Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0169-1-1.ps.gz Дата изменения: Fri Sep 10 12:36:55 2004 Дата индексирования: Mon Oct 1 22:29:26 2012 Кодировка: Поисковые слова: barnard 68

XMM-Newton CCF Release Note
XMM-CCF-REL-169
EPIC MOS low energy response and Quantum Eфciency
R. D. Saxton, Steve Sembay
5 August 2004
1 CCF components
Name of CCF VALDATE EVALDATE Blocks changed XSCS ag
EMOS1 QUANTUMEF 0016.CCF 2000-01-01 QE TOTAL NO
EMOS2 QUANTUMEF 0016.CCF 2000-01-01 QE TOTAL NO
EMOS1 REDIST 0035.CCF 1999-12-10 2000-07-15 CCD REDISTRIBUTION-n NO
EMOS1 REDIST 0036.CCF 2000-07-15 2000-11-09 CCD REDISTRIBUTION-n NO
EMOS1 REDIST 0037.CCF 2000-11-09 2001-04-18 CCD REDISTRIBUTION-n NO
EMOS1 REDIST 0038.CCF 2001-04-18 2001-08-18 CCD REDISTRIBUTION-n NO
EMOS1 REDIST 0039.CCF 2001-08-18 2001-09-26 CCD REDISTRIBUTION-n NO
EMOS1 REDIST 0040.CCF 2001-09-26 2001-11-25 CCD REDISTRIBUTION-n NO
EMOS1 REDIST 0041.CCF 2001-11-25 2002-11-07 CCD REDISTRIBUTION-n NO
EMOS1 REDIST 0042.CCF 2002-11-07 - CCD REDISTRIBUTION-n NO
EMOS2 REDIST 0035.CCF 1999-12-10 2000-07-15 CCD REDISTRIBUTION-n NO
EMOS2 REDIST 0036.CCF 2000-07-15 2000-11-09 CCD REDISTRIBUTION-n NO
EMOS2 REDIST 0037.CCF 2000-11-09 2001-04-18 CCD REDISTRIBUTION-n NO
EMOS2 REDIST 0038.CCF 2001-04-18 2001-08-18 CCD REDISTRIBUTION-n NO
EMOS2 REDIST 0039.CCF 2001-08-18 2001-09-26 CCD REDISTRIBUTION-n NO
EMOS2 REDIST 0040.CCF 2001-09-26 2001-11-25 CCD REDISTRIBUTION-n NO
EMOS2 REDIST 0041.CCF 2001-11-25 2002-11-07 CCD REDISTRIBUTION-n NO
EMOS2 REDIST 0042.CCF 2002-11-07 - CCD REDISTRIBUTION-n NO
2 Changes
As part of the calibration program for the SWIFT project, which is using the same CCDs
as Epic-MOS, the quantum eфciency (QE) of the detectors has been recalibrated (Fig. 1).
These extra ground calibration data have given new information about the low-energy QE
of the MOS CCDs and allow us to improve the QE curves below 1.0 keV for the central
CCD.
In-orbit calibration data has been reanalysed in the light of the improved QE and small
tweaks have been made to the redistribution function for the two MOS cameras to improve
the agreement at low-energies.
1

Figure 1: The measured quantum eфciency from EPIC ground based synchrotron measurements of
a CCD from the same batch as the central CCD of MOS-2. Overlayed are the model QE derived
from tting to the ground based EPIC data (EMOS2 QUANTUMEF 0015.CCF) and also the model
QE (EMOS2 QUANTUMEF 0016.CCF) derived from calibration of an EPIC type device used in the
SWIFT project. Note that the ground based EPIC measurements below 1 keV show large non-statistical
variations indicating a considerable systematic uncertainty in individual measurements.
3 Scientic Impact of this Update
The change has been tested on a very soft source, the isolated neutron star RX0720.4-
3125, to investigate whether the extracted spectral parameters agree with those found
by other instruments. An observation of the source, from revolution 534, was t with
an absorbed black-body spectrum and the resultant t parameters (Table 1.) are found
to be in reasonably good agreement with those measured by the RGS from the same
observation [1]. A Chandra LETG observation made three years ealier shows a consistent
value of NH but a lower temperature [2]. This agrees with the RGS discovery that the
temperature of this target is evolving with time [1].
The residuals to the MOS spectral ts show no major deviations from the model at low
energies (Fig. 2), whereas the same model folded through MOS response matrices gener-
ated with the previous CCF shows an excess below 0.2 keV. This excess was responsible
for small underestimates of the Galactic absorption column.
2

Figure 2: The residuals to the best t of an absorbed black-body spectrum to the MOS spectra of
RX0720.4-3125 (rev 534) with the old (red) and new (black) QE and redistribution CCF les. The
improvement in the MOS-1 (top) and MOS-2 (bottom) ts below 0.2 keV is clear.
3

Table 1: Mos and Chandra ts to RX0720.4-3125
Detector N h Temperature
10 20 cm 2 eV
MOS-1 1:33  0:07 89:0  4:0
MOS-2 1:30  0:07 88:5  4:0
RGS 1:41  0:07 84:0  0:4
Chandra/LETG 1:32  0:14 81:4  1:3
4 Estimated Scientic Quality
The change in QE and rmf has the biggest impact for spectral analysis on soft or low
column density sources. For sources with a column density less than 5.010 20 cm 2 , it
is recommended that the MOS spectra be t down to around 150 eV. The accuracy of
the derived column in these instances is expected to be a few times 1.010 19 cm 2 . For
sources with column densities higher than around 5.010 20 cm 2 it is recommended that
the minimum included energy in spectral tting be determined by the low energy cut-o
of the probable incident spectrum. Observed counts below the cut-o energy are almost
all redistribution from higher energies.
5 Test procedures and results
The new CCF les have been used in an extensive cross-calibration excercise [3] and have
been found to be compatible with the SAS.
References
[1] Vries et al. 2004, A&A, 415L, 31D
[2] Kaplan et al. 2003, Apj, 590, 1008m
[3] Altieri, B., Kirsch, M., Pollock, A., Saxton, R., Chen, B., Smith, M. "XMM-Newton
instruments cross-calibration", XMM-SOC-CAL-TN-0052.
4