What is Infrared Radiation?
The Latin prefix "infra" means "below"
or "beneath." Thus "infrared" refers to the region beyond
or beneath the red end of the visible color spectrum. The infrared region is
located between the visible and microwave regions of the electromagnetic
spectrum. Because heated objects radiate energy in the infrared, it is often
referred to as the heat region of the spectrum. All objects radiate some energy
in the infrared, even objects at room temperature and frozen objects such as ice.
The higher the temperature of an object, the higher the
spectral radiant energy, or emittance, at all wavelengths and the shorter the
predominant or peak wavelength of the emissions. Peak emissions from objects at
room temperature occur at 10 µm. The sun has an equivalent temperature of 5900
K and a peak wavelength of 0.53 µm (green light). It emits copious amounts of
energy from the ultraviolet to beyond the far IR region.
Much of the IR emission spectrum is unusable for detection
systems because the radiation is absorbed by water or carbon dioxide in the
atmosphere. There are several wavelength bands, however, with good
transmission.
- The long
wavelength IR (LWIR) band spans roughly 8-14 µm, with nearly 100%
transmission on the 9-12 µm band. The LWIR band offers excellent
visibility of most terrestrial objects.
- The medium
wavelength IR (MWIR or MIR) band (3.3-5.0 µm) also offers nearly 100%
transmission, with the added benefit of lower, ambient, background noise.
- Visible and
short wavelength IR (SWIR or near IR, NIR) light (0.35-2.5 µm) corresponds
to a band of high atmospheric transmission and peak solar illumination,
yielding detectors with the best clarity and resolution of the three
bands. Without moonlight or artificial illumination, however, SWIR imagers
provide poor or no imagery of objects at 300K.
Infrared Detectors
An infrared detector is simply a transducer of radiant
energy, converting radiant energy in the infrared into a measurable form.
Infrared detectors can be used for a variety of applications in the military,
scientific, industrial, medical, security and automotive arenas. Since infrared
radiation does not rely on visible light, it offers the possibility of seeing
in the dark or through obscured conditions, by detecting the infrared energy
emitted by objects. The detected energy is translated into imagery showing the
energy differences between objects, thus allowing an otherwise obscured scene
to be seen. For example, the left image below is what you may see in ordinary
light on a dark night. The image at right is the same scene but as seen with an
Infrared camera. Hot objects such as people stand out from the typically cooler
backgrounds regardless of the available visible light.
Under infrared light, the world reveals features not
apparent under regular visible light. People and animals are easily seen in
total darkness, weaknesses are revealed in structures, components close to
failure glow brighter, visibility is improved in adverse condition such as
smoke or fog.
Infrared Detector Types
There are two fundamental methods of IR detection, energy
and photon detection. Energy detectors respond to temperature changes generated
from incident IR radiation through changes in material properties. Photon
detectors generate free electrical carriers through the interaction of photons
and bound electrons. Energy detectors are low cost and typically used in single
detector applications; common applications include fire detection systems and
automatic light switches. However, the simplicity of fabricating large 2D focal
plane arrays in semiconductors has lead to the use of photon detectors in
almost all advanced IR detection systems. Recent advances in micromachining and
materials science have lead to the exciting field of uncooled detectors which
promise lower system and operation costs.
Energy Detectors
The absorption of IR energy heats the detection element in
energy or thermal detectors, leading to changes in physical properties which
can be detected by external instrumentation and which can be correlated to the
scene under observation. Energy detectors contain two elements, an absorber and
a thermal transducer. The following are examples of energy detectors.
Thermocouples / Thermopiles
Thermocouples are formed by joining two dissimilar metals
which create a voltage at their junction. This voltage is proportional to the
temperature of the junction. When a scene is optically focused onto a
thermocouple, its temperature increases or decreases as the incident IR flux
increases or decreases. The change in IR flux emitted by the scene can be
detected by monitoring the voltage generated by the thermocouple. For sensitive
detection, the thermocouple must be thermally insulated from its surroundings.
For fast response, the thermocouple must be able to quickly release built up heat.
This tradeoff between sensitivity of detection and the ability to respond to
quickly changing scenes is inherent to all energy detectors.
A thermopile is a series of thermocouples connected together
to provide increased responsivity.
2 Pyroelectric Detectors
Pyroelectric detectors consist of a polarized material
which, when subjected to changes in temperature, changes polarization. These
detectors operate in a chopped system; the fluctuation in the exposure to the
scene generates a corresponding fluctuation in polarization and thus an
alternating current that can be monitored with an external amplifier.
Ferroelectric Detectors
Similar to pyroelectric detectors, ferroelectric detectors
are based on a polarized material which, when subjected to changes in
temperature, changes polarization.
Thermistors / Bolometers / Microbolometers
In thermistors, the resistance of the elements varies with
temperature. One example of a thermistor is a bolometer. Bolometers function in
one of two ways: monitoring voltage with constant current or monitoring current
with constant voltage.
