ANNUALREPORT2006 Forschungs-Neutronenquelle HeinzMaier … · 2015. 3. 4. · LAND) joint European project took broad room in our activities. Detectorreadoutelectronics During the

ANNUAL REPORT 2006Forschungs-NeutronenquelleHeinz Maier-Leibnitz (FRM II)

cover image: An electromagnetic levitation furnance from the Institut für Materialphysik im Weltraum of the DeutschenZentrums für Luft- und Raumfahrt (DLR). For details see section 9.5.

Contents iii

Contents

Directors’ report 2

The year in pictures 3

I Instrumentation 7

1 Central services and reactor 81.1 Detector and electronics lab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.2 HELIOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91.3 Calibration of the control rod . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111.4 Sample environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131.5 Neutron flux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131.6 High density fuel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2 Thin films and large scale structures 192.1 KWS-1 and KWS-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192.2 MIRA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212.3 N-REX+ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262.4 SANS-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272.5 TREFF@NOSPEC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

3 Structure research 313.1 HEIDI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313.2 POLl-HEiDi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333.3 RESI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363.4 SPODI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373.5 STRESS-SPEC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

4 Inelastic scattering, high resolution 414.1 DNS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414.2 J-NSE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 434.3 RESEDA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444.4 SPHERES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454.5 TOFTOF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

5 Three axes spectroscopy 495.1 PANDA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 495.2 PUMA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

6 Imaging at ANTARES and NECTAR 546.1 Imaging the cold source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 546.2 Beam filters for imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 566.3 Phase contrast tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 576.4 NECTAR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

7 Nuclear and particle physics 627.1 NEPOMUC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 627.2 Positron remoderation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 637.3 MEPHISTO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 647.4 UCN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

iv Contents

8 Industrial applications 698.1 Irradiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 698.2 MEDAPP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

II Science 73

9 Scientific highlights 749.1 Solvent content in thin spin-coated polymer films . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 749.2 CDBS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 759.3 PAES-measurements of pure Cu and Cu coated Si(100) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 779.4 ANTARES – Investigation of an early medieval sword by neutron tomography . . . . . . . . . . . . . . . . . . . . 789.5 Containerless sample processing at TOFTOF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 809.6 Larmor diffraction at TRISP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 829.7 Molecular dynamics in pharmaceutical drug delivery systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

10 Events 8710.1 Workshop - Residual stress analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8710.2 Workshop - Neutrons for Geoscience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8710.3 Workshop industrial applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8810.4 Workshop biological and soft matter interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8910.5 Farewell colloquium Prof. Klaus Schreckenbach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

III Facts and Figures 91

11 Experiments and user program 92User office . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92User support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94Industrial applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

12 Public relations and visitor service 95

13 People 9813.1 Committees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9813.2 Staff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10413.3 Partner institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

14 Figures 10714.1 Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

Imprint 111

Contents 1

2 Directors’ report

Directors’ reportPlenty of Neutrons

The past year’s outstanding event wasthe 260-day operation at full power ofFRM II. Already in its second year ofroutine operation our neutron sourceachieved this record, which certainlyis exceptional for high flux neutronsources. Congratulations to both ournuclear staff, who managed the smoothoperation in 2006, and to our experi-mentalists, who served users over thisperiod.

By the end of 2006 fifteen beam holeinstruments served in routine opera-tion. Essential progress was made in thelicensing of MEDAPP, our clinical can-cer irradiation facility, and we all hopeto get the final approval from the Bavar-ian Ministry of Environment, Healthand Consumer Protection early 2007.The Jülich Centre for Neutron Science(JCNS) transported its first instrument- the Neutron Spin Echo Spectrometer -to FRM II in March 2006. With the endof the 7th reactor cycle MEPHISTO, theinstrument for nuclear physics, had tomove from its original position at neu-tron guide NL3 to NL2a, in order tomake room for the installation of thethree small angle scattering machinesof JCNS. The new east building with thenew guide hall on its ground floor andthe offices and labs for JCNS on its firstfloor made significant progress over theyear. We expect JCNS to move into thatbuilding in February 2007. However,the extension of the neutron guide hallwest towards the Atomic Egg advancedslowly. The Jülich small angle scatteringmachines are waiting for this extensionimpatiently in order to start their rou-tine operation.

Important advances have been madein the attempts to develop high den-

sity fuels for future use in FRM II.Within a collaboration of CEA (Com-missariat à l’Energie Atomique), CERCA(the French producer of the FRM IIfuel element), AREVA NP and TUMlarge fuel plates with a 8 g/cm3 denseU8wt%Mo alloy have been irradiated tointegral neutron fluxes beyond the needof FRM II. So far conservative swellinghas been observed. However, there isstill a long way to pursue until thesehigh density fuels will be qualified forthe use in high flux neutron sources.

On 26 June 2006 the Bavarian Min-ister of Science, Dr. Thomas Goppel,invited the members of the EuropeanParliament to the Bavarian Embassy, amarvellous place in the heart of Brus-sels, in order to present FRM II as aBavarian and German contribution tothe European Research Area. Expertmeetings held at Garching in Septem-ber presented the opportunities of theneutron source to geologists, to indus-trial researchers, and to membrane spe-cialists.

The Wolfram-Prandl prize for out-standing work of young neutron sci-entists was awarded conjointly to Dr.Oliver Stockert and Dr. Thomas Keller,both members of different Max-PlanckInstitutes. Dr. Thomas Keller was giventhis prize for the development of TRISP,one of the FRM II instruments withunique measuring capabilities. Amongall the beautiful experiments at FRMII in the course of 2006 we wouldlike to highlight the progress whichwas made by Prof. Keimer’s groupin finding experimental evidence forelectron-phonon and electron-magnoncoupling, just with that instrument putup by Dr. Keller.

Politicians are curious to see how auniversity managed to build up andsuccessfully operate a large scale fa-cility. And of course the TU Munichand FRM II are grateful to politiciansfor their support in the never endingfight for financial and human resources.The Minister President of North Rhine-Westphalia visited FRM II with his en-tire cabinet on 25 September 06. Mr.Yves Caristan, head of CEA Centre ofSaclay, took ideas home on how FRM IIand the Garching campus are buildingup synergies. Prof. Fidel Castro Diaz-Balart, the son of the Cuban leader, vis-ited the research reactor in November inorder to learn more about tumor ther-apy by particle irradiation. On the occa-sion of their visit in May Prof. Mlynek,President of the Helmholtz Gemein-schaft, and Dr. Herbert Diehl from theFederal Ministry of Science promisedtheir further support to FRM II.

Guido Engelke, who first headed theLurgi project controller team of FRMII, was appointed Administrative Direc-tor of FRM II in the beginning of 2001and retired by the end of 2006. Hissuccessor, Dr. Klaus Seebach, startedon 2 January 2007. Like Guido En-gelke he has a strong industrial back-ground. After having studied brew-ing technology and having achieved hisPhD in business administration, both atthe TU München, he started his careerin the brewing business. He becameCEO of several breweries, and for a cer-tain while also CEO of a glass manufac-turer. Colleagues, users and friends ofFRM II expressed their thanks to GuidoEngelke during a farewell party on 20December 2006.

Guido Engelke Winfried Petry Ingo Neuhaus


The year in pictures

17 March 06: Successful relocation of the first of eight instruments to betransferred from Jülich to FRM II.

15 May 06: Official farewell ceremony for Prof. Schreckenbach, TechnicalDirector during the start-up and commissioning phase of FRM II fromMay 1999 until the end of 2005. (Photo: Joachim Hospe)

16 May 2006: Dr. Sebastian Schmidt, Prof. Jürgen Mlynek, the top managersof the Helmholtz Association of German Research Centres visiting FRM IItogether with Dr. Herbert Diehl and Mrs. Ursula Weyrich, Federal Min-istry of Education and Research(from right to left,dark coats, front row).(Second row) Dr. Ulrike Kirste, Bavarian Ministry of Science, Guido En-gelke and Dr. I. Neuhaus accompanied the guests on their tour throughthe research reactor.

4 The year in pictures

23 May 2006: Visitors from the Federal Ministry of Economics and Technol-ogy: Dr. Dorothee Mühl und Mr. Jochen Süßenberger, together with G.Engelke (far left) and I. Neuhaus (far right).

9 June 06: Chief executives from industries and banking invited by theBavarian State Chancellery to visit the Neutron Research Reactor aheadof the start of the Football World Championship 2006.Centre group (touching the footballs, from left to right): Dr. FranzB. Humer, F. Hoffmann-La Roche AG ; Alessandro Profumo, UniCreditS.p.A.; Erwin Huber, Minister, Head of the Bavarian State Chancellery;Hans Spitzner, Vice-Minister in the Bavarian Ministry of Economic Af-fairs. David Martin, CEO of Arriva PLC. and Dr. Andreas Rummelt, CEO ofSandoz International GmbH.Group at the left: Monika Oberndorfer, Invest-in-Bavaria, Bayern Interna-tional GmbH, Dr. Thies Claussen, Bavarian Ministry of Economic Affairs.(Photo: Frank Röthel)

26 June 06: "The New Neutron Source Heinz Maier-Leibnitz - a contributionto strengthen the European Research Area" - Scientists and politicians at-tended the presentation kindly hosted by the Representation of Bavaria inBrussels.Dr. Thomas Goppel, Bavarian Minister of Science, Research and Arts, andProf. Dr. Ernst Winnacker, former President of Deutsche Forschungsge-meinschaft, presently Secretary General of European Research Council.


26 June 06: "The New Neutron Source Heinz Maier-Leibnitz - a contributionto strengthen the European Research Area"Dr. Beatrix Vierkorn-Rudolph (centre), Federal Ministry of Education andResearch, and Anna Lechner, Bavarian Representation in Brussels, talk-ing with Dr. Kuch, TUM Management, and Prof. Petry (far left).(Photo:Alexander Louvet)

12 September 06: On the occasion of the VDI Expert Conference at FRM II:Dr. Yan Gao, General Electric Global Research, and Dr. R. Gilles, FRM II.

25 September 06: Dr. Jürgen Rüttgers, Minister-President of North Rhine-Westphalia, and his cabinet were welcomed by the TUM President, Pro-fessor Dr. Wolfgang A. Herrmann.

6 The year in pictures

25 September 06: The tour guided by Prof. Petry, Scientific Director of FRMII, resulted in lively discussions in which also Dr. Thomas Goppel, Bavar-ian Minister of Science, Research and Arts, took part.

28 November 06: High-ranking Cuban politicians and scientists, membersof the Castro family, at FRM II.

20 December 06: The entire staff says good-bye to Guido Engelke, who hadworked at FRM II for 13 years; from 2002 to 2006 in his capacity as Admin-istrative Director.

Part I

Instrumentation

8 1 Central services and reactor

1 Central services and reactor

1.1 Detector and electronics lab

I. Defendi 1, A. Kastenmüller 1, M. Panradl 1, T. Schöffel 1, K. Zeitelhack 1

1ZWE FRM II, TU München

In 2006 most of the scientific instru-ments went into routine operation thusreducing the effort for commission ofdetector systems and electronics. Whileservice and maintenance still were animportant task of the lab, more empha-sis could be gradually given to the im-provement and further development ofexisting systems.

The launch of two new detectorprojects - the development of a 1×1 m2

fast two-dimensional position sensitivedetector for the new small-angle scat-tering instrument SANS1 and the de-velopment of a 1 mm resolution 2D-detector for the diffraction instrumentStressSpec - as well as the developmentof an individual channel readout sys-tem for a fast Multi-Wire-Proportional-Chamber within the NMI3-JRA2 (MI-LAND) joint European project tookbroad room in our activities.

Detector readout electronics

During the previous years we devel-oped various PCI/cPCI-based detec-tor DAQ-boards using the M-modulestandard for applications like neutroncounting, Time-of-Flight measurementor Multi-Channel scaling [1]. In 2006a further module implementing a 2-channel Multichannel analyzer (MCA)was added. The board features:

• 12 bit peak sensing ADC• 2 input channels; range 0 -

4096mV• selectable input polarity• selectable modus: gated / un-

gated / samplingFigure 1.1 shows a pulse height spec-

trum of a 3He-neutron detector pro-duced by thermal neutrons from a252Cf-source recorded with the MCA-module. We gradually will equip the

scientific instruments using our DAQ-system with the board for the pur-pose of monitoring the installed 3He-detectors and beam monitors.

The NMI3-JRA2 (MILAND) projectis dedicated to the development of afast, medium resolution ('1 mm) 2D-neutron detector. The lab contributeswith the development of a readout sys-tem for 320×320 individual channels ofa fast MWPC depicted in Figure 1.2.

Each cathode wire is connected to atransimpedance amplifier followed bya comparator that generates a Time-Over-Threshold (ToT) signal. 32 chan-nels are integrated into a MILAND32Card developed at ILL [2], that is di-rectly plugged to the rear side of theMWPC. The TTL-output signals aresent to the digital processing unit con-sisting of 2×3 Pre-Processing Mod-ules (PPM) and a COrrelation Mod-

Figure 1.1: Pulse height spectrum of thermal neutrons recorded by the MCA M-module

ule (COM) implemented in a standardVME64x crate with custom backplaneutilization. Both PPM and COM consistof a common FPGA-based board devel-oped at FRM II with specific FPGA de-sign and a piggy board shown in Fig-ure 1.3. The PPM-piggy board con-sists of digital front-end buffers, whilstthe COM-piggy board implements a fastEthernet interface. The ToT-signals areacquired by the processing block incharge of recognizing a cluster of sig-nals and determining the position ac-cording to a well defined algorithm (sig-nal arrival time, longest ToT, ToT basedaveraging). The X and Y coordinates arethen correlated in time and formatteddata are sent to the DAQ system by thetransport block. A set of 3 prototypeboards has been built and successfullypassed the hardware tests. Presently,the VHDL-implementation of the algo-

1 Central services and reactor 9

Figure 1.2: Readout scheme of the MILAND detector

rithm to determine the neutron impactposition is in progress.

Detectors

A 2D-position sensitive detector with1×1 m2 active area, 8 mm×8 mm po-sition resolution and 1 MHz globalcount rate capability is planned for thenew small-angle scattering instrumentSANS1 which is presently under con-struction. After an evaluation of variousalternatives, a linear array of 1 m longposition sensitive 3He-detectors (diam-eter d = 8 mm) is regarded as most fa-vorable configuration similar to the D22detector at ILL. With very helpful sup-port by the ILL detector group the

following two prototype detectors havebeen investigated end of 2006 in neu-tron beams of λ= 9.7 Å and λ= 22 Å re-spectively at the instrument MIRA.

• 8 Reuter Stokes P4-0341-201 3He-PSD’s (L = 1 m, d = 8 mm, pHe =15 bar) arranged as linear array

• A "multitube" 3He-PSD consist-ing of 32 tubes (L = 1 m, d =7.4 mm, pHe = 15 bar) with com-mon gas volume built by the ILLdetector group

Figure 1.4 shows the prototype built ofindividual PSDs mounted in the col-limated beam at MIRA with a remotecontrolled slit in front of the device.

A fast electronics for PSD’s developedby the ILL detector group

Figure 1.3: The MILAND Pre-Processing(PPM) / COrrelation Module(COM)

Figure 1.4: A SANS1 prototype detector built

of individual 3He PSD’s mounted at MIRA

was used to read out both prototype de-tectors. As a first preliminary result,both devices fulfill the requirement of aposition resolution ∆x < 8 mm for neu-trons of both wavelengths and a highrate capability. A more elaborate anal-ysis of the measurements is in progress.[1] Zeitelhack, K., et al. FRM-II annual

report 2003, 6–7.

[2] Guerard, B., et al. NMI3-JRA2 an-nual report 2005, 19–21.

1.2 HELIOS – polarized 3He gas for neutron instrumentation

S. Masalovich 1, O. Lykhvar 1, G. Borchert 1, W. Petry 1,2

1ZWE FRM II, TU München2Physik Department E13, TU München

In 2006 HELIOS, the facility for large-scale production of highly polarized3He gas, has been fully operationalthroughout the year in spite of the lackof one laser for a longer period. Usu-ally the optical pumping at HELIOS is

performed with two high-power lasers,but because of faulty performance onelaser has been sent back to the com-pany for revision. To be able to polar-ize 3He gas to a high polarization statewith only one laser, the optical scheme

of the setup was modified and somecomponents were improved. In par-ticular, the problem of degradation ofthe gas polarization because of instabil-ity of the laser beam polarization hasbeen identified and new optical compo-


nents have been installed to eliminatethis problem. As a result of all of thesemodifications, 3He gas polarization waskept at a high level of around 76% withonly slight reduction in the productionrate. Finally, in autumn 2006 the secondlaser has been put into operation againand since then the 3He gas polarizationabout 82% can be reached in the opticalpumping volume.

Experiments at neutroninstruments

This year experiments with the use ofpolarized 3He gas have been performedat three instruments at the FRM-II:TRISP, MIRA and HEIDI.

At TRISP (the spin echo triple axisspectrometer) a Neutron Spin Filter(NSF) with polarized 3He gas was usedas an analyzer in the measurementswith neutrons of wavelength 1.795Å.The NSF cells with 80 mm neutron flightpass and 2 bar gas pressure were pro-vided for these measurements during 8days with daily replacement. The cellswere positioned in the center of a rect-angular solenoid developed at TRISPand designed to implement the polar-ized 3He gas.

This solenoid was located in one ofthe zero-field arms at the instrument.The gas polarization in the freshly re-placed cell on the neutron beam wasmeasured to be about 64% and the re-laxation time of gas polarization was es-timated as 60 hours.

At MIRA (the beam line for verycold neutrons) the test measurementshave been performed with neutrons ofwavelength 9.7Å to verify the ability touse NSF cells in neutron reflectome-try as well as in small angle scatteringexperiments with polarized neutrons.The aim of these measurements was tocheck the pattern of neutrons scatteredby an empty quartz cell, which sup-posed to be used with polarized 3He gasat MIRA. The requirement for such NSFat any reflectometer (SANS as well) isquite obvious – it has not to scatter neu-trons at small angles. To check howthe quartz cell fits this requirement twomeasurements have been performed atMIRA.

Figure 1.5: Measured intensity distributions across the neutron image with and withoutempty quartz cell in the beam at MIRA

First, a slit mask has been used to pre-pare a narrow neutron beam (3x25mm)and the image of the profile of thisbeam has been recorded with a 2Ddetector located at the distance 1.2mfrom the mask. Then an empty quartzcell was placed into the beam behindthe mask and the image was recordedagain. Two curves in Fig.1 represent thescans across the neutron beam imagesin both cases. One can see a typicalcount rate decrease because of scatter-ing by the quartz cell (measured trans-mittance is 0.90 at this wavelength),but there is no visible small angle scat-tered neutrons. So, one may concludethat neutrons are scattered in a largesolid angle (isotropic scattering) andone may neglect their contribution tothe angular distribution measured inthe real experiments at MIRA (at leastdown to the order of 10−4).

To provide the ability to use polarized3He gas at MIRA a circular solenoid withcompensated ends has been built at theNeutron Optics group. Design of thissolenoid allows installing it inside themu-metal box at MIRA. This box servesto screen the internal area against envi-ronmental magnetic fields and thus en-sures a long lifetime of a NSF. The re-

laxation time related to the magnetic in-homogeneity of the assembled unit wasestimated to be T1 = 200 hours at 1 bargas pressure. A preliminary imaging ex-periment with polarized neutrons hasbeen done using a NSF with 0.25 bargas pressure. The cell with such pres-sure yielded a relaxation time of only35 hours. For future experiments it wasdecided to build special thin cells opti-mized for the neutron wavelength rangeat MIRA.

Whereas TRISP and MIRA have theirintegrated polarizers and NSF with po-larized 3He gas can be used mostly asanalyzer, an experiment with polarizedneutrons at HEIDI (the single crystaldiffractometer with hot neutrons) reliesentirely on the use of polarized 3He gas.Two NSF cells are presumed to be inuse. The first NSF cell serves as a po-larizer and the second one as an ana-lyzer. To check the feasibility of this ap-proach, scientists at HEIDI developedtwo magnetostatic cavities to ensure along lifetime of the polarization in theNSF cells at the environmental mag-netic field of the instrument. Some pre-liminary measurements with NSF cellshave been performed in the summerwith neutrons of wavelength 0.55 Å and


0.87 Å. The whole setup assembled withMuPAD has been tested in the autumnof 2006. The experiment was runningfor a week with neutrons of wavelength1.165 Å. Two NSF cells with 130 mmneutron flight pass and 1.65 bar gaspressure were provided for this experi-ment. A few replacements of the cellshave been done in the course of mea-surement. The gas polarization in thefreshly replaced cells on the neutronbeam was measured to be about 70%and the relaxation time in each cell wasestimated as 100 hours.

Neutron spin filter cells

Seven new cells for Neutron Spin Filtershave been built in the year 2006. Thus,on the whole there are nine cells of dif-ferent sizes currently available for neu-tron experiments. The cells are madeof high purity quartz glasses HSQ 300and HOQ 310. All cells have been fin-ished (cleaned and coated with puremetallic cesium) by the Neutron Optics

group and showed relaxation time rang-ing from 100 up to 200 hours.

It is known that NSF cells suffer fromthe leaks that may appear in the grease-sealed glass valves attached to the cells.Such leaks lead to an oxidation of themetallic cesium inside the cell with sub-sequent degradation of the relaxationtime. The risk of leakage increases withtime since the grease has a tendency toget thinner with increasing number ofvalve operations. Finally the grease filmmay be broken and leaks arise. To beable to change the grease in advance,a special vacuum-tight glove box (seeFig.2) has been designed and built bythe Neutron Optics group. This glovebox can be evacuated and filled withpure 4He gas to ensure oxygen-free en-vironment for valve reinstallation.

In addition, a systematical study hasbeen made by the Neutron Optics groupwith the aim to improve the reliability ofthe valves. As a result it was proposedto use a special wax for vacuum sealinginstead of grease. The wax has a soften-ing point about 50 – 60 °C and the valve

Figure 1.6: Glove box in the cell preparationlab

sealed with such wax can only operateafter slight heating. At room tempera-ture the wax becomes solid and ensurehigh reliable sealing even for the cellswith the elevated gas pressure. It wasshown that effect of the wax on gas po-larization is almost identical with thatof the generally accepted grease.

The work on the preparation of newcells will be continued next year alongwith designing of new magnetostaticcavities for better performance of Neu-tron Spin Filters at neutron instru-ments.

1.3 REACTOR - Calibration of the control rod of the FRM II in thesubcritical regime

K. Schreckenbach 1

1Physik Department E21 and ZWE FRM II, TU München

For the performance and safety of anuclear reactor the reactivity worth cal-ibrations of the control and shut downrods are of major importance. Theexperimental verification of the calcu-lated values are also a valuable benchmark for the validation of the programsused.

The FRM II is equipped with a con-trol rod in the centre of the light watercooled fuel element and five shut-downrods in the surrounding heavy water re-flector. For routine operation all shutdown rods are withdrawn and the reac-tor is stabilized by the control rod fromthe top. With the control rod fully in-serted the reactor is subcritical with ahigh safety margin. In the present re-port the safety margin is deduced from

measured data and compared with cal-culations.

The shut down reactivity for the in-serted control rod was determined dur-ing the start up of the reactor by extrap-olation of the reactivity worth measuredabove the critical position. A morequantitative measurement is the droprod method or the calculation from thesubcritical neutron counting rates. Inthe present report the subcritical count-ing rates are evaluated.

In the subcritical regime the fuel el-ement serves as a neutron amplifier ofan external neutron source. During thefirst criticality a Cf-252 neutron sourcewas used. For an already irradiated corethe photo neutrons serve as primaryneutron source.

For a neutron source S0 with the

same distribution as the neutron in-duced fission reactions and for thepoint kinetic approximation the neu-tron density n(x) as function of controlrod position x is related to the neutronmultiplication factor k by

n(x) = S0

1−k(x)

k(x) = 1− S0

n(x)

S0 =−dk

d x/

d( 1n )

d x

The reactivity ρ is related to k by ρ =(k −1)/k. In one of the cases discussedhere, the reactor core was already usedand photo neutrons were produced bygamma rays from the core in the heavywater. Criticality occurred at a higher


Figure 1.7: Schematic cross section of the reactor with start-up neutron source and the widerange neutron detectors WR.

Figure 1.8: Multiplication factor as function of the control rod position for the photo neu-tron case.

control rod position (384 mm) com-pared to a fresh fuel (340 mm). Forthis case the above prerequisites are ful-filled approximately and the unknownparameter S0 (here photo neutrons)can be determined from the derivativedk/d x as measured for the fresh corewith the critical reactor at 340 mm.

The neutrons were counted in widerange detectors (WR) surrounding theheavy water vessel (see Fig.1.7), thecounting rate ci (x) being proportionalto n(x) in the point kinetic approxima-tion. The result is shown in fig.1.8. Themiddle plane of the core is at level 410mm.

With the Cf-252 neutron sourceplaced outside the core the geometry(see Fig. 1.7) does even not approx-imately fulfil the above prerequisites.On the other hand the counting ratesci (dummy) in the wide range detectorsWR were measured with a fuel dummyand the neutron source in place. Thusan approximate evaluation is possibleby modifying the above equations to:

ci (x)− ci (dummy) = ci (0) ·k/(1−k)

The counting rate ci (0) correspondsto the part from the neutron sourcewhich reaches the fuel element and isamplified. The value of this parametercan again be deduced from the deriva-tive dk/d x close to the critical point.

For the photo neutron case the dataagree within the uncertainty with thecalculation of A. Röhrmoser (FRM II)(estimated uncertainty for calculationand measurement of dk/d x in total15%). For the radioactive neutronsource the maximum deviation was lessthan 0,02 in k at x = 0 (shut down). Itmust be emphasized that the shut downreactivity margin of the control rod fora fresh fuel element is about ρ = (k −1)/k = −0,12 and high compared to -0,02 required by regulation. Thus thepresent evaluation is a methodical in-vestigation and not a necessary prooffor safety of the FRM II.


1.4 Sample environment

J. Peters 1, H. Kolb 1, A. Schmidt 1, A. Pscheidt 1, J. Wenzlaff 1, P. Biber 1


In March 2006, ACCEL Instruments,Germany, delivered our cryogen freesuper-conducting magnet. The magnetprovides a maximum magnetic field of7.5 T and a room temperature bore of100 mm diameter. Further specifica-tions are: split 30 mm, horizontal beamaccess 320°, split angle 3°. A motorizedrotation stage located at low stray fieldlevel is available which allows rotationof sample environment. A second ro-tation stage developed in house is un-der construction and available in early2007. Subsequent to intense laboratorytests a test run was accomplished on thecold three axes spectrometer PANDA inNovember 2006. All tests revealed a sta-ble and reliable running system. Somefurther improvement must be done e.g.rotary feedthrough adapters to uncou-ple circular motion of magnet and cryo-stat from stationary flex lines.

Figure 1.9: 7.5 cryogenfree magnet

1.5 Neutron flux values/spectra at operating FRM II

A. Röhrmoser 1


Neutrons designed for usageat FRM II

The Year 2006 was the first year full withroutine operation of FRM II. Five cy-cles of full burn-up with 1040 MWd eachwere reached.

History within a few words

Core

The neutron physical behavior of thecore during this first full operationshowed up to be exactly or at least ex-tremely close to the theoretical descrip-tions of the reactor design studies, that

were done now more than a decadeago (in the 1980’s [1, 2]). The inte-gral properties of the reactor were cal-culated with a cylinder-symmetric de-scription of the core and its surround-ing. For the core reactivity and burn uppredictions one can reach excellent re-sults that way as verified by reality.

The design was done with the as-sumption of a disturbance by the at thattime unknown user installations (UI) of4% absolute in keff and a necessary re-serve in keff of 2% to cover all uncertain-ties of the calculation or even fabrica-tional tolerance factors.

User installations (UI)

Mid of the 90’s the design task shiftedtotally to the need of secondary sourcedesign. That way one tried to find themost appropriate arrangement for thebeam tubes and the cold (KQ) and hotsource (HQ) in the D2O-moderator ves-sel [3]. All the 10 horizontal tubes andone of the two vertical ones where real-ized very close to the outlines of this de-sign phase. More remarkable changeswere only made later at the tubes fac-ing the cold source in height and width.Further thinner walls have been real-ized at the close core neighborhood.

With this design model of user in-


stallations (3Dmodel96) there were pre-dicted:

1. the decisive flux of neutrons atthe noses that can enter the beamtubes for the usage at experi-ments (s. below) and

2. the disturbance on the core reac-tivity itself.

This design model 3Dmodel96 had arather simplified core, thus being fullyqualified for the ‘beam tube layout’.But also reactivity disturbances couldbe evaluated pretty well. It was calcu-lated a disturbance of the reactor in keff

of 5.6%, thus practically consuming thebulk of reserve in reactivity of the coredesign calculations.

Figure 1.10: Thermal n-flux Φth, cuts at const y-value. The dots are the centers of flux tallybowls, lines are given to guide the eye.

As-built model of userinstallations and core

At the beginning of the new millen-nium the core was realized with all itsbeam and irradiation tubes. TUM cre-ated a quite new 3d model of the coreand D2O-tank content. This model3Dmodel2k took care of all the ‘as-built’-data, that were comprehendedvery diligently. Again the core itselfwas modeled 3d with the involute platesstructure. The output of these new ‘as-built’ calculations was now:

1. The absolute core reactivity isvery close to the results of the 2d-core design calculations withoutUIs and thus confirming late 3d-results of ‘99 [4], too.

2. The disturbance of the reactor inkeff was at 5.0% with ‘as-built’ UI-structures, thus consuming ex-actly half of the design reserve toreach certainly the aimed 52 fullpower (FPDs) operation at 20MWof the core design studies. Theregain of reserve was mainly be-cause of the thinner beam tubefront parts and less expressive as-sumption on absorption of un-known structure additives. Be-sides, this is in absolute agree-ment to the reserve that the re-actor shows in reality with the re-maining control rod margin afteroperating a full cycle of 52 FPDs.

Thermal flux distribution inthe D2O-tank

As a side product of the new core modelone could even produce a 3d-flux mapof the core surrounding. The pro-gram version MNCP5, that inherentlysupports this feature, was not obtain-able to European users till 2007, but itcould be done an adaptation to versionMCNP4C. Vertical asymmetries can bealso obtained in 2d-symmetry, but hereone can especially predict azimuthalasymmetries in the flux distribution,too. Again the axial flux distribution inthe D2O-tank will be describable mostexact.

To obtain a best representable valueover the cycle a calculation was donewith the 3Dmodel2k at about mid ofthe operations cycle (MOL). Figures1.10(and 1.11) show Φth in the core sur-rounding 5 cm below core mid plane.There are given five cuts at constant yposition in the reactor. As structureeasily conceivable is mainly the coreitself, which is cut two times at y=0cm (sharp flux dips). The maximumflux in the D2O is obviously of exactly7.0·1014cm−2/s. At the opposite sideof this maximum point the cold source


Figure 1.11: Thermal n-flux Φth obtained with a 3Dmodel2k calculation 5 cm below coremid plane at about mid of the operations cycle (MOL). Same data are used as in Figure1.10, but now a full 2d-map is drawn. The hot source HQ, the cold source KQ, the beamtubes SR1 and SR5 are shown, the latter showing nearly exact onto the maximum of thethermal n-flux Φth at FRM II. All four installations influence the thermal flux distribu-tion in their surrounding. The very strong flux depression in the center results from thecylindrical fuel element itself.

with its huge beam tubes has depressedthe flux to a lower flux level, what ismore impressive shown in the 2d-fluxmap figure. The cold (and hot) sourceand some beam tubes, which are cuthere, are sketched also in the picture.Some of the beam tubes produce a kindof a fingerprint in the flux map as bestconceivable here at SR1 and SR5. Thelatter tube penetrates nearly direct intothe flux maximum (cmp. below).

As a rough estimation from the for-mer 2d-symmetric contemplations atfull thermal power Po=20 MW it waspredicted a flux level of Φmax,az.-avg.

th =

6.5·1014 cm−2/s, averaged at r=22 cm

Φth (bowl, r=5cm) 10cm thick before nose of beam tube nose [1014cm−2/s]

BOL,Φth rel. error BOL,Φth rel. error EOL,Φth rel. error

model 96 as-built as-built

SR3 3,27 3,1% 3,06 0,28% 3,57 0,26%

SR5 5,66 2,6% 5,49 0,21% 5,63 0,20%

SR7 5,45 2,1% 4,91 0,22% 4,45 0,23%

SR8 4,82 2,8% 5,06 0,22% 4,58 0,23%

Table 1.1: n-flux Φnoseth before the beam tubes obtained with the 3Dmodel96 for BOL and

the ‘as-built’-model 3Dmodel2k for both BOL and EOL. The ‘rel. error’ is the statisticalone.

over the 360° azimuthal angle. This isfully confirmed here.

Flux at thermal beam tubesSR3/5/7/8

For any user the output at the beamends is decisive, not the flux at the noseΦnose

n . But in fact the output must be re-garded rather proportional to the inputΦnose

n .With the design model 3Dmodel96

and the fresh fuel element (begin oflife, BOL) the following Φnose

th valuesare calculated: 5.5·1014cm−2/s for thetwo beam tubes SR5 and SR7, nearly5·1014cm−2/s for beam tube SR8 andmore than 3·1014cm−2/s for tube SR3.

With the new ‘as-built’-model thecorresponding values are expected tobe rather close to the old calculations.The values of the table confirm this andonly SR7 is clearly outside the statisticalerror (3%) of the former ones. The val-ues can be given as follows:

SR5 with 5.5 · 1014, SR3 with 3 ·1014cm−2/s and both SR8/7 at 5·1014.An explanation for the lower value forSR7 must be searched in the modi-fied beam tube, especially at the nose,which is built more stocky; this has amajor effect only on the flux in the con-trol volume at the nose and does notmean less streaming into the beam tothe user exit.

Because the situation BOL is a some-what extraordinary situation with thecontrol rod in the lowest position andmoving strongly upward over the first 2days (Xe poisoning) of nominal poweroperation, the averaged values Φnose

thover the typical cycle life time are ofmore concern for the user. The Figure


1.12 shows the values from Table 1.1 forBOL and compares them to the corre-sponding valuesΦnose,EOL

th from a calcu-lation with a control rod positions STSas for the operational state EOL.

SR3 must obtain a significant gainin flux during reactor operations cycle.SR5 gains again a few %, but the twolow lying tubes SR7 and SR8 have theirhighest flux at BOL (10% higher than atEOL).[1] Böning, K., Gläser, W., Röhrmoser,

A. In RERTR Conference 1988(1988).

[2] Röhrmoser, A. Neutronenphys.Kernauslegung FRM II. Ph.D. thesis,TUM (1990).

[3] Gaubatz, W. Rechnerische Op-timierung der sekundären Neutro-nenquellen des FRM-II. Ph.D. thesis,TUM (1999).

[4] Petrov, Y., Onegin, M., Böning, K. InGatchina-Garching (2001).

Figure 1.12: Φnoseth after the calculations

with the ‘as-built’-model for the extremeoperational states with control rod posi-tions STS time for BOL and EOL. The av-eraged value must be seen a little bit moreclose to the state EOL than BOL. A sketchof SR5 is also shown in Figure 1b. SR3 isabove and SR7 and SR8 are below the coremid plane.

1.6 Reduced enrichment for FRM II

A. Röhrmoser 1, R. Jungwirth 1, W. Petry 1, W. Schmid 1, N. Wieschalla 1


There are intense and widespread in-ternational efforts to develop very highdensity fuels for research reactor con-cepts with reduced enrichment. TUMis fully engaged here, too, and becamea member in the ‘International work-ing group for high density fuel develop-ment’ in 2006.

There exist mainly two pathways thatare investigated intensively.

1. Dispersive fuel in Al matrix and Alcladding: instead of the currentlyused powders as U3Si2 in case ofFRM II one regards mainly UMoalloys. The powder can be pro-duced with the grinding methodor with an atomizing process. Thepercentage of Mo can be chosenfree with the alloy and varies be-tween 6 - 10 wt ‰.

2. The more progressive pathwaywith the same material UMo al-loys but not dispersed or mixedwith a matrix but in form of amonolithic foil. The foils have

to be cladded as usual. One ofthe main challenges with this veryhigh density fuel is the fabricationof full size plates with fine and re-sistable bonding.

By means of collaboration with theFrench CEA, the fuel fabricant AREVA-CERCA, TUM contributes in both fieldsto find ways of progressing the fabrica-tion and the irradiation behavior of thisfuel.

Besides the technical problems withfuel fabrication and plate production itwas followed a quite new conceptionalstudy with the monolithic fuel on baseof a thickness variation of the fuel layer.The results (s. below) were presented inApril and June on RRFM-06 and at anIAEO-meeting on invitation by the Nor-wegian government.

Fuel irradiation program

For the first irradiation program itwas decided to irradiate four test

plates of full size at the MTR reac-tor OSIRIS/Saclay. For this purposesix plates were produced by CERCAwith UMo-Al dispersive fuel (8% wt.Mo) at 50% enrichment. The uraniumdensities were 7 and 8g/cm3. Two ofthe 8g/cm3 plates contain a Si addi-tive, a possible considerable “diffusionblocker”. Before starting irradiationOSIRIS needed an extension of its li-cense for heat flux values in the order of300 W/cm2 for the TUM plates to reacha targeted cladding surface tempera-ture of 100˚C. After some delay with thenuclear licensing the irradiation couldfinally start Sept. 2005 with four platesin two different IRIS devices. Mean-while all four plates have reached theaimed density of 2.3 · 1021fissions/cm3.This value covers any obtainable levelin a future fuel element for FRM II.

Motivated by the conservativeswelling behavior two plates wereirradiated to higher fission densi-ties. Meanwhile, one of the plates


has reached a fission density of 3.4 ·1021fissions/cm3 and still no anoma-lous swelling is observable.

Due to the radioactive load post-irradiation examinations (PIEs) of theirradiated plates have to wait untilmid of 2007. The plates now are al-ready shipped for investigations to CEACadarache.

UMo-monolithic fuelresearch cooperation

For the more progressive pathway therewere collected a lot of new results bythe RERTR program in 2006 so that thevery thin data base till 2005 on irradia-tion behavior is now extended remark-ably. Major results can be summarizedas follows:

1. The swelling of the ‘monolithic’fuel itself shows up on a verycomparable scale to the disper-sive variant.

2. The problem interface switchedfrom the particle surface to theUMo/Al-clad interface. Here avariety of problematic behaviorsshowed up and obviously verydependant on the different pro-cesses for bonding the UMo-foilto the clad.

