Long-term study of low energy counting rate with the Large Volume - - PowerPoint PPT Presentation

Long-term study

• f low energy counting rate

with the Large Volume Detector

Gianmarco Bruno – LVD collaboration

Gran Sasso National Laboratory & L’Aquila University

July 3, 2009

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 1 / 16

Large Volume Detector

Main Features: Active Mass: M = 1kton 840 tanks: (1.0×1.0×1.5 m3) 2520 pmts: 15 cm diameter Liquid Scintillator: Cn H2n+2 n = 9.6, + 1g/l PPO + 0.03 g/l POPOP, ρ = 0.8 g/cm3 Thresholds: EH ≃ 4MeV & EL ≃ 1MeV Goal: The detector is mainly designed to measure low energy ¯ νe from stellar core collapse.

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 2 / 16

Trigger modes

LVD can operate at 2 different thresholds:

1

The signals of Each PMT are discriminated at two thresholds resulting in two possible levels of coincidence between a counter PMTs: H and L, corresponding to EH ≃ 4MeV and EL ≃ 1MeV. The H coincidence, in any counter, represents the scintillator trigger condition.

2

The single tank low threshold rate is monitored by a system of 840 scalers. The counting rate of each tank is measured during a time window of 10 s. The read out of this low priority data channel is enabled every 10 minutes by the: asyncronous monitoring trigger. At energies near EL the single tank counting rate is mainly due to: Rock radioactivity Building materials radioactivity Secondary particles generated by muons

222Rn

(α, n)−reactions

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 3 / 16

Correlation LTCR – Radonmeter

The Low Threshold Counting Rate (LTCR) and the radonmeter data are clearly

• correlated. The last 500 days of data are shown in the picture.

01/01/08 01/04/08 01/07/08 01/10/08 31/12/08 01/04/09 counts per counter (Hz) 10 20 30 40 50 60 70 80 90 100 )

3

Rn concentration (Bq/m

222

50 100 150 200 250 300 350 400 LTCR Rn-meter

2 questions: LVD sensitivity to Rn contamination. What are we counting?

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 4 / 16

Calibration of a single counter

13/07/08 14/07/08 15/07/08 16/07/08 17/07/08 18/07/08 LTCR (Hz) 10 20 30 40 50 60 70 80 90 100 )

3

Rn concentration (Bq/m

222

50 100 150 200 250 300 350 400

Lag (hours)

• 4
• 3
• 2
• 1

1 2 3 4 Correlation coefficient 0.4 0.5 0.6 0.7 0.8 0.9

)

3

Rn-meter (Bq/m 50 100 150 200 250 300 350 400 LTCR (Hz) 30 40 50 60 70 80

For each counter we determine the time lag at which the maximum of the cross-correlation function occurs, and the sensitivity in terms of radon

• activity. Sensitivities have been

evaluated by fitting the bivariate distribution at the lag corresponding to the maximum correlation.

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 5 / 16

Distribution of the calibration parameters

The distribution of the calibration parameters (angular coefficient of the straight line) and the distribution of the delays obtained for all the counters are shown.

Entries 791 Mean 0.264 RMS 0.1415 )

3

m

• 1

Radon sensitivity (Hz Bq

• 0.4
• 0.2

0.2 0.4 0.6 0.8 1 counters 10 20 30 40 50 60 70 80 Entries 791 Mean 0.264 RMS 0.1415 Entries 791 Mean 1.046 RMS 0.3246 delay (hours)

• 1
• 0.5

0.5 1 1.5 2 2.5 3 counters 20 40 60 80 100 120 140 160 180 200 Entries 791 Mean 1.046 RMS 0.3246

On average, a variation in Rn activity of 1 Bq/m3, corresponds to a variation in the single counters low threshold counting rate, of 0.3 ± 0.1 Hz. The average delay between a Rn-meter peak and the corresponding peak in the counter rate is: 1 ± 0.3 hours.

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 6 / 16

222Rn and its decay products

Radium series:

222Rn α

− − − →

3.82d 218Po α

− − − − →

3.10min 214Pb β

− − − − →

26.8min 214Bi β

− − − − →

19.9min 214Po α

− − − →

164µs 210Pb

The parent radionuclide A decay according to exponential law producing atoms of type B: A = A0e−λAt B change with a rate depending on:

1

the parent decay

2

the decay of B itself dB dt = AλA − BλB

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 7 / 16

Decay chain

The more generale case of a n members chain is described by a system of differential equation:

8 > > > > > > > > < > > > > > > > > : dA dt = −AλA dB dt = AλA − BλB dC dt = BλB − CλC . . .

50 100 150 200 0.0 0.2 0.4 0.6 0.8 1.0 Time HminL Activity H%L

214Bi 214Pb 218Po 222Rn

The time evolution of the signal has been calculated assuming that a certain quantity of radon (as measured by the Rn-meter) persists in the environment during 10 minutes.

