Peak Shock Detection Sensor System Fabricated on Flexible Substrate

by Dr. Narendra Lakamraju

Topics Covered

Abstract
Introduction
Design
     Calculations
     Simulations
Fabrication
     Process Flow
     Silicon Sacrificial Etch
Test Results
Conclusions

Abstract

Continuous exposure of personnel to explosion shock waves has been known to cause internal damage that can lead to irreparable damage if not detected early and hence the need to design sensors that can detect, record and display shock information. We present work resulting in the design and fabrication of passive shock sensors on flexible substrate for increased portability enhanced functionality.

Introduction

Closed-head brain trauma is difficult to diagnose and treat both in the field where quick decisions are required, as well as in a hospital environment where early decisions can impact the long-term prognosis for recovery and rehabilitation. Field decisions are crucial since appropriate immediate action for treatment can have a large effect on the long-term prognosis. Moreover, knowledge of the type and severity of traumatic brain injury sustained is critically important in developing and prescribing the appropriate longer term rehabilitation strategies. Not to be ignored is the psychological importance of being able to provide realistic expectations to the patient and his or her family and caregivers^1,2.

A widely deployed, cost effective solution to provide accurate cumulative peak blast dose measurement will be effective in directly providing better patient care and in enabling the development of an accurate experiment-based model for the modes and severity of traumatic brain injury due to specific types, magnitudes and durations of blast dose. The technological rationale is to leverage the recent innovations in flexible substrate electronics and display technologies by integrating a MEMS-like sensor fabrication process for blast (pressure) sensing. A single batch fabrication process integrating sensors, electronics and displays will provide the lowest cost required for wide-scale deployment.

The sensor tag will need to be passive in nature to eliminate the need for a constant power supply to record the information³. Integration of a display element to the tag will enable triage medics to read and possibly diagnose Traumatic Brain Injury (TBI) in the field.

This sensor system can also be used in mining application to detect the amount of shock experienced by personnel. A modified version of the system can also be used to gauge the integrity of a structure exposed to continuous shock waves and possibly prevent mining accidents. Tags that measure the intensity of shock waves sticking buildings during demolitions in busy areas can provide useful data toward better control and acceptable shock levels. Intensity of waves emanating from audio sources can be measured without the use of expensive equipment and setup. This information can be used to determine safe audio levels for listeners and prevent auditory damage due to loud noises.

Design

The sensor tag consists of a sensor connected to an electrophoretic display element. The sensor has a capacitor like structure with a collapsible membrane suspended above a fixed electrode. The spacing between the membranes and the thickness of the flexible membrane are used to control the collapse point. When a pressure wave strikes the movable membrane, it deflects across the spacing between the electrodes and makes contact with the fixed electrode. Upon making contact the electrode, Van der Waal's and/or Casimir's forces prevent the membrane from moving back to its original position. The change in impedance between the two electrodes is then used to detect a collapse and activate the display element through a resistor network.

Calculations

Collapse pressure for a sensor is related to the thickness of the membrane, spacing between, radius of the sensor and the property of the metal as shown in (1).

ω (r) = [ω₀{1 - (r/a)²}] ----- (1)

Where ω₀ is the deflection at the center of the membrane, a is the radius of the sensor. The deflection at the center of the membrane ω₀ is given by (2),

ω₀ = (p •a⁴) / (64 • D) ----- (2)

Where p is the applied pressure and D is the flexural rigidity of the membrane (3).

D = (E • t²) / 12[1-μ²] ----- (3)

E is the Young's Modulus, t is the thickness of the membrane and μ is Poisson's ratio⁴.

To reduce the complexity of the fabrication process and the number of masks required for fabrication, the radius of the sensor and the spacing between the membranes is fixed and the thickness of the membrane varied to achieve different sensitivities. Also, the fourth order dependence of the pressure sensitivity to the radius of the sensor demands ultra fine etch control defining the sensor radius which leads to increased costs. The spacing between the membranes is set at 0.5µm, radius of the sensor is set to 70µm and thickness of the Aluminum film is varied from 0.6µm to 1µm to vary the sensitivity from 100kPa to 450kPa.

Simulations

The designs are tested for functionality and operation using Coventorware®, a simulation tool commonly used for MEMS simulation. Fig. 1 and Fig. 2 show simulations of the sensor before and after actuation. The model is exaggerated in the z-axis to show detail. Displacement of the membrane due to a pressure is shown in the figures and results from simulations are in agreement with calculated values.

Figure 1. Model of sensor with etch holes in the middle before activation.

Figure 2. Model showing deflection in the sensor membrane after activation.

Simulations also help with designing and testing different configurations and materials for the top membrane. Array of sensors are connected in parallel to improve sensitivity of the tag and help with fault tolerance from random faulty sensors.

Fabrication

The sensor tags are fabricated using standard Thin Film Transistor (TFT) processes to ensure compatibility with VLSI processes and reduce fabrication costs. The sensors are fabricated on a flexible substrate to ensure conformality with the mounting surface which can be the back of a helmet or a shoulder patch. All the processes used for the fabrication of the devices are low temperatures to protect the integrity of the substrate.

Process Flow

First step in the fabrication of the sensors involves bonding the flexible Polyethylene naphthalate (PEN) substrate to a carrier substrate⁵. The bonding process is performed using a proprietary compound that can endure all the processing steps performed on the substrate.