Since 2005 TUM is in a ‘Cooperation indevelopment for UMo-monolithic fuelplates’ with French CEA and CERCA. Inthe frame of the so called IRIS5 experi-ment full size plates shall be producedand a common TUM/CEA/CERCA irra-diation experiment will be launched assoon as available at OSIRIS with twoplates. To overcome the bonding prob-lems, the CERCA/TUM cooperation in-vestigates new ways of bonding theUMo foil to the clad.

Core design calculations withUMo monolithic fuel

With the availability of UMo monolithicfuel the uranium density in the fuellayer can be increased drastically to themaximum of 15-16g/cm3, only slightlydependent on the Mo percentage varia-tion.

Regardless of the technical feasibil-ity to produce large monolithic fuel

plates, former studies with continued.The conditions for a possible conver-sion scenario are that the power levelhas to be kept at 20 MW, the cycle lengthhas to be maintained at at least 52 days,only marginal losses in neutron flux andneutron quality, similar safety marginsand no essential increase of fuel costs.

A first trial with monolithic fuelplates, the thickness of which changedstep-like at a certain radial width in or-der to avoid unacceptable heating at theouter fuel side, showed up to be veryunfavorable. At a level of 34 % enrich-ment more than 40 kg Uranium (14 kgU-235) had to be used and the localmaximum power turned out to be largerby 20 % compared to the actual U3Si2

element.The idea was now to introduce a con-

tinuous gradient in fuel thickness. Onekey reason is, that a step gradient in Udensity as currently used is not feasiblewith monolithic fuel at all. Again theU235 amount has to be led to a practicalvalue at about 10 kg or even below. Allthe demanded aims were achieved now.

With those studies of radial continu-ous variation of fuel layer thickness thepower distribution could be progressedvery much. Fig. 1.13 displays the ra-dial thickness variation of the uraniumfoil and the radial heat form factor in acomparison to the currently used fuelelement. The heat form factor is re-markably better than even with the cur-rent core of relative low density. Only9.1 kg of fissionable material U235 areneeded. The maximum thermal flux isdepressed by 9% over the whole cyclewhen compared to the actual HEU fuelat the same thermal power of 20MW.

The solution opens some other op-portunities for parameter studies. Notall of them were followed up to last ex-tent in the moment, since the solutionis afflicted with to many unknowns, butto give an example:With a thicker meat, lets say 300µm in-stead of 250µm and the same thick-ness shaping in this calculation, the to-tal output on FPDs would be extendedby about 20%.

Metallographic studies

The build up of the harmful UMo-Al in-terdiffusion layer (IL) at the surface ofthe UMo powder or foils during irradi-ation has to be better explained.

In a quite new approach TUM irra-diated UMo samples by the domesticheavy ion facility in order to distinguishbetween physical processes, that are re-garded to be responsible for the harm-ful buildup. In contrast to in-pile ir-radiations the samples can be studiedeasily by all kind of metallurgical meth-ods and without time delay, what makesthe procedure fast and less expensive.First results of measurements on ther-mal conductivity of those samples be-fore and after beam bombardment weregathered in a diploma thesis, too.

The approach was followed by theFrench CEA in their efforts to explainand distinguish between different UMofuel powder and additives.


Figure 1.13: Calculated radial power form factors (left scale) in FRM II at BOL with U3Si2-Al HEU fuel as currently and with the newlystudied UMo monolithic MEU fuel. The thickness of the monolithic fuel layer is variable and up to 250µm in the central region (rightscale).

2 Thin films and large scale structures 19

2 Thin films and large scale structures

2.1 KWS-1 and KWS-2 Small-Angle Neutron Scattering Diffractometers

H. Frielinghaus 1, A. Radulescu 1, P. Busch 1, V. Pipich 1, E. Kentzinger 2, A. Ioffe 1, D. Schwahn 2, R. Hanslik 3, K. Dahlhoff 3,M. Heiderich 2, G. Hansen 3, D. Richter 1,2

1JCNS, Research Centre Jülich, outstation at FRM II2IFF, Research Centre Jülich3ZAT, Research Centre Jülich

With the shut-down of the Jülich re-actor FRJ-2 all high-performance in-struments started to move to the Mu-nich reactor FRM II. The two Small-Angle Neutron Scattering diffractome-ters found a space at the very end of theneutron guide hall and partially in thering laboratory of the old FRM-I reac-tor. The following setup should split thelarge number of possible experimentsto the two SANS diffractometers. KWS-1 should be optimized for high reso-lution, GISANS, and polarized neutronexperiments, whereas KWS-2 should beoptimized for high intensity. Thus, allextraordinary requirements should beserved by KWS-1, whereas conventionaland high intensity requirements shouldbe served by KWS-2. A continuous up-grading will bring KWS-1 and KWS-2to the same state in a moderate future.While the old state of the instrumentswas rather conventional, the new tech-niques of this refurbishment will meetthe requirements of world leading sci-entists. These things are described inthe following:

Polarization and He3 filterfor analysis

Transmission neutron polarizers areoptimized for maximum transmissionat 4.5, 7 and 14 Å. To keep the neu-tron beam polarization along the col-limator the neutron guiding elementsare coated with non-magnetic m =1.4 NiMo-Ti supermirrors. Polarizationanalysis of the scattered beam will becarried out by a He3 neutron spin filterof 15 cm diameter immediately behind

the sample. The filter will be placedin vacuum and the direction of the He3

nuclear polarization will be controlledby the adiabatic fast passage method.

Neutron lenses

Neutron lenses will become a standardfor SANS diffractometers now. Quitefrequently spherical biconcave MgF2

lenses with a diameter of approx. 3 cmare being used [1]. We will install MgF2

parabolic lenses of 5.2 cm diameter inorder to avoid spherical aberration andmake use of the whole beam area givenby the neutron guide [2]. The lenses willbe placed in vacuum inside the collima-tor just in front of the sample and can bemoved in and out of the neutron beamby a motor controlled drive similarly asthe neutron guide elements. Further-more, the lenses will be cooled to ap-prox. 70 K in order to avoid thermal dif-fuse scattering. This cooling allows for again of intensity by a factor of approx. 3for 7 Å neutrons and the maximum ir-radiated area. The whole constructionallows for a lens operation with ease.

High resolution detector andchopper

A small high resolution detector willbe installed in front of the conven-tional detector, and can be moved toany position covered by the large detec-tor. When measuring at larger scatter-ing angles the resolution is dominatedby the wavelength distribution. There-fore, a chopper placed behind the selec-tor in combination with a time-of-flight

analysis will select ∆λ channels of 1%and below. The small detector can alsobe placed to the beam centre, and thena USANS experiment is carried out withno need for a narrow wavelength distri-bution. In combination with the lensesa scattering vector of 10−4 Å−1 will be-come available with a less stringent Q−2

restriction compared to the Q−4 law ofconventional SANS [3].

GISANS

For the investigation of magneticnanoparticles on surfaces or heteroge-neous polymer films classical SANS intransmission geometry suffers mainlyfrom the low scattering volume. If al-ternatively the beam impinges under aglancing angle (typically 1 or lower) thelarge footprint of the beam increasesthe scattering volume as well as the pro-jection of the coherence volume ontothe sample surface. In this way infor-mation about lateral as well as verticalcorrelations on length scales between10 and 2000 Å can be obtained simulta-neously at KWS-1. Due to the very smallincident angle this technique has beenestablished as grazing incidence smallangle neutron scattering (GISANS). Ad-ditionally, as for many materials belowa certain incident angle total externalreflection occurs, GISANS can also beused for the investigation of surfaces orinterfaces.

The positioning of the sample will berealized by a hexapod system, allowinga high accuracy of the incident angleas well as the rotation around the sur-face normal for anisotropic films. Thepolarization analysis described above

20 2 Thin films and large scale structures

will allow for probing also the mag-netic properties of magnetic nanoparti-cles or layered magnetic materials. Toresolve also structures in the small an-gle regime, where the in-beam compo-nent of the scattering vector contributessignificantly, the small detector will beused.

Present status

The KWS-1 and KWS-2 instrumentscurrently under reconstruction at theFRM II reactor are positioned at theend of the NL3a-o (KWS-2) and NL3b(KWS-1) neutron guides. Both instru-ments are identical from the conceptand consist of a 20 m collimator and a20 m detector tube. Due to the fore-seen upgrades, which are described inthe previous section, the positioningof instruments within the FRM II neu-tron guide hall and the particular beamcharacteristics at FRM II (beam height,beam size, etc.) the constituent parts ofthe instruments were subject of majorchanges. The first change relates to theincrease of the beam size from 30x45mm2 (in Jülich) to 50x50 mm2 (at FRMII). Therefore, both instruments will beequipped with a new system of neutronguides (for each instrument 18 piecesx 1m length, with an m = 1.4 NiMo-Ticoating) inside the collimation case andnew collimation apertures. The use ofthe old collimation housing required amodification of the working principleof the guide segments and apertures.

Figure 2.1: View along the KWS-2 colli-mation housing and mechanical compo-nents of the collimation system; to the leftthe collimation housing of the KWS-1 in-strument can be seen.

Thus, the old system of one carriagesupporting a neutron guide and a colli-mation aperture alternatively moved inor out of beam was replaced to that ofdistinct carriers for the neutron guideand aperture allowing them to move in-dependently of each other Fig.2.1). Cur-rently, the collimation housing and themechanical parts of the collimation sys-tem are installed and adjusted for bothinstruments.

Due to implementation of the opti-cal focusing elements (parabolic lenses)the end segment of the collimation sys-tem towards the sample position wasmodified both technically and func-tionally. Thus, the old variable aper-ture defining the 1 m collimation lengthwas cancelled and the first 1 m segmentof the collimation housing was trans-formed into a separate chamber des-ignated to host the carrying and thecooling system of lenses. In this way,the smaller collimation distance avail-able will be of 2 m and defined by anadjustable aperture fixed on the sep-aration window between the collima-tion and the housings of the lenses.Concerning the detection system, it isplanned that, due to the limited spaceavailable for KWS-1 and KWS-2 instru-ments, the 20 m long detector tubeswill be pushed out of the present neu-tron guide hall and located partially intothe ring laboratory surrounding the oldFRM-I reactor. Thus, the installation of

Figure 2.2: Inside view of the KWS-2 detector housing.

the complete detector housing will bepossible only after joining the FRM IIguide hall with the FRM-I ring labora-tory (estimation: summer of 2007). Forthe currently available space into theFRM II guide hall a 14 m long detectortube was installed for the KWS-2 instru-ment.

The mechanical elements for guidingand transporting the detector (Fig.2.2)and cables inside the housing havebeen installed and adjusted. The vac-uum system (pumps and controllers)was also installed and the tightnessof the detector tube was successfullytested. The work of refurbishing theKWS-1 and KWS-2 instruments at FRMII reactor is currently in progress.[1] Choi, S.-M., Barker, J. G., Glinka,

C. J., Cheng, Y. T., Gamme, P. L. Jour-nal of Applied Crystallography, 33,(2000), 793–796.

[2] Alefeld, B., Schwahn, D., Springer, T.Nucl. Inst. a. Meth. in Physical Re-search A, 274, (1989), 210–216.

[3] Lengeler, B., Schroer, C. G., Richwin,M., Tümmler, J., Drakopoulos, M.,Snigirev, A., Snigireva, I. Appl. Phys.Lett., 74, (1999), 3924–3926.


2.2 MIRA – The beam line for very cold neutrons at the FRM II

R. Georgii 1, N. Arend 1, P. Böni 2, H. Fußstetter 1, M. Janoschek 3, T. Hils 2, A. Mantwill 1, S. Mühlbauer 2, C. Pfleiderer 2,R. Schwikowski 1, Shah Valloppilly 2

1ZWE FRM II, TU München2Physics Department E21, TU München3Laboratory for Neutron Scattering ETHZ& PSI, CH-5232 Villigen

MIRA is a versatile instrument forvery cold neutrons (VCN) using neu-trons with a wavelength λ > 8 Å (seeFig. 2.3). The flux at the sample po-sition is 5 · 105 neutrons/(cm2 s) unpo-larised. It is situated at the cold neutronguide NL6b in the neutron guide hallof the FRM II. As the instrument set-upcan be changed quickly, MIRA is ideallysuited as a testing platform for realizingnew instrumental set-ups and ideas. Inparticular, MIRA is unique in its possi-bilities of combining different neutronscattering methods as:

• Polarized or non-polarized reflec-tometry.

• Spherical Polarimetry• Polarized or non-polarized small

angle scattering (SANS).• Classical NRSE (Neutron Reso-

nance Spin Echo) setup as well asusing the MIEZE principle.

This year MIRA was successfully oper-ated for 5 reactor cycles, it means for260 days (!). In total, 28 external and33 internal proposals, several test andservice measurements were performed.One example is shown in section 9.1.One Ph.D. thesis (Multi-Mieze measure-ments by Nikolas Arend (FRM II)) andserval measurements for Diploma the-sis were finished using mainly data fromMIRA.

Figure 2.3: MIRA equipped with the MIEZEoption, a closed cycle cryostat and a mag-net.

A new polarising multilayer monochro-mator was taken into operation. Thisallows now full polarisation analysis,3D-polarimetry and MIEZE measure-ments. The MSANS principle was testedfor later operation on MIRA. The exist-ing magnet was upgraded for automaticadjustment of the field perpendicularto the neutron beam and independentfrom the sample movement.

Polarised reflectometry

Fe/Cr thin films and multilayers remainone of the most extensively investigatedsystems in thin film magnetism. Somepioneering discoveries in thin film mag-netism like interlayer exchange cou-pling, giant magneto-resistance (GMR)etc. have been made on this system.Complex magnetic structure in Cr ren-der intriguing magnetic properties inbulk as well as in thin films especiallywhen Cr is used as a sandwich layer be-tween two ferromagnetic layers. Earliermodels of interlayer exchange in Fe/Crmultilayers were based on an oscilla-tory RKKY-type exchange coupling andquantum-well behaviour of the elec-trons in the Cr spacer layer. Later,calculations based on the commensu-rate and incommensurate spin densitywaves (SDW) in Cr, and direct observa-tion of SDW by neutron scattering sug-gested that Fe/Cr multilayers displaymore rich variety of magnetic phenom-ena.

A variety of experimental techniqueshave been employed for the investiga-tions that extend from bulk to layerresolved magnetometry like in Polar-ized Neutron Reflectometry (PNR). PNRreveals the vectorial layer magnetiza-tion and is very sensitive to the par-allel and perpendicular components ofmagnetization with respect to the neu-tron polarization. Thus, bilinear and bi-quadratic types of layer configurations

can be easily resolved with depth sen-sitivity. The origin of bi-quadratic ex-change has been attributed to intrin-sic properties of the spacer layer, dipo-lar fields resulting from rough surfaces,super-paramagnetic impurities withinthe spacer, spacer thickness fluctua-tions etc. It has also been pointedout that the bi-quadratic exchange cou-pling has a dependence on the SDW.

The objective of the present workis to examine the magnetic configura-tions of the layers as a function of mag-netic field and temperature of epitaxi-ally grown [Fe(4 nm)/Cr(1.5 nm)]10 onMgO single crystal by PNR, to corrob-orate with bulk magnetization resultsand to find a model to explain the tem-perature dependence of hysteresis be-haviour. Epitaxial, Fe (4 nm)/Cr (1.5nm)]10 multilayers were grown by mag-netron sputtering on a MgO(100) sin-gle crystal wafer of 1 x 1 cm. Polar-ized neutron reflectometry with full po-larization analysis was implemented atthe MIRA reflectometer by using flip-per coils in conjunction with polariz-ing benders. Due to small size of thesample, a clear total reflection regionwas not obtainable marred by the largefootprint of the beam. Nevertheless,clear effects of PNR are immediately vis-ible in the corrected data as illustratedfor a typical case, T=10K and H=500 Oein Fig. 2.4(top). The splitting of nonspin-flip (NSF) channels + + and - -,and the presence of half order Braggpeak in the spin-flip (SF) channel + -and - + arising from the double peri-odicity of the magnetic lattice are in-dicative of the PNR with polarizationanalysis at MIRA. The experiments wereperformed around the magnetic (half-order) peak and its intensity as a func-tion of applied magnetic field and tem-perature was measured. Fig. 2.4(bot-tom) summarizes the SF scattering in-tensity (integrated intensity of the halforder peak) at selected fields and tem-


1 2 3 4 5 6

1E-3

0.01

0.1

PN

R (a

rb. u

nits

)

theta (deg)

+ + - + - - + -

T=10 K, H ~ 500 Oe

-100 0 100 200 300 400 500 600 700 800 900 1000 1100 12000.00

0.05

0.10

0.15

0.20

0.25

0.30

0.5

• (I-+ +

I+-)

H (Oe)

Integrated SF intensity T10K T100K T150K T200K T300K

Figure 2.4: (top) PNR of Fe/Cr multilayers at10K and H ≈ 500 Oe. (bottom) Depen-dence of SF intensity as a function of fieldand temperature.

peratures.In general, at remanence after expos-

ing the sample to a high negative field,the Fe layers are found to order in anantiferromagnetic configuration as in-dicated by a high NSF intensity and alow SF intensity. By the applicationof a small magnetic field of 50 Oe, theNSF intensity drops to low values andthe SF intensity increases. With fur-ther high magnetic fields, the SF in-tensity reaches a maximum at a cer-tain field and thereafter decreases grad-ually. The highest SF scattering in-dicates the highest deviation of mag-netization from the collinear configu-ration and its field dependence indi-cates the evolution of magnetizationin an exchange coupled magnetizationreversal process. Clearly, the magni-tude and behaviour of the field depen-dence of SF intensity at various temper-atures, suggest the role of Cr on inter-layer exchange and its temperature de-pendence.

3D-Polarimetry

MnSi crystallises in the cubic spacegroup P213 with the lattice parame-ter a = 4.558 Å. MnSi is an itinerant-electron ferromagnet with a Curie tem-perature TC of about 29 K and an or-dered magnetic moment of 0.4 µB oneach Mn site. Its magnetic structurein zero field is a long-period ferromag-netic spiral with the propagation vector(2π/a)(ζ,ζζ) with ζ = 0.017 resulting ina period of approximately 180 Å [1, 2]along the [111] direction.

MnSi attracts presently high interestdue to the speculation about an inter-mediate phase between the helical andthe paramagnetic phase which may betriggered by the soft longitudinal fluc-tuations in MnSi. Roessler et al. [3]calculated that the magnetic state mayform skyrmion patterns in this phase,e.g. the pinning of the spiral is cancelledby the fluctuations. This skyrmion con-densate may only be stable when a vari-ation of amplitude of the magnetisationis allowed. This intermediate magneticground state should in principle be ob-servable by means of Spherical NeutronPolarimetry (SNP) as this technique isvery sensitive to changes in the mag-netic structure [4, 5, 6].

For the experiment the MuPAD op-tion [5, 6] was installed on the verycold neutron instrument MIRA. In thisSNP setup we used MIRA in small an-gle diffraction mode with a wavelengthof 9.7 Å. The sample was adjustedon the instrument to fulfil the Bragg-condition on the magnetic satellite re-flection (0.017, 0.017, 0.017). In orderto have enough intensity to perform fullpolarisation analysis in the vicinity ofTc we adjusted MIRA to use the fullbeam divergence and thus, sacrificedQ-resolution (Slit after monochromator

Measurement TheoryPout x y z x y z

x 0.836(7) 0.07(1) 0.09(1) 1 0 0Pi n y 0.880(1) 0.062(3) 0.091(3) 1 0 0

z 0.880(1) 0.059(3) 0.083(3) 1 0 0

Table 2.1: Polarization matrix observed on the magnetic satellite reflection (0.017 0.0170.017) at T ≈ 20 K is shown together with the expected matrix for this magnetic struc-ture. The elements where the x-component of the final polarization vector is measuredare reduced from one due to the flipping efficiency of the polarisation analysator.

15mmx50mm, circular slit in front ofsample ® = 13 mm).

When the optimum crystal positionwas adjusted we started to measure thetemperature dependence on all matrixelements of the polarisation matrix (s.Fig 2.5). A polarisation value of about 92% (s. e.g. Table 2.1) was observed on thexx, yx and zx terms of the matrix in thehelical magnetic phase. This is due tothe chiral term i (M×M) which is gener-ated by the anisotropic Dzyaloshinski-Moriya (DM) interaction [7, 8] thatarises because of the non-central ar-rangement of the Mn moments in theunit cell. The chiral term additionallypolarizes the neutron beam. All otherterms are small and interpreted as zero.This is the polarization signature of ahelical spin arrangement if the scatter-ing vector Q is parallel to the propa-gation vector of the helix. However,when the temperature is increased overthe transition temperature Tc1 the threeterms xx, yx and zx slowly start to de-crease until they go to zero at approx-imately Tc2 = 31.5 K. This seems to in-dicate an intermediate magnetic phasebetween the helical and the paramag-netic phase as for a direct magneticphase transition from a helical arrange-ment to an unordered state the heli-cal signature is expected to be observedjust until the transition temperature Tc1

is reached, whereas a matrix with zeropolarisation on all elements is expectedjust directly above Tc1. This is shown inthe graphs in Fig 2.5 marked as ’Theory’.

Thus, the result supports the propo-sition of an intermediate phase in MnSiby Rößler et al. [3].The experimentcan be regarded as very successfuleven though special care has to be un-dertaken to analyse the data keepingin mind the effects of the coarse Q-resolution.


Figure 2.5: The temperature dependence of the matrix measured on the magnetic satellitereflection (0.017, 0.017, 0.017) is shown. Remarkable is the smooth transition from thehelical below Tc1 into the paramagnetic phase above Tc2. The lines show the expecta-tion for the behaviour in the case that there would be no intermediate magnetic phase.

Spin echo option - MIEZE

In order to establish the MIEZE andNRSE techniques as standard measure-ment options at the instrument MIRA,new resonance spin echo hardwarehas been built and tested during thelast few years. In 2006 several differ-ent measurements using MIEZE/NRSEhave been performed, among those are

• Verification of the Multi-levelMIEZE principle. The time-dependent, high-frequency si-nusoidal signal of a MIEZE instru-ment can, just like the signal of anNSE/NRSE instrument, be usedfor quasi- and inelastic measure-ments with a wide range of appli-cations. The MIEZE instrument,however, has a strong connec-tion with a time-of-flight inter-ferometer and is therefore poten-tially well-suited for fundamentalphysics experiments. One suchapplication is the verification of

the longitudinal Stern-Gerlach ef-fect, which manifests itself in thetemporal splitting of the spin-up and spin-down states of acold neutron beam when passingthrough resonantly tuned fieldsof an NRSE flipper coil. To actu-ally see the splitting of a cold neu-tron pulse, such pulses must havea sharp width and a sufficientseparation in time. These pre-requisites are difficult to achieveby conventional beam chopping.A Multi-level MIEZE instrument,which could provide those kind ofpulses, consists of several stackedsingle MIEZE parts, all tuned tohave a common focusing point.This principle and the predictedpulse form and sharpening [9]was successfully verified at MIRAwith a two-level setup. Fig. 2.6shows the single and two-levelsignals and the respective non-linear fits.

Figure 2.6: MIEZE and Multi-level MIEZEsignal data with fits.

• MIEZE measurements withstrong magnetic fields and acryostat at the sample region.One of the strengths of MIEZEis the freedom the experimenterhas in arranging the sample re-gion compared to conventionalNRSE. Since the MIEZE signal isalready prepared after the sec-ond analyzer, it is easily possibleto e.g. apply strong magneticfields, do measurements on fer-romagnetic samples, or realizelong scattering geometries. Sincethese advantages do not seem tohave been fully recognized in thepast, we performed a showcasemeasurement: A MIEZE setupat MIRA was equipped with a 2kG solenoid and a cryostat (seeFig. 2.3). The magnetic field atthe sample region was graduallyincreased while monitoring thecontrast of the MIEZE signal. Ifthe strong magnetic field (i.e. itsstray fields) would have destroyedthe beam polarization, the initialcontrast of ≈ 80% would have de-creased significantly. The 2D scanin Fig. 2.7 shows that this was notthe case, proving the suitability ofMIEZE for such measurement en-vironments. An interesting sam-ple to investigate with MIEZE ise.g. MnSi with its chiral magnetic


Figure 2.7: 2D scan of MIEZE signal vs.solenoid current. The signal contrast ofapprox. 80% is more or less unaffected.

structure that is revealed at lowtemperatures and high magneticfields.

Multiple Small AngleNeutron Scattering(MSANS)

Research on polymers, colloid systems,cements, microporous media, are ex-amples of a rising field, where µm-correlations play a crucial role. Smallangle X-ray and neutron scattering(SAXS and SANS) typically measure lat-eral correlation lengths in the 0.01 to1 µm range, the q-resolution rangingup to 10−3Å−1. To measure largerµm correlations with neutrons, the q-resolution has to be improved and vari-ous specific instruments have been de-signed. The technique is commonlyknown as USANS (ultra small angleneutron scattering). However thesemethods are sensitive to scattering onlyin one dimension and often suffer fromintrinsic small angle scattering due tostructure material in the beam.

Here we propose MSANS (Multi holeSANS), a new USANS option for a stan-dard long baseline SANS instrument. Ituses the common SANS infrastructureexcept for the detector, which requiresenhanced spatial resolution. We aimat improving the q-resolution to about10−5 Å−1 at 10 Å, so correlations up to60 µm should be possible.

Using multi-hole apertures at theentrance (Me ) of the collimator andnear the sample (Ms ) with lattice con-stants ae , as and hole diameters de , ds

Figure 2.8: (Top) Number and Intensity of spots depends on number of holes of sample andentrance mask (see Fig. 2.9). (Bottom) Width of spots at detector depends on width ofentrance and sample aperture.

respectively and with the choice

Ge,s = 2π

ae,s

Ge ·L1 = (Gs −Ge ) ·L2

Gd =Gs −Ge

an intensity pattern of well separatedpeaks with lattice constant ad in the de-tector plane is observed (ad = 2π/Gd )(see Fig. 2.8).

Short range correlations in the sam-ple may lead to significant overlap,however typical SANS intensities dropvery rapidly with increasing q, and over-lap will not be fatal in many cases. Setsof apertures with different relations ae,s

/ de,s (d diameter of hole apertures)can be used to adapt the pattern to thedemand. In MSANS, resolution is de-coupled from intensity, as long as thetransmission of apertures is kept con-stant. The increase in q-resolution inMSANS is typically one order of magni-tude, compared to SANS at equal inten-sity. The gain originates from the reduc-tion in q-range in MSANS and the in-crease of the input guide cross sectionand its divergence. Diffraction fromthe aperture holes of typically 1mm are

not yet crucial, as the beam correlationlength is only in the µm range.

Prototypes of a set of multihole-apertures based on cadmium andcoated with typ. 13 µm 10B were pro-duced with the following properties:hole diameter de,s = 0.5/1mm; latticeconstant ae,s = 2.5/5mm. The proposedconfiguration leads to FWHM ≈ 1mm.

Figure 2.9: (top) MSANS setup, (bottom)Multihole apertures


!

!"#$%"&'()*+,-").'$"&

!""!!"!#"!$"!%"!&"!'"!("!)"!*"#""

" ' !# !) #% $" $'

!"#$%"&'()*/$-'(&0,*1223

4#(5-0(),

+,-.,/0-1

Figure 2.10: (top) 2d-Intensity map (arbitrary units) showing resolution in real-space, (bot-tom) Approximation of horizontal resolution due to direct superposition of intensity iny-dir.

In 2006 a first setup on the MIRA beam-line at FRM II was accomplished andpreliminary measurements (Fig. 2.10)mainly concerning the basic principlewere made.[1] Shirane, G., Cowley, R., Majkrzak,

C., Sokoloff, J., Pagonis, B., Perry,C. H., Ishikawa, Y. Physical ReviewB, (1983).

[2] Roessli, B., Böni, P., Fischer, W., En-doh, Y. Phys. Rev. Lett., (2002).

[3] Rößler, U. K., Bogdanov, A. N., Pflei-derer, C. Nature, 442, (2006), 797–801.

[4] Tasset, F. Physica B, 156-157, (1989),627.

[5] Janoschek, M. "". Master’s the-sis, Technical Universtiy Munich,http://mupad.wired-things.de(2004).

[6] Janoschek, M., Klimko, S., Roessli,B., Medarde, M., Böni, P. Annual Re-port E21, 21.

[7] Dzyaloshinskii, L. J. Phys. ChemSolids, 4, (1958), 241.

[8] Moriya, T. Phys. Rev., 120, (1960),91.

[9] Arend, N., Gähler, R., Keller, T.,Georgii, R., Hils, T., Böni, P. PhysicsLetters A, 327/1, (2004), 21–27.


2.3 N-REX+ – The Neutron/X-ray contrast reflectometer for materialsscience

A. Rühm 1,2, M. Nülle 1, F. Maye 1, J. Franke 1,2, U. Wildgruber 1,2, J. Major 1,2, H. Dosch 1

1Max-Planck-Institut für Metallforschung, Stuttgart2Max-Planck-Institut für Metallforschung, Stuttgart, outstation at FRM II

During the year 2006 the basic com-missioning of the neutron reflectome-ter with add-on X-ray option, N-REX+[1], has been completed. The major-ity of instrument operation modes havebeen implemented and undergone firstexperimental tests on real samples. Thisincludes conventional neutron reflec-tometry, X-ray reflectometry, and spin-echo resolved grazing incidence scat-tering (SERGIS) on solid samples. Theimplementation of SERGIS includes po-larization control as it is also requiredfor polarized neutron reflectometry. Atilt of the neutron beam, as requiredfor the study of free liquid surfaces, hasbeen realized, but is not yet automa-tized.

Results

In 2006 we have focussed on the mainnovelties provided by N-REX+, i.e. in-situ neutron/X-ray contrast variationand the novel SERGIS technique [2].

The results of a combined neutronand X-ray reflectivity measurement ona nickel/titanium multilayer on silicon

-0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

100

101

102

103

104

Inte

nsity

[cps

]

!"#

0.0 0.5 1.0 1.510-4

10-3

10-2

10-1

100

101

102

103

Inte

nsity

[cps

]

! "#$

Figure 2.11: Neutron and X-ray reflectivity curves measured on a [Ni(5 nm)/Ti(5 nm)]8/Si multilayer before and after annealing and inter-diffusion at 300 C.

substrate are shown in Fig. 2.11 Thecorresponding experimental setup isshown in 2.12. The multilayer was mea-sured i) after sputtering at room tem-perature and ii) after subsequent an-nealing in vacuum for 5 hours at 300 C.From the electron and neutron scatter-ing length density profiles derived fromthe reflectivity curves we conclude, thatafter annealing the nickel layers havecompletely disappeared and Ni4Ti3 lay-ers have formed instead due to inter-diffusion. Only 1.4 nm thick titaniumlayers remain, interlaced between theNi4Ti3 layers. Interestingly, this alloyphase is instable in the bulk, but itcan be stabilized in the reduced dimen-sion of a thin-film geometry. We areplanning a more detailed investigationof the stability of these alloy layers, aswell as an extension of the combinedneutron/X-ray reflectometry techniqueto magnetic systems and to in-situ stud-ies.

Regarding the second focus area ofN-REX+, we have installed and testedthe spin-echo setup and conducted firstSERGIS experiments on reference sam-

ples. First scientific studies on realsamples have also been started recently.Fig. 2.13 shows N-REX+ in spin-echooperation mode. Visible is the mumet-all housing extending over the sam-ple region and two neutron resonancespin-echo coil assemblies, upstreamand downstream of the sample. A firstSERGIS data set obtained on an opticalgrating with 3600 lines per mm is shownin Fig. 2.14. The polarization P of the in-plane integrated diffuse scattering sig-nal from this sample at fixed incidenceand exit angle is plotted (after correc-tion for instrumental imperfections) asa function of the spin-echo length lSE.P (lSE) is proportional to the in-planepair-correlation function g (r = lSE) ofthe sample in real space [2]. The po-sition of the first peak in the SERGISdata corresponds well with the knowngroove period of 280 nm of the grating.This experiment demonstrates the suit-ability of SERGIS for the characteriza-tion of mesoscopic in-plane structuresin the 100 nm to 1000 nm range, whichare for intensity reasons unresolvableby conventional neutron reflectometers


Figure 2.12: The N-REX+ sample table withcompact X-ray source and detector unitsmounted for simultaneous neutron/X-ray reflectivity experiments. The neu-trons are impinging from the monochro-mator hutch on the right, the neutronarea detector (not shown) is located onthe left.

Figure 2.13: N-REX+ in neutron resonancespin-echo operation mode, with the neu-tron monochromator hutch on the rightand the detector tower on the left.

with the GISANS technique.

Outlook

In the next months, the various opera-tion modes of the instrument N-REX+will be further optimized and automa-tized for routine application. Figures tobe optimized are the background level,the neutron flux, the required mea-surement times, and the measurementspeed.[1] Rühm, A., Wildgruber, U., Franke,

J., Major, J., Dosch, H. In NeutronReflectometry, A Probe for Materi-als Surfaces, Proceedings of a Tech-nical Meeting organized by the Inter-national Atomic Energy Agency andheld in Vienna, 16-20 August 2004,Vienna, Austria, 161–175 (2006).

[2] Major, J., Dosch, H., Felcher, G. P.,Habicht, K., Keller, T., te Velthuis, S.G. E., Vorobiev, A., Wahl, M. PhysicaB, 336, (2003), 8–15.

-100 0 100 200 300 400 5000

20

40

60

80

100

120

Pola

rizat

ion

[%]

Spin-echo length [nm]

Figure 2.14: First successful SERGIS mea-surement at N-REX+ on an optical grat-ing. An AFM image of the grating is shownon top. In the SERGIS experiment, thegrooves were aligned parallel to the inci-dent beam direction. The SERGIS datacurve is explained in the text.

2.4 SANS-1, the Small-Angle Scattering instrument

R. Gilles 1, B. Krimmer 1, A. Ostermann 1, O. Nette 2, A. Vogel 2, R. Kampmann 1,2, H. Türck 1, P. Jüttner 1, J. Krüger 1, K. Zeit-elhack 1, A. Schreyer 2, W. Petry 1

1ZWE FRM II, TU München2GKSS Forschungszentrum Geesthacht

Major tasks in 2006 were test andinstallation of the Z-translation andomega tilt (Z/tilt) stage for the selec-tor, the first start of the medium res-olution velocity selector delivered byEADS/Astrium and the test and installa-tion of the instrument shutter togetherwith a monitor at the end of the S-shaped neutron guide (see Fig. 2.15).A small z-translation moves this moni-

tor into and out for measuring the whitebeam. The Z/tilt stage enables the userto choose between a high intensity ve-locity selector with medium resolutionof ∆λ/λ = 10 %, a neutron guide forwhite beam experiments or a high reso-lution velocity selector (optional). Tilt-ing of the complete stage allows an ad-ditional adjustment of the selector res-olution in combination with one of the

velocity selectors [1]. Monitor 2 is fore-seen as one of the instrument monitorswhich gives a value for the neutrons en-tering the polarizer and collimation sec-tion (Fig. 2.16).

Another significant step has been theconstruction and set-up of the veloc-ity selector shielding cube. Fig. 2.17shows the Z/tilt stage mounted and analuminum cage around it. This cage


Figure 2.15: Schematic drawing of the new SANS-1 instrument.

Figure 2.16: Schematic set-up of the velocity selector area.

works as a carrier for the lead shielding(thickness 100 mm). B4C mats are cov-ering the inner lead surface.

A first neutron distribution measure-ment at the end of the S-shaped neu-tron guide was performed [2]. In Fig.2.18 is a comparison of the simulatedneutron distribution and a measure-ment of the white beam just before theselector position. Fig. 2.18a showsthe first measurement of the beam pro-file (line of sight against the beam) di-rectly after the S-shaped neutron guide.Both, the simulation and the measure-ment show in the upper half of thetwo-dimensional pattern higher inten-sity than in the lower half (see Fig.2.18b) which is caused to the S-shapedneutron guide positioned in vertical.

The result of a Monte Carlo simulationfor this position is given in Fig. 2.18b.The averaged beam profile in horizon-tal and vertical direction is shown in Fig.2.18c and Fig. 2.18d respectively. Asexpected there is slight decrease fromthe right to the left side (compare Fig.2.18c). This is due to the horizontalcurvature (R=2100 m) of the last 12.2 msection of the S-shaped neutron guide.The vertical intensity distribution (seeFig. 2.18d) is not reproduced by thesimulation. Further Monte Carlo simu-lations have shown that the position ofthe intensity maxima are strongly corre-lated with the illumination of the beamtube nose and therefore with the fillingstate of the cold source. The filling stateof the cold source slightly differs from

one reactor cycle to another. For theshown simulation results a completelyfilled cold source with a nearly emptydisplacement body was assumed.

The next component which will beinstalled is the collimation set-up. Thevacuum chambers are currently builtand most of the internals (optics, trans-lation systems etc.) are under construc-tion. At the moment the containers ofthe vacuum chamber will be preparedfor vacuum tests. One of the seven con-tainers is shown in Fig. 2.19.[1] Gilles, R., Ostermann, A., Schanzer,

C., Krimmer, B., Petry, W. Physica B,385-386, (2006), 1174–1176.

[2] Gilles, R., Ostermann, A., Petry, W. J.Appl. Cryst., (2006). Submitted.

Figure 2.17: First selector installed on site ina Z/tilt stage. (Shielding not shown in thispicture).


Figure 2.18: Beam profile at the end of the S-shaped neutron guide: (a) Profile measuredwith neutron imaging plate, (b) Monte Carlo simulation, (c) Averaged horizontal beamprofile of the measured data (red) and the Monte Carlo simulation (blue) and (d) Aver-aged vertical beam profile.

Figure 2.19: Vacuum chamber.

2.5 TREFF@NOSPEC : A facility dedicated to test neutron opticaldevices

U. Rücker 1, A. Ofner 2, S. Mattauch 3, A. Ioffe 3, T. Brückel 1, G. Borchert 2

1IFF-Streumethoden, Forschungszentrum Jülich2ZWE FRM II, TU München3JCNS, Forschungszentrum Jülich, outstation at FRM II

The Temporary REFlectometer Facil-ity (TREFF) (Fig.2.20) is a dedicated re-flectometer and diffractometer for thetest of neutron optical devices. It isa joint project of the Jülich Center forNeutron Science (JCNS) and the Neu-tron Optics Group of FRM II. The in-strument is equipped with polarizedneutrons, polarization analysis and a2D position sensitive detector, whichis completely covered by the polariza-tion analyzer (see Fig. 2.21). A zero-field chamber with 3D vector polariza-tion analysis of the transmitted beamwill be added soon. TREFF will serveas a flexible and high-intensity instru-ment for investigation of neutron op-tical devices as supermirrors (polariz-ing and non-polarizing), monochroma-tor crystals, spin turners etc. Further-more, it will host the research activitiesof FZ Jülich on magnetic thin films un-til the startup of MARIA (Magnetism re-

Figure 2.20: TREFF@NOSPEC: From the cylindrical shielding of the first monochromator(right) the neutrons are brought to the shielding box of the second monochromator.Then follows the collimation (with the resonance spin flipper coil around the vacuumtube), the sample table and the detector arm (middle left).


flectometer with variable Incident An-gle), which is currently being designed.TREFF is not a regular scheduled userinstrument of FRM II.

