Dynamic Neutron Imaging

Transcription

1 Wir schaffen Wissen heute für morgen Pierre Boillat Neutron Imaging & Activation Group (NIAG) and Electrochemistry Laboratory (LEC), Paul Scherrer Institute, Switzerland Dynamic Neutron Imaging PSI, 8. Oktober 2015

2 OUTLINE 1. Introduction 2. Basics of dynamic neutron imaging a. Typical experiment setup b. Signal to Noise Ratio c. Spatial/temporal resolution tradeoff d. Detectors e. Image processing specificities 3. Examples from fuel cell research 4. Conclusion

3 Introduction

4 Why dynamic imaging?

5 Why dynamic imaging? - Make observations which are not possible in steady state

6 Why dynamic imaging? - Make observations which are not possible in steady state - Correlate processes using their time scale

7 Does «dynamic» mean «fast»? 10-2 s 10 4 s Neutron s s Imaging

8 Typical experiment setup

9 Classical «non dynamic» application 3D reconstruction

10 Dynamic radiography 3D reconstruction

11 Dynamic radiography 3D reconstruction

12 Dynamic radiography Typical time resolution: 1 60 s Down to a few ms in specific cases

13 Signal / Noise Ratio (SNR)

14 Classical setup: scintillator + camera 3) Detector noise 2) Photon shot noise Neutrons Light 1) Neutron shot noise

15 Shot noise Average: 2 n/s 3 1

16 Neutron statistics Intensity [-] For a pixel flux of 1000 n/pix/s: Probability (normalized) [-] SNR = Position [pix] (e.g n/cm 2 /s, 100µm pixels) Contrast: 5% Exposure time 10 ms 100 ms 1 s 10 s Intensity [-]

17 Neutron statistics Intensity [-] For a pixel flux of 1000 n/pix/s: Probability (normalized) [-] SNR = Position [pix] (e.g n/cm 2 /s, 100µm pixels) Contrast: 80% Exposure time 10 ms 100 ms 1 s 10 s Intensity [-]

18 Spatial / temporal resolution tradeoff

19 Scaling laws Effect of pixel size, for a given neutron flux: px Neutrons px N α 1/px 2 Φ = constant Neutrons px px

20 Scaling laws Effect scintillator resolution, for a given neutron flux: e Neutrons Φ = constant e effα1/e Neutrons

21 Scaling laws For a constant sample-detector distance r g Neutrons d Φα1/r g 2 r g Neutrons

22 Scaling laws Combination of all: Neutrons Nα1/r 5!!!!! Neutrons

23 Tradeoff mitigation: reduction of distance d r g Neutrons d Φα1/d 2 r g Neutrons

24 Tradeoff mitigation: anisotropic resolution µm 1 mm Source: µm

25 Tradeoff mitigation: anisotropic resolution Φ ~ π/4 D 2 Φ ~ D x D y

26 Tradeoff mitigation: anisotropic resolution Detector Beam d obj = d det Radiogram Cell Beam d obj Detector, tilted Radiogram Cell d det P. Boillat, D. Kramer, B.C. Seyfang, G. Frei, E. Lehmann, G.G. Scherer, A. Wokaun, Y. Ichikawa, Y. Tasaki, K. Shinohara, Electrochem. Commun. 10, 546 (2008)

27 Dynamic imaging Effective resolution 20 µm Exposure time 10 s Readout time 4 s P. Boillat, PhD Thesis No 18397, ETH Zürich. (2009)

28 Tradeoff mitigation: cyclic processes Stochastic process Cyclic process

29 Tradeoff mitigation: cyclic processes 1) Stroboscopic imaging

30 Tradeoff mitigation: cyclic processes 2) Cumulative imaging

31 Detectors

32 CCD vs CMOS CCD Exposure Readout Exposure Readout CMOS Exposure Exposure Exposure Time

33 CCD vs. CMOS Assuming a total readout time of 2 seconds for the CCD: 100% Fraction used for exposure [%] 80% 60% 40% 20% CCD CMOS 0% Total acquisition time [s]

