in-situ measurement of heat-bumps at the chess · 2010-06-03 · nsls-ii xbpm mini workshop, 2009...

NSLS-II XBPM mini workshop, 2009

In-Situ Measurement of Heat-bumps at the CHESS A2 Beam-line

Peter Revesz, Alexander Kazimirov, Cornell University, CHESSIthaca, NY 14850

X-ray beam

Monochromator’s1st crystal

Beam footprint-- What is the heat bump?

-- What is the consequence of the heat bump?Distorted crystal – less flux

-- How to fight it?improved heat withdrawal: cryogenics, better substrate material


Thermalslope errors FWHM from (peak-to-valley) rocking curves at 20keV:FEA: (�) geometrical ray tracing: (�) Takagi–Taupintheory: (�) from and Experiment: (x) as a function of the total power.

Effect of high heat load on Si (111) performance

[5] J. Hoszowskaet.alNIM A 467–468 (2001) 631–634).


Mirror, mirror on the wall..

Distorted image…tells more about the mirrorthan about my face.

Reconstruct the change in surface relief of the mirror by analyzing the changes in the reflected image.

The Idea


The change in the angle of reflection ��where �r (μm) is shift of dot’sposition, and L (m) is the camera light source distance.

,Lr� ��

� ��

�

Measurement principle

�X�Y

The gradient map matrices for an array of dots:

Ly j i g

Lx j i gy x

��

��) , ( , ) , (

Slope vector field of a heat bump measured by using Centroidcalculations for each dot.[1] P.Reveszet.al.540,NIM-A 2-3, 470-9 (2005)

CCD


The problem:Given a vector field of gradients {gx, gy}Recover the surface Swhose gradient fieldbest approximates :

g�

g�

g S S�

� � min

Applying the divergence operator, we can transform this into a Poisson equation:

g S g S� ��

Measurementcentroids (gx, gy)

data saved

Matlab code to reconstruct surface,save calculated data

Surface reconstruction


The Moon: mare ridge south-west of Aristarchus

Surface reconstruction from shade. Shade is basically ~gradient of the reflected light (Lambertianreflection of light)


Shack-Hartmann wave front analyzer

…works using the same principle to measure aberration of the lens. It is compact device, so its accuracy is limited by the short image distance.

Lensletsare Fourier transform devices: X �


Experimental Setup

Camera on top of the monochromator

Inside the monochromatorViewportcrystal holderlight source

Light source

AstrovidCCD camera,55 mm telecentriclens

50 �m thick steel mask,4x4 in. area with holesø0.2mm, 1.5mm ain a rectangular grid withfluorescent light behind.


Centroidprogram screen images

Screen capture of the light dotsreflected form the crystal. A mm paper is placed on the crystal for scaling. The arrow indicates the beam direction

With this calibration we determinethe dot-spacing on the Si surface

Screen capture of the light dotsreflected form the crystal with the grid array used for CoGcalculations. Each CoGisshown as a small cross.

The gradient vector field measured with Centroidbefore X-ray exposure.[gx(i,j)= gy(i,j)=0 ]The surface is assumed to be flat.

d d’


Slope error calibration procedureto determine the correspondence between change in CoGand slope

Change in gradient field

Reconstructed surfaces [2]

051015202530

05

1015

200

0.02

0.04

0.06

0.08

X (mm)

si-15-calConstCalHV

Y (mm)

Height (m

icrons)

051015202530

05

1015

200

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

X (mm)

si-15-calConstCalHV

Y (mm)

Height (m

icrons)

No crystal tilt.50 ��Radian crystal tilt

Optical paths of the measurement

CCD

(un)tiltedcrystal

Light source

Virtual image

Instead of using absolute geometrical data

we use the gonoimeteras a calibrations tool

, /L r� �

��


Slit:1x6 mm 171 mA, 90

010

2030

40

0

10

20

30-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

X (mm)

si171mA10x6mm9deg.txtHV #H(mRads)V(mRads)HxV=40x26 grids20.39sec

Y (mm)

Height (m

icrons)

0102030405060-0.5

0

0.5

1

1.5

X (mm)

Height (m

icrons)

0102030405060-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

X (mm)

Slope error (m

Rad)

Typical heat bump measurement resultCHESS A2 wiggler,

Total power on Si ~90W (Bolometric measurement)


