Microelectronics Medical Devices

Microelectronics have had a profound impact on medical devices - reducing the size and power requirements of sensors, actuators, cameras and microprocessors used in surgical tools and day-to-day patient care.

One electronic component, the Hall effect sensor, has enjoyed relatively few applications in mission-critical monitoring and diagnostic tools, but that is about to change. Micromem Technologies Inc., is actively pursuing development of non-invasive medical sensing and diagnostic devices based on advanced, microfabricated microsensors.

MRAM AND THE HALL EFFECT

Magnetic random access memory (MRAM) is a non-volatile, radiationhardened memory chip that uses gallium arsenide (GaAs) instead of the silicon found in conventional computer chips. MRAM relies on micrometerscale sensors to detect the magnetic state of the region that stores the computer's 1s and 0s, which represent "on/off," respectively. For years, researchers have attempted to develop MRAM technology as a candidate for "universal memory," an exceedingly fast, low-power, highly-compact, nonvolatile memory system.

While working with MRAM, Micromem engineers noted the extraordinary properties of the on-chip sensors, a new twist on an old design known as the Hall effect sensor.

The Hall effect was discovered in 1889 by Edwin Hall, while he was a doctoral student at Johns Hopkins University. The sensor's design is straightforward, but its functions and operation are quite elegant.

Imagine a square or rectangular sheet of thin metal with a current passing through from one side to the opposite side. Under normal conditions electrodes placed at 90° to the input current will measure zero potential (voltage). However, i f a magnet ic f ield is made t o pass through the semiconductor, perpendicular to the applied voltage, it perturbs the current flowing through the metal such that the output voltage will be non-zero. The magnitude of this voltage is directly proportional to both the applied magnetic field and the applied voltage.

The Hall effect, as this phenomenon is called, had little application outside of theoretical physics for 70 years after Hall's initial demonstration. The discovery of semiconducting materials, such as silicon in the 1950s, opened the door to the first Hall effect sensor applications, but these were severely limited by the cost of the devices.

In 1965 engineers at Micro Switch created the first practical miniaturized Hall effect sensors that were contained on a single silicon chip. These devices were the first low-cost, high-volume application of the Hall effect.

When packaged into a "chip" Hall sensors are protected and extremely durable, last ing for bi l l ions of measurements. They are fast, low-cost, have no moving parts, are protected from sometimes harsh environments, switch at high frequencies, and have hundreds of applications. Uses include sensing position, proximity, motion, rotation, electrical current, magnetic fields and many others. Industries such as defense, aerospace, automotive, power, mining, security/biometrics, oil, manufacturing, consumer, and healthcare rely heavily on Hall effect sensors.

BREADTH OF APPLICATIONS

Miniaturization is one of the most impor tant trends in electronics because one can squeeze a great number of smaller objects into a given volume than larger objects.

Serendipitously, the smaller Micromem made its sensors, the greater their sensitivity. This led to the idea that the sensors could be used for specialized measurements and sensing applications.

Micromem's initial application for Hall sensors was metals and mineral exploration, a logical choice since all metal-containing substances have a distinct, unique magnetic signature.

Hall effect sensor arrays could provide answers virtually instantly, compared with sending samples off to a remote laboratory for analysis.

It fact metals are not the only materials with magnetic signatures. Molecules such as cholesterol, glucose, and others all bear a characteristic electronic signature that can be detected and quantified using miniaturized arrays of ultra-sensitive sensors. Within the healthcare sensing market, Hall cross sensing makes the most sense in diagnostics; for example, in blood/ tissue testing and as specialized magnetic imaging devices.

When Micromem began thinking of life science applications of Hall effect sensors there were no real precedents for the type of devices it envisioned, namely sensors and sensor arrays for non-invasive diagnostic testing of blood, body fluids, and tissue for medical conditions including diabetes and hypoxia.

Commercializing the glucose sensor will be an engineering challenge as well. Since the magnetic signature from blood components is minuscule, the sensing surface must contain approximately 5,000 Hall sensors arrayed in a manner that concentrates the signal. Each sensor has either four or six leads, two of which connect to a signal processor. That is a lot of "wiring" for such a small device.

Medical sensors also pose unique regulation challenges, as devices used in human diagnostics in America must be approved by the U.S. Food and Drug Administration.

While medical sensors are held to very high standards of sensitivity and reliability, historically their acceptance in the marketplace has been performance-driven rather than cost-driven. While performance requirements remain high, such devices will increasingly compete on the basis of value due to ever-rising healthcare costs.

From a product development perspective, however, Hall effect sensors hold great potential based on their lower cost compared with conventional Hall sensors.

NEXT-GENERATION SENSING

Micromem's Hall sensor design, a next-generation "cross sensor," is a byproduct of more than a decade of R&D on MRAM. Instead of silicon, the Hall cross sensor uses gallium arsenide (GaAs), an advanced, highly-efficient semiconductor that maximizes the device's sensitivity and accuracy in one of the smallest packages possible.

Table 1 shows that GaAs is the ideal material for Hall sensor fabrication. Compared with other common semiconductor materials, GaAs optimizes both electron mobility and energy band gap. The former provides greater sensitivity, while the latter reduces temperature drift. This is significant because traditionally, Hall sensors required some type of temperature stability in order to provide reliable measurements. GaAs also provides enhanced stability and sensitivity across a very large operating temperature (up to 300° C), and allows for self-calibration.

Sensitivity of an un-optimized Hall cross sensor is shown in Figure 1, compared with a typical miniaturized Hall sensor and an optimized version of the cross sensor.

The cross sensor offers significant improvements over existing magnetic sensors, including enhanced sensitivity, accuracy and noise-reduction; an extended temperature range; resistance to harsh envi ronments; simplified implementation; significantly higher device density; and lower cost.

Recently, the Hall cross design has been implemented in a novel format, the "sensor lamina," where the sensors are embedded between very thin layers of a film material such as glass, plastic or metal, lowering costs while simplifying environmental sealing.

TWO UNMET MEDICAL NEEDS

Ongoing partnerships with medical device companies will hopefuly lead to commercialization of two exciting medical sensing applications of Hall cross sensors: hypoxia and diabetesrelated molecules in the blood.

The first potential product, for hypoxia (low blood oxygen), is based on the fact that hemoglobin is diamagnet ic ( repel led by a magnetic field) when oxygenated, but paramagnetic (magnetized only in the presence of a magnetic field) when it is not. A Hall-based hypoxia sensor would need to incorporate a circuit for emitting a short magnetic pulse to "excite" the hemoglobin molecules, and measure the magnetic decay in non-oxygenated hemoglobin over time using the Hall circuit. By also measuring the diamagnetic proper t ies of the ox ygenated hemoglobin molecules, the device can determine the ratio of oxygenated to non-oxygenated species, and thus the degree of oxygenation.

The company believes it can also measure concentrations of hemoglobin A1C (HA1c) using Hall cross sensor arrays. HA1c is a useful measure of the degree of glycemic control over time (usually several weeks) for diabetics. Still another medical device par tner is investigating the possibility of quantifying blood glucose through a similar mechanism.

The non-invasiveness of medical testing with Hall sensors is a significant benefit, as any diabetic will attest.

Taking blood samples for routine glucose monitoring could one day be a thing of the past. Another huge plus is the ability to take measurements on the fly, at any time during the day, between or even during meals.

Micromem Applied
Sensor Technologies Inc.
Toronto, Ontario, Canada
micromeminc.com