21EE742 – MICRO & NANO SENSORS &

Regulation – 2021 TRANSDUCERS

MODULE - 4

MODULE - 4

Biological, Chemical, and “Lab on a Chip” Sensors: Lab on a Chip Sensor, Other Biochemical Micro-
and Nano-Sensors.
Electric, Magnetic, and RF/Microwave Sensors: Magnetic Field Sensors, Other Important
Electromagnetic/RF Micro- and Nano-Sensors

“Lab on a Chip” Sensors

Chemical and biological sensors have advanced tremendously during the past two decades with the
advancement in nanotechnology and nanofabrication. The state of the art of chemical and biological
sensors is an integrated sensor known as a “Lab on a Chip.” The market for Lab on a Chip product is
expected to reach more than $14 billion in the year 2023. The Lab on a Chip concept is currently believed
to be the ultimate solution for poorly equipped medical facilities and mobile healthcare platforms that
lack the capabilities of a fully equipped laboratory. The Lab on Chip solutions that currently exist can
easily perform diagnostic operations such as microorganism detection and characterization, flow
cytometry applications, polymerase chain reaction (PCR), in addition to many others.

A Lab on a Chip is essentially composed of three sections: actuation, sensing, and electronic interface
circuitry. A block diagram of such a chip is shown in Figure 6.1.

Figure 6.1: General structure of “Lab on a Chip,”

A scanning electron microscope (SEM) micrograph of an actual Lab on a Chip is shown in Figure 6.2. In
the actuation section, the chip generates electrical or mechanical forces that act on the biological sample
(cells/fluid). In the sensing section, sensors that are embedded in the chip measure responses that may be
electrical, optical, thermal, or magnetic, and route the detected signals to the electronic section for final
processing. The electronic interface circuitry finally performs traditional signal processing functions such
as amplification and noise reduction.

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

Figure 6.2: SEM micrograph of the “Lab on a Chip,”

Figure 6.3: Typical interface of a “Lab on a Chip” to a computer data acquisition system

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

Figure 6.4: Neuron of a snail analyzed by a “Lab on a Chip,”

Figure 6.5: Typical packaging of a “Lab on a Chip,”

Figure 6.6: “Lab on a Chip” implemented directly on a printed circuit board

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

Figure 6.3 shows a typical interface of a Lab on a Chip to a computer’s data acquisition system. Figure
6.4 shows a typical advanced application, where the neuron of a snail is tested for various responses by a
Lab on a Chip. Figure 6.5 & 6.6 shows two different methods of packaging Lab on a chip.

Other Biochemical Micro- and Nano-Sensors

Other technologies for chemical and biological sensing have also been introduced recently. Figure 6.7
shows a CMOS on-chip sensor for measuring the dielectric constant of organic chemicals.

Figure 6.7: CMOS on-chip sensor for measuring the dielectric constant of organic chemicals

Figure 6.8: Porous silicon based sensor for chemical gas vapor detection

Figure 6.8 shows a concept for a porous silicon based sensor for chemical gas vapor detection

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

ELECTRIC, MAGNETIC, AND RF/MICROWAVE SENSORS

Magnetic Field Sensors

 A new type of solid-state magnetic field sensor that is similar in size to a Hall effect sensor
and that offers a sensitivity that is approximately an order of magnitude better than a Hall
effect sensor was recently introduced.
 The significant advantage of the new sensor, however, is that it consumes no power. The
sensor consists of a radioactive β -particle source and a silicon p-n junction. If no magnetic
field is applied, the β particles enter the p-n junction and generate a steady DC voltage.
 Under the influence of a magnetic field, however, the β particles (or secondary electrons
generated therefrom) follow a curved path and miss the p-n junction, and the magnitude of
the output voltage drops.
 The new sensor will be very advantageous in battery-powered consumer products
applications since it consumes no power.

INTRODUCTION AND PRINCIPLE OF OPERATION

 Magnetic field sensing is an active research area and the recent advances include techniques
for detection of the magnetic force that acts on MEMS components such as magnetic-tunnel
junctions, Ferro-fluids that are embedded inside optical fibers, giant magneto-resistance
sensors in addition to improvements to the traditional Hall effect technique.
 At the present time, however, the magnetic field sensor of choice in most industrial
applications is still the traditional Hall effect sensor due to its high sensitivity, small size, and
low cost. The Hall effect sensor is also increasingly being used in portable electronic devices,
such as digital cameras and smart tablets.
 Unfortunately, the traditional Hall Effect sensor requires a current on the order of a few tens
of mA to achieve its typical sensitivity, or about 10−4T.
 This current requirement is undesirable in portable equipment, as it results in a shorter
battery life.
 Although new “micro-power” Hall Effect sensors have been introduced recently by a number
of manufacturers, the magnetic field detection sensitivity is typically sacrificed as a result of
the reduction in the current consumption of the device.

