...${\textit{z}>0.4}$

Table 5 and Appendix are only available in electronic form at http://www.edpsciences.org

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... used

In principle this may not apply correctly for objects far from the CCD array centre, as the detector response redistribution matrices may diverge slightly from the on-axis case. It is probably also true that mirror vignetting calibrations become less secure with increasing off-axis angle.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... filter

The Lumb et al. (2002) background templates are only appropriate for observations made using the Thin filter; three of our clusters were observed through the Medium filter (Table 1).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... cluster

We caution the reader not to use Fig. 5 as a general template for XMM-Newton analysis, as both the spectrum and the absolute count rate of the proton background are likely time dependent.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... region

The radius of the excised region was 50  h₅₀^-1 kpc except in the case of RX J1701.3 and RX J1325 where $\sim$120 kpc was used, see Sects. 3.6 and 3.7.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.