Social Media

A small angle neutron scattering (SANS) experiment using very cold neutrons (VCN)

Description
A small angle neutron scattering (SANS) experiment using very cold neutrons (VCN)
Categories
Published
of 4
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  A small angle neutron scattering (SANS) experiment using very coldneutrons (VCN) M. Bleuel a,  , J.M. Carpenter a , B.J. Micklich a , P. Geltenbort b , K. Mishima c , H.M. Shimizu c , Y. Iwashita d ,K. Hirota e , M. Hino f  , S.J. Kennedy g , J. Lal a a Intense Pulsed Neutron Source, Argonne National Laboratory, Argonne, IL 60439, USA b Institut Laue Langevin, 6 Rue J. Horowitz, Grenoble 38042, France c Neutron Science Laboratory KEK, 1-1 Oho, Tsukuba, Ibaraki 305-0801, Japan d Institute for Chemical Research, Kyoto University, Kyoto 611-0011, Japan e RIKEN, 2-1 Hirosawa, Wako, Aitama 351-0198, Japan f  Research Reactor Institute, Kyoto University, Kumatori, Osaka 590-0494, Japan g Bragg Institute, ANSTO, Lucas Heights, NSW 2234, Australia a r t i c l e i n f o Keywords: Very cold neutronsSmall angle neutron scatteringMagnetic neutron lensPolarizing monochromator super-mirror a b s t r a c t This paper describes the results of SANS measurements of small samples using the very cold neutron(VCN) beam of the PF2 instrument at the Institut Laue Langevin (ILL), France. In addition to a classicalSANS pinhole collimation, the experiment used a polarizing supermirror as a monochromator and amagnetic sextupole lens to focus the neutron beam in order to gain intensity and avoid any material inthe neutron beam besides the sample.Published by Elsevier B.V. 1. Introduction Experiments at X-ray spectrometers can be done with samplemasses of the order of milligrams, whereas neutron experimentstypically require several grams of sample. One way to optimize aneutron instrument towards smaller sample volumes is toperform the experiment with neutrons of a longer wavelength.This requires a reduction of the sample thickness in the beamsince the total scattering cross section increases approximatelylinearly with the neutron wavelength (as does the absorptioncross section) [1].It is difficult to use conventional SANS instruments for high-performance measurement around  q min ¼ 0.001A˚   1 and below.We performeda series of tests as a preliminary demonstration of aVCN SANS suitable for small samples. However, due to the smallsize of the detector area it was necessary to pick a very short flightpath downstream of the sample (about 63mm) to be able tomeasure roughly 0.1A˚   1 , in which region the structural peak of the test samples provided a high neutron scattering intensity. Theminimum  q  for these tests was roughly 0.01A˚   1 . 2. Setup The experiment described in the following uses a magneticlens to focus a VCN beam on the sample aperture in order tomeasure the small angle scattering from a very thin sample (aschematic and a photo of the setup used are shown in Fig.1). Thissection gives an overview of the components and theircharacteristics in the order in which the neutrons interact withthem while travelling downstream towards the detector.  2.1. Polarizing monochromator supermirror (PM-SM) The first component in the neutron path is the polarizingmonochromator supermirror consisting of Fe/SiGe3-double layersseparated by a Si layer for magnetic decoupling between thedouble layers (for more details see Ref. [2]). The incoming neutronbeam hits it under an angle of 18 1  relative to the long axis of theexperimental table (see photo in Fig.1). After careful alignment of the mirror it was possible to extract a 44A˚ neutron beam with a D l / l ffi 17% parallel to the experimental table.Fig. 2 compares the time-of-flight (ToF)-spectrum of the VCNbeam hitting this polarizing monochromator supermirror withone when a regular non-polarizing  m ¼ 3 supermirror is used. Inorder to obtain these spectra a chopper was placed upstream themirror. The chopper was locked in the open position for all otherexperiments reported in this paper. ARTICLE IN PRESS Contents lists available at ScienceDirectjournal homepage: www.elsevier.com/locate/physb Physica B 0921-4526/$-see front matter Published by Elsevier B.V.doi:10.1016/j.physb.2009.06.048  Corresponding author. Tel.: +16302525346; fax: +16302524163. E-mail address:  mbleuel@anl.gov (M. Bleuel).Physica B 404 (2009) 2629–2632  Besides using the reflection of the mirror to align the neutronbeam parallel to the experiment table the use of a