34 Photon detection efficiency Assuming a quantum efficiency of 95% for the CCD and 60% for the CMOS 100% Total detection efficiency [%] 80% 60% 40% 20% ~5 seconds CCD CMOS 0% Total acquisition time [s]

35 Image processing: specificities of dynamic imaging

36 Referencing and registration

37 Referencing and registration

38 Referencing and registration Without registration Registration, Without FF correction Registration with FF correction

39 Scattered neutrons 1) By the sample («sample scattering) 2) By the setup («background scattering») Neutrons Sample Setup

40 Correction of scattered background Black bodies body Neutrons Setup

41 Correction of scattered background Sample or background scattering? Black bodies Neutrons Setup

42 Golden ratio tomography 4D imaging

43 Golden ratio tomography 4D imaging

44 4D imaging Golden ratio tomography: each subset is well distributed

45 Examples from fuel cell research

46 Water management what s the problem?

47

48 Water management what s the problem?

49 Examples from fuel cell research 1) Impact of water accumulation on performance

50 Water in GDL Materials 0% 5% 20% Case 1 Case 2 Case 3 Cathode GDL No PTFE Optimal PTFE content Too high PTFE content

51 Water in GDL Results Full humidification 1 A/cm 2 2 bar abs /2bar abs Voltage loss / mv 5%

52 Mass transport losses 0% 20% 5% Voltage loss / mv

53 Water distribution 0% 5% 20%

54 Water distribution 0% 20% 5%

55 Examples from fuel cell research 2) Startup at sub-freezing temperatures

56 Voltage Sub-zero startup Motivation + + O 2 + H 2 + H 2 O Time

57 Voltage Sub-zero startup Motivation + + O 2 + H 2 + H 2 O Current Time

58 Voltage Sub-zero startup Motivation + + O 2 + H 2 + H 2 O Current Time

59 Voltage Sub-zero startup Motivation + + O 2 + H 2 + H 2 O Current Time

60 Sub-zero startup Motivation Voltage + t work + O 2 + H 2 + H 2 O Current Time

61 Sub-zero startup Motivation Voltage T > 0 o C + + O 2 + H 2 + H 2 O Current Time

62 T = -10 o C, i = 0.2 A/cm 2 Sub-zero startup Results

63 Sub-zero startup Results T = -10 o C, i = 0.2 A/cm 2 U [V] i [A/cm2] Water content [% vol] Anode Cathode t [min] t [min] Membrane + catalyst Cathode GDL

64 Sub-zero startup Results T = -10 o C, i = 0.2 A/cm 2 U [V] i [A/cm2] Water content [% vol] Anode Cathode Voltage Current density t [min] t [min] Membrane + catalyst Cathode GDL

65 Sub-zero startup Results T = -10 o C, i = 0.2 A/cm 2 U [V] i [A/cm2] Water content [% vol] Anode Cathode Voltage Current density t [min] t [min] Membrane + catalyst Cathode GDL

66 Sub-zero startup Results T = -10 o C, i = 0.2 A/cm 2 U [V] i [A/cm2] Water content [% vol] Anode Cathode Voltage Current density t [min] t [min] Membrane + catalyst Cathode GDL

67 Sub-zero startup Interpretation How can liquid water be in there? Super-cooled water? Source: watch?v=dpiuzi_3o8s

68 Sub-zero startup Interpretation Repeated experiments without neutron imaging Mechanical shock on the cell? Yes. U [V], i[a/cm 2 ] RH = 30%, T = -10 o C, i = 0.1 A/cm U [V] i [A/cm2] Time [min]

69 Sub-zero startup Interpretation Repeated experiments without neutron imaging Mechanical shock on the cell? Yes.

70 Sub-zero startup Interpretation T = -10 o C, i = 0.2 A/cm 2 U [V] i [A/cm2] Water content [% vol] Anode Cathode Voltage Current density t [min] t [min] Membrane + catalyst Cathode GDL

71 Conclusions Find good tradeoff between spatial and temporal resolution Some methods (anisotropic, stroposcopic) can help you improve the tradeoff Prepare well you experiment! In particular: sample-detector distance