Home
Search
Collections
Journals
About
Contact us
My IOPscience
Cooling rate effects in thermoluminescence dosimetry grade lithium fluoride. Implications for practical dosimetry
This content has been downloaded from IOPscience. Please scroll down to see the full text. 1976 Phys. Med. Biol. 21 60 (http://iopscience.iop.org/0031-9155/21/1/005) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 132.236.27.111 This content was downloaded on 03/08/2015 at 17:17
Please note that terms and conditions apply.
PHYS.MED.
BIOL.,1976, VOL. 21, NO. 1,
60-66. 0 1976
Cooling Rate Effects in Thermoluminescence Dosimetry Grade Lithium Fluoride. Implications for Practical Dosimetry E. W. MASON, and I. CLARK
PH.D.,
A. F . McKINLAY,
PH.D.
National Radiological Protection Board, Scottish Centre, 11 West Graham Street, Glasgow G4 9LF, Scotland Received 30 A p r i l 1975, inftnal f o r m 22 A u g u s t 1975 ABSTRACT. A systematic investigation of the effects of cooling rates in the range 10" to 2 x lo5 'C min-lapplied to TLD-700, LiF thermoluminescence dosemeters has shown that the 'transfer sensitivity' effect observed by Booth, Johnson and Attix (1972) is only of importance for cooling rates > lo3 "Cmin-l. Although it is concluded that for practical dosimetry purposes the effect may be ignored it is not clear why Booth et al. observed such large changes and until this discrepancy is explained it is recommended that a low temperature pre-irradiation anneal should be used.
1. Introduction
The thermoluminescentproperties of dosimetrygradelithium fluoride thermal (Harshaw Chemical Co.) arecriticallydependentuponprevious history. Routine annealing procedures have been devised (Cameron, Suntharalingam and Kenney 1968) to overcome this problem. LiF in the form of solid blocks, rods or powder is typically annealed for 1 h a t 400 ' c and subsequently a t either 100 ' c for 1 or 2 h or a t 80 ' c for 20 h before exposure to radiation. The rate at which the phosphor is cooled after the high temperature anneal is exceedingly importantanddirectlydeterminesthe thermoluminescence sensitivity and the form of the glow pattern. It may, however, be desirable in certain circumstances to avoid this complex It requires timeand effort, and where pre-irradiationannealingprocedure. thermoluminescence dosemeters are used on a large scale a significant economic advantage may be gained by its elimination. Two reasons for the retention of a pre-irradiation anneal of LiF dosemeters are usuallyadvanced.First,the high temperatureannealandsubsequent cooling regenerates the original thermoluminescence sensitivity. Second, the low temperature anneal considerably decreases the concentration of shallow traps. These give rise to loss of the storedthermoluminescence signal (fading) as electrons escape during use a t normal temperatures. However, it has been demonstrated by Marshall, Shaw and Mason (1971) that the original thermoluminescence sensitivity of LiF, which has received no more than a few tens of rads of accumulated dose in its lifetime, may be faithfully regenerated by holding the dosemeter at an elevated temperature
Cooling Rate Eflects inTLD Lip
61
(300 "c)for a few seconds during the read cycle and then cooling to ambient temperature in the required reproducible manner. Furthermore, the fading problem may be eliminated (Marshall et al. 1971) by introducing a hold into the heating cycle during read-out for about 10 S a t 130 'c. This results in the removal of the low temperature glow peaks. The light output arising from the deeper dosimetry traps, which are not affected by fading, may then be integrated for the remainder of the heating cycle or for a selected part of it. Unfortunately, this procedure does not represent a universal solution to the annealingproblem.Dosimetrysystemshandlinglargenumbers of thermoluminescentdosemetersrequire that the readerthroughput(i.e.number of dosemeters processed per minute) should be maximized. The relatively lengthy holds in the read cycle may conflict with this requirement. I n addition an effect has been observed by several groups of workers, but particularlyprominentlybyBooth,Johnson andAttix (1972) which casts doubts upon the advisability of abandoning the pre-irradiationanneal.We shall refer to this effect as 'sensitivity transfer'. They observed that those thermoluminescence glow peaks in LiF normally used for dosimetry show a marked growth as a function of storage time. The magnitude of this effect was greatest when the initial concentration of shallow traps was greatest. Increases amounting to 30 or 40% in the dosimetry peaks were observed during storage a t both room temperature and 100 'c. The effect occurred more rapidly at the higher temperature but the mechanism appeared to be similar in both cases. These effects cannot be eliminated by a hold in the read cycle since they occur before read-outand could give rise to significant uncertaintyinthe thermoluminescence sensitivity. Since themagnitude of the effect appears to dependupon the initial prominence of the low temperature glow peaks which in turn depends upon the rate of cooling after a high temperature anneal (or hold in the reader), it is important that thevariation of the effect as a function of cooling rate should be determined. The present paper reports the results of a study of the variation of thermoluminescence sensitivity of LiF as a function of cooling rate after storage a t 100 or 80 'c. The general conclusions of thestudyare considered to be applicable for storage a t ambient temperature. We shall deal only with the implication of the results for practicalthermoluminescencedosimetry.A more detailed analysis of the data will be the subject of another report. 2.