The solid-state magnetic field sensor that is characterized by:

 Zero power consumption. The sensor, however, must be interfaced to signal conditioning
circuitry like a traditional Hall Effect sensor. Such circuitry can suitably be low-power
circuitry.
 Size that is comparable to a typical Hall effect sensor, with a magnetic field detection
sensitivity that is at least an order of magnitude better than that of a typical Hall effect
sensor (or about 10−5 T).

The new sensor depends on the beta-voltaic principle. Figure 7.1 shows the fundamental
principle of operation of the new sensor. In Figure 7.1(a), a source of low energy b particles
is positioned above a p-n junction. As the β particles enter the p-n junction, electron-hole
pairs are produced inside the junction, and a voltage appears across the terminals of the
junction. In Figure 7.1(b), a magnetic field is applied in a direction that is perpendicular to

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

the direction of motion of the β particles (or high-speed electrons). As a result, the β
particles follow a curved path and miss the p-n junction, and the voltage across the
terminals of the junction therefore drops in magnitude.

Figure 7.1: Fundamental principle of operation of the magnetic field sensor

Two important general remarks about the principle shown in Figure 7.1 are now in order:

The β-particle source used in the new sensor is tritium, with a half-life of 12.3 years. Such a
half-life will be acceptable for portable consumer products, which typically have
comparable lifetimes. β particles are high-speed electrons.
Even if the source is a low-energy source such as tritium, the velocity of the β particles is
substantially high (exceeds 10 7 m/s). According to the Lorentz force equation, however,
such a high-speed particle cannot respond quickly to a magnetic field. For this reason, the
fundamental principle shown in Figure 7.1 must actually be implemented with some
important changes. Figure 7.2 shows the actual structure of the sensor.

Figure 7.2: Cross-sectional view of the actual sensor (upper) and a projection view of the
two n-doped regions (lower)

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

 As shown in Figure 7.2, the β-particle source is mounted immediately above a

semiconductor body consisting of a p-type substrate and two n-doped regions.
 The two n-doped regions have the shape of concentric circles (as shown in the projection
view in the lower part of the figure) and are physically separated from each other.
 Two separate p-n junctions therefore exist in the device. The p-region has one metal contact,
as shown in the figure, while the two n-regions have separate metal contacts.
 β particles emitted from the decay of tritium have a maximum energy of 18.6 keV and an
average energy of 5.7 keV. These energetic β particles enter the p-type region, where they
lose their kinetic energy within a distance of less than 1 μm.
 If no magnetic field is applied, the newly created EHPs drift toward the depletion region in
the center, where the internal electric field in the depletion region forces the electron
toward the n-type material and forces the hole back into the p-type material.
 As a result, a voltage V1 appears across the terminals of the inner p-n junction, while the
voltage V2 across the outer p-n junction will be equal to 0. If, however, a magnetic field is
applied in a direction that is perpendicular to the direction of motion of the β particles, the
newly created EHPs (which have slow, nonrelativistic velocities) follow curved paths and
accumulate on the sides of the p-substrate, as shown in the figure.
 In reality, the path of an electron inside a semiconductor is very irregular, due to the
numerous elastic and inelastic collisions that the electron encounters within the lattice.
However, the overall effect of the magnetic field on the path will be as shown in Figure 7.2.
 Accordingly, the outer p-n junction receives a larger portion of the EHPs and the voltageV 2
increases, while V1 falls. The interface circuit of the sensor is designed to detect the voltage
differential V2−V1.

Figure 7.3: The two components of the sensor: the tritium b source and the two concentric silicon
p-n junctions. The diameter of the larger component is 12 mm. One US penny is shown in the
photograph for size comparison.

Figure 7.3 shows a photograph of the two main components of the sensor: the tritium β-particle
source and the two concentric silicon p-n junctions. The component consisting of the two p-n
junctions has a diameter of 12 mm and a total thickness of approximately 1 mm.

