• NEAR INFRARED IMAGING
Process improvements boost NIR quantum efficiency
Capturing images at very low light levels
is critical for many medical, scientific,
military and security imaging applications. Doing so requires image sensors that
exhibit a high quantum efficiency (QE), a
low read out noise and dark current, as well
as a wide dynamic range that allows high
spatial resolution imaging across a range of
Recent developments in CMOS imagers
have led to an expanding array of low-light-level cameras. Many of which, are finding
use in night-vision applications that require
color image sensors during the day and near
infrared (NIR) illumination at night.
Because there are more NIR photons
than visible photons at night, NIR technology enables higher-resolution image
capture with fewer power-hungry LEDs,
extending the life of battery-powered security camera applications.
Having said that, since the wavelength
of NIR illumination is outside of the visible
spectrum, developers of machine vison and
imaging systems also rely on NIR lights in
applications where avoiding interference with
the surrounding environment or minimizing
human visual distraction are important.
Other growing application areas include
automotive driver-side monitoring applications for detection of drowsy or distracted
drivers, as well as biometric applications
such as facial and iris recognition systems.
However, as useful as these capabilities might be, existing approaches to NIR
detection lack the NIR sensitivity required
for many emerging next-generation mobile,
embedded, AR/VR and security camera
Understanding this, OmniVision Technologies, Inc. (Santa Clara, CA, USA; www.
ovt.com) has developed a new approach to
NIR imaging dubbed Nyxel technology,
which leverages novel silicon semiconductor architectures and processes.
Nyxel combines thick-silicon pixel archi-
tectures with careful management of wafer
surface texture to improve QE, along with
extended deep trench isolation (DTI) to
help retain (modular transfer function) MTF
without affecting the sensor’s dark current.
Increase QE 3 to 5x
After applying the technology as a test case
to an OmniVision OV2770 2. 8 µm back-
side illuminated CMOS image sensor,
results show a QE increase of up to 3x at
850 nm and 5x at 940 nm, when compared
to the same sensor without Nyxel. One sig-
nificant change was doubling the epitaxial
layer thickness from about 3 to 6 microns,
according to Boyd Fowler, OmniVision’s
Chief Technology Officer.
“QE is the probability of collecting an
OV2770 image sensor OV2770 with Nyxel
electron given a photon enters the sub-
strate,” Fowler explains. “It’s a function of
photon path length, so the thicker the sili-
con the better the absorption.”
One drawback to this approach, how-
ever, is that when electrons are absorbed
very deep, they tend to diffuse to their local
neighbors, which can
Tested under 850 nm IR LED illumination, the OV2770 sensor with Nyxel technology is signifi-
cantly brighter, with improved signal to noise ratio and no reduction in image sharpness.
Image comparison under 850nm IR LED illumination
OV2770 with Nyxel
OV2770 image sensor
Zoomed in on the yellow rectangles in the figure above shows that Nyxel technology (top)
provides greater sensitivity with little-to-no reduction in spatial resolution.