Grain analysis with High-Resolution microscope

1
Grain analysis with High-Resolution microscope
Bologna group (G.Sirri)

• An High Resolution microscope was developed in
  parallel with the ESS and is working in BOLOGNA
  (MIC1 PROTO).
• The microscope is used for manual checks, fog
  measurements and (recently) scanning with sysal.
  

• The microscope is equipped with
• ZEISS 100x objective n.a. 1.3
• ZEISS tube lens hosted in a custom tube
• Stage with higher load capacity MICOS HPS-170
• Legs with air suspension mechanisms to reduce
  vibrations.
• Custom CMOS camera (same sensor as mikrotron
  mc1310)
• Illumination, horizontal stage, motor controller,
  frame grabber, workstation are the same as ESS

2
Measurements of the optical resolution star test
Star test diagnostic tool which consists in the
observation of the image of a point source of
light (also known as Point Spread Function (PSF)
of the objective) from the specimen projected by
the microscope objective onto the camera sensor
plane.
In a perfect aberration-free optical system the
image will be a very close approximation to the
diffraction pattern rings
Aberrations enlarge the Airy disk and produce a
spot with variable form and photometric profile
(circle of least confusion). This enlargement
reduces the resolving power both in the lateral
and in the axial directions.
3
Nikon 50x
custom setup camera 50 mm below the specific
coniugate
ZEISS 100x
4
3
Z (micron)
2
1
Circle of least confusion
Circle of least confusion
0
?Z 5 micron
-1
?Z 1 micron
?X, ?Y 1 micron
-2
-3
-4
?X, ?Y 0.2 micron
-6
NEGATIVE IMAGES
-8
4
Image formation
The grain image is the space convolution between
the grain surface and the P.S.F. .

• If our object size (grain Ø 0.6-0.8 micron) is
  smaller than the circle of least confusion
• the geometrical image of the grain is smaller
  than the circle of least confusion and
  consequeltely its size is not proportional to the
  object (impossible to observe little differences
  in size)
• as smaller the object, as brighter the image
• Since the horizontal dimension of the circle of
  least confusion is 0.2 microns, this microscope
  is able to measure the grain size.

5
Acquisition, analysis and credits

• We used libACQ, a test library developed by I.
  Kreslo (Bern) and installed with the help of V.
  Tioukov (Naples).
• Each image grabbing is triggered by stage
  controller
• Output in libEdb format (FEDRA)
• Mean grey level of the clusters is also stored
• 100 levels per emulsion layer
• Image size 0.5 Megapixel
• Magnification 0.118 micron/pixel
• Several views in the whole emulsion sheet
• Analysis with an improved version of the Fedra
  Grain Analysis library (by V. Tioukov).
• limit for the clusters area gt 5
• Vertical length of a grain gt 3 layers
• Two samples
• Nov. 2004 (vertex location CERN test beam)
• June 2006 (cosmic rays exposure at LNGS)
• Grain Size cluster area at center grain

6
Stability of the algorythm (Vertex Loc.
nov04,br7,pl8)
sublayer near the surface
sublayer near the base
Red top Blue bottom
7
Mean grain area profile
100X
ESS (expansion filter)
ESS
? Nov 2004 ? June 2006
8
Cluster area profile (cut areagt2)
Red top Blue bottom
(Vertex Loc. nov04, br7, pl8)
(cosmic rays, jun06, pl33)
9
Grey level horizontal profile
(cosmic rays, jun06, pl33)
FWHM 5 pixel 0.6 micron
10
Grey level vertical profile (negative image)
(cosmic rays, jun06, pl33)
(Vertex Loc. nov04, br7, pl8)
Red top ? ? sublayer near the surface Blue
bottom ? sublayer near the base
11
Grain Size vertical profile
Red top Blue bottom
(Vertex Loc. nov04, br7, pl8)
Ø 0.72 micron
Ø 0.62 micron
(cosmic rays, jun06, pl33)
Ø 0.69 micron
Ø 0.59micron
12
Conclusions

• The grains of two sample were measured with a
  high resolution microscope equipped with 100x
  Zeiss objective.
• Nov. 2004 (vertex location test beam exposure at
  CERN)
• June 2006 (cosmic rays exposure at LNGS)
• Grain size maybe -5 smaller, but we shuold
  better estimate the uncertainty related to
  threshold set-up and illumination uniformity )
• At present the only evidence is for different
  thickness of the emulsion layers (35-40 micron
  for nov. 2004 and 55-60 for june 2006). Further
  investigation are needed.

13
Further test with sysal
MIC3 (sysal) Thr 400 500 600
700 800 900 GS (2004) 6.2 5.7 5.3
5.0 4.7 4.3 GS (2006) 5.8 5.4 5.0
4.7 4.4 4.1 GS (2004) 11.3 10.3
9.7 9.1 8.5 (expansion filter) GS (2006)
4.9
(27 development)