XYNETIQ Sensor Physics

Sensor physics focuses on converting physical, chemical, or biological parameters, such as pressure, motion, light, or temperature into measurable electrical signals, often utilising semiconductor technology. Modern sensor technology is characterised by miniaturisation, high-speed signal processing, and AI integration for self-calibration, enabling applications in IoT, wearable devices, and biomedical fields.  Sensors based on the actions and interactions of atoms at an electrical interface leverage atomic-scale phenomena to detect, measure, and image physical quantities such as electric and magnetic fields and molecular structures with extreme sensitivity. These sensors, often categorised as atomic or quantum sensors, operate by manipulating atomic states with light and translating environmental changes into readable electrical or optical signals.

E-Nose Sensor Types Summary

Key types of sensors operating at this interface include Rydberg atom sensors, atomic vapour magnetometers, and molecular electronic sensors. 

Rydberg Atom Electric Field Sensors

Rydberg sensors utilise alkali atoms (such as Rubidium or Caesium) excited to high principal quantum numbers (n>50), making them highly sensitive to external electric fields.  An external electric field induces Stark shifts or Autler-Townes (AT) splitting in the Rydberg energy levels. This shift is detected via Electromagnetically Induced Transparency (EIT) using laser spectroscopy.

Metal Oxide Semiconductor (MOS) Sensors

The most common type of these sensors uses metal oxides like tin oxide.

Conducting Polymers (CP)

These utilise organic polymers, such as polypyrrole, to detect gases via changes in conductivity, typically operating efficiently at room temperature.

Mass-Sensitive Sensors (Piezoelectric)

These measure the mass of gas adsorbed onto a sensor surface.

Quartz Crystal Microbalance (QCM)

Measures frequency shifts in a quartz crystal.

Surface Acoustic Wave (SAW)

Highly sensitive, low-noise sensors suitable for VOC detection, often used in medical breath analysis.

Field Effect Transistor (FET) Sensors

These modulate current flow through a semiconductor channel based on the adsorption of target compounds, offering high sensitivity for detecting low concentrations of VOCs.

Optical Sensors

These detect odours by measuring changes in light properties, such as fluorescence or absorbance, upon exposure to gas samples.  Lasers can be used in these sensors as well as NIR, IR, UV, and VIS.& 

Electrochemical Sensors

These convert chemical reactions occurring at an electrode surface into measurable current or voltage changes. 

Key Considerations

Array Composition

E-noses frequently use a combination of different sensor types to enhance classification accuracy, often reaching >99% in experimental studies.

Performance Metrics

While MOS sensors are strong in sensitivity and fast response times, they require high operating temperatures, unlike CP sensors that work at room temperature.

Advanced Materials

New developments include nanotechnology-based sensors (such as carbon nanotubes or graphene) to further improve detection limits. This is an ongoing research program.

Engineering

E-Nose design is equally critical of electronics and mechanical hardware, dimensionally critical component parts and the locating of sensor elements. Airflow past the sensor array must be controlled and measured during sampling, where pumped systems are used, and atmospheric sensing must be volume/time controlled; again, precision is needed to obtain accurate results.

Xynetiq favour precission engineered internal structure from a solid aluminium block to enable the flow-paths to\ be dimension exact and thermally stable, inert to the gas flow. This is obtained by coating, anodising, or film on the block, which achieves maximum integrity for the sensor block.  The electronics are then located within the block beneath the sensor array, so the main block also offers shielding from electrical noise. The unit is then coated with an RFI/EMI layer to increase the shielding and further increase the SNR (signal-to-noise ratio), improving the sensitivity of the sensing electronics.

Physical Properties
 

E-Noses vary in physical size depending on the application and type of sensor array. Some have a single technology array, and others can have multiple sensor arrays for a broadband sensing range of odours (gases).  E-Noses can measure from 50mmx50mmx30mm, up to 200mmx150mmx35mm, and tubular types can vary between 10mmODx30mmL, up to 60 mm OD x 300 mm L. Virtually any size and shape can be accommodated in a custom housing option to meet customer requirements.

Connection Interface

Connection options include wireless communications data link, Satellite, 4G/5G, LoRa IoT, BT, and GPS location.  Cabled using POE, Ethernet, RS485, CAN, USB, and RS232 Serial,  cloud connection and M2M. Other sensors such as accelerometers, temperature, humidity, CAM, Drone uplink, and remote controls.

Other Sensor Technologies

Carbon nanotube (CNT) 

Sensors are highly sensitive devices utilising the unique electrical and structural properties of cylindrical graphene sheets to detect gases, chemicals, and biological molecules. Due to their high surface-to-volume ratio, they excel at detecting low-concentration analytes, offering applications in environmental monitoring, healthcare (e.g., breath analysis), and flexible wearable devices.

Breath Analysis

Researchers have developed CNT-based sensors to detect volatile organic compounds, acting as an "artificial olfactory system" to diagnose diseases.

Light-Activated Sensing

Recent developments include high-performance oxygen sensors activated by light, useful for environmental and medical monitoring.

Wearable Health Monitoring

CNTs are increasingly used for flexible sensors that track bodily functions, offering potential for early health disorder detection

Sensing Mechanism

CNT sensors generally function as resistors or field-effect transistors (FETs). When molecules interact with the CNT surface, they induce charge transfer, altering the electrical conductance of the sensor.