Advances in the micromachining of silicon have lead to the
exciting field of microbolometers. A microbolometer consists of an array of
bolometers fabricated directly onto a silicon readout circuit. This technology
has demonstrated excellent imagery in the IR. Although the performance of
microbolometers currently falls short of that of photon detectors, development
is underway to close the performance gap. Microbolometers can operate near room
temperature and therefore do not need vacuum evacuated, cryogenically cooled
dewars. This advantage brings with it the possibility of producing low cost
night vision systems for both military and commercial markets.
Microcantilevers
Microcantilevers are based on the bimetal effect to measure
IR radiation. This effect utilizes the difference in thermal expansion
coefficients of two different bimetals to cause a displacement of a
microcantilever. In combination with a reference plate, this cantilever forms a
capacitance. When infrared light is absorbed by the microcantilever, the
microcantilever deflects and thus alters the capacitance of the structure. This
change in capacitance is a measure for the incident infrared radiation.
Photon Detectors
Light interacts directly with the semiconductors in photon
detectors to generate electrical carriers. Because these detectors do not
function by changing temperature, they respond faster than energy detectors.
However, these detectors will also pick up the IR radiation generated by their
own mountings and accompanying optics and thus must be cooled to cryogenic
temperatures to minimize background noise. The following are examples of photon
detectors.
Intrinsic Detectors
Photovoltaic Intrinsic Detectors
Photovoltaic (PV) detectors generate photocurrents which can be monitored with
a trans-impedance amplifier. These photocurrents are created when incident
light with energy greater than or equal to the energy gap, or diode junction,
of the semiconductor strikes the detector causing excited, minority, electrical
carriers to be swept across the photodiode's electrical junction.
PV devices operate in the diode's reverse bias region; this
minimizes the current flow through the device which in turn minimizes power dissipation.
In addition, PV detectors are low noise because the reverse bias diode junction
is depleted of minority carriers. The highest performance PV detectors are
fabricated from Si, Ge, GaAs, InSb, InGaAs, and from HgCdTe (MCT).
Photoconductive Intrinsic Detectors
Photoconductive (PC) detectors function similarly to PV detectors. Incident
light with energy greater than or equal to the energy gap of the semiconductor
generates majority electrical carriers. This results in a change in the
resistance, and hence conductivity, of the detector. Examples of PC detector
materials are Lead sulfide (PbS), Lead selenide (PbSe) and MCT.
Extrinsic Detectors
Extrinsic detectors are based on Si (SiX) or Ge (GeX) doped
with impurities such as Boron, Arsenic and Gallium. They are similar to
intrinsic detectors. However, in extrinsic detectors carriers are excited from
the impurity levels and not over the bandgap of the basic material. Both
photovoltaic and photoconductive types exist.
Photo-emissive Detectors
Photo-emissive detectors are based on the emission of
carriers from a metal into a semiconductor material through the absorption of
light. A typical example is Platinum Silicide (PtSi) on Si.
Quantum Well Infrared Photodetector
The Quantum Well Infrared Photodetector (QWIP) is an
infrared detector that consists of multiple alternating thin gallium arsenide
(GaAs) and aluminum gallium arsenide (AlGaAs) layers. Carriers are generated by
absorption of IR light inside quantum wells.
Detector Types and Materials Overview
The table below summarizes the main detector types and
materials.
Infrared Detector Formats and Architectures
Infrared detectors are available as single element detectors
in circular, rectangular, cruciform, and other geometries for reticle systems,
as linear arrays, and as 2D focal plane arrays (FPAs).
Single element detectors are normally
frontside illuminated and wire bonded devices. Linear and 2D arrays may be
fabricated with a variety of device and signal output architectures.
First generation linear arrays were usually
frontside illuminated, with the detector signal output connected by wire
bonding to each element in the array. The signal from each element was then
brought out of the vacuum package and connected to an individual room
temperature preamplifier prior to interfacing with the imaging system display.
Gain adjustments were usually made in the preamplifier circuitry. This approach
limited first generation linear arrays to less than two hundred elements.
Second generation arrays, both linear and 2D,
are frequently backside illuminated through a transparent substrate. Several
alternative focal plane architectures are illustrated in the graph below.
The diagram below illustrates a detector array which is
electrically connected directly to an array of preamplifiers and/or switches called
a readout. The electrical connection is made with indium "bumps"
which provide a soft metal interconnect for each pixel. This arrangement,
commonly referred to as a "direct hybrid", facilitates
the interconnection of large numbers of pixels to individual preamplifiers
coupled with row and column multiplexers.
Indirect hybrid configurations (b) may be used
with large linear arrays to interface the detector with a substrate having a
similar thermal coefficient of expansion. These hybrids may also be used for
serial hybridization, allowing the detector to be tested prior to committing
the readout, and/or to accommodate readout unit cells having dimensions larger
than the detector unit cell, increasing the charge storage capacity and thereby
extending the dynamic range. Readouts and detectors are electrically
interconnected by a patterned metal bus on a fanout substrate.