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

Theory
Range of the β particles in silicon

β particles emitted from the decay of tritium have a maximum energy of 18.6 keV and an average
energy of 5.7 keV. The maximum range of penetration of a b particle into any material can be found
from the Katz–Penfold formula

𝑅𝑚𝑎𝑥 = 0.412𝐸𝛽
1.265−0.0954ln⁡(𝐸𝛽 ) (7.1)

Where Rmax is the material-independent maximum range (in g/cm2) and Eβ is the initial energy of
the β particle (in MeV). The maximum penetration distance in silicon will therefore be given by

𝑅𝑚𝑎𝑥 (7.2)
𝑅𝑠𝑖𝑙𝑖𝑐𝑜𝑛 = ⁡
𝜌

Where 𝜌 = 2.33 g/cm3 is the density of silicon. For an average kinetic energy E β = 5.7 keV, the
formula predicts a penetration depth in silicon of 0.2 μm. The penetration depth is therefore very
small. However, each b particle generates numerous EHPs during its short journey, as is well-
known from the theory of betavoltaics.

Bending radius of the generated free electrons

 The focus now is on the free electrons that will be generated in the p-region of the junction
(see Figure 7.2).
 We shall obtain expressions for the minimum initial kinetic energy of the electron, its final
kinetic energy, and the radius of the arc along its trajectory. These expressions will then be
used to calculate the expected displacement of the electron in the horizontal direction.
 The electron mainly loses kinetic energy as it encounters inelastic collisions inside the
material. Kinetic energy is also lost due to the Lorentz forces that will be acting on the
electron in the direction opposite to the direction of motion.
 This last portion of the kinetic energy is the only portion that can be practically calculated,
and we shall designate it as the “minimum” kinetic energy that will be needed in order for
the electron to reach the depletion region. This minimum kinetic initial energy will be given
by
1 (7.3)
𝑚𝑣 2 = ⁡ ∫ 𝐹𝑑𝑙
2 𝑜

Where F is the Lorentz force, given by 𝐹 = 𝑞(𝐸⃗ + ⁡ 𝑣 × 𝐵⃗ ). Here, as usual, m is the electron’s
mass, 𝑣 is the electron’s initial velocity, q is the electron’s charge; 𝐸⃗ is the steady-state
electric field that will be present between the terminals of the device. 𝐵⃗ ⁡is the magnetic flux
density of the applied magnetic field, and l designates the path of integration. The above
equation can therefore be written as

1 (7.4)
𝑚𝑣 2 = ⁡ ∫ 𝑞 (⃗⃗𝐸 + ⁡𝑣
⃗ × ⃗⃗⃗𝐵) 𝑑𝑙 = ⁡𝑞 ∫ (⃗⃗𝐸) 𝑑𝑙 + 𝑞 ∫ (⁡𝑣
⃗ × ⃗⃗⃗𝐵) 𝑑𝑙
2 𝑜

Since the magnetic force is always perpendicular to the path, it can be immediately seen
that the second integral vanishes, and the above equation reduces to
Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

1 (7.5)
𝑚𝑣𝑜2 = ⁡𝑞𝑉
2

Where, V is the steady-state electrostatic potential or voltage between the terminals of the
device. From Eq. (7.5), the minimum initial velocity v 0 will be given by

(7.6)
2𝑞𝑉
𝑣𝑜 = ⁡ √
𝑚

The terminal (or final) kinetic energy of the electron as it approaches the depletion region
can be found as follows: at steady-state the sum of the electron and hole currents must be
equal to zero (see Figure 7.2). By using the well-known equation

𝐽 = ⁡𝑛𝑞𝑉𝑒 (7.7)

Where J is the current density, n is the free-electron density in the p-region, and ve is the
terminal velocity of the electron, we must therefore conclude that

𝑛𝑞𝑉𝑒 + 𝑝𝑞𝑉ℎ = 0 (7.8)

Where p is the hole density and Vh is the hole’s terminal velocity. Hence,
𝑝 (7.9)
𝑉𝑒 = 𝑉
𝑛 ℎ

Holes will be directly driven by the steady-state electric field present between the terminals
of the device, and Vh will be therefore given by

𝑉ℎ = 𝜇ℎ 𝐸 (7.10)

Where 𝜇ℎ ⁡is the hole’s mobility. Hence,

𝑝 𝑝 𝑝 𝑉 (7.11)
𝑉𝑒 = 𝑉ℎ = ⁡ 𝜇ℎ 𝐸 = ⁡ 𝜇ℎ ( )
𝑛 𝑛 𝑛 𝐿

Where, L is the thickness of the device. In this application, electrons and holes are created in
pairs. Since only these mobile carriers contribute to the current in the device, n = p,
accordingly,