TREFF is located at the beamportNOSPEC (Neutron Optics’ SPECtro-meter) of the Neutron Optics Group us-ing the lower part of the neutron guideNL5-S, which it shares with RESEDA(REsonance Spin Echo for Diverse Ap-plication). TREFF uses main compo-nents of the former HADAS reflec-tometer [http://www.fz-juelich.de/iff/wns_hadas] that used to beinstalled at the reactor FRJ-2 in Jülich.The neutrons are extracted from NL5-S with a double monochromator at aBragg angle of 2θ=90.

In a first step, TREFF will oper-ate at λ=4.75Å, delivered by PG (002)monochromator crystals. A verticallyfocussing monochromator and a col-limation with two pairs of slits in adistance of 170 cm deliver a beamto the sample position, which is 4cm high, has a vertical divergence of1.9 and can be collimated horizon-tally between 0.1 and 16 mrad. To re-duce the background, a vacuum tubeis inserted between the collimationslits. A transmission polarizer can beplaced automatically between the sec-ond monochromator and the first slitand offers polarized neutrons with a

maximum divergence of 4 mrad. A res-onance spin flipper is used to controlthe direction of the incident neutrons’spins. In a second step, we will installNb(002) monochromator crystals to beused alternatively to P.G.(002) (PyrolyticGraphite). This will allow to use λ=2.5Å for the investigation of single crystalquality, mosaic spread and orientation.

The sample table can take and adjustheavy loads of up to 300 kg in all 6 di-mensions (3 translation and 3 rotationdegrees of freedom). An electromagnetwith µ0H=450 mT over 10 cm gap be-tween the poles or 1.9 T over 2 cm gapis available to expose the samples to amagnetic field.

The detector arm carries a lead andB4C shielding tunnel to protect the de-tector from gamma and neutron back-ground, that is not coming from thesample. It contains an evacuated beamtube with spin flipper, a beamstop forthe part of the primary beam, whichis not reflected by the sample, a po-larization analyzer (see Fig. 2.21) anda 2D detector. The inclination of thepolarization analyzer can be adjustedautomatically either to be polarizingor transparent (for maximum transmis-sion without polarization analysis). Itis designed to offer good polarizationefficiency without compromising angu-lar resolution and sensitivity over the

whole range of scattering angles, whichthe detector covers. The detector has acircular sensitive area with 80 mm di-ameter and a spatial resolution of 1.5mm. It is mounted at a distance of 1.6m from the sample, so that it covers arange of scattering angles of 2.2. Scat-tering angles of more than 60 can bereached by moving the whole arm on aircushions.

The first parts arrived in Garching inMarch 2006 after having been assem-bled and tested in Jülich. During thepast months the instrument has beenassembled in the Neutron Guide Hallof FRM II. Today, we are ready to startthe commissioning of the instrument assoon as the modification of the neutronguide NL5-S is accomplished and neu-trons are available.

Figure 2.21: The polarisation analyzer ofTREFF.

http://www.fz-juelich.de/iff/wns_hadas

http://www.fz-juelich.de/iff/wns_hadas

3 Structure research 31

3 Structure research

3.1 HEiDi – Single crystal diffractometer with hot neutrons

M. Meven 1, V. Hutanu 1,2, G. Heger 2

1ZWE FRM II, TU München2Institut für Kristallographie, RWTH Aachen

General

HEiDi is one of the two single crystaldiffractometers at the neutron sourceHeinz Maier-Leibnitz (FRM II). It isplaced at beam line SR9B in the exper-imental hall of the reactor building (fig.3.1). It was developed in collaborationbetween the RWTH Aachen (Institut fürKristallographie) and the TU München(ZWE FRM II) to cover a broad range ofscientific cases in the area of structuralresearch on single crystals in the follow-ing fields of interest:

• Crystal structure analysis (har-monic and anharmonic meansquare displacements, hydrogenbonds, molecular disorder).

• Investigation of magnetic or-dering (magnetic structure, spindensity).

• Structural and magnetic phasetransitions.

Beam line SR9 uses the hot source of theneutron reactor. This leads to a remark-able increase of the neutron flux below1 Å with a gain factor of 7.7 at 0.44 Å(fig.3.2).

Figure 3.1: Overview of HEiDi

The hot source itself consists ofa graphite cylinder with temperaturearound 2300 K at the maximum reac-tor power of 20 MW and shifts the ther-mal neutron spectrum to shorter wave-lengths. Diffraction experiments withthe focus on structural and magneticdetails profit at HEiDi from the accessto a very large reciprocal space (Q =|~Q| = sinΘmax /λ) with Q > 1.5 at 0.55 Å.Other advantages are the

• reduction of extinction effects oflarge and very perfect single crys-tals and

• reduction of absorption effects incompounds with highly absorb-ing elements like samarium orgadolinium.

Although we had to focus our effortson the development of the new BMBFsupported polarized diffractometer atSR9 and in spite of the vacant positionof the technician for HEiDi since Au-gust 2006, we managed to guaranteethe high availability of our instrumentfor experiments of internal and interna-tional users through the whole year. De-tails of the instrument and its applica-tions were presented on the DGK meet-ing in Freiburg and on the SNI 2006 inHamburg.

0

200

400

600

800

1000

0 2 4 6 8 10 12

inte

nsit

y [

cp

s]

Time [hours]

gain factor = 7.7 at 0.44 Angstrom

reflection intensity, I(max)=913cpscalc. intensity w/o hot source

Figure 3.2: Gain factor of hot source at SR9

Sample environment

Since January 2006 a small closed cyclecryostat (Sumitomo SDK101) is avail-able at HEiDi. The minimum tempera-ture of 2.2 K makes the instrument veryattractive for detailed investigations onmagnetism at low temperatures.

In the last cycle of 2006 the cryostatwas successfully tested with substan-tially more flexible connectors betweenthe stiff He hoses and the rotating cryo-stat in the Eulerian cradle. This changereduces the angular limits of HEiDi atlow temperatures. Therefore, the num-ber of observable reflections could beincreased significantly. In figure 3.3 theold and the new design of the cryostat isshown.

An air cooled furnace (developed atFZ Jülich) was successfully tested with a

Figure 3.3: Cryostat in cradle: without (top)and with (bottom) rotating connectors

32 3 Structure research

maximum temperature of about 1100 K(fig. 3.4). Currently optimizations ofthis furnace design as well as alterna-tive techniques (mirror furnace for Eu-lerian cradles) are discussed to improvethe maximum temperature significantlyaccording to the users needs. Experi-ments with slightly increased temper-atures up to about 420 K can be per-formed with a small furnace devel-oped at the Institut für Kristallographie,RWTH Aachen (fig. 3.4).

Scientific output

In the last three proposal rounds HEiDiwas always substantially overbooked.On average only less than half the timerequested altogether in the proposalscould be assigned. In 2006 15 propos-als were accomplished at HEiDi with al-together 180 days. Additionally, sometime was used to check crystal orien-tations and homogeneities of samplesthat were used for experiments on otherinstruments at FRM II, especially thetriple axes spectrometers. Typical sci-entific cases of the proposals were

• magnetic superstructures at lowtemperatures,

Figure 3.4: Large (top) and small (bottom)furnace in Eulerian cradle

• order/disorder phase transitionsat low and high temperatures,

• ionic conductors and• local disorder or vacancies, esp.

of H bonds.A typical result from a temperaturedependent measurement of a mag-netic superstructure is shown in figure3.5, where the evolution of the mag-netic (110) reflection in the antiferro-magnetic state of the artificial olivineCo2SiO4 – starting at Tc = 49.5 K – canbe seen down to 2.2 K.

This measurement is part of a Ph.Dthesis (A. Sazonov from the Institutfür Kristallographie, RWTH Aachen)where the influence of the symmet-rically nonequivalent Co2+ sites onthe magnetic superexchange cross sec-tions is of large interest. In Decem-ber 2006 a data set of this compoundwas measured at HEiDi at 2.2 K andup to Q=sinΘ/λ=1.1 to determine pre-cisely the nuclear structure separatelyfrom the magnetic contributions in thereflection intensities which vanish atQ=0.7.

Polarized diffractometer atSR9

The development of the polarizeddiffractometer at SR9 is supposed toextend the capabilities of single crys-tal investigations with hot neutrons atFRM II to zero field 3d polarizationanalysis and to spin density analysisusing the spin flip method. To polarizethe hot neutrons 3He spin filter cells areused which get their helium from theHELIOS facility at FRM II.

In 2006 we focused our efforts onthe development and testing of com-ponents for polarization analysis, e.g.3He spin filter cells with large relaxationtimes, magnetic cavities for stabilisingthe cells and shielding them from ex-ternal fields and zero field polarizationanalysis devices (PAD). One of the mostimportant experiments in this frame-work was the test of the muPad, a de-vice for zero field polarization analyses

Figure 3.5: Temperature dependence of themagnetic (110) reflection of Co2SiO4

developed at PSI and E21/TUM usingthe complete HEiDi diffractometer unitas polarization analyser and detector(fig. 3.6). This was also the first time,3He spin filter cells were used as polar-izers and analysers as well. More de-tails about the polarized diffractometerat SR9 can be found in the section 3.2 inthis annual report.

Outlook

The successful extension of HEiDi withsample environments like cryostat andfurnaces was an important step forwardto fulfil the needs of our scientific users.Developments to solve minor problemslike the limited Tmax =850 °C of our largefurnace are in progress and should besolved in 2007. Altogether, in 2006HEiDi has proven to be an extremely ef-ficient tool for multiple scientific casesin the area of detailed structural analy-sis.

The development of the new polar-ized neutron diffractometer took a bigstep forward in 2006. A large numberof new components were successfullytested (3He spin filter cells, magneticcavities, etc.), others are currently un-der construction. For 2007 the assem-bly of all these components and the suc-cessful commissioning of this new de-vice for zero field 3d polarization analy-sis is a major goal.


Figure 3.6: Test of components for polarization experiments

3.2 Progresses in the development of the polarized neutron diffractometerPOLl-HEiDi

V. Hutanu 1,2, M. Meven 2, G. Heger 1

1Institut für Kristallographie, RWTH Aachen2ZWE FRM II, TU München

In fall 2004, the Institute of Crystal-lography at the RWTH Aachen starteda BMBF supported project to extendthe existing new single crystal diffrac-tometer HEiDi at the new researchneutron source Heinz Maier-Leibnitz(FRM II) with a polarized neutron op-tion to allow detailed investigations onthe magnetic order of single crystals.The evaluation of the problems con-nected to a direct implementation ofthe polarized neutron option on theexisting instrument lead to the deci-sion to build a new instrument espe-cially designed for this purpose, calledPOLI-HEiDi (Polarisation Investigator –HEiDi). Both options, zero-field spher-ical polarization analysis in complexmagnetic structures and spin densitydistribution analysis using a high mag-netic field for flipping ratio measure-ments should be available at the new

instrument. Several components forthis new polarized diffractometer weredeveloped, produced and successfullytested in 2006.

3He spin filter cells

Taking into account the existing polar-ized 3He production facility “Helios” atFRM II and the advantages of 3He spinfilters for hot neutrons [1, 2], it wasplanned to use NSF cells as polarizersand analyzers as well. Two spin filtercells specially optimised by maximis-ing the quality factor for neutrons withwavelengths of 0.68 Å and 0.87 Å wererealised. The dimensions of the cells are60 mm in diameter and 130 mm innerlength. The maximum thickness of thewindows is 4 mm. The working pres-sure can be chosen up to 3 bar, while the

burst pressure is not less than 5 bar. Inthe Fig. 3.7 a picture of the cells Heidi1and Heidi2 is shown. Both cells aremade of HOQ 310 quartz glass and werecoated inside with Cs.The final preparation and Cs coat-ing was done by S. Masalovich andO. Lykhvar from the neutron opticsgroup at FRM II. After the preparationthe relaxation time constants T1 mea-sured in the cells were 137(3) h for cellHeidi2 and 114(3) h for cell Heidi1. Mi-nor variations of T1 in the cells are prob-ably due to exposures to slightly dif-ferent magnetic environments as ob-served during the tests in 2006. Never-theless repeated degaussing of the cellsand “refreshing” of the Cs coating in-side the cells even improved the relax-ation time constant of the cell Heidi1


Figure 3.7: a) Spin filter cell Heidi2: Smallmetallic Cs droplets in the lower part ofthe cell and in the appendix are visible; b)Cell Heidi1: The dark spots inside the cellare traces of Cs sub-oxide which is liquidat room temperature.

up to about 165 h. The transmission ofour empty quartz glass cells at a neu-tron wavelength of 0.55 Å was found tobe between 0.85 and 0.88 depending onthe window thicknesses.

Magnetostatic cavities

Two compact magnetic cavities basedon permanent magnets and mu-metalsheets were constructed in 2006 andare dedicated for 130 mm cells use.The design of our cavities is similarto that of the “magic box” cavities re-cently developed at the ILL [3] but morecompact. The outer dimensions are(L×B×H) 380× 250× 100 mm. Calcula-tions by means of a finite element mag-netic simulation software promised afield homogeneity of (dB/dr )/B < 2 ×10−5 cm−1 in the central region. In spiteof the smaller experimental value ofabout 2×10−4 cm−1 obtained from our3He relaxation measurements, this re-sult guarantees a T1mag n. value of about400 hours, representing the relaxationtime constant due to magnetic field gra-dients in the cavity. The deviation be-tween the simulation and the experi-mental result can be explained by im-perfections of the mu-metal parts andby inhomogeneities of the used perma-nent magnets. In Fig. 3.8 “magic box”one MB1 is presented.

To assure the parallelism between thepolar pieces and to avoid deformationsof the sensible mu-metal sheets the box

is fixed in an aluminium frame, whichserves also for easy installing in the Eu-ler cradles in the front of detector onthe instrument HEiDi during our ex-periment. In Fig. 3.9 “magic box” two(MB2) is presented.

The frame is fixed on an aluminumfront flange which can be rotated 180by pneumatic drives and a tooth belt.The time for a 180 rotation (flip) is lessthan 0.5 seconds. Such a device withreliable fixed 3He spin filter cell insidepermits the transversal polarization ofthe transmitted neutron beam in anydirection in the plane perpendicular tothe propagation direction of the beam.We call it “HERO-Pol” [4] as abbrevia-tion for (Helium Rotating Polarizer).

Non-magnetic support unit

The concept and design of the new non-magnetic diffractometer basic unit wascompleted. The support unit is con-

Figure 3.8: Compact magnetostatic cavityMB1 during the test measurements fixedin the Euler cradles of instrument HEiDi

Figure 3.9: Magnetostatic cavity MB2 fixedon the rotating flange permitting 180“flipping”

ceived as a single massive part for tworeasons:

One is to reduce the total dimensionsand to assure shortest drive-times forthe detector, the other one is the capa-bility to support bulky and heavy com-ponents like cryomagnet or MuPAD.

It consists of a massive Al-alloy mademounting plate with two concentricallyfixed rotating tables, one for the an-alyzer/detector and one for the sam-ple. The first one moves a detectorplate which is designed to bear two ex-changeable setups, one for an analyzer-detector unit for 3D polarimetry andone for a lifting-counter for flipping ra-tio measurements with a magnet. Be-tween the two rotating tables a supportplate for the sample environment de-vices as well as for the polarizer unit isforeseen. A 2D tilting table is positionedon top of the sample rotation table. InFig. 3.10 a draft of the support unit withlifting counter is presented.

Figure 3.10: Non magnetic support unit andit parts: 1 – Air cushions (load capacity≥ 1.5 t); 2 – Mounting plate; 3 – Turn ta-ble/goniometer for detector plate (min.load — 700 kg); 4 – Detector plate (load150 kg); 5 – Support plate (load 100 kg); 6 –Sample turn table; 7 – 2-direction tiltingtable and manual flat turn table with ver-tical axis min. load per axis: 600 kg up to±5 tilt max. tilt ≥ ±10; 8 – Mountingunit for the detector mechanics; 9 – Lift-ing counter system +5/−25 vertical tiltand about 0.60 m radius to sample posi-tion.


First test experiments

Two experiments with neutrons forcomponents to be tested for the newpolarized instrument POLI-HEiDi weredone in 2006 at the diffractometerHEiDi. The characterization of the 3Hespin filter cells and the test of the mag-netostatic cavity MB1 at different wave-lengths were performed in June 2006(Exp. Nr. 523 at FRM II). Fig. 3.11 showsthe experimentally measured time de-pendence of the transmission in the3He filter cell Heidi2 situated in MB1.

From the data of this experiment allimportant parameters of the filter celllike 3He polarization rate, polarizing ef-ficiency, relaxation time constant, cellopacity, etc. could be extracted. A re-laxation time constant of about 90 h wasmeasured for this setup.

Another important experiment wasperformed in August to September 2006in order to test the feasibility of the Mu-PAD in combination with a 3He spin fil-ter polarizer and analyzer for 3D spher-ical polarimetry with the hot neutrons.During this experiment all three com-ponents of the polarization vector forthe incident and scattered beam weremeasured. Polarization matrices fortwo Bragg reflections of an oriented Sisample were determined.

Fig. 3.12 presents a picture of the ex-perimental setup.

The unpolarized neutron beam witha wavelength of 1.165 Å from the fo-cused Ge-(311) monochromator camefrom the right hand side in fig. 3.12. Therotating magnetic box HERO-Pol withfilter cell Heidi1 was used as polarizer.The MuPAD was fixed on a support ta-ble specially designed for this experi-ment and was used as zero field sam-ple environment for a cubic silicon sin-gle crystal (size about 120mm3). Theexisting diffractometer HEiDi was con-nected to the outgoing arm of the Mu-PAD and worked in combination withthe MB1 and second spin filter cell fixedin the Eulerian cradle as an analyzer-detector unit.

In both cells relaxation time con-stants T1 about 100 h were found. Thisassures in combination with the typi-cal 3He start-polarization of the “freshfilled” cells of 70% a mean neutron

Figure 3.11: Time dependence of the transmission of cell Heidi2 filled to 1.7 bar, measuredin the direct beam

Figure 3.12: Instrumental setup of the MuPAD test experiment. 1 – aperture of the unpo-

larized monochromatic beam, 2 – HERO-Pol (rotated polarizer with 3He spin filter cellinside), 3 – fixed incoming arm of the MuPAD with tree coils, 4 – Si sample in the zero-field chamber of MuPAD, 5 – support table, 6-movable outgoing arm of the MuPAD withthree coils, 7 – analyzer (with 3He filter cell inside), 8 – diffractometer basis HEiDi, 9 –detector.

polarization efficiency of about 0.8 dur-ing the one day before the new fillingis performed. Both analyzer and polar-izer cells were repeatedly refilled dur-ing the eight days of experiment. Highreproducibility of the cells parameters

were experimentally proved. Moreoverthe function of the rotated polarizer as amechanical spin flipper was tested. Theflipping efficiency of this novel flipperin combination with the coupling coil of


MuPAD was found to be as good as thatof the precession coil of the MuPAD.

Further detailed investigations areplaned for 2007 (Exp. Nr. 808, FRM II).

Acknowledgements

We would like to thank S. Masalovichand A. Lykhvar from the FRM II neu-tron optics group for the cooperationin the cell preparation and the engage-ment during the test experiments. Wealso would like to thank G. Langen-

stück and F. Tralmer from the construc-tion department and S. Egerland for thework on the MuPAD experiment. Spe-cial thanks to J. Krüger for substantialsupport in the all automation and soft-ware questions.

The work is supported by the GermanFederal Ministry for Education and Sci-ence (BMBF project 03HE6AA3).[1] Lelievre-Berna, E., Tasset, F. Physica

B, 267, (1999), 21.

[2] Cussen, L. D., Goossens, D. J., Hicks,

T. J. Nucl. Inst. Meth. Phys. Res. A,440, (2000), 409.

[3] Petoukhov, A. K., Guillard, V., An-dersen, K. H., Bourgeat-Lami, E.,Chung, R., Humblot, H., Jullien, D.,Lelievre-Berna, E., Soldner, T., Tas-set, F., Thomas, M. Nucl. Inst. Meth.Phys. Res. A, 560, (2006), 480.

[4] Hutanu, V., Meven, M., Heger, G.PNCMI 2006, Berlin, to be publishedin Physica B, (2007).

3.3 RESI – The single crystal diffractometer

B. Pedersen 1, G. Seidl 1, W. Scherer 2, F. Frey 3

1ZWE FRM II, TU München2Inst. f. Physik, Universität Augsburg3Sektion Kristallographie, GeoDepartment, LMU München

Status

During the last year RESI was operatingwithout major problems. We installedand commissioned the heavy goniome-ter with Eulerian craddle option suc-cessfully. This allows the use of differ-ent sample environments. In the Eu-lerian craddle geometry the RDK-101coldhead can be used to reach temper-atures down to 2K. Used with the tiltinghead, the FRM II standard furnace andthe sample tube crystat have been usedsuccessfully. This crystat also allowedthe use of a piston-cylinder type pres-sure cell.

Results

In the last year, commissioning of thesingle counter option has been started.To gain optimum performance we per-formed extensive tests on our countertube together with the detector group.The tests showed a high efficiency of thechosen 17 cm end-window 3He countertube. The linearity has been tested anda low dead time can be achieved up tocount rates of about 30kHz.

With the analyser option installed,first scans on some selected test sam-ples could be performed. The resultsshow a reflectivity of 30% for the Ge-(111) analyzer.

0

10

20

30

40

50

60

70

80

90

100

6 6.5 7 7.5 8 8.5 9 9.5 10

’primbeamana.dat’ u 1:4f(x)

Figure 3.13: Analyzer rocking scan of the primary beam.

0

5000

10000

15000

20000

25000

30000

80.5 81 81.5 82 82.5 83 83.5

’333t2t.dat’ u 2:7f(x)

Figure 3.14: θ−2θ-scan of the [333] reflection of quartz with analyzer


In the last year we also performedan intensive study on quasi crys-tal structures together with T. We-ber and W. Steurer (ETH Zürich). A3*3*3 mm3 sized sample of a decago-nal Al72Co8Ni20 quasicrystal was usedand 7277 reflections could be inte-grated. The data was used for higher-dimensional structure refinement (Th.

Weber, ETH Zürich). Structure solutionby the charge-flipping method gave aclear real space picture of the atomicdistribution, in particular of the transi-tion metals Ni and Co. A second spec-imen of the sample was prepared andmeasured by x-rays. A comparison ofthe neutron and x-ray data gave for thefirst time a clear indication of transition

metal order, which was unknown so far,as Co and Ni are hard to discriminatewith x-ray-methods.

Outlook

The next big step to finish is the com-missioning of the single counter optionwith analyzer.

3.4 Structure Powder Diffractometer SPODI

M. Hoelzel 1,2, A. Senyshyn 1,2, R. Gilles 2, H. Boysen 3, H. Fuess 1

1TU Darmstadt, Material- und Geowissenschaften2TU München, ZWE FRM II3Ludwig-Maximilians-Universität, Depart. für Geo- und Umweltwissenschaften, München

Results

In the beginning of 2006 two majorimprovements lead to a significant in-crease in the efficiency of the StructurePowder Diffractometer SPODI. A sec-ondary wavelength contamination (λ =1.983 Å with relative intensity of ca. 4%), existing together with the nominalwavelength, has been eliminated by thereplacement of a faulty wafer stack crys-tal. Thus, a purely monochromatic neu-tron beam is now available for Ge(551)setting. For the standard configurationwith a monochromator take-off angleof 155° the Ge(551) yields a wavelengthλ = 1.548 Å. At the same monochro-mator take-off angle, additional wave-lengths can be selected using differ-ent (hkl) planes by a ω-rotation of themonochromator: Ge(331) results in λ =2.537 Å, Ge(771) in λ = 1.111 Å. Further, the installation of additional shieldingfor each of the 80 detectors resulted indrastic improvements in the signal tobackground ratio.

All components of the sample envi-ronment available at SPODI have beenset into operation in the frame of usermeasurements. The closed-cycle cryo-stat ("Kaltkopf-Kryostat"), the closed-cycle refrigerator ("Probenrohrkryo-stat") as well as the high temperaturevacuum furnace are now in routineoperation. In summer 2006 first ex-periments using a high-pressure cellof the Paris-Edinburgh VX3 type havebeen successfully carried out, allowing

pressures up to 10 GPa. Additionally,a load frame for materials testing hasbeen set into operation and used for in-situ analysis of shape memory alloys.The load frame allows flexible program-ming for materials testing, e.g. in thedetermination of stress-strain curves.Its application at SPODI enables thesimultaneous study of mechanical be-haviour and structural changes, such asphase transformations, lattice expan-sion/contraction and evolution of mi-crostresses. About 2/3 of the total beamtime was given to external users. Withinthis beam time a number of structuralstudies have been carried out for a vari-ety of different materials. Some resultshave either appeared in publications orhave been submitted to pertinent jour-nals. For example, in a ZrO2 sample,partially substituted by Sc and N, thenitrogen position and amount of phasefractions have been determined as afunction of temperature together withprobable diffusion pathways [1]. Simi-larly, for zeolite-like sogdianite the dif-fusion processes of lithium [2] were de-termined. In the oxygen ionic conduc-tor mayenite (Ca12Al14O32O)the posi-tions of "free" mobile oxygen and otherions have been localized within theframework structure using differenceFourier maps, and the diffusion wasfound to proceed via a site exchangeprocess with framework oxygen [3]. Theproper tilt systems have been deter-mined for the high-temperature quan-tum paraelectrics Na0.5Nd(Pr)0.5TiO3

and (Na0.5Nd0.5)1−x Srx TiO3, whose per-ovskite structures undergo very weakdistortions from the prototype cubicstructure [4, 5]. In order to reveal thecorrelations between high ionic con-ductivity and disorder in AgCuSe andAgCuS, high-temperature structuralstudies have been done in the tem-perature range 298-723 K using neu-trons and synchrotron radiation. Theobtained data were analysed by a com-bination of whole-pattern decomposi-tion procedure and Rietveld refinement[6]. Proton localization in cement-classmaterial 4CaO·3Al2O3·3H2O has beenperformed [7] by analysis of differentialFourier maps for diffraction patternscollected at room temperature and at 5K.[1] Lerch, M., Boysen, H., Rödel, T.,

Kaiser-Bischoff, I., Hoelzel, M.,Senyshyn, A. J. Sol. State Chem.Submitted.

[2] Park, S.-H., Hoelzel, M., Senyshyn,A., Boysen, H., , Schmidbauer, E. J.Sol. State Chem. Submitted.

[3] Boysen, H., Lerch, M., Stys, A.,Senyshyn, A., Hoelzel, M. Acta Cryst.B. Submitted.

[4] Ranjan, R., Agrawal, A., Senyshyn,A., Boysen, H. J. Phys.: Condens.Matter, 18, (2006), 9679.

[5] R. Ranjan, A. A., Senyshyn, A., Boy-sen, H. J. Phys.: Condens. Matter, 18,(2006), L515.


[6] Trots, D., Senyshyn, A., Mikhailova,D., Knapp, M., Hoelzel, M., Fuess, H.J. Phys.: Condens. Matter. Accepted.

[7] Peters, L., Knorr, K., Evans, J.,Senyshyn, A., Rahmoun, N.-S., Dep-meier, W. Z. Kristal. Submitted.

Figure 3.15: Observed, calculated and differential powder diffraction patterns for mayeniteat 700 °C. Bragg reflection positions are indicated by vertical bars.

3.5 STRESS-SPEC Materials science diffractometer

M. Hofmann 1, U. Garbe 1,2, G.A. Seidl 1, J. Rebelo-Kornmeier 1,3, J. Repper 1, R. Schneider 3, H.G. Brokmeier 4

1ZWE FRM II, TU München2GKSS, Geesthacht3BENSC, Hahn-Meitner-Institut, Berlin4TU Clausthal, Clausthal-Zellerfeld

Introduction

The Materials Science diffractometerStress-Spec is located at the thermalbeam tube SR3 and is optimized forresidual stress analysis and texturemeasurements of new materials andengineering components. Its maincharacteristics and components havebeen described elsewhere [1, 2] andhere only the most relevant develop-ments of the last year will be reviewed.

New hardware

In the beginning of the user cycles2006 the faulty PG monochromatorhas been repaired, reinstalled and re-aligned. Since then it works accordingto specifications and is now used rou-tinely for texture measurements and ex-periments which require only mediumresolution but a very high neutron fluxon the sample. During commission-ing experiments it was found that ec-centricity of the original z-stage of theheavy sample table was too large withheavy loads. It was therefore replacedby a new z-stage from Huber GmbH

allowing now 320 mm travel (formerlyonly 220 mm) of the sample in ver-tical direction. In course of this re-placement the distance from the top ofthe sample table to the beam centre-line was increased to 520 mm which al-lows now to accommodate even largersamples. A 4-point bending test ma-chine and 50 kN load frame (figure3.16) were specifically developed andbuild for the diffractometer. The loadframe environment kit can be used forin-situ tension/compression strain andtexture measurements and low-cycle fa-tigue experiments.


Figure 3.16: 50 kN load frame with grips forflat samples attached.

Experiments

A total of 21 user experiments have beenconducted during the year . They cov-ered a wide area of applications in resid-ual stress analysis and texture measure-ments. Examples in the field of residualstress analysis included residual stressanalysis in welds to improve model cal-culations of postweld heat treatments(J. Francis, University of Manchester),studies of the influence of surface treat-ments on the strain development insteel rods for construction work (J.Ruiz, University of Madrid) and inves-tigations of strain distribution in ge-ometrically complex components (C.Krempaszky, TUM). Other experimentsshowed the potential of STRESS-SPECto investigate large samples such as verythick aluminum plates (M. Fox, Univer-sity of Manchester) or large steel pipes(L. Edwards, Open University). Alsohighlighted during 2006 was the abilityto produce complete strain maps dueto the high neutron flux of the diffrac-tometer. For instance figure 3.17 showsthe through thickness transverse resid-ual strain distribution mapped alongthe centreline of a weld bead on a steelplate (M. Turski, Open University). Herethe interest was to investigate weld startand stop effects, where evidence existsfor further increases in residual stressfor those regions. The results show in-deed such an increase and are thoughtto be significant as such features arecommon in welded structures. The datawill be used to evaluate and improve ex-isting FEM models.

Roughly one third of all experiments

on STRESS-SPEC have been texturemeasurements. This included the tex-ture development in Mg-Si-Al alloysunder ECAP pressing (W. Gan, TUClausthal), quantitative texture analysisof deformed natural quartz samples (H.Sitepu, Virginia Tech) and texture de-velopment and its influence on shapememory properties of Fe-Mn-alloys (R.Bolmaro, University of Rosario). Theenhanced flux due to the realigned PGmonochromator allowed to continuewith the development of methods tomeasure locally resolved textures. Fig-ure 3.18 shows as an example the localtexture of the VAMAS TWA 20 shrink-fit ring and plug sample used before asround robin standard for neutron resid-ual stress analysis [3, 4]. The global tex-ture of this sample was already deter-mined during the stress round robin,however, no information of the localtexture was available. For the local tex-

Figure 3.17: Map of transverse residual strain measured in the through thick-ness/longitudinal plane, along the centre line of the weld bead. Note: The weld metalregion has been highlighted with a red box.

ture analysis on STRESS-SPEC three po-sitions within the sample were chosento measure the (222) pole figure usinga gauge volume of 2x2x2mm3. The firstpoint is in the centre of the whole sam-ple giving texture information of thecylinder. Point 3 shows the texture ofthe ring, at the border between ring andplug (point 2) a mixture of both texturecomponents will be found. Becauseof the negligible absorption cross sec-tion of aluminum it was possible to getreasonable results for pole figures evenwithout a proper absorption correction.The pole figures in figure 3.18 observedat the three points show very similarintensity distributions. Comparing thefirst and the third pole figure one canrecognise a small shift of the maximaaround the pole figure normal. Pole fig-ure 2 shows, as expected, a mixture ofthe texture components in point 1 and3 [5].


Besides the user experimentsSTRESS-SPEC was also used commer-cially to investigate strain profiles indifferent industrial components suchas crankshafts, wheel bearings or satel-lite nozzles.

Outlook

In the new year first experiments withthe new sample environment kit areplanned to measure samples in-situwith applied loads.

For texture measurements the time toposition samples has to be decreased.Developments on a new instrumentcontrol software are ongoing and it isplanned to test it for the first time witha routine to take texture data continu-ously while rotating the sample aroundthe rotation angle phi at the earliestpossible date in 2007. The data anal-ysis software is also further developedwith the aim to accommodate textureand strain analysis in one package.[1] Hofmann, M., Seidl, G., Schneider,

R. Ann. Report FRM-II, (2003).

[2] Hofmann, M., Seidl, G., Rebelo-Kornmeier, J., Garbe, U., Schneider,R., Wimpory, R., Wasmuth, U., Nos-ter, U. Mat. Sci. Forum, 524-525,(2006), 211.

[3] Webster, G. Proceedings ICRS-6, 1,(2000), 189.

[4] Hofmann, M., Garbe, U., Seidl, G.,Schneider, R., Rebelo-Kornmeier, J.Ann. Report FRM-II, (2005).

[5] Garbe, U., Hofmann, M., Brokmeier,H.-G. Z. Krist., (2006).

Figure 3.18: (222)-pole figure of the VAMAS sample measured at three different positions as discussed in the text.

4 Inelastic scattering, high resolution 41

4 Inelastic scattering, high resolution

4.1 DNS - A versatile multi-detector time-of-flight spectrometer withpolarization analysis

Yixi Su 1,2, Werner Schweika 2, Eckhard Küssel 2, Klaus Bussmann 2, Thomas Brückel 1,2

1JCNS, outstation at FRM II, Forschungszentrum Jülich2Institut für Festkörperforschung (IFF), Forschungszentrum Jülich

Introduction

The diffuse neutron scattering (DNS)instrument is among several selectedinstruments being transferred from theJülich research reactor to FRM II. Mean-while, a substantial upgrade and mod-ernization on most of its major com-ponents is being carried out. This willtransform DNS into a worldwide com-petitive instrument in its kind. DNSis a cold neutron multi-detector time-of-flight (TOF) spectrometer with bothlongitudinal (i.e. XYZ-method) and vec-tor polarization analysis (LPA/VPA) [1,2, 3]. This allows the unambiguous sep-aration of nuclear-coherent, nuclear-spin-incoherent and magnetic scatter-ing contributions simultaneously over alarge range of scattering vector Q andenergy transfer E . With its compactsize (only 80 cm distance from sam-ple to detector), DNS is optimized as ahigh intensity instrument with mediumQ and energy resolutions. DNS istherefore ideal for the studies of elas-tic and inelastic diffuse scattering thatmay arise from short-range magneticand structural correlations and disor-dering phenomena in a wide rangeof emergent materials, such as frus-trated magnets, highly correlated elec-trons, molecular magnets and complexnano-structured compounds. Further-more, interesting applications to softmatter physics by separating coherentand spin-incoherent scattering can alsobe found at DNS.

Instrument description

DNS uses a vertically and horizon-tally adjustable double-focusing py-rolytic graphite monochromator, in-stalled in the cold neutron guide NL6a.A schematic layout of DNS at FRM II isshown in Fig. 4.1. Monochromatic neu-tron beams with a wavelength rangingfrom 2.4 to 6 Å are available at DNS.The neutron beam is polarized using am = 3 Schärpf bender-type focusing su-permirror polarizer. The neutron spinsare manipulated using a Mezei-type π-flipper, followed by a set of orthogo-nal XYZ-coils situated around the sam-

Figure 4.1: The schematic layout of DNS at FRM II

ple position for providing guide fields.The polarization analysis (PA) is per-formed by using m = 3 supermirror ana-lyzers in focusing arrangement in frontof each 3He detector. 128 new position-sensitive 3He detector tubes will be in-stalled in additional detector banks fornon-polarized experiments. This willincrease the covered solid angle up to1.9 sr. Two new high-frequency (ro-tation frequency up to 300 Hz corre-sponding to a repetition rate of 900 Hzfor 3 slits) chopper systems will be usedat DNS, as shown in Fig. 4.1. Chopper-1is used for selecting a single harmonicfrom the reflected orders of neutrons

42 4 Inelastic scattering, high resolution

from the monochromator and also forreducing the background. Chopper-2 isthe TOF chopper. Various sample en-vironments e.g. cryostat, furnace, di-lution cryostat and cryomagnet can bemounted on a heavy loading Huber go-niometer on the sample rotation table.The two perpendicular arcs of the go-niometer can be used for the orien-tation of single-crystal samples. Theinstrument data is summarized in Ta-ble 4.1.

In addition to high intensity, theunique strength of DNS lies on its ex-treme versatility. DNS can be op-erated in a number of modes for awide range of samples. There arethree PA modes at DNS: uni axi al −PA for separation of coherent and spin-incoherent scattering in non-magneticsamples; long i tudi anl − PA (XYZ-method) for separation of magneticscattering in paramagnetic and an-tiferromagnetic samples; (spher i calor) vector − PA for determination of

complex magnetic structures. All ofthese PA set-ups can be operated ei-ther in diffraction or in a TOF mea-surement. With installations of 128position-sensitive 3He tubes and dou-ble choppers, the performance of DNSas a TOF spectrometer will be drasti-cally improved, thus allowing S(Q,E)to be measured simultaneously over alarge range of Q and E on single-crystalsamples. As a high intensity cold neu-tron single-crystal TOF spectrometer,DNS is well suitable for investigationsof magnon and phonon excitations inmagnetic model systems and complexoxides.

Current status and outlook

In 2006, an increasing number of activ-ities of the DNS project took place atFRM II. The Tanzenboden was built inspring. The monochromator shieldingwas successfully installed and passedfor radiation protection during the

summer break. The improvementof the monochromator shielding wasmade. Another major development isthe permanent placement of the in-strument responsible at Garching. Theyear of 2007 will be crucial for DNS.The installation of the double-focusingmonochromator and the relocation ofthe secondary spectrometer are ex-pected in spring 2007. The delivery ofthe first neutrons and the starting of thecommissioning phase are targeted be-fore the next summer break. The instal-lations of new double choppers and po-sition sensitive 3He detector tubes areexpected to follow suit.[1] Schweika, W., Böni, P. Physica B,

297, (2001), 155–159.

