Institute for Experimental Physics E21 · Institute for Experimental Physics E21 Annual Report 2002...

Institute for Experimental Physics E21 Annual Report 2002 1

Institute forExperimental Physics E21

3 cm


Annual Report 2002of the Institute for Experimental Physics E21published: 12.3.2003Edited by Florian Grünauer, E21http://www.physik.tu-muenchen.de/lehrstuehle/E21

Physik-Department E21Technische Universität MünchenJames-Franck-StrasseD-85747 Garching, Germany

Phone: +49 89-289-14711Fax: +49 89-289-14713

Copyright:Inquiries about copyright and reproduction etc.should be addressed to the authors.

Annual Report 2002of the Institute for Experimental Physics E21

Prof. Dr. P. BöniTechnische Universität München


INDEX

Preface...................................................................................................................................................4

1 Neutron Scattering and Magnetism .......................................................................................6

1.1 Focusing parabolic guide for very small samples.................................................................71.2 Horizontally focusing Heusler analyzer for TASP without depolarising zero field region..........91.3 Observation of soliton chirality in the Ising-chain system CsCoBr3......................................111.4 Phonon density-of-states in intermediate valence Ce2Ni3Si5...............................................131.5 Crystal field effects on Er3+ and Nd3+ ions in intermediate valence Ce2Ni3Si5.......................151.6 Search for the crystal field splitting in TmB12 .....................................................................171.7 Mapping of magnetic excitations in single-Q chromium......................................................191.8 Specific heat study of the spin-chain compounds Ca2+xY2-xCu5O10......................................211.9 Frustrated phase separation in the overdoped regime of ‘123’ high-Tc superconductors ......231.10 X-ray and magnetization studies on FeCoV/Ti multilayers..................................................251.11 Preparation and Characterization of FeCoV/Ti Multilayers: a Comparison...........................271.12 Critical spin dynamics in CsMnBr3....................................................................................291.13 Polarized SANS studies from the weak itinerant helimagnet MnSi......................................311.14 Design of an inelastic neutron spectrometer for catalytic applications at FRM-II ..................321.15 MIRA -- Very cold neutrons for new methods ....................................................................341.16 Testing the shielding of the monochromator of Mira by Monte Carlo Method .......................371.17 RESEDA Spectrometer: Jahresbericht 1.1.-31.12.02.........................................................391.18 MUPAD Spectrometer: Annual Report 1.1.-31.12.02 .........................................................421.19 Multiple Small Angle Neutron Scattering...........................................................................44

2 Neutron Radiography and Tomography ..............................................................................46

2.1 The design and construction of the neutron tomography facility ANTARES .........................472.2 Steps toward a dynamic neutron radiography of a combustion engine ................................512.3 Phase contrast radiography using thermal neutrons ..........................................................532.4 Neutron Topography at TOPSI.........................................................................................552.5 Influence of gaps in shieldings against neutron radiation....................................................572.6 Estimation of the imaging quality of Munich´s new neutron radiography station ...................59

3 Sample environments..........................................................................................................61

3.1 Minimization of the background radiation of a sample environment ....................................62

4 Positron Physics.................................................................................................................64

4.1 Experimental facilities at the intense positron source at FRM II ..........................................654.2 Investigation of the Annealed Copper Surface by Positron Annihilation Induced Auger

Electron Spectroscopy ....................................................................................................674.3 High Resolution Gamma Spectrometer for Doppler Broadening Measurements

of the 511 keV Positron Annihilation Line..........................................................................704.4 Investigation of Laser Treated AlN and Silica by Positron Lifetime Measurements...............72

5 Reactor Physics...................................................................................................................75

5.1 Nuclear Licensing of the FRM-II.......................................................................................765.2 Fuel Element with Reduced Enrichment for the FRM-II......................................................77

6 Activities 2002.....................................................................................................................78

6.1 Publications ....................................................................................................................796.2 Conference, Workshop and Seminar Contributions ...........................................................826.3 Lectures, Courses and Seminars .....................................................................................856.4 Committee Memberships.................................................................................................866.5 E21 Members .................................................................................................................876.6 Associated Members at FRM-II .......................................................................................886.7 Guests ...........................................................................................................................88


PREFACE

After the retirement of Prof. Gläser onSeptember 30, 2001, the joining of the differentsubgroups of E21 proceeded in a rathersmooth way and we managed to create somesort of “corporate identity”. This endeavour wassuccessful despite various difficult boundaryconditions. One of them was caused by aserious illness of our secretary since August2002, one of the key persons in E21. Thanksto the generous help of B. Russ and thesupport of the faculty administration wemanaged to overcome this difficult situation. Inaddition, we had difficulties with budgeting dueto problems we experienced with SAP (a newaccounting system at TUM). Last but not least,our scientific work was seriously hampered bythe ongoing delay by the Ministry ofEnvironmental Protection (BMU) to provide thefinal permission to run the FRM-II. We are verygrateful that other facilities, mostly in Franceand Switzerland, helped us to compensate forthe loss of beam time.

Let us now come to some of the highlights ofE21 in 2002. The installation of the beam linesof E21 at the FRM-II progressed very well. Thedesign of the shielding for theradiography/tomography (ANTARES) beamline was finalized and the demanding task ofdesigning the shielding blocs in the form of a“three dimensional puzzle” came to an end.The blocs are ordered and will be installedearly in spring 2003. The primary componentsof the beam line MIRA for very cold neutronsare installed. Here we mention in particular themonochromator shielding that was installedwithin 4 months after the order was placed.The neutron spin-echo spectrometer RESEDAis still in a very advanced state and waiting forneutrons… E21 was strongly engaged indeveloping new techniques and methods forneutron scattering. Due to the lack of neutronsthis work was mostly done at the ILL inGrenoble, France, and at PSI in Switzerland. Incollaboration with F. Demmel and R. Gählerwe investigated at the ILL methods to improvethe coil geometry for neutron spin echo (ZETA)and for mapping of inelastic scattering inreciprocal space in itinerant magnets (IN3),respectively. At PSI we have investigated anew focusing technique using parabolicmirrors. In addition, in collaboration with PSI anew type of mechanics for focusing withHeusler monochromator crystals wasdeveloped.

The laboratories of E21 in the PhysicsDepartment have improved rather dramaticallyafter we succeeded to obtain most of thepromised laboratory space after fighting for

almost two years! We acquired two Bruker X-ray diffractometers, namely a D5000 forreflectivity measurements from multilayers anda D500 for powder diffraction. In addition weupgraded the Physical Property MeasurementSystem PPMS with an option to measurethermoelectric properties of samples. TheD500 is presently in its commissioning phaseand will be equipped soon with a closed cyclerefrigerator to attain temperatures as low as10 K.

Of course, all the installations are being usedfor doing science. In this respect, the year2002 was very successful. Most neutronscattering measurements of E21 wereperformed at SINQ at PSI due to the lack ofneutrons in Garching. One of the highlightswas the first observation of single-handed,chiral fluctuations in very low magnetic fields inthe helical ferromagnet MnSi using polarizedneutrons. The results were published on thecover page of Physical Review Letters. Inaddition, we investigated the chirality of field-induced solitons in low-dimensional spinsystems were we observed rather dramaticeffects. In a E21-ILL collaboration we managedto improve the quality of time-resolvedradiography by measuring the movement ofthe pistons in a car motor with very highresolution. These results may lead to detailedinvestigations of the combustion process inmotors. In the field of artificial magneticstructures we investigated the morphology andmagnetic properties of multilayers that werefabricated at facilities at PSI and TUM. Thegoal is the engineering of the hystereticbehaviour of multilayers.

The research group around the technicaldirector of the FRM-II, Prof. K. Schreckenbachthat belongs scientifically to E21, continuestheir investigation in fundamental physics withthe set-up TRINE searching for a possibleviolation of the time reversal symmetry in thesystem of the free neutron decay. The till bestlimit for the D-coefficient was obtained, whichis an important information for particle physicsbeyond the standard model. Furthermore, theresearch with positrons in collaboration withthe Universität der Bundeswehr in Munich waspursued. The novel type of a slow positronsource was successfully tested at a coldneutron beam at the ILL, Grenoble, and will beplaced at the foreseen FRM-II in-pile positionbefore the nuclear start-up. An experiment onAuger-spectroscopy induced by positronannihilation gave first signals with a laboratorypositron source. This set-up will be one of the


instruments at the FRM-II intense source ofslow positrons.

A major field of work in 2002, in particular forProfs. K. Schreckenbach and K. Böning, wasto take the – hopefully – final actions in thenuclear licensing procedure of the FRM-II. InJanuary 2002 the BMU issued its FederalSupervisory Assessment containing some 65questions or conditions to the BavarianLicensing Authority, the Ministry ofEnvironment StMLU. The groups at the TUMand its general contractor Framatome-ANPhave carefully discussed all these topics untilJuly 2002. At the end of October 2002 the

BMU sent an updated list of questions, whichstill contained some 10 topics with a focus on“beyond design basis accident scenarios”.After all the documents resulting from thathave been submitted by about the end of 2002,we are pretty confident now to receive thegreen light to start up and operate the FRM-IIwithin a few more months only. – The planningof a new fuel element of the FRM-II withreduced enrichment represented another fieldof activity that, however, can actually start onlyonce the FRM-II has received its licence tooperate.

Garching, in February 2003

Peter Böni Klaus Böning Klaus Schreckenbach


1.1 Focusing parabolic guide for very small samples

P. Böni1 and J. Stahn2

1Technical University of Munich, Physics Department E21, D-85747 Garching, Germany2Laboratory for Neutron Scattering, ETHZ & PSI, CH-5232 Villigen PSI.

We show that neutrons can be very efficiently focused by using parabolic neutron guides leading togains of more then 50 on areas as small as 1 mm2. The advantages are i) ease of implementation, ii)improved signal to noise ratio, and iii) largest gain at focal point. (Boni_02_Focusing.doc)

The interest in modern materials likemultilayers, superconductors, GMR and CMRmaterials as well as biological samples hasincreased significantly. These samples canoften only be grown in small quantities andneutron scattering experiments are difficult toperform. Therefore, it is necessary to focusneutron beams to as small areas as possible toincrease the flux at the sample position. Ofcourse, due to the theorem of Liouville thedivergence of the beam is increased tooleading to a coarser q-resolution.

For PGAA is has become common practice touse a Kumakhov lens [1] that consists ofthousands of glass capillaries focusing thebeam to a spot with a diameter as small as 0.7mm. Until recently, such a lens was installed atSINQ providing an intensity gain of ≅16. It wasalso used for CEF-measurements on the triple-axis spectrometer TASP [2]. The majordisadvantage is that due to the identical areaat the entrance and the exit of the capillariesthe whole exit area of an end position of a

neutron guide is wasted leading also to anincreased background.

In order to overcome these problems we haveinvestigated the possibility of using a smallguide with an entrance area of 16 mm × 16mm that focuses the beam on an area < 1mm2. The length of the whole device is 455mm. The interior surfaces (coated withsupermirror m = 3) were parabolically shapedto focuse the parallel beam onto a single spotat the focal point. The performance wassimulated by means of the program packageMcStas from Risø National Laboratory.

We show in Fig. 1 the beam profiles at the exitand at the focal point of the device. For theincident beam we assumed a flat spectrum 0.5Å ≤ λ ≤ 11.5 Å and a rather large divergence of0.760 corresponding to the situation at beamline SXD at ISIS. It is seen that gains of afactor of 50 can easily be realized on a spot ofless than 1 mm2. This is a rather exciting resultwhen compared with a Kumakhov lens.

Fig. 1: Intensity distribution of the neutron beam at the exit and the focal point of a 455 mm longparabolic neutron guide. The gain (G = 51) at the focal point is higher than at the exit of the guide (G =45).

A prototype of the device was built andcharacterized on TOPSI at SINQ usingneutrons with λ = 4.88 Å. In a simple setup wemounted the device on a translation table andrecorded the neutrons that were focused on an

aperture with a quadratic cross section 2 mm ×2 mm that was positioned before a 3Hedetector. The distance between exit andaperture was varied between 7 mm ≤ L ≤ 80mm. Fig. 2 shows the measured intensity when

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Gain: 45

Exit

0

7.5E-4

0.001500

0.002250

0.003000

0.003750

0.004500

0.005250

0.006000

width (mm)

heig

ht (

mm

)

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Gain: 51

Focus

0

8.75E-4

0.001750

0.002625

0.003500

0.004375

0.005250

0.006125

0.007000

width (mm)

heig

ht (m

m)


the focusing unit is translated through theneutron beam. The maximum intensity isobserved at a translation uty = 12 mm. The twominima in intensity occur when the glass wallsintersect the beam. The shoulder around uty =25 mm corresponds to the intensity of the non-focused direct beam. The measured gain G ≅ 6is reduced when compared with the calculationbecause the measured intensity correspondsto the total number of neutrons that istransmitted through the aperture of 4 mm2 thatis much larger than the spot size at the focal

point of ≅ 0.49 mm2. Hence, the result iscompatible with the McStas simulation. Wepoint out that the intensity does not change for7 mm < L < 40 mm indicating that the point ofthe maximum flux is indeed at the focal pointand not at the exit of the device.

Because the input area of the focusing deviceis so small one may consider filling up the fullcross section of the beam tube with parabolicfocusing units. This way, gains of up to 500may be realized for PGAA applications.

0 10 20 300

2000

4000

6000

8000

10000

0 10 20 300

2000

4000

6000

8000

10000

0 10 20 300

2000

4000

6000

8000

10000

0 10 20 300

2000

4000

6000

8000

10000

0 10 20 300

2000

4000

6000

8000

10000

0 10 20 300

2000

4000

6000

8000

10000

0 10 20 300

2000

4000

6000

8000

10000

0 10 20 300

2000

4000

6000

8000

10000

TOPSI851

1/15/2003File Trumpet_TOPSI_uty.opj

Focusing trumpet: λ = 4.88 Å / 2.44 Ådistanceexit - aperture:

Inte

nsity

(co

unts

/200

0 m

on)

uty (mm)

80 mm

TOPSI849

Inte

nsity

(co

unts

/200

0 m

on)

uty (mm)

50 mm

TOPSI847

Inte

nsity

(co

unts

/200

0 m

on)

uty (mm)

60 mm

TOPSI845

Inte

nsity

(co

unts

/200

0 m

on)

uty (mm)

40 mm

TOPSI838

Inte

nsity

(co

unts

/200

0 m

on)

uty (mm)

20 mm

TOPSI837

Inte

nsity

(co

unts

/200

0 m

on)

uty (mm)

7 mm

TOPSI853

Inte

nsity

(co

unts

/200

0 m

on)

uty (mm)

7 mm

TOPSI843

Inte

nsity

(co

unts

/200

0 m

on)

uty (mm)

30 mm

Fig. 2: Intensity of the neutron beam versus position of the beam tube.

References[1] M. A. Kumakhov and V. A. Sharov, Nature 357, 390

(1992).[2] P. Böni, Physica B 276-278, 6 (2000).


1.2 Horizontally focusing Heusler analyzer for TASP without depolarisingzero field region

B. Roessli1, A.Podlesnyak1, R. Thut1, Ch. Kägi1, T. Mühlebach1, R. Schneider1, R. Bürge1, P. Böni2

1Laboratory for Neutron Scattering, ETH Zürich & PSI, CH-5232 Villigen PSI, Switzerland2Physik-Department E21, Technische Universität München,

D-85747 Garching, Germany

In the actual full-polarization analysis set-up, the spin-state of the scattered neutron beam is analyzedby a bender installed before the analyzer. With that set-up, however, it is not possible to takeadvantage of the horizontally focusing PG002 analyzer. For that reason, a new horizontally focusingHeusler has been constructed for TASP. Due to a clever arrangement of the magnetic field there is nozero field region and no correction coils are required.

The triple-axis spectrometer TASP is located atthe end of a cold neutron guide at SINQ andhas been in operation since 1997. An optionwith full polarization analysis is available. In thecourse of the past years the demand forpolarized neutrons has steadily increasedreaching now an average of 25% of thescheduled beam-time. To polarise the neutronbeam, a remanent bender is installed after themonochromator. The advantage of such adevice is that no spin-flipper is needed toselect the neutron polarization [1]. To analysethe polarisation of the scattered beam, asecond bender is usually inserted in thecollimator insert before the analyzer. Becausethe benders have a collimation of about 80', itis not possible to operate the spectrometer inthe focusing mode on the analyzer side.Therefore it was decided to build a newhorizontally focusing Heusler analyser.

The new analyzer consists of 15 Cu2MnAlHeusler single-crystals (d spacing = 3.43) ofdimensions 2.5cm x 5cm produced at theInstitute Laue-Langevin, Grenoble. Thereflectivity and the mosaic of the single crystalshave been tested on the 3-axis spectrometer3AX at the late reactor Saphir at PSI (Fig. 1).

Fig. 1: Rocking curves of on of the bestHeusler (open circles) and graphite (filledcircles) monochromators. The solid lines arefits to a Gaussian.

The Heusler and graphite monochromatorshave the same size and we used the samespectrometer configuration for thecharacterisation (λ= 2.276, perfect Ge(111)monochromator). The comparison shows thatthe samples have a similar mosaic while thepeak intensity of Heusler is reduced by a factorof 4405/13170=0.33. Taking into account thatthe measurements were done in zero field andthat it was not magnetized the factor willactually go up to about 0.4. Note, that a factorof two in intensity is lost due to the polarizationof the beam. The data shows that we canexpect a reasonably high performance of theHeusler analyzer.

Each single crystal is mounted on an individualsmall goniometer, that allows aligning thescattering plane both in the horizontal andvertical directions. Three such pieces aremounted on top of each other to build one ofthe columns of the analyzer blades (Fig. 2).The top and bottom crystals are inclined by anangle of 3° to focus the beam in the verticaldirection onto the detector. To allow for a fullypolarized beam, the Heusler crystals have tobe magnetized in the vertical direction withrespect to the scattering plane. In contrast toprevious designs we have built individualyokes for each column of monochromator andoptimized the permanent magnets such that nochange in sign of the magnetic field occursbefore the analyser. This design is not onlymechanically very stable it also facilitates theuse of guide fields around the analyser. Fiveanalyzer columns form then the completeanalyzer. Each of these blades can bemechanically rotated to achieve horizontalfocusing. The mechanics are fully motorizedand computer controlled, so that the radius ofthe horizontal curvature can be adjusted as afunction of energy.


Fig. 2: Picture of the new Heusler analyzerassembly. As explained in the text the 3columns of the analyzer are magnetizedindividually to allow horizontal focusing.

References[1] F. Semadeni, B. Roessli, and P. Böni, Physica B297, 152-154 (2001).


1.3 Observation of soliton chirality in the Ising-chain system CsCoBr3

H.B. Braun1,2, J. Kulda1, P. Böni2, B. Roessli3, K. Krämer4, and D. Visser5

1Institut Laue-Langevin, Grenoble, France2Technical University of Munich, Physics Department E21, D-85747 Garching, Germany

3Laboratory for Neutron Scattering, ETHZ & PSI, CH-5232 Villigen PSI, Switzerland4ISIS, Rutherford Appleton Laboratory, Chilton, Didcot OX11 0QX, United Kindom

5Department of Chemistry, University of Bern, Switzerland

Solitons are the elementary excitations in the quasi 1D Ising compound CsCoBr3. We haveobserved a novel chiral asymmetry associated with the quasielastic Villain mode and the 2-solitoncontinuum in the inelastic neutron scattering cross section. This proves that Ising-type solitonsacquire an intrinsic chirality through quantum fluctuations.

The emergence of chiral correlations is one ofthe central themes in science ranging from theexistence of chiral molecules to chiralsymmetry breaking in particle physics. Inmagnetism, static helical structures have beenstudied with Ho as a prominent example. Onthe other hand, it is well known that quantumeffects play an important role in low-dimensional magnetic systems, which maybehave differently depending on their spinquantum number. Spin chains with integer spinexhibit a Haldane gap while half-integer spinchains are gapless, hinting towards a highersymmetry in the latter systems.

Theoretical arguments have shown thatchirality emerges as a new degree of freedomof solitons in spin-1/2 systems. This isintimately connected with the decay of amagnon into two solitons, most easily seen inthe Ising limit, where a state with two well

separated solitons ↓⇑⇑↓↑↓↑↓⇑⇑↓ ... has

the same energy as a single spin flip. Suchstates are exact eigenstates of the IsingHamiltonian prop. Jz.The transverse exchange

∑ ++ +i

yi

yi

xi

xit SSSSJ ][ 11 induces quantum

dynamics of the solitons and gives rise to theone-soliton dispersion [1] as shown in Fig.1

kBgkJJk zBtz cos2cos)2/()( µε ++= (1)

Here we have included an external magneticfield Bx in transverse direction. It should bestressed that the k2cos dispersion is not aconsequence of a unit-cell doubling due tolocal antiferromagnetic order – the same effectalso occurs for ferromagnetic solitons. Ratherthe two bandminima at 2/π±=k havedifferent chirality as illustrated in Fig. 1. Anexternal field transverse to the Ising directionlifts the chiral degeneracy, and gives rise to anonvanishing polarization dependent part ofthe structure factor

∫ −− ∧⋅ )0()( qq

ti tdtei SSP ω (2)

where P is the polarization of the incidentneutrons. We have measured the inelasticneutron response at the triple-axisspectrometers IN14, IN20 at ILL in Grenoble

working at IN14 with fixed -1Å 5.1=fk with

an energy resolution of 0.2 meV. Fig. 2 showsthe detected asymmetry between differentpolarization directions with respect to theexternal field, together with the theoreticalpredictions with no free parameters, theparameters being determined from theunpolarized measurements.