polarizingmirror has the advantage that the background of the focusedneutron beam behind the magnetic lens becomes lower (sinceonly one spin state is focused it is possible to filter the other spinstate upstream of the lens). Due to its almost constant d spacingthe mirror acted as a wavelength filter (monochromator). Themonochromating coating was chosen to reduce the wavelengthspread. It also experimentally proves the feasibility to install thiskind of setup in a polychromatic beam.  2.2. Apertures In this experiment cadmium pinholes are used to define thebeam at two places — immediately after the PM-SM and at thesample position. Both apertures consist of a 1mm thick cadmiumsheet with a 4mm diameter pinhole and together they form aclassical SANS pinhole collimator system.  2.3. Adiabatic gradient spin flipper  The spin flipper used in this setup is identical to the one usedin tests on the pulsed Asterix beam line at LANSCE [3] and wasused to check the polarization of the setup.  2.4. Sextupole lens Neutron lenses in a SANS geometry can be used either to focusneutrons on the sample in order to increase the intensity of theexperimentat the costof a biggerdirectbeam spoton the detectoror to focus on the detector in order to decrease the minimum  q  of the setup. For both cases gain factors between 10 and 100 havebeen reported in literature [4,5]. In the present setup thesextupole magnetic lens was used to focus the neutrons on thesample in order to increase the intensity. The lens has beenstudied in detail at the VCN beamline of the ILL instrument PF2[6]. For the wavelength used in this experiment (roughly 44A˚ ) itsfocusing length is  f  ffi 0.4m.  2.5. Samples Two kinds of samples were used in these tests: a 0.3mm thickVycor glass sample (made by Corning) with a transmission of about 64% which has a structure peak at roughly 0.023A˚   1 [7] anda liquid 1mm thick sample containing 1M sodium dodecylsulphate (SDS) micelles in D 2 O with a structure peak at about0.1A˚   1 [8]. The transmission of the micelles sample in a siliconglass holder was roughly 20% while the empty holder had atransmission above 98%. Both samples are known to be strongneutron scatterers.  2.6. Detector  The detector used in this experiment was a Li-dopedscintillator combined with a position sensitive photo-multiplier-tube (PMT) [9]. The active area of the detector was about 90mmin diameter and the pixel resolution roughly 0.8mm. In order tobe able to measure momentum transfers up to 0.1A˚   1 with suchlong wavelength neutrons it was necessary to put the detectornear the sample. In the following the sample–detector distancewas 63mm. 3. Results Fig. 3 shows results of the polarization tests of the setup.Switching the spin flipper on reduces the intensity of the directbeam (integrated over a 10  10 pixel square centered on themaximum intensity) by roughly one order of magnitudeindicating that the polarization of the setup was above 80%.The lens is capable of creating an image of the first 4mmpinhole 1.6m downstream with roughly the same diameter. Thisindicates that the focusing is working properly. However, the ARTICLE IN PRESS Fig. 1.  (Color online) Setup used for the tests. The distance between the supermirror and the detector was about 2m; both apertures are in the focus points of the lens(about 80cm from the lens for the 44A˚ wavelength neutrons used). 40020000 20 40 60WaveLength [A]    C  o  u  n   t   R  a   t  e MonochroSuperMirror Fig. 2.  (Color online) Neutron spectrum after (black) a regular  m ¼ 2.9pol. SM and(red) a polarizing monochromator SM. M. Bleuel et al. / Physica B 404 (2009) 2629–2632 2630  image is not perfect as a significant part of the non-flipped beamintensity is not focused correctly and shows up on the small angledetector as non-focused (left part of  Fig. 3). In other setups wheremagnetic lenses are used the polarization direction is turned intothe flight direction of the neutrons before entering the lensleading to a more adiabatic transition of the polarization andtherefore reducing the defocussing [4–6].In all experiments the setup consisted of the same componentsdescribed above: the polarizing monochromator mirror, a 4mmentrance aperture, the spin flipper, the magnetic lens at roughly0.8m distance from each aperture, the 4mm sample aperture andthe detector 63mm downstream the sample.Figs. 4 and 5 show the small angle scattering pattern of the twosamples. Each samplewas measured for roughly 48h. Fig. 4 showsthe calibrated 