[2] Schweka, W. Physica B, 335, (2003),157–163.

[3] Schweka, W., Easton, S., Neumann,K.-U. Neutron News, 16, (2005), 14–17.

Monochromator horizontal- and vertically adjustable double-focusing

PG(002), d = 3.355 Å

crystal dimensions 2.5 × 2.5 cm2, 5 × 7 crystalswavelengths 2.4 Å≤ λ ≤ 6 Å

Double-chopper sys-tem

chopper frequency < 300 Hz

repetition rate < 900 Hzchopper discs Titanium, 3 slits, φ = 420 mm

Expected neutron fluxat sample (n/cm2s)

non-polarized ∼ 108, λ = 3 Å

polarized ∼ 107, λ = 3 ÅDetector banks fornon-polarized neu-trons

position-sensitive 3He detector tubes 128 units, φ = 1.27 cm, height 101 cm)

total solid angle covered 1.9 srcovered scattering angles 0 < 2θ ≤ 135

Detector banks for po-larized neutrons

polarization analyzers 24 units, m = 3 supermirrors)

3He detector tubes 24 units, φ = 2.54 cm, height 15 cm)covered scattering angles 0 < 2θ ≤ 120

Qmax λ = 2.4 Å (Ei = 14.2 meV) 4.84 Å−1

λ = 6 Å (Ei = 2.28 meV) 1.93 Å−1

Expected energy reso-lution

λ = 2.4 Å (Ei = 14.2 meV) 1 meV

λ = 6 Å (Ei = 2.28 meV) 0.1 meVSuitable samples single crystals, powders, soft matters (e.g. polymer, liquid etc.)Sample environments closed-cycle cryostat, orange-type L-He cryostat, furnace, dilution cryostat, cryomagnet (up to 4T)

Table 4.1: Instrument data of DNS at FRM II


4.2 J-NSE – The Jülich neutron spin echo spectrometer

O. Holderer 1, M. Monkenbusch 1, R. Schätzler 1, D. Richter 1

1JCNS, outstation at FRM II, FZ-Jülich

Introduction

The Neutron Spin Echo Spectrometerwas successfully in operation at the FRJ-2 since 1996. Routine use there was lim-ited to neutron wavelengths around 8 Åand up to a magnetic field integral of0.25 Tm, giving access to a max. fouri-ertime τ of 22 ns. It has been movedfrom the Jülich Research reactor FRJ-2to the FRM II at the beginning of 2006.Since then, mechanical, electrical andelectronical installations have been car-ried out.

Activities

To adapt to the beam height at theFRM II which is about 24 cm belowthat at Jülich, the carrier structure ofthe spectrometer arms has been recon-structed. The correction elements arenow mounted on motor controlled po-sitioning devices enabling more preciseadjustment. The neutron guide systemhad been designed and installed as re-ported earlier [1]. Extra lead shield-ing around the neutron entrance intothe instrument has been installed, firstmeasurements indicate that –after posi-tioning mobile lead walls on the bound-ary to the Mephisto area– the radia-tion levels outside the Tanzboden staywithin 3µSv/h. The electrical installa-tions, cabling of all coils, sensors andmotors, took place in 2006. Electricalpower and water cooling that comply tothe 200KW maximum power consump-tion of the main coils has been passedto the instrument position. The elec-tronics of the spectrometer has beenmodernized. In particular new powersupplies (except that for the main coils)have been installed, all current sourcesas well as the SPS modules for hardwarecontrol are equipped with profibus in-terfaces. The hardware (powersupplies,motors, sensors) is now controlled withTaco servers according to the "JülichMünchner Standard".

Performance of J-NSE

The performance of the spectrometerwill largely improve at the FRM II dueto several reasons. The spectrometeris now located at the end of the neu-tron guide NL2a-o[1], which enables usto select the wavelength in the range ofabout 4.5 to 16 Å. Short wavelengths (upto about 8 Å) will be polarized in a bentFeSi polarizing neutron guide inside thecasemate. For larger wavelengths, apolarizer will be employed that is in-stalled at the end of the neutron guide.The cross section of the guide at exit is60x60 mm. New correction coils havebeen designed and manufactured, thenew backplate avoids loss of geometri-cal precision and yields better heat re-moval, use of pure Al instead of AlMg3results in smaller electrical resistanceand better heat conduction. This al-lows the use of higher currents corre-sponding up to a maximum field inte-gral of 0.5 Tm of the main precessioncoils. Tests in Jülich during the last NSEcycle there showed the path to a fur-ther necessary modification in the cor-rection coil shape that will improve thecorrection efficiency. The thus designedcoil generation with oblique cuts is un-der production at the FZJ and will re-place the current generation after thefirst operation cycle.

The neutron flux at the exit of theneutron guide NL2a-o has been mea-sured with gold foil activation analy-sis (in collaboration with K. Zeitelhack,TUM). The results are presented in Fig-ure 4.4. At λ = 7 Å, the flux is about15 times higher than it was at the neu-tron guide exit at the FRJ-2, in additionthe beam cross section at FRM II is 2.7larger than that in Jülich. The flux hasbeen compared to McStas simulations(although made with a simplified coldsource). The simulated flux agrees rea-sonably well with the measured values.At higher wavelengths, the neutron fluxis smaller than predicted, an observa-tion made also with more sophisticatedmodels of the cold source.

The larger neutron guide exit, thehigher flux and the better correction el-ements push the performance of the in-strument to fouriertimes λ = 1 ps to ∼350 ns, with a q-range of q=0.02−1.5 Å−1

Outlook

First test with neutrons have been per-formed at the end of 2006. Currentlywe work on the final function testsand parametrization of the new soft-and hardware modules which then willbe integrated into the operation soft-ware according to predefined and pre-pared routes. The first experimentswith neutrons will be used to completethe parametrization and testing workand will allow final adjustment of thecorrections. In addition circuitry formaintaining He-atmosphere along theneutron flight path has to be installedand the detector shielding must be op-timized. If the thus obtained resolutionfunctions are satisfactory first test ex-periments may be performed.[1] Breunig, C., et al. Annual Report

2005 FRM-II, 18–19.


Figure 4.2: One of the two main precessioncoils of the J-NSE spectrometer enteringthe neutron guide hall

Figure 4.3: The NSE spectrometer during in-stallation in the neutron guide hall.

4 6 8 10 12 14 16)A(

0

1

2

3

4

5

)s2mc(

/n(

xul

Fnort

ueN

x1e+7

Figure 4.4: Neutron flux at the end of the neutron guide NL2a-o. For comparison, the flux atthe FRJ-2 is shown (red square).

4.3 First quasielastic measurements at RESEDA

Wolfgang Häußler 2, Dominik Streibl 1, Reinhard Schwikowski 2, Andreas Mantwill 1, Bodo Gohla-Neudecker 2, RodrigoCaballero 2, Peter Böni 1

1Physik-Department E21, TU München2ZWE FRM II,TU München

In 2006, first quasielastic scatteringmeasurements at the Resonance SpinEcho (NRSE) Spectrometer RESEDAhave been performed. In order to pre-pare RESEDA for these experiments, in-tensity and polarization of the primarybeam were optimized. At first, highradiation background produced in thefirst spectrometer arm and at the sam-ple region (4.5) was shielded with mov-able lead walls. Now, usage of the fullbeam is possible at RESEDA, in con-trast to previous measurements usingattenuators in front of the spectrometer.Test measurements of the polarizationas a function of the wavelength of theprimary beam and its divergence wereperformed. Thereafter, the guide field

around the polarizing guide was en-forced from 80 G to 250 G. These mea-sures improved the polarization. How-ever, it was still not perfect, especially athigh beam divergence.

Nonetheless, the performance ofRESEDA could be demonstrated byfirst quasielastic test experiments. Thesample consisted of the protein Cy-tochrome C (140 mg/ml) in aqueoussolution. The velocity selector wasoperated at 22000 rpm, providing themean wavelength 5.3 Å with a wave-length spread of 12 % (FWHM). We usedthe NSE setup at small spin echo times(0.01-0.064 ns), the NRSE setup at inter-mediate (0.16-1.2 ns) and the BNRSEsetup at large spin echo times. The RF

frequency was tuned to values between35 kHz and 371 kHz leading to spin echotimes up to 4 ns. Typical intermediatescattering functions are shown in 4.6,together with exponential fit functions.All data have been normalized by theresolution function determined by us-ing a standard elastic scattering sample(graphite). As seen from 4.6, the decayrate of the relaxation function increaseswith increasing scattering vector val-ues Q, as expected for center-of-massdiffusion. The decay rates are in goodagreement with the Cytochrome parti-cle size (15x17x17 Å3). The size of theerror bars is comparable to the sym-bol size, and this is due to the still notoptimum primary beam polarization.


Figure 4.5: The instrument RESEDA with itstwo secondary spectrometer arms. Themu-metal shielding of the left arm is re-moved and gives free view on the NSE andNRSE coils.

Further optimization of the polariza-tion was abandoned, however, becausethe whole guide system was anyhowremoved after the quasielastic test ex-periments, due to non-tolerable acti-vation of the polarizing guide. A newguide system is in preparation, and anew polarizer (cavity) will be installedin early 2007 in front of RESEDA. There-fore, the measurements at RESEDAwere stopped in July 2006 and, conse-quently, the progress at RESEDA during2006 was only possible thanks to effi-cient use of available neutrons.

During the second half of 2006,several improvements have been

Figure 4.6: Intermediate scattering functions of Cytochrome C measured by means of NSE,NRSE and BNRSE. The scattering angle is varied between 2.5 and 7 degree, leading to thescattering vector values as indicated in the plot.

performed at RESEDA. Almost all axeswere supplied with encoders, in orderto prevent uncontrolled movement ofinstrumental components, for exam-ple of the sample environment and thedetectors. In addition, the NRSE coilsare going to be supplied with motorizedgoniometers and turntables, in order toprovide fast and efficient positioning ofthe coils. Finally, new NSE coils, usedfor measurements at small spin echotimes are in preparation. The previ-

ously used coils produced relatively in-homogeneous magnetic field integralsprovoking a resolution gap between theNSE and the NRSE spin echo times. Thenew coils possess better field integralhomogeneity, and, in addition, are de-signed more compact.

As soon as the new guide system,including the polarizer, is installed inearly 2007, RESEDA will be finally com-missioned and usable for routine userexperiments.

4.4 Commissioning of the backscattering spectrometer SPHERES

Joachim Wuttke 1, Peter Rottländer 1, Wilhelm Bünten 1, Peter Stronciwilk 1, Alexander Ioffe 1, Michael Prager 2, DieterRichter 2, Hans Kämmerling 3, Matthias Drochner 4, Franz-Josef Kayser 4, Harald Kleines 4, Frank Suxdorf 4

1JCNS, outstation at FRM II, FZ Jülich2Institut für Festkörperforschung, FZ Jülich3Zentralabteilung Technologie, FZ Jülich4Zentralinstitut für Elektronik, FZ Jülich

A signal-to-noise ratio of 165:1 isachieved. The energy resolution withunpolished Si[111] crystals is δEr es =0.69µeV . As a test experiment, the tun-neling spectrum of m-xylene was mea-sured. A preliminary comparison withthe performance of the HFBS instru-ment at the NIST, USA, is made.

Starting point

The commissioning of the backscat-tering spectrometer SPHERES (SPectro-meter with High Energy RESolution)was continued in 2006. The initial sta-tus was:

• A neutron flux of 1.9·109n/cm−2 sec−1

was determined at the exit of theneutron guide. This is more thanestimated originally.

• First test experiments were done.• The data contained a huge neu-

tron background.• A large γ background allowed to

use the beam only in the presenceof radioprotection personnel.

• Shielding restricted the availablesolid angle.


Technical changes anddevelopments in 2006

• A standard FRM II beamshutterreplaced the old one. This shut-ter is placed more upstream theneutron guide. This operationwas necessary to minimize spatialconstraints for another Jülich in-strument, MARIA. The new shut-ter improves also the safety in-strumentation. It closes automat-ically on power failure or on inter-ruption of the pressurised air sup-ply.

• The selector has got its dedicatednew lead shielding.

• Lead shieldings were installedaround the main γ sources in thespectrometer, the PST chopper,the beamstop and the Dopplermonochromator.

• A vacuum flight path betweenPST chopper and Dopplermonochromator was imple-mented, Fig. 4.7. This flightchamber is a prerequisite for alater argon flooding of the in-strument, as it reduces argonactivation by about an order of

Figure 4.7: Newly installed vacuum flight chamber between the PST chopper (right side)and the Doppler monochromator (left side).

magnitude. The vacuum cham-ber also increases the neutronflux at the sample by around 20%which were otherwise lost by airscattering.

• Collimations between chopperand monochromator shall pre-vent primary neutrons from by-passing the chopper.

• A similar collimation is ready tobe installed between chopper andsample.

• Diaphragms restricting theview from the sample onto themonochromator were installed.

• Most inner surfaces were cov-ered by boron absorber, either asB4C/PE composite, or in form ofa paint. The result of these mea-sures was, that the γ backgroundoutside the spectrometer went al-most everywhere below the re-quired 3 µSv/h. This allows au-tonomous working with the in-strument.

• At the expense of flux a signal-to-noise ratio of 165:1 is obtained.

Systematic measurements revealedtwo main sources of the remainingbackground:

• An instantaneous, sample-independent component due tofast neutrons originating fromthe LiMg absorber in the chopperrotor.

• A retarded background propor-tional to the scattering power ofthe sample from neutrons thatsomehow bypass the chopperwhile it is closed.

For a 20% scatterer the two sources areof similar importance. The fast neu-trons background can be avoided if theabsorber on the PST will be exchangedagainst boron. This will be a majoraction but will improve the signal-to-noise ratio by a factor of almost 2 forreasonably thin samples. A further fac-tor 2 can be gained if the chopper whichfor technical reasons actually runs atonly 1/3 of its final speed reaches itsfull operation frequency. Finally, an ar-gon atmosphere will allow about 20%more analyzed neutrons to reach thedetectors. With these factors in mindand cautious extrapolation, we can ex-pect for the final state of the instru-ment a signal-to-noise ratio above 500.Ways of further improvement are ob-vious, though technologically challeng-ing, e.g. coating the backside of thechopper with absorber. The effects ofsuch actions on the signal-to-noise ra-tio cannot be estimated reliably, how-ever.

Progress is also achieved on the sideof the data acquisition software. Inaddition to the Doppler velocity his-tograms we are ultimately interested in,we now also collect chopper phase his-tograms. The acquisition program wasdeveloped into a daemon that can becontrolled and parametrized through asimple TCP interface. Similarly, sam-ple temperature can be set and readby a simple daemon. Parameters likechopper frequency etc are now regu-larly written to a log file. Another pro-cess logs human interventions and er-ror states. The entire software is de-signed such that a graphical user inter-face can easily be added on top of it.

The actual status of the instrumentis presented by two spectra. The elas-tic spectrum, Fig. 4.8, documentsthe signal-to-noise ratio of 165:1 at anenergy resolution of δEr es = 0.69µeV


FWHM. The measuring time was ∼12h.Fig. 4.9 shows a tunneling spectrum.

M-xylene was used as test samplesince it was previously measured onthe equivalent HFBS instrument at theNIST, USA. In both experiments thesignal-to-noise ratio was rather simi-lar and below 100:1. The energy res-olution, FWHM, is better on SPHERESwhile the low intensity wings of theresolution function is less pronouncedon HFBS. For SPHERES the energy res-olution looks rather independent onenergy transfer. This is due to ourDoppler monochromator drive basedon air bearings, which allows an almostfriction free movement compared to themore vibration sensitive cam drive ofHFBS.

The improvements of the instrumentare going on.[1] O. Kirstein, R. D., M. Prager,

Desmedt, A. J.Chem.Phys., 122,(2005), 14502.

-10 -5 0 5 10

energy (μeV)

10

100

1000

counts

(s

ec-1

)

.25mm PET

cylinder

detector j=3

SNR = 165:1

13/14 oct 2006

Figure 4.8: Resolution curve, measured on a 0.25mm thick polyethyleneterephtalat sample.

-10 0 10

E (μeV)

0.001

0.01

0.1

1

I /

I ma

x

m-xylene

JCNS

NIST

Figure 4.9: Tunneling spectrum of m-xylene measured on SPHERES (blue) and HFBS (red)[1]. Normalization to same peak intensities. Sample temperature T=5 K. Momentumtransfer: Q∼1.7Å−1. Open symbols: raw data; closed symbols: after subtraction of a flatbackground.


4.5 TOFTOF back in successful operation

T. Unruh 1, J. Ringe 1, J. Neuhaus 1, W. Petry 1,2

1ZWE FRM II, TU München2Physik Department E13, TU München

After the successful startup of useroperation of the TOFTOF spectrometerin July 2005 a damage of the veryfirst part of the neutron guide, whichfeeds TOFTOF and the reflectometerREFSANS with neutrons from the coldsource of the FRM II, interrupted theoperation of both instruments for halfa year. With the reactor cycle no. 6,which started at the end of March 2006,

Figure 4.10: TOFTOF spectrum of the compound Fe6O2(OH)2(O2CCMe3)12·(THF)2 at 0.5 K.The measurement was performed to investigate the spin dynamics of an Fe6–Cluster. Forcomparison a vanadium and the empty can measurement are displayed too. The sampletemperatures and the recording times are given in the legend for all spectra. The data isshown here by courtesy of Tatiana Guidi and Grigore Timco.

Figure 4.11: Panorama of the detector bank of the TOFTOF spectrometer. The detectors areoriented tangentially to the intersection line of the Debye–Scherrer cones and a virtualsphere with a radius of 4 m around the sample position. Beside the detectors only cad-mium sheets are visible shielding the detectors against neutrons not coming from thedirection of the sample position.

TOFTOF came back to full operation.A lot of exciting experiments were per-formed already in this cycle, as e.g. thefirst TOF experiment on highly super-cooled Ni–melts held and heated byelectromagnetical levitation (cf. section9.5).

Although TOFTOF restarted only inMarch 28 different experiments couldbe performed in 2006 in addition to

the work on instrument maintenanceand optimization. A major task dur-ing the experiments was to optimizethe sample environment for the needsof the experimentalists. Thus a “bio–furnace” for the medium temperaturerange (-30°C–180°C) was developed andsuccessfully tested. With this furnace itis now possible to control the temper-ature of a sample with an accuracy ofabout 0.1°Cin the specified temperaturerange. The furnace can be pre–cooledor heated and therefore the sample canbe inserted at a well defined tempera-ture. The space around the sample canbe evacuated, flushed with gas and con-trolled for humidity, which has, how-ever, not yet been tested.

First experiments on magnetic sys-tems were performed. Using the 3He–insert inside the standard closed cyclecryostat temperatures below 0.5 K couldbe achieved. A measurement at 0.5 Kon the spin dynamics of an Fe6–Clusteris displayed in figure 4.10. Accordingto the incident neutron wavelength of1.8 Å energy transfers up to -23 meVcan be detected on the neutron energyloss side of the spectrum. It should bepointed out here that within only 4 h ofmeasuring time very good statistics canbe achieved due to the particular highneutron flux of the TOFTOF spectro-meter in the wavelength range between1.6 Å and 3Å.

At the beginning of 2007 it was de-cided to provide TOFTOF with 400 addi-tional detectors, which will increase theperformance of the spectrometer by afactor of about 1.7. The detectors willbe purchased in 2007. As it can be seenin figure 4.11 there is space left for theadditional detectors on the racks in the3rd and 4th raw, respectively. Thus asignificant upgrade of the instrument isalready under way.

5 Three axes spectroscopy 49

5 Three axes spectroscopy

5.1 PANDA – cold neutron three-axes spectrometer

P. Link 1, A. Schneidewind 1,2, D. Etzdorf 1, M. Frontzek 2, O. Stockert 3, M. Loewenhaupt 2

1ZWE FRM II, TU München2Inst. f. Festkörperphysik, TU Dresden3Max-Planck-Institut für Chemische Physik fester Stoffe Dresden

Introduction

In the year 2006 PANDA entered intoroutine operation. Already a broaduser community has benefited from thespecial characteristics of the PANDAspectrometer. Apart from numer-ous user experiments performed, quitesome time on PANDA was dedicated toimprove the instrument and the sam-ple environment. In the following abrief summary of the main characteris-tics and improvements on PANDA willbe given. This will be complementedby examples of scientific research per-formed on PANDA in 2006.

Instrument

Current status and characteristics

Starting with the first neutrons onPANDA in 2005 and the experience withdifferent working options improve-ments were initiated. Sets of param-eters for different working conditionsand instrument setups were collected tohelp users to decide on the best suitableoperation mode of PANDA. In February2006 the monochromatic flux at thesample position was determined bygold-foil activation analysis and using acalibrated monitor for ki = 1.5 Å−1 andki = 2.662 Å−1. The measurements werecarried out with the monochromatorbeing vertically focused and horizon-tally flat, the virtual source set to 40 mmwidth using no collimation and the ap-propriate higher order filters in the in-coming beam (170 mm Be Filter, LN2

cooled and 60 mm PG Filter; 3.5 mo-saicity, respectively). The obtained val-ues of 1.9×107 n/cm2/s(ki=1.5 Å−1) and

5.5 × 107 n/cm2/s(ki = 2.662Å−1) agreewell with our expectation for this opera-tion mode. The measured experimentalbackground was decreased to less than1 count/min for small ki ≤ 1.5 Å−1 andstandard TAS scattering angles. To fur-ther reduce the background especiallyat small scattering angles a vacuumbox with a diameter of 60 cm aroundthe sample was built and successfullytested during an experiment on a multi-layer system. This vacuum box substan-tially reduces air scattering which oth-erwise would give rise to an increasedbackground at small q.

To increase the incoming intensitythe PG monochromator focus driveswere reconstructed and prepared forautomatic tracking with ki changes. Auser test experiment was performedwith a prototype of an elliptic neutronguide to focus the neutron beam ona tiny sample size suitable e.g. forhigh pressure cells. The peak inten-sity in the obtained central spot of 2×2mm2 showed an increase by a factorof ten in comparison to the absence ofthe elliptical guide. Measuring nuclearand magnetic Bragg peaks of a NiS2 [1]single crystal with dimensions 1 × 1 ×0.1mm3 the signal gain was about a fac-tor of two.

Two magnets are now available forexperiments on PANDA, the 7.5 T andthe 15 T magnet. The special sampleenvironment of PANDA, the 15 T cryo-magnet including the dilution low tem-perature insert, was commissioned andput into user operation mode. It wassuccessfully used for about 25% of thetotal beam time (in close arrangementwith the positron source group). Veryrecently, the 7.5 T cryogen free verti-

cal magnet of the FRM II sample en-vironment was tested for the first timeon PANDA and necessary technical im-provements for inelastic scattering ex-periments were initiated.

The alignment of the instrument andof the sample in its cryogenic environ-ment was strongly simplified by adapt-ing the DELCam 2D neutron camera de-veloped by the FRM II detector group.Such a camera is now permanentlyavailable on PANDA.

Future upgrades

Apart from routine operation for userexperiments the first half of 2007 willsee the commissioning and test of thepolarisation analysis option using bothHeusler monochromator and analyser.In addition a BeO filter has been or-dered. Such a BeO filter option was re-quested by many users and will be usedat kf ≤ 1.3 Å−1 when high energy resolu-tion is required. As a new sample envi-ronment component a VariOx cryostatwill be delivered end of March to allowexperiments with 1.5 K base tempera-ture and to serve as a "host cryostat" forthe Kelvinox dilution insert.

Science

The PANDA spectrometer has seen in2006 an already broad user communitywhich is still growing. So far, 17 externaluser experiments have been performed.Not only German user groups but also alarge number of European experimen-talists and even guests from the UnitedStates were hosted. The scientific sub-jects covered by the users range fromferro-electric substances, low dimen-

50 5 Three axes spectroscopy

sional quantum magnets, frustratedmagnets to strongly-correlated electronsystems like High-Tc superconductorsor heavy-fermion systems. For most ofthe experiments the special sample en-vironment available on PANDA, i.e., the15T magnet together with the dilutioninsert, was indispensable for the suc-cess of the experiments. The two exam-ples given below are in the fields of frus-trated magnetism and spin dynamics inheavy-fermion systems where the mea-surements on PANDA revealed new in-sight into the complex behavior of mag-netism in condensed matter physics.

Temperature and field dependentmagnetic structures in Tb2PdSi3

The series of R2PdSi3 (R = rare earth)compounds has been in the focus ofinterest for over 15 years [2]. The com-pounds crystallize in an AlB2-derivedhexagonal structure (P6/mmm) with alatent geometric frustration. Tb2PdSi3

orders antiferromagnetically belowTN = 23.6 K and exhibits in addition aspinglass like (SGL) phase transitionaround T2 ≈ 8 K observed predomi-nantly with ac- and dc-susceptibilitymeasurements [3]. Neutron scatteringwas performed at E2, HMI Berlin, andPANDA to study the magnetic struc-tures of Tb2PdSi3 and to shed light onthe origin of the SGL phase transition.In Tb2PdSi3 the antiferromagnetic LROis squared-up in the whole temperaturerange below TN. All magnetic satellitesof the LRO have a similar temperaturedependence. No features of the LROcan be correlated with the tempera-ture T2 of the additional phase transi-tion. However, antiferromagnetic SROwith a characteristic temperature ofT2 ≈ 8 K was found. Thus, the disor-der to SRO transition correlates to theobserved SGL phase transition. TheSRO is likely to originate from geomet-ric frustration of magnetic momentswhich are not able to fulfil the antifer-romagnetic coupling to all neighbourssimultaneously. This frustration effectseems to affect the LRO arrangementonly slightly. Applying magnetic fields,the intensities of the LRO reflectionsdecrease dramatically and change fully

0.6 0.8 1.0 1.2 1.4 1.60

50

150 Tb2PdSi3(0 0 L) scan10.000 Mon

T = 2 K; 0H = 0 T T = 2 K; 0H = 2.4 T T = 2 K; 0H = 4.5 T

Inte

nsity

[1kc

ts]

L [r. l. u.]

Figure 5.1: (00L) section of Tb2PdSi3 at T = 2 K at different magnetic fields. In zero field(black) the magnetic reflections are on n/16 position and relative intensity according to asquared-up arrangement. In intermediate magnetic fields (red) coexistence of two mag-netic propagations is observed. Above 3 T (blue) only the antiferromagnetic propagationwith the L component of 1/4 remains.

into a different LRO structure aboveµ0H = 3 T (see Fig.5.1). The SRO struc-ture vanishes at µ0H = 1.5 T, but onequal positions reflection due to an in-commensurate arrangement are found.The preliminary results of the investi-gations in Tb2PdSi3 encourage to con-tinue the investigation in the (H ,T )-phase space to understand the richmagnetic phase diagram.

Magnetism and superconductivityin the heavy-fermion systemCeCu2Si2 close to quantumcriticality

The heavy fermion compoundCeCu2Si2 attracts special attentiondue to the interplay between antifer-romagnetic order and superconduc-tivity. Both phenomena exclude eachother on a microscopic scale. Thesystem is located close to a quan-tum phase transition at the disappear-ance of the antiferromagnetic order.By small differences in crystal growthCeCu2Si2 samples can be producedwhich show either S(uperconductivity),

A(ntiferromagnetism), or both (A/Stype). The antiferromagnetic order waspreviously determined on an A-typesingle crystal to be an incommensuratespin-density wave below TN = 800 mK[4]. The aim of a first experiment onPANDA was to study the supercon-ducting phase of an S-type crystal.Since superconductivity appears belowTc ≈ 600 mK, a dilution refrigerator hadto be used to reach temperatures wellbelow Tc . Surprisingly, in elastic scansshort-range magnetic correlations havebeen found at the same q positionswhere in the antiferromagnetic phaseof A-type crystals superstructure peakshave been observed. Fig. 5.2 showselastic scans taken well inside the su-perconducting phase of CeCu2Si2 atT = 50 mK and in the paramagneticphase at T = 800 mK. In the paramag-netic regime these correlations havecompletely disappeared. The linewidthof these magnetic correlations point toa correlation length of 50 − 60 Å. How-ever, it is so far an open question howthese correlations are related to the su-perconductivity in CeCu2Si2. Further


measurements are planned to look forthe response and temperature depen-dence of these newly discovered corre-lations in detail.[1] Niklowitz, P. G., Mühlbauer, S., Link,

P., Böni, P. FRM II Experimental Re-port, (2006).

[2] Kotsanidis, P. A., Yakinthos, J. K.,Gamari-Seale, E. J. Magn. Magn.Mat., 87, (1990), 199.

[3] Frontzek, M., Kreyssig, A., Doerr,M., Schneidewind, A., Hoffmann, J.-U., Loewenhaupt, M. J. Phys: Cond.Matter, 18, (2006), 1.

[4] Stockert, O., Faulhaber, E., Zwick-nagl, G., Stüsser, N., Jeevan, H. S.,Deppe, M., Borth, R., Küchler,R., Loewenhaupt, M., Geibel, C.,Steglich, F. Physical Review Letters,92, (2004), 136401. 184 186 188 190 192

Sample rotation (deg.)

150

200

250

300

350

400

Neu

tron

inte

nsity

(co

unts

/mon

5

min

)

T = 50 mKT = 800 mK

CeCu2Si2Q (0.226 0.226 1.467)

Figure 5.2: Elastic scans taken well inside the super- conducting phase of CeCu2Si2 atT = 50 mK and in the paramagnetic phase at T = 800 mK

5.2 PUMA – The thermal triple axis spectrometer

K. Hradil 1,2, R.A. Mole 2, H. Schneider 1,2, J. Neuhaus 2, G. Eckold 1

1Inst. f. Physikal. Chemie, Universität Göttingen2ZWE FRM II, TU München

The thermal triple axis spectrometeris installed at the beam tube SR7. Dur-ing the last year PUMA in its basicversion with the single detector beganroutine operation. Approximately 150days have been assigned to externalusers (for results see experimental re-ports). Additionally the first time re-solved measurements in dependence ofan applied electrical field both on satel-lite and phonon scattering could be per-formed.

New hardware andcharacterisation

The new Cu(220) monochromatorbending device built in the workshop ofthe University Göttingen could be im-plemented and allowed measurements

of energy transfers up to 20 THz. Fig. 5.3shows the monochromator within theexchange unit of the four monochroma-tors (left) and on the test device (right).The high quality of the monochromatorproven by rocking scans with a FWHMof 0.4° measured on the PUMA instru-ment illuminating a horizontal andvertical area of the monochromatorof 70 by 150 mm respectively, for ki =2.662 Å. The performance of the indi-vidual Cu crystals as mounted on thebending device were analysed on theSTRESS SPEC instrument by rockingscans which revealed a mean FWHM of0.37° +/-0.03°. High energy excitationsinvestigated during a test measurementon a La1.95Sr0.05CuO3 crystal resultin comparatively short measurement

times and excellent resolution charac-teriscs.

During the start-up phases of the in-dividual reactor cycles we continuedto characterise both the PG(002) andthe Cu(220) monochromator. The en-ergy resolution characterisation as de-termined by using a vanadium standard

Figure 5.3: left: Cu(220) monochromatorwithin the change unit; right: with thebending test unit

52 5 Three axes spectroscopy

sample for different instrument config-urations are shown for PG(002) in fig.5.4 and Cu(220) in fig. 5.5 respectively.The comparison to resolution calcu-lations clearly demonstrates the out-standing quality of the monochromatordevices.

The focussing options for the Cu(220)monochromator were also investi-

Figure 5.4: energy resolution of the PG(002) monochromator for different instrument con-figurations

Figure 5.5: energy resolution of the Cu(220) monochromator for different instrument con-figurations

gated by using a CCD camera devel-oped for tomography. The neutronbeam profiles on sample position ob-tained for both a flat and focussedCu(220) monochromator are shown infig. 5.6 and fig. 5.7,respectively. Thehomogeneity of the intensity distribu-tion on the sample position for the

Figure 5.6: CCD camera picture of thebeam profile on sample position for flatmonochromator

Figure 5.7: CCD camera picture of the beamprofile on sample position for focussedmonochromator

flat monochromator as well as the fo-cussing properties prove also the ex-cellent quality of the monochromatorsetup.

Beside the 2 cryostates (3-300K, 15-800K) and a furnace (up to 1200K),we succeeded in putting the Paris Ed-inbourgh pressure cell into operationwhich allows us to investigate excita-tions with an applied pressure up to 10GPa. The cell mounted on the goniome-ter of PUMA for an external user exper-iment is shown in fig. 5.8. Although theexperiment results in data with reason-able statistics, the intensity for higherpressures was reduced by a factor 3-4due to the small opening of the anvil cellin comparison to the large dimensionsof the beam. To overcome this problemwe plan to install special focussing (m >5) in front of the sample.

Apart from the user program, sev-eral measurements using the strobo-scopic measurement technique couldbe performed on the instrument. Fig.5.9 shows the intensity distribution fordifferent time channels for a satel-lite reflection in K2SeO4 at 94.7 Kwith/without applied electrical field.


The reaction time for the lock-in tran-sition was found to be temperature de-pendent and varies between 0.5 and 2ms.

Outlook

Further developments of the instru-ment will include the full implementa-tion of the stroboscopic measurementtechnique and the implementation ofthe multianalyser/-detector.

Figure 5.8: Paris Edinbourgh pressure cell onthe gonimeter table

Figure 5.9: time resolved intensity distribution for the first order satellite reflection (2 0 4/3-δ) in dependence of an applied electrical field

54 6 Imaging at ANTARES and NECTAR

6 Imaging at ANTARES and NECTAR

6.1 Observation of the filling level of the FRM II cold source by pinholeneutron radiography

B. Schillinger 1, Elbio Calzada 1, Christian Müller 1


Abstract

The ANTARES facility for neutron imag-ing is situated at beam port SR4 facingthe cold source. By using a 2 mm pin-hole, the surface of the cold source wasimaged onto the detector. At low powerduring reactor startup when the liquidD2 was not boiling yet, the filling level ofthe cold source was observed. From thegeometrical parameters of this projec-tion, the filling level of the cold sourcewas calculated.

The cold source and theANTARES facility

During reactor shutdown, the liquid D2

filling of the cold source of the FRM IIreactor is transferred to a hydride stor-age system. The refilling of the coldsource before startup is a difficult pro-cess, where the exact amount of deu-terium filling of the cold source is noteasy to determine.

The ANTARES facility is situated onbeam port SR4b facing the cold source(fig. 6.1).

ANTARES possesses an external sec-ondary shutter outside of the drumshutter in the biological shielding. Be-hind this secondary shutter, a selectorwheel is mounted containing differentpin hole apertures for phase contrastimaging (Fig. 6.2).

The cold source is a slightly tiltedcontainer filled with liquid D2. To ho-mogenize the flux and to minimize self-absorption, it contains a hollow dis-placement body shaped like an invertedcup (Fig. 6.3, Fig. 6.4). After filling,the container and the cup contain liq-uid D2 to a certain filling level. When

the reactor reaches power, the D2 startsto boil. The displacement body fillswith gas and displaces the liquid in-side, thus raising the liquid level out-side the cup. The beam tube SR4 feed-ing the ANTARES facility faces the sur-face of the cold source. In normal op-eration, the effective pinhole camerageometry of ANTARES images an un-sharp projection of the surface of thecold source onto the sample area, lead-ing to a a homogenous flux distribution.If one of the small pinhole diaphragmsintended for phase contrast imaging isused, a sharp well-defined projectionof the surface itself is imaged onto thecamera.

Figure 6.1: Schematic view of the ANTARESfacility. The whole imaging facility is builtin a pin hole camera geometry, imagingthe surface of the cold source onto thesample area.

Figure 6.2: Selector wheel for various pin-holes.

Figure 6.3: Drawing of the displacementbody inside the cold source.

6 Imaging at ANTARES and NECTAR 55

Figure 6.4: Photo of the displacement body.

Visualization of the fillinglevel of the cold source

During a first reactor startup, reactorpower was halted at several power lev-els between 300 kW and 5 MW. Pin-holes with 1, 2, and 7 mm diameterwere tested. Best results were obtainedwith the 2 mm pin hole at 3 MW reactorpower. The very small pinhole requiredexposure times between 10 and 30 min-utes.

Since the pinhole camera shows aninverted image, the cold source isshown upside down with the liquid levelfrom top to bottom. The liquid isseen as darker (less intensity) becauseneutrons emitted from the liquid D2

are moderated to lower average wave-length. On the way to the detector,the neutrons have to pass several wallsand windows: The ZircAlloy wall of thecold source, the Aluminium beam tubenozzle and the vacuum windows of thebeam tubes. The absorption and scat-tering cross sections increase with in-creasing wavelength, causing higher at-tenuation for cold neutrons than forthermal neutrons.

Fig. 6.5 shows the upside-down im-age of the liquid level in the cold sourceat 3 MW reactor power, Fig. 6.6 showsthe same field of view at 5 MW, withthe visible liquid level gone as the coldsource boils at this power.

Figure 6.5: The cold source at 3 MW with thefilling level visible.

Figure 6.6: At 5 MW, the visible level is gone.

Measurement of the fillinglevel of the cold source

At a following reactor startup, the ex-periment was repeated for attemptedquantitative measurement. For gaug-ing the image scale, and marking thebeam center, two strips of boratedpolyethylen of known width were fas-tened at the beam exit. The horizon-tal strip was tilted to make the imageasymmetric so there would be no doubtabout the correct orientation of the im-age. Fig. 6.7 shows the image of the coldsource with the polyethylen strips in thebeam, Fig. 6.8 has the dimensions in-serted.

Figure 6.7: The cold source at 3 MW with po-lethylen strips in the beam, with the geo-metric center marked by the upper cornerof the horizontal strip.

Figure 6.8: Gauging of scale and projectionratio from measured distances.

With the strip width of 33 mm, thescale of the image was determined. Theliquid level was thus measured at 77mm distance from the center line (Fig.6.8). With the known distances betweenthe cold source and the pinhole and thepinhole and the detector, the projec-tion ratio was calculated. The 77 mmdistance at the detector correspond to28 mm at the cold source. From theCAD model of the cold source, the fillingvolume of the cold source at the cen-ter line of the beam was determined as10636 cubic centimeters or roughly 10.6liters, with one centimeter height differ-ence corresponding to 704 cubic cen-timeters. The determined filling level


was then calculated as 12626 cubic cen-timeters, or roughly 12.6 liters.

Only after these calculations, weasked the operators for the cold sourcefor their estimate of the actual filling.They assumed that the best filling theyobtained was 12.5 liters.

Conclusions

The surprising apparent accuracy ofthis measurement in spite of so manyinaccuracies involved may be pure co-incidence and luck. At the time of writ-ing, there had been no opportunity yet

to repeat the measurement on a consec-utive reactor startup. The results will beverified by a new experiment as soon aspossible.