Fig. 1: (i) One soliton dispersion giving rise to a(ii) 2-soliton continuum due to soliton paircreation an a Villain-mode continuum due totransitions between thermally activated statesof the one-soliton band.


Fig. 2: (i) Total cross section of the 2-solitoncontinuum at qc=0.4 as a function of energy, cf.cut a) in Fig. 1. Blue dots correspond to spin-flip, red dots to non spin-flip intensitiesindicating opposite soliton chiralities (ii) Totalpolarization of the cross-section for the Villainmode at qc=0.5, together with the theoreticalprediction at a field of B=3.3T.

References[1] J. Villain, Physica D 79, 1 (1975); F. Devreux and

J.P. Boucher, J. Physique 48, 1663 (1987); S.E.Nagler et al., Phys.Rev.Lett. 49, 590 (1984)Sharov, Nature 357, 390 (1992).

[2] H.B. Braun, in “Complementarity betweenNeutron and Synchrotron X-Ray Scattering” ed.By A. Furrer, World Scientific (1998)


1.4 Phonon density-of-states in intermediate valence CE2NI3SI5E.Clementyev1, I.Sashin2, P.A.Alekseev3, V.N.Lazukov3 and L.C.Gupta4

1Technical University of Munich, Physics Department E21, D-85748 Garching, Germany2FLNP, Joint Institute for Nuclear Research, 141980 Dubna, Russia

3RRC “Kurchatov Institute”, 123182 Moscow, Russia4Tata Institute of Fundamental Research, 400005 Bombay, India

Inelastic neutron scattering measurements of the phonon density-of-states in intermediate valenceCe2Ni3Si5 and its nonmagnetic isostructural reference compound Y2Ni3Si5 have been performed attemperatures 10K to 300K. The phonon density-of-states in Y2Ni3Si5 is temperature-independent. Thetemperature dependence of the phonon spectrum in Ce2Ni3Si5 is indicative of a coupling of valencefluctuations and lattice vibrations.

Intermediate valence (IV) rare earthcompounds, in which a lanthanide ion can beregarded as fluctuating between two differentstates, are prone to display anomalies in theirvibrational properties because changes in the4f-shell occupancy are associated with largedifferences in the ionic radius. But whereassignificant phonon anomalies have been foundin Sm and Tm-based IV compounds [1] only afew indications have been reported so far inthe case of Ce [2].

The intermetallic compound Ce2Ni3Si5 exhibitstypical IV properties [3]. The characteristicKondo energy is about 100K. A very specialfeature of this compound is the strongtemperature dependence of the Ce valence [4].The energies of valence fluctuations and latticevibrations are close to each other in Ce2Ni3Si5and, therefore, sizable effects of the electronicinstability on the lattice dynamics can beanticipated.

0 10 20 30 40 50 60 700

1

2

3

4

5

6

7

8

9

10

Ce2Ni

3Si

5

300K 200K 130K 70K 10K

S(Q

,E)

(a.u

.)

Energy (meV)

Fig. 1: Background-corrected inelastic neutron scattering spectra of Ce2Ni3Si5 measured attemperatures 10K to 300K.


The inelastic neutron scattering (INS)measurements were carried out in thetemperature range 10K to 300K on the time-of-flight spectrometer KDSOG-M of invertedgeometry located at the IBR-2 pulsed reactorin Dubna. The final energy was fixed at Ef=4.9meV. A pyrolitic graphite analyzer was used incombination with a beryllium filter to suppresshigher-order contaminations.

Fig.1. shows the background-corrected INSspectra of Ce2Ni3Si5 measured in thetemperature range 10K to 300K. The datataken at different scattering angles (80 to 140degrees) were averaged to get a goodapproximation of the incoherent neutronscattering function. Several phonon bandsshow up in the INS spectra. The temperatureeffects (caused not only by the Bose

temperature factor but also by a shift of thephonon bands) are observed in the low-energypart of the spectra. The heavier rare-earth andnickel ions contributions dominate for thisenergies.

Fig.2 depicts the generalized phonon density-of-states (GDOS) of Y2Ni3Si5. Clearly all themajor peaks of the GDOS don’t show anysizable temperature dependence as expectedfor a nonmagnetic reference compound.

The INS spectra of Ce2Ni3Si5 combined withthose of Y2Ni3Si5 will allow a careful estimationof the magnetic contribution to the neutronscattering function to get the phonon GDOS ofthe former compound. This work is in progressnow.

0 10 20 30 40 50 60 700.0

0.5

1.0

1.5

2.0

2.5

3.0

Y2Ni3Si5

11K 130K 290K

GD

OS

(a.

u.)

Energy (meV)

Fig. 2: Generalized phonon density-of-states of Y2Ni3Si5 obtained from INS spectra.

References[1] H.A.Mook and R.M.Nicklow, Phys. Rev.

B 20, 1656 (1979).[2] E.S.Clementyev, P.A.Alekseev,

M.Braden, J-M.Mignot, G.Lapertot,V.N.Lazukov and I.P.Sadikov, Phys. Rev.B 57, R8099 (1998).

[3] C.Mazumdar, R.Nagarajan,S.K.Dhar, L.C.Gupta,R.Vijayaraghavan andB.D.Padalia, Phys. Rev. B 46,9009 (1992).

[4] C.Mazumdar, R.Nagarajan, C.Godart,L.C.Gupta, B.D.Padalia andR.Vijayaraghavan, J. Appl.. Phys. 79,6347 (1996).


1.5 Crystal field effects on Er3+ and Nd3+ ions in intermediate valenceCe2Ni3Si5

E.Clementyev1, I.Sashin2 and L.C.Gupta3

1Technical University of Munich, Physics Department E21, D-85748 Garching, Germany2FLNP, Joint Institute for Nuclear Research, 141980 Dubna, Russia

3Tata Institute of Fundamental Research, 400005 Bombay, India

Magnetic spectral response of Er3+ and Nd3+ ions in intermediate valence host matrix Ce2Ni3Si5 hasbeen measured by inelastic neutron scattering. A full crystal field splitting scheme was established forNd3+. Experimental information allows unambiguous determination of the crystal field potential for Er3+.

The hybridization of the 4f shell withconduction electrons is believed to give rise toanomalous properties in rear-earth basedcompounds. A study of cerium-basedcompounds is of particular interest as in thiscase the 4f shell contains just one electron,which may help us to explain this system withrelative ease.

Ce2Ni3Si5 has been reported to demonstrateproperties typical for intermediate valence (IV)systems [1]. Prior to measurements of thedynamic spectral response in pure Ce2Ni3Si5compound it's important to get an estimation ofa strength of the crystal field (CF) interactionvs the Kondo energy. The ratio of these twobasic interactions plays a crucial role in theformation of the ground state and magneticproperties in cerium-based systems.Unfortunately, cerium ions do not provide ussuch an opportunity due to a substantialdegree of delocalization of the cerium 4f shell,expected in Ce2Ni3Si5. That is why using aparamagnetic impurity ions with a stablevalence is the only possible way to learnsomething on the CF interaction.

The first measurements of the CF effects inCe2Ni3Si5 were performed using the Er3+ andNd3+ ions as paramagnetic impurities. ManyCF transitions show up in the inelastic neutronscattering (INS) spectra measured on the tripleaxis spectrometer Drüchal (SINQ neutronsource, Villigen PSI). However the energytransfer range was limited by 10 to 12 meV sonot all the CF transitions were visible [2].

The main aim of the present experiments wasto enlarge the energy transfer range to get afull CF splitting scheme for Er3+ and Nd3+ ionsin Ce2Ni3Si5. INS measurements have beenperformed on the time-of-flight spectrometerKDSOG-M of inverted geometry located at theIBR-2 pulsed reactor (Dubna). The finalneutron energy was fixed Ef=4.9 meV. Such ageometry allows getting the INS spectra in theenergy transfer range up to about 200 meV.

Fig.1. shows the INS spectra of Ce2Ni3Si5sample with Er3+ paramagnetic impurity. Themagnetic contribution dominates in the energytransfer range E<10 meV.

0 2 4 6 8 10 20 30 40 50 60 70 80 90 1000

2

4

6

8

10

12

14

16

(Er0.15

Ce0.85

)2Ni

3Si

5

T=10K T=50K

S(Q

,E)

(a.u

.)

Energy (meV)

Fig. 1: Background-corrected inelastic neutron scattering spectra of (Er0.15Ce0.85)2Ni3Si5.


Its temperature dependence is indicative of theCF transitions out of the ground state. Nosizable magnetic scattering was detected athigher energies. In the case of Nd3+ (see Fig.2)two CF transitions from the ground state weredetected at 10meV<E<20 meV. At higherenergies only the phonon contribution to theneutron scattering function is visible. Thus a

full CF splitting scheme is established for Nd3+,namely E=0, 6.2, 9.2, 14.8 and 17.4 meV.

As to Er3+ ions in Ce2Ni3Si5, a few CF statesare invisible but the INS experiment providesenough data to calculate the CF parametersand to restore a full CF splitting scheme.

0 5 10 15 20 25 30 40 50 60 700

2

4

6

8

10

12

14

(Nd0.15

Ce0.85

)2Ni

3Si

5

T=10K T=73K

S(Q

,E)

(a.u

.)

Energy (meV)

Fig. 2: Background-corrected inelastic neutron scattering spectra of (Nd0.15Ce0.85)2Ni3Si5.

References[1] C.Mazumdar, R.Nagarajan, S.K.Dhar,

L.C.Gupta, R.Vijayaraghavan and B.D.Padalia,Phys. Rev. B 46, 9009 (1992).

[2] E.Clementyev and L.C.Gupta, unpublished.


1.6 Search for the crystal field splitting in TMB12

A.Murasik1, A.Szopnik2, E.Clementyev3, S.Janssen4 and N.Shitsevalova5

1Institute of Atomic Energy, Swierk, Poland2Institute for Low Temperature Wroclaw, Poland

3Technical University of Munich, Physics Department E21, D-85748 Garching, Germany4Paul Scherrer Institute Villigen, Switzerland

5Institute for Problems of Material Science Kiev, Ukraine

Neutron inelastic scattering experiments performed on the time of flight spectrometer FOCUS using apowder sample TmB12 enriched with 11B isotope show at low temperatures two clearly visibile CFtransitions at 7.5 and 14 meV, which basing on the temperature behaviour of the spectrum, allowedus to determine the crystal field splitting scheme in this compound with the CF parameters:W = 0.104 meV and x = - 0.071.

We have shown in the previous work [1] thatTmB12 orders antiferromagnetically at lowtemperatures. Its magnetic structure has beendescribed as a modulation of magneticmoments, propagating along three crystallo-graphic direction. The heat capacitymeasurements allowed us to determine theSchottky contribution to the specific heat fromwhich two crystal-field parameters x ≅ - 0.15and W ≅ 1.1 K yielding the Γ5

(1) triplet as alowest level was found. This result agreeswell with neutron data but the crystal fieldsplitting scheme needed further confirmation.However, there were some doubts concerningthe possibility of performing successful neutroninelastic experiments. First of all, (as proved bythe diffraction experiment) the enrichement ofthe sample with low absorbing 11B isotope wasnot 100% yielding in effect poor signal-to-background ratio. Secondly, due to a largeamount of nonmagnetic ions in the sample,one could expect also the significant phononscattering. In these circumstances we havedecided to perform a testing measurement onthe time of flight instrument "FOCUS" (PSI,Villigen, Switzerland). To optimise theexperimental conditions a flat geometry wasused, i.e. the sample was contained in twosheets of aluminum foil of 50 x 30 mm sizecovering 60 x 30 mm beam aperture. Thesample was placed at an angle of ≅ 30degrees with respect to the incoming beam tofavour reflection geometry applicable for ahighly absorbing material.

Several neutron spectra have been collectedincluding the variation of incoming energy Ei,energy transfers ∆E and temperature. Fig. 1displays the main feature of the INSexperiment. It shows two principal maxima (cf:spectum (b) and (c)) taken at differentgeometry and different incoming energies.

The spectrum (b) displays the intensityvariation with temperature characteristic for

the ground state CF transition. On (c), the de-excitation of two levels at about 7.5 meV and14 meV is visualized by two sizable peaks. Thecrystal field splitting scheme displayed on(a) is based on performed experiments aswell as a simulation of calculated intensitiesfrom which one can conclude that the maincontribution to the scattering, results fromtransitions between three lowest levels.


-5 0 5 10 15 20 25 30 35

(a)

Crystal field splitting scheme in TmB12

CF parameters: W = 0.103 meV, x = -0.074

Γ5(1)

Γ3

Γ4

Γ1

Γ5

(2)

Γ2

Energy (meV)

-4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 120123456789

1011

(b)

12K 50K 100K

TmB12

Ei=17 meV

Inte

nsity

(a.u

.)

Energy (meV)

(c)

-30 -25 -20 -15 -10 -5 00.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

TmB12

T=100K Ei=3.8meV

Inte

nsity

(a.

u.)

Energy (meV)

Fig. 1 Crystal field splitting scheme and observed CF transitions in TmB12.References[1] A.Czopnik, A.Murasik, L.Keller, N.Shitstevalova and Y.Paderno, Phys. Stat. Sol. (b) 221, R7 (2000)


1.7 Mapping of magnetic excitations in single-Q chromiumP. Böni1, E. Clementyev1, F. Demmel2, and G. Shirane3

1Technical University of Munich, Physics Department E21, D-85748 Garching, Germany2Institute Laue Langevin, F-38042 Grenoble Cedex 9, France

3Physics Department, Brookhaven National Laboratory, Upton, New York 11973

The low-energy excitations of antiferromagnetic Cr have been measured by means of inelastic neutronscattering using a novel multi detector setup of the triple-axis spectrometer IN3. A huge area in the[100]/[010] scattering plane was investigated around a few reciprocal lattice positions. The observedintensity maps indicate that the low-energy excitations for E = 6 meV are arranged on a ring in the Qh-Qk scattering plane around the commensurate positions. The multi detector system on the IN3spectrometer has proved to be a very powerful “excitation scanner”.

Cr is one of the most thoroughly studieditinerant antiferromagnets. The richness of thephenomena observed in Cr derives from itsspin-density-waves (see [1] and referencestherein). Recently interest has been focused onthe unusual spectral features of chromium,namely the low-energy Fincher-Burke (FB)excitations around the commensurate positions[2]. The origin of these modes and theirrelevance with regard to incommensurableorder is unclear. A series of high-resolutionmeasurements was performed recently tocharacterize the behavior of the FB modes[3,4]. However experimental information israther scarce since contour maps of intensitywere collected only along two directions in Q-space around the commensurate wavewectorQ = (1,0,0).

The primary goal of the measurements was toexplore a large region in the Qh-Qk scatteringplane around the commensurate positions andproduce contour maps of excitation spectra atdifferent neutron energy transfers. Aconventional triple-axis spectrometer (TAS) isnot well suited for such a study due to theenormous number of points in Q-space to bemeasured. That is why a novel multi detectorTAS was used to map out the magneticexcitation spectrum of Cr. The TAS IN3 (ILL)was equipped with a set of 32 analyser crystalspositioned on an arc around the sample (see [5]for details). Constant-E scans were performedin the [100]/[010] scattering plane at T = 230 Kand fixed final energy Ef = 31 meV. Fig. 1depicts the points in the Qh-Qk plane covered bya single constant-E scan.

The (radial) Q-resolution along the h-direction israther coarse due to the size of the individualanalyser crystals, while the (tangential) Q-resolution along the k -direction can be madesignificantly smaller because it is mostly givenby the step size of the orientation of the singlecrystal on the sample table and the samplemosaic.

0.6 0.8 1.0 1.2 1.4 1.6-0.3

-0.2

-0.1

0.0

0.1

Qk (

r.l.u

.)

Qh (r.l.u.)

Fig. 1: Points in the Qh-Qk plane covered by atypical constant-E scan.

The scans were performed around thereciprocal lattice positions Q = (1,0,0), (0,1,0)and (1,1,0) at several energy transfers 0 meV <E < 20 meV. The scattering intensity measuredaround Q = (1,0,0) at E = 6 meV is shown inFig. 2.

Fig.2: Contour map of the excitation spectra ofCr as measured at E = 6 meV around Q =(1,0,0).


The strong incommensurate scattering is clearlyseen at Q = (1+ δ,0,0) as white spots. Thesemodes emerge from the magnetic satellites. Astriking feature is a ring of intensity between theincommensurate peaks with a hole at Q =(1,0,0) at E = 6 meV. This feature representsthe FB excitations. In addition, we also see anincrease of the intensity at the silent positions(1,±δ,0).

Fig. 3: Contour map of the excitation spectra ofCr at E = 8 meV measured around Q = (1,1,0).

It is worth to note that no magnetic intensitywas observed at the reciprocal lattice position Q= (1,1,0), which is not accompanied bymagnetic satellites. Only the phononcontribution is present in the spectrum as arectangle (see Fig. 3). It is directly seen that thelongitudinal phonons along have a largervelocity of sound than the transverse phonons.

The most salient feature of the present study isthe ring of intensity around Q = (1,0,0) thatconfirms the results on the new excitations in Cr[4]. We point out that the present experimentshave been performed on a different Cr crystaland very different experimental conditions thanin reference [4]. Similar scans have beenperformed around the (0,1,0) Bragg peak,where the Q-resolution along the h-direction isexcellent, while along the k -direction it iscoarse.

The TAS IN3 with a novel multi detector setupproves to be a useful inelastic spectrometer toinvestigate large areas in reciprocal space. Theexperiment was carried out in a view toencourage such experimental methods in theTAS spectroscopy. In principle, it would berather trivial to increase the performance of themulti-detector setup by a factor of 10 by movingthe multidetector from the thermal 58Ni guideposition to a thermal beam port at the ILL.

References:

[1] E. Fawcett, Rev. Mod. Phys. 60, 209(1988).

[2] S. K. Burke, W. G. Stirling, K. R. A.Ziebeck, and J. G. Booth, Phys. Rev. Lett.51, 494 (1983).

[3] P. Böni, B. Roessli, E. Clementyev, Ch.Stadler, G. Shirane, and S. A. Werner,Applied Physics A 74 (suppl.), S716 (2002).

[4] H.Hiraka, Y.Endoh, P. Böni, M.Fujita,K.Yamada, and G.Shirane, Phys. Rev. B67, 064423 (2003).

[5] F.Demmel, The ILL Millenium Symposium &European Users Meeting, Grenoble, 305(2001).


1.8 Specific heat study of the spin-chain compounds Ca2+xY2-xCu5O10

A. Mirmelstein 1,2, V. Kargl 1, A. Amato 3, J. Karpinski 4, S. Kazakov 4, P. Böni 1, A. Erb 5

1Technical University of Munich, Physics Department E21, D-85748 Garching, Germany2Institute for Metal Physics, Russian Academy of Sciences, 620219 Ekaterinburg GSP-170, Russia3Laboratory for Muon Spin Spectroscopy, Paul Scherrer Institute, CH-5232 Villigen PSI, Switzerland

4Laboratory for Solid State Physics, ETH Hönggerberg, CH-8093 Zürich, Switzerland5Walther Meissner Institute, Bavarian Academy of Sciences, D-85748 Garching, Germany

The specific heat of the spin-chain compounds Ca2+xY2-xCu5O10 as a function of x (0 ≤ x ≤ 1.2) wasmeasured in magnetic fields up to 9 Tesla.

Recently, the new system Ca2+xY2-xCu5O10 hasbeen synthesized which consists only of linearchains and allows a variable doping levelranging from x = 0 to x = 2 (formal coppervalences from +2 to +2.4) [1]. This wide rangeof hole doping makes these compoundsparticularly interesting for a detailed study ofdoping-induced dimensional crossoverbetween 3D and 1D. In fact, the undopedcompound exhibits long-range AF order withTN = 29 K [2]; however, as holes are dopedinto the chains, the magnetic susceptibility andthe specific heat data indicate a change from3D long-range order to 1D chain behavior [3].Therefore, these compound series are anexcellent model for linking experimental andtheoretical studies of spin and chargedynamics in doped low-dimensional copperoxides. The aim of the present study is tocharacterize the magnetic state of Ca2+xY2-

xCu5O10 as a function of x by specific heatmeasurements.

The ceramic samples of Ca2+xY2-xCu5O10 (0 ≤ x≤ 1.5) were prepared at high oxygen pressure(> 200 bar) as described in [1]. The specificheat was measured by the relaxationtechnique using a conventional PPMS(Quantum Design) in magnetic fields up to 9Tesla. The magnetic state of the samples withx = 0 and x = 1.5 was also probed by µSR.As expected, the highly doped compositionCa3.5Y0.5Cu5O10 shows no long-range magneticorder. Instead, the x = 1.5 data are indicativeof a 1D Heisenberg chain behavior (Fig. 1).The specific heat results are in agreement withthe magnetization measurements. Absence ofa static magnetic order down to 4 K for thiscomposition is confirmed by the µSRexperiments.