Experimental procedure
For convenience, hot pressed LiF chips (Harshaw Chemical Co.) were used throughout the study but the results may be considered valid for other forms of LiF dosemeter. A high temperature anneal of 30 min a t 300 'c was selected to allow direct application of the results to the particular cases of LiF : PTFE dosemeters which cannot be annealed much above this temperature due to
62
E . W . Mason
et
al.
softening of the PTFE. LiF : PTFE discs form the basis of the TLD Individual Monitoring Service which is to be introduced by the National Radiological Protection Board in the near future. All thermoluminescence measurements employed reader a previously described (Mason and Linsley 1971). A linear heating rate was used during read-out. or spurious thermoluminescence effects were Non-radiation induced minimized by reading in a dry nitrogen atmosphere. The thermal annealing of the chips was in all cases performed on a heated nichrome plate, the temperature of which was controlled to within & 0.5 ' c by a specially designed controller. This unit, used in conjunction with a steady flow of nitrogen gas, also allowed the chips to be cooled a t a controlled linear rate up t o 3000 "C min-l. Faster, non-linear cooling rates were achieved either by theuse of nitrogen gasflow or by quenchingthe chips in carbon tetrachloride. I n these cases the cooling rate was measured as the average rate between the annealing temperature and 60 "c. After annealing at 300 "c for 30 min and cooling at the selected rate the chips were either (a) given no further thermal treatment, (b) annealed at 100 'c for 2 h, or (c) annealed a t 80 "c for 20 h before irradiation and read-out. was by means of a Irradiationto a standard dose of 5 rad(inair) 2 mCi 90Sr/90Yp-source. Read-out was effected within about15 min of irradiation and the chips were read with the irradiated face toward the photomultiplier tube to avoid variation in thermoluminescence response due to self-shielding from the P-rays. 3. Results and discussion It has become conventional to assign numbers to the peaks appearing in the glow pattern of thermoluminescence grade LiF above ambient temperature. In the present series of experiments up to five glow peaks were observed although under different conditions it is possible to observe other peaks at higher temperatures. Fig. 1 shows a glow pattern for chips cooled a t lo3 'cmin-l. Peaks 1-5 are numbered. Peak 1 occurs a t 60 "c and is of no dosimetric interest since it fades very rapidly. cooling from the high Peak 2 (120 "C)isnormallyveryprominentafter temperature of the first anneal and is relatively well resolved from peak 3. Peaks 3 and 4 (170 and 190 'c respectively) are usually not well resolved and it is normally only possible to discuss their behaviour in qualitative terms. Peak 5 occurs in the region of 210 "c and is the peak normally used in dosimetry although some unresolved contribution from peak 4 is also likely in many cases.
Cooling Rate Effects in L i F T L D
63
Temperature l "C I
Fig. 1 . LiF extruded ribbons. Typical glow curve for dosemeter cooled at lo3 ' C min-l.
3.1. N o low temperature anneal
Thevariationsin glow peakheight of t'he resolved peaks 2 and 5 asa function of the cooling rate after annealing at 300 'c for 30 min are presented in fig. 2 . The effect of cooling rate is distinctly different for the two peaks, Peak 2 increases continuously with increasing cooling rate over the whole range, 10-1 t o 2 x lo4 Ocrnin-l, examined.
I 10-1
lo'
lo2 lo3 IO' lo5 Cooling rate ("C rnin-l) Fig. 2. LiF e'xtruded ribbons. Effect of cooling rate on peak 5 and peak 2 for different
anneal regimes.