01020304050

0.0

0.5

1.0

1.5

0

1020

Height (m

icrons)

Y (mm)

X (mm)

2 mm

010

2030

4050

0.0

0.5

1.0

1.5

0

510

1520

25

Height (m

icrons)

Y (mm)

X (mm)

10 mm

010

2030

4050

0.0

0.5

1.0

1.5

0

510

1520

25

Heigh t (m

icr ons)

Y (mm)

X (mm)

4 mm

010

2030

4050

0.0

0.5

1.0

1.5

0

510

1520

25

Z Axis

Y (mm)

X (mm)

6 mm

010

2030

4050

0.0

0.5

1.0

1.5

0

510

1520

25

Heigh t (m

icr ons)

Y (mm)

X (mm)

8 mm

010

2030

4050

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

0

510

1520

25

Heigh t (m

icron s)

Y (mm)

X (mm)

3x6 mm

010

2030

4050

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

0

510

1520

25

Heigh t (m

icron s)

Y (mm)

X (mm)

1.8x6 mm

010

2030

4050

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

0

510

1520

25

Heigh t (m

icron s)

Y (mm)

X (mm)

2.2x6 mm

010

2030

4050

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

0

510

1520

25

Heigh t (m

icron s)

Y (mm)

X (mm)

1x6 mm

01020304050

-1.0-0.50.00.51.01.52.02.5

0

1020

Heigh t (m

icrons)

Y (mm) X (mm)

0.6x6 mm

Upstream slit opening

Slit width

Slit height

X-ray beam incidence is along X-axis, tilt: 9o


051015202530354045

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

Slope error at varoius slit heights

Slope error (mR

adian)

Y Distance across the beam directions (mm)

0.6 mm 0.8 mm 1.0 mm 1.2 mm 1.4 mm 1.6 mm 1.8 mm 2.0 mm 2.2 mm 2.4 mm 2.6 mm 2.8 mm 3 mm

slit height

051015202530354045

0.0

0.5

1.0

1.5

2.0

2.5 Heat bump height at various slit heights

Bum

p height (microns)

Distance across the beam directions (mm)

0.6 mm 0.8 mm 1 mm 1.2 mm 1.4 mm 1.6 mm 1.8 mm 2 mm 2.2 mm 2.4 mm 2.6 mm 2.8 mm 3 mm

slit height

Heat bump on Si as a function of vertical slit opening

010203040506070

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

Slope error at various slit heights

Slope error (m

Radians)

Distance aling the beam direction (mm)

0.6 mm 0.8 mm 1.0 mm 1.2 mm 1.4 mm 1.6 mm 1.8 mm 2.0 mm 2.2 mm 2.4 mm 2.6 mm 2.8 mm 3 mm

Along the beam

Across the beam

010203040506070

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

Heat bump heights at various slit heights

Bum

p heigt (microns)

X Distance along the beam direction (mm)

0.6 mm 0.8 mm 1 mm 1.2 mm 1.4 mm 1.6 mm 1.8 mm 2 mm 2.2 mm 2.4 mm 2.6 mm 2.8 mm 3 mm

Slit height

Y

X


051015202530

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

Slope error across Si at 9o 173 mA

Slope error (m

Radian)

distance across Si (mm)

2 mm 4 mm 6 mm 8 mm 10 mm

01020304050

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

Slope error along the beam, 9o, 173 mA

Slope error (m

Radian)

distance along beam (mm)

2 mm 4 mm 6 mm 8 mm 10 mm

Heat bump on Si as a function of horizontal slit opening

Y

X


3D simulation of heat bumps using ANSYS

MATLAB code to generate 3D powerdeposition in Si (heatbumpgui.m)

•ANSYS Workbench with engineering model of Si and Holder with cooling,

•Setting ANSYS parameters,interfaces, meshing,

•Steady State Thermal Simulation andStatic Structural Simulation,

•Output surface deformation mesh

MATLAB code to generate Cartesian deformation surface from the ANSYS surface mesh

Matlab work with the help fromE.Windischand C. MacGahan


Simulation resultsSi, copper crystal holder with water cooling ANSYS simulation of

deformation at A2 beam at 171 mA, slit 1.x2 mm

Comparison of raw experimental data of slope error and bump height along the beam direction with ANSYS simulations.

The slope error from ANSYS surface was obtained by numerical differentiation of the surface along the maximum of the heat bump.