𝑉 (7.12)
𝑉𝑒 = ⁡ −𝜇ℎ ( )
𝐿

The negative sign merely indicates that the direction of motion of the electrons will be
opposite to that of holes. The kinetic energy of the electron takes the form

1 1 (7.13)
𝑚𝑣 2 = ⁡ 𝑚𝑣𝑜2 𝑒 −𝛼𝑙
2 2

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

Where α is an attenuation constant. Since the total length l of the electron’s path is only
slightly different from the overall thickness L of the device, then, to a very good
approximation,

1 𝑣𝑒2 (7.14)
𝛼 ≈ ⁡ −⁡ ⁡𝑙𝑛 ( 2 )
𝐿 𝑣𝑜

where v0 and ve are the final and the terminal velocities of the electron as previously
indicated. Equation (7.13) therefore allows the calculation of the velocity v at any point
along the electron’s path by knowledge of v 0, ve, and the arc length l. We are finally ready to
obtain an expression for the bending radius of the electron’s path. Along its path, the
electron is in equilibrium due to the equality of the magnetic and the centrifugal forces, that
is,

𝑚𝑣 2 (7.15)
𝑞𝑣𝐵 = ⁡
𝑅

Where, R is the arc’s radius and where only the magnitudes are shown in the equation.
Hence, from the above equation and by using the results of Eqs. (7.13) and (7.14), R will be
given by

𝑚𝑣 𝑚𝑣𝑜 1 𝑣𝑒 (7.16)
𝑅 =⁡ =⁡ ⁡𝑒𝑥𝑝 [ ln ( )]
𝑞𝐵 𝑞𝐵 𝐿 𝑣𝑜

The bending radius (or radius of curvature) can therefore be calculated at any point by
knowledge of the initial velocity v0, the final velocity ve, and the arc length l. Since only the
minimum value of v0can be estimated, then the radius of curvature so calculated will be the
minimum radius of curvature. However, the final value of that radius can actually be
determined with good accuracy: it can be immediately seen from the above equation that
when l = L, the final radius of curvature will be given by
𝑚𝑣𝑒 (7.17)
𝑅 =⁡ ⁡
𝑞𝐵

Numerical Data and Experimental Results

Basic data:
The prototype sensor shown in Figure 7.3 was experimentally tested and was found to have
the following parameters:

 Concentration of radioisotope in the source silicon layer: 8.64 mCi / cm 2

 Nominal voltage across the inner p-n junction: 0.8 V
 Max current at rated voltage: 2.5 nA
 Minimum detectable magnetic flux density B: approximately 10 −5 T

Figure 7.5 shows the voltage V1 that appears across the inner p-n junction as a function of time
(with no applied magnetic field). The voltage V2 across the outer p-n junction was determined to be
practically equal to 0 while no magnetic field is present. It is to be pointed out that the magnetic flux
density of the earth is typically in the range of 25–65 μT, which is above the threshold of sensitivity

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

of the sensor. Therefore, the magnetic field of the earth was shielded by placing the experimental
setup inside a nickel-iron enclosure.

Figure 7.5: The voltage V1 produced by the inner p-n junction as a function of time (with no applied
magnetic field). The magnetic field of the earth was shielded by placing the experimental setup
inside a nickel-iron enclosure.

As Figure 7.5 shows, noise with a magnitude of about 3.5% is present in the voltage produced. This
noise is due to the normal fluctuation in the β-particle flux that is reaching the p-n junction (at
lower voltages, the noise was found to still be about 3.5% of the magnitude of the voltage
produced). The bandwidth of the noise was found to be approximately 500 Hz.

Response of the sensor to DC magnetic fields

A magnetic flux with a density B ranging from 10−6 T to 0.1 T was applied to the sensor.
The magnetic field was generated with a simple, custom-made electromagnet that was placed in the
vicinity of the sensor.
Once again, the magnetic field of the earth was shielded by placing the experimental setup inside a
nickel-iron enclosure. The experimental setup is shown in Figure 7.7. Figure 7.6 shows the
measured voltages V1 and V2 across the inner and outer p-n junctions, respectively, as the field was
increased from 10−6 T to 0.1 T.
The theoretically predicted value of V1 was also calculated by calculating ymax from Eq. (7.18) and
assuming that V1 will be linearly proportional to that deviation. The theoretical and the
experimental results are both shown in Figure 7.6. The data in Figure 7.6 was collected at room
temperature (25◦ C).