In agreement with previous results [2], theundoped Ca2Y2Cu5O10 compound experiencesan AF ordering at TN = 29 K. The Cu ions forman orthorhombic sublattice within theorthorhombic Fmmm structure. The magneticpeaks could be indexed within this chemical

unit cell with k = [0,0,1]. The magneticmoments lie along the b direction of the unitcell, i.e. perpendicular to the chain direction[100]. Since the low temperature orderedmoment (0.95 ± 0.03)µB is close to the free ionvalue for Cu2+, spin fluctuations are expectedto be negligible. However, Fig. 1 demonstratesthat the magnetic specific heat does not vanishabove TN and, moreover, reveals the sametemperature dependence as highly dopedcompositions. Essentially, the anisotropiccharacter of the magnetic excitations inCa2Y2Cu5O10 is probably responsible for themagnetic field-induced broadening of the AF(Fig. 2). Systematic investigation of thethermodynamic properties of the Ca2+xY2-

xCu5O10 compounds as a function of doping isin progress.

Fig. 1: The magnetic contribution to thespecific heat Cmag = Cmeasured – Clattice ofCa2+xY2-xCu5O10 for x = 0 and 1.5.

0

0,5

1

1,5

2

2,5

0 10 20 30 40 50 60 70 80

T (K)

Cm

ag =

C -

Cla

t (J/

K g

at)

Ca2Y2Cu5O10

Ca3.5Y0.5Cu5O10


Fig. 2: The specific heat of Ca2Y2Cu5O10 as a function of magnetic field in the region of the AFtransition.

References:

[1] A. Hayashi et al., Phys. Rev. Lett. 58, 2678(1998)

[2] H.F. Fong et al., Phys. Rev. B 59, 365 (1999)

[3] M.D. Chabot et al., Phys. Rev. Lett. 86, 163(2001)

1,6

2

2,4

2,8

26 27 28 29 30 31T (K)

C (J

/K g

at)

0 T

2T

3 T

5 T

7 T

9 T


1.9 Frustrated phase separation in the overdoped regime of ‘123’ high-Tc

superconductors A. Mirmelstein 1,2, V. Bobrovskii 2, N. Golosova 2 A. Podlesnyak 3, A. Furrer 3

1Technical University of Munich, Physics Department E21, D-85748 Garching, Germany2Institute for Metal Physics, Russian Academy of Sciences, 620219 Ekaterinburg GSP-170, Russia

3Laboratory for Neutron Scattering, ETH Zurich & PSI, CH-5232 Villigen PSI, Switzerland

The systematic analysis of the concentration dependance of the crystal-field (CF) parameters for bothHo- and Er-based “123” high-TC compounds shows that the local charge inhomogeneity in the CuO2planes of high-TC cuprates is a characteristic feature of the doping process, which depends neither onthe way of introducing doping nor on the doping level.

For some time, superconductivity in thecuprates has been believed to occur in ahomogeneous system. However, theexperimental evidence for spatial charge andspin inhomogeneities and lattice effects hasbeen accumulated over the last years. Theexperiments show that the distribution ofdoped charges in the CuO2 planes is veryinhomogeneous in the underdoped regime andbecomes more homogeneous when goingabove the optimal doping. Theoretical modelsemerging in response to these data emphasizethe strong involvement of lattice degrees offreedom in mediating superconductivity incuprates as well as a direct relation betweenthe pseudogap phenomenon and an intrinsicelectron inhomogeneity of the metallic phase[1]. Thus the experimental study of the spatialdistribution of the doping-induced charges inthe CuO2 planes is of crucial interest.

The crystal-field (CF) interaction measured bythe inelastic neutron scattering (INS) techniqueallows direct observation of the clusterformation upon doping of high-TC cuprates withcharge carriers [2]. This phenomenon (whichmay be called “frustrated phase separation”)reflects a remarkable difference between theaverage carrier concentration and the localcharge density in the CuO2 planes near thedoping centers. In this work [3] wedemonstrate that the frustrated phaseseparation exists also in the overdoped regimeof the “123” cuprates.

The INS experiments performed on the tripleaxis spectrometer DruehaL at SINQ (PSI,Switzerland) revealed the CF spectra of theoverdoped R1-yCayBa2Cu3O7 (R=Ho, Er;0<y<0.25) to consist of two spectralcomponents associated with the optimallydoped and the overdoped domains (Fig. 1).Increase of the Ca concentration does notaffect the local charge density of domains, butchanges the spectral weight of thecomponents. Similar “two-phase” picture was

established earlier for the underdoped regionof the phase diagram.

We have demonstrated that the CF parametersas a function of the in-plane hole concentrationfollow the same dependence for both Er- andHo-based 123-compounds, below and aboveoptimal doping and independently of the way tointroduce doping (by variation of the oxygenstoichiometry or by the partial Ca substationsfor rare-earths). The established systematicbehavior of the CF parameters proves thecharge origin of the superposition effect.Therefore, it exists a smooth crossoverbetween the under- and overdoped parts of thephase diagram. It seems, that the resultsobtained are compatible with the scenariosuggested recently in Ref. 1.


-0.5 0.0 0.5 1.0 1.5 2.0 2.50

20

40

0

40

800

4080

120160

0

50

100

150

0

25

50

75

4 6 8 10 12 140

25

50(c)

(b)

BB'A'A

B'A' BA

Neu

tron

coun

ts

(a)

Energy transfer (meV)

BA

(a)

FEDC

(b)FED

D'C'

C

F'E' (c)FED

D'C'

C

FEDCBA

Γ4

(2)Γ2

(2)Γ1

(2)Γ

3

(2)Γ1

(1)Γ

2

(1)Γ4

(1)Γ3

(1)

Fig. 1: Energy spectra of neutrons scattered from Ho1-yCayBa2Cu3O7 at T = 1.5 K for y = 0, 0.1, 0.25(in panels (a), (b), (c), respectively). Left: high-resolution scans of the lowest CF transition at Q = 0.85Å-1, Ef = 3.5 meV. Right: scans at Q = 1.8 Å-1, Ef = 7 meV. The top of the figure shows the low-energyCF splitting of Ho1Ba2Cu3O7. The observed transitions are indicated by arrows.

References

[1] Yu. Ovchinnikov, S.A. Wolf, and V. Kresin, Phys.Rev. B 63, 064524 (2001)

[2] J. Mesot et al., Phys. Rev. Lett. 70, 865 (2001)

[3] A. Mirmelstein et al., J. Superconductivity:Incorporating Novel Magnetism 15, 367 (2002)


1.10 X-ray and magnetization studies on FeCoV/Ti multilayersM. Senthil Kumar1, Shah Valloppilly2, Christian Schanzer2, Peter Böni2, M. Horisberger3

1Department of Physics, Indian Institute of Technology Bombay, Mumbai 400 076, India2Physik Department E21, Technische Universität München

3Laboratory for Neutron Scattering, ETHZ & PSI, CH-5232 Villigen PSI, Switzerland

A series of FeCoV/Ti multilayers have been investigated by x-ray reflectivity, high angle x-raydiffraction and magnetization measurements in order to understand how the microstructural andmorphological parameters influence the magnetic behavior of the multilayers. Interface roughnessincreases with the total thickness and the existence of lateral roughness correlation is observed fromthe x-ray diffuse scattering background. A strong magnetic coupling is observed between theferromagnetic layers that depends on the Ti spacer layer thickness. For thicker Ti spacer layers agradual magnetization reversal is observed.

Spin-dependent scattering of neutrons is beingutilized as the working principle of the polarizerand analyzer in a neutron scatteringexperiments and this concept can be realizedin FeCoV/Ti multilayers since the scatteringlength for polarized neutrons is:

mcoh bbb ±=±

where the total scattering length has a nuclearand a polarization dependent magnetic part.Therefore, very large contrast between the

+ and − eigen states of the neutrons can

be achieved by selectively choosing the non-magnetic material (e.g. TiNx) and the angle oftotal reflection can be increased by increasingthe number of layers [1]. However, theorientation of magnetic moments in themultilayers with respect to the neutronmagnetic moment is an important factor thatdetermines the efficiency of the polarizer,especially in the remanent state.Misorientation of the layer magnetic momentand the neutron moment can cause significantspin-flip scattering. In multilayers, sharp andflat interfaces are exceptional and in practice,they can be topographically and chemicallyrough due to morphology in the growth processand the interdiffusion. Such interfaces mayinfluence the magnetic interlayer coupling andcan affect magnetic properties like coercivity,anisotropy etc [2]. We have investigated theinterface of FeCoV/Ti multilayers by specularx-ray reflectivity and off-specular scans. Wehave also studied the magnetization reversalas a function of varying the spacer layerthickness.

[FeCoV(100 Å)/Ti (x Å)/ FeCoV(30 Å)/Ti (xÅ)]20 (x = 30, 50, 80, 100, 120 and 150 Å)multilayers were sputter deposited by dcmagnetron sputtering at PSI, Switzerland.High angle x-ray diffraction reveals that thefilms are textured and polycrystalline, and aregrown in the orientation of highest in-planeatomic packing density, as deduced from [110]and [002] reflections for FeCoV and Ti

respectively. The analysis of the peak widthusing the Scherrer equation shows that thegrain size is of the same order as the layerthickness. Specular reflectivity as a function ofmomentum transfer perpendicular to the filmplane (qz) show the Bragg peaks due to theperiodically repeated FeCoV and Ti layers.The fast decay of the specular intensity with qz

is due to the roughness developingprogressively during the film growth. Thereflectivity curve simulated using appropriatefilm thickness, x-ray scattering factors, volumedensity of the materials and the roughness arecompared with the experimental data and atypical example is shown in Fig.1.

Fig.1 Specular x-ray reflectivity of Ti = 30 Åmultilayer sample. The roughness is modeledto 0.1 t (t is in units of nm).


Fig.2. Switching of soft and hard layermagnetizations in FeCoV/Ti multilayers

From the specular reflectivity analysis it is

observed that the roughness scales with t ,where t is the thickness of the film innanometer. Since the specular reflectivitycannot distinguish the roughness frominterdiffusion, we carried out off-specularreflectivity measurements by using rockingscans of the sample. A broad diffusebackground and a sharp specular peak arevisible in the diffuse scan. Since the width ofthe diffuse profile is inversely related to theeffective in-plane roughness correlation length,it indicates the existence of vertically correlatedroughness apart from the possibleinterdiffusion.

There exists an in-plane anisotropy in thesemultilayers of magnetoelastic origin arisingfrom the in-plane stress distribution caused bythe preparation conditions [3]. In fig.2, isshown the second quadrant of the hysteresisloop, where the magnetization reversals of thethin (soft) and thick (hard) FeCoV layers arefound to be strongly dependent on the Tispacer layer thickness. For thin Ti layers, thehard and soft layers switch almost at the samereverse field, whereas at higher Ti spacerthickness, the soft layers switch earliercompared to the hard layers. However, weobserve that the switching fields are differentirrespective of the same soft layer thicknessand also the magnetization reversal becomesless sharp compared to their bilayercounterparts [4]. When the Ti spacer thicknessis small, the exchange can be mediatedthrough RKKY type mechanism or mediatedthrough pinholes. At higher Ti spacerthickness, the coupling is mainly due to theNeél type (Orange-peel) coupling originatingfrom the dipolar fields created by the rough

interfaces. The topographical roughness and‘magnetic roughness’ are very importantparameters in the performance of the neutronpolarizer supermirrors and presently, we areinvestigating the origin of this ‘magneticroughness’ by polarized neutron reflectivityexperiments.

[1] P. Böni, Physica B 234 – 236, 1038 (1997)[2] Chung-Hee Chang and Mark H. Kryder, J. Appl.

Phys. 75 6864 (1994)[3] D. Clemens, A. Vananti, C. Terrier, P. Böni, B.

Schnyder, S. Tixier and M. Horisberger, PhysicaB 234 – 236 , 500 (1997)

[4] M.Senthil Kumar and P. Böni, J. App. Phys. 91,3750 (2002)


1.11 Preparation and Characterization of FeCoV/Ti Multilayers: a ComparisonC. Schanzer1, S. Valloppilly1, B. Russ1, P. Böni1, C. Breunig2, A. Urban2

1Physik Department E21 der Technischen Universität München2Forschungsreaktor München II der Technischen Universität München

The physical properties of materials as a thinlayer can be rather different from their bulkcounterparts. Especially the magnetism of thinlayers shows interesting behaviour. The aim ofour investigations is to study the dependenceof the magnetic properties on microstructuralparameters like interdiffusion and interfaceroughness. In order to address such questions,metallic multilayers were prepared consistingof ferromagnetic layers separated by non-magnetic layers. The multilayers were grownby DC magnetron sputtering. The structuralproperties were characterized by specular andoff-specular X-ray reflectometry. Furthermorewe performed magnetization measurements toinvestigate the magnetic properties. Therewere clear differences from known similarsystems [1], most likely originated fromdifferent growth conditions.

We investigated multilayers consisting of thealloy Fe50Co48V2 as the magnetic layer andtitanium as the non-magnetic layer. Suchsamples with different layer thicknesses wereavailable from the Paul-Scherrer-Institut (PSI).Some of those results are discussed in [2].Furthermore, we prepared similar samplesusing the sputter unit at ForschungsreaktorMünchen II (FRM-II). Here we equipped anexisting target head with two DC magnetrontargets (viz. Fe50Co48V2 and Ti) as shown infig. 1. A first series of samples was depositedat a pressure of 1.6 µbar. A deposition rate of≈ 28 Å/s at the power of 0.8 kW for the FeCoVtarget and a rate of ≈ 16 Å/s at the power of0.7 kW for the Ti target was obtained. Theaccuracy of the layer thickness was estimatedto be 1% of the nominal thickness.

Fig. 1: Target head for the FRM-II sputter unitequipped with two DC magnetron targets:FeCoV and Ti

Specular and off-specular X-ray reflectometrywere performed to characterize the structuralproperties of the multilayers. Themeasurements were carried out using aSiemens D5000 diffractometer. Originally, theunit was a dedicated powder diffractometer.After upgrading with a sample table equippedwith tilting facility, a high precision knife-edgecollimator, an automatic absorber and amonochromator, the instrument was enabled toperform high quality reflectometrymeasurements. The knife-edge collimator isespecially beneficial for the off-specularreflectometry since, to a good approximation,the angular resolution is independent of theangle of incidence. A comparison of samplessputtered at FRM-II and PSI is shown in Fig. 2.The specular reflectivity showed more intenseBragg peaks for the FRM-II sample, whichwere also prominent at higher qz. As aqualitative result one can deduce that theinterface roughness of the FRM-II sample ismuch smaller than that of the PSI sample. Thiswas also supported by the off-specularreflectometry results where the diffuse profilebackground for FRM-II sample is found to bemuch less than that for the PSI sample.


1,E-08

1,E-07

1,E-06

1,E-05

1,E-04

1,E-03

1,E-02

1,E-01

1,E+00

0 0,05 0,1 0,15 0,2 0,25 0,3 0,35 0,4

qz [A-1]

Ref

lect

ivity

FRM-II sample

PSI sample

a)

0,000001

0,00001

0,0001

0,001

0,01

0,1

1

-0,001 -0,0008 -0,0006 -0,0004 -0,0002 0 0,0002 0,0004 0,0006 0,0008 0,001

qx [A- 1]

norm

aliz

ed In

tens

ityI(

q z) /

I(q z

= 0

)

FRM-II sample (qz=0.0743A-1)PSI sample (qz=0.0787A-1)

b)

Fig. 2: X-ray measurements of multilayers: [Ti(100 Å) / FeCoV (100 Å) / Ti (100 Å) / FeCoV(30 Å)]20 + Ti (50 Å), a) specular reflectivity b)diffuse scattering (rocking scan)

Magnetization measurements were performedusing a Quantum Design PPMS. Acomparison, like in the x-ray data, between thetwo types of samples showed a largedifference in the magnetic properties althoughthe thicknesses of the layers were identical.Basically, one would expect a dependence ofthe coercive field of the FeCoV layers to beproportional to the layer thickness in a certainthickness region [1]. This has been observedalso for the FRM-II samples. But here, thecoercivity was quite large compared to that ofthe PSI counterparts. This difference and someother magnetic properties like interlayercoupling have to be understood further inconjunction with the structural differences ofthe layers as seen from the x-ray studies.

-1-0,9-0,8-0,7-0,6-0,5-0,4-0,3-0,2-0,1

00,10,20,30,40,50,60,70,80,9

1

-500 -400 -300 -200 -100 0 100 200 300 400 500

ext. field [Oe]

M/M

s

FRM-II sample

PSI sample

Fig. 3: Magnetization measurements ofmultilayers: [Ti (100 Å) / FeCoV (100 Å) / Ti(100 Å) / FeCoV (30 Å)]20 + Ti (50 Å)

References

[1] M.Senthil Kumar, P. Böni, J. App. Phys. 91,3750 (2002)

[2] S. Valloppilly, this annual report


1.12 Critical spin dynamics in CsMnBr3

C. Reich1, P. Böni1 and B. Roessli2

1Technische Universität München, Physik-Department E212Laboratory for Neutron Scattering, ETH Zürich & Paul-Scherrer-Institut, CH-5232 Villigen, Switzerland

Inelastic magnetic neutron scattering studies were performed on CsMnBr3 in the critical regime aboveTN= 8.3 K. We observe an upward renormalization of the spin-wave dispersion at the magnetic zonecenter (1/3 1/3 1) with increasing temperature. The different dispersion branches were preciselydistinguished by applying polarization analysis techniques .

CsMnBr3 represents a good realization of atriangular Heisenberg antiferromagnet on astacked hexagonal lattice. A strong magneticexchange coupling Jc exists between thestacked hexagonal net planes, but in the planeitself only a relatively weak coupling Jab ispresent. Additionally, an anisotropy due tolong-ranged dipole-dipole interactions confinesthe spins into the hexagonal plane. Three-dimensional magnetic ordering breaks down inthe critical temperature region above TN = 8.3K, but strong short-ranged magneticcorrelations still persist along the c-axisbecause of the large value of Jc (see figure 1).Thus CsMnBr3 shows quasi-one-dimensionalantiferromagnetic behaviour for a wide criticalregion above TN.

Fig. 1: Sketch of the spin chains along the c-axis. The dashed boxes represent one-dimensional AF-domains.

We investigated the spin wave dispersionspectra of CsMnBr3 in the critical regime andespecially the renormalization behaviour of theenergy gap at the magnetic zone center(1/3 1/3 1). CsMnBr3 shows three dispersionbranches in the ordered phase: Two branchesof in-plane spin fluctuations (acoustic andoptic) and one optic branch, corresponding toout-of-plane fluctuations (figure 2). The opticalcharacter of the latter is due to dipolaranisotropy, see Hummel et al. [1]. Previouscritical scattering measurements [2] indicate adownward renormalization of the acoustic in-plane mode, but the behaviour of the twooptical modes above TN still remainedunknown.

Our measurements were performed using thetriple-axis-spectrometer TASP at the neutronspallation source SINQ, located at the Paul-Scherrer-Institute in Switzerland. In a firstexperiment, we clearly observed a linearrenormalization upwards of at least one opticalmode with rising temperature, see figure 2.

Since only one optical excitation peak wasobserved in the spectra of figure 2, weconcluded that both optical branches mayappear at very similar energies and mergetogether to one single scattering signal. Thusin a second experiment polarization analysiswas applied with aim to resolve bothcomponents by means of polarized neutronscattering. The results confirmed that bothmodes linearly renormalize upwards withincreasing temperature (figure 3).

Fig. 2: Spin wave dispersion in the hexagonalplane. The lines represent the calculateddispersion branches in the ordered phasebelow TN. The black points are values gainedat T = 9 K, the circles correspond to T = 17.5 Kand the triangles to T = 40 K. The arrowsindicate the observed renormalization of themodes (see also figure 3).


Fig. 3: Temperature dependence of the opticalin-plane (XY, circles) and out-of-plane (ZZ,triangles) modes.

This very peculiar behaviour indicates that theinteractions between the spin chains are notnegligible. In fact, dipole effects seem to play asignificant role by building up interchain-correlations. Especially the spin dynamics ofthe optical out-of-plane mode underlies strongcompeting influences by dipole and exchangeforces.