IO0
E.
64
W. Mason et al.
I n contrast,the maindosimetrypeak 5 , increases initiallyin the range 10-1 to about lo2 '~min-l and then declines by approximately 20% as the cooling rate is further increased. For a thermoluminescence dosimetry system based on LiF and employing no separate low temperature anneal the effects of fading or of 'sensitivity transfer' from shallow to deep trapsmust be minimized. If, forexample, peak 2 is a large proportion of peak 5 significant 'sensitivity transfer' is to be expected according to Booth et al. (1972). Fig. 3 (curve A) indicates that the ratio of peaks 2 and 5 has a minimum value in the cooling range 1-10 'c min-l forlinear cooling. Other workers, Dhar, De Werdand Stoebe (1973) and Harris and Jackson (1968), have reported minima a t 50-100 'c min-l for nonlinear cooling. r
BO-
- 70S
Low temperature anneal A no LT anneal
-
IO 1
la'
100
I
IO'
I
IOZ
I
I
Id
IO'
IOS
Cooling rate ( C rnin"~
Fig. 3. LiF extruded ribbons. Percentage ratio of peak 2 : peak 5 as a function of cooling rate for different anneal regimes.
The values of cooling rate in the region of the minimum are not only too slow for practical application, but the ratio is never sufficiently small to allow 'sensitivity transfer' to be ignored. It was expected that storing at 80 or 100 'c which results in a decrease in peak 2 (and 3) might give rise to significant increases in peak 5 as observed by Booth et al. (1972). 3.2. Low temperature pre-irradiation storage
The data for peaks 2 and 5 obtained after storage of cooled LiF chips at 'C for 20 h are included in fig. 2. In bothcases storage results in a dramaticdecrease in peak 2 (and in peak3) over almost the whole cooling range. Peak 2 constitutes only about 5-6% of peak 5 for cooling rates in excess of 10 Ocmin-l (fig. 3, curves B and C). The main dosimetry peak 5 , however, does not show the expected complementary increase in intensity over the range 10-1 to 10-3 ' c min-l. For faster cooling rates substantial increases in peak 5 are observed but only for the 100 'c for 2 h or 80
Cooling Rate Effects in
LiF TLD
65
fastest cooling does this approach the 30-40% increases observed by Booth et al. a t 100 ‘c. It is clear that the ‘sensitivity transfer’ effect is only significant for very rapid cooling and on the basis of the present results should not constitute a problem for most practical dosimetry systems which usually employ cooling rates of no more than lo3 ‘c min-l. The results of Booth et al. are not readily explained on the basis of the present data. Themost rapid cooling which could be expected for their particular reader and dosemeters seems unlikely to be in the region of the value of 2 x lo4 Ocmin-l required by us inorder t o observe comparable (30-40y0) increases in peak 5 during storage. It may be that thiseffect is batch dependent, changing with slight variations in the added impurity levels. Conclusions This detailed study of TLD-700 LiF has confirmed the importanceof cooling rate as a determining factor of thermoluminescence sensitivity for dosemeters annealed a t 300 ‘c. Ultimately the factors which determine the feasibility of using LiF without high and low temperature annealing are the throughput requirements of the system and the magnitude of the ‘sensitivity transfer’ effect as described by Booth et al. On the basis of the presentresults it would seem that for allpractical dosimetry purposes the latter would not arise since cooling in the reader is not normally excessively rapid. Indeed at least one modern reader (Pitman Ltd 1974) includes a facility for controlling the cooling rate and the problem may be deliberately avoided by this means. However, it is not clear why Booth et al. observed such large changes with their system and further work in this area with different batches of TLD-700 and different dosemeter forms is required. Until thisdiscrepancy is eliminated it is recommended that a low temperature pre-irradiation anneal shouldbe used to remove shallow traps thereby avoiding fading, ‘sensitivity transfer’ and throughput problems. 4.
This work was part funded under Euratom ContractNo. 129-74-1 B10 U.K.