Following bonding, a thin layer of Aluminum that is sputter deposited is used to form the bottom electrode. Aluminum is chosen as the electrode material as it offers good etch selectivity to xenondifluoride (XeF₂) sacrificial etch release process^6,7 that is performed at the end. 0.5µm thick silicon is sputter deposited to form the sacrificial layer between the two electrodes. The top electrode is also formed by patterning a sputter deposited Aluminum layer whose thickness is chose to get the desired sensitivity. Though the tests were performed using Aluminum films other materials such as metals with different Young's Modulus may be chosen. The spacing between the two electrodes is created by etching away silicon thorough etch holes placed across the area of the sensor. Final step in the fabrication involves debonding the PEN substrate with fabricated devices from the carrier substrate. Fig. 3, 4 and 5 show the process flow used for the fabrication of the shock sensors. After the sensors are debonded from the carrier substrate a strip of electrophoretic material is attached to indicate the activation of the sensor.

Figure 3. Deposit and pattern bottom electrode for the sensor

Figure 4. Deposit silicon sacrificial layer followed by top metal.

Figure 5. Pattern top metal with etch holes and perform sacrificial etch to release structure.

Silicon Sacrificial Etch

A timed XeF₂ gaseous silicon etch process is used to define the size of the sensor and eliminate the need for an additional mask to pattern the silicon layer which in turn lowers the unit cost of the sensor tag.

A basic XeF₂ etch system consists of a XeF₂ source connected to an expansion chamber which in turn is linked to the device chamber. XeF2 is a solid but has a low vapor pressure causing the solid to change to gas at room temperature and atmospheric pressure. To control the etch process, the solid is allowed to expand to a set pressure in an expansion chamber. The gas is then allowed to enter the device chamber which holds the sample. The gas is allowed to react for a predetermined time also referred to as the cycle time, after the etchant is completely used up the chamber is pumped out and the process repeated for a set number of cycles. An etch rate of 2µm/min is observed when the expansion pressure is set to 2.7mTorr. Gaseous nature of the etchant helps overcome stiction common in structures released using a wet etchant. 8 etch cycles each 60 seconds long was found to be adequate for releasing the sensors.

Pictures of sensors fabricated on PEN substrates are shown in Fig. 6 and Fig. 7. Fig. 6 shows a sensor with pads for resistance measurement and Fig. 7 shows an integrated sensor tag with display element.

Figure 6. Image of a fabricated sensor showing the sensor array in the middle and measurement pads around the edge.

Figure 7. Sensor array integrated with display element.

Test Results

Optical and scanning electron images of the sensors after fabrication confirm the release of the membranes and help refine the fabrication process. Image of sensors before and after activation validate the design. Optical images of the sensors are shown in Fig. 8.

Figure 8. Optical image of sensor array showing sensors before and after activation.

Figure 9. SEM image and a FIB cut of sensor showing deformed membrane.

Fig. 9 shows an SEM image of a sensor indicating the change in membrane before and after activation. A Focused Ion Beam (FIB) cut across one of the membrane is performed to confirm the spacing between the membrane and the bottom electrode. Upon verification of the membrane collapse, resistance measurements across the two electrodes are recorded for various sensors and the readings listed in Table 1.

Table 1. Resistances for different sensors before and after activation.

Preliminary results from devices fabricated on flexible PEN substrates are very promising and testing of devices in a calibrated shock tube are under way at US Army Natick Soldier Research, Development, & Engineering Center in Natick, MA.

Conclusions

Passive shock wave pressure sensors capable of detecting and recording intensity of explosions have been fabricated on flexible PEN substrates. MEMS capacitor like structures that have collapsible membranes with integrated electrophoretic display elements have been fabricated to detect and record pressures from 100kPa to 450kPa. Data resulting from initial testing is being used to refine the sensor design and integrate multiple sensors for reduced foot print and increased range and resolution.

References

1. Douglas S. DeWitt and Donald S. Prough, "Blast-Induced Brain Injury and Posttraumatic Hypotension and Hypoxemia," pp. 877-887, June 2009.
2. D. Warden, "Military TBI During the Iraq and Afghanistan Wars," J. Head Trauma Rehabil., vol. 21, pp. 398-402, 2006.
3. T. Ng, A. C. Arias, J. H. Daniel, S. Garner, L. Lavery, S. Sambandan and G. L. Whiting. Flexible printed sensor tape for diagnostics of mild traumatic brain injury. Presented at IDTechEx Printed Electronics Asia.
4. S. P. Timoshenko, "Theory of Plates and Shells", McGraw-Hill, 1940.
5. S. M. O'Rourke, S. M. Venugopal, G. B. Raupp, D. R. Allee, S. Ageno, E. J. Bawolek, D. E. Loy, J. P. Kaminski, C. Moyer, B. O'Brien, K. Long, M. Marrs, D. Bottesch, J. Dailey, J. Trujillo, R. Cordova, M. Richards, D. Toy and N. Colaneri, "Active Matrix Electrophoretic Displays on Temporary Bonded Stainless Steel Substrates with 180 [degree]C a-Si:H TFTs," SID, vol. 39, pp. 422-424, May 2008, 2008.
6. L. R. Arana, N. d. Mas, R. Schmidt, A. J. Franz, M. A. Schmidt and K. F. Jensen, "Isotropic etching of silicon in fluorine gas for MEMS micromachining," J Micromech Microengineering, vol. 17, pp. 384-392, 2007.
7. I. W. T. Chan, K. B. Brown, R. P. W. Lawson, A. M. Robinson, Yuan Ma and D. Strembicke, "Gas phase pulse etching of silicon for MEMS with xenon difluoride," Electrical and Computer Engineering, 1999 IEEE Canadian Conference on, vol. 3, pp. 1637-1642 vol.3, 1999.

Copyright AZoNano.com, MANCEF.org