References[1] Hummel et al., Phys. Rev. B 63, 094425 (2001)[2] Mason et. al., Phys. Rev. B 39, 586 (1989)


1.13 Polarized SANS studies from the weak itinerant helimagnet MnSiR. Georgii1, P. Böni2, H. Eckerlebe3, S. Grigoriev4, N. Kardjiloy2,

A.Okorokov4, K.Pranzas3, B. Roessli5 and W. Fischer5

1Technische Universität München, Germany2Technische Universität München, Physik-Department, E21, Germany

3FRG-1, GKSS, Geestacht, Germany4PNPI, St. Petersburg, Russia

5PSI, Villigen, Switzerland

The intermetallic compound MnSi shows at Tc=28.5 K a magnetic phase transition from theparamagnetic phase to an ordered phase witha long-period ferromagnetic spiral [1]. Theantisymmetric exchange interaction, the so-called Dzyaloshinsky-Moria (DM) interaction,being responsible for this structure, arises fromthe lack of inversion symmetry in the cubicstructure of MnSi. A unique feature of the DMinteraction is that the spiral should be eitherright or left handed depending on the sign ofthe DM interaction. Polarized SANS is ideal forthe investigation of chirality and the criticalscattering in such a system. Furthermore in aSANS experiment, where the MnSi isorientated along its (111) direction, themagnetic peaks should be visible close thenuclear diffraction point, which is itself blockedby the beam stop. It also offers the advantageof better q-resolution compared to for examplea three-axis spectrometer. Another interestingpoint is how the critical scattering around Tc

develops from the ordered to the paramagneticphase [2]. Here the critical indices can bedetermined.

Therefore we performed a polarised SANSexperiment at the SANS-2 of the researchreactor FRG-1 of the GKSS in Geestacht. TheSANS-2 polarizes the neutrons after they havebeen monocromatised in a velocity selectorwith a multilayer supermirror such, that onlyone spin state can pass the mirrorunperturbed. A RF spin flipper from PNPI,which has no material in the beam, can thenflip this state for all neutrons without beingsensitive to the their velocity. Therefore the fullwavelength spread of ∆ λ / λ of ≈ 5% at 5 Åcan be used for the experiment. The disk-shaped sample is mounted perpendicular tothe beam in a cryomagnet and its temperatureis being controlled by a Lake-shore controller.The position sensitive detector (55 cm x 55 cmsensitive area) can be moved from 1 m to 16 mfrom the sample offering a q-range of 10-3 Å-1 <q < 0.3 Å-1. For our experiment the magneticpeak is expected to be at 0.035 Å-1, thus thedetector was positioned at a distance 1mbehind the sample. The data were recorded foreach spin state separately allowing for apolarisation sensitive evaluation. This is

ongoing, but some preliminary results aregiven here:

MnSi shows exactly the expected behaviour(see Figure 1) as described above. Attemperatures slightly above the criticaltemperature a ring around the nuclear peakcan be observed, which narrows whenapproaching Tc. Below the critical temperaturetwo magnetic peaks form, which are sensitiveto the incoming neutron polarisation. It shouldalso be noted that the cross sections are huge.We get more then 500000 counts in 100 secper magnetic peak! The next step will be toevaluate from the 2-dimensional intensitypattern the susceptibility and to determine thecritical exponents for the phase transition.

Fig.1: The transition from the disorderedmagnetic phase (left) to the spiral orderedphase (middle and right) in MnSi in the criticalscattering regime around Tc = 28.5 K. Thecrystal was oriented along its (111) direction.The two pictures below Tc are shown for bothpolarisation states of the incoming neutrons.

References[1] G. Shirane et al., Physical Review B, 28, 6251

(1983)[2] B. Roessli et al., Physical Review Letters 88,

237204-1 (2002)


1.14 Design of an inelastic neutron spectrometer for catalytic applications atFRM-II

A. Jentys1, J. A. Lercher1, E. Steichele2, W. Petry2

1Technische Universität München, Department of Chemistry2Technische Universität München, Physics Department

BackgroundVibrational spectroscopy represents a versatiletechnique for material characterisation and thestudy of dynamics at the atomic scale. Besidesthe use of photons to excite vibrational androtational transitions (e.g. in the IR- andRaman spectroscopy) the study of theinteraction between neutrons and matter gainsincreasing importance. Inelastic neutronscattering (INS) excites transitions betweenvibrational levels and, thus, allows to study themotions of atoms and molecules. Theadvantage of INS over IR- and Ramanspectroscopy results from (i) the absence ofselection rules, (ii) the spectral range from 5 to500 meV (40 to 400 cm-1) can be observed,(iii) optically opaque samples are typicallyalmost transparent for neutrons, (iv) the largeincoherent cross-section of protons make INSvery sensitive for hydrogen atoms, and (v)frequencies and intensities can be calculatedwith high accuracy, which supports theinterpretation of the spectra obtained.Examples in the field of catalysis range frombulk samples such as Raney- metals, overhighly dispersed oxide supported metalclusters, (micro-) porous metal oxides, metalsulfides and carbons. The use of (partially)deuterated molecules or the use of D insteadof H should allow the identification of reactivespecies and the differentiation between thereactive and non-reactive intermediates.

Special Sample EnvironmentAs most of the samples will consist of highsurface area powders, specially designedreaction cells have to be developed, whichshould allow the complete removal of waterand other contaminants such as hydrocarbonsbefore the measurements. Aluminum, stainlesssteel and quartz can be used as materials forthe in situ reaction cells.

The successful description of catalyticprocesses requires that the catalyst ischaracterized in direct contact with thereactant/product molecules under conditionsclosely related to those of the reactions withrespect to solvent, pressure and temperature.Therefore, in most cases it will be necessary tointercept the reaction by rapid cooling(quenching), which has two advantages: (i) thewidths of the bands in the INS spectra will bereduced and (ii) the reaction will be interruptedand, thus, the reaction kinetics can be

uncoupled from the time required to collect thespectra. This will allow to follow rapid kineticreactions, because the time necessary tocollect a single spectrum will be in the order of60 minutes. The inherent disadvantage of thisapproach is that the reaction once stopped,cannot be restarted to observe a later timeframe. Therefore, the development of asystem, which allows collecting samples atgiven time interval and to store them for thelater (asynchronous) analysis by neutronscattering has to be developed.

Design of a Catalysis Spectrometer at theFRM-IIThe study of vibrational states needs tomeasure energy transfers of about 500 meV,what, at the FRM-II, is most advantageouslydone with hot neutrons from the Hot Source infront of beam tube SR-9. This beam tube hastwo channels ( 8 cm x 12 cm channel width),one of which (SR-9b) is already used forHEIDI, a hot neutron single crystaldiffractometer. The second channel (SR-9a)can be used for the catalysis instrument, butthe locality is confined by the neighbourinstruments. INS needs energy determinationof the neutrons before and after scattering,what is done normally either by Bragg reflexionfrom crystalline materials or by the Time-of-Flight (TOF) technique. Out of the manifold ofpossible spectrometer types we have studiedthe feasibility of a TOF spectrometer, of acrystal monochromator Be filter instrument andof a spectrometer with a PG graphite analyzer.

The Time-of Flight spectrometer design was aTOSCA- like instrument [1] with a fast chopper,a super mirror guide for hot neutrons (m = 3.5or even m = 4) and a beryllium filter analyzer.The length of the guide to an external neutronlaboratory was 30 – 40 m , the pulse-widthabout 100 µs for an energy resolution of about5 % for primary neutrons of 300 meV. Chopperrotors of a mass of 20 – 40 kg had to berotated with typically 10 000 rpm. The relativeenergy resolution varied between 4% and 7 %for a primary neutron energy between 200 and600 meV, the intensity of the instrument wascomparable to a crystal monochromator andberyllium filter analyzer instrument of about thesame resolution. Because of the hightechnological challenge and the resulting highcosts the TOF-spectrometer was rejected.


As an alternative a crystal monochromatorinstrument in the experimental hall of thereactor building was studied. The distancefrom the hot source to the monochromatorposition at beam tube SR9-a is 8 m, allowing amaximum primary vertical divergence of about1° and a horizontal divergence of about 0.5°.The maximum illuminated area of amonochromator can be about 22 cm x 12 cm,which can be focused down to about 7 cm x 4cm at the sample. In such an „open“configuration one gets a high intensityinstrument with a poor energy resolution ofabout 6 % to 12 % for neutrons with an energybetween 200 meV and 600 meV. A betterenergy resolution could be obtained by use ofnarrow Soller collimators in combination withlarge monochromator angles. High take-offangles are, however, not possible because ofspatial limitations in the horizontal plane.

Fig. 1: Sketch of the catalytic applicationspectrometer

To overcome these limitations another designis proposed as shown in figure 1. Thescattering plane at the monochromator isturned from the horizontal into the verticalplane. Neutrons from the beam tube areentering the shielding block (1) and arereflected by a monochromator crystal, whichcan be rotated around the horizontal axis (2).The monochromatic neutrons are reflectedthrough a radial channel in the shielding block(3), which can rotate also around the horizontalaxis (2) in a theta-two-theta coupling to themonochromator crystal. The instrument is thusoperated with variable primary energy, whichcan be varied by rotation of themonochromator and the shielding block. Thesample is fixed to the rotating shielding block(3) at the exit of the channel. Scatteredneutrons are analyzed by pyrolytic graphitecrystals, which are arranged in two rotational

ellipsoids (4) with a horizontal axis. Such ananalyzer, typically 70 – 80 cm long with 40 –50 cm diameter, was proposed in the„Millennium Programme“ at ILL by Ivanov [2]and its principle is shown in figure 2. Thesample is located in one focus and the detector(5) in the other. Direct radiation withoutreflection at the graphite crystals is blocked byan absorber. Second order reflections from thegraphite can be eliminated by a Be filterwindow at the entrance of the ellipsoids. Theimportant advantage of such an analyzer is anunusually big acceptance angle of nearly 3Steradian per ellipsoid. These analyzers arecoupled to the rotating shielding block (3). WithPG crystals of ~3° mosaic spread an energyresolution of ~2 % for energy transfers of 100meV can be obtained.

The last proposal is certainly the mostattractive one as it is a high resolution and highintensity instrument, which fits well into thecrowded surrounding. Nevertheless it is achallenge as that type of analyzer has not yetbeen realized before.

Fig. 2: Ellipsoidal energy analyzer of scatteredneutrons [2]

References:

[1] D. Colognesi et al., Appl. Phys. A 74 (Suppl.)S64 – S66 (2002)

[2] A. Ivanov: „Beryllium Filter Spectrometer on IN1in the Millennium Programme of ILL“ (2000)


1.15 MIRA -- Very cold neutrons for new methodsR. Georgii1, N. Arend1,2, T. Hils2, P. Böni2, H.Fußstetter1, G.Langenstück1

1Technische Universität München2Technische Universität München, Physik-Department, E21

A beam line for VCN has being built at the newneutron source FRM-II. Situated in the neutronguide hall at the end of a curved neutron guidelooking to the cold source, MIRA will operatebetween 7.7 Å and 30 Å. The beam line isdesigned to develop, test and demonstratenew methods to use very cold neutronstherefore the instrument will be equipped withdifferent instrument options.

The instrument consists of a neutron guide (1cm x 12 cm) with a shutter in the neutron guidebunker. The monochromator mechanics issituated at the end of a 7 m long curvedneutron guide (R = 84 m) inside the shielding.The differential flux will reach a maximum atthe wavelength λ = 8.5 Å and will be around4.107 neutrons/(Å cm2). A mica (Phlogopite)monochromator with d200 = 9.9 Å and amosasicity of 0.9° will be used in thewavelength range from 7.7 Å, the cut-off of thecurved neutron guide, up to 14 Å. A ∆ λ/λ inthe range of 2-8 % will be obtained, dependingon the choice of the divergence of the beamsbefore and after the monochromator. Forwavelengths of 15 Å, 20 Å and 30 Å multilayermonochromators on glass or silicon substratesare foreseen. The multilayer monochromatisesthe neutrons with a ∆ λ/λ of about 2 – 10 %depending on the bandwidth of the multilayersequence. A low-band pass mirror in front ofthe monochromators can be brought into thebeam for suppressing the long wavelengthcontributions (from the regime of totalreflection), which would otherwise contaminatethe beam after monochromatisation with amultilayer monochromator. The 4monochromators can be individually rotatedinto the beam. The whole monochromator unit,with different tables for rotation and x-y-ztranslation, an aperture for the incomingneutrons and a monitor counter is surroundedby the shielding. It consists of 10 cm lead and25 cm concrete ρ = 4.7 g/cm3 with a 10 mmneutron absorbing layer of B4C (see figure 1).

During 2002 the design of the instrument wasfinished. The neutron guide with the shutter inthe bunker was installed and is now ready forroutine operation. The shielding was designedand manufactured at Swiss Neutronics. Theshielding is now installed at its final position inthe neutron guide hall. The mechanics andcontrol electronics for the monochromator iswaiting for the installation into the shielding.The mica monochromator is finished and wascharacterised with neutrons of 5 Å at PSI in

Switzerland. The data evaluation isproceeding.

Fig. 1: The shielding for the monochromator ofMIRA

The very cold neutron reflectometer optionof MIRA

One of the instrument options is a classicalreflectometer in horizontal geometry (seefigure 2). Such a reflectometer for very coldneutrons has the advantage over most existingreflectometers that the angles of reflection arelarger and alignment errors near to the criticaledge are less critical, yet the accessibility ofthe phase space is comparable toreflectometers with cold or thermal neutrons.


Fig. 2: MIRA in the reflectometer option with itsneutron guide

After monochromatisation in the shielding theneutrons will enter a vacuum tube and hit asample. Neutron mirrors above and below thebeam for the transport of the verticaldivergence from the monochromator to thesample are foreseen to be installed in thistube. In the beginning the instrument willmainly operate as a classical reflectometer,with a θ−2θ scan for a vertical samplegeometry. It will reach a q-transfer of 7.0.10-4

Å-1 < q < 1.1 Å-1 within the wavelength range ofthe instrument. Apertures will be used to adaptthe wavelength spread of the beam to thebeam divergences from the differentmonochromators. The reflected radiation fromthe sample will be detected in the detector unit,which will be a 3He-counter. Later the use of a2-dimensional position sensitive detector isplanned.

For using the reflectometer with polarisedneutrons a multilayer polarisation filter shortlyafter the monochromator can be brought intothe beam. After the sample there will be asupermirror-based analyser. There will be alsospin-flippers before and after the sample. Theidea is to perform specular and non-specularpolarised neutron reflectometry in the classicalway, i.e. measurement of the four moduli of thereflection matrix [1]. For this only 4 reflectivitiesneed to be measured, namely R++, R+-, R-+,and R--, where the + and - refers to thepolarisation of the beam before and after thesample.

In 2002 the design for the reflectometer optionwas finished and a performance analysis of theoption was performed. First components,especially of the sample table have beenordered and are expected to be installed insummer 2003.

The Multiple SANS Option

Normal SANS machines are, due to the smallangles being measured, relative long and needtherefore long measurement times since thesample is seen from the detector only under avery small solid angle. Using a setup as shownin figure 3 two-dimensional small anglescattering with several partial-beams can beperformed. For special cases (q-transfer) thismight allow measuring with higher intensitiesand smaller machines as standard SANS orcould be used as resolution enhancementoption at a conventional SANS. In this projectthe main properties like resolution and intensityof a setup consisting of two periodicalmultihole-apertures (typical Øhole 1 mm, d =2,5mm) that lead to an incoherent superposition ofa multitude of "little (ultra-high-resolution)SANS machines" are derived. The number ofsuperpositions results from the product ofilluminated aperture-holes. The basic ideatherefore is to compensate the intensity lossdue to the small apertures by incoherentaddition of many partial solid angles. Theworking principle of such a device will betested at MIRA and a comparison toconventional SANS is planned. In 2002 thebasic design for this option was completed and2 prototypes of the multihole-apertures weremanufactured.


Fig. 3: The multi-SANS option

Multi level-MIEZE setup

This option of MIRA will be used for theinvestigation of longitudinal coherenceproperties of neutron beams. This can beachieved by using the multi level-MIEZEvariant of the Neutron Resonance Spin Echo(NRSE) technique [2] with VCN.

Fig. 4: Scheme of a multi level-MIEZE

The longitudinal coherence properties of thetwo neutron spin states can be studied by avery short temporal modulation of thepolarisation of the neutron beam. Stacking ofseveral MIEZE setups with a common detectorposition results in a signal, which is the productof the sinusoidal signals from all individualsetups, see fig. 4. A time resolution of 10-7 ispossible and the respective coherence lengthwould be of the order of cm. It should bementioned that the multi level-MIEZE principlehas not been verified so far and shall beverified with this setup of MIRA.

Crucial parts of the instrument are the spinflippers, which will be constructed as bootstrapcoils [3]. They must be suitable for thewavelength range of VCN. Since the neutronbeam has to penetrate the coils, interaction(scattering, absorption) with the coil windings isa serious problem, especially in the cold

neutron range. Therefore material testmeasurements to find new adequate materialcombinations have been performed at theSANS machines at PSI.

Fig. 5: Signal of a 4-level multi MIEZE setup

First evaluation of the data shows that thecommonly used material for bootstrap flippercoils (Al2O3 coated Al band) will mostpresumably cause problems due to relativelyhigh absorption and scattering features fromthe isolation layer.

To optimise the design concerning heatproduction, magnetic field homogeneity, etc,the coils are being simulated using finiteelement methods. Furthermore efforts on ray-trace simulations using McStas are planned.

[1] G.P. Felcher, Physica B, 198, 1595 (1981)[2] Scattering and Inverse Scattering in Pure and

Applied Science, edited by Roy Pike, PierreSabatier. pp. 1264-1286, San Diego, CA,Academic Press (2002)

[3] R. Gähler and R. Golub, J. Physique, 49:1195(1988).


1.16 Testing the shielding of the monochromator of Mira by Monte CarloMethod

Nico Wieschalla1, Florian Grünauer1, Robert Georgii1,2

1TU-München, Physics Department, E 212TU-München, ZWE FRM-II

Munich´s new research neutron source FRM-2will provide a variety of different neutronbeams for many different applications. One ofthese applications will be ´Mira´ - a station forexperiments with cold neutrons.

At the entrance of this station the neutronbeam will have a Maxwell energy distribution ofabout 40K. The neutron flux will be

2.5·108 sec -1cm-2 at this position. For mostexperiments it is necessary to have amonochromatic beam. The wanted energy willbe selected from the ´white´ incoming beam byBragg reflection in a phlogopit crystal. A Li-6Beamcatcher will absorb neutrons, which passthis crystal.

Fig 1: horizontal cut on beam level

In addition a large amount of neutrons will bescattered or absorbed (causing secondaryradiation) in structure material of theexperiment. The resulting unwanted radiationmust be attenuated by a shielding, in order tomeet radiation protection purposes and toreduce background radiation for otherexperiments.

Fig 2: dose rate at the outer detector fordifferent heights. The beam level is on 120 cm.

The efficiency of this shielding was estimatedby Monte Carlo Method, considering neutronsand generated gamma radiation. The geometryof the shielding is shown in fig. 1. It consists oftwo layers. The inner one is made of boroncarbide. Neutrons are captured in boron by a(n,α)-reaction. This reaction is accompanied by0.48 MeV gamma radiation. This energy israther low compared to gamma radiation fromneutron capturing in other materials like iron,aluminum or hydrogen; attenuation of theproduced gamma radiation is therefore mucheasier. Nevertheless it is necessary to have agamma shielding as second layer. In our casewe use lead for this purpose, because of itshigh atomic number. This layer will be 10 cmthick and it will be surrounded by a steel liner(wall thickness: 0.6cm). At FRM-2 the doselevel outside shieldings should not exceed3µSv/h.

The simulated horizontal dose distributionoutside the shielding at different heights is


shown in fig.2. The beam level is on 120 cm.For this simulation an angle of 120° betweenbeam entrance and exit was assumed. Ofcourse the highest dose values are reached atthe beam entrance and the beam exit.

Fig 3: maximum dose rate for different anglesof the beam exit

However the values at this locations are aconservative estimation:

• In reality there is an additional shieldingsurrounding the incoming beam, whichwas omitted in the Monte Carlo modell.

• The wideness of the beam after the Braggreflection was assumed to be bigger than itcan be expected in reality. Therefore moreneutrons hit the steel liner that surroundsthe exit channel. The resulting gammaproduction by neutron capturing in iron isincreased by this effect.

Even in this last case, the dose peak can belimited by covering the steel structure of thebeam exit by boron carbide layer (thickness0.5cm)

Another problem appears when the beam exitis at 190°. A direct view from the source to theexit is possible due to the divergence of thebeam. Now the neutron flux in the exit is muchhigher and the gamma production too, due toincreased neutron capture in iron. Therefore itis not advisable to work at this angle neither formonochromating the neutron beam nor for thedose rate (see fig. 3).

Also this shielding was compared with anotherone: The boron carbide of the first layer of theshielding was replaced by Li-6. This neutronabsorber produces no gammas by neutroncapture, but the dose rate at the outer detectorwas exactly the same compared to boroncarbide. This behaviour shows, that theinfluence of the outer steel liner of the shieldingis dominant for the dose rate outside theshielding. Therefore it is not necessary to useLi-6 (much more expansive compared to boroncarbide) for neutron capturing inside theshielding.

Another interesting point was the estimation ofthe neutron flux dependance on the thicknessof the crystal. Figure 4 shows the neutron fluxat the inner detector (fig. 1) for differentthickness and without the crystal.

Fig 4: Neutron flux at the inner detector fordifferent thickness of the crystal

Most neutrons were scattered by a crystal of athickness of 10 mm. However the difference tothe other investigated thicknesses are ratherlow: The thickness of the crystal can beoptimized without regarding radiationprotection purposes.