R~SUM~ Influence du taux de refroidissement dans la nuance du fluorure de lithiumpour la dosimbtrie par thermo-luminescence. Implications pour la dosimktrie pratique Une Qtudesystkmatique de l’influence qu’ont les taux derefroidissement dans la plage de 10” L 2 x lo5 “C min-l appliques a des dosembtres a thermo-luminescence LiF du modele TLD-700 a montr6 que l’effet de ‘sensibilitb de transfert’ observe par Booth, Johnson et Attix (1972) n’est important que pour destaux derefroidissement > lo3 “C min-l. Bien que 1’Qtudeconclue que dans des buts de dosimetrie pratique on peut ignorer cet effet, il n’est pas clair pourquoi Booth et al. observerent des changements aussiimportants et,jusqu’h ce que cette anomalie soit expliqube, on recommande l’emploi d’un recuit avant irradiation B faible tempbratwe.
3
E. W . Mason et al.
66
ZUSAMMENFASSUNG Auswirkungen derKuhlrate aufLithiumfluoridin der Thermoluminiszenz-Dosimetrie. Implikationen fur die praktische Dosimetrie Eine systematischeUntersuchungder Auswirkungen von Kuhlraten im Bereich 10-1 bis 2 x lo5 "C min-l auf TLD-700 LiF-Thermoluminiszenz-Dosimeterhat erbracht, dass der von (1972) beobachtete'Transfersensitivitatseffekt' nur bei Kuhlraten Booth,JohnsonundAttix > lo3 "C min" von Bedeutung ist. Obwohl man zu dem Schluss kommt, dass in der praktischen Dosimetrie der Effekt ignoriert werden kann, ist nicht klar, warum Booth et al. derart grosse Veranderungenbeobachteten. Bis diese Diskrepanz aufgelost ist, wird empfohlen, dass eine Siedrigtemperatur-Prairradiations-Gluhanlage benutzt wird.
Pes~o~e 3@(&KT
CKOPOCTH OXJIaXgeHUR Ha @TOPHA JIHTUR B TepMOJIIoMIIHeClleHTHOfi A03HMeTpUH. 3HaYeHUe AJIR IIpaKTHYeCKOfi AOJUMeTPUU
~a c K o p o c T e f i oxnamema B ~ ~ a n a s OT o ~10" e C u c T e M a T m e c K o e H c c n e A o B a H H e s @ @ e r pa3~b1x 2 X io5 MUH-l, IIpHMeHReMOrO Ha TLD-700 C TepMOnIoMIIHeCqeHTHblMU A03UMeTpaMH LiF n O K a 3 a J I 0S, T 03 @ @ e K T 'YYBCTBHTenbHOCTU DO IIepeAaYe',OTMeYeHHbIfiEYTOM,&KOHCOHOM U 3TTEIXC (1972) RBJIReTCR 3Ha'MTeJIbHbIMTOJIbXO npII CKOPOCTRX OXJIaXAeHIIRBXIpeAeJIaX MeHee MIIH". XOTR B 3aKJIH)YeHUH yKa3blBaeTCR, YTO AJIR IIpaKTUYeCKOPA03AMeTPHUMOXHO I I p e H e 6 p e r a T b 3 T U M 3@@KTOM, OCTaeTCR HeRCHblM, llOYeMy 6 y T U dp. 0 6 H a p y X H B a J I U TaKMe 3Ha'MTeJIbHbIeKOJIe6aHUR II DOKa 3 T 0 HeCOOTBeTCTBUe H e 06€.RCHeHO, PeKOMeHAyeTCRDOJIb30BaTbCR HU3KOTeMIIepaTypHbIMOTXUTOM A 0 IIpeABapHTeJlbHOrO O ~ ~ ~ Y ~ H H R . A0
"c
1O3'c
REFERENCES BOOTH, L. F., JOHNSON, T. L.,and ATTIX,F. H., 1972, Health Phys., 23, 137-142. CAMERON,J. R., SUNTHARALISGAM, W.,and KESNEY,G. N., 1968, Thermoluminescent Dosimetry (Wisconsin University Press). DHAR,A., DE WERD,L. A., and STOEBE,T. G., 1973, Health Phys., 25, 427-432. HARRIS, A. M., and JACKSOX, H., J. 1968, Health Phys., 15, 457-461. MARSHALL,T. O., SHAW, K. B., and MASON, E. W., 1971, in Proc. 3rd Int. Conf. on Luminescence Dosimetry, R i m Report 249 (Danish Atomic Energy Agency, Rim) pp. 530-549. MASON, E. W., and LINSLEY,G. S., 1971, in Proc. 3rd Int.Conf. o n Luminescence Dosimetry, Risa Report 249 (Danish Atomic Energy Agency, Rim) pp. 164-182.
