The BRE work (which I did!) is in the public domain in that it was work for the Home Office Fire Research and Development Group. Cath Reynolds was their contact and she published a summary on behalf of FRDG. This was circulated to fire brigades, I understand. She is now in charge of DCLG's fire research and may still be able to supply copies of this publication.
The work however is nearly 20 years old, and I'm not certain quite how comparable today's detectors are with the ones available then.
That study gave an indication that with small fires with low energy, the optical detectors were slow to respond. It was thought this was due to the laberinthyn entry to the detecting chamber, needed to keep out light, which also needs more energy to push smoke through into the chamber compared with the ion detectors. This was based on the fact that such detectors in a wind-tunnel with a gentle flow of air showed very little delay in responding to changes in smoke level.