At least we can say, that the built shielding is agood protection against the radiation.


1.17 RESEDA Spectrometer: Jahresbericht 1.1.-31.12.02M. Bleuel

project team: TU München E21project leader: Dr. R. Gähler,instrument responsible: M. Bleuel

Activities in 2002:• Continuous cabeling of the instrument (Motors, air cushions, air sensors, water, DC- and HF-

supply for the Bootstrapcoils)• Additional detector for MIEZE is in development• Further tests with the C-box for the resonance-HF-circuit leading to an improver resonance-HF-

circuit with a ceramic ring ferrit• Continuous programming on the instrument-control-program (component_test_ panels, automated

adjustment of the coils and the spectrometer, test of parameters before execution)• True up of the NSE-Z-Coils in the spectrometer• Winding of additional B0-Coils and test for the best• Further development on the multiplexer-platine together with the Elektronik-Labor• Up-building of the guide field for the N-guide together with Christian Schanzer (4m are at the final

position)• Further tests of the Mu Metallshielding at the final position, model calculations of the shielding

factor, first tests of the automated demagnetising switching mechanism• Three-dimensional modeling of RESEDA and surrounding installations with the construction

program solid works• Conception of the „Selektorburg“ and „Meßkabine“• Detailed simulations of the neutron beam with the Program Mcstas-1.6• Removement of damage done to the spectrometer, as much as possible• Sergey Prokudaylo left the RESEDA-team


Actual project state (12.02):The estimated start of the FRM2 will not be before middle of 2003. Because of this delay of at least 2years compared to the early time table, all RESEDA activities are streched. It is important to keep theknow how and not to break up the team.

DoppelterMu-Metall-Schild

Glasleiter

NSE-, NRSE-Spulen undBlenden

Flansch undPositionier-einheit für dieSpektrometer-arme

Entriegelung


1.18 MUPAD Spectrometer: Annual Report 1.1.-31.12.02M.Schulz, M. Janoschek

project team: TU München E21project leader: Dr. R. Gähler,instrument responsible: M.Schulz, M. Janoschek

Project Description:MUPAD is a Spectrometer made for 3-D polarization-analysis with polarized neutrons, e.g. formeasuring magnetic lattices of matter or magnetic excitations in matter. It uses two specially designedflipper coils with nearly no outer stray fields to turn the spin of neutrons in arbitrary directions in front ofthe measured sample and two after it for analyzation. Our goal is that the final setup of the instrumentwill make less than 1° turn error with all four coils together.

Activities in 2002• First of all Sergey Prokudaylo performed some simulations of the coils in order to know whether

the design for the coils should implement active or passive guides of the stray fields. It turned outthat passive guides constructed out of mu-metal would be the best solution.

• The design for the first prototype of coils were made. The coils were build and measured. Themeasurement told us, that the coils were not meeting our guidelines. We decided to build morecompact coils, which were winded with lot of tension. Additionally we enhanced our coils by usingsecondary outer mu-metall-screens out of thin foil, and by using a lot of time on demagnetizationof all the mu-metal-parts of our design.

• With that design we could prove that our coils would match our given guidelines. Here are the fieldintegrals in each direction in front and behind our coils and the from experimentally data calculatedturn errors for 1 Angstrom neutrons (Measurements made with coil current of 1.8A correspondingto B=18Gs inner field which is a turn of 180° for 1 Angstrom neutrons). Our results showed thatthe sum of outer field integral is smaller than 1/500 of inner field integral!!!

• Also our measurements and experience with demagnetization showed us some ways for furtherimprovement of our design. The outer secondary screens will be made thicker. Also we came upwith a new idea for using the connections of the coil to enhanced the field. We think that we canimprove our results again by factor two which means that that the sum of outer field integral will besmaller than 1/500 of inner field integral.

Outlook for 2003In the next year we will hopefully build and test our new designed of coils. We are looking forward tointegrate them in the instrument.

Direction x y zapprox.

Field-Integral 47,9 -84,995 12,43[mGs cm]

turn error [°] 0,063 -0,113 0,016

Direction x y zapprox.

Field-Integral -101,655 86,445 90,805[mGs cm]

turn error [°] -0,135 0,115 0,120


1.19 Multiple Small Angle Neutron ScatteringT. Hils, R. Gähler

Technical University of Munich, Physics Department E21

Measuring density correlations in the µm rangeis a domain of light scattering, however it oftenfails due to low contrast or opacity of thesample. Research on polymers, colloidsystems, cements, microporous media, areexamples of a rising field, where µmcorrelations play a crucial role. Small angle X-ray and neutron scattering (SAXS and SANS),on the other hand, commonly measures lateralcorrelation lengths in the 0.01 to 0.1 µm range.To measure µm correlations with neutrons,various specific instruments have beendesigned, and the technique is commonlyknown as USANS (ultra small angle neutronscattering)1. However these methods aresensitive to scattering only in one dimensionand suffer from intrinsic small angle scatteringdue to structure material in the beam.

Here we propose MSANS (Multi hole SANS), anew USANS option for a standard longbaseline SANS instrument on pulsed or steadystate neutron sources. It uses the commonSANS infrastructure except for the detector,which requires enhanced spatial resolution.The system is sketched in Fig1.

In a first attempt the main properties likeresolution and intensity of a setup consisting oftwo periodical multihole apertures that lead toan incoherent superposition of a multitude of"little (ultra high resolution) SANS-machines" isto be derived.

⇒Neutr.⇒

⇒

Fig. 1: For MSANS, the entrance- and sample apertures ofa common SANS instrument may be replaced by multi-hole apertures. With proper lattice constants of theseapertures, superposition of individual patterns at thedetector can be achieved. The q-resolution is determinedby the size of individual holes, the q-range is determinedby the distance between holes.

1 Instead of the lateral correlation length, often the q-resolution is used, which is in the order of 10-3 to 10-5 Å-1

for USANS.

It is based on multi-hole apertures at theentrance (Me) of the collimator and near thesample (Ms) with lattice constants ae and asand hole diameters de and ds respectively.With the choice

( ) esdesese

se GGGLGGLGa

G −=−=⋅= ;;2

21,

,π

an intensity pattern of well separated peakswith lattice constant ad in the detector plane isobserved (ad=2π/Gd). The factor 2π is addedfor convenience. The equations holdsseparately for x and z-direction, i.e. the latticeconstants in both dimensions can be chosenaccording to the requirements. The validity ofthe above equations may be directly seen fromFig. 2.

Fig.2: Condition for the lattice constants ae and as anddistances L1 and L2 for ‘focussing’ in the detector plane d.

Intensity and q-resolution:The n-flux Φs at the sample can be estimatedas follows:

ess0

2

s TTdd

d Ωλ∆Ωλ

Φ=Φ

where the first factor is the double differentialflux of the entrance guide (assumed asisotropic), ∆λ the mean wavelength bandtransmitted by the selector, Ωs the solid angleof the entrance aperture seen from the sampleaperture and Te,s are the transmission factorsof both apertures Te,s = de,s

2π/4ae,s2. Ωs is

assumed to be smaller than the beam solidangle in the n-guide. For optimum spatialresolution at the detector the hole diameters∅e,s and lattice constants ae,s should be relatedby

21

2

e

s

e

sLL

Laa

dd

+==

which follows directly from Fig. 2. In this casethe peak shape in the detector plane can bederived from the convolution of two roundapertures of equal size. The peak shape can

N

L1 L

Eingangs-Lochmaskeca. 300Lochblenden bei14 ×7 cm2 (h×w)Strahlquerschnittde = 2 mmae ≥ 5 mm

Abbildung am Detektor,wenn:Gitterkonstanten

Proben-Lochmaske25Lochblendenbei1.5 ×1.5 cm2

(h×w)a

lateraldistance

ad

ad

asae

1

21

s

d

LLL

aa +=

1

2

e

d

LL

aa

=

d

L2L1

With G = 2π /a ⇒ Gd = Gs - G e

GeL1 = (Gs - Ge)L2


be approximated by a triangular function ofwidth

)FWHM(;dL

Ld e

1

2d ≅

and both transmission factors Ts and Te will bethe same (T). The relation ae/2de (or as/2ds)

determine T, and also the dynamic range, i.e.the max. q-value divided by the q-resolution,without overlap between neighboring peaks.Short range correlations in the sample maylead to significant overlap, however typicalSANS intensities drop very rapidly withincreasing q, and overlap will not be fatal inmany cases. Sets of apertures with differentrelations ae,s/∅e,s can be used to adapt thepattern to the demand. In MSANS, resolution isdecoupled from intensity, as long as T is keptconstant. The increase in q-resolution inMSANS is typically one order of magnitude,compared to SANS at equal intensity. The gainoriginates from the reduction in q-range inMSANS and the increase of the input guidecross section and its divergence.

Diffraction from the aperture holes of typically1mm are not yet crucial, as the beamcorrelation length is only in the µm range.

In 2002 a basic test-design for this option wascompleted and 2 prototypes of the multihole-apertures were manufactured.


2.1 The design and construction of the neutron tomography facilityANTARES

B. Schillingera,b, E. Calzadaa, F. Grünauerb

a) Technische Universität München, FRM-II, 85747 Garching, Germanyb) Technische Universität München, Physik Department E21, 85747 Garching, Germany

The neutron tomography facility ANTARES will have an unprecedented combination of high flux andhigh beam collimation with two selectable beam-adjusted collimators. The space for the installationwas assigned very late during the planning of the new reactor FRM-II. This lead to several harshconstraints due to surrounding experiments, e.g. the site is not accessible by crane. We describethe engineering solutions as well as the current progress in the construction.

The facilityIn the early planning stage of the reactor, thetomography facility was foreseen at a neutronguide leading out of the reactor hall to anexternal building.

When it became clear that the beam geometryof a neutron guide was unsuited fortomography, a classical flight tube design hatto be fitted into the remaining space in thereactor hall. The tomography facility at FRM IIshares the beam port (with two channels) withthe UCN source and is surrounded by a spin-echo spectrometer and the platforms of thepositron source at the inclined beam tubeabove its beams tube (Fig.1a). The UCNsource will insert a nozzle through the secondchannel of the drum shutter in the biologicalshielding, so an additional secondary shutteroutside the reactor wall is required.

Fig. 1.: Tomography facility surrounded byother experiments

Shutter and collimatorThe beam geometry resembles a classicalpinhole camera imaging the square sourcearea onto the sample area. A special beam-adjusted collimator with a cross section withvarying shape follows exactly the contours ofthe pin-hole imaging of the source area. Thiscollimator also limits the penumbra region of

the beam. The first part of this collimator isintegrated into the drum shutter in thebiological shielding, the second part consists oftwo separate channels above each other insidethe vertical secondary shutter (Fig.2). Withthese two collimators, the beam can beadjusted for high resolution (L/D=800, 3x107

n/cm²s) or high flux (L/D=400, 1.2x 108

n/cm²s). The secondary shutter is driven by aregulated hydraulic positioning system with apressure storage and fail-safe mechanism thatshuts the beam off even in case of powerfailure.

Fig. 2: Cross section of the vertical beamshutter

Flight tube and shieldingThe collimator feeds a divergent 12m longflight tube. It consists of six sections made ofaluminum with square cross section (Fig.3) .Each section can be removed in order to installinstruments, for example a velocity selector.The tube size and geometry are adapted to thebeam generated by the collimator, thereforethe penumbra region of the beam does nottouch the beam tube wall, avoiding excessivegamma generation in the aluminum walls.

Fig. 3: The flight tube.


The flight tube is shielded by L-shaped wallsand a roof made of heavy concrete-filled steelcontainers (Fig. 4a). As the access by crane isblocked by the positron source platform, thewall elements have to be positioned by aircushions, the roof components will bepositioned onto the walls in the smallaccessible space between the platforms andwill be rolled on ball bearings into their finalposition (Fig. 4b).

Fig. 4a,b: The flight tube shielding.

The neutron beam enters the measurementcabin through a square hole of 400 mm x 400mm, the penumbra being absorbed by the wall.The walls consist of several segments with athickness of 800 mm and less than 10 t weightwhich will be stacked vertically in the craneaccessible area. The walls will then bepositioned by air cushions.

As there is no space for a conventional door toswing open, a whole 20 t wall section will bemoved on rails parallel to the UCN experiment(Fig.5).

Fig. 5. Measurement cabin

Manipulator and beam size limiterThe motor driven sample manipulator (Fig.6)allows for samples of up to 500 kg and 1 mdiameter to be rotated and translated in heightand width across the neutron beam.

FIG. 6: SAMPLE MANIPULATOR

The maximum beam cross section is 400 mm x400 mm. It can be reduced by a variable beamsize limiter (Fig.7), installed at the end of theflight tube outside the cabin. It consists of aaluminum frame with four sliding platescovered with BorAl layers. A fast pneumaticbeam shutter with B4C, situated at thebeginning of the flight tube, captures thethermal flux in intermissions betweenmeasurements (e.g. for data transfer time) tokeep the activation of the sample as low aspossible.

Fig. 7: Variable beam size limiter

THE COMPLETE LAYOUTThe complete layout of the facility is explainedin Fig.8.

Fig. 8: Complete layout of the facility

Collimator inside of the biological shielding External vertical beam shutterIris diaphragmPneumatic shutterFlight tubeVariable beam size limiterMeasurement cabinWall and beam stopDetector and camera systemSample manipulatorSliding doorFlight tube shielding


Status in January 2003In January 2003, the status was as follows:The primary collimator has been installed inthe drum shutter, the secondary shutter insertsare near completion (Fig.9).

The sample manipulator and control arecompleted, cabling and programming are inprogress (Fig.10).

The beam size limiter is almost completed,waiting for the Boral plates to be mounted(Fig.11).

All shielding walls except for the secondaryshutter are completed and are beingtransported to an assembly hall to be filled withheavy concrete in the fully assembled state(Fig. 12,13,14).

The camera system (2048 x 2048 pixels) forthe detector has been delivered and awaitsfirst tests.

The hydraulic control system is being deliveredin parts for assembly with the secondaryshutter.

The mechanical mounting of the completefacility will be finished by March 2003, withelectrical installations to be completed.

Fig. 9: The collimators

Fig. 10: The sample manipulator

Fig. 11: The frame of the beam size limiter

Fig. 12: Part of the flight tube shielding andentrance to the measurement cabin

Fig. 13: The beam entrance of themeasurement cabin


Fig. 14: A side wall of the measurement cabinmade from stacked segments

References[1] B. Schillinger: Estimation and Measurement of L/D

on a Cold and Thermal Neutron Guide. World Conf.Neutron Radiography, Osaka 1999, in: Nondestr.Test. Eval., Vol. 16, pp.141-150

[2] B. Schillinger, E. Calzada, F. Grünauer, E.Steichele: The design of the neutron radiographyand tomography facility at the new research reactorFRM-II at Technical University Munich, 4th

International Topical Meeting on NeutronRadiography, Pennsylvania 2001, acc. pub Journalof Radiation and Isotopes, 2002.

[3] F. Grünauer, B . Schillinger: Optimization of theBeam Geometry and Radiation Shieldings for theNeutron Tomography Facility at the New NeutronSource in Munich. 4th International Topical Meetingon Neutron Radiography, Pennsylvania 2001, acc.pub Journal of Radiation and Isotopes, 2002.

[4] E. Calzada, F. Grünauer, B. Schilllinger,“Engineering solutions for the new radiography andtomography facility at FRM-II“, World Conf. NeutronRadiography, Rome 2002


2.2 Steps toward a dynamic neutron radiography of a combustion engineJ. Brunnera, G. Freib, E. Lehmannb, B. Schillingera

a) Technical University of Munich, Physics Department E21, 85747 Garching, Germanyb) Paul Scherrer Institut,5232 Villigen, Swizzerland

The Dynamic NR is a new established technique, which is applied very successful for non destructivetesting for time dependent phenomena with a similar resolution and a similar contrast as inconventional NR. Of special interest is the investigation of the injection process in a runningcombustion engine under real conditions .

For obtaining a 16 bit conventional radiographyimages one has to integrate scintillator light onthe CCD chip typically for some seconds. Ofcourse the sample thickness, the samplematerial as well as the neutron flux of thesource determines the exact time. Additionallythe readout time, depending on the dynamicrange and the pixelnumber of the CCD puts alower limit to the highest reachable framerate.The observation of objects in the millisecondtime scale is impossible with standardmethods. A possible solution for repetitiveprocesses can be stroboscopic imaging. Thatmeans a synchronization of the detector (theCCD) with the repetitive process and anintegration of 1000 images taken at the samemillisecond timewindow of the cycle. That waya virtual opening time of some seconds can berealized as long as the noise is constant withintime (i.e. negligible readout noise) and notdeterming the system.

With this method a combustion engine towedby a electric motor was observed in the firstexperiments testing the method with aselfconstructed detector. The reached spatialresolution was 1mm.

3 cm

3 cm

Fig. 1: A single frame (left) contains only 10graylevels and 500 summed NR-imagescontain 3000, image after processing (right),t=4ms.

This year it was possible to perform dynamicneutron investigations at PSI with an excellentnew detector system and at ILL at a very highflux test beam line with a detector built by agroup of the University of Heidelberg.

At the NEUTRA facility a dynamic neutronradiography experiment of a running modelaircraft engine was performed. The detectorconsisted of a CCD camera using a multi

channel plate as image intensifier coupled onthe CCD-chip with fiberoptics.

With this equipment and the thermal neutronbeam of 107 n/cm2s an extremly high detectionsensitivity was possible hardly reached by anyother system. As neutron to light converter aLi6F/ZnAg scintillator of a millimeter thicknesswas used. The reached spatial resolution was0.15 mm.

3 cm

Fig. 2: 4000 summed frames of the neutronradiography movie of a running model aircraftengine (1 PS), one frame = 0.25 ms, running at4800rpm

At ILL NR-measurements on an air brushnozzle with a very high flux of 109 n/cm2s wereperformed. In the false color image Fig. 3 thered cloud of injection liquid is clearly visible.Boron water and a Gadolinium emulsion weretested as a contrast liquid. Fig. 3 was obtainedwithout synchronizing with the detector.


3 cm

Fig. 3: A false color NR-image of boron waterinjection of an air brush nozzle, t=20ms, on thetop of the image is the nozzle, the injection isthe red strip. The spatial resolution is 0.5 mm.

References[1] B. Schilllinger, “Neue Entwicklungen zu

Radiographie and Tomographie mitthermischen Neutronen“, Doktorarbeit an derTechnischen Universität München

[2] M. Balasko, “Neutron radiography visualizationof internal processes in refrigerators”, PhysicaB: Condensed Matter, Volumes 234-236, 2June 1997, Pages 1033-1034


2.3 Phase contrast radiography using thermal neutronsN. Kardjilov 1, E. Lehmann 2, E. Steichele 1, P. Vontobel 2

1 Fakultät für Physik E21, Technische Universität München, D-85747 Garching, Germany2 Paul Scherrer Institute, CH-5232 Villigen, Switzerland .

An alternative method to the standard radiography is the phase contrast imaging, where the intensitydistribution in the radiography picture is proportional to the phase variations induced by thetransmission of a coherent radiation through a medium. To achieve a phase contrast in radiographyimages usually a high spatial and chromatic beam coherence is required. It can be shown that aneutron phase contrast imaging can be performed without monochromatisation of the beam.

In case of high transversal spatial coherenceall rays emanating from a point source, definede.g. by a small pinhole, are associated with asingle set of spherical waves that have thesame phase on any given wavefront. As shownin [1] the variations of the thickness and theneutron refractive index n of the sample causea change in the shape of a neutron wavefronton passing through the sample. The angulardeviation of the normal to the wavefront, ∆α,can be expressed as:

∫ −∇=∇≈∆ |]1),,([||),,(|1,, dzzyxnzyx

k yxyx ϕα , (1)

where the optical axis is parallel to z and ϕ isthe phase change for a ray path through anobject relative to vacuum. This means that thedeformation of the wavefront after thetransmission depends on the gradient of therefractive index n in a plane perpendicular tothe direction of the wave propagation, k

r.

polychromatic beamsample

loss line

I

k1>k>2 k3

Z

Fig. 1 Schematic illustration of the energy-dependent deflection of a polychromatic beamfrom a sample area with a strong variation ofthe refractive index.

To show that rapid variations in refractive indexcan lead to strong phase-contrast effect evenwith polychromatic radiation, the formalrepresentation shown in Fig. 1 can be used [1].The intensity variations due to the gradient ofthe refractive index in the sample area can becharacterized by a very sharp minimum in thepropagation direction (loss line) followed by abroad maximum beside it, Fig. 1. Such highvalues of the refractive index gradient existmainly on the sample edges. Therefore thephase contrast imaging provides a high edge-enhancement in the resultant images.

The experiments were performed at thethermal neutron radiography setup NEUTRA.

To produce a high spatial coherence a pinholewith a size of 0.5 mm in a Gd foil (thickness of0.1 mm) was used. It was placed at a largedistance of 7 m from the sample. The sampleto detector distance was fixed to 1 m. Theexposure time at the used thermal neutronbeam was 90 min per image.

