INTRODUCTION
The Advanced CCD Imaging Spectrometer (ACIS) is a CCD camera that operates in photon-counting mode to record both spatial and spectral information from celestial X-rays.  Each photon that is stopped by photoabsorption produces a cloud of secondary electrons; this charge cloud is pixelized and recorded as an ``event'' with well-defined characteristics, including position, amplitude, and a measure of its spatial concentration (``grade'').

We have developed a Monte Carlo CCD simulator for use in characterizing and calibrating the X-ray CCD detectors in the ACIS instrument on the Chandra X-ray Observatory.  It contains machinery for simulating all typical kinds of X-ray CCDs:  bulk front-illuminated, epitaxial front-illuminated, and back-illuminated.

The work discussed here was conducted by Leisa Townsley and Patrick Broos (with help from several others) to support their involvement in the analysis of ACIS data.  This work is NOT an official product of the ACIS Team and has NOT been endorsed by the ACIS Team or the Chandra X-ray Center (CXC).

A paper describing these techniques is in press in NIM.  A preprint is available:  astro-ph/0111003.

THE TECHNIQUE
Our Monte Carlo simulations are used to augment ACIS calibration data; once the simulator is tuned to reproduce monochromatic calibration data, it can be used to predict the instrument's spectral response at energies unavailable in the laboratory.  This ability is used to generate the ACIS response matrix, the mathematical representation of the ACIS spectral redistribution function.  The tool can also be used to simulate photon pile-up, which can occur when bright sources are observed.  It also incorporates a model for charge transfer inefficiency (CTI), a phenomenon that spectrally and spatially degrades events from ACIS detectors.

The basic aspects of the data that we are attempting to reproduce are the detected CCD event spectrum, the quantum efficiency, and the event grade distribution (branching ratios) resulting from a monochromatic incident X-ray flux.  The current model relies on new solutions of the diffusion equation to predict the radial charge cloud distribution in field-free regions of CCDs (Pavlov and Nousek 1999, "Charge Diffusion in CCD X-ray Detectors," Nuclear Instruments and Methods in Physics Research A, 428, 348).  By adjusting the size of the charge clouds, we can attempt to reproduce the grade distribution seen in ACIS calibration data event lists.  We have built into our code a model for channel stops and an interpretation of the MIT/ACIS group's model for the insulating layer under the gate structure (Prigozhin et al. 2000, "The Physics of the Low Energy Tail in the ACIS CCD.  The Spectral Redistribution Function," Nuclear Instruments and Methods in Physics Research A, 439, 582).  These components are necessary to explain subtle redistribution features in the spectra.
 

THE CODE
The simulator software is written in IDL and is available as part of a larger package we have developed for correcting CTI in ACIS CCDs .  This code is provided in the hope that it may assist other investigators interested in simulating ACIS devices or other X-ray CCDs.
 

DETAILS
Here is an example IDL script (called "on-orbit") that runs the simulator at several monochromatic calibration energies:

;; Run with     nice idl on_orbit >& log1 &
temperature = -120
num_events = 10000L
suffix  = '.evt0'
ccd_id  = 2

props   = ['AMP_ID','CHIPX','CHIPY','PHAS','pha','energy','instrument_grade','asca_grade']

bev_grades =[24,66,107,214,255]

energies=  [1.486, 2.771, 4.155, 4.511, 4.932, 5.895, 6.490, 9.711]
 

;; Simulate with uniform illumination, CTI on, exclude Bev grades, then correct for CTI.

for ii=0,n_elements(energies)-1 do $
runsims, ccd_id, energies[ii], num_events, NAME_SUFFIX=suffix, TEMP=temperature, PROP_NAMES=props,VERBOSE=2, /MODELCTI, GRADES_EXCLUDED=bev_grades, /CORRECT_CTI, PHOTON_GENERATOR='rmf_photon_generator'

exit