Update on R&D

Sumanta Pal

Center for Neutrino Physics, Virginia Tech.

AAP2016, 1 Dec 2016

CHANDLER

(Carbon Hydrogen Anti-Neutrino Detector with a Lithium Enhanced ROL)

The CHANDLER detector concept

S. Pal, Virginia Tech

2

e⁺

p

n

Detected Light

Positron (e⁺)

Neutron

Capture

⁶Li

Time

Geant4 Simulation of light transport by total-internal-reflection

Photon Ray Tracing in GEANT4

Cubes: WLS plastic scintillator

Sheets: Li6 loaded ZnS for thermal neutron detection

⁶Li sheet

Research and Development Effort

S. Pal, Virginia Tech

3

MiniCHANDLER is a fully funded system test (8×8×5) which is now being commissioned and will be deployed at North Anna Nuclear Power Plant, Virginia.

Cube String Studies have been used to study light production, light collection, light attenuation, energy resolution and wavelength shifter concentration.

MicroCHANDLER is a 3×3×3 prototype which we are using to test our full electronics chain, develop the data acquisition system, study neutron capture identification and measure background rates.

S. Pal, Virginia Tech

4

MicroCHANDLER R&D

Light tight mechanical set up of the MicroCHANDLER

Light guide used with HAMAMATSU PMTs

• Old : 2 inch PMT (xp2202)
• New : Hamamatsu (R6231-100), 2 inch diameter, high Q.E. and linearity over large dynamic range.
• Readout resolution of two types of PMTs are being tested in this set up.
• Easy access to PMTs. No possibility of over heating the PMT bases.
• No hassle to put the radioactive source on top of the box.
• No light leakage observed.

Fully functioning MicroCHANDLER

S. Pal, Virginia Tech

5

MicroCHANDLER : Compton edge study

• ²²Na gamma source : 0.511 MeV and 1.275 MeV gamma lines.
• Collimated source was placed approximately at the center of each cube on the top layer.

S. Pal, Virginia Tech

6

Mechanical design of MiniCHANDLER

Goal of the MicroCHANDLER mechanical set up was to test the light tightness of the box.

​

After successful operation of the MicroCHANDLER, same mechanical set up has been prepared for the MiniCHANDLER.

S. Pal, Virginia Tech

7

Mechanical design of MiniCHANDLER

Use of O-ring, ring spacer and PMT O-ring clamp plate is to make the box light tight.

​

Easy access to any PMT channels as before.

​

S. Pal, Virginia Tech

8

Commissioning of MiniCHANDLER

Assembling WLS scintillator cubes

Assembling PMTs

​

Li sheets (white) are also visible between layers.

Goal: scale up the technology of the MicroCHANDLER towards CHANDLER.

Electronics & DAQ set up

S. Pal, Virginia Tech

9

Anode

PreAmp

Gain x 2

Shaper (25ns)

Digitizer (16ns)

Data

Storage

VME shaper module

(16 channels)

DAQ system

Readout Electronics

• 2 inch PMTs.
• MicroCHANDLER : xp2202 PMTs (9) + R6231-100 PMTs (9)
• MiniCHANDLER : xp2202 PMTs (80)
• CHANDLER : R6231-100
• VME 16 channel shaper (charge integrating):
• Analog Preamp and Shaper circuit (VT + CREMAT)
• VME digitizer : 64 channels, 16 ns, 12 bit (CAEN V1740)

Readout Software

• VT customized DAQ readout.
• Each digitizer connected by dedicated single optical link to the computer (80 MB/s).
• Multi-Board readout tested on MicroCHANDLER and successfully working.
• Trigger scheme : trigger on a plane.
• Apply zero suppression before writing to the disk.
• Data streams are combined and sorted to identify events via event builder process running on raw data.

Scalable set up : successfully tested in MicroCHANDLER. MiniCHANDLER will use same set up.

External Trigger

S. Pal, Virginia Tech

10

Electronics & DAQ set up

Dual

Timer

Shaper

Board

Digitizer

Linear FIFO

CAEN SY527 HV Mainframe

&

CAEN A734N HV cards

MicroCHANDLER DAQ performance

S. Pal, Virginia Tech

11

MicroCHANDLER : preliminary calibration scheme

S. Pal, Virginia Tech

12

Use muons, simple and they are present in the detector, avoiding complicated calibration scheme (not so effective in past experiments while considering the complication of the calibration system and of the risk of the deployment.