Steel capillaries with diameters between 0.3mm and 0.9 mm were investigated with phasecontrast and conventional radiography. Theobtained profiles through the capillaries areshown in Fig.2.

5 10 15 20 25 300,9

1,0

0.9 mm

conventional radiography

Nor

mal

ized

inte

nsity

, I/Io

mm

0,9

1,0

0.3 mm 0.5 mm 0.6 mm 0.8 mm

phase contrast radiography

Fig. 2 Intensity profiles of steel capillariesobtained by phase contrast and conventionalradiography.

The edge-enhancement on the needle bordersin the case of phase contrast radiographyfollows the behaviour shown in Fig. 1 – aconsequence of minimum and maximum in theintensity at rapid variations of the refractiveindex. In the case of conventional thermalradiography such an edge enhancement ismissing.

An example of phase-contrast imaging withthermal neutrons of a small toothed wheel isshown in Fig. 3.


1 mm

Fig 3 Conventional thermal radiography of atoothed wheel (top) and phase-contrastradiography (bottom).

The improved contrast in case of phase-contrast radiography is obvious. The observededge-enhancement is due to a very sharpchange of the intensity on the borders.


2.4 Neutron Topography at TOPSIN. Kardjilov1, S. Baechler2,3, M. Dierick4, B. Masschaele4, J. Stahn3

1 Fakultät für Physik E21, Technische Universität München, D-85747 Garching, Germany2 Physics Department, University of Fribourg, Ch-1700 Fribourg, Switzerland

3 Paul Scherrer Institute, CH-5232 Villigen, Switzerland4 Vakgroep Subatomaire en Stralingsfysica, University of Gent, B-9000 Gent, Belgium

The topography technique allows the imaging of crystalline materials using the whole beam diffractedby the individual crystallites. In comparison with the conventional micro beam techniques where only asmall region of the sample is irradiated, in the topography experiments a large beam of the order ofsquare centimeters is used.

To prevent overlapping of the rays diffractedfrom different areas of the sample, which willcause a loss of spatial information, sollercollimators in vertical and horizontal directionswere placed between the sample and thedetector. In that case each of the rectangularcells defined by the collimators lamellae willprovide a diffracted radiation only from a well-defined sample region onto the correspondingdetector element. The advantage of thetopography technique is the exploitation of thescattering information as complete as possiblein reasonable time. The detector system wasoptimized for detecting low signals with aresolution not higher than determined by thesoller collimators and the sample to detectordistance. The used position sensitive detectorwas based on a conventional CCD radiographysystem. The visible light emitted from the6LiF/ZnS(Ag) scintillator was imaged with anoptical lens onto a Peltier-cooled CCD camerathrough a mirror at a 45o angle. Thecollimator/detector arrangement was alignedon the 2θ arm of the two axis diffractometerTOPSI at SINQ, (Fig. 1).

Topography detector

Soller collimators

Figure 1: A view on the two axis neutrondiffractometer TOPSI at SINQ, arranged fortopography measurements.

The results from a topography investigation ofHeusler crystals Cu2MnAl with d111=3.53 Åintended for neutron monochromators areshown in Fig. 2. The measured diffractionspectra are shown together with thetopography images. An example of a well-

orientated single crystal is presented in Fig. 2a. In a small angular interval of ∆ω = ± 0.2o

around the maximum intensity of ω = 42.7o thewhole sample volume gives a contribution tothe Bragg reflection intensity. Opposite to thatcase a topography investigation of a crystalwith a mosaic structure is presented in Fig. 2 b.To obtain a maximal intensity of the diffractedbeam the crystal was tilted at ϕ = -1.8o. Clearlytwo complimentary reflecting regions can berecognized on the images at ω = 43.0o and ω =43.5o. The transition between them is smoothas it can be seen from the diffraction spectrum.An example of an abrupt transition betweentwo crystallites (domains) disorientated by ∆ω~ 1o is shown in Fig. 2 c. The areascorresponding to the two domains can belocated easily on the topography images takenat ω = 42.8o and ω = 43.7o. The agreementbetween the topography images and thediffraction spectra is excellent.


41,5 42,0 42,5 43,0 43,5 44,0 44,5 45,0 45,50

100

200

300

400

500

600

700

a)

ϕ = 0O

2θ = 85.95O

Cu2MnAl 102-4A K7

Cou

nts

ω,o

ω = 42.7o

41,5 42,0 42,5 43,0 43,5 44,0 44,5 45,0 45,50

1000

2000

3000

4000

5000

b)

ϕ = -1.8o

2θ = 85.95o

Cu

2MnAl 104-4 K1

Cou

nts

ω,o

ω = 43.0o

ω = 43.5o

41,0 41,5 42,0 42,5 43,0 43,5 44,0 44,5 45,0 45,5 46,00

1000

2000

3000

4000

5000

6000

7000

c)

ϕ = 0O

2θ = 85.95O

Cu

2MnAl 101-4

Cou

nts

ω,o

ω = 42.8o

ω = 43.7o

Figure 2: A topography investigation of Heusler crystals (right) together with the diffraction spectra(left)


2.5 Influence of gaps in shieldings against neutron radiationF. Grünauer

Technische Universität München, Physik-Department E21

In most neutron experiments it is neitherdesirable nor possible to construct radiationshieldings in one piece. There would ariseproblems with the assembly, the accessibility,and flexibility. Therefore a modularconstruction has to be preferred. The singlemodules cannot be shaped to a precision thatthere will be no gaps between the elements.Gaps of up to some millimeters may occurinside the final shielding construction.

For the new neutron radiography station atMunich's new neutron source (FRM-II), wehave the requirement that parts of the shieldingcan be moved on top of the lateral walls. Gapsbetween these parts and the walls are needfultherefore. It is necessary to estimate theleakage of radiation through these gaps. Theestimation was done by Monte Carlo Method.The neutrons are emitted from an area sourceon one side of the shielding. The dose on theopposite side of the shielding is calculated witha spatial resolution of 2mm. The shielding isinfinite long and high. The gaps are repeatedwith a period of 50cm. The shielding material isheavy concrete (density: 3.5g/cm3) withcolemanite, hematite, and steel resin asadditives.

50cm

detector

y=0cm

y=+25cm

y=-25cmy

x

sourcearea

heavyconcrete

gap

60cm

Fig. 1: Monte Carlo Model for the estimation ofneutron radiation transport through gaps

Three different shapes of gaps were examined(Fig. 2):• Curved• Angular: one bend with an angle of 120°• Labyrinth: two 90° bends, 20cm offsetFor all shapes a gap width of 1cm wasassumed. The shielding elements are coatedby a steel liner of 1cm thickness along the gap.

For comparison two additional ídealćonfigurations were calculated (Fig. 2):• Monolithic: homogeneous, no gaps• Simple brick: coated by 1cm steel, no gaps

Fig. 2: Gap configurations in radiationshieldings: A: monolithic, B: simple brick, C:labyrinth (two 90° bends), D: angular (one 120°bend), E: curved gap

Factors of increase of the neutron dose outsidethe different shielding types compared to themonolithic shielding are shown in Fig. 3 for anisotropic source. The áverage´ values showthe average dose increase in the total detectorarea, and the peak values the increase at thegap mouth.

The curved structure shows the lowest and thelabyrinth shows the highest dose peak at thegap mouth. However, the dose integral in theentire detector area is in all cases 20% abovethe integral dose without gap. The shieldingpower does not so much depend on the type ofgaps.

Increase of dose outside different shielding types with gaps compared to the monolithic case

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

no gap simplebrick

labyrinth angular curved

fact

or o

f in

crea

se average

peak

Fig. 3: Factors of neutron dose increaseoutside the different shielding types comparedto the monolithic shielding for an isotropicsource.


The situation changes completely for a parallelbeam (neutrons enter the shieldingperpendicular). The dose outside of the

different gap systems for a parallel beam isshown in Fig. 4.

Fig. 4: spatial neutron dose distribution outside of shieldings with different gap configurations for aparallel beam (gap width=1cm)

The labyrinth with two 90° bends increases thedose considerably more than all other gapsystems (78% in the total detector area). Twodistinct dose peaks appear at the positions ofthe gap parts, that are parallel to the beam.The bad shielding quality of the labyrinth for aparallel beam is obvious: The transmissionthrough both parallel parts of the gap (gapentrance and gap exit) is enormous.

All other gaps show even better results than forthe isotropic beam.

Increase of dose outside different shielding types with gaps compared to the monolithic case

0,00,5

1,01,5

2,02,5

3,03,5

4,04,5

no gap simplebrick

labyrinth angular curved

fact

or o

f inc

reas

e

average

peak

Fig. 5: Factors of neutron dose increaseoutside the different shielding types comparedto the monolithic shielding for a parallel source.The average values show the average doseincrease in the total detector area, and thepeak values the increase at the gap mouth.


2.6 Estimation of the imaging quality of Munich´s new neutron radiographystation

F. Grünauer

Technische Universität München, Physik-Department E21

The imaging properties of the new neutronradiography and tomography station atMunich´s new neutron source FRM-II wereestimated and optimized by simulation of thewhole facility by Monte Carlo method.

Homogeneity of open beam imageAn important question is the influence of thestructure of the cold source and its surrounding(Fig 1.) with regard to homogeneousillumination of our detector.

other experiments

to neutronradiographystation

reactorcore moderator

(D2O)

cold source(D2; 25K)

displacement body

Fig. 1: horizontal cut through the cold sourceand its surrounding inside the moderatorvessel

The cold source itself is not homogeneous. Adisplacement body without deuterium isincluded inside the cold source. This is tooptimize flux and spectrum, especially for theother two beam tubes, that are connected tothe cold source. In addition, the other beamtubes are arranged in an asymmetric way. Theresult is a flux depression on one side of thecold source, due to missing backscatteringfrom the volumes of the tubes. Theseinfluences are estimated by calculation of openbeam images in the detector plane for differenttheoretical arrangements inside the moderatorvessel using our 4.3 cm aperture:

• a cold source without displacement bodyand without other beam tubes. Thevolumes of the tubes are replaced byheavy water

• cold source with displacement body but noother beam tubes

• the real arrangement.

Neutron flux profiles in the energy range below1eV in our detector plane are displayed in Fig.2 for different theoretical arrangements in themoderator vessel. Without displacement bodya bias in the flux profile occurs, because bothsides of the entrance of our beam tube arepopulated by neutrons that crossed differentthicknesses of moderator. The situation isreversed with included displacement body: the

total flux increases. The resulting bias is nearlycompensated by including the beam tubes forthe other experiments. Inhomogeneities arerather low for neutrons in the interestingenergy range below 1eV, whereasinhomogeneities in higher energy rangesoccur.

neutron flux profile in the detector plane (E<1eV)

6,0E+7

7,0E+7

8,0E+7

9,0E+7

1,0E+8

1,1E+8

1,2E+8

1,3E+8

-25 -20 -15 -10 -5 0 5 10 15 20 25

horizontal distance from beam axis [cm]flu

x [s

ec-1

cm-2

]

no displacement body; no other experimentswithout other experimentsreal configuration

Fig. 2: Neutron flux profiles (E<1eV) in thedetector plane for different theoreticalarrangements in the moderator vessel.

Point spread functionOne of the most important imaging propertiesfor of a neutron radiography station is the pointspread function. The point spread functiondefines the image of a point in the object onthe detector plane. It is influenced by severalfactors: beam geometry, transmission throughthe collimator material, scattering insidestructure material, scattering inside the object,etc. Of course for the last point no predictionscan be made as long as the object is notknown. The resulting point spread function(PSF) from all other factors was estimated foran object plane at a distance of 50cm beforethe detector. The results for our two differentapertures are displayed in Fig.3.

Fig. 3: Point spread functions for an objectplane at a distance of 50cm before the detectorplane for the 4.3 cm aperture (left-hand side)and for the 2.1 cm aperture (right-hand side)

Both PSFs have the shape of a pill-box, thatcan theoretically be expected for the case of ahomogeneous and isotropic source, that isobserved through a round aperture in ´black´


material. That means our beam geometry,especially our ´beam adjusted´ collimator isvery close to the ideal configuration. The PSFscan be used for image deconvolution in thefuture.


3.1 Minimization of the background radiation of a sample environmentNico Wieschalla1, Florian Grünauer1, Jürgen Peters2

1TU München, Physics Department E212TU München, ZWE FRM-II

At the new neutron source FRM-2 many tools for neutron scattering experiments are provided. One ofthem will be a sample environment for experiments at low temperatures and high magnetic fields. Oneimportant question during construction of this tool was the amount of background radiation from structurematerial. This radiation was estimated by Monte Carlo simulations for different configurations of the tool.

Because of mechanical reasons most of thestructure material of the tool is steel. Thermalneutron capturing in iron causes aconsiderable flux of generated high energeticgammas, which is unwanted because of a bigdose outside the experiment. In order to avoidthis additional radiation, all iron structures haveto be covered by a neutron absorber layer.Thickness and choosing of the material of thislayer was optimized by Monte Carlo Method. Athree dimensional cut through the Monte Carlomodel and a horizontal cut is shown in fig.1. Inthe Monte Carlo simulations the neutronsstarted on an area of 1cm x 1cm locatedoutside the tool (fig.1 0°). The energy

distribution is a Maxwell spectrum of 40K or300K (this tool will be used for cold neutronsas well as for thermal neutrons). The assumedsource yield was 109 sec -1 and the maximumangle between the direction of a neutron andthe beam axis is 4°.

Because the super conducting coils cause ahigh secondary radiation, it is essential to findthe best shielding positions and materials toreduce the neutron flux in the coil bodies. Forthis reason we cover surfaces, which areclosest to the neutron beam (shielding 1 infig.1) with different materials of 1mm thickness.

Fig. 1: three dimensional cut through the Monte Carlo model (left hand side) and horizontal cut onbeam level (right hand side

The covering of this surface reduces theneutron flux in the outer coil by two orders ofmagnitudes. By adding the shielding 2 (fig. 1)the neutron flux in the inner coil is reduced bythe same factor. In spite of the differentmacroscopic cross sections of the shieldingmaterials, the neutron flux in the coils is alwaysnoted to be approximately the same. Thereason is amazingly simple. The biggest part ofthe neutrons inside the coils comes throughthe spacer rings, which are necessary becausethe coils will attract with a force of 100kN andonly a small part comes through the shielding.Unfortunately, it is not possible to bring someshielding material between the coils and thesespacers due to the very high forces. Theresulting leakage is not avoidable.

In contrast to the neutron flux inside the coilbodies, the noise to signal ratio (scatteredneutrons from structure material compared tounscattered neutrons) at the sample position inthe center of the tool, depends on the shieldingmaterial. In this case gadolinium is the bestchoice (5·10-7:1) followed by cadmium (10-5:1)and lithium (2·10-5:1).

But if we have a look at the gamma dose at thereal experimental detector outside the tool,gadolinium is the worst of all. It is even worsethan no shielding (see fig. 2). One might guesslithium must be the best choice in this case,because it doesn't produce secondary gammaradiation by neutron capturing. But it iscomparable to cadmium. The reason is the notnegligible gamma production by neutron


capturing in the steel structure of the coil body.As shown above, the neutron flux inside thecoil body is dominated by leakage neutronsfrom the spacer rings, that is independent fromthe shielding material.

However the lowest gamma background canbe reached by a cadmium layer.

Cadmium has the following advantagescompared to the other materials:

• lowest gamma dose at detector position• low noise to signal ratio at sample position• easy to mount (but toxic)• low costs• available in appropriate thickness• because of the high cross section, a thin

layer is sufficient (0.2mm). This importantbecause thicker layers could disturb themagnetic field.

Other simulations showed also that otheradditional shieldings at other positions don´tdecrease background radiation much more.

Fig 2: gamma dose rate at the detectorposition outside the tool for different layers ofabsorber materials (1mm thickness)


4.1 Experimental facilities at the intense positron source at FRM IIC. Hugenschmidt1, G. Kögel2, R. Repper1, K. Schreckenbach1,

P. Sperr2, B. Straßer1, W. Triftshäuser2

1 Physik Department E 21, FRM-II, Technische Universität München, D-85747 Garching2 Institut für Nukleare Festkörperphysik, Universität der Bundeswehr München, D-85577 Neubiberg

The Positron Source at FRM IIAt the new Munich research reactor FRM-II anin-pile positron source based on neutroncapture is installed. After thermal neutroncapture in cadmium the absorption of thereleased high-energy γ-radiation in platinumgenerates positrons by pair production [1].

The positron source is placed in the tip of thedeclined beam tube SR11 at FRM-II. Thedesign of the source is schematically shown infigure 1. Inside the tip of the beam tube a capof cadmium (∅ = 11 cm, l = 9.5 cm) isencapsulated in aluminium. Besides thefunction as γ source after neutron capture,cadmium is a perfect shielding for thermalneutrons. Consequently, it also protects innersource components from neutron activation.The inner part of the source consists of ahoneycomb structure at the tip and rings ofplatinum acting as converter and positronmoderator. Electrical acceleration lenses andmagnetic coils are used for beam formation.

The source is placed at a position where thecalculated unperturbed thermal neutron fluxamounts to 21014 ncm-2s-1. Taking intoaccount the neutron capture rate, theabsorbing mass, the geometry, the moderationefficiency, and a recent test run at an externalneutron beam at the ILL high flux reactor inGrenoble [2], an intensity of the order of 1010

slow positrons per second in the primary beamis expected.

Experimental Facilities at the PositronSourceThe beam line guides the positrons to anexperimental platform where it will beconnected via a beam switch to threeexperiments: A pulsed low energy positronbeam system (PLEPS) [3], a scanning positronmicroscope [4] and a facility for positronannihilation induced Auger electronspectroscopy (PAES) [5].

The PLEPS and the positron microscope wereconstructed and built at the university of theBundeswehr in Munich (figs. 2 and 3). Atpresent, both facilities are under operation with22Na positron sources with beam intensities upto 103 positrons per second. In order to getdepth dependent information from the surfaceto the bulk of the material, the positron energycan be varied from a few 100 eV up to 18 keV.The lateral resolution is about 3 mm at PLEPSand less than 10 µm at the microscope.

Fig. 1: Cross-sectional view of the in-pile positron source at FRM-II.


The PAES facility was built at E21 at TUM andis designed for extremely sensitive studies insurface science [5]. The Auger process isusually initiated by photo- or impact ionizationof core electrons by irradiation with X-rays orkeV-electrons. An alternative technique toinduce core shell ionization is the annihilationof core electrons with slow positrons. Due tothe low positron energy of some 10 eV, nobackground of secondary electrons isproduced in the higher energy range ofreleased Auger electrons. Besides thisadvantage, PAES is an exceptionally surfacesensitive technique since most of theimplanted positrons annihilate with electrons ofthe topmost atomic layer. In first experimentson polycrystalline copper the spectrometer wastested and optimized with a low intensepositron beam at TUM.

Fig 2: Pulsed low energy positron beamsystem (PLEPS) under operation at theuniversity of the Bundeswehr in Munich.

Fig. 3: The scanning positron microscopeunder operation at the university of theBundeswehr in Munich.

Fig. 4: Facility for positron annihilation inducedAuger electron spectroscopy (PAES) at E 21,TUM.

References1. C. Hugenschmidt, G. Kögel, R. Repper, K.

Schreckenbach, P. Sperr, B. Straßer, andW. Triftshäuser; Appl. Phys. A 74, S295(2002)

2. C. Hugenschmidt, G. Kögel, R. Repper, K.Schreckenbach, P. Sperr, and W.Triftshäuser; Nucl. Instr. Meth. B 198, 220(2002)

3. A. David, G. Kögel, P. Sperr, and W.Triftshäuser; Phys. Rev. Lett. 87, 067402(2001)

4. W. Bauer-Kugelmann, P. Sperr, G. Kögel,and W. Triftshäuser; Mat. Sci. For. 363-365, 529 (2001)

5. B. Straßer, C. Hugenschmidt, and K.Schreckenbach; Mat. Sci. For. 363-365,686 (2001)


4.2 Investigation of the Annealed Copper Surface by Positron AnnihilationInduced Auger Electron Spectroscopy

B. Straßer, C. Hugenschmidt and K. Schreckenbach

Fakultät für Physik E21, Technische Universität München,D-85748 Garching, Germany

Auger electron spectroscopy (AES) is a wellestablished technique in surfaces physics. TheAuger process is usually initiated by photo- orimpact ionization of core electrons due toirradiation with X-rays or keV-electrons. Therecorded spectra show a high background ofsecondary electrons which are released by theX-rays or the primary electron beam. Analternative technique to induce core shellionization is the annihilation of core electronswith slow positrons. Due to their low energy ofsome 10 eV, no background of secondaryelectrons is produced in the higher energyrange of released Auger electrons [1]. Besidesthis advantage, positron annihilation inducedAuger electron spectroscopy (PAES) is anextremely surface sensitive technique sincemost of the implanted positrons annihilate withelectrons of the topmost atomic layer [2].

Experimental SetupA PAES facility comprises a slow positronbeam and an analysis unit, which detectsAuger electrons, that are emitted from thesample surface.