2D-images and Fig. 5 the isotropic averages. Forthese measurements the spin flipper was used and the intensitywas modulated. However, since the data presented here are timeaveraged, the SANS scattering is the same as for unpolarizedunmodulated measurements and the modulation is irrelevant forthese results.It is clearly visible that the samples show distinguishablescattering patterns. The isotropically averaged data yield a peakfor Vycor which is in good agreement with the value reported inliterature (0.023A˚   1 , see for example Ref. [7]). The structuremaximum for the SDS-Micelles at 1M concentration can beexpected at 0.1A˚  1 which is just outside the range of this testsetup.However, the rising shape of the SDS-curve looks reasonable,especially since later experiments revealed problems in thesensitivity for the outer pixels of the detector. Fig. 5 also indicatesthat  q min  is below 0.01A˚   1 for this setup. However, there is asignificant background at lower  q , most likely from scattering of VCN in air. 4. Concluding remarks To our knowledge this is the first ever demonstration of a SANSmeasurement using very cold neutrons. The prototype setup can ARTICLE IN PRESS Fig. 3.  (Color online) 2D image of the raw data intensity at the detector without the sample pinhole; spin flipper turned off (left) and spin flipper (right) turned on. Fig. 4.  (Color online) 2D-SANS pictures of the Vycor (left) and the SDS-micelles sample (right). Fig. 5.  (Color online) Isotropic average of the SANS scattering raw data for Vycor(black circles), SDS-micelles (red triangles) and background (blue diamonds). M. Bleuel et al. / Physica B 404 (2009) 2629–2632  2631  now be used as a test bed for further ideas in instrumentation andneutron optics  all of which work better at long neutron wavelengths .For example quasi-elastic scattering from the SDS micelles wasobserved with the same setup by modulation of the current in theflipper and thus chopping the neutron beam. First promisingresults will be published later. The count rate in the experimentsdescribed above was about 10n/s; however, there are hugepossible gain factors using this wavelength band.The combination of the lens with a focusing collimator or aneutron guide system instead of the single entrance pinhole couldbring gain factors of roughly 100. There is also room to furtheroptimize the instrument towards either resolution or flux, sincewith the lens and the collimator the umbra and penumbra of thebeam can be controlled independently. A more optimized sourceand moderator have the potential to increase the flux at thesample by an additional factor of 20 [10]. And even with themodest setup used in the tests reported in this paper a factor of 4–10 could be gained just by using a broader band of wavelengthsat a slightly shorter mean wavelength (as indicated by the non-monochromating supermirror in Fig. 2) with a polarizing super-mirror with a higher  m -value.A state-of-the-art magnetic guide field system and mountingthe whole beamline in a vacuum container would reducesignificantly the background of the instrument.  Acknowledgments The support of the management and the staff of the IPNS atArgonne National Laboratory and the Institut Laue-Langevin isgrateful acknowledged. T. Brenner and S. Klimko kindly offeredessential support with the setup. The Kristallabor of the TUMunich helped preparing the thin Vycor sample. This work wasfunded by the US Department of Energy, BES-Materials Science,under Contract DE-AC02-06CH11357. References [1] J.R.D. Copley, Neutron News 18 (1) (2007) 30.[2] M. Hino, et al., Physica B 385–386 (2006) 1187.[3] M. Bleuel, et al., Nucl. Instruments Methods Phys. Res. Sect. A 592 (1–2)(2008) 100.[4] W. Paul, in: Proceedings of International Conference on Nuclear Physics andPhysics of Fundamental Particles, Chicago, 1959, p. 172.[5] H.M. Shimizu, et al., Nucl. Instruments Methods Phys. Res. A 430 (1999) 423.[6] M. Yamada, et al., in: Proceedings of EPAC08, Genoa, Italy, 2008, pp.2392–2395.[7] P. Wiltzius, et al., Phys. Rev. A 36 (1987) 2991.[8] V.Y. Bezzobotnov, et al., J. Phys. Chem. 92 (1988) 5738.[9] K. Hirota, et al., Phys. Chem. Chem. Phys. 7 (2005) 1836.[10] B.J. Micklich, J.M. Carpenter (Eds.), in: Proceedings of the Workshop onApplications of a Very Cold Neutron Source, Argonne National LaboratoryReport ANL-05/42, August 2005. ARTICLE IN PRESS M. Bleuel et al. / Physica B 404 (2009) 2629–2632 2632
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks