The characteristics of a low background germanium gamma ray spectrometer at China JinPing underground laboratory

Author's Accepted Manuscript

The characteristics of a low background germanium gamma ray spectrometer at China JinPing underground laboratory Zhi Zeng, Yuhao Mi, Hao Ma, Jianping Cheng, Jian Su, Qian Yue

www.elsevier.com/locate/apradiso

PII: DOI: Reference:

S0969-8043(14)00225-5 http://dx.doi.org/10.1016/j.apradiso.2014.05.022 ARI6699

To appear in:

Applied Radiation and Isotopes

Received date: 7 January 2014 Revised date: 18 May 2014 Accepted date: 22 May 2014 Cite this article as: Zhi Zeng, Yuhao Mi, Hao Ma, Jianping Cheng, Jian Su, Qian Yue, The characteristics of a low background germanium gamma ray spectrometer at China JinPing underground laboratory, Applied Radiation and Isotopes, http://dx.doi.org/10.1016/j.apradiso.2014.05.022 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

  The characteristics of a low background germanium gamma ray spectrometer at China JinPing Underground Laboratory Zhi Zenga, Yuhao Mia, Hao Maa*, Jianping Chenga, Jian Sua, Qian Yuea a

Key Laboratory of Particle and Radiation Imaging (Ministry of Education) and Department of

Engineering Physics, Tsinghua University, Beijing 100084, China Abstract: A low background germanium gamma ray spectrometer, GeTHU, has been installed at China JinPing Underground Laboratory (CJPL). The integral background count rate of the spectrometer was 0.629 cpm between 40 and 2700 keV, the origins of which were studied by Monte Carlo simulation. Detection limits and efficiencies were calculated for selected gamma peaks. Some samples of rare event experiments were measured and 137Cs contamination was found in boric acid. GeTHU will be mainly used to measure environmental samples and screen materials in dark matter and double beta decay experiments. Key words: low-background, germanium spectrometer, CJPL

1.

Introduction Low background high purity germanium (HPGe) gamma ray spectrometry has been developed

for many years and applied in different fields such as fundamental physics and conventional sample investigations (Hult, 2007). Recently, as rare event experiments (e.g. neutrino oscillation, dark matter detection and 0νββ search) and monitoring of environmental radioactivity have attracted more and more attention, low background HPGe gamma spectrometers are playing an increasingly important role in material screening in such fields(Arpesella et al., 2002; Budjáš et al., 2009; Kohler et al., 2009). Such spectrometers are trying to achieve a lower background level so as to measure lower concentrations of radioactive contaminations. The background of low background HPGe spectrometers comes from: a) cosmic rays and cosmogenic radionuclides; b) radioactivity from primordial and man-made radionuclides in the surrounding environment of the laboratory i.e. rocks and soil; c) radioactivity in the construction materials and detectors themselves (Heusser, 1995; Hult et al., 2006; Laubenstein et al., 2004). To suppress background contributions, passive graded shields are installed with radiopure materials, e.g low background lead and oxygen-free high purity copper. The veto and coincidence system are also used to reduce the contribution from muons and Compton effect. (Heusser, 1995; Kohler et al., 2009; Laurec et al., 1996; Zastawny, 2003). Now low background HPGe spectrometers are preferred to be located in underground laboratories with thick rock overburden to achieve lower background.                                                                 *Corresponding author: Hao Ma, email address: [email protected].     

  China JinPing Underground Laboratory (CJPL), with a rock overburden of 2400 m (equal to about 6720 m w.e.), is located in the center of a transportation tunnel owned by JinPing hydropower station on Yalong River southwest of China (Cheng et al., 2011; Kang et al., 2010). In the laboratory the muon flux is about 2.0 × 10

−6

8

m-2s-1, reduced by a factor of about 10 compared to flux at sea

level (about 180 m-2s-1) (Wu et al., 2013). The radioactive contaminations in surrounding rocks are: 3.69-4.21 Bq/kg for 238U, 0.52-0.64 Bq/kg for 232Th and 4.28 Bq/kg for 40K (Zeng et al., 2014). A low background HPGe gamma ray spectrometer, called GeTHU, was developed in the laboratory in 2012, and will be dedicated in material screening and selections for dark matter detection experiment CDEX (Kang et al., 2013) and other rare event experiments. In this study, we will present the configuration and working characteristics of GeTHU. 2.

GeTHU facility GeTHU is based on a coaxial n-type HPGe detector of 40% relative detection efficiency and

was manufactured by CANBERRA in France. The germanium crystal has a height of 59.8 mm and a diameter of 59.9 mm. The cryostat is made of ultra-low background aluminum (ULB Al) and of U-shape to avoid direct line-of-sight from outside to the crystal. The preamplifier is positioned outside the shield due to the relatively high radioactivity. The data acquisition system is based on CANBERRA NIM modules including, a high voltage power supply (3106D), a spectroscopy amplifier (2026) and a multi-channel analyzer (Multiport II). CANBERRA Genie 2K software is used to analysis the data. As shown in Fig. 1, GeTHU has a large sample chamber with 30 × 30 × 63cm . The inner 3

shield surrounding the chamber is 5 cm of oxygen-free high purity copper (22 cm for the base plate), which has been polished with sandpaper and cleaned with anhydrous ethanol to remove residual surface contamination. Three layers of ordinary lead with a 210Pb activity of about 100 Bq/kg, each 5 cm thick, are surrounding the copper. All lead bricks were cleaned with anhydrous ethanol before being installed. The outermost is 10 cm of borated polyethylene plates which can prevent the penetration of ambient neutrons. The upper copper plate which supports the upper lead bricks and polyethylene plates is lain on sliding rails so that the sealed measurement chamber can be open and closed with the help of a hydraulic transmission equipment. The design of the shields paid special attention to avoiding direct line-of-sight to the crystal. The measurement chamber, with a dimension of 30 × 30 × 63cm , has a capacity of large volume samples such as an 8 L Marinelli beaker. The 3

entire facility is flushed with boil-off nitrogen from a dedicated LN2 tank to reduce the influence from 222Rn in the air. With N2 flushing, the

222

Rn concentration in the air inside the measurement

3

chamber can decrease from ~100 Bq/m to ~3 Bq/m3, resulting in an apparent background reduction. Routine samples are carefully processed with ultrasonic treatment, cleaned with anhydrous ethanol,  

  sealed in sample bags and well stored in a specified cabinet before measurement.

Fig. 1 The configuration of GeTHU spectrometer in CJPL.

3.

Results and discussion

3.1.

Background characteristics In routine measurements, the measurement chamber is flushed with boil-off nitrogen which

can suppress the influence of 222Rn in the air. To verify the effect of this measure and confirm the interference of

222

Rn existence, both a 12-day background measurement of GeTHU with nitrogen

flushing and a 14-day one without nitrogen flushing were completed in December 2012 and are compared in Fig. 3, in which main gamma peaks induced by

222

Rn are labeled as to the 14-day

spectrum. Obviously, after flushing detector with nitrogen, the background level dropped significantly and the integral count rate between 100~2700 keV was only 4% of that without nitrogen flushing.

 

 

214

214

1765keV: Bi

214

1120keV: Bi

214

609keV: Bi

214

242keV: Pb 214 295keV: Pb

without Nitrogen with Nitrogen

-1

-1

-1

count rate (kg s 0.36keV )

-3

10

352keV: Pb

-2

10

-4

10

-5

10

-6

10

0

500

1000

1500

2000

2500

3000

 

Fig. 2 Background spectra of GeTHU with (blue) and without (red) nitrogen flushing at CJPL. Count rates are in counts per kgGe per second and per channel of the MCA.

When measuring the 12-day background spectrum with nitrogen flushing, the activity concentration of

222

Rn in the experiment hall rose remarkably because of the malfunction of the

ventilation system. Especially in the first 4 days the concentration was up to about 300 Bq/m3, while in the last 8 days the concentration was only about 80 Bq/m3. Accordingly, the corresponding 8-day background spectrum was extracted to study the impact of the high radon concentration in surrounding environment. The count rates of main gamma peaks induced by daughters of 222Rn, are listed and compared in Table 1. Apparently, the count rates of interesting gamma peaks in both 12-day and 8-day spectra are close to each other and there is no obvious fluctuation in the background level. Therefore, the nitrogen flushing is proved to be effective to keep the concentration level inside the measurement chamber relatively stable.

 

222

Rn

  Table 1 Comparison of the count rates of gamma peaks induced by 222Rn with normal/fault condition of ventilation system in CJPL. Uncertainties are purely statistical. Peak count rate (day-1) Energy (keV)

242 295 352 609 1120 1765

Nuclide

12-day

8-day

(from 12/12 to 24/12)

(from 16/12 to 24/12)

214

Pb

<3.55

<4.45

214

Pb

<3.36

<4.04

214

Pb

6.09±0.96

6.95±1.19

214

5.10±0.73

6.30±0.91

214

<1.73

<2.26

214

<1.43

<1.93

Bi Bi Bi

In the 12-day background spectrum with nitrogen flushing, all gamma peaks were induced by primordial radionuclides, as indicated in Fig. 3. No gamma peaks from artificial radionuclides were found, which proved no obvious artificial radioactive contamination in detector and surrounding materials. Some gamma peaks are prominent, such as the 46.5 keV peak of 210Pb mainly from the lead shield; the 295 keV, 352 keV ones of 214Pb and the 609 keV, 1765 keV ones of 214Bi from the shield materials or 222Rn in the air; the 1461 keV one of 40K mainly from rocks and so on. No signs of cosmogenic radionuclides, such as 57Co, 58Co, 65Zn (Kohler et al., 2009; Loaiza et al., 2011), are found in the spectrum, mainly due to 2-year storage of GeTHU in CJPL. The integral background count rate between 40 and 2700 keV as well as background count rates of single gamma peaks all calculated from the 12-day spectrum of GeTHU are listed in Table 2. The integral background count rates of GeTHU were 0.629±0.006 cts/(kg min) from 40 to 2700keV and 0.510±0.006 cts/(kg min) from 100 to 2700keV respectively. From 40 to 2700keV the data of GeTHU was larger than 0.021±0.001 cts/(kg min) of GeMPI in LNGS which is the most sensitive spectrometer in the world and comparable to those of GeMi(~0.385±0.005 cts/(kg min)) and GePaolo (~0.154±0.001 cts/(kg min)) in LNGS. Between 100 and 2700 keV it was comparable to 0.383±0.002 cts/(kg min) of Gator (Baudis et al., 2011; Heusser et al., 2006; Laubenstein, 2005).   (a)

 

 

-5

3.5x10

212

239keV: Pb

-1

-1

-1

count rate (kg s 0.36keV )

-5

3.0x10

214

352keV: Pb

-5

2.5x10

214

609keV: Bi

-5

2.0x10

-5

1.5x10

-5

1.0x10

-6

5.0x10

0.0 0

500

1000

1500

2000

2500

3000

Energy (keV)

  (b)

-5

3.5x10

210

X-rays

46.5keV: Pb -5

-5

2.5x10

X-rays

-5

2.0x10

-1

-1

-1

count rate (kg s 0.36keV )

3.0x10

-5

1.5x10

-5

1.0x10

-6

5.0x10

0.0 30

40

50

60

70

80

90

100

Energy(keV)

  (c)

 

 

-5

3.5x10

212

239keV: Pb

214

3.0x10

222

352keV: Pb( Rn) 214

222

242keV: Pb( Rn)

-5

2.5x10

226

186keV: Ra

214

222

295keV: Pb( Rn)

-5

2.0x10

-1

-1

-1

count rate (kg s 0.36keV )

-5

-5

1.5x10

-5

1.0x10

-6

5.0x10

0.0 100

150

200

250 300 Energy (keV)

350

400

                    (d)

-5

214

609keV: Bi

-5

2.0x10

-5

1.5x10

228

-1

-1

-1

count rate (kg s 0.36keV )

