Advertiser and

Buyer's Guide
Buyers Guide

Chip Shots blog

Greatest Hits of 2005
Greatest Hits of 2005

Featured Series
Featured Series

Web Sightings

Media Kit

Comments? Suggestions? Send us your feedback.


Chipworks Corner

Fujifilm maximizes charge-coupled device’s 0.35-µm, two-metal, double-poly process

One of the fastest-growing consumer product segments is the digital still camera (DSC) market. Digital sales exceeded film camera sales a few years ago, and a number of manufacturers have stopped film camera production entirely. DSC sales were 77 million in 2004 and are projected to be about 100 million in 2005, according to Future Image. Of course, those figures are eclipsed by expected camera-phone sales of close to 500 million in 2005, but that is arguably a different market.

Within the DSC market, there is a split in imaging technology between the two types of image sensors: charge-coupled devices (CCDs) and CMOS imagers. CMOS technology is taking over the low-cost end, while CCDs still dominate at the high end, although not exclusively. Canon has a number of high-end models using CMOS imagers, including the EOS 5D that uses a 12.8-Mpixel full-frame (24 × 36 mm) CMOS sensor.

For this installment of Chipworks Corner, we will analyze the CCD sensor from a Fujifilm Finepix E550 6.3-Mpixel camera, which uses the company’s proprietary Super CCD design. For a point-and-shoot camera, this has impressive resolution as well as ISO 800 sensitivity and 1/2000-second shutter speed.

The Super CCD design rotates the conventional x-y array 45° to align the pixels diagonally, optimizing real estate with comparatively large octagonal photodiodes and allowing closer spacing than other designs. Fujifilm claims “a balanced combination of resolution, sensitivity, dynamic range, signal/noise ratio, and color fidelity” as a result of this layout. In addition, the company’s signal processing technology creates virtual pixels between the photodiodes, so that the stored image is actually a square-grid x-y array of 12.3 Mpixels.

The Fujifilm MS3895A described in this article is a Super CCD HR, the fourth generation of technology originally announced in 1999, with the pixel size reduced to 2.7 µm square to improve resolution. It was manufactured by Fujifilm Microdevices in its fab in Sendai, Japan, in a 0.35-µm, two-metal, double-polysilicon process, using p-wells in an n-substrate. The 7.7 × 9.0-mm (69.3 mm2) die has a total of 6.63 Mpixels, with 6.3 million effective pixels. With signal processing adding virtual pixels, the number of recorded pixels increases to 12.3 million.

The device’s pixels line up in a diagonal orientation. Figure 1 shows a cross section of the structure. The cross section is also oriented diagonally across the die, to show the structure as it is laid out. Organic hemispherical lenses can be seen at the top, then a color filter layer over an organic planarizing layer. Below that layer are silicon nitride lenses, and below those are photodiode and CCD charge-transfer electrodes.

Figure 1

Figure 2 zooms in on the silicon structure, revealing that the nitride lenses are made up of two nitride layers. The lower one is deposited over the CCD structure and then planarized, while the upper one is deposited on the planarized layer and then etched into a hemispherical shape. The two polysilicon transfer electrodes below are covered with a tungsten light shield.

Figure 2

Figure 3’s scanning capacitance microscope (SCM) cross-sectional image illustrates the doping structure in the substrate. From the bottom, the light region is the n-substrate, graduating into the array p-well; the dark regions at the substrate surface are the p-isolations between the photodiodes and the p-pinning layer at the surface of the diodes. The light spots in the p-isolation are the n-doped charge-transfer channels, and the subtly lighter regions below the surface are the buried photodiodes. Above the surface, the transfer electrodes and dielectrics show as blank areas, covered by an unbroken layer of metal. (This section was taken from the dark pixel area of the chip, where the light shield was continuous.)

Figure 3

Switching to plan view, Figure 4 shows the chip with the organic layers removed and the nitride lenses exposed. The diagonal layout is clear, and some circles have been added to illustrate the positions of the virtual pixels created by the signal processing.

Figure 4

Figure 5 reveals the array polished to show the polysilicon transfer electrodes running horizontally across the array; the line right above the diamond-shaped apertures is the poly-1 layer. In two of the apertures, some residual tungsten remains, defining the circular light access to the photodiodes.

Figure 5

Figure 6 is a plan-view-bevel SCM image of the array. The silicon surface is the blank area at the lower right, and higher up, the zigzag n-doped transfer channels can be seen surrounding the faint diamonds of the p-pinning layers in the pixel. Farther up (deeper into the substrate), rhombic buried photodiodes are exposed; the structures are not octagonal, which is contrary to Fujifim’s own description. The collected charge is clocked out of each photodiode by the transfer electrodes using the charge transfer channel to the edge of the CCD array, where a secondary CCD transfers the charge to the first-stage amplifier and then to the signal processing circuitry.

Figure 6

As can be seen, Fujifilm’s Super CCD sensor is a relatively simple structure in this nanotech era, but nicely optimized to do exactly what is required in a 12-Mpixel camera. The unit has a street price of $325, numbers that have impressed digital photography reviewers on the Web.—Dick James

This report is one of a regular series of device-level process analyses, written exclusively for MICRO by Chipworks’ senior technology adviser, Dick James, a 30-year veteran of the semiconductor industry. Chipworks is an Ottawa, Canada–based specialty reverse-engineering company that gets inside technology and takes apart ICs and electronics systems in order to provide engineering information for its customers. The technical intelligence customers are usually within manufacturing companies, performing product development, or doing strategic marketing or benchmarking studies. The patent intelligence clients are usually patent lawyers or intellectual property groups within manufacturing companies.

MicroHome | Search | Current Issue | MicroArchives
Buyers Guide | Media Kit

Questions/comments about MICRO Magazine? E-mail us at

© 2007 Tom Cheyney
All rights reserved.