Search
Collections
Journals
About
Contact us
My IOPscience
Cooling rate effects in thermoluminescence dosimetry grade lithium fluoride. Implications for practical dosimetry
This content has been downloaded from IOPscience. Please scroll down to see the full text. 1976 Phys. Med. Biol. 21 60 (http://iopscience.iop.org/0031-9155/21/1/005) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 132.236.27.111 This content was downloaded on 03/08/2015 at 17:17
Please note that terms and conditions apply.
PHYS.MED.
BIOL.,1976, VOL. 21, NO. 1,
60-66. 0 1976
Cooling Rate Effects in Thermoluminescence Dosimetry Grade Lithium Fluoride. Implications for Practical Dosimetry E. W. MASON, and I. CLARK
PH.D.,
A. F . McKINLAY,
PH.D.
National Radiological Protection Board, Scottish Centre, 11 West Graham Street, Glasgow G4 9LF, Scotland Received 30 A p r i l 1975, inftnal f o r m 22 A u g u s t 1975 ABSTRACT. A systematic investigation of the effects of cooling rates in the range 10" to 2 x lo5 'C min-lapplied to TLD-700, LiF thermoluminescence dosemeters has shown that the 'transfer sensitivity' effect observed by Booth, Johnson and Attix (1972) is only of importance for cooling rates > lo3 "Cmin-l. Although it is concluded that for practical dosimetry purposes the effect may be ignored it is not clear why Booth et al. observed such large changes and until this discrepancy is explained it is recommended that a low temperature pre-irradiation anneal should be used.
1. Introduction
The thermoluminescentproperties of dosimetrygradelithium fluoride thermal (Harshaw Chemical Co.) arecriticallydependentuponprevious history. Routine annealing procedures have been devised (Cameron, Suntharalingam and Kenney 1968) to overcome this problem. LiF in the form of solid blocks, rods or powder is typically annealed for 1 h a t 400 ' c and subsequently a t either 100 ' c for 1 or 2 h or a t 80 ' c for 20 h before exposure to radiation. The rate at which the phosphor is cooled after the high temperature anneal is exceedingly importantanddirectlydeterminesthe thermoluminescence sensitivity and the form of the glow pattern. It may, however, be desirable in certain circumstances to avoid this complex It requires timeand effort, and where pre-irradiationannealingprocedure. thermoluminescence dosemeters are used on a large scale a significant economic advantage may be gained by its elimination. Two reasons for the retention of a pre-irradiation anneal of LiF dosemeters are usuallyadvanced.First,the high temperatureannealandsubsequent cooling regenerates the original thermoluminescence sensitivity. Second, the low temperature anneal considerably decreases the concentration of shallow traps. These give rise to loss of the storedthermoluminescence signal (fading) as electrons escape during use a t normal temperatures. However, it has been demonstrated by Marshall, Shaw and Mason (1971) that the original thermoluminescence sensitivity of LiF, which has received no more than a few tens of rads of accumulated dose in its lifetime, may be faithfully regenerated by holding the dosemeter at an elevated temperature
Cooling Rate Eflects inTLD Lip
61
(300 "c)for a few seconds during the read cycle and then cooling to ambient temperature in the required reproducible manner. Furthermore, the fading problem may be eliminated (Marshall et al. 1971) by introducing a hold into the heating cycle during read-out for about 10 S a t 130 'c. This results in the removal of the low temperature glow peaks. The light output arising from the deeper dosimetry traps, which are not affected by fading, may then be integrated for the remainder of the heating cycle or for a selected part of it. Unfortunately, this procedure does not represent a universal solution to the annealingproblem.Dosimetrysystemshandlinglargenumbers of thermoluminescentdosemetersrequire that the readerthroughput(i.e.number of dosemeters processed per minute) should be maximized. The relatively lengthy holds in the read cycle may conflict with this requirement. I n addition an effect has been observed by several groups of workers, but particularlyprominentlybyBooth,Johnson andAttix (1972) which casts doubts upon the advisability of abandoning the pre-irradiationanneal.We shall refer to this effect as 'sensitivity transfer'. They observed that those thermoluminescence glow peaks in LiF normally used for dosimetry show a marked growth as a function of storage time. The magnitude of this effect was greatest when the initial concentration of shallow traps was greatest. Increases amounting to 30 or 40% in the dosimetry peaks were observed during storage a t both room temperature and 100 'c. The effect occurred more rapidly at the higher temperature but the mechanism appeared to be similar in both cases. These effects cannot be eliminated by a hold in the read cycle since they occur before read-outand could give rise to significant uncertaintyinthe thermoluminescence sensitivity. Since themagnitude of the effect appears to dependupon the initial prominence of the low temperature glow peaks which in turn depends upon the rate of cooling after a high temperature anneal (or hold in the reader), it is important that thevariation of the effect as a function of cooling rate should be determined. The present paper reports the results of a study of the variation of thermoluminescence sensitivity of LiF as a function of cooling rate after storage a t 100 or 80 'c. The general conclusions of thestudyare considered to be applicable for storage a t ambient temperature. We shall deal only with the implication of the results for practicalthermoluminescencedosimetry.A more detailed analysis of the data will be the subject of another report. 2.