2.5x10

969keV: Ac

511keV -5

1.0x10

228

911keV: Ac

-6

5.0x10

0.0 400

500

600

700 800 Energy (keV)

900

1000

  (e)

 

 

-6

6.0x10

40

1461keV: K 214

-6

1120keV: Bi

4.0x10

214

-1

-1

-1

count rate (kg s 0.36keV )

-6

5.0x10

1765keV: Bi

-6

3.0x10

208

2614keV: Tl

-6

2.0x10

-6

1.0x10

0.0 1000

1500

2000 Energy (keV)

2500

3000

  Fig. 3 Background spectra of GeTHU measured for 12 days in CJPL: (a) 30~3000keV, (b) 30~100keV, (c) 100~400keV, (d) 400~1000keV, (e) 1000~3000keV. Count rates are in counts per kgGe per second and per channel of the MCA.

 

 

  Table 2 The integral background count rate and background count rates in the ±3σ regions of single gamma peaks . Uncertainties are purely statistical. Energy (keV) 239 352 511 609 662 911 1120 1173 1332 1461 1765

Peak/Integral count rate (day-1)

Nuclide 212

214

Pb

5.11±1.20

Pb

6.09±0.96

β+

3.85±0.71

214

Bi

5.10±0.73

137

Cs

<1.49

Ac

<1.86

228

214

Bi

<1.73

60

Co

<1.54

60

Co

<1.26

40

K

<1.90

214

Bi

<1.43 815.6

40~2700

In order to further understand the origins of the remaining background, a simulated background spectrum was obtained with Monte Carlo methods, in which gamma rays resulting from the

238

U

series, 232Th series and 40K in the lead and copper shields were taken into consideration. Meanwhile, beta radiation both from 210Bi, the daughter of 210Pb, in the lead bricks and from natural radioactive series in copper shields, was included to simulate possible bremsstrahlung contributions. The simulated result is shown in Fig. 4 and compared to the 12-day measured spectrum. It is clear that in the measured background spectrum, the 352 keV peak and the 609 keV one (both induced by

222

Rn)

are pronounced with more net counts, while in the simulated background spectrum such two peaks are lower with less net counts. This difference could indicate that 222Rn in the air should contribute a lot to the background. Besides, the 46.5 keV gamma peak (induced by 210Pb) is also prominent in the measured spectrum, while it can not be seen in the simulated one. This gamma peak could probably be generated by

210

Pb from the decay of

222

Rn in the air, which also helps demonstrating that

background from primodial radionuclides in the shields is not the dominant part and 222Rn in the air around the detector must contribute significantly. Additional measures will be developed to further reduce the background from 222Rn in the future.  

 

 

-5

3.5x10

210

46.5keV: Pb X-rays 212 239keV: Pb 226 186keV: Ra

-1

-1

-1

count rate (kg s 0.36keV )

-5

3.0x10

-5

2.5x10

214

352keV: Pb 214 295keV: Pb

-5

2.0x10

214

609keV: Bi -5

1.5x10

-5

1.0x10

-6

5.0x10

0.0 0

500

1000

1500

2000

2500

3000

Energy (keV)

  -5

3.5x10

-1

-1

-1

count rate (kg s 0.36keV )

-5

3.0x10

-5

2.5x10

-5

2.0x10

-5

1.5x10

214

214

352keV: Pb

609keV: Bi

X-rays

295keV: Pb

-5

1.0x10

214

-6

5.0x10

0.0 0

500

1000

1500

2000

2500

3000

Energy (keV)

  Fig. 4 Comparison of the measured background spectrum of GeTHU (top) with Monte Carlo simulated spectrum (bottom). Count rates are in counts per kgGe, per second and per channel of the MCA.

3.2.

Detection limits The detection limit, Ld, was calculated by the following equation with a 95% confidence level

(Gilmore et al., 2008):

Ld = 2.71 + 3.29 × 2 BR × tC × n

[1]

where BR represents the interference-free background count rate per energy channel which takes no account of background produced by samples themselves, tC stands for counting time and n is the channel width of the “ROI” which covers 2.54 FWHM around a specific peak. Results are given in  

  Table 3, where Ld was calculated both for the 12-day background spectrum with nitrogen flushing and the 14-day one without nitrogen flushing. The counts are normalized to 1 day for comparison.