The experimental bump was obtained by surface recovery procedure from the gradient field.

footprint

01020304050607080

-0.10

-0.05

0.00

0.05

0.10

Slope error, Measurement vs. ANSYS171 mA, Slit:1x6mm, 9

o

Slope error (mR

adian)

X (mm)

ANSYS Experiment

010203040506070800.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

Slope error measurement Slit: 1x6 mm, 9o

Bum

p height (microns)

X (mm)

ANSYS Integrated raw

experimental slopes

footprint


Heat bump on multilayers

Why multilayers?•high/variable band-width•choice of substrate materials

Multilayers used at CHESSBeam- line

SourceSagittal focusin g

Usaged-spacing/ /materials

�E/E,%

Experimental techniques / application areas

A2W, e-y50%15 Å – Mo/B4C20 Å – Mo/B4C28 Å – W/B4C

0.30.61.9

In-situ diffraction/scatteringRadiographyMicrobeam fluorescenceMicrobeam diffraction

D1BM, e+n100%15 Å - W/B4C25 Å – W/C30 Å – Mo/B4C

0.551.71.4

Small- and Wide-angle scatteringFluorescence imagingRadiography

F3BM, e+n50%27 Å – W/C24 Å – W/B4C22 Å – Mo/B4C

2.20.80.7

Macromolecular crystallographyRadiation damage in protein crystals

G1W, e+y100%24 Å – W/B4C28 Å – W/B4C

1.31.9

Small-angle scatteringMicrobeam fluorescenceMacromolecular crystallography

G3W, e+y100%28 Å – W/B4C1.9In-situ diffraction/scattering

[4] A. Kazimirov et.al.J. Synchrotron Rad. (2006). 13, 204–210


X-ray flux through the 0.25 mm x 0.25 mm slit in the hutch as a function of the horizontal width of the white beam slit. The vertical size of the slit was 1 mm.

Heat bump on Multilayers

PropertySiCu(GlidCop�)

CVD SiC

thermalconductivity�, [W/m�K]

150365300

thermal expansion�, [x10-6 K-1]

2.516.62.2

figure of merit�/�[x10-6W/m]

604136

Young’s modulu[Gpa]

107124450

meltingtemperature,oC

141210832500

As more X-ray power is absorbed in the crystal heat bump develops that leads to flux loss.



Heat bump on Multilayers (WB4C)

The effect of substrate material is quite evident.As a result of high heat conductivity and low thermal expansion coefficient of SiCthe heat bump is much reduced (by a factor of 3).



Conclusion and Summary

[1] P. Revesz and J.A. WhiteAn X-ray beam position monitor based on the photoluminescence of helium gas Nuclear Instruments & Methods in Physics Research, Section A540, n 2-3, 470-9 (2005)

[2] P. Revesz, A.Kazimirovand I.BazarovIn-situ Visualization of Thermal Distortions of Synchrotron Radiation OpticsNuclear Instruments and Methods in Physics Research A 576(2007) 422–429

[3] P. Revesz, A.Kazimirovand I.BazarovOptical Measurement of Thermal Deformation of Multilayer Optics under Synchrotron RadiationNuclear Instruments and Methods in Physics Research A 582(2007) 142–145

[4] Alexander Kazimirov, Detlef-M. Smilgies, Qun Shen, XianghuiXiao, QuanHao, Ernest Fontes, Donald H. Bilderback, Sol M. Gruner, YuriyPlatonovband Vladimir V. MartynovMultilayer X-ray optics at CHESS,J. Synchrotron Rad. (2006). 13, 204–210

[5] J. Hoszowska, V. Mocella, L. Zhang, J.-S. Migliore, A.K. Freund, C. FerreroPerformance of synchrotron X-ray monochromators under heat load Part 3: Comparison between theory and experiment, Nuclear Instruments and Methods in Physics Research A 467–468 (2001) 631–634)

References

•A 3D In-Situ measurement of heat bumps on crystals and multilayers at high heatload has been developed.

•The experimental results show good agreement with ANSYS calculations.•Results show that for multilayers SiCsubstrate material is much favorable in comparison

with Si substrates.

Acknowledgement: Thanks due to Jim Savino for ANSYS simulations

in-situ measurement of heat-bumps at the chess · 2010-06-03 · nsls-ii xbpm mini workshop, 2009...

Documents