6.2 Implementation of the new multi filter at ANTARES

Klaus Lorenz 1, Elbio Calzada 2, Martin Mühlbauer 2, Michael Schulz 2, Burkhard Schillinger 2, Karl Zeitelhack 2

1Physics Department E21, TU München2ZWE FRM II, TU München

Figure 6.9: The new multi-filter (schematic(top) and photograph (bottom).

Design efforts

A new multi-filter is ready for appli-cation at the ANTARES facility at FRMII. Between fast shutter and aperturewheel [1] a very compact filter pack-age was installed (fig. 6.9) allowing thequick and precise positioning of fourdifferent crystal filters in the neutronbeam. Due to the limited space atthis position, the maximum thicknessfor the filters was 50mm, which is thethickness of all but the beryllium filter(only two 20mm thick beryllium plateswere available, which were combinedto a 40mm filter). The effect of thedifferent filters on the neutron spec-trum was investigated with TOF mea-surements (fig. 6.10). The cause for theoffset in the raw TOF data are epither-mal and also a fraction of the thermalneutrons, that penetrate the Gd coatingof the chopper wheel. Sapphire singlecrystals are well known as good filtersfor epithermal neutrons [2, 3] and theTOF measurement confirms that the in-stalled filter efficiently blocks epither-mal neutrons without remarkable mod-ifications of the spectrum in the ther-mal an cold range. This is e. g. veryuseful for phase contrast imaging with aLiF-scintillator [1, 4], where epithermalneutrons are a major cause for noise.Another reason for background noise isX-ray and gamma radiation. Becauseof its high atomic number and trans-parency for cold and thermal neutrons,bismuth is a very good gamma filter.

Beside the single crystal gamma fil-ter a polycrystalline bismuth filter wasinstalled to investigate the utilization of

certain bragg edges in neutron imag-ing. Finally a beryllium filter was in-stalled to suppress thermal neutronsbelow 3.96Å[5]. This modification of thespectrum is on the one hand useful formeasurements, where only cold neu-trons contribute to the measured signal(e. g. phase contrast imaging with grat-ings), on the other hand it can be usedin a primitive way for energy selectiveradiography (fig. 6.11).

Applications

In fig. 6.11 neutron radiographies ofstep wedges of different materials areshown. The upper was done withouta filter, the one below with the beryl-lium filter and the lowest picture dis-plays the result of the division of theupper two. In the radiography withouta filter the attenuation of the neutronbeam by iron is only slightly higher thanof the lead. With the beryllium filterthis changes and the lead attenuates thebeam stronger than the iron. The con-trast is still not high, but if the radio-graphy without filter is divided by theradiography with beryllium filter, thecontrast becomes clearly visible. Thismethod is an easy way to increase thecontrast for certain materials in neu-tron radiographies.[1] Lehmann, E., Lorenz, K., Steichele,

E., Vontobel, P. Nucl. Instr. andMeth. in Phys. Res. A, 542, (2005),95–99.

[2] Born, R., Hohlwein, D., Schneider,J. R., Kakurai, K. Nucl. Instr. andMeth. in Phys. Res. A, 262, (1987),


359–365.

[3] Stamatelatos, I. E., Messoloras, S.Rev. Sci. Instr., 71, (2000), 70–73.

[4] Cowley, J. M. In Diffraction Physics(Elsevier, Amsterdam, Netherlands,1995), 3 edition.

[5] Webb, F. J. Nucl. Instr. and Meth., 69,(1969), 325–329.

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 150

5000

10000

15000

20000

25000

diff.

neu

tron

flux

dΦ/d

λ [a

rb. u

nits

]

λ

1 2 3 4 5 6 7 8 9 10 11 12 13 14 150

1x106

2x106

3x106

4x106

5x106

6x106

7x106

8x106

Φ

λ

λ

Figure 6.10: TOF measurements of the neutron spectrum atANTARES with the different crystal filters (top: raw data, bottom:corrected).

Figure 6.11: Method to increase the contrast for certain materials inneutron radiographies with a beryllium filter.

6.3 Neutron phase contrast tomography

Klaus Lorenz 1, Burkhard Schillinger 2


Experimental setup

In neutron phase contrast imaging thevariation δ of the real part of the refrac-tive index n = 1 − δ+ iβ from unity isused to get an additional contrast be-sides absorption contrast. The easiestway to get this phase contrast is to use

a propagation based method. Unlikeinterferometric measurement methods,no complicated experimental setup isnecessary. Basically the same setupas for conventional neutron radiogra-phy can be used with two additional re-quirements: The neutron beam musthave a high transversal spatial coher-

ence at the sample position and thedetector must have a certain distancefrom the sample [1]. The high co-herence is achieved by the introduc-tion of pinhole apertures with a diam-eter of 2mm and less in the beam ina distance of 14m to the sample po-sition (see fig. 6.12). Due to the


Figure 6.12: Implementation of the pinholeapertures at ANTARES.

drastically reduced neutron flux, expo-sure times in phase contrast measure-ments are much longer than in con-ventional neutron radiography, whatmakes this method much more sensi-tive to background noise.

Main causes for noise are epithermalneutrons and gamma radiation in thebeam, secondary radiation due to ac-tivation and inherent noise of the de-tector system. Until recently, the noiselevel in long time exposure radiogra-phies with a CCD detector system wasmuch too high for phase contrast imag-ing. Good results were only achievedwith image plate detectors [2], whichhave the big disadvantage of not allow-ing normalization and tomographies.After some modifications of the pin-hole apertures, the improvement of theshieldings in the sample and detectorarea and the installation of the newmulti filter (see corresponding report),the signal-to-noise-ratio in long timeexposure measurements with the CCDdetector improved significantly and al-lowed the step from phase contrast ra-diography to phase contrast tomogra-phy.

Applications

In figure 6.13 the results of a conven-tional tomography are compared withthose using the phase contrast effect.The effect is only used in a qualitativeway for contrast enhancement at edgesand interfaces, the phase shift is notmeasured directly (no phase retrieval).

The test sample is a step wedge outof AlMg4.5Mn with another aluminumalloy (AlSi9Cu4) cast around it (see fig.6.14). Both alloys have very similar at-

tenuation coefficients. Because of thisand the fact, that the sample has aconstant thickness in beam direction,there is no contrast between the dif-ferent steps in conventional neutronimaging. Under phase contrast condi-tions, due to the fact that Manganesecauses a negative and Copper a strongpositive phase shift, phase contrast oc-curs at the inner interfaces which areparallel to the direction of the neutronbeam.[1] Cowley, J. M. In Diffraction Physics

(Elsevier, Amsterdam, Netherlands,1995), 3 edition.

[2] Lehmann, E., Lorenz, K., Steichele,E., Vontobel, P. Nucl. Instr. andMeth. in Phys. Res. A, 542, (2005),95–99.

Figure 6.13: Comparison of conventional and phase contrast radiography (left) and tomog-raphy (reconstructed slice, right).

Figure 6.14: Schematic of the test sample.


6.4 NECTAR – Neutron Computer Radiography and Tomography facility

T. Bücherl 1, Ch. Lierse von Gostomski 1

1Institut für Radiochemie, TU München

General

The NECTAR facility is designed for ra-diography and tomography using fis-sion neutrons. Its general layout andfirst results have already been pub-lished [1, 2, 3]. In this report im-provements and measurement resultsachieved during the last months of 2006are presented.

Improvements

In 2006 a significant improvement ofthe image quality was achieved by re-placing the liquid nitrogen cooled CCD-camera (512 x 512 pixels) by a newCCD-camera with 1024 x 1024 pixels(ANDOR DV424-BV). With its thermo-electrical cooling system the detectionefficiency now remains constant dur-ing a complete measurement cycle, be-ing an essential requirement for tomo-graphic measurements which take sev-eral hours.

The control software was modified,now integrating the control of the ob-ject manipulator and the detector sys-tem within one program, i.e. improvingits usability.

Results and discussion

In a number of radiographic measure-ments possible fields of applicationwere investigated. In principle thesemeasurements can be subdivided in thecategories

• standards,• technical objects and• wooden objects.

Examples of each category are given be-low.

During all measurements a largenumber of scattered gamma rays andneutrons directly hit the CCD chipcausing randomly distributed signals(sparks, glitches) in individual pixels.Thus, for each radiograph a set of typ-ically five frames in total was measured.The measuring time per frame was 60 s.

These five frames were filtered applyinga median filter and then summed up. Bydark image subtraction and normaliza-tion with the open beam image the re-sulting radiograph of the object was cal-culated.

Standards

Step wedges are often used to inves-tigate the penetration, scattering andcontrast characteristics of materials.The results for step wedges made oflead, iron, aluminum and polyethylenehaving a maximum thickness of 50 mmwere already presented [4]. The highpenetration power of fission neutronsfor polyethylene (as an example of a

Figure 6.15: Left: Photograph of a step wedge set up of ten plates made of plexiglas of 10mm thickness. Right: Normalized radiograph.

Figure 6.16: Left: Photograph of a standard set up of a plexiglas block containing (hollow)cylinders partly filled with other materials and a lead shield of 1 mm thickness that isplaced in front of the block. Right: Normalized radiograph.

material having a high content of hy-drogen) was investigated in a further ex-periment using ten plates made of plex-iglas of 10 mm thickness each, i.e. hav-ing a maximum thickness of 100 mm tobe penetrated (figure 6.15 left). In thenormalized radiograph (figure 6.15 left)all 10 steps can be separated clearly,thus proofing that fission neutrons arean extremely valuable tool for inspec-tion thick materials having an high hy-drogen content, like oils, water etc.

The contrast properties were investi-gated in more detail by means of a spe-cial test object. In a plexiglas block of30 mm thickness five (hollow) cylinders,partly filled with other materials, wereinserted. Additionally a lead shield of


1 mm thickness was placed on the outerside of the block (figure 6.16 left). Thecorresponding normalized radiographis shown on the right side of figure 6.16.All cylinders are clearly visible as well asthe fillings of the hollow cylinders. Thelead shield does not effect the contrast.

Technical objects

The non-destructive inspection of tur-bine blades often fails when using X-rays, gamma-rays or thermal neutronsdue to low contrast and/or penetra-tion power of these probes. Here fis-sion neutrons can be applied as demon-strated in figure 6.17. A massive turbine

Figure 6.17: Left: Photograph of a turbine blade (height about 260 mm). Middle: Normalized radiograph. Right:A two-dimensional tomo-graph of the turbine blade.

Figure 6.18: Left: Photograph of a timber (150 mm x 150 mm x 1000 mm). Middle: Normalized radiograph of a part of the timber. Right:Two-dimensional tomograph of the timber.

blade of about 260 mm in height wassuccessfully radiographed for differentangular positions. In figure 6.17 , mid-dle, one of these radiographs is shown.All cooling channels as well as innercavities can be seen clearly. A tomo-gram for one height position of the tur-bine blade is shown in figure 6.17(right).As it is one of the first results of ap-plying tomography at NECTAR, the im-age still shows many artefacts (e.g. ringand line artefacts). An improvement ofthe measurement and the reconstruc-tion parameters will drastically enhancethe image quality for the next measure-ments. Nevertheless, the position of thecooling channels and their dimension

can be determined from the tomogram.Taking into account the good pene-

tration of hydrogen containing materi-als as shown by the measurements ofthe standards, these results suggest it-self the application of fission neutronradiography/tomography on dense ma-terials containing oil distributions.

Wooden objects

Figure 6.18 shows on the left side thephotograph of a timber with the ap-proximate dimensions 150 mm x 150mm x 500 mm. Slightly off centred adrilling of about of 40 mm diameter wasrealized. For the radiography mea-


surement the timber was placed with itsedge facing the incoming fission neu-trons. The normalized radiograph (fig-ure 6.18, middle) shows the section ofthe timber containing the knob. Theknob and the annual rings are clearlyvisible. The drilling is visible, too, whenchanging the gray scale of the image. Atomograph of a section above the knobis shown in Figure 6.18(right). Althoughthe annual rings and the drilling areclearly visible, the image quality of thetomograms will be further optimizedfor the next measurement period beginof 2007.

Outlook

The results achieved at the NECTAR fa-cility during the last months of 2006demonstrated its great potential fornon-destructive inspection of manytypes of objects, but also detected areasfor improvement. Especially image pro-cessing for tomography needs some fur-ther developments, which will be final-ized at the beginning of 2007, enablinghigh quality three-dimensional tomog-raphy.

Several new areas of application willbe investigated in more details. For ex-ample, first investigations on the sen-sitivity of fission neutron radiography

on the time dependent distribution ofwater in (thick) wooden objects showedpromising results.[1] Bücherl, T., Kutlar, E., et al. Applied

Radiation and Isotopes, 61, (2004),537–540.

[2] Bücherl, T., Lierse v. Gostomski,C., et al. Proceedings of the Sev-enth World Conference, Rome, Italy,September 15-21, 401–410.

[3] Bücherl, T., Lierse v. Gostomski,C. International Workshop on FastNeutron Detectors and ApplicationsProceedings of Science, (2006).

[4] Annual Report FRM II 2005, 44.

62 7 Nuclear and particle physics

7 Nuclear and particle physics

7.1 The positron beam facility NEPOMUC and instrumentation forpositron physics

Christoph Hugenschmidt 1, Thomas Brunner 2, Günther Dollinger 3, Stefan Legl 2, Benjamin Löwe 2, Jakob Mayer 2, PhilipPikart 2, Christian Piochacz 1, Reinhard Repper 1, Martin Stadlbauer 1, Klaus Schreckenbach 2

1ZWE FRM II, TU München2Physik Department E21, TU München3Institut LRT2, UniBw München

The low-energy positronbeam of high intensity atNEPOMUC

In 2006 various experiments were per-formed at the high intensity positronbeam facility at NEPOMUC –NEutroninduced POsitron source MUniCh– inorder to improve the beam characteris-tics such as intensity, available energyrange and beam brilliance. The maxi-mum positron yield is up to 5 ·108 mod-erated positrons per second at a kineticenergy of 1 keV. At present, the lowestavailable beam energy is 15 eV with anintensity of 4 · 107 positrons per sec-ond. The maximum beam energy islimited to the high voltage, that can beapplied at the platinum foils of the in-pile source components and amountsto about 3 kV. A survey of the beam per-formance and positron beam experi-ments can be found in [1, 2].

A decrease of the positron intensitywas observed within about 20 h whichis attributed to surface contaminationsadsorbed at the platinum moderationfoils. The regeneration of these foils isachieved by exposure to a small amountof oxygen (≈ 10−1 mbar) for a few min-utes.

In the longitudinal magnetic guidefield of 7 mT the diameter of thepositron beam amounts typically to15-20 mm. Several efforts have beenmade to develop devices for beam en-hancement, since a more brilliant beam(lower divergence and reduced beamdiameter of about 2 mm) is desirablefor a variety of experiments. For this

reason an additional remoderation unitbased on a tungsten single crystal in re-flection geometry was tested at NEPO-MUC (section 7.2). Another approachis presently developed, which benefitsfrom inelastic positron scattering in agas-filled drift-chamber, in order to im-prove the beam brilliance.

Experiments at NEPOMUC

Fig. 7.1 shows experiments connectedto the positron beam facility NEPO-MUC.

The coincident Doppler-broadening

Figure 7.1: The positron beam facility NEPOMUC and positron spectrometers in the exper-imental hall of FRM II: CDB spectrometer (TUM), PAES-facility (TUM, not shown), ap-paratus for Ps−-production (MPI nuclear physics, Heidelberg), and components of thepulsed-beam facility PLEPS (UniBW, Munich).

spectrometer (CDBS) is routinely op-erated with a primary beam energy of1 keV. The beam is focused to about1 mm and can be accelerated to 31 keVonto the sample in order to allow spa-tially resolved defect studies. Vari-ous experiments have been performedin order to investigate the chemicalsurrounding at open-volume defectsof ion-irradiated metal samples (sec-tion 9.2) or defects in metals after me-chanical load.

Great efforts have been made in orderto reduce the γ-induced electron back-ground for studies with positron anni-

7 Nuclear and particle physics 63

hilation induced Auger-electron spec-troscopy (PAES). With new parametersettings of the electrostatic beam guid-ance at the entrance of the analysischamber PAES-spectra were recordedwithin only 3 h acquisition time. Low-energy positrons (15 eV) were focusedonto Au-covered surfaces of single crys-talline silicon and poly-crystalline cop-per in order to study sample surfaceswith highest sensitivity (section 9.3).

Two additional experimental setupswere installed at the multi-purposebeam port:

An apparatus for the productionof the negatively charged positronium

(Ps−) was developed at the Max-Planck-Institute for nuclear physics and con-nected to the open beam port of thepositron beam line. This year first mea-surements were performed in order toimprove the energy dependent produc-tion rate of Ps− and the signal-to-noiseratio. It was shown that the produc-tion rate is at least a factor of 25 higherthan in previously performed lab exper-iments which would allow experimentsfor QED-tests.

The angular correlation and theDoppler-shift of the annihilation pho-tons were detected in coincidence witha second experimental device at the

open beam port of NEPOMUC. For thisreason two segmented high-resolutiongermanium detectors were installed incollaboration with E12 of the physicsdepartment. At present the data analy-sis is in progress in order to reconstructthe electron momenta in three dimen-sions for each annihilation event.[1] Hugenschmidt, C., Schreckenbach,

K., Stadlbauer, M., Straßer, B. Nucl.Instr. Meth. A, 554, (2005), 384–391.

[2] Hugenschmidt, C., Stadlbauer, M.,Straßer, B., Schreckenbach, K. Appl.Surf. Sci., 252, (2006), 3098–3105.

7.2 Positron remoderation facility for the slow positron beam at FRM II

Christian Piochacz 2, Christoph Hugenschmidt 2, Gottfried Kögel 3, Klaus Schreckenbach 1, Günther Dollinger 3

1Physics Department E21, TU München2ZWE FRM II, TU München3Institut LRT2,UniBw München

In order to enhance the brightnessof the positron beam produced by theNEPOMUC source, a positron remoder-ator was developed. It has been in-stalled at the first accessible point of thebeam facility and first measurementshave been done in order to obtain theefficiency of the setup and the qualityof the remoderated beam. The idea ofusing remoderation for brightness en-hancement was first described by [1]and has been realized in several tabletop setups [2][3]. The remoderationunit described here was designed ac-cording to the ideas of the remoderatorutilized in [3] but with improvementsto accept a beam with a greater phasespace volume.

For the remoderation process,positrons are focused on a solid, wherethey stop and thermalize. There is acertain possibility that the thermalizedpositrons diffuse back to the surfacewhere they can leave the solid with asharp energy and a small angular di-vergency. The whole process dependson the properties of the solid, which isused for moderation. Materials, suchas tungsten, nickel and platinum areknown to be efficient positron mod-erators. There are basically two pos-

sibilities for remoderating a positronbeam, depending on which surface thepositrons are emitted: the reflection orthe transmission geometry. The pre-sented remoderatoration device worksin reflection geometry with a W(100)single crystal. The moderated positronsleave the crystal surface with an energyof about (3 ± 0.03)eV and an angularspread of about 0.1 eV.

Experimental setup

The remoderation setup is shown in Fig.7.2. The positrons from the NEPOMUC

source are guided by magnetic solenoidfields to the entrance of the device. Atthe last 60 cm in front of the setup amagnetic field gradient can be variedin order to adjust the ratio of longitu-dinal and transversal momentum of thebeam. The longitudinal magnetic guid-ing field is terminated by a novel fieldtermination, in order to avoid a compli-cated and unwished superposition withthe electric and magnetic fields insidethe remoderator. The field terminationis build up of 30 10 µm thick and inthe center 2 mm broad metglass stripes.This solution has the advantage of ahigh transparency even for a beam with

a great diameter up to 60 mm and en-sures nevertheless an abrupt termina-tion of the magnetic field. After pass-ing this device the positrons fly withouta guiding field till they enter the fieldof the magnetic lens and get focusedon the tungsten crystal. The remoder-ated positrons can pass the field of themagnetic lens adiabatically because oftheir low energy and are formed elec-trostatically to a beam. This beam isbent to the outlet by a perpendicularmagnetic dipole field which is locatedat the beam separator. The electrodesand the dipole have nearly no effecton the primary beam due to its muchhigher energy in the range between 0.5and 2 keV. On the outlet two variableapertures permit an adjustment of theremoderated beam onto the center ofthe axis and allow the determination ofthe beam diameter. After the remoder-ation unit the positrons are guided bymagnetic solenoidal fields to the differ-ent experiments installed at the NEPO-MUC facility. Additional electric lensesare installed at the outlet to ensure anoptimal injection of the remoderatedbeam into the magnetic field. Thehole setup is magnetically shielded bya mu-metal housing because the slow


Nepomucpositronbeamrem

oderated

positronbeam

pole shoeof magnetic

lens

remoderatorhousing

beamseparator

drives forapertures

port forpump magnetic

shielding

magnetic fieldtermination

Figure 7.2: Schematic view of the positronremoderator. The magnetic dipole andmagnetic correction coils are not shown.

positrons are very sensitive on magneticdisturbances.

Results

In order to measure the remoderationefficiency two equivalent and remov-

able annihilating targets in front of theremoderator and behind the outlet ofthe remoderator were connected. Thequality of the primary and the remod-erated beam could be evaluated due topictures taken by two systems consist-ing of a MCP, phosphor screen and CCDcamera.

We succeeded to produce very wellformed remoderated positron beams atseveral energies of the primary beam(see fig. 7.3). At an energy of 1.5keVa remoderating efficiency of nearly 2%was attained and the beam diameter ofthe remoderated beam was about 2mm(FWHM). This beam could be guidedto the open beam port of the NEPO-MUC facility and to the PAES experimentfor further measurements. The shapeof the beam could be conserved per-fectly due to the small diameter and thesmall energy spread of the remoderatedbeam.

Figure 7.3: The picture of the profile of theremoderated positron beam taken witha system of MCP, phosphor screen andCCD camera. The beam has a diameterof 2 mm (FWHM).

[1] A. P. Mills, J. Appl. Phys., 23, (1980),189.

[2] Frieze, W., Gidley, D. W., Lynn, K.Phys. Rev. B, 31(9), (1985), 5628.

[3] Kögel, G., SPM-Group. Appl. Surf.Sci., 116, (1997), 108–113.

7.3 Physics at the cold neutron beam facility MEPHISTO

H. Angerer 1,5, F. Ayala Guardia 3,5, S. Baeßler 3,5, M. Borg 3,5, L. Cabrera Brito 3,5, K. Eberhardt 3,5, B. Franke 1,6, F. Glück 3,5,W. Heil 3,5, I. Konorov 1,5, G. Konrad 3,5, N. Luquero Llopis 3,5, R. Muñoz Horta 3,5, M. Orlowski 3,5, C. Palmer 3,5, G. Pet-zoldt 1,5, D. Rich 4,5, P. Schmidt-Wellenburg 2,6, M. Simson 1,5, Y. Sobolev 3,5, H.-F. Wirth 1,5,6, O. Zimmer 1,5,6

1TU München, Physik-Department, E 182Institute Laue-Langevin, Grenoble3Universität Mainz, Institut für Physik4TU München, ZWE FRM II5aSPECT collaboration6Helimephisto group

During 2006, three different experi-ments where performed successfully atthe MEPHISTO beamline: aSPECT,cubeD2, and Helimephisto. An intro-duction to the cubeD2 experiment andits results is given elsewhere in this an-nual report. The other two experimentswill be specified below. Furthermore,many efforts went into moving our fa-cility from NL3a to NL1 within the neu-tron guide hall in the middle of 2006.The properties of both beam line posi-tions are described in the following sec-tion.

Movement from NL3a toNL1

The neutron beam at NL3a had a crosssectional area of 116 × 50 mm (height× width). It offered a cold spectrumwith a thermal equivalent flux of 2 ×1010 cm−2s−1, measured by gold foil ac-tivation [1]. The beam shutter wasplaced inside the casemate. aSPECTand cubeD2 were performed at NL3a.

At its new position MEPHISTO isplaced downstream of N-REX+. Themonochromator of N-REX+ is placedinside a concrete bunker, which nowcontains 4 m of additional neutronguide and a shutter for MEPHISTO. Theneutron guide cross section of NL1 is ta-

pered down from 120×62 mm2 to 100×62 mm2. The new exit of NL1 is situatedat the end of the N-REX+ bunker. Thecenter of the beam is 118 cm above theground level of the neutron guide hall.An additional beam attenuator can de-crease the beam intensity down to 60,20, 4, and 2 %, using borated aluminumplates with different apertures. The at-tenuator plates are situated in the shut-ter box. The spectrum of NL1 was char-acterized by K. Zeitelhack et al. in 2004;its intensity maximum is at 4.5 Å [1].MEPHISTO also provides a beam stopwhich was already used at NL3a.


aSPECT

An accurate measurement of the neu-trino/electron angular correlation coef-ficient a in free neutron decay is impor-tant in order to determine precisely theelement Vud of the CKM quark-mixingmatrix. A unitarity test of this matrixchallenges the Standard Model of par-ticle physics. aSPECT provides a preci-sion measurement of the proton spec-trum in free neutron decay. For kine-matic reasons its shape depends on theangular correlation between the mo-menta of the electron-antineutrino andthe electron [2, 3].

The setup of aSPECT is shown inFig. 7.4 at top. The neutron beam (greenarrow) is guided through the spectro-meter. Some neutrons decay in the de-cay volume, and the decay protons areguided by a magnetic field to the pro-ton detector. At the analyzing plane,an electrostatic potential barrier is ap-plied. Therefore, only protons with suf-ficient energy can overcome the barrierand be detected. By varying the barrierpotential from 0 to 750 eV one can mea-sure the integrated proton spectrum offree neutron decay. At bottom, mea-sured pulseheight spectra are shown fordifferent applied barrier voltages U atthe analyzing plane. At about channel80, the protons are visible. The pro-ton count rate decreases when increas-ing the analyzing plane voltage. Sincethe endpoint of the proton spectrumis at about 750 eV, by setting the an-alyzing plane to 780 V no events aredetected, apart from background fromdecay electrons and gamma radiation.The events below channel 55 are due toelectronic noise of the detector. Fromthe dependence of the total protoncount rate on the barrier voltage we willextract the correlation coefficient a.

Helimephisto – asuperthermal source ofultracold neutrons

Helimephisto is a prototype of a su-perthermal source for ultracold neu-trons (UCN). Inside the converter ves-sel made from stainless steel filled with

Neutrons

aSPECT

Decay volume

Beam Stop

Beam line

aSPECT spectrometer

Analyzing Plane0 - 800 V

Proton detector

Protons

Pulseheight [ADC channels]

20 40 60 80 100 120

Co

un

ts[s

]-1

0

10

20

30

40

50

AP Voltage: 50 V

200 V

400 V

780 V

500 V

Pulseheight spectrum (with background)

Figure 7.4: Left: Setup of aSPECT at NL3a. Right: Pulseheight spectra for different appliedbarrier voltages U at the analyzing plane (AP).

superfluid 4He at temperatures below1.3 K incoming cold neutrons are down-scattered to UCN energies (below ap-prox. 170 neV). The main UCN produc-tion process is single phonon excitationby neutrons with wavelength 9 Å [4, 5].The UCN are then extracted via evacu-ated stainless steel guides for detection.The helium is liquefied within the setupitself, thus being independent of sup-ply with cryogenic liquids. The cool-ing power is provided by a commer-cially available Gifford McMahon coldhead with 1.5 W at 4.2 K and also bytwo Joule-Thomson evaporation stages(4He, 3He) (for further information onthe cold head type and the liquefac-tion process see Ref. [6]). To produceUCN with our source, we performedmaesurements in two different ways: Ina storage type experiment we irradiatedthe converter with cold neutrons for adenominated time period in order tobuild up a UCN density in the helium.Then we closed the beam shutter andsimultaneously opened the UCN outlet-valve. UCN time histograms obtainedfor different irradiation times are shownin Fig. 8.4. Integrated count rates upto 1450 counts were achieved (with thebeam attenuated to 20 %). The secondway was to irradiate the converter whilethe UCN outlet-valve was open, suchthat the produced UCN could continu-ously leave the converter vessel to thedetector. Count rates up to 190 s−1 wereobtained.

Note that the goal of the experimentwas not to obtain maximum UCN out-put but to investigate the properties ofthe source and, in particular, to demon-strate for the first time efficient ex-traction of stored UCN from a super-fluid helium converter. The beam wasstrongly collimated in order to avoidactivation by scattered neutrons. AUCN production rate of 0.7 cm−3s−1

was measured for a cold flux of 1.5 ×109 cm−2s−1, close to theoretical expec-tation. The results show that there isno principle obstacle to build an ex-pile UCN source using a cold neutronbeam, which may provide UCN densi-ties of several 1000 cm−3.[1] Zeitelhack, K., et al. Nuclear Instru-

ments and Methods in Physics Re-search A, 560, (2006), 444.

[2] Glück, F., et al. Europhys. Journ. A,23, (2005), 135.

[3] Zimmer, O., et al. Nucl. In-strum. Meth. A, 440, (2000), 548.

[4] Schott, W., et al. Europhys. Journ. A,16, (2003), 559.

[5] Golub, R., Pendlebury, J. M. Phys.Lett. A, 62, (1977), 337.

[6] Schmidt-Wellenburg, P., Zimmer, O.Cryogenics, 46, (2006), 799.


Figure 7.5: Left: Setup of Helimephisto at NL1. Right: Plots of measured UCN count rates in the storage type experiments. From the timedependence of the integral UCN numbers at 1.0 K one derives a storage time of 44(3) s of UCN in the converter vessel.

7.4 Production of ultracold neutrons at the FRM II and at the TRIGAMainz

I. Altarev 1, P. Amos 1, A. Frei 1, E. Gutsmiedl 1, F.J. Hartmann 1, A. Müller 1, S. Paul 1, H.F. Wirth 1, D. Tortorella 1, D. Rich 4,K. Eberhardt 2, G. Hampel 2, J.V. Kratz 2, T. Lauer 3, N. Wiehl 2, W. Heil 3, Y. Sobolev 3, Y. Pokotilovski 6, R. Hackl 5, L. Tassini 5

1Physik Department E18, TU München2Inst. f. Kernchemie, Universität Mainz3Inst. f. Physik, Universität Mainz4ZWE FRM II, TU München5WMI Garching6Joint Institute for Nuclear Research, Dubna, Russia

Introduction

During the last year two basic exper-iments with solid deuterium (sD2) asconverter material for production of ul-tracold neutrons (UCN) have been per-formed at the FRM II and the TRIGA re-actor in Mainz. The goal of these exper-iments was to study the main importantparameters for an optimized strong sD2

source for UCN at the FRM II [1]. Theseparameters are:

1. The way of freezing out the sD2,2. The optimum temperature of the

sD2,3. The lifetime of the UCN within

the sD2,4. Dependence of the UCN produc-

tion on the cold neutron temper-ature,

5. Optical structure of frozen sD2,6. Comparison of measured and cal-

culated UCN production rates

Both experiments delivered com-plementary information on the abovementioned parameters. For example inthe TRIGA experiment only solid deu-terium frozen from the gas phase atlow temperature (6K), was studied. Inthe cubeD2 experiment the solid deu-terium was mainly produced from liq-uid deuterium.

Experimental setup at theFRM II - cubeD2

At FRM II an experimental setupcubeD2 was installed at the MEPHISTOcold neutron beam line. The setup iscomposed of two main sections, thecooling stage and the gas handlingsystem. Auxiliary parts are the para-to-ortho deuterium converter and theremote control system. The main com-ponent of cubeD2 is the converter cell.Deuterium is frozen in a cubic chamber

of aluminum (46 mm outer side length).Two sapphire windows (d=12 mm) aremounted at the side walls to allow opti-cal inspection. The bottom of the cell isin contact with a two-stage cold head.The entrance window for the cold neu-trons (thickness 0.5 mm) on the frontside is machined directly from an alu-minum block. The produced UCN leavethe target through the rear-side win-dow, a glued 100 µm thick aluminumfoil. Two diode sensors monitor thetemperature of the cell. Temperaturesdown to 6 K at the surface are easilyreachable. A heating resistor (20 Ohm),together with a temperature controller,allows to vary the temperature over awide range [5 - 300 K].

In Figure 7.7 a picture of solid frozendeuterium is shown. The cell was halffilled with liquid deuterium and after-wards slowly frozen out below 18.7 K.The crystal was of good quality and verytransparent.


Figure 7.6: cubeD2 setup

Figure 7.7: Picture of frozen solid deuterium

Figure 7.8: TRIGA Mainz setup

Experimental setup at theTRIGA reactor Mainz

A schematic view of the test facility forinvestigating the UCN production at theTRIGA reactor, Mainz, is shown in Fig-ure 7.8. So far, all test measurementswere performed at the tangential beamtube C where the thermal heat-load isrelatively small. The very end of the in-pile part with the helium cooled con-verter head is positioned just in frontof the graphite reflector. During thereactor pulse the thermal neutron fluxamounts to 1015 n/cm2/s. The UCNproduced in the solid-D2 converter areguided outside the biological shield andfinally detected. This setup is, simi-lar to the cubeD2 setup, composed oftwo main sections, the liquid He cool-ing system and the gas handling sys-tem. Auxiliary parts are the para-to-ortho- deuterium converter and the re-mote control system. With this setup itis possible to freeze out 200 cm3 of solidD2 (sD2).

Measurements with thecubeD2 setup

Figure 7.9 shows the measured UCNcount rates for two different sD2 samplecells (0.5 cm and 3 cm sD2 thickness).The UCN production was measured af-ter freezing from the liquid phase (24K -19K). The solid was produced carefullybelow 18.7K and cooled down slowlyto 8K/10K. The UCN production withinthe sD2 is roughly 50-60% higher thanthat with the liquid deuterium. TheUCN count rates from the 0.5 cm sD2

sample are higher than the rates of the3 cm sample. This effect is under-stood, if one takes into account that themean free path of the UCN in sD2 isin the range of 5-8 mm (active layer).This value can be determined by ana-lyzing the temperature dependence ofthe UCN production rate at fixed sD2

thickness, exposed to the cold neutron-beam of MEPHISTO. The key point inthis analysis is the determination of theloss cross section for UCN inside thesD2. A model for this temperature de-pendent loss cross section was used in acalculation of the measured UCN countrates and the free parameters where fit-ted. With the aid of this loss cross sec-tion it is possible to determine the av-erage UCN life time inside sD2 and alsothe UCN mean free path. If the thick-ness of the sD2 is larger than the activelayer, the effective cold neutron flux,which produces UCN inside the activelayer, is weakened by the inactive sD2

behind the active layer. The incomingcold neutron flux at the entrance of thesD2 cell was measured by gold-foil acti-vation to be 1.55 x 109 cm−2s−1.

The measured UCN count rates maybe confirmed with the common modelfor UCN production in sD2 [2] theDebye model for one phonon down-scattering of cold neutrons and theknowledge of the loss cross section forUCN inside sD2. This cross sectionfor sD2, frozen from the liquid phase,was determined by an other group byan UCN transmission measurement [3]and agrees quite well with our results.


Measurements at the TRIGAMainz setup

Figure 7.10 shows the measured UCNcount rates for two different experi-mental setup. In one arrangementthe sD2 was exposed directly to thethermal neutron flux of the TRIGA re-actor. In the second setup the sD2

was surrounded by the frozen premod-erator 1,3,5-trimethylbenzene (Mesity-lene). This premoderator was keptfrozen at a temperature of 21 K. Mesity-lene down-scatters the thermal neu-trons to the cold neutron regime. Theproduction of UCN in sD2 is most effi-cient for incoming cold neutrons withan equivalent temperature of 30K [2]. Inboth setups the amount of frozen deu-terium was varied (0 - 200 cm3 / 0 - 9mol). For smaller quantities of sD2 (0- 4 mol) the setup with the premodera-tor has roughly a gain of more than two,compared to the setup without a pre-moderator. The UCN count rate for thepremoderator setup starts to saturatefrom 4 -5 mol on. This behavior is notseen with the setup without premoder-ator. This is may be explained by thefact that sD2 itself acts also as premod-erator for thermal neutrons, but it is notas efficient as Mesitylene or other coldmoderator (C D4 or H2).Therefore thesaturation of the extracted UCN fromsD2 is compensated by adding addi-tional sD2 at higher quantities of sD2,which acts as a premoderator.

From the measurements with thepremoderator it is possible to extractthe average mean free path for UCN in-side sD2. This value is about 4 - 5 cm.The mean free path of UCN, obtainedfrom the TRIGA experiments, is a fac-tor ten larger than the mean free pathin the cubeD2 experiment. The conclu-sion of this is that the inelastic and elas-tic cross section of sD2 for UCN in sD2

ice, which is frozen out slowly from thegas phase, is much smaller than the sD2

ice cross section, produced from the liq-uid phase. This result has to be investi-gated in more detail.

Outlook

The measurements on the UCN pro-duction at the FRM II and the TRIGA re-actor in Mainz have demonstrated, thatit is possible to use solid deuterium as astrong UCN source. The measured UCNcount rates are understood and confirmthe theoretical production rates of UCNinside sD2. The UCN losses for sD2

crystals produced in different ways hasto be further investigated, though themethod of freezing from the gas phaseseems the best method for producing

Figure 7.9: UCN count rate as function of temperature

Figure 7.10: UCN count rate as a function of the amount of frozen sD2 (with and withoutpremoderator) normalized to the reactor power (10 MJ at full reactor power puls)

an efficient UCN source. Transform-ing the results to the situation of hav-ing this kind of UCN source exposedto the strong cold neutron flux of theFRM II inside the heavy water vesselshows that for a typical UCN experi-ment one can achieve a UCN density of5 -10x103cm−3.[1] Trinks, U., et al. NIM A, 440, (2000),

666.

[2] Golub, R., Böning, K. Z. Phys. B:Cond, Matt., 51, (1983), 95–98.

[3] Atchison, F., et al. PRL, (2005).

8 Industrial applications 69

8 Industrial applications

8.1 Irradiation facilities

X. Li 1, H. Gerstenberg 1, J. Favoli 1, V. Loder 1, M. Oberndorfer 1, A. Richter 1, H. Schulz 1


General

The irradiation service of the FRM IIcontinued its routine operation suc-cessfully during the five reactor opera-tion cycles in 2006. By means of

• the pneumatic rabbit system RPA,• the capsule irradiation facility

KBA,• the test rig of the silicon doping

facility SDA,• the position for short term irra-

diations of medium volume sam-ples (SDA1) and

• the high flux irradiation positionin the central control rod (RS).

altogether 334 irradiations were carriedout for different research projects andcommercial purposes. Table 8.1 showsthe irradiation numbers on each sys-tem. The silicon doping was the mostfrequently used rig and amounted toabout 50% of all irradiations carried outin 2006. Some other important and in-teresting projects supported by our irra-diation systems are described in the fol-lowing text.