Table 3 Detection limits of GeTHU for selected gamma lines. Nuclide

Energy (keV)

Ld (counts) 12-day spectrum

14-day spectrum

239

17.27

44.21

214

352

16.55

174.13

214

609

13.96

146.98

137

662

7.24

22.63

212

727

6.41

21.56

228

Ac

911

8.56

24.38

60

Co

1173

7.43

19.43

60

Co

1332

6.41

18.58

40

K

1461

8.70

19.89

1765

7.05

59.52

212

Pb Pb Bi Cs Bi

214

Bi

After flushing with nitrogen, detection limits for gamma lines related to 222Rn were reduced by ten times maximumly, demonstrating remarkable improvement by nitrogen flushing again. 3.3.

Energy resolution calibration The energy resolution of the detector, defined as the ratio of the full width of half maximum

(FWHM) to the energy of the gamma line, is calibrated with some prominent gamma peaks in the 14-day background spectrum without nitrogen flushing and shown in Fig. 5. The FWHM of the detector at 1332.5keV is 2.04keV.

 

 

0.012 experimental data fitting curve

FWHM/E

0.010 0.008 0.006 0.004 0.002 0.000 0

200 400 600 800 1000 1200 1400 1600 1800 2000 2200 Energy (keV)

Fig. 5

GeTHU’s energy resolution. The solid curve is a fit with the function

FWHM = 0.5472 + 0.0408 × E + 4.218 × 10−7 × E 2 .

3.4.

Efficiency calibration Monte Carlo simulation methods are used to calculate efficiencies in most of our

measurements. With the formula below, efficiencies can be calculated:

ε=

Anet N

[2]

where ɛ stands for the efficiency of a specific energy, Anet represents the net counts of the corresponding full energy peak and N correspond to the number of simulated events. Fig. 6 shows a schematic view of the detector of GeTHU. In this preliminary work, point sources are employed in the simulation of efficiencies and finally the energy dependency of efficiencies is shown with a curve (see Fig. 7). In simulations, point sources are located 25 cm above the endcap assuming isotropic emissions and a series of gamma rays from 238U series, 232Th series, 40

K and some artificial radionuclides such as

152

Eu is used. In Fig. 7, the curve is fitted with the

following function (Guan and Zhou, 2006):

ln ε = a1 × E + a2 + a3 × ln E

E

+

a4

E

+

a5

E2

+

a6

E3

+ a7 × ln E

[3]

where ɛ represents efficiencies and E stands for the energy of gamma rays. The efficiency of 1.33 MeV gamma ray from 60Co is calculated using the final fitting function and transformed into the relative efficiency, which is about 0.45 while the standard data from the manufacture is 0.4. The main cause of this error between two data is the inaccuracy of the detector model in the simulation process. Further research will be implemented on the efficiency simulation of complex geometries and corresponding measurements will be done at the same time.  

 

Ge crystal ULB Al

Fig. 6 Schematic view of the detector of GeTHU.

 

0.01

efficiency

simulated data fitting curve

1E-3

1E-4 10

100

1000

10000

Energy (keV)

  Fig. 7 The efficiency as a function of energy. The solid line represents a fit using equation [3].

3.5.