Experimental procedure
For convenience, hot pressed LiF chips (Harshaw Chemical Co.) were used throughout the study but the results may be considered valid for other forms of LiF dosemeter. A high temperature anneal of 30 min a t 300 'c was selected to allow direct application of the results to the particular cases of LiF : PTFE dosemeters which cannot be annealed much above this temperature due to
62
E . W . Mason
et
al.
softening of the PTFE. LiF : PTFE discs form the basis of the TLD Individual Monitoring Service which is to be introduced by the National Radiological Protection Board in the near future. All thermoluminescence measurements employed reader a previously described (Mason and Linsley 1971). A linear heating rate was used during read-out. or spurious thermoluminescence effects were Non-radiation induced minimized by reading in a dry nitrogen atmosphere. The thermal annealing of the chips was in all cases performed on a heated nichrome plate, the temperature of which was controlled to within & 0.5 ' c by a specially designed controller. This unit, used in conjunction with a steady flow of nitrogen gas, also allowed the chips to be cooled a t a controlled linear rate up t o 3000 "C min-l. Faster, non-linear cooling rates were achieved either by theuse of nitrogen gasflow or by quenchingthe chips in carbon tetrachloride. I n these cases the cooling rate was measured as the average rate between the annealing temperature and 60 "c. After annealing at 300 "c for 30 min and cooling at the selected rate the chips were either (a) given no further thermal treatment, (b) annealed at 100 'c for 2 h, or (c) annealed a t 80 "c for 20 h before irradiation and read-out. was by means of a Irradiationto a standard dose of 5 rad(inair) 2 mCi 90Sr/90Yp-source. Read-out was effected within about15 min of irradiation and the chips were read with the irradiated face toward the photomultiplier tube to avoid variation in thermoluminescence response due to self-shielding from the P-rays. 3. Results and discussion It has become conventional to assign numbers to the peaks appearing in the glow pattern of thermoluminescence grade LiF above ambient temperature. In the present series of experiments up to five glow peaks were observed although under different conditions it is possible to observe other peaks at higher temperatures. Fig. 1 shows a glow pattern for chips cooled a t lo3 'cmin-l. Peaks 1-5 are numbered. Peak 1 occurs a t 60 "c and is of no dosimetric interest since it fades very rapidly. cooling from the high Peak 2 (120 "C)isnormallyveryprominentafter temperature of the first anneal and is relatively well resolved from peak 3. Peaks 3 and 4 (170 and 190 'c respectively) are usually not well resolved and it is normally only possible to discuss their behaviour in qualitative terms. Peak 5 occurs in the region of 210 "c and is the peak normally used in dosimetry although some unresolved contribution from peak 4 is also likely in many cases.
Cooling Rate Effects in L i F T L D
63
Temperature l "C I
Fig. 1 . LiF extruded ribbons. Typical glow curve for dosemeter cooled at lo3 ' C min-l.
3.1. N o low temperature anneal
Thevariationsin glow peakheight of t'he resolved peaks 2 and 5 asa function of the cooling rate after annealing at 300 'c for 30 min are presented in fig. 2 . The effect of cooling rate is distinctly different for the two peaks, Peak 2 increases continuously with increasing cooling rate over the whole range, 10-1 t o 2 x lo4 Ocrnin-l, examined.
I 10-1
lo'
lo2 lo3 IO' lo5 Cooling rate ("C rnin-l) Fig. 2. LiF e'xtruded ribbons. Effect of cooling rate on peak 5 and peak 2 for different
anneal regimes.
