2D Arrays of Quantum Well Hall Effect Sensors for Picotesla Magnetometry of Inorganic and Organic Materials.

The principle aim of my PhD project is investigating whether Quantum Well Hall Effect (QWHE) sensors can have their dynamic range extended into the pico-tesla range.

Current technologies and techniques allow the QWHE sensors, developed at the University of Manchester, to reach the very low nano-tesla range. There is a wealth of applications in pico-tesla magnetometry such as; Magneto cardiograms, Archaeology, Geophysical surveying, Body-position tracking, and many more.

Currently all these applications are done using other sensors (all bulky and power hungy). If my project is successful, then QWHE sensors will be usable in this diverse range of applications leading to improvements in their respective systems as QWHE sensors are: light, small, linear over a large range, more sensitive, and low power when compared to other sensors. This could, therefore, lead to improved performance in these different and important fields.

The main limitation on the lower end of the dynamic range is electronic noise. For QWHE sensors this comes in the form of flicker noise, thermal noise, shot noise, and generation-recombination noise.

There are numerous techniques that can circumnavigate noise such as superheterodyne mixing which works by shifting the measuring frequency out of the realms of flicker noise. However, this project seeks to find another method that can be done before super-heterodyne mixing.

The question this project asks is, 'can the sensor performance be improved by parallel connections of QWHE sensors?' This will create an array of sensors that behave as one with a lower limit on its dynamic range. This relies on a very simple part of physics, parallel connections. Connecting components in parallel usually causes their resistance to decrease by a factor of 1/N, where N is the number of sensors.
The question this project then asks is 'Does this apply to the QWHE sensors?' and 'What effect does this have on the measurements?'. The theory then is simple, if the resistance goes down by a factor of 1/N then the overall noise should decrease, as all 4 types of noise previously mentioned are affected by the resistance.

The objectives to achieve this are as follows.

Create a small test circuit of a 2x2 array (4 sensors) and see if the phenomena can be observed.
Extend the test circuit to a larger array. See if the phenomena hold true for larger arrays.
Move the array from the PCB scale to the crystalline level (will allow several thousand sensors to be paralleled together).
Apply superheterodyne mixing and other noise reduction techniques to the arrays to further reduce noise and enter the pico-tesla range.
Find the balance between power demands/dynamic range/sensitivity that creates a useable device for implementation in pico-tesla magnetometry.


The next question is why is this novel? Why can this only be done with QWHE effect sensors? This is mostly due to size and power requirements, sensors such as fluxgates are very large, and it difficult, if not impossible, to array very many. There are sources where groups have used arrays of GMRs in a similar fashion to what this project suggests but they only demonstrate N values up to a couple of hundred, before size and power requirements become prohibitive. QWHE sensors being novel in of themselves are, unlike other sensors, semiconductor devices. This means they boast very small die size (approximately 200 microns by 200 microns with sensing area as low as 5 microns x 5 microns), very low power requirements (milliwatts). QWHE technology therefore lends itself well to the concept of large number arraying.

In conclusion the inherent properties of QWHE sensors lend themselves to large scale parallel arraying, which will greatly reduce their electronic noise. This should increase the dynamic range of the sensors into the pico-tesla range. The PhD intends to investigate these phenomena.