The positron beam consists of a 4 mCi 22Nasource and an annealed 1 µm tungstenmoderation foil, in which the primary positronsare slowed down to thermal energy intransmission geometry (figure 1). Subsequentelectric lenses accelerate the moderatedpositrons up to the desired energy (30 eV) intobeam direction. A longitudinal magnetic field of2 mT guides the positrons along a curvedbeamline to the entrance of the analysischamber, where a µ-metal structure is installedto extract the positron beam out of magneticguide field. After that, the beam is electricallyfocused onto the sample, where an intensity of1.1104 positrons per second is achieved [3].

Electrons released from the sample aredetected by an electrostatic hemisphericalenergy analyzer. In addition, the analysischamber is equipped with a heatable sampleholder, an argon sputter source, an electrongun for conventional AES and a massspectrometer for rest gas analysis. Moreover, atungsten shielded NaI-detector is directed ontothe sample at the center of the analysischamber (figure 2) in order to detect theannihilation radiation. A negative bias of -10 V

was applied at the sample in order to focus thepositron beam onto the sample surface.Consequently, the kinetic energy of thepositrons that enter the specimen amounts to40 eV. Due to the negative potential of thesample, the emitted electrons are acceleratedonto the grounded entrance of the energyanalyzer.

Measurements and ResultsEAES (electron induced Auger electronspectroscopy) and PAES measurements wereperformed for an Ar+ -sputtered and annealedpolycrystalline copper sample. An EAESspectrum for the as-received sample wasrecorded as well.

At the beginning, the as-received copper foilwas analyzed by EAES (lower curve infigure 3). Besides the Auger peaks accordingto copper, the adsorbates sulfur, chlorine andcarbon can clearly be identified. After Ar+ -sputtering and the annealing procedure, thepronounced peaks of the surfacecontamination vanish. The peaks according tothe Auger transitions Cu M2,3VV and Cu M1VVbecome more significant.

The PAES spectrum is shown in figure 4. Thebackground due to secondary electrons endsat 40 eV according to the kinetic energy of theimplanted positrons (blue line). Hence, the CuM2,3VV transition (60 eV) can clearly beobserved at the Ar+ -sputtered and annealedcopper sample. The background at higherenergy depends on annihilation radiationinduced photo and Compton electrons and onbeam independent noise of the energyanalyzer. Due to the low intensity of thepositron beam, the recording time of the PAESspectrum was 20 days.

It is planned to install this PAES facility at theintense positron source at the Munich researchreactor FRM-II [4]. Taking into account theexpected intensity of 109 slow positrons persecond, the measuring time should decreaseby at least four orders of magnitude.


References1. A. Weiss, R. Mayer, M. Jibaly, C. Lei, D. Mehl, K. G. Lynn, Phys. Rev. Lett. 61 (1988) 22452. D. Mehl, A.R. Köymen, K.O. Jensen, F. Gotwald, A. Weiss, Phys. Rev. B, 41 (1990) 7993. B. Straßer, C. Hugenschmidt, and K. Schreckenbach; Mat. Sci. For. 363-365 (2001) 6864. C. Hugenschmidt, G. Kögel, R. Repper, K. Schreckenbach, P. Sperr, B. Straßer, and

W. Triftshäuser; Mat. Sci. For. 363-365 (2001) 425

UHV pump

moderation foil

Ti foil

Helmholtz coils

ß+ source

e+

electric lenses

solenoid coil

gate valve

magnetic field termination

analysis chamber

argon sputter

gun

electron gun

sample

electron energy

analyser

electric lenses

UHV pump

Fig. 1: Experimental set-up of the positron beam and the PAES analysis chamber (schematically). ThePhotography of the entire experimental set-up is shown in report 4.1 "Experimental facilities at theintense positron source at FRM-II", fig. 4.


Fig. 2: View into the analysis chamber. Theincoming positrons are focused by electriclenses (top left) onto the sample in the centerof the analysis chamber. Top: entrance of theelectron energy analyzer. Top right: tungstenshielded NaI-detector.

0 50 100 150 200 250 3004

5

6

7

8

9

10

11

12

13

14 EAES on polycrystalline Copper

above: Ar+ -sputtered and annealedbelow: as-received

CClSCu

CuCu M

1VV

Cu M2,3

VV

coun

ts/e

V [a

.u.]

energy [eV]

Fig. 3: EAES at polycrystalline copper (primaryelectron energy: 1 keV; measuring time: 10minutes at each case). The pronounced peaksof the surface contamination vanish after Ar+ -sputtering and annealing.

0 10 20 30 40 50 60 70 80 90 100 110 1200

10

20

30

40

50

60

70

80

90

100

PAES on polycrystalline CopperAr+ -sputtered and annealed

Cu M2,3

VV

x 100

coun

ts/e

V [

a.u.

]

energy [eV]

Fig. 4: PAES at Ar+ -sputtered and annealedcopper: The secondary electron spectrum endsat 40 eV according to the kinetic energy of theimplanted positrons (blue line). The Auger-electron peak of the Cu M2,3VV transition issignificantly detectable.


4.3 High Resolution Gamma Spectrometer for Doppler BroadeningMeasurements of the 511 keV Positron Annihilation Line

C. Hugenschmidt, M. Rykaczewski and K. Schreckenbach

Fakultät für Physik E21, Technische Universität München,D-85748 Garching, Germany

In solid state physics positron annihilation isused as a high sensitive and non-destructivetechnique for the investigation of themicrostructure and the momentum distributionof electrons. In these experiments, positronsare implanted into the sample material wherethey thermalize in a few picoseconds. Afterdiffusion (typical diffusion length 0.1 µm), thepositron together with an electron annihilatesinto two 511 keV quanta, which are emitted atan angle of 180° in the center-of-mass-system.In the laboratory system one measures aDoppler shift of the annihilation radiation due tothe momentum of the annihilating electron –the momentum of the thermalized positron isnegligible.

ExperimentalA new spectrometer was constructed with twofacing germanium detectors for themeasurement of the annihilation radiation withhigh energy resolution. In between, a positronsource (22Na, 3.0 MBq) is sandwiched betweentwo identical samples in order to use almostthe whole solid angle of the emitted positrons(figure 1). The 180°-geometry of thegermanium detectors allows the coincidentdetection of both annihilation photons, i.e. inthese events the whole annihilation gammaenergy of 1022 keV is deposited in the Ge-detectors. This 2D-Doppler broadeningtechnique leads to a better energy resolutionand much lower background.

Fig. 1: Experimental set-up

Measurements and ResultsThe contour plot (fig. 2, left) and the 3D-intensity profile (fig. 2, right) show the result ofthe first test run with the new spectrometer.The peak intensity is located at the annihilationenergy E = mec

2 = 511 keV. The dark bluelines at E1 = 511 keV and at E2 = 511 keVrespectively are due to events where onedetector detects the whole energy of oneannihilation photon and the other records theenergy of Compton scattered gammas mainlyof the 1275 keV deexcitation of 22Ne. Thediagonal from top left to the down right at

E = E1 + E2 = (511 keV - ∆E)1 + (511 keV + ∆E)2 = 1022 keV

corresponds to the whole annihilation energyand contains the information of the Dopplershift ∆E. The resulting broadening is clearlyobservable in the contour plot in figure 2.

Figure 3 shows the conventionally recordedspectrum of one Ge-detector in single modeoperation (red). The energy resolution of oneGe-detector at 511 keV is FWHM = 1.2 keVwhich is broadened to FWHM = 2.42 keV dueto the Doppler effect. For comparison, the datacut diagonally from down left to the top rightthrough the 511 keV peak of the contour plotrecorded with the spectrometer in coincidencemode of both Ge-detectors is also plotted(blue). One can observe a significantly betterenergy resolution of FWHM = 0.97 keV andeven more obviously the reduction ofbackground events of several orders ofmagnitude.

Al - sample

511keV511keV

22Na - source

Ge-detector 2Ge-detector 1

e+


Fig. 2: Contour plot (left) and 3D-intensity profile (right) of the gamma energy (E1, E2) recorded withtwo Ge-detectors in coincidence. The peak intensity is located at the γ energyE = mec

2 = 511 keV. The diagonal from top left to the down right atE = E1 + E2 = (511 keV-∆E)1 + (511 keV + ∆E)2 = 1022 keV corresponds to the wholeannihilation energy and contains the information of the Doppler shift ∆E.

500 505 510 515 520 525

101

102

103

104

105

106

Events

Eγ

FWHM: 0.97 keV

FWHM: 2.42 keV

Detector 1 Coincidence Det.1 & Det.2

Fig. 3: Comparison between the spectrum in single mode (red) and coincidence mode of both detectors (blue)

E1

E2

Events

E2

E1


4.4 Investigation of Laser Treated AlN and Silica by Positron LifetimeMeasurements

C. Hugenschmidt1, L. Liszkay2, W. Egger2

1 Physik Department E 21, Technische Universität München, D-85747 Garching2 Institut für Nukleare Festkörperphysik, Universität der Bundeswehr München, D-85577 Neubiberg

Aluminumnitride has become a technicalmaterial of great interest that simultaneouslycomprises two important macroscopicproperties: an uncommonly high thermalconductivity (180 Wm-1K-1 at RT) and goodelectric insulation properties (ρ > 1014 Ωcmat RT). For this reason, AlN is used in a widerange of technical applications such as high-temperature electronics, substrates and heatsinks for power electronics equipment as wellas for pans for various molten-metal baths.

The sensitive positron annihilation technique iswell suited to study defects and changes inmatter on a microscopic scale. The presentedmeasurements have profited particularly fromthe surface sensitivity of a low energy positronlifetime facility. In this work sintered AlN wasirradiated with 532 nm laser pulses in order tostudy damage effects of the material due tooptical induced changes or thermal load. Inaddition, amorphous SiO2 was laser treated forcomparison.

ExperimentalThin plates (0.6 mm thickness) of AlN andsamples of amorphous SiO2 (HERASIL2) wereirradiated with intense laser light at the WalterSchottky Institute at the TUM. The wavelengthof the Nd:YAG laser light was 532 nmaccording to a photon energy of 6.33 eV. Themaximum energy per 8 ns pulse of 272 mJleads to a total power of 34 MW, which permitslocalized heat- or photo-treatment of materials.The diameter of the laser beam at the samplesurface was 7 mm with a Gaussian intensityprofile. Consequently, the mean energy densityamounted to 7 mJ/mm2 per pulse. For the AlNsamples the laser energy was varied between10 mJ and 272 mJ and one specimen wasexposed to ten laser pulses of maximum poweraccording to a total energy of 2720 mJ. Twoas-received samples of silica were studied as areference and one silica sample was irradiatedwith one 272 mJ laser pulse.

The positron lifetime measurements wereperformed with the pulsed low energy positronbeam system (PLEPS) at the university of theBundeswehr in Munich [1]. This facility will betransferred to the intense positron source atFRM-II [2]. In order to get depth dependentinformation from the surface to the bulk, the

positron energy was varied from 1 keV up to18 keV.

ResultsThe positron lifetime spectra of AlN wereanalyzed by fitting two lifetime components τ1

and τ2 (fig. 1). With increasing positron energythe lifetimes τ1 and τ2 decrease. Due to thesmall positron diffusion length, surface effectsbecome negligible above a positronimplantation energy of 3 keV. Hence, at 3 keV– corresponding to a penetration depth ofabout 60 nm – the bulk value of τ1 of about160 ps is reached. In the near surface region(Ee+ < 3 keV) an increase of τ1 and τ2 isobserved with increasing laser light intensity.

The data analysis for the recorded SiO2-spectra was performed with three lifetimecomponents. The corresponding lifetime ofpara-positronium (p-Ps) was fixed at τ1 =125 ps. τ2 with the intensity I2 is attributed tothe free positron annihilation in the sample andseems not to be affected by the laserirradiation (not shown). The long componentwith τ3 describes the lifetime of ortho-positronium (o-Ps). The intensity I3corresponds to the o-Ps formation rate inmatter. At the surface (Ee+ = 1 keV) the valueof τ3 drops down from 1450 ps for the as-received material to 1200 ps after lasertreatment. The intensity I3 is also much lowerfor laser irradiated silica in particular at lowpositron energy.

Discussion and ConclusionDue to the high molecular stability of N2 whichforms on and desorbs from the surface, mostheated metal nitrides evaporate primarily bydecomposition to their elements. Therefore,the thermal impact of a laser pulse leads to asublimation of nitrogen molecules and to smallgrains of aluminum at the AlN surface. Theincrease of τ1 and τ2 with increasing laserenergy is attributed to larger defects in AlN andin Al at the surface.

The prompt decrease of the o-Ps lifetime τ3and the corresponding intensity I3 show thatalmost no o-Ps-formation occurs at the surfaceand the near surface region after lasertreatment. It is known that silica contains OH --groups and hence dangling bonds. For this


reason, the decrease of Ps formation may beattributed to a lower concentration of danglingbonds, i.e. less “free” electrons, which may beoptically activated if silica is irradiated byintense laser light.

The present results show that the damaging inAlN and in fused silica due to laser irradiationis completely different. In AlN the thermalimpact of the laser pulse creates positrontrapping sites at the surface, whereas in silicaless Ps-formation is observed due to lessdangling bonds after laser irradiation.

AcknowledgementThe author likes to thank CeramTec andHeraeus which made the sample materialavailable. The assignment of the laser facilityand the technical help by Christoph Nebel ofthe Walter Schottky Institute at the TUM isgratefully acknowledged.

References[1] W. Bauer-Kugelmann, P. Sperr, G. Kögel, and

W. Triftshäuser, Mat. Sci. For. 363-365 529(2001)

[2] C. Hugenschmidt, G. Kögel, R. Repper, K.Schreckenbach, P. Sperr, B. Straßer, and W.Triftshäuser; Appl. Phys. A 74 (2002) S295-S297

1 2 3 4 6 8 10 12 14 16 18200

300

400

500

600

700

800

900

1000LASER-irradiated AlN

τ 2 [ps]

E e+ [keV]

energy [mJ] 0 10 90 120 150 210 272 2720

1 2 3 4 6 8 10 12 14 16 18140

160

180

200

220

240

260

280

300

320

LASER-irradiated AlN

τ 1 [p

s]

E e+

[keV]

energy [mJ] 0 10 90 120 150 210 272 2720

Fig. 1: As-received and laser treated AlN: The positron lifetimes τ1 and τ2 as a function of positronenergy. The significant increase of the lifetimes below 3 keV – according to a positronpenetration depth of 60 nm – is attributed to the creation of defects.


0 2 4 6 8 10 12 14 16 1 80

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

amorphous SiO2

I 3 [%]

E e+

[keV]

as-received 1 as-received 2 laser-treated: E=272mJ

0 2 4 6 8 10 12 14 16 181150

1200

1250

1300

1350

1400

1450

1500

1550

1600

as-received 1 as-received 2 laser-treated: E=272mJ

amorphous SiO2

τ 3 [p

s]

E e+ [keV]

Fig. 2: As-received and laser treated amorphous SiO2: τ3 and the intensity I3 describe the o-Ps lifetimeand the o-Ps formation rate.


5.1 Nuclear Licensing of the FRM-IIProf. Dr. K. Böning

Technische Universität München

The first draft of the Third Partial NuclearLicense of the FRM-II, which includes thelicense to start up and operate the reactor, hadbeen sent by the Nuclear Licensing Authority,the Bavarian State Ministry of StateDevelopment and Environment (StMLU), to theFederal Ministry of Environment and ReactorSafety (BMU) already in August 2000. Afternumerous meetings of the representatives ofthe Technische Universitaet Muenchen (TUM)and its general contractor Framatome-ANPwith the technical experts and advisors of theBMU, the Reactor Safety Commission (RSK)and the Commission on Radiation Protection(SSK), the BMU issued its Federal SupervisoryAssessment to the StMLU in January 2002.This document listed some 65 questions orconditions – more and including additionaltopics than contained in the positiveassessments of both RSK and SSK - which theBMU wanted to get further discussed in moredetail. This represented very hard work for

both TUM and Framatome-ANP until we coulddeliver all the relevant documents to theStMLU and the StMLU could submit anupdated draft of the Third Partial NuclearPermit to the BMU in July 2002.

At the end of October 2002 the BMU sent anupdated List of Questions which still containedsome 10 items with a focus on Beyond DesignBasis Accident scenarios. This again meanthard work for both TUM and Framatome-ANPuntil we could submit the last documents in midJanuary 2003. Having answered all thesequestions we are pretty confident now that theStMLU will receive a positive FederalSupervisory Assessment of the BMU in early2003. The final, Third Partial Nuclear Licenseof the StMLU would probably follow shortlylater and the transport of the first fuel elementsand the nuclear startup of the FRM-II could beinitiated immediately after that.


5.2 Fuel Element with Reduced Enrichment for the FRM-IIProf. Dr. K. Böning

Technische Universität München

The outstanding performance of the FRM-II toprovide a high thermal neutron flux (8 × 1014

cm-2s-1 unperturbed) in a large usable volumeat a relatively low reactor power (20 MW) isdirectly related to the design of a single, verycompact fuel element in a large tank filled withheavy water moderator. This compact fuelelement is cooled by light water and containsuranium-silicide (U3Si2) aluminum dispersionfuel with high enrichment (93 % U235) andwith a relatively high uranium density (3.0gU/cm3, graded).

After its election in 1998 the red-green FederalGovernment – following its nonproliferationphilosophy - decided to investigate if the FRM-II could be converted to use low-enriched fuel(with 20 % U235 only). The Federal Ministry ofEducation and Research (BMBF) establishedan experts’ committee which – after manysessions – presented a final report listing anumber of alternatives in June 1999. Theposition of the TUM was always to maintain thereactor power and the compact dimensions ofthe fuel element – so that the FRM-II reactorfacility would not need to be modified and builtonce again and that the penalties in reactorperformance would be marginal; this means

that the enrichment of U235 could be reducedonly so far as this could be fully compensatedby an increase in the uranium density of newfuels (i.e. without making the fuel elementlarger). An agreement following this line hasbeen negotiated between the Bavarian StateMinistry of Science, Research and Arts(StMWFK) and the BMBF; this agreement wasfinalized in October 2001 but will be signed bythe ministers only once the green light hasbeen given by the Federal Government to startup and operate the FRM-II. This agreementsays that the FRM-II will be permitted tooperate with fuel elements using highlyenriched uranium for about ten years, i.e. untilan advanced fuel with a higher uranium densityhas been developed and tested so that a newfuel element with reduced enrichment can bedesigned, fabricated, tested and licensed.

The TUM has already established aninternational working team to develop such anadvanced fuel and a modified fuel element forthe FRM-II. This work will actually begin oncethe green light for the FRM-II has beenobtained and the agreement has been signedand the necessary budget has been allocated.


6.1 Publications

1. P.A. Alekseev, E.Clementyev, P.Allenspach, Yu.Chumlyakov, V.N.Lazukov and I.P.Sadikov,“Soft mode and magnetic phase transition in PrNi”, JETP Lett. 76, (2002) 110

2. P.A. Alekseev, J.-M.Mignot, K.S.Nemkovski, R.Kahn, N.Yu.Shitsevalova, R.I. Bewley,R.S.Eccleston, E.Clementyev, V.N.Lazukov, “Yb-Yb correlations and crystal field in the Kondo-insulator YbB12”, Appl. Phys. A 74, (2002) s562-s564

3. P.A. Alekseev, J.-M.Mignot, A.Ochiai, E.V.Nefeodova, I.P.Sadikov, E.Clementyev, V.N.Lazukov,M.Braden and K.S.Nemkovski, “Collective magnetic excitations in mixed-valence Sm0.83Y0.17S”,Phys. Rev. B 65, (2002) 153201

4. P.A. Alekseev, J.-M.Mignot, U.Staub, A.Ochiai, A.V.Golubkov, M.Braden, R.Bewley,E.V.Nefeodova, I.P.Sadikov, E.Clementyev, V.N.Lazukov, K.Nemkovski, “f-electron excitations inthe neutron spectra of mixed-valence Sm1-xYxS”, Physica B 312-313, (2002) 333

5. S. Baechler, N. Kardjilov, M. Dierick, J. Jolie, G. K.uhne, E. Lehmann, T. Materna, "New featuresin cold neutron radiography and tomography Part I: thinner scintillators and a neutron velocityselector to improve the spatial resolution", Nuclear Instruments and Methods in PhysicsResearch A 491 (2002) 481-491.

6. M. Bleuel, F. Demmel, R. Gähler, R. Golub, K. Habicht, T. Keller, S. Klimko, I. Köper, S.Longeville, and S. Prokudaylo, "Future Developments in Resonance Spin Echo", Springer-VerlagLNP 601 (2002), p. 176 ff.

7. P. Böni, "Experimental Methods: Neutron Scattering", in `Magnetism', Lecture Notes of the 1st

Summer School on Condensed Matter Research, Zuoz, Switzerland, August 10-17 2002, PSI-Proceedings 02-02, ISSN 1019-6447, 1-21 (2002).

8. P. Böni, E. Clementyev, Ch. Stadler, B. Roessli, G. Shirane, and S. A. Werner, "Fincher-BurkeExcitations in Single-Q Chromium", Applied Physics A 74, S716 - S718 (2002).

