The human brain (volume=1200cm3) consumes 20W and is capable of performing > 10^16 operations/s. Current supercomputer technology has reached 1015 operations/s, yet it requires 1500m^3 and 3MW, giving the brain a 10^12 advantage in operations/s/W/cm^3. Thus, to reach exascale computation, two achievements are required: 1) improved understanding of computation in biological tissue, and 2) a paradigm shift towards neuromorphic computing where hardware circuits mimic properties of neural tissue. To address 1), we will interrogate corticostriatal networks in mouse brain tissue slices, specifically with regard to their frequency filtering capabilities as a function of input stimulus. To address 2), we will instantiate biological computing characteristics such as multi-bit storage into hardware devices with future computational and memory applications. Resistive memory devices will be modeled, designed, and fabricated in the MESA facility in consultation with our internal and external collaborators.
Columbia University has developed a sensitive highly multiplexed system for genetic identification of nucleic acid targets. The primary obstacle to implementing this technology is the high rate of false positives due to high levels of unbound reporters that remain within the system after hybridization. The ability to distinguish between free reporters and reporters bound to targets limits the use of this technology. We previously demonstrated a new electrokinetic method for binary separation of kb pair long DNA molecules and oligonucleotides. The purpose of this project 99864 is to take these previous demonstrations and further develop the technique and hardware for field use. Specifically, our objective was to implement separation in a heterogeneous sample (containing target DNA and background oligo), to perform the separation in a flow-based device, and to develop all of the components necessary for field testing a breadboard prototype system.
Two different Sandia MEMS devices have been tested in a high-g environment to determine their performance and survivability. The first test was performed using a drop-table to produce a peak acceleration load of 1792 g's over a period of 1.5 ms. For the second test the MEMS devices were assembled in a gun-fired penetrator and shot into a cement target at the Army Waterways Experiment Station in Vicksburg Mississippi. This test resulted in a peak acceleration of 7191 g's for a duration of 5.5 ms. The MEMS devices were instrumented using the MEMS Diagnostic Extraction System (MDES), which is capable of driving the devices and recording the device output data during the high-g event, providing in-flight data to assess the device performance. A total of six devices were monitored during the experiments, four mechanical non-volatile memory devices (MNVM) and two Silicon Reentry Switches (SiRES). All six devices functioned properly before, during, and after each high-g test without a single failure. This is the first known test under flight conditions of an active, powered MEMS device at Sandia.
The objective of this LDRD was to develop a uniquely capable, novel droplet solution based manufacturing system built around a new MEMS drop ejector. The development all the working subsystems required was completed, leaving the integration of these subsystems into a working prototype still left to accomplish. This LDRD report will focus on the three main subsystems: (1) MEMS drop ejector--the MEMS ''sideshooter'' effectively ejected 0.25 pl drops at 10 m/s, (2) packaging--a compact ejector package based on a modified EMDIP (Electro-Microfluidic Dual In-line Package--SAND2002-1941) was fabricated, and (3) a vision/stage system allowing precise ejector package positioning in 3 dimensions above a target was developed.