Sample measurements Boric acid and silica sand are both crucial raw materials in today’s industrial production, and

they are involved in some rare event experiments. Thus it is critical to investigate the radioactivity in such materials and ensure that no unexpected contamination occurs. Collected boric acid (0.384 kg) and silica sand (0.262 kg) samples were measured on GeTHU for about 3.2 days and 2 days respectively. Activity concentration for selected radionuclides was calculated with the following equation:

 

  [4]

· · ·

where Am represents activity concentration in mBq/kg, Nnet stands for net counts of the selected full energy peak, ɛ is the full energy peak efficiency from Monte Carlo simulation, t is the living time in second, γ is the emission probability and m is the sample mass in kg. During calculations, if Nnet is larger than the relevant critical limit, the activity can be given as a certain value; on the contrary, it can only be shown as smaller than the relevant upper limit (Gilmore et al., 2008). With the measurements of samples, minimum detectable activities (MDAs) were also calculated by Ld (Gilmore et al., 2008) in order to show the performance of GeTHU: ·

[5]

· ·

Results are listed in Table 4. As to 40K in common silica sand, the low activity concentrition can be 8808 Bq/kg, compared to 304±103 mBq/kg in the silica sample here, which proves the relatively high radiopurity of our sample (Community Material Assay Database). Table 4 shows that the radioactivity of

238

U and

232

Th series in boric acid is higher than that in silica sand, which

indicates the relatively radio-impurity of the boric acid. Moreover, 137Cs contamination was found in boric acid sample with an activity concentration of 21±6 mBq/kg. It is also clear that neither the daughters from the 238U decay chain nor those from the 232Th one reach a secular equilibrium as one always expects about natural decay chains. This deviation from a secular equilibrium is attributed to the fact that different daughters in one decay chain have different chemical properties and they are lost with different rates during all kinds of physical and chemical treatment, which finnaly breaks the secular equilibrium. Especially, gaseous 222Rn can escape from samples, which results in smaller radioactivity of its daughters remaining in the sample, and this can be confirmed here.

 

  Table 4 Radioactivity and MDAs of some radionuclides in the boric acid and silica sand samples. Uncertainties are purely statistical. Activity (mBq/kg) Series 238

U

238

U

238 232

U

Th

232

Th

232

Th

Nuclide 226

Ra

Boric acid

Silica sand

Boric acid

Silica sand

186.221

<209

<151

303

308

214

Pb

295.21

<30

<35

66

78

214

Pb

351.92

<21

<24

45

53

212

Pb

238.632

17±8

<10

26

23

Ac

911.07

34±16

26±12

44

14

Tl

583.14

11±4

<6

12

13

K

1460.8

311±111

304±103

331

290

Cs

661.65

21±6

<3

13

9

228

208 40 137

4.

MDA (mBq/kg)

Energy (keV)

Summary A new low background germanium spectrometer, GeTHU, has been installed and run at CJPL.

The integral background count rate from 40 to 2700 keV is about 0.63 cpm and detection limits for selected gamma lines are all less than 20 counts normalized to 1-day counting based on a 12-day background spectrum. It can reach a sensitivity on activity concentration of about ~mBq/kg when the measuring time is within 2 to 3 days. Although nitrogen flushing was used and the background was reduced observably,

222

Rn and its daughters still contribute significantly to the remaining

background. To reduce possible 222Rn related contribution, an air tight glove box and a set of copper bricks will be added into the facility. The latter will be arranged around the detector when measuring normal size samples, i.e. excluding Marinelli beakers, PMTs or some other large size samples. Also, to achieve lower background, low background lead bricks with 3 mBq/kg of 210Pb will be arranged in graded shielding to replace the inner layer of ordinary lead bricks. Boric acid and silica sand samples have been screened on GeTHU and

137

Cs contamination was found in the boric acid. At

present, GeTHU is operated under stable condition in CJPL and used to prior screening and selection of materials from CDEX experiment. In the future, it will be further validated different kinds of reference materials and also open to other rare event experiments and environmental monitoring.

 