μ

• Each PMT has an individual HV channel.
• Better for gain calibration and optimization.

Landau

+

Polynomial

Landau

+

Exponential

Amplitude of pulse

Neutron and positron Pulse shape

S. Pal, Virginia Tech

13

Neutron PSD : area over amplitude

S. Pal, Virginia Tech

14

A clear indication of neutron signal in the detector was presented before.

Neutron like events (left plot) in absence of neutron source are most likely from cosmic background. Can we identify that ?

⁶Li sheet

Cosmic Source

Neutron Source

Front view of the detector

Improved neutron selection

S. Pal, Virginia Tech

15

• Neutron like events : Lower amplitude and larger area under pulse relative to positron like events.
• The different classes of events are clearly separated into distinct bands.
• In the overlap region positron like events clearly out numbers neutron candidates.
• Is there any way to recover those neutrons ?

Resolving neutron position to a specific sheet

• In CHANDLER we can see lights from both sides of the sheets.
• So, we can use that light to determine which sheet tag the neutron.
• When a good neutron is tagged in the middle layer (above both red lines), we looked at the signals from the top and bottom layers.

S. Pal, Virginia Tech

16

• The sheet is determined from the layer of larger amplitude.

The Other Layer Distributions

• We see a clear neutron band extending into the overlap region.
• So, neutrons in the overlapping regions can be recovered by looking into the other sides.

S. Pal, Virginia Tech

17

When a good neutron is tagged in the middle layer (above both red lines), we looked at the signals from the top and bottom layers.

What is the background to the neutron signal?

S. Pal, Virginia Tech

18

• If we see neutron in the middle layer, it should come from middle-bottom ⁶Li sheet.
• We collected large number events over the period of 4012 min (66.87 hr = 2.78 days).
• Number of triggers obtained: 62M

?

• Are neutron like events coming really from cosmogenic neutrons ?
• To investigate that we use a combination of Li and Li free scintillating sheets to obtain clear neutron PSD.

Front view of the detector

Neutron misidentification

S. Pal, Virginia Tech

19

• When neutron candidates are first identified in the middle layer, we see very few neutron like events in the top layer coming from the Li free sheet.
• Cosmogenic neutrons out number fake neutrons by more than 50 to 1 when we compare events between middle-top and middle-bottom layers.
• So, the neutron tag is extraordinarily pure in this detector.

1.6% of events w.r.t. the middle layer

89.3% of events w.r.t. the middle layer

Neutron misidentification

S. Pal, Virginia Tech

20

• 28 events are pedestal fluctuation where sheet is assigned incorrectly.
• Less than 1% of time we choose wrong sheet in this method.

28 events

2.4% of events w.r.t. the middle layer

97.6% of events w.r.t. the middle layer

Future plan : Deployment at the reactor site

S. Pal, Virginia Tech

21

North Anna Nuclear Power Plant, Virginia

​

Around 4 hours drive from the Virginia Tech.

MiniCHANDLER will be placed inside this trailer at the power plant.

Summary

• Chandler R&D is on track:
• Read out and DAQ fully designed and tested.
• Demonstration of a highly pure neutron tag.
• Demonstration of significant gain in energy resolution with light guides and high Q.E phototubes.
• MiniCHANDLER construction is complete and commissioning is ongoing.
• Fully funded plan for the deployment at the North Anna Nuclear Power plant.
• Expected to see IBD events by AAP2017.

S. Pal, Virginia Tech

22

S. Pal, Virginia Tech

23

Backup

S. Pal, Virginia Tech

24

MicroCHANDLER : DAQ : offline Event builder

1. Identify trigger pattern : external or PMTs.
2. Linux time stamp has been written into the data stream when external trigger fires. External trigger freq. 1 Hz.
3. Identify external trigger into the data,
4. Match Linux time stamp between boards,
5. Estimate time offset between boards ( slave – master ). [All timings are global time in each board]
6. Correct time in the slave board.
7. Event merging :
1. Identify coincidence data between boards (time window 40 ns): merge the data.
2. If no coincidence found, save data as separate events.

Linux Time Stamp