Irradiations for the fissiontrack dating

Fission track dating is a radiometricdating technique based on analysis ofthe damage trails, or tracks, left by fis-sion fragments in certain uranium bear-ing minerals and glasses. The num-ber of tracks correlates directly with theage of the sample and the uranium con-tent. To determine the uranium con-

Table 8.1: Irradiations at the FRM II in 2006

position PRA KBA SDA SDA1 RS totalirradiation No. 62 66 173 32 1 334

tent the sample is annealed by heat-ing and exposed to a barrage of ther-mal neutrons. More than 20 apatitesamples from different geological insti-tutes were irradiated in our irradiationchannels in 2006. Due to the high ura-nium content in the geological sam-ples, the samples were irradiated nor-mally in the position SDA-1 with a rela-tively low neutron flux between 1×1015

and 1× 1016 (cm−2), which is achievedwithin few minutes. Standard neu-tron flux monitors (Au/Co) were usu-ally irradiated simultaneously with thesamples together and analyzed sepa-rately, in order to obtain the infor-mation about the local neutron fluxeswithin the sample sets, which typicallyconsist of more than 10 single sam-ples. Figure 8.1 shows an apatite sam-ple with fission tracks under the micro-scope (Max-Planck-Institute for nuclearphysics).

Production of isotopes forthe nuclear medicine

In the 8th reactor cycle of the FRMII, the very high flux irradiation posi-tion in the control rod was used for thefirst time for the production of 188W .It was the first successful radioisotopeproduction in the FRM II with a verylong irradiation time of an entire reac-tor cycle. The sample was loaded be-fore the reactor was started and un-loaded after the reactor was shut down,i.e. after completion of 52-days. Thetarget sample contained tungsten en-

riched to 99.9% in 186W . 188W is pro-duced via a double neutron capture re-action with thermal neutrons and de-cays to 188Re: 186W (n,γ)→187W (n,γ)→188W (β−)(T1/2=69d)→188Re(T1/2=17h).

The irradiation product is used inthe so-called 188W /188Re generatorfor medical purposes. 188Re emitsbeta particles (Emax = 2.12 MeV ) hav-ing an ideal range for intravascu-lar brachytherapy and certain cancerbrachytherapies. The specific activityof 188W in our first sample was approx-imately 155mCi/g after a cooling time of15 days. This project is a cooperationwith the Institute for Radiochemistry ofthe TU München. Further irradiationsfor the production of 188W are alreadyplanned in the next reactor cycles in2007.

Another interesting radioactive iso-tope in the nuclear pharmacy, whichis also prepared at FRM II right now,is 177Lu. This radioisotope may helpcreate the first successful radiophar-maceutical for solid tumors; it emitsa low beta energy, which reduces ra-diation side effects and produces atissue-penetration range appropriatefor smaller tumours. As a bonus, 177Luemits gamma radiation, which allowsphysicians to also use it for both imag-

Figure 8.1: apatite sample with fission tracksunder the microscope (Max-Planck-Institute for nuclear physics.)

70 8 Industrial applications

ing and therapeutic purposes. In or-der to prepare the 177Lu, Lutetium sam-ples with enriched 176Lu were irradi-ated on the capsule irradiation system(KBA) typically for 2 to 7 days. TheInstitute for Radiochemistry of the TUMünchen is trying to develop a new andmaybe a more effective way to prepare177Lu via activation of 176Y b at FRM II.Several samples were already irradiatedand the method of the isotope separa-tion will be optimized. Totally about 30samples were irradiated at FRM II forthe production of 177Lu.

Measurements of neutronfluxes and γ-dose

On the pneumatic rabbit system RPA,cadmium-ratio of neutron flux wasmeasured directly by the irradiation ofCd-shielded neutron flux monitors. Inorder to avoid unacceptable heating ofthe specimens, the irradiations wereperformed at a low reactor power of300kW, mostly at the beginning phaseof a reactor cycle. The results are com-parable to the results measured by us-ing the standard flux monitors (Au, Co,Zr, U) in the earlier reactor cycles.

High γ-dose at the irradiation posi-tions in reactor can be determined byusing thermal luminescence detectors.For irradiation of many biological sam-ples, the information about the γ-doseat the irradiation positions is very im-portant. Three thermal luminescencedetectors consisting of pure quartz wereirradiated as test samples for short timeat the RPA and the KBA. The measureddata will be analyzed by the GSF - Na-tional Research Centre for Environmentand Health, GmbH. Due to the influ-ence of the high neutron dose and theheat effect on the detectors, the mea-surement needs to be optimised.

Silicon doping facility (SDA)

The silicon doping for a commercialproduction had already begun by us-ing the prototype of the doping facil-ity (SDA-opt) at the end of 2005. Al-though only this simplified system wasavailable we entered into a business re-lation with 5 companies from Europeand Asia. A total of 173 batches sum-ming up to almost 3 tons silicon ingotswere irradiated in the 5 reactor cycles in2006.

To reduce the axial inhomogeneity ofthe neutron flux within the ingot below5% as required by the specification ofmost of clients, a special Ni-absorberis mounted on the outside of the irra-diation container. Based on the valuesof the flux measurements, its specialshape was calculated and optimised bymeans of Monte-Carlo calculations. Ac-cording to the feed-backs of the clients,our doping quality is extremely positive.Compliance with the target resistivityand axial resistivity variation is less than4 ∼ 5%. The radial variation is below 3 ∼4% and seems to be completely hiddenin measurement noise. The lifetime offree charges is high, around 400-700 ms,what means a low content of irradiationdefects. After adjustment of the calibra-tion factors connecting the irradiationdose to the target resistivity, our dopingquality can fulfill the conditions of thedoping specifications for target resistiv-ities between 20Ωcm and 750Ωcm verywell.

Most of the irradiated ingots have di-ameters of 4, 5 and 6 inches. 2 batchesof 8 inches ingots were irradiated as testsamples for the determination of thecalibration factor. Generally the start-ing material was predoped n-type Si ex-hibiting a resistivity of several thousandΩcm, but also some predoped p-typeingots were irradiated at FRM II.

Figure 8.2: lifting unit of the doping facilitymounted on the handling bridge.

Figure 8.3: new design of the coupler for thesample container.

The final silicon doping facility offer-ing a semi automatic operation was al-ready successfully tested in the factoryand verified by the responsible surveyorTÜV Süd in summer of 2006. The liftingunit of the doping facility mounted onthe handling bridge is shown in Figure8.2. Figure 8.3 shows the new design ofthe coupler for the sample container.

The whole system will be completelyinstalled at FRM II during the reactormaintenance period in January of 2007.A bunch of commissioning programscontaining about 30 tests for the dop-ing system were verified by the respon-sible surveyor TÜV Süd and the regula-tory authority. After a successful com-missioning, the new doping system willstart operations in the next reactor cyclein 2007.

8 Industrial applications 71

8.2 Fission neutron source - MEDAPP

F.M. Wagner 1, S. Kampfer 1, A. Kastenmüller 1, B. Loeper-Kabasakal 1, P. Kneschaurek 2

1ZWE FRM II, TU München2Klinikum Rechts der Isar, München

The irradiation facility

The fast reactor neutron beam at beamtube SR10 is the successor of the Reac-tor Neutron Therapy beam RENT at theformer research reactor FRM at Garch-

Figure 8.4: Along the fission neutron beam:the beam tube SR10, filters, collima-tor and the heavily shielded irradiationrooms for medical and technical use.

Figure 8.5: Medical irradiation room withtreatment couch. The fixed horizontalbeam enters from left, 1.45 m above thefloor.

ing. The last patients out of 700 at RENTwere irradiated in July 2000. The follow-up of patients is still going on and canhave an impact on the choice of suit-able cases at FRM II. Generally, neutrontherapy is most favourably applied toslowly growing and well differentiatedtumours. Most suitable sites turned outto be tumours of the head and neck,where curative treatments are possi-ble. All shallow tumour lesions, likelymph node metastases or skin metas-tases from various cancer diseases, aswell as chest wall metastases of breastcancer are suited for neutron irradia-tion, especially when pre-treated withlow-LET-radiations.

The beam at SR10 has been de-signed for multiple use, e.g., not only forclinical neutron therapy, radiobiology,and dosimetry, but also for materials-testing, and for computerized tomogra-phy and radiography. Therefore, thereare two irradiation rooms along thebeam (Fig. 8.4); in the first one (Fig. 8.5)medical applications including physicaland biological dosimetry are performed(MEDAPP), the second room is reservedfor a permanent set-up, i.e. the neu-tron computed tomography and radio-graphy facility NECTAR.

During the fission of uranium, notonly fast neutrons are produced, butalso hard gamma radiation. In orderto establish the effect of neutrons, thegamma dose rate is decreased by a filterconsisting of 3.5 cm Pb. Furthermore,a B4C-filter suppresses contaminatingthermal neutrons and reduces epither-mal neutrons. The beam is equippedwith a multi leaf collimator (MLC). TheMLC allows of the conformation of theradiation onto the contour of the tu-mour up to an area of 30 ·20cm2.

Physical beam quality

The fast neutron flux is up to 7.108 s-1cm-2 (depending on filters and colli-mation); the mean neutron energy is 1.6

MeV. At 5 cm depth of a water phantom,the neutron and gamma dose rates ofthe medical beam with collimation 9x9cm2 are 0.33 Gy/min and 0.18Â Gy/min,respectively [1, 2]. An example of depthdose curves is shown in Fig. 3 8.6. Thedose distribution in the depth of a wa-ter phantom is the basis of the physi-cal treatment planning. In order to de-termine the neutron and gamma com-ponents separately, two chambers areused with different sensitivities to neu-trons and gammas.

With the combination of filters fortherapeutic irradiations (1 cm B4C-PEand 3.5 cm Pb), the neutron-to-photonratio decreases from about 3.6 nearto the surface to 1.8 at 5 cm depthof the phantom. The half-maximumdose rate of the neutrons is at about54 mm depth in water. The depthdose measurements have been certifiedby the Physikalisch-Technische Bunde-sanstalt Braunschweig (PTB). The steepdecrease of the dose rate with depthis the reason that only superficial andnear-surface tumours can be irradiated.

MCNP calculations of the neutronspectra and of the corresponding doserates were carried out for various phan-toms. Calculated neutron spectra for aPE phantom using an MCNP code areshown in Fig. 8.7.

Two current projects deal with the ex-perimental determination of the neu-tron spectra at SR10. One method usesa set of passive Bonner spheres withgold probes. The activation data of theprobes activated in the center of thespheres with sizes up to 15” undergo adeconvolution procedure with respectto the spectral response of the spheres.The project is carried out together withthe Institut für Strahlenschutz at GSFNeuherberg, Germany (H. Paretzke, W.Rühm, S. Studeny), and is part of an on-going PhD thesis.

For radiation protection purposes itis necessary to know the neutron spec-tra also outside of the heavily shieldedirradiation rooms. For this sake, 16

72 8 Industrial applications

Figure 8.6: Neutron and gamma dose rates Dn and Dg , resp., vs. depth in a water phantomon the beam axis using the medical filter combination.

Figure 8.7: Monte-Carlo simulation of the neutron spectrum at the patient site. Closedrhombi: unfiltered beam; open squares: beam filtered by 1 cm B4C-PE and 3.5 cm Pb

active PE spheres with sizes rangingfrom 2.5” to 15” were placed at 3 sites.In this case, the more sensitive He-3-counters are placed amidst of thespheres instead of gold probes.

Measurements of lineal energy trans-fer spectra by use of the microdosimeterAMIRA (from Strahlenbiologisches In-stitut, Universität München, LMU), anddose measurements with the newly de-veloped electronic personnel dosimeteron a PMMA phantom (M. Wielunsky)accomplished the action with GSF.

Towards the end of 2006, anothermethod to determine fast neutron spec-tra has been started within a diploma

thesis using threshold reactions of Rh-103, In-115, Ti-47, P-31, Zn-64, Fe-54,Al-27, and others (H. Breitkreutz). Incontrast to the Bonner sphere method,the use of threshold probes allows formeasurements within a phantom andfor spatial resolution.

Biological beam quality

In context with the clinical applicationof the beam, the determination of theRelative Biological Effectiveness, RBE, isof great importance. In a master thesisby V. Magaddino (University of Naples,

Italy), the dependence of the RBE of fis-sion neutrons on dose and on gammacontamination has been investigated inmegacolonies of human squamous cellcarcinoma [3]. This project was con-ducted within the course “EuropeanMaster of Science in Radiation Biol-ogy” under the auspices of the Univer-sity College London (K.-R. Trott), andin co-operation with the Radiobiolog-ical Institute of the University of Mu-nich, LMU (J. Kummermehr). It wasshown that fractionation and depth inphantom give rise to a variation of theRBE from 1.4 to 4.7. The investigationwill be continued.

Preparation for clinicalapplications

In order to be applied to humans,MEDAPP had to be checked accordingto the Medical Devices Directive, MDD93/42/EEG. After an effort of about 10man-years, the CE mark has been fixedat the facility. The permission for pa-tient treatment is pending.

For the administration of the pa-tients’ personal data, the program RIS-SKA was developed in the medical fac-ulty of TUM (Thamm) [4]. This programalso allows for controlling the adminis-tered single doses, the number of frac-tions, and for documenting the appliedirradiations. The parameters of each ir-radiation field (i.e. doses, position ofthe multi leaf collimator, position of thepatient) can be reported by words andby implemented photo documentation.RISSKA offers also data base functionsand statistical features for the evalua-tion of patient and irradiation data.[1] Kampfer, S., Wagner, F., Loeper,

B., Kneschaurek, P. MedizinischePhysik, 318–319.

[2] Loeper-Kabasakal, B., Thamm, R.,Kampfer, S., Wagner, F., Kasten-müller, A., Lange, W., Kneschaurek,P. ESTRO 25, poster 892.

[3] Magaddino, V. Thesis for the De-gree of Master of Science in Radia-tion Biology. Master’s thesis, Uni-versity College London (2006).

[4] Thamm, R., Loeper, B., Wagner, F.,Molls, M. DEGRO, (2006).

Part II

Science

74 9 Scientific highlights

9 Scientific highlights

9.1 Solvent content in thin spin-coated polymer films

J. Perlich 1, V. Köstgens 1, L. Schulz 1, R. Georgii 2, P. Müller-Buschbaum 1

1Physik-Department E13, TU München2ZWE FRM II, TU München

The detection of remaining solvent inthin polymer films is of importance dueto its effect on chain mobility and filmhomogeneity. Moreover, it gives an es-timate on possible aging effects causedby the reduction of the solvent content,which typically yield an increased brit-tleness.

In the present investigation, we fo-cus on a well controlled model sys-tem, which consists of protonatedpolystyrene (PS) with different molec-ular weights Mw of 7, 27, 207, 514, 908,1530 kg/mol, spin-coated out of proto-nated or deuterated toluene (solvent)onto silicon (Si) wafer substrates. Di-rectly after spin-coating the thin PSfilms were investigated with neutronreflectivity (NR) at the MIRA instru-ment at a wavelength of 16 Å. A narrowqz range (0 Å−1 to 0.02 Å−1) around thecritical edge was probed with high res-olution. Due to the high sensitivity ofthe position of the critical edge on theratio of protonated PS and deuteratedtoluene (toluene-d8), the exact positionof the critical edge enables to deter-mine the solvent content. In additionto the deuterated solvent samples, sam-ples with thin PS films spin-coated outof protonated toluene were also pre-pared for a direct comparison of thedifferent sample types [1]. In order torule out all kinds of measurement er-rors precautions on the sample part,e.g. preparation and repetitions withdifferent but identical samples, are veryimportant. With respect to the instru-ment, a very accurate alignment of theexperimental set-up is crucial for thosemeasurements.

Experimental results

The recent experiment focuses on twodifferent key parameters which influ-ence the solvent content: the molecu-lar weight of PS and the film thicknessinvestigated in the range of 10 to 100nm. Thus, two respective sample se-ries were prepared. Focussing on themolecular weight as key parameter, thinPS films with a fixed thickness of 50nm are investigated: The expected shiftof the critical edge position, which isobserved in neutron reflectivity simu-lations on this model system, is veri-fied by the MIRA measurements. Figure9.1a shows one particular example for amolecular weight of 7 kg/mol. In directcomparison the reflectivity of the sam-ple prepared out of deuterated and pro-tonated solvent is plotted. The intensityis shown on linear scale to emphasizeon the region of total external reflection.

The obtained reflectivity data ofthin PS films with different molecularweights and a fixed film thickness are

Figure 9.1: (a) Neutron reflectivity data obtained from spin-coated thin polystyrene (PS)films with a molecular weight Mw = 7 kg/mol and fixed film thickness. Depending onthe use of protonated and deuterated (d8) toluene, the position of the critical edge shiftssignificantly. (b) Neutron reflectivity data of a selection of thin PS films prepared of dif-ferent molecular weights Mw of 7, 27, 907 and 1530 kg/mol and fixed film thickness. Withincreasing molecular weight the position of the critical edge shifts towards higher qz val-ues.

shown in figure 9.1b. The data show ashift of the critical edge with increasingmolecular weight, although the mea-sured critical edges of much higher Mwshift not completely but rather indicatea slightly changed slope of the criticaledge. For the investigation of the filmthickness as the second key parameter,PS with a molecular weight of Mw =207 kg/mol is dissolved in toluene-d8.The thickness of the thin PS film is de-pending on the viscosity of the solutionand thus the concentration of PS in thesolution. Therewith the concentrationis chosen in such a way to achieve adesired film thickness [2]. The thick-ness series comprises thin PS films of athickness of 10, 30, 50, and 100 nm. Fig-ure 9.2a shows the obtained reflectivitydata for films with thickness of 10, 30and 100 nm at fixed molecular weight.The data of the critical edge indicatean influence on the film thickness, butprevent a definite conclusion about thebehaviour in dependence of the filmthickness without fitting the data.

9 Scientific highlights 75

Figure 9.2: (a) Neutron reflectivity measurements from spin-coated thin films with PSmolecular weight Mw =207 kg/mol and film thicknesses of 10, 30 and 100 nm. An in-fluence of the film thickness is indicated by the reflectivity data. (b) Simulated neutronreflectivity data for a single PS film on top of bulk Si. Simulated for a film SLD of pro-tonated PS and deuterated toluene. Since the measured data is a mixture of both, themeasured reflectivity lies within the simulated borders.

Data analysis

The simulation and data analysis ofthe neutron reflectivity measurementsis performed with Parratt32, a simula-tion tool for neutron and x-ray reflec-tivity. The model layer stack in ambientair consists of bulk Si with a single layeron top, representing the PS film. In or-der to match the simulated curve withthe measured curve of the PS/toluene-d8 film only the scattering lenght den-sity of the PS film is adjusted, whereasthe magnitude of the adjustment repre-sents a measure for the ratio of proto-nated PS and deuterated toluene. Sincethe relation SLDtol < SLDPS < SLDtol−d8

applies for the sample materials andsince the film will be a mixture of PS

and toluene-d8, the SLD for deuteratedtoluene is a reasonable maximum forthe adjustment. To rule out the inde-termination of the PS film thickness, x-ray reflectivity (XRR) measurements areperformed, since the NR measurementsin the narrow qz range provide no esti-mation of the PS film thickness. In fig-ure 9.2b the NR simulations for a PS filmare shown, whereas the measurementsshould lie within the regions defined bythe extreme values of the SLD.

Summary and outlook

In summary, the experiment was verysuccessful. A direct comparison indi-cates a clear distinction between thinfilms prepared of PS dissolved in pro-tonated or deuterated toluene. This di-

rectly transforms into the amount ofsolvent remaining in the polymer film.Depending on the molecular weight ofPS the shift of the critical edge is pro-nounced to a greater or lesser extent,but clearly visible. In total, the actualsample preparation conditions are ofimportance and affect the amount ofsolvent inside the PS films. As a conse-quence, well defined experimental con-ditions of the PS film are essential towork out the solvent content. Due tothe close vicinity of the MIRA instru-ment and the polymer preparation lab-oratories, this is perfectly fulfilled.

Following MIRA experiments will beexpanded to a technically more relevantsample system consisting of nanocom-posite films, prepared by the combina-tion of an amphilic diblock-copolymer,which acts as the templating agent, andan inorganic sol-gel chemistry. Thisparticular nanocomposite films are em-ployed in new photovoltaic devices. Theinvestigation of such a system is of im-portance, since starting from the natu-rally remaining solvent content in thefilm after preparation and the subse-quent decrease during aging alters thestructural properties of the nanocom-posite film and thus results in loss ofperformance.[1] Spangler, L., Torkelson, M., Royal, J.

Polym. Eng. Sci., 307, (1990), 644–653.

[2] Schubert, D. Polymer Bulletin, 38,(1997), 177.

9.2 Investigation of AZ31 and ion irradiated Mg with the coincidentDoppler broadening spectrometer CDBS

M. Stadlbauer 1, C. Hugenschmidt 1, K. Schreckenbach 2

1ZWE FRM II, TU München2Physik-Department E21, TU München

The chemical vicinity of open vol-ume defects in alloys is of great inter-est in material science since it is cru-cial for the stiffness and tensile strength.The coincident Doppler broadeningspectroscopy (CDBS) with positrons al-lows very sensitive measurements ofthe electron momentum distribution in

defects due to the efficient trappingtherein [1]. After annihilation with coreelectrons from the neighboring atoms,the Doppler shifted 511keV annihila-tion radiation reveals the momentumdistribution of the involved electronswhich depends in particular on theperiodic number of the element. A

detailed description of CDBS and theCDB-spectrometer at NEPOMUC hasbeen published in [2].

Sample Preparation

Samples with 20 × 20 × 3mm3 consist-ing of pure, polished and annealed Mg


have been irradiated with Mg-, Zn-and Al-ions at the 3 MeV-Tandetron inRossendorf. The energy of the ionswere chosen between 1.4 and 3MeV ac-cording to 2.3µm mean implantationdepth. Implantation of Mg-ions intoa Mg-sample ensures that only defectschange the CDBS-signature. Zn and Alions were implanted into Mg in orderto study the influence of these elementsto the CDBS-signature in combinationwith defects. Single defects and disloca-tions created by this treatment in mag-nesium anneal already below 200K butvacancy clusters survive up to 400K [3].For each ion type a set of 4 samples wasproduced with doses between 3 × 1013

and 3× 1016 ions/cm−2 in order to finda sensitivity threshold for the low doseson the one hand and to get into the re-gion of saturation trapping of positronsfor the high doses on the other hand.

The diameter of the ion beam fromthe tandetron was reduced to 5mmsince spatially resolved positron scanswith a resolution of 2mm should imagethe ion beam spot on the sample.

Additional samples consisting ofpure, polished and annealed AZ31 wereirradiated with Mg-ions of 1.4 MeV en-ergy. The same doses were applied asdescribed above.

Measurements and results

The samples with the maximum iondose were measured up to now. First,the width of the annihilation line vs. po-sition and energy was recorded with aspacial resolution of 2mm and positronenergies between 1 and 9keV, since

6 8 10 1214

0,5000,5020,5040,5060,5080,5100,512

68

1012

14

S-pa

ram

eter

[a.u

.]

Y [mm]

X [mm]

Figure 9.3: Width of the annihilation line ina.u. vs. position on the Mg-ion irradi-ated AZ31-sample measured with 4.5keVpositron energy.

the most obvious variation of this pa-rameter was expected to occur at lowenergies due to results from previousmeasurements of the annihilation linewidth in the irradiated area (see fig.9.3)which can be clearly detected between1 and 6keV. For higher energies the ir-radiated region could not be separatedfrom the untreated one which is an un-expected result, since the mean pene-tration depth of the Mg-ions was 2.3µmcorresponding to 17keV positrons. Apossible explanation for this is the veryhigh ion dose (cion = 35% in the irradi-ated region), which may have led to lo-cal annealing effects due to local heat-ing during the irradiation procedure.

The coincident spectra of pure Mgirradiated with Mg-, Al- and Zn-ionsshowed a clear deviation from thepure and untreated Mg-samples (seefig.9.4). Nevertheless the signatures ofthe irradiated samples are statisticallynot distinguishable among each otherwithin the error bars between 511 and513keV. In particular there is no Zn-signature in the Zn-ion irradiated Mg-sample detectable. This signature isshifted to lower values between 513 and514.5keV in contrary to the Zn-curve

511 512 513 514 515 5160,85

0,90

0,95

1,00

1,05

1,100 2 4 6 8 10 12 14 16 18

Zn in Mg 17kV Mg Ref 17kV Mg in Mg 17kV Zn Ref 25kV

ratio

to M

g R

ef [a

.u.]

energy [keV]

electron momentum [10-3m0c]

Figure 9.4: CDB ratio curves of pure Mg and irradiated Mg with Zn- and Mg-ions. The Zn-ionirradiated Mg-sample shows no Zn-signature and is clearly shifted towards the signatureof the Mg-ion irradiated sample.

which shows a large deviation in thehigh momentum region.

Outlook

Since the high ion dose may have led tolocal annealing effects in the samples,the next important step is to measurethe samples with lower ion dose. Firstinvestigations on these samples werevery promising.[1] Puska, M. J., Nieminen, R. M. Rev.

Mod. Phys., 66, (1994), 841–897.

[2] Stadlbauer, M., Hugenschmidt, C.,Piochacz, C., Strasser, B., Schreck-enbach, K. Appl. Surf. Science, ac-cepted for publication.

[3] Ehrhart, P. In Ullmaier, H., edi-tor, Atomic Defects in Metals, 242(Springer-Verlag GmbH, San Diego,CA, 1991).


9.3 PAES-measurements of pure Cu and Cu coated Si(100)

Jakob Mayer 1, Christoph Hugenschmidt 2, Klaus Schreckenbach 1


Experimental setup

A spectrometer for positron annihi-lation induced Auger electron spec-troscopy (PAES) has been installedat the high intensity positron beamNEPOMUC. PAES is based on thesame principle as conventional electroninduced Auger electron spectroscopy(EAES), but with different precedingionization process. Since electrons inan atom are bound with high energy,the incoming electrons need an energyof at least a few keV to ionize the atomsby impact. At PAES the ionization is re-alized by electron-positron annihilationand hence positrons with a very low ki-netic energy are sufficient (Ee+ ≤ 40eV).Due to the low positron energies, thereis no background in the higher ener-getic regions of the Auger peaks. (n,γ)-reactions of surrounding experimentsin the experimental hall at the FRM IIinitially led to a higher background thanexpected (see fig. 9.5). With the help of alead shielding this external backgroundhas been reduced considerably.

40 60 80 100 1200

2

4

6

8

10 new settings old settings

Cu M2,3

VV

Cu M2,3

VV

I [cp

s]

energy [eV]

Figure 9.5: Old PAES-measurements of Cu incomparison with the new measurements(without background). The backgroundwas reduced by lead shielding and im-proved settings of the electric lenses atthe entrance of the analysis chamber.

Characteristics of thepositron beam at the samplesite

The energy of the positrons is definedby the electric lenses and platinumstructure in the tip of the NEPOMUC-beamtube [1]. In order to measurethe energy distribution of the primarypositron beam, the energy of the 30eVpositrons at the entrance of the analysischamber was measured using a retard-ing grid in the longitudinal magneticguiding field and a NaI-detector. In ad-dition the geometric dimensions of thebeam have been specified with a MCPand a CCD.

Since only a little fraction (< 0.6%)of the positrons has an higher energythan the expected 30eV the influence ofthem to the background is almost neg-ligible. The diameter of the beam wasdetermined to 20mm.

Measurements and results

Figure 9.5 shows the reduction of thebackground due to the lead shieldingand enhanced settings of the electriclenses at the entrance of the Augerchamber. The signal to noise ratio withPAES was increased from 1/2 to 2 andis hence a factor of 6 higher than withconventional EAES.

Another feature of PAES, the veryhigh surface sensitivity, was demon-strated with a measurement of clean sil-icon and of the same sample coveredwith 1.5 monolayer of copper (see fig.9.6). At the coated sample almost 1/5 ofthe whole Auger-signal, i. e. the sum ofall detected Auger electrons, originatesfrom copper. With EAES this quotient isonly 1/13, since the electrons penetrate

deeper into the Si-substrate. This showsthe high surface sensitivity of PAES.

Outlook

The results of the recent year show thehigh potentials of this unique facility.For a further reduction of the measure-ment time a new electron energy ana-lyzer is required, since the current ana-lyzer has an extremely low detection ef-ficiency. With a new analyzer the de-tectable area will be enlarged and withnine channeltrons instead of one oreven a MCP the measurement time isexpected to be reduced at least by afactor of 50. With this investment anda new sample holder for the heatingand cooling of the samples the PAES-spectrometer enables temperature de-pendent surface investigations.[1] Hugenschmidt, C., Kögel, G., Rep-

per, R., Schreckenbach, K., Sperr,P., Triftshäuser, W. Nuclear Instru-ments and Methods in Physics Re-search B, 198, (2002), 220–229.

40 60 80 1000

2

4

6

8

10

12

50 60 70 80 90 100 110

0,4

0,5

0,6

0,7

0,8

0,9 clean Si (100) Si (100) with 1.5ML Cu

I [cp

s]

energy [eV]

Cu M2,3

VV

Si L2,3

VV

Figure 9.6: Single crystalline clean siliconand the same sample covered with 1.5atomic layers of copper.


9.4 ANTARES – Investigation of an early medieval sword by neutrontomography

Martin Mühlbauer 2, Elbio Calzada 2, Brigitte Haas-Gebhard 3, Rupert Gebhard 3, Klaus Lorenz 1, Burkhard Schillinger 2,Michael Schulz 2

1 Physics Department, E21, TU München2ZWE FRM II, TU München3Archäologische Staatssammlung, München, Germany

We have studied a sword of the 6thcentury AC from Pforzen near Kempten,Germany. During a complex restora-tion parts of the scabbard that was con-structed of different layers of leatherand wood could be conserved. It waspresumed that these parts still containorganic material. Therefore it was ofinterest for us to compare the resultsof the x-ray analysis to neutron radiog-raphy and tomography measurements,as neutrons are sensitive to the hydro-gen in the remaining organic material,which is completely invisible for x-rays.This assumption could be verified bythe examination results.

Introduction

The preservation of archaeological ar-tifacts depends on the various corro-sion processes of the artifacts duringburial. Since wetland finds are scarcein Bavaria and dry or cold preserva-tion is missing, organic material is veryrare. There are however a few cases ofpreservation of organic parts sticking tometallic artifacts in early medieval buri-als. They are mainly conserved by ironoxides diffusion into the organic parts.In most cases the organic material wastotally converted into iron or copper ox-ides, sometimes however organic mate-rial could remain.

The results of the neutron tomogra-phy allow a reconstruction of the or-ganic scabbard. It was constructed oftwo wooden parts that have been cov-ered at the inner side with leather orskin and leather at the outer side. Thisconstruction has been strengthened bywinding a string around near the toppart of the scabbard. Additionally somebronze parts are mounted at the tip ofthe scabbard and where it is attached tothe belt. The center part of the swordblade was decorated in damask tech-

nique. This however could also be ob-served by x-ray techniques.

The organic parts are shown very wellby neutron tomography. The attenua-tion of the neutron beam caused by theorganic material is enhanced due to theconservation treatment by resins. Thediffusion of these resins within the ob-ject is uneven. Neutron tomographycould therefore be used to check the

Figure 9.7: Right: overview of the four tomography data sets taken from the sword and thescabbard, left: different slices and cuts of the reconstructed data.

conservation process as well. The in-vestigated sword is property of the FreeState of Bavaria and kept in the Archae-ological State Collection (Archäologis-che Staatssammlung) in Munich [1].


The neutron radiographyand tomography facilityANTARES at FRM II

The measurements were carried out atthe neutron radiography and tomogra-phy facility ANTARES, which is oper-ated by the Institute for experimentalphysics E21.

By exchanging a part of the main col-limator the effective aperture diame-ter of the collimator system can be re-duced from 4.1 cm to 2.15 cm. Thustwo main beam geometries are avail-able at ANTARES, which result in L/Dratios of 400 and 800 respectively. Thecorresponding neutron flux at the sam-ple position after a 16 m flight path is1.0 · 108 s−1cm−2 or 2.5 · 107 s−1cm−2 .Additional apertures made of cadmiumallow to increase the L/D ratio up toabout 16000. These apertures are usedfor phase contrast measurements pre-dominately. The energy spectrum ofthe neutron beam can be described bya Maxwellian distribution for thermalneutrons with an enhancement in thecold energy range due to the cold sourcefilled with liquid deuterium at 25 K infront of the radiography beam tube.The maximum beam size at the sampleposition is 40 cm by 40 cm.

The detector system we used for themeasurement was a CCD-camera incombination with a scintillation screen.The field of view was set to 14 cm by14 cm and a L/D ratio of 800 was used.One scan showing the complete swordand four tomography data sets show-ing only parts of the sword were taken.Two of the data sets display an upperpart and two form a lower part of thescabbard and the sword (fig. 9.7). Thedistance of the axis of rotation of thesample to the scintillator was 10 cm

Figure 9.8: Sketch of the structure of thescabbard. From the inside to the outsideit is built from fur, wood, leather, cord,bronze.

and the minimum distance of the sam-ple surface to the scintillator was about8 cm. For each radiograph the expo-sure time was 20 seconds. As 400 pro-jections over 180 degrees were taken foreach tomography data set, it required 3hours to acquire the raw data of one to-mography (readout time included). Theachieved resolution was limited by thescintillator to about 0.5 mm.

Results

The scabbard is composed from theinside to the outside of different lay-ers: fur, wood, leather, cord, bronze.The structure of the composition of thescabbard is visible in cuts through thethree dimensional dataset. So it is pos-sible to check the existence and con-dition of the different layers over thewhole length of the sample.

These layers are not existent onthe whole scabbard, e.g. the bronzestrengthening is only placed at the tipof the scabbard and where it is attachedto the belt of the warrior. For a sketchshowing the different layers of the scab-bard see figure 9.8.

The two edges of the sword are stillin good condition (fig. 9.9). The cor-rosion process starts in the center re-gion of the sword where different ironmaterial was used for the damask tech-nique. A thin intermediate layer isfound between the sword and the scab-bard. Here small connections are foundbuilt from the fur on the inner side ofthe scabbard and the corrosion prod-ucts of the sword.

Cracks in the wooden parts of thescabbard or in the sword can be foundand followed. The same is true forthe damask structure in the core of thesword.

At the tip of the sword a small part hasbroken away from the rest of the sword(fig. 9.10). This is not visible from out-side because of the bronze cover.

We found a heavily attenuating partat the cone end of the scabbard. Atfirst we were wondering if it could be apart of the sword, which had been sep-arated from the blade. We could not ex-plain why its position was so close tothe border of the scabbard. It seemed

Figure 9.9: Cut through the reconstructed3D data set showing a crack in the scab-bard.

Figure 9.10: Small piece of the blade apartfrom the rest of the sword.

not to belong to the sword, because itwas too far away from the rest of thesword and did not merge to it on ei-ther end. Then we compared the x-ray radiograph with the neutron radio-graph again, and it turned out that ithad to be an organic part, because itwas only visible in the neutron radio-graph (see fig. 9.11). The existence ofthis part was not known before and itwas unclear why it was placed there. Itis likely that it was used to fix the tipof the sword when it was placed in thescabbard. This special construction wasdetected on early medieval swords forthe first time. It takes further investiga-tions to verify this construction elementon other early medieval scabbards. Ithas to be discussed if this part was ageneral construction element or it wasonly applied at the sword of this study inconnection with a repair after the loss ofthe bronze plate at the back side of thetip of the scabbard.

Conclusions

The application of neutron radiographyand tomography methods in this fieldof archeology is a very useful non de-structive technique for the study of theinner structure of archaeological arti-facts. Comparable results could onlybe achieved by mechanical removal ofthe different layers of the object. This


however would destroy the objects irre-versibly. Neutrons are not the only toolto investigate archeological objects, butthey give further information to com-plete the research results. In combina-tion with x-ray measurements archeo-logical objects can be investigated with-out damage or loss of material. X-rays show primarily the remains of themetallic sword while neutrons show thesurrounding organic material in addi-tion. Thus it is easy to focus on thesword using x-rays. The organic mate-rial of the scabbard is suppressed. Thiswas very important to ensure that thestrong attenuating part found at thecone end of the scabbard was wooden.Looking at the scabbard from the out-side, it does not give the impression oforganic material due to the heavy up-take of metal oxides during the decom-position process.

The wooden part to fix the tip of thesword is an example for a perfect com-plement of x-rays and neutrons.

Acknowledgment

We thank the restoration department ofthe Archeological State Collection Mu-nich for sample preparation and pro-viding the x-ray data.[1] Archeologische Staatssammlung.

http://www.lrz-muenchen.de/~arch/start.html.

Figure 9.11: Comparison of x-ray radiography (b) and neutron tomography (c): A strip ofwood could be detected under the bronze metal sheet that covers the front side of thescabbard. The bronze metal sheet of the backside has been lost in antiquity.

9.5 Containerless sample processing in combination with quasielasticneutron scattering

A. Meyer 1,2, S. Stüber 2, D. Holland-Moritz 1, O. Heinen 1, T. Unruh 3

1Institut für Materialphysik im Weltraum, Deutsches Zentrum für Luft- und Raumfahrt, Köln2Physik Department E13, TU München3ZWE FRM II, TU München

The study of structure and dynamicsin liquid metals is often prevented by achemical reaction of the high tempera-ture melt with its sample holder. Thiscan be overcome by the use of an elec-tromagnetic levitation apparatus (EML)that allows for a containerless process-ing of electrically conducting samples.At the DLR in Cologne a compact andmobile EML was designed and built for

the use on neutron diffractometers [1].The absence of a sample holder madeit not only possible to extend the acces-sible temperature range for liquid met-als research up to 2300 Kelvin, but alsoto undercool liquid metals and alloysseveral hundreds of Kelvin below theirmelting points. This led to new findingsconcerning the short range order in liq-uids [2].

The roughly spherical, electricallyconductive samples, 6-8 mm in diame-ter, are levitated within an inhomoge-neous electromagnetic radio frequencyfield.