𝑙 1 𝑙 (7.18)
𝑥 (𝑙 ) = 𝑎 [1 − ⁡ ]⁡cos ( 𝑙𝑛 [1 − ⁡ ])
𝑅0 𝑏 𝑅0
𝑙 1 𝑙
𝑦(𝑙 ) = 𝑎 [1 − ⁡ ]⁡sin ( 𝑙𝑛 [1 − ⁡ ])
𝑅0 𝑏 𝑅0

From Figure 7.6, it is of course apparent that the voltage will be a nonlinear which is as a function of
the applied magnetic flux density (generally, magnetic field sensors are non-linear devices). In
practical applications, the magnetic flux density can be inferred from the measured voltage (V 1 or
V2) by inspection of the corresponding curve in Figure 7.6. Alternatively, in an automatic

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

application (where the voltage must be connected to an interface/signal conditioning circuit), a

“lookup table,” which may be stored in a microprocessor, can be used for estimating the value of B.

Figure 7.6: Voltages measured across the inner and outer p-n junctions as a function of magnetic
flux density, at a temperature of 25°C.

Figure 7.7: Diagram of the experimental setup for testing the sensor.

Figure 7.8 shows a typical plot of the difference V1−V2 of the two voltages. It is to be observed,
however, that the sensor is inherently not a differential sensor

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

Figure 7.8: The voltage difference V1−V2 as a function of magnetic flux density, at a temperature of
25°C.

Response to AC magnetic fields

The tests described above were repeated with AC fields. The AC signal was generated with an
Agilent model N9310A signal generator. It was interfaced to the electromagnet with a simple
transistor driver circuit. The output voltage was observed with an oscilloscope. Figure 7.9 shows
the response of the sensor to an alternating field with a peak value of 0.1 T and a frequency of 600
Hz. The low frequency was chosen for better visualization; however, the sensor was tested at
frequencies of up to 100MHz and the shape of the waveform was the same.

Figure 7.9: Oscilloscope plot of the voltage V1 across the inner p-n junction (upper trace) and of the
applied magnetic field (lower trace) at a frequency of 600 Hz.

The response in Figure 7.9 can be understood as follows: as the magnetic flux density increases in
value, the EHP current through the inner p-n junction is “chocked off” until it reaches 0 at the point
of maximum applied field (0.1 T). The voltage across the inner junction increases again to 0.8 V as
the field weakens, and, as the field reverses polarity, the EHP current decreases once again and the
same cycle is repeated. The measurement shown in Figure 7.9 was taken at room temperature
(25°C).

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

Temperature effect on the response of the sensor

The sensor and the magnetic field source were placed inside a controlled temperature chamber,
and the tests described above were repeated. Figure 7.10 shows the results of two tests that were
conducted with a DC field at temperatures of −40◦ C and +120◦ C. Clearly, lower temperatures have
only a minimal effect on the sensitivity of the sensor, while a significant deviation in the output
voltage occurs at high temperatures (above +80°C). The suitable operating range of the sensor is
therefore −40°C to +80°C (this is typically the preferred operating range of a silicon p-n junction).
Outside that range, an electronic circuit will be needed to compensate for the temperature
dependence of the sensor.

Figure 7.10: Voltages measured across the inner and outer p-n junctions as a function of magnetic
flux density, at temperatures of −40°C and +120°C.

Figure 7.11: A thermocouple-based self-heating RF power sensor with GaAs MMIC-compatible

micromachining technology

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.
 21EE742 – MICRO & NANO SENSORS &
Regulation – 2021 TRANSDUCERS
MODULE - 4

Other Important Electromagnetic/RF Micro- and Nano-Sensors

The field of RF, electric, and magnetic field sensing is advancing rapidly. Figure 7.11 shows an SEM
micro view of a self-heating radio frequency (RF) power sensor which senses RF power by
detecting the heat generated in the sensor by means of two thermocouples.

Conclusion

The new magnetic field sensing technology introduced in this chapter will be very advantageous in
battery-powered consumer products applications since it requires no power. The sensor described
is comparable in size to a Hall Effect sensor and offers a sensitivity that is at least an order of
magnitude better than a typical Hall Effect sensor. The radioactive β-particle source that is used in
the sensor, namely, tritium, exists in numerous other consumer products (such as toys and glow-in-
the dark products) and is quite safe in small quantities. One disadvantage of the new technology is
the cost, since it is more complex than the Hall effect based sensing technology. With mass
production, however, the cost can be substantially reduced. The sensor is expected to become
available commercially within a time span of 1 to 3 years.

Prepared by: Dr. RAMYA. K, Dept. of EEE Sri Sairam College of Engineering Anekal.