9. P. Böni, B. Roessli, D. Görlitz, and J. Kötzler, "Damping of Spin Waves and Singularity of theLongitudinal Modes in the Dipolar Critical Regime of the Heisenberg-Ferromagnet EuS", Phys.Rev. B 65, 144434-(1-9) (2002).

10. K. Böning: „Use of Highly Enriched Uranium at the FRM-II“. Proceedings of the 6th InternationalTopical Meeting on Research Reactor Fuel Management (ENS RRFM 2002), Ghent, Belgium, 17.– 20. 03. 2002, Seite 73 – 77.

11. K. Böning, A. Axmann, K. Schreckenbach, M. Köhler, H.-J. Didier: „Die neue Forschungs-Neutronenquelle FRM-II“. Jahrbuch der Atomwirtschaft 2002, Manuskript 17 Seiten

12. K. Böning: „Ausschluß einer Rekritikalität der abgebrannten Brennelemente und Konverterplattendes FRM-II im Endlager und Konzept einer möglichen Konditionierung“. Technische UniversitätMünchen, Bericht OPA00290 vom 25. 01. 2002 (22 + 10 Seiten).

13. K. Böning: „Zur Endlagerung der abgebrannten Brennelemente und Konverterplatten des FRM-II:Radionuklide und Aktivitätsinventar“. Technische Universität München, Bericht OPA00292 vom05. 03. 2002 (3 + 30 Seiten).

14. Braun, K. M. Briggs, and P. Böni, "Analytical Solution for Matthews'and Blakeslee's CriticalDislocation Formation Thickness of Expitaxially Grown Thin Films", Journal of Crystal Growth241/1-2, 231-234 (2002).

15. E.Clementyev, P.A.Alekseev, P.Allenspach and V.N.Lazukov, “Soft-mode-driven magneticordering in the singlet ground-state system PrNi”, Appl. Phys. A 74, (2002) s589-s591

16. R. Georgii, S. Plüschke, R. Diehl, G.G. Lichti, V. Schönfelder, W. Hermsen, J. Ryan and K.Bennett, “COMPTEL upper limits for the 56Co ?-ray emission from SN1998bu”, A&A 394, (2002),517 - 523

17. N. Golosova, V. Bobrovskii, E. Mitberg, A. Podlesnyak, I. Zhdakhin, A. Mirmelstein, "Superpositionstructure of crystal spectra in high-Tc superconductors: overdoped mode", JETP 94, (2002) 1013-1025


18. N. Golosova, A. Podlesnyak, E. Mitberg, V. Bobrovskii, A. Mirmelstein and A. Furrer, "The effect ofCa and Th substitution on the crystal-field spectrum in the high-Tc superconductor HoBa2Cu3Ox",J. Phys.: Condens. Matter 14, (2002) 1923-1936

19. C. Hugenschmidt, G. Kögel, R. Repper, K. Schreckenbach, P. Sperr, B. Straßer, and W.Triftshäuser, "Monoenergetic Positron Beam at the Reactor Based Positron Source at FRM-II",Nucl. Instr. Meth. B 192, 1-2 (2002) 97-101

20. C. Hugenschmidt, G. Kögel, R. Repper, K. Schreckenbach, P. Sperr, B. Straßer, and W.Triftshäuser, "Intense Positron Source at the Munich Research Reactor FRM-II", Appl. Phys. A 74(2002) S295-S297

21. C. Hugenschmidt, K. Schreckenbach and B. Straßer, "Investigation of Positron Work Function andModeration Efficiency of Ni, Ta, Pt and W (100)", Appl. Surf. Sci. 194, 1-4 (2002) 283-286

22. C. Hugenschmidt, G. Kögel, R. Repper, K. Schreckenbach, P. Sperr, and W. Triftshäuser, "FirstPlatinum Moderated Positron Beam Based on Neutron Capture", Nucl. Instr. Meth. B 198 (2002)220-229

23. N. Kardjilov, S. Baechler, M. Bast=FCrk, M. Dierick, J. Jolie, E. Lehmann, T. Materna, B.Schillinger, P. Vontobel, "New features in cold neutron radiography and tomography Part I:Applied energy-selective neutron radiography and tomography", accepted for publication inNuclear Instruments and Methods in Physics Research A (2002)

24. N. Kardjilov, E. Lehmann, P. Vontobel, "Representation of the image formation in applied neutronradiography in terms of a PSF superposition", Appl. Phys. A 74 [Suppl.], S228-S230 (2002).

25. M. Koppe, M. Bleuel, R. Gaehler, R. Golub, P. Hank, T. Keller, S. Longeville, U Rauch, J. Wuttke,"Prospects of resonance spin echo", Physica B266, (1999) 75

26. V.N.Lazukov, P.A.Alekseev, N.N.Tiden, Ch.Beck, E.Clementyev and I.P.Sadikov, “The SpecialFeatures of the Ground State in CeAl3”, JETP Lett. 76, (2002) 295

27. A. Mirmelstein, V. Bobrovskii, N. Golosova, A. Podlesnyak, and A. Furrer, "Frustrated phaseseparation in the overdoped regime of "123" high-Tc superconductors", Journal ofSuperconductivity: Incorporating Novel Magnetism 15, (2002) 367-371

28. A. Mirmelstein, A. Podlesnyak, V. Bobrovskii, N. Golosova, "Crystal-field spectrum of R1-

yCayBa2Cu3Ox high-Tc superconductors in the overdoped regime", Applied Physics A 74, [Suppl.](2002) S1019-1021

29. P. Müller-Buschbaum, P. Böni, A. Meyer, J. Neuhaus, W. Petry, and E. Steichele, Editors ofProceedings of the International Conference on Neutron Scattering 2001, 9-13 September,Munich, Germany, Applied Physics A 75 (Springer, Berlin, 2002).

30. M. Nuding, K. Böning: „Bericht über das Brennstoff-Bestrahlungsexperiment am OSIRIS-Reaktor“.Technische Universität München, Bericht OPA00297 vom 15. 04. 2002 (49 Seiten).

31. Pirogov, A. Podlesnyak, Th. Strassle, A. Mirmelstein, A. Teplykh, D. Morozov, A. Yermakov,"Neutron diffraction investigation of the metamagnetic transition in ErCo2", Applied Physics A 74,[Suppl.] (2002) S598-600

32. Podlesnyak, A. Mirmelstein, N. Golosova, E. Mitberg, I. Leonidov, V. Kozhevnikov, I. Sashin,F. Altorfer, A. Furrer, "Magnetic properties and crystal-field excitations in the LnxSr1-xCoO3 (Ln =La, Pr, Ho)", Applied Physics A 74, [Suppl.] (2002) S1746-1748

33. Podlesnyak, Th. Strässle, A. Mirmelstein, A. Pirogov, R. Sadykov, "Magnetization and neutronscattering studies of the pressure effect on the magnetic transition in Er0.57Y0.43Co2 ", Eur. Phys. J.B 29, (2002) 547- 552

34. Podlesnyak, Th. Strässle, J. Schefer, A. Furrer, A. Mirmelstein, A. Pirogov, P. Markin, andN. Baranov, "Magnetic transition in Er1-xYxCo2 (x = 0, 0.4) single crystals probed by neutronscattering in magnetic fields", Phys. Rev. B 66, (2002) 012409 (4 pages)

35. S. Prokudaylo, R. Gähler, T. Keller, M. Bleuel, M. Axtner, A. Selvachev, "Calculations onresonance spin-echo coils", Appl Phys A 74 (2002) [Suppl1], s317-s319

36. B. Roessli and P. Böni, "Polarized Neutron Scattering", in 'Scattering and Inverse Scattering inPure and Applied Science', edited by E. R. Pike and P. Sabatier, S. 1242-1263, Academic Press2002.


37. B. Roessli, P. Böni, W. Fischer, and Y. Endoh "Chiral Fluctuations in MnSi above the CurieTemperature", Phys. Rev. Lett. 88, 237204 (2002).

38. M. Senthil Kumar and P. Böni, "Influence of Interstitial Nitrogen on the Structural and MagneticProperties of FeCoV/TiNx Multilayers", J. Appl. Phys. 91, 3750 (2002).

39. M. F. Thomas, G. S. Case, J. Bland, A. D. F. Herring, W. G. Stirling, S. Tixier, P. Böni, R. C. C.Ward, M. R. Wells, and S. Langridge, "Studies on Rare-Earth/Iron and Actinide/Iron Multilayers",Hyperfine Interactions 141/142, 471-476 (2002).


6.2 Conference, Workshop and Seminar Contributions

P. Böni"Spindynamik in magnetischen Materialien: Ein Spielplatz für Neutronen", Kolloquium UniversitätBraunschweig, 29. Januar 2002.

P. Böni, B. Roessli, D. Görlitz, and J. Kötzler"Spin waves and singularity of the longitudinal modes in the dipolar critical regime of ferromagneticEuS", European Conference on ESS, Bonn, 16-17 May 2002.

P. Böni"Chirale Anregungen in magnetischen Systemen", Workshop 'Magnetismus und Streumethoden',Forschungszentrum Jülich, 7. Juni 2002.

P. Böni"Grundlagenphysik: Experimente am FRM-II", Treffen Bund der Freunde der TUM, TU-München, 19.Juli 2002.

P. Böni"Experimental Methods: Neutron Scattering", Lecture, 1st Summer School onCondensed Matter Research, Zuoz, Switzerland, August 10-17 2002.

P. Böni"Spindynamik in magnetischen Materialien – Ein Spielplatz für Neutronen", Tag der offenen Tür,Alumni-Forum, Technische Universität München, 23. November 2002.

K. Böning„Der steinige Weg zum FRM-II“. Vortrag auf dem Festkolloquium der Münchner Physiker aus Anlaßder Emeritierung von Prof. Dr. Wolfgang Gläser, Physik-Department der Technischen UniversitätMünchen, Garching, 18. 02. 2002

K. Böning„Use of Highly Enriched Uranium at the FRM-II“. Vortrag auf dem 6th International Topical Meeting onResearch Reactor Fuel Management (ENS RRFM 2002), Ghent, Belgium, 18. 03. 2002.

K. Böning„Forschungsreaktor München 2 – Risiko oder Chance?“ Vortrag und Diskussion mit Dipl. Phys. C.Pistner (Gruppe IANUS an der Technischen Universität Darmstadt), veranstaltet von derVolkshochschule im Norden des Landkreises München, Bürgerhaus Garching, 16. 04. 2002.

J. Brunner, B. Schillinger"Dynamic Neutron Radiography of a Combustion Engine", 7th World Conference on NeutronRadiography, Rome, 15-20 September 2002

E. Calzada, F. Grünauer, B. Schillinger,"The Design of the tomography facility at the new research reactor FRM-II at TU München", EuropeanConference on ESS, Bonn, 16-17 May 2002

E. Calzada, F. Grünauer, B. Schillinger"Engineering solutions for the new radiography and tomography facility at FRM-II", 7th WorldConference on Neutron Radiography, Rome, 15-20 September 2002

F. Grünauer, P. Böni"Optimization of the composition of heavy concrete for neutron shielding purposes", EuropeanConference on ESS, Bonn, 16-17 May 2002

F. Grünauer, B. Schillinger"Shielding aspects for the neutron tomography facility at FRM-II", 7th World Conference on NeutronRadiography, Rome, 15-20 September 2002


C. Hugenschmidt"Positron Generation and Extraction at FRM-II", EPOS-02, The 1st Annual ELBE Positron SourceMeeting, Rossendorf, September 2002

C. Hugenschmidt"Der neue Forschungsreaktor FRM-II und die intensive Positronenquelle",Seminar im Helholtz-Institut für Strahlen- und Kernphysik, Bonn, Dezember 2002

C. Hugenschmidt"The In-Pile Positron Source at the Munich Research Reactor FRM-II", Jahrestagung Kerntechnik,Stuttgart, Mai 2002

C. Hugenschmidt"The Positron Source Based on Neutron Capture", PPC-7, 7th International Workshop on Positron andPositronium Chemistry, Knoxville, USA, June 2002

C. Hugenschmidt"Investigation of Laser Treated AlN by Positron Lifetime Measurements",PSSD-2002, The 3rd International Workshop on Positron Studies of Semiconductor Defects, Sendai,Japan, September 2002

C. Hugenschmidt, G. Kögel, K. Schreckenbach, P. Sperr, B. Straßer, and W. Triftshäuser;"Intensive Positronenquelle am neuen Münchner Forschungsreaktor FRM-II",DNJ, Deutsche Neutronenstreutagung, Bonn 2002

C. Hugenschmidt, B. Straßer and K. Schreckenbach"Investigation of the Annealed Copper Surface by Positron Annihilation Induced Auger ElectronSpectroscopy", PPC-7, 7th International Workshop on Positron and Positronium Chemistry, Knoxville,USA, June 2002

E.H. Lehmann (Paul Scherrer Institut, CH), P. Vontobel (Paul Scherrer Institut, CH), B. Schillinger,"Stand und Perspektiven der Tomographie mit Neutronen als Werkzeug derzerstörungsfreien Materialanalyse", Vortrag am Fraunhofer Institut für Produkttechnik undAutomatisierung (IPA) in Stuttgart

N. Kardjilov, S. Baechler, M. Bastürk, M. Dierick"Energy-selective neutron radiography and tomography with cold Neutrons", 7th World Conference onNeutron Radiography, Rome, 15-20 September 2002

N. Kardjilov, S. Baechler, E. Lehmann, G. Frei"Applied energy-Selective neutron radiography and tomography with cold neutrons", 7th WorldConference on Neutron Radiography, Rome, 15-20 September 2002

N. Kardjilov, F.C. de Beer, M.F. Middleton, B. Schillinger"Corrections for scattered neutrons in quantitative neutron radiography", 7th World Conference onNeutron Radiography, Rome, 15-20 September 2002

B. Schillinger, M. Bleuel, P. Böni, E. Steichele, "Design and simulation of a neutron guide plus flighttube for neutron radiography and tomography at the new research reactor FRM-II atTechnical University Munich", 7th World Conference on Neutron Radiography, Rome, 15-20September 2002

B. Schillinger"Neutron computed tomography", invited lecture, Korean Atomic EnergyResearch Institute in Daejon,Korea

B. Schillinger"Neutron computed tomography as an industrial tool", invited lecture, Korean Atomic EnergyResearch Institute in Daejon,Korea


B. Schillinger"Neutronentomographie", Eingeladener Vortrag am Hahn-Meitner-Institut

B. Schillinger"Die Neutronentomographie-Anlage am neuen Forschungsreaktor FRM-II", Eingeladener Vortrag amHahn-Meitner-Institut

Shah Valloppilly"Influence of interfaces on the magnetism of ultra thin filmsand multilayers: 3 case studies", Neutrons in research and industry seminar,Technische Universität München, October 28, 2002.


6.3 Lectures, Courses and Seminars

M. Bleuel: Tutorial "Experimental Physics for Civil Engineers 1"Tutorial "Experimental Physics for Civil Engineers 2"Tutorial "Introduction to Solid State Physics I"

P. Böni: Lectures "Experimental Physics for Civil Engineers 1"Lectures "Experimental Physics for Civil Engineers 2"Lectures “Introduction to Solid State Physics I"Seminar "Neutrons in Research and Industry"

K. Böning: Lectures "Reaktorphysik I: Grundlagen“Seminar "Neutrons in Research and Industry"

J. Brunner: Tutorial "Elektronikpraktikum"

R. Georgii: Seminar "Neutrons in Research and Industry"

F. Grünauer: Tutorial "Experimental Physics for Civil Engineers 1"Tutorial "Experimental Physics for Civil Engineers 2"Tutorial "Introduction to Solid State Physics I"

N. Kardjilov: Tutorial "Elektronikpraktikum"

T. Keller: Tutorial "Elektronikpraktikum"

B. Schillinger: Tutorial "Elektronikpraktikum"

K. Schreckenbach: Seminar "Neutrons in Research and Industry"


6.4 Committee Memberships

P. Böni• Nutzerausschuss deutsches Kontingent, Institut Laue Langevin, Grenoble• Instrument Subcommittee, Institut Laue Langevin, Grenoble• SINQ Scientific Committee• Projektbegleitender Beirat FRM-II• Instrumentierungsausschuss FRM-II• Direktorium FRM-II• TUM-Beirat für den FRM-II• Coordinator of Work Package on Neutron Optics, Joint Research Projects in EU: FP6

K. Böning• Chairman (President) of the International Group on Research Reactors IGORR.• Member of the Technical Review Committee (TRC) of the Taiwan Research Reactor System

Improvement and Utilization Promotion Project (TRR-II).• Member of the Scientific Committee of the ENC (European Nuclear Conference) of the European

Nuclear Society ENS, Track Leader (together with Alain Ballagny, CEA, France) for „Experimental,Research Reactors and Neutron Sources“ at the ENC 2002 Scientific Seminars.

B. Schillinger• Vice President of the European Neutron Radiology Association (ENRA)• Board Member of the International Society for Neutron Radiography (ISNR)

K. Schreckenbach• Komitee für Neutronenforschung (KFN)• Instrumentierungsausschuss FRM-II• Direktorium FRM-II• TUM-Beirat für den FRM-II• Arbeitgemeinschaft Forschungsreaktoren (AFR)

E. Steichele• International Conference on Neutron Scattering ICNS 2001, Munich: Co-Editor Committee• Instrumentierungsausschuss FRM-II


6.5 E21 Members

Name Phone+49-89-289-

Fax+49-89-289-

Room email

Arend Nikolas, Dipl. Phys. -14723 - PH 1, 2205 [email protected] Markus -12125 -13776 RS, 140 [email protected] Markus, Dipl. Phys. -13777 -13776 RS, 124 [email protected]öni Peter, Prof. Dr. -14711 -14713 PH 1, 2215 peter.boeni@

frm2.tum.deBöning Klaus, Prof. Dr. -12150 -12191 FRM-II UBA 0325 [email protected] Hans-Benjamin, Dr. -12512 -14713 PH 1, 2367 [email protected] Johannes,Dipl. Phys.

-12106 -13776 RS, 102 [email protected]

Clementyev Evgeni, Dr. -14722 -14713 PH 1, 2207 [email protected]äser Wolfgang,Prof. emerit.

-12476 -12474 PH 1, 2227 [email protected]

Grünauer Florian,Dipl. Phys.


Hils Thomas, Dr. -14721 - Flachbau, 17 [email protected] Nikolay, Dipl.Phys.


Kargl Verena, Dipl. Phys. -12515 - PH 1, 2373 [email protected] Alexey, Dr. -14718 -14713 PH 1, 2207 [email protected] Christian,Dipl. Phys.

-12299-12199

-13776 Flachbau, 13 [email protected]

Prokudaylo Sergey,Dipl. Phys.


Reich Christian,Diploma Student

12515 - PH I, 2373 [email protected]

Russ Barbara, Dipl. Ing. -14717 -14713 PH 1, 2207 [email protected] Christian, Dipl.Phys.

-14725 - PH 1, 2214 [email protected]

Schreckenbach Klaus, Prof.Dr.

-12183 -12191 RS [email protected]

Skorski Cornelia, Secretary -14712 -14713 PH 1, 2217 [email protected] Christoph,Diploma Student

- - - -

Steichele Erich, Dr. -12141 -12112 RS [email protected]ßer Benno, Dipl. Phys. -12137 -12191 RS [email protected] Tejic -12656 - FB PH. T. ABT,

1321-

Valloppilly Shah, Dr. -14723 -14713 PH 1, 2205 [email protected] Nico,Diploma Student

-12111 -13776 RS 146b [email protected]

PH: Physics DepartmentRS: Reactor Station


6.6 Associated Members at FRM-II


Room email

Calzada Elbio, Dipl. Ing. -14611 RS [email protected] Robert, Dr. -14986 RS [email protected] Christoph, Dr. -14609 RS, 4 [email protected]ück Guido, Dipl. Ing. -14915 - [email protected] Burkhard, Dr. -12185 RS [email protected]


6.7 Guests


Room email

Gähler Roland, Dr. -12111 RS 146b [email protected] Thomas, Dr. -12164 RS, 4 [email protected]



1 – Shah Valloppilly2 – Christoph Hugenschmidt3 – Nico Wieschalla4 – Benno Straßer5 – Markus Bleuel6 – Elbio Calzada

7 – Verena Kargl8 – Wolfgang Gläser9 – Barbara Russ10 – Peter Böni11 – Christian Reich12 – Burkhard Schillinger

13 – Markus Axtner14 – Klaus Lorenz15 – Christian Schanzer16 – Florian Grünauer17 – Evgeni Clementyev

not on the picture:

Nikolas ArendKlaus BöningJohannes BrunnerRobert GeorgiiThomas Hils

Daniel LamagoGuido LangenstückAlexey MirmelsteinChristian PlonkaKlaus Schreckenbach

Cornelia SkorskiErich Steichele

Institute for Experimental Physics E21 · Institute for Experimental Physics E21 Annual Report 2002...

Documents

Transcript of Institute for Experimental Physics E21 · Institute for Experimental Physics E21 Annual Report 2002...