Types

Sensors use either single-walled (SWCNTs) or multi-walled (MWCNTs) carbon nanotubes, often functionalized with metal oxides or nanoparticles to enhance selectivity and sensitivity.

Performance Benefits

These sensors offer high sensitivity to target gases like NO2  & HO3 and acetone, and they can operate at room temperature.

Applications

Key uses include environmental gas detection, industrial safety monitoring, food safety checks, and medical diagnostics.

Fabrication

Often created by depositing a network of nanotubes onto a silicon substrate with interdigitated electrodes.

Challenges

Main limitations for widespread adoption include high-cost manufacturing and integrating CNTs into large-scale industrial devices. Progress is being made in this technology, and it will be available soon.

Quantum & Magnetic Sensor Technologies 

Other sensing technologies include Magnetic, SQUIDS, Magnetostrictive technology and nuclear electrostatic resonance (NER), which is a method for manipulating nuclear spins using electric fields rather than magnetic fields. It allows for addressing single nuclei in quantum devices while maintaining long coherence times, overcoming limitations of traditional nuclear magnetic resonance.  It works by using localised electric fields to interact with the nuclear quadrupole moment, which exists in nuclei with spin I ≥ 1. Unlike NMR, which affects a large, ensemble sample, NER enables the targeting of individual nuclear qubits within a quantum register. used in quantum information processing, particularly for controlling spin in single molecules, and can be driven via pulsed laser excitation, known as optically detected nuclear electric resonance (ONER). 

Nuclear Quadrupole Resonance (NQR)

This technique uses the electric quadrupole moment to study molecular structure, which is similar in principle to NER but often used for chemical analysis in bulk samples.  NQE can only detect solid state, not liquids, fluidic or gas. One approach for liquids is to use cryogenics to freeze to solid, prior to sampling.

Nuclear Acoustic Resonance (NAR)

Another resonance method that uses acoustic waves to interact with nuclear spins is a very sensitive method but not commonly used except for specialised applications.

RF Sensors

RF-based sensors include proximity, TDR, Doppler, interference, and modulated sensors, all of which have specific application areas and circumstances, including atmospheric, soil, water, movement, level, position, material detection and monitoring.  

Data Processing & Analysis

Sensor arrays can generate lots of information; this is mostly an analogue signal (changing voltage) directly from the sensor.  To obtain any meaningful data from a sensor, it is necessary to clean and amplify the signal(s) and convert them into a digital format, where it is easier to analyse the signals and deduce what they are and what they mean.   A single signal may not be sufficient to enable an accurate analysis because of signal drift, sample variation and other factors, so we take many samples over a short period of time, averaging the outcome and passing it to ML (machine learning).  The machine learning processor, the data is mathematically operated on using algorithms self-generating from a master math file.

This routine is a continuous one in a feedback or circulating core, where AI (artificial intelligence) matches the data and to another library and learns what to output to the user when 'signatures' are received from continuous samples and processed in the ML. As the process continues with more samples over time, the analysis becomes more accurate and grows its signature database from many different odours (samples), either learning or identifying which becomes more accurate over time.  A Base master reference database is placed into the library to start the process, ensuring that it can work to identify from initial switch-on use, it will become faster and more accurate over time as it gathers intelligence.

Sensor design technologies used in some of our sensors incorporate nano technology, hybrid ceramic substrates, SOC, and ML63Q2537/ML63Q2557,32-bit Arm Cortex-M0+ CPUs with an AxlCORE-ODL AI accelerator, for low-power, real-time sensing solutions; other types are used according to application.  Only the most recent developments are applied to our sensors to ensure longevity and high efficiency for the best green ultra-low carbon footprint and best recycling at the end of life.  Materials are a major design consideration and the best green choice for specific designs and applications.

User Data Output

Post data processing, the data will attach headers (with identification data) and send the output to the user's chosen method to receive or display the data on a GUI (graphical user interface) either on a dedicated local screen, PC, mobile or tablet, or directly to the cloud or other server.  The data protocol can be used as a standard type or a customer-specific one that displays or outputs data to a system where it is further processed.

Extreme Operating Environments Capabilities

Our standard sensors are designed for an automotive operating temperature range of -40°C to +125°C, or for extended ranges of Military & Aerospace  -55°C to 125°C. 

Operating environments are key to design, and component choices must follow this. Outside of the Automotive, Military, and Aerospace range, we design for extreme temperatures, for these High-temperature TEKMOS ASICs and customised microcontrollers rated for +200°C to +250°C and VORAGO Technologies VA10800/VA10820 ARM Cortex-M0 based MCU designed for sustained operation at +200°C.  At the lower temperature ranges down to -200°C, and other extreme ranges can be achieved using cooling and cryogenics technology, and ceramic heating technology for applications that are below this.. Few applications require such extreme temperature ranges, and some require Radiation-hardened (rad-hard) ICs, which are specialised electronic components designed to withstand high-energy radiation, such as in space, nuclear, and high-altitude environments. They ensure reliability by resisting Total Ionising Dose (TID) and Single Event Effects (SEE) through hardened design techniques (RHBD), specialised fabrication processes (RHBP).  Environmental considerations usually only apply to a few applications, but we are able to work in some very extreme environmental operating conditions, including vibration and shock, high pressure, vacuum, and ATEX for flammable or explosive environments, all of which are possible.

If you require specific operating features, please specify in the CONTACT form, and we will match your requirements.