As a result of the RF-field, eddy cur-rents are induced in the specimen. Onthe one hand, this leads to an induc-tive heating of the sample which al-lows melting of the specimen. On the

http://www.lrz-muenchen.de/~arch/start.html

http://www.lrz-muenchen.de/~arch/start.html


Figure 9.12: Electromagnetically levitatedliquid Ni droplet at a temperature of1810 K. The water cooled copper coil isshielded with a neutron absorbing rub-ber. The radial collimator of Tof-Tof is onthe right.

other hand, the interaction of the eddycurrents with the inhomogeneous mag-netic field of the levitation coil leads to aforce that acts on the sample. Accordingto Lenz’s rule this force points into thedirection of low magnetic field strength.If it is adjusted such that gravity is com-pensated, the sample is levitated. Theconvective stirring induced by the in-ductive currents in combination with alarge heat conductivity of the sample re-sults in a homogenous sample temper-ature that is contact-free measured witha two-color pyrometer. The tempera-ture is controlled via the flow of ultrapure cooling gas (He/4%H2 in the caseof Ni) which is injected by a nozzle thatis installed below the sample.

For the use on neutron time-of-flightspectrometers the EML was equippedwith a newly developed Cu coil that pro-vides an 8 mm gap between the upperand the lower part of the coil. Thisenhanced the visibility of the samplefor the incoming beam and for thedetectors significantly. In combina-tion with the high flux and the excel-lent signal to noise ratio of the neu-tron time-of-flight spectrometer ToF-ToF quasielastic measurements on lev-itated metallic droplets are now fea-sible. In a first measurement cam-paign we investigated liquid Ni. Fig-ure 9.13 displays the quasielastic signalthat is dominated by incoherent scat-tering below about 2.2 Å−1. Please notethe semilogarithmic representation of

the scattering law S(q,ω). The linesare fits with a Lorentz function that isconvoluted with the instrumental en-ergy resolution function. The q de-pendence of the quasielastic signal con-tains information on the mechanism ofatomic transport. Via the width of thequasielastic line the self diffusion coef-ficient D can be derived on an abso-lute scale with high precision. Over alarge temperature range – from morethan 200 K above to more than 200 K be-low the melting point at 1727 K – thetemperature dependence of the diffu-sion coefficients can be described withan Arrhenius law. The change from aliquid in thermodynamical equilibriumto a metastable liquid at large under-coolings is not reflected in its atomictransport [3].

With this new sample environmentwe are now able to study the interplay ofstructure, dynamics and properties e.g.of Fe and Ti based melts that form inter-metallic phases and of Zr based meltsthat form bulk metallic glass at large un-dercoolings. In a next step we are de-veloping an electrostatic levitator. Sucha device will give access to even highertemperatures and to non metallic liq-uids.

It is a pleasure to thank StephanJanssen and the FOCUS team of thePaul Scherrer Institute for their supportduring a first test experiment on FOCUSand Suresh Mavila Chathoth, Fan Yang,Tarik Mehaddene, Jürgen Brillo, Helena

-1 0 1 2

10-2

10-1

10 0

hω (meV)

S (

q,ω

) (

meV

-1)

1514 K1870 K

0.9 A-1

0.50 0.55 0.60 0.65

2000 K Tm 1500 K

2

3

4

5

1000 / T (K-1)

D (

10-9

m2

s-1)

liquid Ni

Figure 9.13: Quasielastic spectra of liquid Ni (left). Measuring times ranged from 1 h to 3 h.Ni self diffusion coefficients obtained from the width of the quasielastic line (right). Theopen circles are results from QENS measurements on liquid Ni in an alumina crucible[4].

Hartmann and Thomas Volkmann fortheir help. We acknowledge finan-cial support by the German DFG (SPPPhasenumwandlungen in mehrkompo-nentigen Schmelzen) under Grant No.Me 1958/2-3 and Ho1942/6-3.[1] Holland-Moritz, D., Schenk, T.,

Convert, P., Hansen, T., Herlach,D. Meas. Sci. Technol., 16, (2005),372–380.

[2] Schenk, T., Holland-Moritz, D., Si-monet, V., und D.M. Herlach, R. B.Phys. Rev. Lett., 89, (2002), 075507–075510.

[3] Meyer, A., Stüber, S., Holland-Moritz, D., Heinen, O., Unruh, T.Phys. Rev. Lett., (eingereicht).

[4] Chathoth, S. M., Meyer, A., Koza,M., Juranyi, F. Appl. Phys. Lett., 85,(2004), 4881–4883.


9.6 Larmor diffraction at TRISP

T. Keller 2, K. Buchner 2, C. Pfleiderer 3, P. Böni 3, B. Keimer 1

1Max-Planck-Institute for Solid State Research, Stuttgart2Max-Planck-Institute for Solid State Research, outstation at FRM II3Physics Department E21, TU München

The TRISP (triple axis spin-echo)spectrometer at the FRM II is a ther-mal triple axis specrometer incorpo-rating the NRSE (neutron resonancespin-echo) technique. TRISP is opti-mized for high resolution spectroscopyof phonons and magnons with an en-ergy resolution down to the µeV range[1, 2]. Here we report on the high res-

G

B

L

q

kG

=/2

k1

k2

B

L 0

polarizeranalyzer

detector

L

G

qa

k2

k1

(b)

k1k

2

(a)

rf spinflip coil

Figure 9.14: Larmor diffraction setup (a) ba-sic principle: a polarized neutron beamtravels through a field B, where the neu-tron spins undergo Larmor precession.The precession phase is proportional tothe component of k perpendicular to thefield boundaries, which is the same forall Bragg diffracted neutrons (k⊥ = G/2)if the field boundaries are parallel to thelattice planes of the diffracting crystal.Tilting the lattice planes by an angle α

leaves the precession phase constant tofirst order (k1⊥ + k2⊥ = const). Thus themethod also works for powder samplesusing beam divergencies in the order ofone degree. (b) Larmor diffraction con-figuration at TRISP. The DC field B is re-placed by two pairs of rf spinflip coils.The field boundaries are defined by theprecise windings of these coils.

olution Larmor diffraction (LD) modeof TRISP, which was first proposed byRekveldt [3, 4] and is also based onthe NRSE technique with inclined fieldboundaries. LD at TRISP typically of-fers a resolution∆d/d ' 2×10−6 for sin-gle crystal and powder samples, i.e. atleast one order of magnitude more ac-curate than conventional neutron or x-ray diffractometers. LD is operated withopen collimation and thus with a highintensity beam. Additionally the spreadof the lattice spacing can be measuredwith high accuracy (typically ∆d/d '10−5 FWHM for a Gaussian distribu-tion). In a second mode, LD allows foraccurate determination of the mosaicspread of single crystals.

During the last year, LD at TRISPwas used for a series of thermal expan-sion measurements under extreme con-ditions with pressures up to 20kbar andtemperatures down to 500mK (lowertemperatures will be available) [5, 6].This parameter range is interesting fora variety of systems showing uncon-ventional phases or phase transitions.So far, no experimental technique wasavailable to determine thermal expan-sion under these conditions with suf-ficient accuracy: Dilatometry has ex-cellent resolution, but doesn’t work inpressure cells. High resolution high en-ergy x-ray diffraction offers typically aresolution around ∆d/d ' 10−5, whichis in most cases not good enough formeaningful thermal expansion mea-surements. Additionally it causes toomuch sample heating, thus the temper-atures are limited to values above 4K.The resolution of conventional singlecrystal neutron diffraction in the orderof 10−4 is also insufficient.

The basic idea of the LD techniqueis to mark each single neutron by aLarmor precession phase such that thephase only depends on the lattice spac-ing d and is independent of the Braggangle or the velocity of the single neu-tron. Thus the resolution is inde-

pendent of the beam collimation andmonochromaticity. The principle isshown in fig. 9.14a: an initially polar-ized neutron runs through a homoge-neous magnetic field B with the fieldboundaries oriented parallel to the lat-tice planes of the diffracting crystal. Theinitial polarization is perpendicular toB. The total Larmor phase after passingB twice (incident and scattered beam) isgiven by the time T the neutron spendsin the field. T only depends on the com-ponents of k1,2 perpendicular to thefield boundaries:(9.1)

φtot =ωLT =ωLLm

ħ(

1

k1⊥+ 1

k2⊥

)ωL = 2πγB with γ = 2.916kHz/Gauss isthe Larmor frequency. The Bragg con-dition requires that all diffracted neu-trons have the same k-component per-pendicular to the lattice planes:

(9.2) k sinθB =G/2

where G = 2π/d is the modulus of areciprocal lattice vector. If the latticeplanes are oriented parallel to the fieldboundaries, k1⊥ = k2⊥ =G/2 and the to-tal phase is

(9.3) φtot = 2mωLL

πħ d

A change of the lattice spacing gener-ates a phase shift

(9.4) ∆φ=φtot∆d

d

φtot can be as large as 104rad. With anormalized distribution function f (∆d)of the lattice spacing variations, thecomponent of the neutron beam polar-ization measured by the analyzer is(9.5)

P (φtot ) = ⟨cos(∆φ)

⟩= ∫

f (∆d/d)cos(φtot

∆dd

)d

(∆dd

)i.e. the cos-Fourier transform of f (∆d).

To test the performance of LD atTRISP, we measured the thermal expan-sion of a copper single crystal using the


[111] reflection. Cu is a calibration stan-dard for dilatometers and accurate dataon the thermal expansion are available[7]. Fig. 9.15 shows the data obtained atTRISP and reference data from capaci-tive dilatometry (no fit!), where the ac-curacy of the dilatometric data is typ-ically three orders of magnitude bet-ter than the TRISP data. The excellentagreement shows that the LD techniqueworks reliable on a 2× 10−6 relative er-ror level. The accuracy of LD can be in-creased at least by one order of magni-tude by increasing the length of the pre-cession regions, by using longer wave-length neutrons and by careful con-trol of the spectrometer stability, whichis currently mainly limited by thermaldrifts of the length of the precession re-gions.[1] Bayrakci, S., Keller, T., Habicht, K.,

Keimer, B. Science, 312, (2006),1926.

[2] Keller, T., Aynajian, P., Habicht, K.,Boeri, L., Bose, S., Keimer, B. Phys.Rev. Lett., 96, (2006), 225501.

[3] Rekveldt, M. T. Mater. Sci. Forum,321-324, (2000), 258.

[4] Rekveldt, M. T., Keller, T., Golub, R.Europhys. Lett., 54, (2001), 342.

[5] Pfleiderer, C., Böni, P., Keller, T.,Rößler, U., Rosch, A. To be pub-lished.

[6] Niklowitz, P., Pfleiderer, C., Keller, T.,Mydosh, J. To be published.

[7] Kroeger, F. R., Swenson, C. A. J. Appl.Phys., 48, (1977), 853.

0.0

5.0x10-4

1.0x10-3

1.5x10-3

2.0x10-3

2.5x10-3

3.0x10-3

3.5x10-3

0 50 100 150 200 250 300-1.0x10-5

0.0

1.0x10-5

thermal expansion of Copper

dilatometry TRISP Larmor diffraction

L(T

)/L(

5K)

- 1

diffe

renc

e

T[K]

Figure 9.15: Thermal expansion of single crystal Cu [111] measured both with Larmordiffraction (dots) and dilatometry. In the lower part the difference between the LD anddilatometry data is plotted. The larger error bars above 150K result from thermal drifts inthe order of 0.5K of the sample during the measurement.


9.7 Molecular dynamics in pharmaceutical drug delivery systems

C. Smuda 1, G. Gemmecker 2, H. Bunjes 3, T. Unruh 1

1ZWE FRM II, TU Muenchen2Chemie Department, Organische Chemie II, TU Muenchen3Institut fuer Pharmazeutische Technologie, TU Braunschweig

Introduction

In recent years it has been shown thatthe development of new drugs alone isnot sufficient to ensure progress in drugtherapy. Today there are many drugswhich in spite of their high pharma-cological potential cannot be adminis-tered using standard formulations be-cause of e.g. side effects, targetingproblems or adverse release effects. Apromising strategy to overcome theseproblems is the development of suit-able drug carrier systems [1]. Emulsionsof liquid oils are under investigation forthese purposes. During the develop-ment process it turned out that the ma-terial properties of the lipids used asmatrix material may change dramati-cally upon dispersion into the colloidalstate. One particular phenomenon isthe highly extended supercooling rangewhich has been observed among oth-ers for coenzyme Q10. It has beenproposed to use corresponding super-cooled emulsions as drug carrier sys-tems [2, 3]. For the dispersion of thissubstance an enhanced bioavailabilityhas been demonstrated in cell cultureexperiments [4].

The dynamics of the carrier and theemulsifier molecules influence impor-tant technological properties of the for-mulation as the drug release rate ordrug leakage. But only few investi-gations on the molecular dynamics oflipid drug delivery systems have beenpublished so far. Quasi-elastic neutronscattering (QENS) is a quite unknownmethod in pharmaceutical science butit is a powerful tool to gain a deeperinsight in the dynamics of drugs andexcipient molecules leading to a betterunderstanding of e.g. release behaviourof the drug.

In this article we present first re-sults of QENS investigations on Q10-nanodispersions which are of interestas drug carriers and exhibit an uncom-mon supercooling behaviour.

Further investigations on the dy-namics of phospholipid molecules in-side a monolayer stabilizing emulsiondroplets are presented. Although phos-pholipids are used as stabilizers inmany pharmaceutical dispersion for-mulations the stabilization mechanismespecially during formation and crys-tallization of the particles, respectively,is not well understood.

Experimental details

The Q10-nanodispersions were pre-pared by high-pressure homogeniza-tion of the molten Q10 (10 weight%)and D2O containing 1.6% phospho-lipid Lipoid S100 and 0.4% sodiumglycocholate (SGC) as emulsifiers.The corresponding reference sam-ple was prepared by the same pro-cedure but contained only the two

Figure 9.16: Observed half linewidth Γ of the first Lorentzian as a function of Q2 for pure

Q10 at two temperatures. According to D = Γ/(ħQ2) the self diffusion coefficients couldbe determined from the slopes of the regression lines. For 293 K a diffusion constantD = 8.7·10−7 cm2/s and for 323 K a diffusion constant D = 1.9·10−6 cm2/s were calcu-lated [5].

emulsifiers and D2O. The particle sizewas analyzed using photon correla-tion spectroscopy (PCS). The averageradii (z-averages) for all Q10-dispersionswere in the range from 60 nm to 120 nm.

Four n-hexadecane-d34 (HD) na-noemulsions were prepared by meansof high-pressure homogenization or ul-trasonic homogenization. They con-tained 5% HD, 2% dimyristoylphos-phatidylcholine (DMPC); 5% HD, 2%Lipoid S100; 5% HD, 1.6% DMPC, 0.4%SGC and 5% HD, 1.6% Lipoid S100, 0.4%SGC dispersed in D2O, respectively. Thereference sample was prepared by ultra-sonic homogenization of a mixture of5% HD and 2% sodium dodecylsulfate-d25 (SDS) in D2O. The particle sizes (z-average) determined by PCS were in therange from 31 nm to 56 nm.

Neutron scattering measurementswere performed with the time-of-flight


spectrometer TOFTOF at FRM II. Sam-ples were measured with an incidentwavelength of 6 Å and a chopper speedof 16000 rpm leading to an instrumentresolution of about 35µeV. PFG-NMRdiffusion measurements of Q10 disper-sion and pure Q10 were performed on aBruker DMX NMR Spectrometer oper-ating at 600 MHz 1H frequency.

Results

Self diffusion measurements of Q10

molecules and Q10-nanodroplets ina dispersion

We have studied nanodisperse liquidQ10 (melting point: 320 K) in the moltenand supercooled state and neat Q10 inthe same temperature range.

The QENS spectra were fitted witha scattering function Sinc(Q,ω) whichcomprises of two Lorentzian functions,of which the first one describes thelong-range diffusion and the secondfunction approximates internal mo-tions like methyl group rotation in Q10,rotational motions of the whole Q10

molecule and chain motions of the Q10

molecule. The model function was con-voluted with the resolution function ofthe instrument achieved from a vana-dium measurement. From the firstLorentzian function a Q2 dependenceof the linewidth was obtained. The cor-responding motion was assigned to along-range diffusion. Exemplarily theQ dependence of the linewidth for twotemperatures is displayed in Fig. 9.16.A physically meaningful interpretationof the Q dependence of the linewidthof the second Lorentzian function, how-ever, could not be found so far, althougha good description of the experimen-tal data using this simple model wasachieved.

Testing other models which includedetailed expressions for superimposedlong-range diffusion and isotropic ro-tational diffusion, or long range dif-fusion and uniaxial rotational diffu-sion and different models including re-stricted diffusion, respectively, it couldbe extracted as a result that the onlyconstant and reproducible parameterachieved was the linewidth of the first

Lorentzian function, i.e. the derived selfdiffusion coefficients. No reasonablephysical results for the other parame-ters of the scattering function could beobtained: e.g. the rotational diffusioncoefficient has shown an unphysical Qdependence. Obviously we do not knowenough about the dynamics which in-fluences the scattering function. There-fore we started to investigate the localmotions of Q10 in more detail which isdescribed later in this article.

Q10 self diffusion coefficients deter-mined at the same temperatures fornanoparticles and neat Q10, respec-tively, exhibit no differences within thestatistical error. Hence no ”nanoeffect”owing to the dispersed state could befound. But this was not expected in theinvestigated size range of the nanopar-ticles since the size of the droplets(∼ 120 nm) is still large in relation tothe size of a Q10 molecule (∼ 6 nm).However, it could be demonstrated thatsubtracting QENS spectra of the matrix(D2O) from the spectra of the dispersionlead to nearly identical QENS spectra ofdispersed and bulk Q10. This is an im-portant result for further investigationsand not self-evident taken into accountpotential interactions of the dispersedphase with the dispersion medium andthe scattering of the emulsifier.

PFG-NMR diffusion experiments for

Figure 9.17: Temperature dependence of the most probable jumping rate Γ0 obtained fromfits of amorphous Q10 in the emulsion droplets.

a Q10-nanodispersion and liquid Q10

were performed. This method pro-vides self diffusion constants on a mil-lisecond timescale. Interestingly themeasured NMR values for neat Q10 aresmaller compared to the QENS diffu-sion constants. For 293 K the self diffu-sion constant determined by PFG-NMRis DQ10 = 3 · 10−8 cm2/s and this is 29times smaller than the QENS value andfor 323 K we found DQ10 = 1 ·10−7 cm2/swhich is by a factor of 20 smaller thanthe QENS diffusion constant.

The diffusion constant of the nan-odroplets inside the dispersion mediumat 293 K yielded a diffusion constant of4 · 10−8 cm2/s. This diffusion constantis in perfect agreement with values esti-mated from PCS measurements [5].

Methyl group dynamics in Q10

In order to get more information onthe local motions in pure Q10, therotation of the methyl groups of theQ10 molecules in the nanodroplets wasstudied at low temperatures between50 K and 200 K. This dynamics shoulddominate the scattering function in thegiven temperature range. According tothe different local environment of themethyl groups in the amorphous Q10

samples we applied the rotation ratedistribution model (RRDM) [6] to our


QENS data. In this model a rotationaljump process for the methyl group is as-sumed with variable rate constants dis-tributed according to a logarithmic nor-mal distribution around the most prob-able jump rate Γ0.

A good agreement of the model withthe data was found for temperaturesfrom melting point to 180 K. Fromthe Arrhenius plot which is displayedin Fig. 9.17 an activation energy of4.3 kJ/mol was extracted. This is in goodagreement to values found for simi-lar substances [7] taken into accounta mixture of different types of methylgroups in the Q10 molecule. The de-viation of the data at higher tempera-ture from the linear behaviour seems tobe caused by the soaring recrystalliza-tion of the nanoparticles with increas-ing temperature. A model describingthis effect is currently under investiga-tion.

Phospholipid dynamics inmonolayers

From temperature resolved QENS mea-surements on emulsions of perdeuter-ated n-hexadecane in D2O stabilizedwith different emulsifiers informationof the dynamics of the phospholipidmolecules in the stabilizing monolayercould be extracted. The scatteringcontributions of the continuous phaseand of the interior of the hexadecane-droplets were removed by subtractionof the scattering curves of the referencesample. Therefore the obtained scatter-ing curves are dominated by the scatter-ing of the emulsifier molecules. It wasenhanced even more by using phos-pholipids as the only protonated com-ponent of the formulation.

For our first experiments we selectedDMPC which is a well investigatedphospholipid and Lipoid S100 which isused as a standard emulsifier in phar-maceutical formulations. Sodium gly-cocholate was used as a co-emulsifier inthe same samples.

In spite of a very small amount ofonly 20 mg phospholipid in the beamit was possible to detect changes of thephospholipid dynamics due to temper-ature change and temperature inducedphase transitions. Exemplarily the neu-

Figure 9.18: QENS spectra of DMPC in the monolayer of n-hexadecane-d34 nanoemulsionsat different temperatures. The displayed spectra were obtained after subtraction of thespectra of the fully deuterated reference sample.

tron scattering spectra of DMPC in themonolayer of n-hexadecane-d34 na-noemulsions are displayed in Fig. 9.18.The reduced elastic line intensity of adispersion stabilized with DMPC andSGC compared to a dispersion stabi-lized only with DMPC indicates that theaddition of SGC increases the mobilityof the DMPC molecules. This enhancedmobility of DMPC might be responsi-ble for a better stabilization of lipid nan-odispersions when adding SGC whichcould in fact be observed for similar dis-persions during and after the crystal-lization of their liquid nanodroplets [8].

Furthermore, the temperature de-pendence on the elastic scattering in-tensity is stronger for the samples con-taining DMPC compared to the cor-responding samples containing S100as the phospholipid component. Thiseffect, which is particularly obviousfor the temperature step from 20C to30C can be attributed to the gel/liquid-crystalline transition of the DMPCmonolayer [5]. However, due to thepoor counting statistics a quantitativeanalysis of the QENS-signal will only bepossible from further experiments with

higher emulsifier concentrations andextended beam time.[1] Mehnert, W., Maeder, K. Adv. Drug

Del. Rev., 47, (2001), 165–196.

[2] Bunjes, H., Siekmann, B., Weste-sen, K. Submicron Emulsions inDrug Targeting and Delivery (Har-wood Academic Publishers, 1998).

[3] Siekmann, B., Westesen, K. Pharm.Res., 12, (1995), 201–208.

[4] Stojkovic, M., Westesen, K., Za-khartchenko, V., Stojkovic, P., Box-hammer, K., Wolf, E. Biol. Reprod.,61, (1999), 541–547.

[5] Unruh, T., Smuda, C., Gemmecker,G., Bunjes, H. Mater. Res. Soc. Symp.Proc, 137.

[6] Chahid, A., Alegria, A., Colmenero,J. Macromolecules, 27, (1994), 3282–3288.

[7] Mukhopadhyay, R., Alegria, A.,Colmenero, J., Frick, B. Macro-molecules, 31, (1998), 3985–3993.

[8] Siekmann, B., Westesen, K. Pharm.Pharmacol. Lett., 1, (1992), 123–126.

10 Events 87

10 Events

10.1 Workshop - Residual stress analysis

Michael Hofmann 1

1ZWE FRM II, Technische Universität München

On the invitation of FRM II andthe chair of metal forming of the TUMünchen (Prof. Hoffmann) the an-nual meeting 2006 of the committeefor residual stress (FA13 "Eigenspan-nungen") of the AWT (Working groupof thermal treatment and materials sci-ence - Arbeitgemeinschaft Wärmebe-handlung und Werkstofftechnik e.V.)was held in Garching from the 30.-

31.5.2006. The chairman, Prof. Scholtes(University of Kassel), could welcomemore than 60 participants to the two-day workshop. The talks on the firstday mainly focused on possibilities andprospects of residual stress analysis us-ing neutron diffraction methods. Dur-ing the second day several aspects ofthe virtual institute "Photon and Neu-tron Research on Advanced Engineer-

ing Materials", which was introduced inthe beginning by Prof. Reimers (Tech-nical University Berlin), were presentedand discussed. After the meeting mostof the participants took the opportunityto visit the research reactor and to viewthe experimental halls and the instru-mentation of FRM II.

10.2 Workshop - Neutrons for Geoscience

Jürgen Neuhaus 1


The workshop Neutrons forGeoscience, organized by theForschungsneutronenquelle HeinzMaier-Leibnitz FRM II, took place onJuly 14th in Garching. The aim wasto bring together scientists from thebroad range of earth sciences as geol-ogy, crystallography, mineralogy andothers using neutron beams for theirresearch. Invited talks as well as shortcontributions gave an overview of ac-tual and future applications of neutronsin these scientific fields. Around 45 par-ticipants followed the presentations inthe new lecture hall of the faculty ofmathematics, where the workshop tookplace.

Hans Keppler from the BavarianGeoinstitut, Bayreuth started the pre-sentations with an overview on exper-iments under extreme conditions, i.e.high pressure and temperature simulta-neously. These in-situ experiments cangive new insights in the physics of theearth mantle. Focusing on texture in-vestigations, Bernd Leiss from the Cen-ter of Geoscience, Göttingen gave an

impressive talk starting from the cloudbuilding in the sky to the inner partsof the earth. Especially the large sam-

Figure 10.1: Participants of the workshop including the organizing team.

ple volumes of the rock material neces-sitates the usage of neutrons beams fora quantitative analysis. Simon Redfern

88 10 Events

from the Department of Earth Sciencesin Cambridge gave new insights whatcombined high pressure and tempera-ture research using neutron scatteringcan investigate. Unconventional neu-tron scattering techniques were pre-sented by Bjoern Winkler from the In-stitute of Geosciences, Frankfurt. Hedemonstrated how besides the tradi-tional diffraction experiments inves-tigations using inelastic, quasi-elasticand magnetic scattering as well as SANSand radiography/tomography are ap-plied for research in the earth science.In the last presentation Karsten Knorrfrom the Institute of Geoscience, Kielreviewed the development of high pres-sure cells and presented recent experi-ments taken at the FRM II.

During lunch break the poster ses-sion took place including presentationsof the instruments from the FRM II.Lively discussions about recent and fu-ture experiments pointed out the largeinterest of the geoscience communityto use the experimental facilities of theFRM II.

The workshop closed by a summaryof Winfried Petry who was impressedby the large variety of methods start-ing from neutron diffraction to inelas-tic scattering, activation analysis and

Figure 10.2: Enthusiastic discussion of Winfried Petry (right) with experts from Göttingen(from left to right: Bent Hansen, pensive, Bernd Leiss, astonished and Götz Eckold, notyet convinced).

imaging techniques applied to researchin the earth sciences. He emphasizedthe interest of the FRM II to estab-lish the infrastructure for experimentsunder extreme conditions, especiallyfor the geoscience community. Theseextreme conditions of high temperature

and high pressure will be made avail-able on existing instruments and willplay a major role in the future develop-ment of instruments in the new easternguide hall which is actually under con-struction.

10.3 Potential industrial applications of the research neutron sourceForschungsneutronenquelle Heinz Maier-Leibnitz (FRM II)

Ralph Gilles 1, Jürgen Neuhaus 1


To start further co-operation be-tween TU München Forschungsneutro-nenquelle Heinz Maier-Leibnitz (FRMII) and industrial partners a workshoptook place on September 12th, 2006in Garching. The workshop was orga-nized under the auspices of the VDI-Werkstofftechnik (presented by Dr. H.J.Schäfer) settled in Düsseldorf, Ger-many. About 50 participants were reg-istered at the workshop. The compa-nies as BMW AG, DLR, Siltronic AG,MTU Friedrichshafen, Infineon, Gen-eral Electrics, Hilti and Osram as well assmall companies from the surrounding

of Munich were present and involved incontributions.

The scope of the workshop was tobring together different groups as peo-ple from industry, scientists working inthe field of applied science with neu-tron scattering methods and scientistswho are responsible for neutron instru-ments which are used for industrial rel-evant applications. The main topicwas to work out for which industrialproblems neutrons are a unique probeto solve open questions or provide animportant complementary method toachieve a complete characterization of

the sample. As an introduction tothe workshop a compilation of differ-ent talks were presented. Half of thepresentations were coming from indus-try partners to present first results byusing neutrons as a probe for their in-vestigations. In the second part of thetalks instrument responsibles gave anoverview about the potential of the in-struments and the experience with in-dustrial customers to provide informa-tion especially for novices. The maintopics which were discussed rely onthe possibilities to measure stress andstrain in automotive components, to x-

10 Events 89

ray objects in a non-destructive way byusing tomography or radiography andthe possibility to irradiate samples withneutrons. This includes the produc-tion of radioactive isotopes, the activa-tion analysis for highly sensitive chemi-cal analysis, the production of radioac-tive tracers and doping of semiconduc-tors (transmutation doping).

After the talks a poster presentationtook place where industrial customershad the opportunity to discuss in de-tail how to perform experiments at thedifferent instruments and how to applyfor beam time. For many participantsthe highlight has been the guided tourthrough the FRM II facility at the end ofthe workshop. It enabled them to lookvery close to the real instruments and tocatch a glimpse of the heart of the facil-ity, the reactor pool.

Figure 10.3: Spirited discussions during the lunch break and poster session.

10.4 Workshop for neutron scattering from biological and soft matterinterfaces

Bert Nickel 1

1Department für Physik, Ludwig-Maximilians-Universität München

On 22nd September 2006, the work-shop on neutron scattering from bi-ological and soft matter interfaceswas held at the Forschungsneutronen-quelle Heinz Maier-Leibnitz (FRM II) inGarching, Germany. The workshop wasorganized jointly by the REFSANS teamfrom the LMU München and GKSS,Geesthacht, respectively. During thefirst talks the scientific perspectivesfor neutron scattering in areas suchas membrane physics and surfactantchemistry were highlighted by AdrianRennie (Uppsala U.) and Tim Salditt (U.Göttingen).

The performance of the new neu-tron reflectometer REFSANS at theFRM II was reported by R. Kamp-mann (GKSS). REFSANS is a uniquetime of flight instrument which al-lows switching from a reflectometrymode to a grazing incidence small an-gle neutron scattering mode (GISANS)

Figure 10.4: Participants in front of the FRM II main building.

90 10 Events

by remote control. The first experi-ments in the GISANS mode using het-erogenous polymer films (group of P.Müller-Buschbaum, TUM) obtained re-sults which demonstrate that REFSANSwill be able to compete with the re-spective ILL instruments. Furthermore,first reflectometry experiments indicatethat a dynamic range of eight ordersof magnitude seems to be possible infavourable cases.

In the afternoon session presenta-tions from M. Tanaka (U. Heidelberg),I. Köper (MPI Mainz), R. Willumeit(GKSS), P. Vandoolaeghe (Lund U.)and the group of C. Papadakis (TUM)ranged from the study of biomimeticsurface architectures to bacterial de-fence mechanisms.

A special focus of the workshop wasto discuss infrastructure and sampleenvironment needs for such experi-

ments. In this context, a new micro-fluidic cell which combines fluores-cence microscopy and neutron reflec-tometry was presented by Bert Nickel(LMU). These cells reduce the amountof liquid required for the neutron ex-periments to 3 ml, an important aspectif rare (e.g. deuterated) biomoleculessuch as membrane proteins are to bestudied.

10.5 Farewell colloquium for Prof. Dr. Klaus Schreckenbach

Jürgen Neuhaus 1


On the occasion of the leave ofProf. Klaus Schreckenbach as techni-cal director of the FRM II, the Tech-nische Universität München organizeda farewell colloquium on 15th May2006. Prof. Schreckenbach has di-rected the reactor department fromMay 1999 to December 2005. Dur-ing this time he had to overcome ahuge number of hurdles to obtain thenuclear license for the reactor and toput him final into operation in March2004. The welcome address of the uni-versity was held by the vice presidentProf. Rudolf Schillinger. Prof. Win-fried Petry reviewed in a pleasing pre-sentation the scientific and private lifeof Klaus Schreckenbach, showing as-pects ranging from nuclear physics tohill climbing in the Himalayas moun-tains. The commemorative speech washeld by Prof. Dieter Richter, FZ-Jülichon soft matter research at the JCNS.

The evening was closed by a Bavariandinner, where numerous discussionsconcerning neutrons, reactors and evenhill climbing have accompanied an in-

teresting event. Prof. Klaus Schreck-enbach has retired from his position astechnical director of the FRM II to pur-sue an academic task in the Physics De-

Figure 10.5: Winfried Petry and Klaus Schreckenbach trying to display Himalaya pictures.

partment. He concentrates now on re-search with positrons and education ofstudents.

Part III

Facts and Figures

92 11 Experiments and user program

11 Experiments and user program

J. Neuhaus 2, U. Kurz 2, B. Tonin-Schebesta 2, W. Wittowetz 2, E. Jörg-Müller 2

2Forschungsneutronenquelle Heinz Maier-Leibnitz, FRM II, TU München

The user office

The FRM II is operated as a user facil-ity with international access to the in-struments. The available beam time isshared to 2/3 for proposals which haveto be approved by an international ref-eree committee and 1/3 is given to thescientific groups which maintain the in-struments. By this we ensure well main-tained and technical up to date instru-ments which enable first-class experi-ments.

Business as usual might not be theright expression for the first year of fulloperation and 260 days neutron beams.However, in the year 2006 we had again2 proposal rounds in January and Au-gust. This cycle in between the dead-lines of other facilities seems to be anappropriate timing for our users. Alto-gether we got 301 proposals requesting2700 days of beam time which lead to anaverage overload factor of two. So thisscenario is not different from the one ofthe previous year.

Figure 11.1: Received proposals by country of the main proposer.

More interestingly we performed intotal, i.e. including the internal beamtime for the instrument groups, 687 ex-periments during 3001 days of beamtime, summed up over all instruments.As we were in the first year of full op-eration still 17% of this beam time wasused for instrument development andimprovement. As several instrumentsare still in the commissioning phase thisnumber is expected to decrease duringthe next years.

The proposals originate from a largenumber of different countries. Sum-ming up the statistics from 2005 and2006 we counted 62% of the propos-als from Germany, 30% from Europeancountries, eligible for support from theFP6 NMI3 initiative and the rest of 8%from other countries, mainly from Rus-sia and the United States.

The proposal submission is done on-line in the user office system (http://user.frm2.tum.de). Each proposerhas to register for a personal account inthe system. Using the module Proposal

the user is guided through a question-naire concerning the experimental pa-rameters. In addition a written scien-tific motivation in form of a pdf file witha length of maximal two A4 pages is re-quired. A detailed guide line for pro-posal submission is given in the sectionuser guide in the online system.

In addition to the proposal submis-sion the user office system collects ex-perimental reports which are requestedfor each experiment. The reports will bepublished online on our Internet pages.

Instruments

In the year 2006 the instruments couldprove their performance in a large num-ber of experiments. Nevertheless finetuning and improvements as well aspartly commissioning took about 17%of the beam time as mention above.The increasing number of publica-tions (14.1) shows shows the scientificoutput of the experiments, even if alarge number are still under prepara-tion or to be submitted.

The table 11.1 summarizes all instru-ments, working, commissioning or un-der construction. The number given incolumn position corresponds to the po-sition number in figure 11.2.

http://user.frm2.tum.de

http://user.frm2.tum.de

11 Experiments and user program 93

Instrument position operated by experimentsExperimental hall

Panda SR2 (17) TU Dresden 36, 231 daysStress-Spec SR3 (18) HMI 45, 238 daysAntares SR4b (19) TU München 93, 194 daysTrisp SR5b (21) MPI Stuttgart 22, 260 daysPuma SR7 (22) Univ. Göttingen, TU München 36, 198 daysSpodi SR8a (23) TU Darmstadt, LMU München 69, 231 daysResi SR8b (24) Univ. Augsburg, LMU München 28, 220 daysHeidi SR9b (25) RWTH Aachen 29, 195 daysNectar SR10h (26) TU München 42, 193 daysMedapp SR10v (27) TU München commissioningNepomuc∗ SR11 (28) TU München, Univ. Bundeswehr 21, 139 days

Neutron guide hallN-Rex+ NL1 (1) MPI Stuttgart first experimentsMephisto NL1 (5) TU München 6, 260 days∗∗NSE NL2ao (2) JCNS commissioningTofTof NL2au (3) TU München 28, 160 daysRefsans NL2b (4) GKSS, LMU München first 10 experiments, 208KWS-3 NL3a (6) JCNS under constructionKWS-1 NL3a (7) JCNS under constructionKWS-2 NL3b (8) JCNS under constructionPGAA NL4b (10) Univ. Köln, TU München commissioningReseda NL5ao (11) TU München 3, 21 days†

Spheres NL6a (14) JCNS commissioningDNS NL6a (15) JCNS under constructionMira NL6b (16) TU München 47, 248 days

Table 11.1: List of instruments at the FRM II in December 2006. Experiments in cycles 5-9 including instrument development. (∗Nepomuc

3 Instruments, ∗∗including setup of the experiments,† partly commissioning )

,QVWUXPHQWVQ5(; 0HSKLVWR 6$16 0$5,$ 3$1'$ 75,63 +(,',16( .:6 3*$$ 63+(5( 675(6663(& 380$ 0('$3372)72) .:6 5(6('$ '16 $17$5(6 632', 1(&7$55()6$16 .:6 1263(& 0,5$ 723$6 5(6, 1(3208&

1/ 1/DR 1/DX1/E

1/D1/E1/D

1/E1/DR1/E

1/E 1/D

65E65D

65E

65D 65

7XQQHO

.DVHPDWWH

65E

65

6565

65

1/DX

65

Figure 11.2: Position of the instruments in the experimental and neutron guide hall.

94 11 Experiments and user program

User support

Business as usual, however, might bethe right expression for the user officeteam that helps to organize the stay atthe FRM II for external visitors. Theyprovide support for hotel booking andother organizational help. A centrale-mail ([email protected]) andphone number (+49 899 289 14313) isreachable during normal office hours.A guest house is still not yet availableon site, which unfortunately will notchange in 2007.

European users are supported by theNMI3 project, under which travel andsubsistence costs can be reimbursed forup to two scientists per experiment. Sofar all proposals which have been ac-cepted by the referee committees couldbe supported.

Industrial applications

The presentation of the FRM II forindustrial researcher has a long tra-dition for our institute. Already inthe early days during the constructionphase of the reactor, we organized reg-ularly workshops and seminars. Thisyear we have presented industrial appli-cations with neutrons together with theVDI-W (10.3 in order to get closer con-tacts to engineers.

In the year 2006 we participated atthe symposium Material Innovativ, Au-tomobil – Forschung – Energietechnikwhich took place at the University of

Figure 11.3: Exhibition of the FRM II on the cluster congress.

Bayreuth on 29th March. WinfriedPetry explained in a lecture how neu-tron could be used to design new mate-rials. The symposium was organized byBayern Innovativ, a publicly held com-pany that was set up by the BavarianState Government in 1995. It was initi-ated by politics, science and industry asa corporation for innovation and tech-nology transfer.

