BSI vs FSI Sensors: Which Sensor Suits Your Needs?
The debate between CMOS (Complementary Metal-Oxide-Semiconductor) and BSI (Backside-Illuminated) sensors has long captivated the digital imaging technology community. Each sensor type offers unique advantages and disadvantages, making them suitable for different applications. In this article, we will delve into the primary distinctions between CMOS and BSI sensors, evaluating their strengths and weaknesses, and discussing how these aspects affect image quality, low-light performance, and overall camera functionality.
Image Credit: Sony
What are BSI Sensors?
Backside-Illuminated (BSI) sensors are a type of image sensor where the photodiode layer is positioned on the back of the sensor, allowing light to directly hit the photodiodes without having to pass through the metal wiring and transistor layers. This design maximizes the amount of light captured by the photodiodes, enhancing sensitivity, especially in low-light conditions. BSI sensors are widely used in digital cameras, smartphones, and other imaging devices due to their superior performance in low light.
What are FSI Sensors?
Frontside-Illuminated (FSI) sensors have the photodiode layer positioned at the front, meaning light must pass through the metal wiring and transistors before reaching the photodiodes. This can reduce the amount of light that reaches the photodiodes, especially in smaller pixels, which affects image quality in low-light situations. FSI sensors have been the traditional design for many years but are increasingly being replaced by BSI technology in applications where low-light performance is critical.
BSI CMOS vs. CMOS
Traditional CMOS pixels consist of several essential components: microlenses (on-chip lenses), on-chip color filters, metal interconnects, photodiodes, and substrates.
Image Credit: Wikipedia
Traditionally, CMOS sensors have utilized a "front-illuminated" structure, as shown on the left side of the illustration. When light enters the pixel, it passes through the on-chip lens and colour filter before encountering the metal interconnect layer. The photodiode ultimately captures the light. However, the metal, being opaque and reflective, presents a challenge. Light passing through the metal wiring layer is partially blocked and reflected, resulting in only about 70% or less of the incident light reaching the photodiode. This reflection can also cause cross-talk with neighbouring pixels, leading to colour distortion. In mid-to-low-end CMOS sensors, the metal used in the wiring layer is often aluminium (Al), which has a reflectivity of approximately 90% across the visible light spectrum (380-780nm).
To address these limitations, the "back-illuminated" CMOS configuration was developed. In this design, the positions of the metal wiring layer and the photodiode are reversed compared to the "front-illuminated" type. As a result, light can reach the photodiode directly with minimal obstruction or interference. This significantly enhances light utilization, making back-illuminated CMOS sensors highly efficient at capturing incoming light. Consequently, they perform exceptionally well in low-light environments, offering superior image quality.
In summary, traditional CMOS sensors have faced limitations due to the obstructive nature of metal interconnects. However, the advent of back-illuminated CMOS technology has addressed these challenges, resulting in enhanced image quality, particularly in challenging lighting conditions.
CMOS vs. BSI CMOS: Which is Better?
Let's delve deeper into the advantages and disadvantages of Frontside Illuminated (FSI) and Backside Illuminated (BSI) sensor technologies. Before proceeding, let's revisit their cross-sections. The dashed line indicates the location of the photodiode, providing a clear view of BSI's streamlined optical path, which translates to reduced energy loss within the sensor.
The light receiving surface located at the top does not inhibit the way of light coming in, capturing more photon and allowing for higher sensitivity (Image Credit: Vieworks)
Advantages of FSI:
FSI sensors closely follow semiconductor manufacturing processes. Light passes through the gaps between the metal wiring on the front, converging onto the photodiode within the light-sensitive area. Even with larger pixels, FSI can still deliver satisfactory performance. The relatively modest ratio of optical stack height to pixel area ensures a well-preserved photosensitive area. Additionally, FSI has a simple manufacturing process, leading to lower costs and higher efficiency compared to BSI. FSI also generally has lower manufacturing process requirements than BSI.
Disadvantages of FSI:
However, as pixel sizes decrease, FSI faces limitations such as reduced fill factor, elongated optical paths, and significant reflection and absorption losses within the metal wiring. These factors collectively impede sensor performance as pixel dimensions shrink.
Following the front-end process, BSI sensors undergo a distinctive procedure where the wafer is flipped, bonded, thinned, and subsequently outfitted with colour filters and microlenses on the back.
Advantages of BSI:
The primary advantage of BSI lies in its ability to separate electrical components from light, allowing independent optimization of the light path and reducing absorption and reflection by the FSI's metal wiring layer. The optical stack in BSI pixels is significantly reduced, resulting in an increased fill factor that approaches nearly 100%. Consequently, BSI achieves higher Quantum Efficiency (QE).
BSI's streamlined optical stack significantly reduces thickness, making it more amenable to incident angles. For instance, at a 1.4μm pixel size, the BSI structure can reduce lens shading by 40% compared to FSI.
Disadvantages of BSI:
Nevertheless, the BSI process is complex and challenging. The early stages of development and initial yield rates present significant challenges. However, with ongoing advancements in semiconductor technology, BSI's viability has matured, with corresponding improvements in yield rates.
Additionally, the back-side-up orientation of BSI chips exposes the silicon body material near the photodiode to direct illumination. This can lead to crosstalk issues due to light diffusion into neighbouring pixels or recombination at the back interface. However, industry innovations are addressing these challenges by enhancing processing on the chip's rear.
Learn more on implementing BSI Sensor in high speed applications.
The VT Sense Series' back-side illuminated (BSI) TDI line scan sensors deliver improved quantum efficiency (QE) and signal-to-noise ratio (SNR). These BSI sensors provide enhanced sensitivity and ensure high-quality imaging across visible, UV, and NIR spectrums. With their remarkable sensitivity, the VT Sense Series excels in low-light conditions, offering superior performance.