  Acknowledgements This work is partly supported by National Science Foundation of China through No.11175099, 11075090, 11055002 and 11355001. We are grateful to Mr. Qin Jianqiang, He Qingju and other colleagues of CDEX collaboration for their help during the design and installation of the spectrometer. References Arpesella, C., Back, H.O., Balata, M., Beau, T., Bellini, G., Benziger, J., Bonetti, S., Brigatti, A., Buck, C., Caccianiga, B., Cadonati, L., Calaprice, F., Cecchet, G., Chen, M., Dadoun, O., D'Angelo, D., De Bari, A., de Bellefon, A., De Hass, E., de Kerret, H., Derbin, A., Deutsch, M., Di Credico, A., Elisei, F., Etenko, A., von Feilitzsch, F., Fernholz, R., Ford, R., Franco, D., Freudiger, B., Galbiati, C., Gatti, F., Gazzana, S., Giammarchi, M.G., Giugni, D., Goger-Neff, M., Goldbrunner, T., Golubchikov, A., Goretti, A., Grieb, C., Hagner, C., Hagner, T., Hampel, W., Harding, E., Hartmann, F.X., von Hentig, R., Heusser, G., Hult, M., Ianni, A., Ianni, A.M., Kiko, J., Kirsten, T., Kohler, M., Korga, G., Korschinek, G., Kozlov, Y., Kryn, D., LaMarche, P., Laubenstein, M., Lendvai, C., Loeser, F., Lombardi, P., McCarthy, K., Machulin, I., Malvezzi, S., Maneira, J., Manno, I., Manuzio, G., Martemianov, A., Masetti, F., Mazzucato, U., Meroni, E., Miramonti, L., Monzani, M.E., Musico, P., Neder, H., Niedermeier, L., Nisi, S., Oberauer, L., Obolensky, M., Ortica, F., Pallavicini, M., Papp, L., Perasso, L., Pocar, A., Raghavan, R.S., Ranucci, G., Rau, W., Razeto, A., Resconi, E., Riedel, T., Sabelnikov, A., Salvo, C., Scardoni, R., Schonert, S., Schuhbeck, K.H., Shutt, T., Simgen, H., Sonnenschein, A., Smirnov, O., Sotnikov, A., Skorokhvatov, M., Sukhotin, S., Tarasenkov, V., Tartaglia, R., Testera, G., Trincherini, P.R., Vyrodov, V., Vogelaar, R.B., Vignaud, D., Vitale, S., Wojcik, M., Zaimidoroga, O., Zuzel, G., 2002. Measurements of extremely low radioactivity levels in BOREXINO. Astroparticle Physics 18, 1-25. Baudis, L., Ferella, A., Askin, A., Angle, J., Aprile, E., Bruch, T., Kish, A., Laubenstein, M., Manalaysay, A., Undagoitia, T.M., 2011. Gator: a low-background counting facility at the Gran Sasso Underground Laboratory. Journal of Instrumentation 6, P08010. Budjáš, D., Gangapshev, A., Gasparro, J., Hampel, W., Heisel, M., Heusser, G., Hult, M., Klimenko, A., Kuzminov, V., Laubenstein, M., 2009. Gamma-ray spectrometry of ultra low levels of radioactivity within the material screening program for the GERDA experiment. Applied Radiation and Isotopes 67, 755-758. Cheng, J.P., Wu, S.Y., Yue, Q., Shen, M.B., 2011. A review of international underground laboratory developments. Physics 40, 149-154. Gilmore, G., Hemingway, J.D., Gilmore, G., 2008. Practical gamma-ray spectrometry 2nd Edition. Wiley Chichester. Guan, Y.X., Zhou, G., 2006. The Semi-empirical Formula Fitting of Germanium Detector Efficiency Curve. Journal of Zaozhuang University 23, 77-79. Heusser, G., 1995. Low-radioactivity background techniques. Annual Review of Nuclear and Particle Science 45, 543-590. Heusser, G., Laubenstein, M., Neder, H., 2006. Low-level germanium gamma-ray spectrometry at the μ Bqkg level and future developments towards higher sensitivity, in: P.P. Povinec, J.A.S.-C.E. (Ed.), Radionuclides in the Environment. Elsevier, Amsterdam, pp. 495-510. Hult, M., 2007. Low-level gamma-ray spectrometry using Ge-detectors. Metrologia 44,  

  S87-S94. Hult, M., Preusse, W., Gasparro, J., Kohler, M., 2006. Underground gamma-ray spectrometry. Acta Chimica Slovenica 53, 1. Kang, K.J., Cheng, J.P., Chen, Y.H., Li, Y.J., Shen, M.B., Wu, S.Y., Yue, Q., 2010. Status and prospects of a deep underground laboratory in China. Journal of Physics: Conference Series 203, 012028. Kang, K.J., Cheng, J.P., Li, J., Li, Y.J., Yue, Q., Bai, Y., Bi, Y., Cheng, J.P., Chen, N., Chen, N., 2013. Introduction of the CDEX experiment. arXiv preprint arXiv:1303.0601. Kohler, M., Degering, D., Laubenstein, M., Quirin, P., Lampert, M.O., Hult, M., Arnold, D., Neumaier, S., Reyss, J.L., 2009. A new low-level gamma-ray spectrometry system for environmental radioactivity at the underground laboratory Felsenkeller. Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine 67, 736-740. Laubenstein, M., 2005. The low background laboratory at the Gran Sasso National Laboratories. Laubenstein, M., Hult, M., Gasparro, J., Arnold, D., Neumaier, S., Heusser, G., Köhler, M., Povinec, P., Reyss, J.-L., Schwaiger, M., 2004. Underground measurements of radioactivity. Applied radiation and isotopes 61, 167-172. Laurec, J., Blanchard, X., Pointurier, F., Adam, A., 1996. A new low background gamma spectrometer equipped with an anti-cosmic device. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 369, 566-571. Loaiza, P., Chassaing, C., Hubert, P., Nachab, A., Perrot, F., Reyss, J.L., Warot, G., 2011. Low background germanium planar detector for gamma-ray spectrometry. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 634, 64-70. Wu, Y.C., Hao, X.Q., Yue, Q., Li, Y.J., Cheng, J.P., Kang, K.J., Chen, Y.H., Li, J., Li, J.M., Li, Y.L., 2013. Measurement of Cosmic Ray Flux in China JinPing underground Laboratory. arXiv preprint arXiv:1305.0899. Zastawny, A., 2003. The Measurements of low radioactivities. Zeng, Z., Su, J., Ma, H., Yi, H., Cheng, J., Yue, Q., Li, J., Zhang, H., 2014. Environmental gamma background measurements in China Jinping Underground Laboratory. Journal of Radioanalytical and Nuclear Chemistry, 1-8.

Figures Fig. 1

The configuration of GeTHU spectrometer in CJPL.

Fig. 2

Background spectra of GeTHU with (blue) and without (red) nitrogen flushing at CJPL.

Count rates are in counts per kgGe per second and per channel of the MCA. Fig. 3

Background spectra of GeTHU measured for 12 days in CJPL: (a) 30~3000keV, (b)

30~100keV, (c) 100~400keV, (d) 400~1000keV, (e) 1000~3000keV. Count rates are in counts per

 

  kgGe per second and per channel of the MCA. Fig. 4

Comparison of the measured background spectrum of GeTHU (top) with Monte Carlo

simulated spectrum (bottom). Count rates are in counts per kgGe, per second and per channel of the MCA. Fig. 5

GeTHU’s energy resolution. The solid curve is a fit with the function

FWHM = 0.5472 + 0.0408 × E + 4.218 × 10−7 × E 2 . Fig. 6

Schematic view of the detector of GeTHU.

Fig. 7

The efficiency as a function of energy. The solid line represents a fit using equation [3].

Tables

Table 1 Comparison of the count rates of gamma peaks induced by 222Rn with normal/fault condition of ventilation system in CJPL. Uncertainties are purely statistical.

Table 2 The integral background count rate and background count rates in the ±3σ regions of single gamma peaks . Uncertainties are purely statistical. Table 3 Detection limits of GeTHU for selected gamma lines. Table 4 Radioactivity and MDAs of some radionuclides in the boric acid and silica sand samples. Uncertainties are purely statistical.

  Highlights: 1. The first low background gamma ray spectrometer (GeTHU) was developed at CJPL. 2. It has a large inner chamber which can host large samples for different purposes. 3.The background characteristics are presented and the origin is studied. 4. Detection limits are given for selected radionuclides and efficies are calculated. 5. Some samples were measured and 137Cs contamination was found in boric acid.