Transcript Taratura ed uso di un transistore come dosimetro per il

First Planning and Coordination Meeting - Technical Coordination Project RER/8/010
Quality Control Methods and procedures for Radiation Technology
Warsaw 21-25 February 2005
In-plant calibration and use of power
transistors for process control of
gamma and electron beam facilities
Piergiorgio Fuochi
, Marco Lavalle
, Ugo Corda
, András Kovács
ISOF-CNR, Via P Gobetti, 101 – 40129 Bologna
Institute of Isotope and Surface Chemistry, Budapest, Ungheria
In-plant calibration and use of power transistors for process control of gamma and electron beam facilities
Introduction
A bipolar transistor, previously investigated as a possible radiation
dosimeter, has been tested under industrial irradiation conditions in high
activity gamma and high energy, high power electron beam facilities. Inplant calibrations have been performed against transfer standard alanine
and reference standard ethanol-monochlorobenzene (ECB) dosimeters over
an absorbed dose range of 5 to 45 kGy. Routine irradiations have been done
by placing the transistors, side by side with the routine dosimeters used in
the plants, on product boxes during production runs. The results of the
studies have shown that the agreement between the absorbed dose
measured by the transistors and other well established routine dosimeters is
within ±3%. This indicates that this type of transistors could be used as
routine radiation dosimeters in the dose range investigated.
In-plant calibration and use of power transistors for process control of gamma and electron beam facilities
Objectives
Silicon devices, as radiation monitors, were already
considered since mid-1960s both at low doses, as in
the field of radiation therapy, and at high doses as
for industrial radiation processes. In a previous work
the possibility of using a low cost, small size
commercial bipolar power transistor as a routine
dosimeter was investigated. Attention was focused
on the changes of carrier lifetime with irradiation,
knowing that these changes were proportional to the
absorbed dose.
That study, conducted under standard laboratory
conditions, revealed that this “dosimeter” was
capable of measuring doses in the range 0.1-25 kGy,
both in gamma and electron radiation fields.
A = 7,4 ÷ 7,8 mm
B = 10,5 ÷ 10,8 mm
C = 2,4 ÷ 2,7 mm
D = 15,7 mm
Since the damage to the silicon crystal structure produced by irradiation (which affect
the carrier lifetime) does not anneal at temperatures below 300 °C, the information
about the absorbed dose is not lost during transistor readout and therefore they can be
used for recording dosimetry history. It was also found that the response of the
transistor is energy dependent; therefore if the devices are calibrated and used in
radiation fields with significantly different radiation energy spectra, due to diffused and
scattered photons and electrons, differences in dosimeter response may be expected.
The aim of the present work is to characterize the behaviour and the performance of
these transistors for routine dose measurements in high activity gamma irradiators and
high energy, high power EB facilities by comparing them with other routine dosimeters
used in the plants.
In-plant calibration and use of power transistors for process control of gamma and electron beam facilities
Irradiation sources
Gamma irradiations
Gamma irradiations were performed at the
multipass, product overlap, pallet type
60Co facility of Gammarad Italia, Ca’ de’
Fabbri, Italy, with current activity of 6.9 x
1016 Bq.
Electron irradiations
Electron irradiations were carried out
with the 10 MeV TT100 type Rhodotron
at Gambro Dasco plant, Medolla, Italy,
under operating production conditions,
i.e. on a conveyor system with a beam
width of 83 cm and average pulse
current of 1.72 mA
In-plant calibration and use of power transistors for process control of gamma and electron beam facilities
Dosimetry systems
To minimize the contribution of influence quantities to the overall uncertainty and to ensure
similar irradiation conditions both for calibration and routine dosimetry during the production
run, full in-plant calibration of the transistors was performed.
Field trial irradiations were carried out under operating
conditions to check the applicability and the performance of
the transistors for routine application.
Calibration of transistors
The calibration of transistors in the gamma plant was done
against ECB dosimeters traceable to Risø High Dose Reference
Laboratory. Two vials containing ECB solutions and four
transistors were arranged in line inside a polystyrene phantom.
Several of such phantoms were attached to the sides of
product pallets, parallel to the source plaques and irradiated to
nominal doses from 5 up to 45 kGy.
At the EB facility transistors were calibrated against alanine
transfer standard dosimeter supplied by Risø High Dose
Reference Laboratory. Polystyrene phantoms from Risø
Laboratory holding one pack of alanine dosimeters and four
transistors, were placed in stainless steel trays generally used to
carry product boxes. These phantoms were then irradiated to
different nominal doses from 5 up to 45 kGy.
In-plant calibration and use of power transistors for process control of gamma and electron beam facilities
Field trial irradiations
For the field trial irradiations at gamma plant, eight packs of dosimeters were attached to
product pallets at a reference position and irradiated during production runs. Each
dosimeter pack contained four transistors and a Red 4034 PMMA routine dosimeter in close
contact, without shielding each other. Evaluation of the absorbed dose from the PMMA
dosimeters was done by Gammarad using their own dosimetry system traceable to NPL.
Field trial irradiations at EB facility were performed by placing dosimeter packs, each
consisting four transistors in close contact, on top of nineteen product boxes evenly
distributed during the production run. The routine dose measurements were done with
polystyrene calorimeter, traceable to Risø High Dose Reference Laboratory, according to
the usual irradiation procedure used by the facility.
