Analysis results for KamLAND PPO samples irradiated 7/13/2001 at MITR



Z. Djurcic1, Lin-Wen Hu2, A. Piepke1, J. Ritter3, R. Swinney3 and B. Tipton3

1University of Alabama
2MIT Nuclear Reactor Laboratory
3California Institute of Technology
(Status 9/7/2001)




On July 13, 2001 various KamLAND PPO samples have been activated at the MIT research reactor. The samples consisted of a mix of all PPO batches as delivered by Packard, PPO dissolved in pseudocumene, different blanks and spiked samples. Batches of PPO-pseudocumene mixture taken before and after water extraction were analyzed with the intend to study the effectiveness of the purification system.
The PPO samples obtained from Packard were prepared in the UA clean room. The PPO dissolved in pseudocumene was first evaporated to dryness and then filled into irradiation vials. This work was done in Caltech's new clean room. A total of 18 samples was then sent to MITR for irradiation. Some of the samples were identical and served to cross check UA's and Caltech's analysis consistency.
After the activation each institution chemically processed and counted half of the samples. UA and Caltech performed Np/Pa (the unstable U/Th activation products) extractions by means of ion exchange to enhance the sensitivity. The sample activities were then determined by low background gamma ray spectroscopy, using shielded Ge detectors.
A detailed description of the work done at Caltech can be found on the web.
These measurement were performed to investigate:

Links to detailed analysis description and results

Results


Radiopurity of the PPO


UA analysis
U and Th are not observed in any of the tested PPO samples using acid digestion followed by ion exchange. The following table lists the concentration we derive from the data. A short description of the samples may be found in the chemical analysis page. An overall efficiency of 75% has been used for the extraction of the activation products Np and Pa into ion exchange columns. The result given for the element K has been derived by counting the acid discharged from the ion exchange column. By analyzing 8 different PPO samples we determined that 84% (with a sigma of 6.6) of the chemically similar element Na remains in the acid which we are using as the K efficiency. PPO data is listed in red , blanks in blue.
Sample K
[ng/g] 		Th
[pg/g] 		U
[pg/g]
Packard PPO mix -380±320
-6.8±2.5
4.4±2.0 		-3.3±8.7
19.3±4.4
Unpurified
evaporated PC-PPO 		-1270±740
240±155 		33±16
8.5±2.1
-2.3±11
Purified
evaporated PC-PPO 		82±94
240±155 		-5.3±1.5
8.5±2.1 		0.5±3.6
-2.3±11


Caltech's analysis
The most sensitive determination of the uranium and thorium content of the samples comes from the extraction of activity into actinide columns. The full elemental analysis results are shown below. To summarize, we see no significant signal for U/Th in the samples. A radiochemical extraction efficiency correction of 65% has been applied. The concentrations reported are the concentrations in the irradiation ( liquid evaporated) PPO. The concentrations in the PPO+PC liquid are typically 1/5 of the reported values. For the potassium measurement no specific radiochemical extraction of potassium was tried in this analysis. So the best limits for potassium comes from the acids counting (listed below).
K U Th
Unpurified PPO+PC <26 ppb 18.5 ± 9.5 ppt 2.3 ± 1.7 ppt
Purified PPO+PC(2 samples) <590 ppb 8.9 ± 6.0 ppt 0.83 ± 1.03 ppt


Interpretation of the UA and Caltech data
A K, Th or U concentration significantly above the blank is neither observed at UA nor Caltech. At a concentration of 1.5 g/l the PPO constitutes only 1.9·10-3 of the scintillator mass. From the combined UA and Caltech data (using the most stringent limit) we derive the following 90 % c.l. limits for the contribution of the purified PPO to the radioisotope concentration in KamLAND scintillator:
40K Th U
<5.4·10-14 g/g <4.7·10-15 g/g <1.2·10-14 g/g

We hence conclude that the PPO is suited for KamLAND's reactor phase.

Effectiveness of the water extraction



UA analysis
The following table lists the element concentrations determined for KamLAND PPO. For Na, Br and Au the acid digestion efficiency is known to be 0.86, 0.43 and 0.18, respectively. For the other elements listed it is not known. In the following table we assume an efficiency of 100% for all elements. This approximation does not influence the sample comparison in which the efficiency cancels. Values given in red were determined for the PPO those in blue are the appropriate blanks. All samples were activated in PE vials.
Sample Na
[µg/g] 		Cr
[ng/g] 		Co
[ng/g] 		Zn
[ng/g] 		Br
[ng/g] 		Au
[pg/g]
Packard PPO mix 1.4±0.2
- 		16±2
1.6±0.2 		2±0.2
5.5±0.6 		62±6
33±3 		17±2
5±5 		0.7±0.4
-
Unpurified
evaporated PC-PPO 		0.1±0.01
0.073±0.008 		4.8±0.7
1.5±0.2 		3.1±0.3
- 		258±26
200±20 		8.3±0.8
0.3±0.05 		7.7±0.8
4.3±0.5
Purified
evaporated PC-PPO 		0.13±0.02
0.073±0.008 		5.2±0.5
1.5±0.2 		-
- 		52±5
200±20 		7.2±0.7
0.3±0.05 		4.8±0.5
4.3±0.5


Caltech's analysis
Activities related to 18 elements are found in the post-irradiation germanium detector spectrum. By analyzing the relative concentrations of these elements one may get a general picture of the effectiveness of the purification system, in removing ionic impurities from scintillator. The activities are isolated to the PPO by washing irradiation containers with strong, hot acid. One hopes that only activities related to the PPO impurity are removed in this step. Hence the table below shows element concentrations determined from activities in these acid washes. Some samples were washed more than once, to see how the element extraction changed; if the elements are in the PPO ( and not, say, in our blank materials) the concentration in the second wash should be significantly decreased.
Sample Na Cr Zn Br
Unpurified PPO(Plastic) 70 ± 6 ppb 1900 ± 190 ppt 25 ± 2 ppb 19.5 ± 2 ppb
Purified PPO(Plastic) 178 ± 20 ppb -40 ± 100 ppt 17 ± 1 ppb 32 ± 3 ppb
Purified PPO(Quartz) 72 ± 15 ppb -192 ± 428 ppt 460 ± 56 ppt 81 ± 10 ppb
Plastic Blank 242 ± 21 ppb 1.1 ± 0.3 ppb 100 ± 20 ppb 3.3 ± ppb
Quartz Blank 12 ± 2 ppb 52 ± 22 ppt 29 ± 4 ppt 0.3 ± ppb


Interpretation of the UA and Caltech data
An obvious reduction in the concentration of the listed chemical impurities is observed between the original PPO delivered from Packard and the dissolved-evaporated PPO. The reduction factors observed for Na, Cr and Br vary between 2 to 10. Co, Zn and Au have substantial blanks which make a reliable determination of reduction factors impossible. However, an improvement in chemical purity is not observed when comparing the purified and unpurified PPO samples. Br exhibits this the best because the blank is negligible (in all activations done in the past Br gave always the most stable and predictable results). This observation perhaps suggests that the Br is present in form of an organic compound which is insoluble in water. The ionic Na is only slightly above blank for both evaporated PPO samples. The purification factor might hence be masked by a blank fluctuation. The use of quartz irradiation vials might allow us to improve our sensitivity.
This page is maintained by A. Piepke
Last update September 7, 2001