Magnetometers measure a magnetic field or a magnetic dipole moment. A typical magnetometer is a compass. It measures the direction of a surrounding magnetic field, in this case the Earth’s magnetic field. Various types of magnetic sensors, on the other hand, detect the direction, strength, or relative change of a magnetic field at a particular location.
Magnetometers have been used in the automotive industry and industrial applications for several decades; in consumer electronics, they are standard in smartphones, wearables, and augmented reality/virtual reality (AR/VR) glasses, for example, as well as in drones and robots, smart home devices, and IoT applications. In addition, there are exciting new application areas such as head orientation for 3D audio, improved indoor navigation, positioning, and speed detection. One way to implement these is with Hall sensors.
Hall sensors take up a large market volume
A Hall sensor is a sensing element for detecting the Hall effect or Hall voltage. Hall sensors can be traced back to a discovery made by Edwin Hall. In 1879, the scientist found that a magnet placed perpendicular to a current-carrying conductor pulls the electrons flowing in the conductor to one side, creating a charge difference (i.e. a voltage). The Hall effect is thus an indicator of a magnetic field near a conductor and its strength. It is used in sensors to indicate the presence, absence or strength of a magnetic field based on the resulting Hall voltage. Today’s highly integrated Hall sensors incorporate various sensor signal conditioning functions, such as a differential array of Hall elements, instrumentation amplifiers, A/D converters, and even MCUs (depending on the version). So, although Hall sensors work by detecting a magnetic field, they can be used to measure many parameters, such as position, temperature, current, and pressure.
Due to their highly developed and low-cost production, Hall sensors have long had a significant market volume. They essentially consist of a thin piece of a rectangular p-type semiconductor material, such as gallium arsenide (GaAs), indium antimonide (InSb), or indium arsenide (InAs), through which a continuous current flows. When the sensor is in a magnetic field, the magnetic flux lines exert a force on the semiconductor material that deflects the charge carriers, electrons, and holes to either side of the semiconductor wafer. This movement of the charge carriers results from the magnetic force they experience as they pass through the semiconductor material. The output voltage of the Hall element, known as the Hall voltage (UH), is proportional to the strength of the magnetic field penetrating the semiconductor material (output: a H). However, such silicon-based Hall sensors have limited output power, low accuracy, and a large offset.
AMR sensors with limited applications
An alternative to the Hall sensor is the anisotropic magnetoresistance (AMR) sensor. A magnetoresistance (MR) changes the electrical resistance of a conductor due to a magnetic field. When the electrical resistance decreases due to the magnetic field, it is called negative magnetoresistance.
Two definitions of percent magnetic resistance are commonly used: MR0 is defined as the difference between the resistance with a magnetic field and the resistance without a field divided by the resistance without a field.
MRP, on the other hand, is the difference between the resistance with a magnetic field and the resistance in the saturated field divided by the resistance in the saturated field. The maximum value can be arbitrarily large.
The AMR effect was discovered in 1856 and first used as a transducer to read magnetic tape in 1971. Honeywell developed magnetic random access memory (MRAM) based on the AMR effect.
An AMR sensor can also be used as a compass to measure the Earth’s magnetic field. Other than that, its applications are somewhat limited. That is because although several semiconductor suppliers offer a range of AMR sensors, their magnetoresistance is typically less than five percent. Conventional AMR sensors also require additional circuitry or permanent magnets to restore the magnetization of the thin film after use. This complicates the packaging and adds extra costs.
GMR sensors for many applications
Then there is the giant magnetoresistance (GMR) effect, which Peter Grünberg and Albert Fert independently observed in 1986 as unusual magnetoelectronic behavior in Fe/Cr/Fe layers. Both were awarded the 2007 Nobel Prize in Physics for this discovery.
When two iron layers are ferromagnetically coupled via the non-magnetic chromium layer, the resistance is low because the electrons can transfer to the second iron layer without changing their spin. The MR ratio in the metallic spin-valve structure is typically about ten percent.
IBM soon used GMR sensors as magnetic read heads in hard disk drives to achieve higher storage capacities. GMR sensors are now being used in a variety of other applications as well.
TMR technology drives innovation
However, the development of magnetometers did not end there; tunnel magnetoresistance (TMR) magnetic sensor technology has since been innovated. It is more accurate, exhibits less noise, and consumes less power than previous magnetometer technologies. Thanks to these characteristics, it is set to increasingly replace Hall sensors.
The discovery of the TMR effect also opened up further possibilities for the use of magnetoelectronic phenomena in the computer industry, such as non-volatile data storage based on the MR effect in layered systems. The technical development of this MRAM can be traced back to IBM, among others. The first products came on the market around 20 years ago. Today, all modern hard drives use TMR read/write heads.
MRAMs combine the advantages of semiconductor memory – fast access times – and of magnetic materials – high storage density. In addition, these non-volatile memories are robust, energy-autonomous, and radiation-resistant. MRAMs also offer non-destructive reading and can even store data without power.
Currently, data storage with dynamic random access memory (DRAM) is still the predominant technology. However, it has the disadvantage of losing data in the event of a power failure. In addition, the storage systems require regular refreshing to prevent data loss. Although it appeared that silicon semiconductors in DRAMs were gradually being replaced by TMR technologies, MRAMs are still found only in niche applications and are still waiting for their commercial breakthrough. In recent years, however, their market share in the automotive, consumer, and industrial markets has grown disproportionately compared to other technologies such as Hall, AMR, and GMR.
The TMR effect
The TMR effect is based on an arrangement comparable to the GMR effect. It was first discovered by Michel Jullière in 1975 in Fe/Ge-O/Co junctions at 4.2 K. The relative resistance change was about 14 percent and did not attract much attention. In 1991, Terunobu Miyazaki found a change of 2.7 percent at room temperature. Three years later, Miyazaki found 18 percent in compounds of iron separated by an amorphous alumina insulator. Jagadeesh Moodera measured 11.8 percent in compounds with electrodes of CoFe and Co.