Aluminium carrier
e- beam
palletted product
dosimeters
dosimeters
Conveyor movement
In-plant calibration and use of power transistors for process control of gamma and electron beam facilities
Field trial irradiations
Since the routine calorimeters were located
at different distance from the accelerator
window, as compared to the transistors, a
separate experiment was carried out to
check the influence of this difference on the
dose measurements. In a tray, along with a
routine polystyrene calorimeter another
polystyrene calorimeter was placed at the
same height as the transistors. The tray was
then sent through the irradiation zone
together with other product boxes during the
field trial. Both the calorimeters registered
almost the same dose: 28.44 kGy and 28.48
kGy. This allowed us to compare the dose
measured by the calorimeters and by the
transistors during routine irradiations.
Measurements of the parameter T for the
irradiated transistors were done at constant
temperature (25 °C) and the changes of T
were plotted as
(1/T)=1/T-1/T0,
where T0 and T are the pre-irradiation and
post-irradiation values respectively, against
the dose registered by the reference
standard ECB and transfer standard alanine
dosimeters.
Portable instrument used to measure a
physical parameter T directly related to the
carrier lifetime t.
Resistive load switching test circuit for
measuring the T parameter.
In-plant calibration and use of power transistors for process control of gamma and electron beam facilities
Results 1
Response curves
The results of the calibration at the gamma plant and at the EB facility are
shown in figures respectively. These data were used to establish the calibration
curves following the least squares fitting procedure. The polynomial functions
were selected according to the correlation coefficients and distribution of the
percentage residuals. The calibration curves thus obtained were then used for
response evaluation of the transistors irradiated together with other types of
routine dosimeters used in the plants during actual production runs.
0,12
0,015
0,09
0,06
0,005
 = 1/T - 1/T
0,010
 = 1/T - 1/T
0
0
0,020
0,03
Dose [kGy]
0,000
0
10
20
Dose [kGy]
0,00
30
GAMMARAD response curve
(gamma irradiation plant)
40
50
0
10
20
30
40
GAMBRO DASCO response curve
(electron beam irradiation plant)
50
In-plant calibration and use of power transistors for process control of gamma and electron beam facilities
Results 2
Comparison of dose mesured by transistors and routine dosimeters
Linear regression was applied to compare the absorbed dose results obtained in
the gamma plant, measured by the transistors and by the Red 4034 PMMA
dosimeters (Fig. 10). The overall combined uncertainty associated to the Red
PMMA is 6% at 95% confidence level, as reported by Gammarad. As for the
electron irradiation trial, the dose values determined by the irradiated transistors
placed on the top of nineteen boxes in non consecutive trays are plotted against
the average dose measured by the polystyrene calorimeters during the trial (Fig.
11). The overall combined uncertainty associated with the polystyrene
calorimeter is 3% at 95% confidence level, as reported by Gambro Dasco.
32
30
30
25
DP = 1,0258 × DT
28
2
R = 0,9842
26
Dose [kGy]
15
Dose PMMA [kGy]
20
•
Dose transistori ± 2,5%
media = 29,3 kGy (1 =0,6 kGy)
—Dose media calorimetro (28,5 kGy)
24
— Dose min e max calorimetro
10
22
N° transistore
Dose transistori [kGy]
5
20
5
10
15
20
GAMMARAD results
(gamma irradiation plant)
25
30
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19
GAMBRO DASCO results
(electron beam irradiation plant)
In-plant calibration and use of power transistors for process control of gamma and electron beam facilities
Conclusions
32
30
26
Dose [kGy]
28
•
Dose transistori ± 2,5%
media = 29,3 kGy (1 =0,6 kGy)
—Dose media calorimetro (28,5 kGy)
24
— Dose min e max calorimetro
22
N° transistore
20
1
2
3
4
5
6
7
8
9
30
25
DP = 1,0258 × DT
2
R = 0,9842
20
15
Dose PMMA [kGy]
The tests conducted on this type of transistors show the
good performances of this device in the production
irradiators and if they are calibrated in the plant they are
reliable dosimeters thus confirming that they can be used
as routine dosimeters in the dose range of 5 to 45 kGy
under production irradiation conditions at high activity
gamma and high energy, high power electron beam
facilities. As shown in figure, the slope of the line is close
to unity (0.972) with a correlation coefficient of 0.992,
indicating that the doses measured in the gamma plant
were in agreement within ±2.8%.
10
Dose transistori [kGy]
5
5
10
15
20
25
30
Similarly the results of the transistors irradiated at EB
facility during actual production runs demonstrate the
good agreement among them (0.6 kGy standard
deviation) and with the dose measured by the calorimeters
(only 2.8% higher than the average value measured by
the calorimeter). However, it has to be pointed out that
these transistors are not suitable for dose measurements
in EB irradiation facilities with energy <2 MeV.
10 11 12 13 14 15 16 17 18 19
Based on the response values of the transistors irradiated in the gamma facility for routine
measurements, the coefficient of variation is found to be 4.5% (1) even including a few
outliers. For electron irradiation, this value is much better, namely 2.1% (1) confirming the
results obtained previously under standard laboratory conditions.