IO0
E.
64
W. Mason et al.
I n contrast,the maindosimetrypeak 5 , increases initiallyin the range 10-1 to about lo2 '~min-l and then declines by approximately 20% as the cooling rate is further increased. For a thermoluminescence dosimetry system based on LiF and employing no separate low temperature anneal the effects of fading or of 'sensitivity transfer' from shallow to deep trapsmust be minimized. If, forexample, peak 2 is a large proportion of peak 5 significant 'sensitivity transfer' is to be expected according to Booth et al. (1972). Fig. 3 (curve A) indicates that the ratio of peaks 2 and 5 has a minimum value in the cooling range 1-10 'c min-l forlinear cooling. Other workers, Dhar, De Werdand Stoebe (1973) and Harris and Jackson (1968), have reported minima a t 50-100 'c min-l for nonlinear cooling. r
BO-
- 70S
Low temperature anneal A no LT anneal
-
IO 1
la'
100
I
IO'
I
IOZ
I
I
Id
IO'
IOS
Cooling rate ( C rnin"~
Fig. 3. LiF extruded ribbons. Percentage ratio of peak 2 : peak 5 as a function of cooling rate for different anneal regimes.
The values of cooling rate in the region of the minimum are not only too slow for practical application, but the ratio is never sufficiently small to allow 'sensitivity transfer' to be ignored. It was expected that storing at 80 or 100 'c which results in a decrease in peak 2 (and 3) might give rise to significant increases in peak 5 as observed by Booth et al. (1972). 3.2. Low temperature pre-irradiation storage
The data for peaks 2 and 5 obtained after storage of cooled LiF chips at 'C for 20 h are included in fig. 2. In bothcases storage results in a dramaticdecrease in peak 2 (and in peak3) over almost the whole cooling range. Peak 2 constitutes only about 5-6% of peak 5 for cooling rates in excess of 10 Ocmin-l (fig. 3, curves B and C). The main dosimetry peak 5 , however, does not show the expected complementary increase in intensity over the range 10-1 to 10-3 ' c min-l. For faster cooling rates substantial increases in peak 5 are observed but only for the 100 'c for 2 h or 80
Cooling Rate Effects in
LiF TLD
65
fastest cooling does this approach the 30-40% increases observed by Booth et al. a t 100 ‘c. It is clear that the ‘sensitivity transfer’ effect is only significant for very rapid cooling and on the basis of the present results should not constitute a problem for most practical dosimetry systems which usually employ cooling rates of no more than lo3 ‘c min-l. The results of Booth et al. are not readily explained on the basis of the present data. Themost rapid cooling which could be expected for their particular reader and dosemeters seems unlikely to be in the region of the value of 2 x lo4 Ocmin-l required by us inorder t o observe comparable (30-40y0) increases in peak 5 during storage. It may be that thiseffect is batch dependent, changing with slight variations in the added impurity levels. Conclusions This detailed study of TLD-700 LiF has confirmed the importanceof cooling rate as a determining factor of thermoluminescence sensitivity for dosemeters annealed a t 300 ‘c. Ultimately the factors which determine the feasibility of using LiF without high and low temperature annealing are the throughput requirements of the system and the magnitude of the ‘sensitivity transfer’ effect as described by Booth et al. On the basis of the presentresults it would seem that for allpractical dosimetry purposes the latter would not arise since cooling in the reader is not normally excessively rapid. Indeed at least one modern reader (Pitman Ltd 1974) includes a facility for controlling the cooling rate and the problem may be deliberately avoided by this means. However, it is not clear why Booth et al. observed such large changes with their system and further work in this area with different batches of TLD-700 and different dosemeter forms is required. Until thisdiscrepancy is eliminated it is recommended that a low temperature pre-irradiation anneal shouldbe used to remove shallow traps thereby avoiding fading, ‘sensitivity transfer’ and throughput problems. 4.
This work was part funded under Euratom ContractNo. 129-74-1 B10 U.K.
