One-page Summary report for PPARC PIPSS grant "The spatial
resolution of back-illuminated CCDs"
Louisa J. Bradley, David M. Walton, Christopher J. McFee,
Alan Smith
The aim of this study was to investigate various methods of determining the spatial resolution, and use the chosen method to measure the resolution of several CCDs. These CCDs spanned a range of parameters in order to develop a model predicting spatial resolution as a function of these parameters, paying particular attention to the important role of photo-electron diffusion in the field-free region.
Five methods were assessed and some of these investigated experimentally before being rejected. These were:-
· Interference fringes
· Fe55 X-ray photons
· Single pinhole method
· Pinhole grid method
The method chosen was the Vernier/slit method, which was found to give the best combination of consistency, ease of test and small amount of data to store. Four devices were tested, ranging in nominal thickness from 9 to 13μm, resistivity 20 to 100 ohm-cm, pixel size 13.5 to 16 μm, with one-phase or two phases high during integration and at various substrate voltages. Wavelengths from near UV through visual to near IR were used to give a range of absorption depths to probe the device structure. The data from the Vernier method gives a measure of the Line Response Function (LRF). Modulation Transfer Function (MTF) is a more commonly used measure of spatial resolution, so software was written to convert LRF to MTF. E2v technologies' Excel spreadsheet model predicts MTF as a function of various parameters, and this model was extended to cover more parameters, such as signal level and substrate voltage. The measured variations in MTF were compared with the model predictions.
· Several experimental methods for determining spatial resolution have been investigated. The Vernier/slit method has been found to give the best combination of convenience and consistency.
· An MTF measurement setup has been established at MSSL, including the necessary hardware, software and procedures. The data is consistent and reproducible, in particular following improvements made to the mechanical stability and optics during the course of this study. MTF measurement can now be performed routinely and rapidly.
· Several devices have been tested, exploring a range of different regimes in device physical parameters, operational parameters, and wavelength.
· E2v technologies’ Excel spreadsheet MTF model has been extended to predict behaviour as a function of more parameters.
· Agreement between data and the e2v technologies/MSSL model is good. Both data and model show the expected trends, although there are small departures in the details, which could possibly be explained with the aid of a more complex model, beyond the scope of this work.