Back-illuminated CCD: improve light sensitivity

Infineon / Mitsubishi / Fuji / Semikron / Eupec / IXYS

Back-illuminated CCD: improve light sensitivity

Posted Date: 2024-02-01

What is a back-illuminated CCD?

When we say an image sensor is front-illuminated, we mean that its physical configuration is similar to what we would expect from a typical IC: pins extend from the perimeter of the device down to the mounting terminals, i.e. the "back" of the device faces the PCB surface , the "front" side of the device is exposed to incident light.

Most CCDs do work this way, but it turns out that performance can be enhanced by physically modifying the device and then mounting it so that incident light arrives from the other side. The CCD used for this type of implementation is called a back-illuminated CCD.

Front lighting and reduced sensitivity

The image below, which we saw in Part 4 of this series, reminds us that a CCD is made up of a layer of polysilicon gates (aka electrodes) covering the shift register and potentially light-sensitive parts of the device.

In full-frame and frame-transfer CCDs, the photoactive sites are located beneath the electrodes, so light from the front of the CCD must pass through the polysilicon to create a charge. This can cause quite serious problems, as we'll see later.

Interline transfer CCDs appear to eliminate this problem because the photodiodes are not part of the shift register and therefore do not need to be covered by the polysilicon gate. However, in this case we trade one limitation for another: the interline transfer method reduces the overall sensitivity because the photosensitive sites occupy a relatively small portion of their respective pixels. Microlenses that focus light onto photodiodes can mitigate but not eliminate this effect.

wavelength dependent behavior

The generation of light-induced charges occurs at different physical depths in the silicon, depending on the absorption coefficient, which in turn is a function of wavelength. The picture below expresses this phenomenon. Red waves represent longer wavelength photons (such as red light or infrared radiation), and blue waves represent shorter wavelength photons (such as blue light or ultraviolet radiation).

In front-illuminated systems, the CCD's response to light is significantly altered by the presence of polysilicon electrodes. First, the electrodes are not completely transparent; they scatter and reflect incoming light, reducing overall sensitivity.

Furthermore, the electrodes make detection of certain wavelengths essentially impossible because their thickness exceeds the absorption depth. For example, if the thickness of the polysilicon layer is 500 nm, UV radiation with an absorption depth less than 500 nm will not produce any electrical response in the CCD.

A CCD's ability to detect light is measured by its quantum efficiency (QE), which represents the percentage of incident photons that are actually converted into usable charge. Can be general optical response or wavelength-specific QE. A typical front-illuminated CCD is adversely affected by the polysilicon gate, with a QE of about 50% around 700 nm and an average QE of perhaps 25-30% in the visible spectrum.

Learn about backlighting

If light enters from the other side of the device, we bypass the troublesome polysilicon electrode entirely. This increases overall quantum efficiency and is particularly advantageous in applications that require sensitivity to shorter wavelength radiation.

A back-illuminated CCD equipped with a good anti-reflective coating can have an average quantum efficiency in excess of 70% in the visible spectrum. At certain wavelengths, the theoretical value is close to 100% and the measured value exceeds 90%. The chart below provides an overall comparison of front-illuminated QE and back-illuminated QE.

Backlighting: Disadvantages

As you might guess, this increase in performance comes at a price. The first is the literal price - the thickness of a back-illuminated sensor must be significantly reduced to ensure adequate sensitivity, and this challenging manufacturing process makes the device more expensive.

Additionally, the procedures used to thin the CCD may cause defects that increase noise.

If the CCD is made very thin, so that the longer wavelength absorption distance is closer to the thickness of the device, then we may also trade increased short-wavelength sensitivity for reduced long-wavelength sensitivity.

For example, near-infrared radiation may pass directly through a silicon substrate. Applications requiring infrared detection can benefit from improved forms of backside illumination, where a thicker substrate is combined with a bias voltage to prevent loss of photogenerated charge through diffusion. These types of CCD sensors are cooled to extremely low temperatures during operation; at normal temperatures, dark current is excessive.

in conclusion

We've covered the characteristics of backlighting and why engineers use this technology. Back-illuminated CCDs aren't the kind of thing you can put into a cheap camera to make home videos, but in high-performance systems that require the quantum efficiency the sensor can deliver, they're worth the extra trouble and expense.

#Backilluminated #CCD #improve #light #sensitivity