R~SUM~ Influence du taux de refroidissement dans la nuance du fluorure de lithiumpour la dosimbtrie par thermo-luminescence. Implications pour la dosimktrie pratique Une Qtudesystkmatique de l’influence qu’ont les taux derefroidissement dans la plage de 10” L 2 x lo5 “C min-l appliques a des dosembtres a thermo-luminescence LiF du modele TLD-700 a montr6 que l’effet de ‘sensibilitb de transfert’ observe par Booth, Johnson et Attix (1972) n’est important que pour destaux derefroidissement > lo3 “C min-l. Bien que 1’Qtudeconclue que dans des buts de dosimetrie pratique on peut ignorer cet effet, il n’est pas clair pourquoi Booth et al. observerent des changements aussiimportants et,jusqu’h ce que cette anomalie soit expliqube, on recommande l’emploi d’un recuit avant irradiation B faible tempbratwe.
3
E. W . Mason et al.
66
ZUSAMMENFASSUNG Auswirkungen derKuhlrate aufLithiumfluoridin der Thermoluminiszenz-Dosimetrie. Implikationen fur die praktische Dosimetrie Eine systematischeUntersuchungder Auswirkungen von Kuhlraten im Bereich 10-1 bis 2 x lo5 "C min-l auf TLD-700 LiF-Thermoluminiszenz-Dosimeterhat erbracht, dass der von (1972) beobachtete'Transfersensitivitatseffekt' nur bei Kuhlraten Booth,JohnsonundAttix > lo3 "C min" von Bedeutung ist. Obwohl man zu dem Schluss kommt, dass in der praktischen Dosimetrie der Effekt ignoriert werden kann, ist nicht klar, warum Booth et al. derart grosse Veranderungenbeobachteten. Bis diese Diskrepanz aufgelost ist, wird empfohlen, dass eine Siedrigtemperatur-Prairradiations-Gluhanlage benutzt wird.
Pes~o~e 3@(&KT
CKOPOCTH OXJIaXgeHUR Ha @TOPHA JIHTUR B TepMOJIIoMIIHeClleHTHOfi A03HMeTpUH. 3HaYeHUe AJIR IIpaKTHYeCKOfi AOJUMeTPUU
~a c K o p o c T e f i oxnamema B ~ ~ a n a s OT o ~10" e C u c T e M a T m e c K o e H c c n e A o B a H H e s @ @ e r pa3~b1x 2 X io5 MUH-l, IIpHMeHReMOrO Ha TLD-700 C TepMOnIoMIIHeCqeHTHblMU A03UMeTpaMH LiF n O K a 3 a J I 0S, T 03 @ @ e K T 'YYBCTBHTenbHOCTU DO IIepeAaYe',OTMeYeHHbIfiEYTOM,&KOHCOHOM U 3TTEIXC (1972) RBJIReTCR 3Ha'MTeJIbHbIMTOJIbXO npII CKOPOCTRX OXJIaXAeHIIRBXIpeAeJIaX MeHee MIIH". XOTR B 3aKJIH)YeHUH yKa3blBaeTCR, YTO AJIR IIpaKTUYeCKOPA03AMeTPHUMOXHO I I p e H e 6 p e r a T b 3 T U M 3@@KTOM, OCTaeTCR HeRCHblM, llOYeMy 6 y T U dp. 0 6 H a p y X H B a J I U TaKMe 3Ha'MTeJIbHbIeKOJIe6aHUR II DOKa 3 T 0 HeCOOTBeTCTBUe H e 06€.RCHeHO, PeKOMeHAyeTCRDOJIb30BaTbCR HU3KOTeMIIepaTypHbIMOTXUTOM A 0 IIpeABapHTeJlbHOrO O ~ ~ ~ Y ~ H H R . A0
"c
1O3'c
REFERENCES BOOTH, L. F., JOHNSON, T. L.,and ATTIX,F. H., 1972, Health Phys., 23, 137-142. CAMERON,J. R., SUNTHARALISGAM, W.,and KESNEY,G. N., 1968, Thermoluminescent Dosimetry (Wisconsin University Press). DHAR,A., DE WERD,L. A., and STOEBE,T. G., 1973, Health Phys., 25, 427-432. HARRIS, A. M., and JACKSOX, H., J. 1968, Health Phys., 15, 457-461. MARSHALL,T. O., SHAW, K. B., and MASON, E. W., 1971, in Proc. 3rd Int. Conf. on Luminescence Dosimetry, R i m Report 249 (Danish Atomic Energy Agency, Rim) pp. 530-549. MASON, E. W., and LINSLEY,G. S., 1971, in Proc. 3rd Int.Conf. o n Luminescence Dosimetry, Risa Report 249 (Danish Atomic Energy Agency, Rim) pp. 164-182.
