring. The cell area and x/y cantilever pitch is 92 pm x 92 pm, which results in a total array size of less than 3 mm x 3 mm for the 1024 cantilevers. The cantilever is fabricated entirely of silicon for good thermal and mechanical stability. It consists of the heater platform with the tip on top, the legs acting as a soft mechanical spring, and an electrical connection to the heater. They are highly doped to minimize interconnection resistance and replace the metal wiring on the cantilever to eliminate electromigration and parasitic z-actuation of the cantilever due to the bimorph effect. The resistive ratio between the heater and the silicon interconnection sections should be as high as possible; currently the highly doped interconnections are 400 aand the heater platform is 11 krm. (at 4 V reading bias). The cantilever mass must be minimized to obtain soft (flexible), high-resonantfrequency cantilevers. Soft cantilevers are required for a low loading force in order to eliminate or reduce tip and medium wear, whereas a high resonant frequency allows high-speed scanning. In addition, sufficiently wide cantilever legs are required for a small thermal time constant, which is partly determined by cooling via the cantilever legs. These design considerations led to an array cantilever with 50-pm-long, 10-pm-wide, 0.5-pmthick legs, and a 5-pm-wide, 10-pm-long, 0.5-pm-thick platform. Such a cantilever has a stiffness of 1 Nlm and a resonant frequency of 200 kHz. The heater time constant is a few i microseconds, which should allow a multiplexing rate of 100 kHz. This contradicts the requirement of a large gap between the chip surface and the storage medium to ensure that only the tips, and not the chip surface, are making contact with the medium. Instead of making the tips longer, bent the cantilevers a few micrometers out of the chip plane by depositing a stress-controlled plasma-enhanced chemical vapor deposition (PECVD) silicon-nitride layer at the base of the cantilever. Close-up of a lever's tiny tip Cantilevers are released from the crystalline Si substrate by surface micromachining using either plasma or wet chemical etching to form a cavity underneath the cantilever. Compared to a bulk-micromachined through-wafer cantilever-release process, as performed for 5 x 5 array, the surface-micromachining technique allows an even higher array density and yields better mechanical chip stability and heat sinking. Because the Millipede tracks the entire array without individual lateral cantilever positioning, thermal expansion of the array chip must be either small or well-controlled. Because of thermal chip expansion, the lateral tip position must be controlled with better precision than the bit size, which requires array dimensions as small as possible and a well-controlled chip temperature.