Already in 2005 the Bavarian Min-istry of Economic Affairs, Infrastruc-ture, Transport and Technology initi-ated the Cluster Strategy in order to fur-ther mobilize the cooperation of indus-

try with research institutions and uni-versities. The inauguration of the clus-ter took place at the M,O,C, fair centerMunich on 2nd February 2006. The or-ganization was done by Bayern Innova-tiv in collaboration with the TechnischeUniversität München. More than 1000participants and 80 exhibitors joined inthe successful congress. The FRM II waspresented by Dr. Ralph Gilles and Dr.Jürgen Neuhaus who explained the pos-sibilities of neutron scattering for thedevelopment of new materials and theirradiation facilities of the FRM II for el-ement analysis and isotope production.

[email protected]


12 Public relations and visitor service

J. Neuhaus 3, U. Kurz 3, B. Tonin-Schebesta 3

3Forschungsneutronenquelle Heinz Maier-Leibnitz, FRM II, TU München

Our visitor service could welcome2480 people in 2006. Some of the moreprominent ones are depicted on thefirst pages (3). The demand has againexceeded our possibilities for guidedtours. We are pleased, that we could in-crease the number of visiting studentsat the FRM II by more than a factor ofthree. In 2006 we had 47 groups and intotal 707 students on site visiting our in-stitute. In addition 20 groups of pupils(278 persons) could be welcomed dur-ing the year. Some times it even hap-pens that VIP’s bring along students totheir visit.

The day of open doors 2006 wasjointly organized by the scientific cam-pus in Garching. This day, the 15th Oc-tober 2006 was put together with the in-auguration of the underground U6 withthe terminal station ForschungsgeländeGarching. By this line, the connectionto Garching and Munich city is provided7 days a week until the late eveninghours, which is a big advantage espe-cially for our scientific users workinglate in the evening and during the week-end.

Figure 12.1: Cuban delegation in the neu-tron guide hall. Prof. Fidel Castro(in front) explaining his son (student inMadrid) where the neutrons are.

The still unchanged tremendous re-quest to get informations about theFRM II motivated us to increase ourportfolio of informations at the day ofopen doors. Besides our guided toursthrough the reactor building where 489

Figure 12.2: Great interest on the exhibition of the radiation protection department.

Figure 12.3: Exhibition of the FRM II in the central tent.

visitors got the opportunity to see theinstruments working, 2 talks in the lec-ture hall were given by Prof. Win-fried Petry. The large lecture hall inthe physics department in addition wasused to show two recent films about

96 12 Public relations and visitor service

Figure 12.4: Lively discussions at the reactor pool to explain the unique possibilities of neu-tron beams.

the science of the FRM II. The oneproduced by the Bayerischen Rundfunkgives an overview about the instituteand scientific applications, the otherproduced by a film team from colognewas broadcasted in the TV station Voxin the channel Auto Motor Sport andshowed how radiography with neutronsgive an insight in to car engines.

Lively discussions at the exposition ofnatural radioactivity and instruments tomeasure radiation of our division of ra-diation protection encourage us to con-tinue this dissemination of informationfor the general public.

The FRM II participated as well witha stand in the exhibition tent in the cen-tral area of the campus. Here more gen-eral information about the FRM II wasgiven in numerous discussions. Scien-tists as well as people from the generalpublic were informed about the use ofneutrons for science and industry.

During the summer 2006 Germanywas taken by the overwhelming event ofthe football world championship. Onthe occasion of the opening game inMunich, the Bavarian State Chancelleryinvited CEO’s from industry and bank-ing, which took the opportunity to visitthe FRM II. Accompanied by the Bavar-ian Minister of Economic Affairs, Infras-tructure, Transport and Technology, Er-win Huber, the visitors were welcomedby our President Prof. Dr. Dr. WolfgangHermann, who gave an introduction to

Figure 12.5: Science meets politics in Brussels.

Figure 12.6: Science meets politics in Garching.

the Technische Universität München.Subsequently Dr. Jürgen Neuhaus gavein a short presentation an overviewabout the application of neutrons in sci-ence and industry atthe FRM II. The fol-lowing guided tour through the reactorbuilding provided more insight to theoperation of the reactor and its utiliza-tion.

On the invitation of the BavarianState Minister of Science, Research andArts, Dr. Thomas Goppel, members ofthe European Parliament came to a pre-sentation of the FRM II at the BavarianEmbassy in Brussels on 26th June 2006.Presentations were given from site ofpoliticians by Dr. Beatrix Vierkorn-Rudolph (Federal Ministry of Educationand research) and Dr. Thomas Goppel.

It was pointed out, how the FRM II is


integrated in the European research in-frastructure and how the already estab-lished international use of the FRM II inthe order of 40% can still be increased.

The FRM II itself was presented byProf. Winfried Petry, showing possi-ble applications of neutrons for basicscience as well as for applied research.In further presentations Prof. DieterRichter (FZ-Jülich) showed how neu-tron scattering can contribute to soft-matter research in a large variety of to-days products. The medical applicationof fast neutrons was demonstrated byProf. Michael Molls, Klinikum Rechts

der Isar, who showed at level particlebeams and especially neutrons are fa-vorable for the radiation treatment ofcancers.

One highlight during the year con-cerning the visits of policy-makers atthe FRM II has clearly been the visit ofthe Minister president of North Rhine-Westphalia and nearly his entire cabi-net on the 25th September 2006. On in-vitation of Minister Dr. Thomas Gop-pel and on the occasion of the commonBavarian North Rhine-Westphalia cabi-net meeting in Munich, the FRM II gotthe opportunity to be presented as the

leading national neutron source in Ger-many. After the welcome of our Presi-dent Prof. Dr. Dr. Wolfgang Herrmannand a short introduction of Prof. Win-fried Petry the guest could take own im-pressions on the reactor installation it-self as well on the broad range of appli-cations for neutrons. Special emphasiswas put to explain, how the TechnischeUniversität München operates a largescale facility and how scientific groupsfrom all over Germany participate onthe use of the FRM II by building andoperating scattering instruments.

98 13 People

13 People

13.1 Committees

Strategierat FRM II

Chairman

Prof. Dr. Gernot HegerInstitut für KristallographieRWTH Aachen

Members

MRin Dr. Ulrike KirsteBayerisches Staatsministerium für Wissenschaft,Forschung und Kunst

Dr. Rainer KoepkeBundesministerium für Bildung und Forschung

Prof. Dr. Georg BüldtInstitut für Biologische InformationsverarbeitungForschungszentrum Jülich

Prof. Dr. DoschMax-Planck-Institut für MetallforschungStuttgart

Prof. Dr. Dieter RichterInstitut für FestkörperphysikForschungszentrum Jülich

Prof. Dr. Dirk SchwalmMax-Planck-Institut für KernphysikHeidelberg

Prof. Dr. Helmut SchwarzInstitute of ChemistryTechnische Universität Berlin

Prof. Dr. Dr. Michael WannenmacherRadiologische Klinik und PoliklinikAbteilung StrahlentherapieUniversität Heidelberg

Prof. Dr. Ewald WernerLehrstuhl für Werkstoffkunde und -mechanikTechnische Universität München

Prof. Dr.-Ing. Heinz VoggenreiterDirector of the Institute of Structure and DesignGerman Aerospace Center (DLR) Köln

Prof. Dr. Götz EckoldInstitute of Physical ChemistryUniversität Göttingen (until 22 September 06,succeeded by Prof. Braden)

Prof. Dr. Markus BradenPhysikalisches InstitutUniversität zu Köln

Honorary Members

MDgt i.R. Jürgen Großkreutz Prof. Dr. Tasso Springer

Guests

Prof. Dr. Dr. h.c. mult. Wolfgang A. HerrmannPräsident

Technische Universität München

13 People 99

Dr.-Ing. Rainer KuchBeauftragter der HochschulleitungTechnische Universität München

Dr. Michael KlimkeReferent der HochschulleitungTechnische Universität München

Prof. Dr. Winfried PetryZWE FRM IITechnische Universität München

Dr. Ingo NeuhausZWE FRM IITechnische Universität München

Guido EngelkeZWE FRM IITechnische Universität München

Secretary

Dr. Jürgen NeuhausZWE FRM II

Instrumentation advisory board(Subcommittee of the Strategierat)

Chairman


Members

Prof. Dr. Dirk DubbersPhysikalisches InstitutUniversität Heidelberg

Prof. Dr. Michael GradzielskiInstitut für ChemieTechnische Universität Berlin

Prof. Dr. Rainer HockLehrstuhl für Kristallographie und StrukturphysikUniversität Erlangen

Prof. Dr. Werner KuhsGZG Abteilung KristallographieUniversität Göttingen

Prof. Dr. Stephan PaulPhysik Department E18Technische Universität München

Prof. Dr. Wolfgang SchererLehrstuhl für Chemische PhysikUniversität Augsburg

Prof. Dr. Wolfgang SchmahlDept. für Geo- und UmweltwissenschaftenLudwig-Maximilians-Universität MünchenDeputy of the Chairman

Dr. habil. Dieter SchwahnInstitut für FestkörperforschungForschungszentrum Jülich

Prof. Dr. Andreas TürlerInstitut für RadiochemieTechnische Universität München

Prof. Dr. Albrecht WiedenmannAbteilung SF3Hahn-Meitner-Institut Berlin

Dr. habil. Regine WillumeitGKSS Forschungszentrum Geesthacht

100 13 People

Guests


Dr. Klaus FeldmannBEO-PFRForschungszentrum Jülich

MRin Dr. Ulrike KirsteBayerisches Staatsministerium für Wissenschaft,Forschung und Kunst

Prof. Dr. Gernot HegerInstitut für KristallographieRWTH Aachen

Dr. Jürgen NeuhausZWE FRM IITechnische Universität München




Secretary

Dr. Peter LinkZWE FRM II

Committee for industrial and medical use(Subcommittee of the Strategierat)

ChairmanProf. Dr.-Ing. Heinz VoggenreiterGerman Aerospace Center (DLR) Köln

Members

Automobile industryDr.-Ing. Rainer SimonBMW AG München

Dr.-Ing. Maik BrodaFord Forschungszentrum Aachen

Aerospace industryDr.-Ing. Rainer RauhAirbus Deutschland Bremen

Chemistry and environmentDr. Jens RiegerBASF AG Ludwigshafen

SecretaryDr. Ralph GillesZWE FRM II

13 People 101

Scientific committee - Evaluation of beamtime proposals(Subcommittee of the Strategierat)

ChairmanProf. Dr. Wolfgang SchererLehrstuhl für Chemische PhysikUniversität Augsburg

Members

Prof. Dr. John BanhartAbteilung Werkstoffe (SF3)Hahn-Meitner-Institut Berlin

Dr. Philippe BourgesLaboratoire Léon BrillouinCEA Saclay


Prof. Dr. Heinz-Günther BrokmeierInstitut für WerkstoffforschungGKSS - Forschungszentrum Geesthacht

Prof. Dr. Thomas BrückelInstitut für FestkörperforschungFZ-Jülich

PD Dr. Mechthild EnderleInstitut Laue LangevinGrenoble

Dr. Hermann HeumannMax-Planck-Institut für BiochemieMartinsried bei München

Prof. Dr. Jan JolieInstitute of Nuclear PhysicsUniversität zu Köln

Prof. Dr. Bernhard KeimerMax-Planck-Institut für FestkörperforschungStuttgart

Prof. Dr. Andreas MagerlLS für Kristallographie und StrukturphysikUniversität Erlangen

Prof. Dr. Karl MaierHelmholtz-Institut für Strahlen- und KernphysikUniversität Bonn

Dr. Joël MesotETH Zürich undPaul-Scherrer-Institut Villigen, Schweiz

Prof. Reinhard Dr. Krause-RehbergDepartment of PhysicsUniversität Halle

Dr. Stéphane LongevilleLaboratoire Léon BrillouinLaboratoire de la Diffusion NeutroniqueCEA Saclay

Prof. Dr. Andreas MeyerInstitut für Materialphysik im WeltraumDeutsches Zentrum für Luft- und Raumfahrt (DLR)Köln

Dr. Michael MonkenbuschInstitut für FestkörperforschungForschungszentrum Jülich

Prof. Dr. Werner PaulusStructures et Propriétés de la MatièreUniversité de Rennes 1

Prof. Dr.-Ing. Anke PyzallaMax-Planck-Institut für EisenforschungDüsseldorf

Prof. Dr. Joachim RädlerDepartment für PhysikLudwig-Maximilians-Universität München

Prof. Dr. Günther RothInstitut für KristallographieRWTH Aachen

Prof. Dr. Michael RuckInstitut für Anorganische ChemieTechnische Universität Dresden

Prof. Dr. Wolfgang SchmahlDep. für Geo- und UmweltwissenschaftenLudwig-Maximilians-Universität München

Prof. Dr. Bernd StühnInstitut für FestkörperphysikTechnische Universität Darmstadt

Prof. Dr. Monika Willert-PoradaLehrstuhl für WerkstoffverarbeitungUniversität Bayreuth

102 13 People

Scientific secretaries

Dr. Jürgen NeuhausZWE FRM II

Dr. Christoph HugenschmidtZWE FRM II

Dr. Peter LinkZWE FRM II

Dr. Martin MevenZWE FRM II

Dr. Tobias UnruhZWE FRM II

TUM Advisory board

Chairman

Prof. Dr. Ewald WernerLehrstuhl für Werkstoffkunde und -mechanikTechnische Universität München

Members

Prof. Dr. Peter BöniPhysik Department E21Technische Universität München

Prof. Dr. Andreas TürlerInstitut für RadiochemieTechnische Universität München

Prof. Dr. Markus Schwaigerrepresented by Prof. Dr. Senekowitsch-SchmidtkeNuklearmedizinische Klinik und PoliklinikKlinikum Rechts der IsarTechnische Universität München

Prof. Dr. Bernhard WolfHeinz Nixdorf-Lehrstuhl für medizinische ElektronikTechnische Universität München

Prof. Dr. Arne SkerraLehrstuhl für Biologische ChemieTechnische Universität München

Guests





13 People 103

Scientific steering committee

Chairman


Members

Dr. Hans BoysenSektion KristallographieLudwig-Maximilians-Universität München

Prof. Dr. Bernhard KeimerMax-Planck-Institut für Festkörperforschung Stuttgart

Prof. Dr. Dieter RichterInstitut für FestkörperforschungForschungszentrum Jülich

104 13 People

13.2 Staff

Board of directors

Scientific directorProf. Dr. W. Petry

Technical directorDr. Ingo Neuhaus

Administrative directorG. Engelke

Experiments

HeadProf. Dr. W. Petry

SecretariesW. WittowetzE. Jörg-Müller

Secretary JCNSSt. Mintmans

CoordinationDr. J. NeuhausH. TürckH. Bamberger

InstrumentsN. Arend (JCNS)P. Aynajian (MPI-Stuttgart)Dr. S. Bayrakci (MPI-Stuttgart]Prof. Dr. P. Böni (E21)J. BrunnerK. Buchner (MPI-Stuttgart)Dr. T. BücherlP. Busch (JCNS)M. Bröll (MPI-Stuttgart)W. Bünten (JCNS)E. CalzadaD. EtzdorfJ. Franke (MPI-Stuttgart)Dr. H. Frielinghaus (JCNS)Dr. U. Garbe (GKSS)Dr. A.-M. Gaspar (Visiting scientistfrom PortugalDr. R. GeorgiiDr. R. GillesDr. M.Haese-Seiller (GKSS)Dr. W. HäußlerF. HibschDr. O. Holderer(JCNS)Dr. M. HofmannDr. M. Hölzel (TU Darmstadt)Dr. K. Hradil (Univ. Göttingen)Dr. C. HugenschmidtDr. V. Hutanu (RWTH Aachen)Dr. A. Ioffe (JCNS)S. KampferR. Kampmann (GKSS)Dr. T. Keller (MPI-Stuttgart)P. Kudejova (Universität zu Köln)Dr. V. Kudryaschov(GKSS)

B. KrimmerD. LamagoDr. P. LinkDr. B. Loeper-KabasakalC. LoistlK. LorenzA. Mantwill(E21)J. Major (MPI-Stuttgart)M. Major (MPI-Stuttgart)Dr. St. Mattauch (JCNS)F. Maye (MPI-Stuttgart)T. MehaddeneDr. M. MevenDr. R. MoleS. MühlbauerQi Ning (Visiting scientist from China)M. Nülle (MPI-Stuttgart)Dr. A. OstermannDr. B. PedersenC. PiochaczDr. J. Rebelo-Kornmeier (HMI Berlin)Dr. A. Radulescu (JCNS)J. RepperR. RepperDr. D. RichJ. RingeDr. P. Rottländer (JCNS)Dr. A. Rühm (MPI Stuttgart)A. SazonovDr. B. SchillingerH. Schneider (Univ. Göttingen)Dr. A. Schneidewind (TU Dresden)Prof. Dr. K. Schreckenbach (E21)M. SchulzR. SchwikowskiDr. A. Senyshyn (TU Darmstadt)G. SeidlC. SmudaM. StadlbauerB. StraßerP. Stronciwilk (JCNS)Dr. Y. Su (JCNS)Dr. T. UnruhF. M. WagnerDr. R. Wimpory (HMI Berlin)Dr. H.-F. WirthDr. Joachim Wuttke (JCNS)Prof. Dr. O. Zimmer (E18)

Detectors and electronicsDr. K. ZeitelhackDr. I. DefendiS. EgerlandChr. HesseDr. A. KastenmüllerM. PanradlTh. Schöffel

Sample environmentDr. J. PetersP. BiberH. KolbA. PscheidtA. SchmidtJ. Wenzlaff

Neutron opticsProf. Dr. G. BorchertC. BreunigH. HofmannE. KahleO. LykhvarDr. S. MasalovichA. OfnerA. UrbanR. Valicu

IT servicesJ. KrügerH. WenningerL. BornemannJ. ErtlJ. DettbarnS. GalinskiH. GildeF. HänselJ.-P. InnocenteR. MüllerA. PrellerJ. PulzC. RajaczakS. RothA. SchwertnerM. StowasserB. Wildmoser

ConstructionK. Lichtenstein

13 People 105

Administration

HeadG. Engelke

SecretaryC. Zeller

MembersR. ObermeierB. BendakB. GallenbergerI. HeinathK. Lüttig

Public relationsI. Scholz (ZV TUM)A. Schaumlöffel(ZV TUM)

Visitor’s serviceU. KurzDr. B. Tonin-Schebesta

Reactor operation

HeadDr. Ingo Neuhaus

SecretariesM. NeubergerS. Rubsch

ManagementDr. H. Gerstenberg(Irradiation and fuel cycles)Dr. J. Meier(Reactor operation)R. Schätzlein(Electric and control technology)

Shift membersF. GründerA. BancsovA. BenkeM. DannerChr. FeilM. FlieherH. GroßL. HerdamF. HofstetterK. HöglauerT. KalkG. KalteneggerU. KappenbergF. KewitzM. G. KrümpelmannJ. KundA. LochingerG. MauermannA. MeilingerM. MoserL. RottenkolberG. Schlittenbauer

Technical servicesK. PfaffR. BinschH. Gampfer

W. GlashauserG. GuldB. HeckG. WagnerA. WeberM. WöhnerC. Ziller

SourcesC. MüllerD. PätheA. Wirtz

Electric and control technologyR. SchätzleinG. AignerW. BuchnerR. KrammerK.-H. MayrÜ. SarikayaH. SchwaighoferJ. Wildgruber

IrradiationDr. H. GerstenbergJ.-M. FavoliDr. X. LiM. OberndorferW. LangeV. LoderA. RichterH. SchulzF.-M. WagnerN. Wiegner

Reactor enhancementM. SchmittB. StruthV. ZillR. Lorenz

Technical designF.-L. Tralmer

J. FinkH. FußstetterJ. JüttnerG. LangenstückK. Lichtenstein

WorkshopsC. HerzogU. StiegelA. BegicM. FußA. HuberA. ScharlR. Schlecht jun.

Radiation protectionDr. H. ZeisingS. DambeckW. DollrießH. HottmannD. KuglerB. NeugebauerD. SchrulleH.-J. WerthS. WolffD. BahmetW. KlugeA. SchindlerD. Strobl

Chemical laboratoryU. JaserC. AuerR. BertschS. Kiermaier

Technical safety serviceJ. WetzlR. MaierJ. AignerK. OttoN. HodzicJ. Schreiner

106 13 People

Reactor physics

Dr. A. RöhrmoserC. Bogenberger

R. JungwirthW. Schmid

N. Wieschalla

Security department

L. Stienen J. Stephani

13.3 Partner institutions

GKSS Research Centre GmbHMax-Planck-Straße 121502 GeesthachtGermanyhttp://www.gkss.de/index_e.html

Hahn-Meitner-Institute GmbH (HMI)Glienickerstraße 10014109 BerlinGermanyhttp://www.hmi.de/index_en.html

Jülich Centre for Neutron Science JCNSResearch Centre Jülich GmbH52425 Jülich, GermanyOutstation at FRM II: 85747 Garchinghttp://www.jcns.info

Max-Planck-Institut für FestkörperphysikHeisenbergstraße 170569 StuttgartGermanywww.fkf.mpg.de/main.html

Max-Planck-Institut für MetallforschungHeisenbergstraße 370569 StuttgartGermanyhttp://www.mf.mpg.de/de/index.html

RWTH AachenInstitute of CrystallographyJägerstraße 17 - 1952056 AachenGermanyhttp://www.xtal.rwth-aachen.de/index_e.html

Technische Universität DarmstadtMaterial- und GeowissenschaftenPetersenstraße 2364287 DarmstadtGermanyhttp://www.tu-darmstadt.de/fb/matgeo/

Technische Universität DresdenInstitut für Festkörperphysik01062 DresdenGermanyhttp://www.physik.tu-dresden.de/ifp/ifp.php

Universität AugsburgInstitut für PhysikLehrstuhl für Chemische Physik und Materialwissenschaften86135 AugsburgGermanyhttp://physik.uni-augsburg.de/exp3/home.html

Universität der Bundeswehr MünchenInstitut Angewandte Physik und MesstechnikWerner-Heisenberg-Weg 3985579 NeubibergGermanyhttp://www.unibw.de/lrt2/

Universität GöttingenInstitut für Physikalische ChemieTammannstraße 637077 GöttingenGermanyhttp://www.uni-pc.gwdg.de/eckold/home.html

Universität zu KölnInstitut für KernphysikZülpicherstraße 7750937 KölnGermanyhttp://www.ipk.uni-koeln.de/

Ludwig-Maximilians-Universität MünchenSektion Kristallographie (Prof. Schmahl)and Sektion Physik (Prof. Rädler)Geschwister-Scholl-Platz 180539 MunichGermanyhttp://www.uni-muenchen.de

http://www.gkss.de/index_e.html

http://www.hmi.de/index_en.html

http://www.jcns.info

www.fkf.mpg.de/main.html

http://www.mf.mpg.de/de/index.html

http://www.xtal.rwth-aachen.de/index_e.html

http://www.tu-darmstadt.de/fb/matgeo/

http://www.physik.tu-dresden.de/ifp/ifp.php

http://physik.uni-augsburg.de/exp3/home.html

http://www.unibw.de/lrt2/

http://www.uni-pc.gwdg.de/eckold/home.html

http://www.ipk.uni-koeln.de/

http://www.uni-muenchen.de

14 Figures 107

14 Figures

14.1 Publications

[1] Appavou M.-S., Busch S., Doster W., Unruh T. The Effect of Packing in Internal Molecular Motions of Hydrated Myoglobin.MRS (Materials Research Society) Bulletin, in press, 6 pgs.

[2] Babcock E., Petoukhov A., Andersen K., Chastagnier J., Jullien D., Lelièvre-Berna E., Georgii R., Masalovich S., Boag S.,Frost C., Parnell S. AFP flipper devices: Polarized 3He spin flipper and shorter wavelength neutron flipper; Proceedings ofPCMI 2006, Berlin, Germany. Physica B, submitted, 3 pgs.

[3] Bayrakci S., Keller T., Habicht K., Keimer B. Spin wave lifetimes throughout the Brillouin zone. Science, 312, (2006), 1926– 1929. www.sciencemag.org/cgi/content/full/312/5782/1926/DC1.

[4] Blackburn E., Hiess A., Bernhoeft N., Rheinstädter M., Häussler W., Lander G. Fermi Surface Topology and the Supercon-ductivity Gap Function in UPd2Al3: A Neutron Spin Echo Study. Phys. Rev. Lett., 97 (057002), (2006), 057002–1–4.

[5] Bunjes H., Unruh T. Characterization of Lipid Nanoparticles by Differential Scanning Calorimetry, X- Ray and NeutronScattering. Advanced Review in Pharmaceutical Science, submitted.

[6] Busch S., Doster W., Longeville S., Sakai V., Unruh T. Microscopic Protein Diffusion at High Concentration - QENS Pro-ceeding. MRS Bulletin, in press, 8 pgs.

[7] Cadogan J., Moze O., Ryan D., Suharyana, Hofmann M. Magnetic ordering in DyFe6SN6. Physica B, 385 - 386, (2006),317 – 319.

[8] Gilles R., Hölzel M., Schlapp M., Elf F., Krimmer B., Boysen H., Fuess H. First test measurements at the new structurepowder diffractometer (SPODI) at the FRM II. Z. Kristallogr., 23, (2006), 183–188.

[9] Gilles R., Mukherji D., Hoelzel M., Strunz P., Toebbens D., Barbier B. Neutron and X-ray diffraction measurements onmicro- and nano-sized precipitates embedded in a Ni-based superalloy and after their extraction form the alloy. Actamaterialia, 54 (5), (2006), 1307–1316.

[10] Gilles R., Ostermann A., Petry W. Monte Carlo simulations of the new Small-Angle Neutron Scattering instrument SANS-1at the Heinz Maier-Leibnitz Forschungsquelle. J. Appl. Cryst., in press.

[11] Gilles R., Ostermann A., Schanzer C., Krimmer B., Petry W. The concept of the new small-angle scattering instrumentSANS-1 at the FRM -II; Proceedings of ICNS 2005. Physica B, 385-386, (2006), 1174.

[12] Hoelzel M., Del Genovese D., Gilles R., Mukherji D., Toebbens D., Roessler J., Fuess H. Phase analysis and lattice mis-matches in superalloys DT706 and Iconel 706, Proceedings of ICNS 2005. Physica B, 385-386, (2006), 594.

[13] Hofmann M., Campell S., Link P., Fiddy S., Goncharenko. Valence and Magnetic Transitions in YbMn2Ge2 - Pressure andTemperature. Physica B, 385 - 386, (2006), 330 – 332.

[14] Hofmann M., Schneider R., Seidl G., Kornmeier J., Wimpory R., Garbe U., Brokmeier H. The New Materials ScienceDiffractometer STRESS-SPEC at FRM-II. Physica B, 385 - 386, (2006), 1035 – 1037.

[15] Hugenschmidt C., Schreckenbach K., Stadlbauer M., Straßer B. First positron experiments at NEPOMUC. Appl. Surf. Sci.,252, (2006), 3098–3105.

[16] Häußler W., Gohla-Neudecker B., Schwikowski R., Streibl D., Böni P. RESEDA - the new Resonance Spin Echo Spectrometerusing cold neutrons at the FRM II. Physica B, submitted, 3 pgs.

[17] Kampfer S., Wagner F.-M., Loeper B., Kneschaurek P. Erste dosimetrische Ergebnisse an der neuen Neutronentherapiean-lage am FRM II. Medizinische Physik ; Tagungsband der 37. Jahrestagung der Deutschen Gesellschaft für MedizinischePhysik, 20. - 23. 09.2006 in Regensburg, herausgegeben von Ludwig Bogner and Barbara Dobler, 2 pgs.

www.sciencemag.org/cgi/content/full/312/5782/1926/DC1

108 14 Figures

[18] Keller T., Aynajian P., Habicht K., Boeri L., Bose K., Keimer B. Momentum-resolved electronic-phonon interaction in leaddetermined by neutron resonance spin-echo spectroscopy. Phys. Rev. Lett., 96, (2006), 225501–1–4.

[19] Kögel G., Dollinger G. Planned positron experiments at FRM-II. Applied Surface Science, 252, (2006), 3111–3120.

[20] Lamago D., Georgii R., Pfleiderer C., Böni P. Magnetic-field induced instability surrounding the A-phase of MnSi: Bulkand SANS measurements. Physica B, 385 - 386, (2006), 385–387.

[21] Li H., Schillinger B., Calzada E., Yinong L., Mühlbauer M. An adaptive algorithm for gamma spots removal in CCD-basedneutron radiography and tomography. Nuclear Instruments and Methods in Physics Research A, 564, (2006), 405–413.

[22] Lin X., Henkelmann R., Türler A., Gerstenberg H., De Corte F. Neutron flux parameters at irradiation positions in the newresearch reactor FRM-II. Nuclear Instruments and Methods in Physics Research A, 564, (2006), 641–644.

[23] Magaddino V. The dependence ofthe relative biological effectiveness (RBE) of fission neutrons on dose and on gamma raycontamination in human SCC megacolonies. Master Thesis at University College London (European Master of Science inRadiation Biology).

[24] Maye F. Aufbau und Test einer Neutronen-Röntgen-Kontrast-Reflektometrie-Anordnung. Diploma Thesis at the Max-Planck-Institut für Metallforschung, Stuttgart, und Institut für Theoretische und Angewandte Physik der UniversitätStuttgart.

[25] Mehaddene T., Neuhaus J., Petry W., Hradil K. Phonon dispersions in NiAlMn shape memory alloy. Material Science &Engineering A, submitted.

[26] Mühlbauer Q., Hradil K. Monitoring and Preventing collision for a triple axis Spectrometer. Proceedings of The IEEEInternational Conference on Central Applications, October 4 - 6, 2006, Munich, Germany, 6 pgs.

[27] Palancher H., Martin P., Sabathier C., Dubois S., Valot C., Wieschalla N., Röhrmoser A., Petry W., Jarousse C., Grasse M.,Tucoulou R. Heavy Ion Irradiation as a Method to Discriminate Research Reactor Fuels. Proceedings on the ’InternationalConference on Research Reactor Fuel Management RRFM 2006’, May 04, 2006, Sofia.

[28] Pigozzi G., Mukherji D., Gilles R., Barbier B., Kostorz G. Ni3Si(Al)/a-SiOx core shell nanoparticles: characteriziation, shellformation, and stability. Nanotechnology, 17, (2006), 4195 – 4203.

[29] Pühlhofer T., Baier H., Krämer L., Unruh T. Design, manufacturing and testing of high speed rotating cfrp chopper discs.SAMPE EUROPE International Conference 2006.

[30] Rühm A., Wildgruber U., Franke J., Major J., Dosch H. n/X materials science reflectometer at FRM-II in Garching. NeutronReflectometry, a Probe for Materials Surfaces; Proceedings of a Technical Meeting organized by the International AtomicEnergy Agency and held in Vienna, 16-20 August 2004, Austria,, 161–175.

[31] Schillinger B., Brunner J., Calzada E. A study of oil lubrication in a rotating engine using stroboscopic neutron imaging.Physica B (6th International Conference on Neutron Scattering Sydney, November 2005), 385- 386, (2006), 921 – 923.

[32] Schneidewind A., Link P., Etzdorf D., Schedler R., Rotter M., Loewenhaupt M. PANDA – first results from the cold threeaxes-spectrometer at FRM-II, Proceedings of ICNS 2005. Physica B, 385 - 386, (2006), 1089–1091.

[33] Senff D., Link P., Hradil K., Hiess A., Regnault L., Sidis Y., Aliouan N., Argyriou D., Braden M. Magnetic excitations inmultiferroic TbMnO3. Phys. Rev. Lett., submitted.

[34] Sergueev I., van Bürck U., Chumakov A., Asthalter T., Smirnov G., Franz H., Rüffer R., Petry W. Synchrotron-radiation-based perturbed angular correlations used in the investigation of rotational dynamics in soft matter. Physical Review B,3, (2006), 024203–1–12.

[35] Smirnov G., van Bürck U., Franz A., Asthalter T., Leupold O., Schreier E., Petry W. Nuclear gamma resonance time-domain interferometry: Quantum beat and radiative coupling regimes compared in revealing quasielastic scattering.Physical Review B, 73, (2006), 184126–1–9.

[36] Stadlbauer M., Hugenschmidt C., Schreckenbach K., Straßer B. Spatially Resolved Investigation of Thermally TreatedBrass with a Coincident Doppler-Spectrometer. Appl. Surf. Sci., 252, (2006), 3269–3273.

[37] Stampanoni M., Borchert G., Abela R. Progress in Microtomography with the Bragg Magnifier at SLS. Radiation Physicsand Chemistry, in press.

14 Figures 109

[38] Stampanoni M., Groso A., Borchert G., Abela R. Bragg Magnifier: High Efficiency, High Resolution X-Ray Detector. Proc.Syn. Rad. Instr., in press.

[39] Strunz P., Mukherji D., Näth O., Gilles R., Röser J. Characterization of nanoporous superalloy by SANS; Proceedings ofICNS 2005. Physica B, 385-386, (2006), 626.

[40] Unruh T., Smuda C., Gemmecker g., Bunjes H. Molecular dynamics in pharmaceutical drug delivery systems - the poten-tial of qens and first experimental results. Mat. Res. Soc., in press.

[41] Wagner F., Bücherl T., Kampfer S., Kastenmüller A., Waschkowski W. Thermal Neutron Converter for Irradiations withFission Neutrons. Conference Proceedings: Current Problems in Nuclear Physics and Atomic Energy (NPAE-Kyiv2066),May 29 - June 03, 2006, Kyiv, Ukraine, in press, 9 pgs.

[42] Wieschalla N., Bergmaier A., Böni P., Böning K., Dollinger G., Großmann R., Petry W., Röhrmoser A., Schneider J. Heavyion irradiation of U-Mo/Al dispersion fuel. Journal of Nuclear Materials, 357, 1-3, (2006), 191–197.

[43] Zeitelhack K., Schanzer C., Kastenmüller A., Röhrmoser, Daniel C., Franke J., Gutsmiedl E., Kudryashov V., Maier D.,Päthe D., Petry W., Schöffel T., Schreckenbach K., Urban A., Wildgruber U. Measurement of neutron flux and beamdivergence at the cold neutron guide system of the new Munich research reactor FRM-II. Nuclear Instruments and Methodsin Physics Research A, 560, (2006), 444–453.

[44] Boysen H., Lerch M., Stys A., Senyshyn A., Hoelzel M. Oxygen mobility in the ionic conductor maynite (C a12 Al14O33: ahigh temperature neutron diffraction study. Acta Cryst. B., submitted.

[45] Georgii R., Böni P., Janoschek M., Schanzer C., Valloppilly S. A flexible instrument for VCN. Physica B, in press.

[46] Grigoriev S., Maleyev S., Okorokov A., Chetverikov Y., Böni P., Georgii R., Lamago D., Eckerlebe H., Pranzas K. Magneticstructure of MnSi under an applied field probed by polarized small-angle neutron scattering. Phys. Rev. B, 74 (21), (2006),214414–1–10.

[47] Hofmann M., Seidl G., Rebelo-Kornmeier J., Garbe U., Schneider R., Wimpory R., Wasmuth U., Noster U. The NewMaterials Science Diffraktometer STRESS-SPEC at FRM II. Materials Science Forum Vols., 524 - 525, (2006), 211 – 216.

[48] Hugenschmidt C., Legl S. A Novel Time-of-Flight Spectrometer for the Analysis of Positron Annihilation Induced AugerElectrons. Rev. Sci. Instr., 77, (2006), 103904–1–6.

[49] Hutanu V., Meven M., Heger G. Construction of the New Hot Neutrons Single Crystal Diffractometer POLI-HEiDi at FRM-II (Proceedings of PNCMI 2006, Berlin, Germany). Physica B, in press.

[50] Lerch M., Boysen H., Rödel T., Kaiser-Bischoff I., Hoelzel M., Senyshyn A. High temperature neutron diffraction study ofscandia/nitrogen co-doped zirconia. J. Sol. State Chem., submitted.

[51] Park H.-S., Hoelzel M., Boysen H., Schmidbauer E. Lithium conductivity in a Li-bearing ring silicate mineral, sogdianite.J. Sol. Chem., in press.

[52] Pedersen B., Frey F., Scherer W., Gille P., Meisterernst G. The new single crystal diffractometer RESI at FRM-II. Physica B,385 - 386, (2006), 1046–1048.

[53] Peters L., Knorr K., Evans J., Senyshysn A., Rahmoun H.-S., Depmeier W. Proton positions in and thermal behaviourof the phase 4C aO ∗ 3Al2O3 ∗ 3H2O and its thermal decomposition to (OC a4)2[Al12O24, determined by neutron/X-raypowder diffraction and IR Spectroscopic investigations. Z. Kristallographie, in press.

[54] Ranjan R., Agrawal A., Senyshyn A., Boysen H. Phases in the system N a1/2N d1/2 −Sr T iO3: a powder neutron diffractionstudy. J. Phys.: Condensed Matter, 18, (2006), 9679 ff.

[55] Ranjan R., Agrawal A., Senyshyn A., Boysen H. Crystal structures of high temperature quantum paraelectricsN a1/2N d1/2T iO3 and N a1/2Pr1/2T iO3. J. Phys.: Condensed Matter, 18, (2006), L515–L522.

[56] Ranjan R., Senyshyn A., Frey F., Boysen H. Crystal structure of N a1/2Ln1/2T iO3 (Ln : La,Eu,T b). J. Sol. State Chem., inpress.

[57] Röhrmoser A., Petry W. Reduced enrichment program for FRM II, actual status & principal study of monolithic fuel forFRM II; Proceedings on the 10th International Topical Meeting on Reserach Fuel Management, RRFM 2006, Sofia. 2006,submitted.

110 14 Figures

[58] Yeom J., Defendi I., Takahashi H., Zeitelhack K., Nakazawa M., Murayama H. 1 12- Channel CMOS Preamplifier-Shaper-Discriminator ASIC for APD and Gas Counters. IEEE Transactions on Nuclear Science, 53, No. 4, (2006), 5 pgs.

Imprint

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Publisher:Technische Universität MünchenForschungs-Neutronenquelle Heinz Maier-Leibnitz (FRM II)Lichtenbergstr. 185747 GarchingGermanyPhone: +49 89-289-14966Fax: +49 89-289-14995Internet: http://www.frm2.tum.deemail: mailto://[email protected]

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Editors:J. NeuhausB. PedersenE. Jörg-Müller, TUM

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Photographic credits:All images: TUM, except otherwise noted

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Design:B. Pedersen, TUMJ. Neuhaus, TUME. Jörg-Müller, TUM

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Typesetting(LATEX 2ε):B. Pedersen, TUME. Jörg-Müller, TUM

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

http://www.frm2.tum.de

mailto://[email protected]

Documents

ANNUALREPORT2006 Forschungs-Neutronenquelle HeinzMaier … · 2015. 3. 4. · LAND) joint European project took broad room in our activities. Detectorreadoutelectronics During the