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 8 / 16

Decay chain

The previously obtained curves can be regarded as pulse response functions of our system to a brief Rn injection. Discrete time convolution can be used to determine the output of a sampled data system from its input and pulse response. y(t) =

∞

• n=0

x(t − n)h(n) Thus, appliyng a convolution between each one of that curves and the radonmeter data series we can convert the activity of the radon in activity

• f the corresponding product.

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 9 / 16

Shape

time (days) 196 196.5 197 197.5 198 counts (arbitrary units) 80 90 100 110 120 130 140 150 160 170

Comparison between radonmeter data (red line) and LTCR data (black line), during a time period of intense variation in radon concentration due to a scheduled switch off of the ventilation system in the experimental hall.

time (days) 196 196.5 197 197.5 198 counts (arbitrary units) 80 90 100 110 120 130 140 150 160 170

Now the red line represent the activity of 214Bi calculated from the radon-meter data series. A significantly better agreement is achieved, explaining the delay measured, so we can state that:

• ur counters can measure the Rn concentration

because gammas from 214Bi (Eγ = 609KeV Iγ = 46.1%) are detected.

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 10 / 16

(α, n)-reactions

Time (days) 4 6 8 10 12 14 16 18 20 22 24 26 LTCR (Hz) 100 200 300 400 500 600 700 800 900 Reference tank Gd loaded tank background Radon signal Energy (MeV)

• 1

1 2 3 4 5 6 7 8 9 Events (Hz) 10 20 30 40 50 60 70 80 90 Reference tank Gd loaded tank

Counting rates versus time of a standard counter (red line) and a counter filled with Gd loaded scintillator (blue line) collected during a period of transition between high and low Rn contamination. Since we are studying the Rn contribution the spectra are background subtracted. From the two spectra, after Background subtraction, we can argue that the contribution of n from (α, n)-reactions is negligeable.

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 11 / 16

12 years of data series

Out[26]=

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 20 40 60 80 100 120 years counts per counter HHzL LTCR

Out[26]=

0.0 0.5 1.0 1.5 2.0 500 1000 1500 2000 Frequency Hdays-1L Power Spectral Density

We report the counting rate collected since 1997 up to 2009 and the power spectral density obtained applying DFT algorithm. The discontinuity in the middle of 2003 is related to upgrade in the ventilation system.

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 12 / 16

Weekly modulation

days 1 2 3 4 5 6 7 8

counts per counter (Hz) 44 45 46 47 48 49

MON TUE WED THU FRI SAT SUN

days 1 2 3 4 5 6 7 8

counts per counter (Hz) 62 64 66 68 70 72

The counting rate behaviour of the average week is inverted before and after the discontinuity of 2003.

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 13 / 16

Power Spectral Density (data since 1997 to 2003)

Out[28]=

0.5 1.0 1.5 2.0 200 400 600 800 Frequency Hdays-1L Power Spectral Density

Out[37]=

0.00 0.02 0.04 0.06 0.08 0.10 200 400 600 800 1000 1200 1400 Frequency Hdays-1L Power Spectral Density

Notations: background level, 3σ c.l., 7σ c.l.

1

The 4 peaks in simmetrical position with respect to the 1 day−1 frequency exceeding the 7 σ c.l. are related to the daily signal structure.

2

The remaining two peaks represents the weekly modulation (0,14 day−1) and its higher harmonic.

3

In the low frequency spectrum the higher peak corresponds to a frequency compatible with an annual modulation:

frequency: 365±32 d amplitude: 1.5 – 2.5 Hz

4

The peaks exceeding 3 σ c.l. in the frequency region between 0.03 and 0.04 day−1 correspond to a monthly periodicity and are under study (they could be related to a tidal effect).

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 14 / 16

Annual Modulation

h365

• ffset

46.66 amplitude 0.7347 frequency 332.7 phase

• 0.4744

days 50 100 150 200 250 300 350 counts per counter (Hz) 44 45 46 47 48 49 50 51 52

h365

• ffset

46.66 amplitude 0.7347 frequency 332.7 phase

• 0.4744

Counting rate of 6 years averaged and fitted by sinusoidal function: k + A · sin(2π( t T + φ))

• btaining: T = 333 ± 32 d, maximum = 240 ± 32 d

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 15 / 16

Conclusions

We have analyzed the LVD Counting Rate measured at EL > 0.5MeV during 1997 – 2009. the sensitivity in terms of Rn activity has been measured for each counter. the analysis of the time delay in the radon component conferms that we are counting gammas from 214Bi. The spectral analysis shows evidence (> 7 σ c.l.) of annual modulation during 6 years, with:

1

T = 333 ± 32 d

2

Amplitude ≃ 1.5 − 2.5 Hz = ⇒ 5 − 8Bq/m3

3

Maximum at 28th August (± 32 days)

a candidate signal (3 σ c.l.) compatible with monthly periodicity has been found and could be related to a tidal effect

Gianmarco Bruno (INFN–LNGS) TAUP 2009, Rome July 3, 2009 16 / 16