Unlike GMR, which has a non-magnetic layer, TMR involves inserting a non-conductive layer between two magnetic layers. This is done with a magnetic tunnel junction, a component consisting of two ferromagnets separated by a thin insulator (Fig.1).
If the insulator layer is thin enough (typically a few nanometers), electrons can pass through the tunnel barrier from one ferromagnetic layer to the other. The probability of this also depends on the spin, leading to high MR values for parallel vs. anti-parallel magnetization of the spins in the magnetic tunnel junction layers. The largest effects are expected for materials with fully spin-polarized electrons.
Since the tunneling process is forbidden in traditional physics, TMR is a quantum mechanical phenomenon. The direction of the two magnetizations of the ferromagnetic layers can be changed by an external magnetic field. If the magnetizations are aligned in parallel, the electrons are more likely to tunnel through the insulating layer than if they are aligned in opposite (anti-parallel) directions. This means that it is possible to switch between two states of electrical resistance, one with low resistance and one with high resistance (Fig. 2).
Structure of a thin film stack
The TMR effect can be used for many applications. However, doing so requires building a thin film stack . The trick is to have only one free ferromagnetic layer.
The magnetic tunnel junction (MTJ) in Figure 3 uses what is called exchange coupling. This TMR structure is an MTJ multilayer between two electrodes in a geometry where the current flows perpendicular to the plane. The complex stack consists of double exchange electrodes composed of a bottom electrode, a bottom anti-ferromagnet (AFM), a pinned layer (PL), a spacer, a reference layer (RL), a tunnel barrier, a sensing layer (SL), and the top electrode.
To increase the exchange field and to make the MTJ more thermally stable, a synthetic anti-ferromagnetic (SAF) structure can be used instead of a single ferromagnet (FM) in the pinned layer adjacent to the AFM. The SAF structure consists of two or more FM layers separated by thin ruthenium layers and coupled by the RKKY interaction. To fix the magnetization of the pinned layer in one direction, the exchange coupling between the FM and AFM layers is employed. Only magnetic fields above the exchange field can reverse the magnetization of the pinned layer. The arrows in Figure 3 indicate the direction of the magnetization and the applied magnetic field.
The rate of change of the resistance of a multilayer stack was introduced as the MR ratio. Here, the MR values of conventional AMR and GMR elements are about five and ten percent, respectively. For the much more sensitive TMR element, it is 100 percent or more.
Why is the TMR so sensitive? As described, the GMR element consists of a non-magnetic metal (e.g. copper) sandwiched between two ferromagnetic layers. Electron transfer occurs by electrical conduction in the metal. In a TMR element, however, electron transfer occurs through a quantum mechanical tunneling effect. Therefore, when the pinned layer and the free layer are anti-parallel, a TMR element has an exciting property. This means the electrons are blocked and cannot pass into the tunnel barrier. In a GMR, on the other hand, it is difficult for the electrons to pass through the non-metallic barrier. As a result, a TMR element has an extremely large MR ratio and outputs very clear signals, e.g. yes/no or 1/0, depending on the spin polarization of the metals used.
New magnetometer based on TMR technology
The new BMM350 3-axis magnetometer from Bosch Sensortec is based on this TMR technology. Its much higher sensitivity compared to standard Hall, AMR, and GMR sensors results in significantly greater measurement accuracy. In addition, TMR sensors have better temperature stability and offer a faster response time (Fig. 4).
Thus, wearables and hearables, smartphones and tablets, AR and VR devices as well as vehicle applications can be achieved and improved with the BMM350. Due to its small size, the magnetometer is almost invisible: The wafer-level chip-scale package (WLCSP) measures only 1.28 mm × 1.28 mm × 0.5 mm.
Compared to the previous generation (BMM150), the BMM350 offers significantly improved performance. Its average power consumption is only 200 μA at a data rate of 100 Hz, which is twenty times lower than its predecessor. Noise on the x/y axis is three times lower, and measurement sensitivity is four times more accurate than with the BMM150. Its field shock recovery function makes the BMM350 very robust to external magnetic fields, ensuring high-level accuracy at all times.
The list of possible applications for TMR sensors, like the BMM350, is long. As position sensors (with one, two, or three axes), they can measure rotation or linear motion or the Earth’s magnetic field as a compass.
In hearables, the BMM350 improves head orientation and recognition for 3D audio applications. In this case, the combination with inertial sensors and intelligent fusion software compensates for the rotation rate drift that always occurs. In commercially available AR and VR headsets, it is important that the magnetometer is combined with the accelerometer and the rotation rate sensor to reduce pixel latency. This improves the user experience and prevents nausea.
For indoor navigation where a GPS signal is not available, the BMM350 can act as a digital guide and increase position accuracy.
Its capability for speed measurement is not only of interest for automotive applications – with back-biased magnets or magnetic encoders, the TMR sensor can also measure the wheel speed on e-bikes.
Current measurement is another interesting application for TMR sensors. As non-invasive current measuring elements, they are ideal for many applications including in power distribution, power electronics, and drive technology. This is because they offer higher sensitivity and linearity than Hall, AMR, and GMR sensors.
In addition, they are stable, small, and highly integrable and feature low power consumption and typically a wide frequency bandwidth.
Summary
TMR technology, as used in Bosch Sensortec’s BMM350, enables a better user experience for numerous applications as well as completely new, exciting use cases that cannot be implemented with other technologies.
For more information and a direct ordering option, please visit our e-commerce platform at www.rutronik24.com.
Subscribe to our newsletter and stay updated.