Monolithic detector arrays (c) have integrated
detector and readout functions. Generally, in these arrays, the command and
control signal processing electronics are adjacent to the detector array,
rather than underneath. In this case, the signal processing circuits may be
connected to the detector by wire bonds. In the monolithic configuration it is
not necessary for the signal processing circuits to be on the same substrate as
the detector/readout (as shown in the figure) or at the same temperature as the
detector. Monolithic PtSi detector arrays can be made with signal processing
incorporated on the periphery of the detector/readout chip through the use of
silicon-based detector technology.
Z technology, as illustrated in figure (d),
provides extended signal processing real estate for each pixel in the readout
chip by extending the structure in the orthogonal direction. In the approach
illustrated, stacked, thinned readout chips are glued together, and the
detector array is connected to the edge of this signal processing stack with
indium.
Finally, a "Loophole" approach, as
illustrated in figure (e), relies on thinning the detector material after
adhesively bonding it to the silicon readout. Detector elements are connected
to the underlying readout with vias, which are etched through the detector
material to contact pads on the readout and metallized.
History and Trends of Infrared Detectors
Infrared detectors are in general used to detect, image, and
measure patterns of the thermal heat radiation which all objects emit. Early
devices consisted of single detector elements that relied on a change in the
temperature of the detector. Early thermal detectors were thermocouples and
bolometers which are still used today. Thermal detectors are generally
sensitive to all infrared wavelengths and operate at room temperature. Under
these conditions, they have relatively low sensitivity and slow response.
Photon detectors were developed to improve sensitivity and
response time. These detectors have been extensively developed since the
1940's. Lead sulfide (PbS) was the first practical IR detector. It is sensitive
to infrared wavelengths up to ~3 µm.
Beginning in the late 1940's and continuing into the 1950's,
a wide variety of new materials were developed for IR sensing. Lead selenide
(PbSe), lead telluride (PbTe), and indium antimonide (InSb) extended the
spectral range beyond that of PbS, providing sensitivity in the 3-5 µm medium
wavelength (MWIR) atmospheric window.
The end of the 1950's saw the first introduction of
semiconductor alloys, in the chemical table group III-V, IV-VI, and II-VI
material systems. These alloys allowed the bandgap of the semiconductor, and
hence its spectral response, to be custom tailored for specific applications.
MCT (HgCdTe), a group II-VI material, has today become the most widely used of
the tunable bandgap materials.
As photolithography became available in the early 1960's it
was applied to make IR sensor arrays. Linear array technology was first
demonstrated in PbS, PbSe, and InSb detectors. Photovoltaic (PV) detector
development began with the availability of single crystal InSb material.
In the late 1960's and early 1970's, "first
generation" linear arrays of intrinsic MCT photoconductive detectors were
developed. These allowed LWIR forward looking imaging radiometer (FLIR) systems
to operate at 80K with a single stage cryoengine, making them much more
compact, lighter, and significantly lower in power consumption.
The 1970's witnessed a mushrooming of IR applications
combined with the start of high volume production of first generation sensor
systems using linear arrays.
At the same time, other significant detector technology
developments were taking place. Silicon technology spawned novel platinum
silicide (PtSi) detector devices which have become standard commercial products
for a variety of MWIR high resolution applications.
The invention of charge coupled devices (CCDs) in the late
1960's made it possible to envision "second generation" detector
arrays coupled with on-focal-plane electronic analog signal readouts which
could multiplex the signal from a very large array of detectors. Early
assessment of this concept showed that photovoltaic detectors such as InSb,
PtSi, and MCT detectors or high impedance photoconductors such as PbSe, PbS,
and extrinsic silicon detectors were promising candidates because they had
impedances suitable for interfacing with the FET input of readout multiplexers.
PC MCT was not suitable due to its low impedance. Therefore, in the late 1970's
through the 1980's, MCT technology efforts focused almost exclusively on PV
device development because of the need for low power and high impedance for
interfacing to readout input circuits in large arrays. This effort has been
paying off in the 1990's with the birth of second generation IR detectors which
provide large 2D arrays in both linear formats. These detectors use TDI for
scanning systems; in staring systems, they come in square and rectangular
formats.
Monolithic extrinsic silicon detectors were demonstrated
first in the mid 1970's. The monolithic extrinsic silicon approach was
subsequently set aside because the process of integrated circuit fabrication
degraded the detector quality. Monolithic PtSi detectors, however, in which the
detector can be formed after the readout is processed, are now widely
available.
Second generation devices have now been demonstrated with
many detector materials and device types, including PbS, PbSe, InSb, extrinsic
Si, PtSi, and PV MCT.
It has taken nearly two decades since the invention of the
CCD to mature the integration of IR detectors coupled with electronic readouts
on the focal plane. This progress brought with it the transition from first
generation to second generation device production. The size and complexity of
infrared image detectors corresponds to the evolution of silicon integrated
circuit size and complexity; this can be seen through comparison to dynamic
random access memory chip trends (see graph below). Note that DRAMs require
just one transistor per unit cell, whereas infrared sensor readouts require
three or more, one of which must be a low noise analog device.
