Technology

The neutron guide upgrade of the TOSCA spectrometer

Description
The primary flightpath of the TOSCA indirect geometry neutron spectrometer has been upgraded with a high-m 14.636 m (including 0.418 m of air gaps) neutron guide composed of ten sections in order to boost the neutron flux at the sample position. The
Categories
Published
of 7
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
  Nuclear Inst. and Methods in Physics Research, A 896 (2018) 68–74 Contents lists available at ScienceDirect Nuclear Inst. and Methods in Physics Research, A  journal homepage: www.elsevier.com/locate/nima The neutron guide upgrade of the TOSCA spectrometer Roberto S. Pinna a,b , Svemir Rudić a, * , Stewart F. Parker a , Jeff Armstrong a , Matteo Zanetti a,b ,Goran Škoro a , Simon P. Waller a , Daniel Zacek a , Clive A. Smith a , Matthew J. Capstick a ,David J. McPhail a , Daniel E. Pooley a , Gareth D. Howells a , Giuseppe Gorini b ,Felix Fernandez-Alonso a,c a  ISIS Facility, STFC, Rutherford Appleton Laboratory, Chilton, Didcot OX11 0QX, UK  b CNISM, Universita’ degli Studi di Milano-Bicocca, Piazza della Scienza 3, 20126 Milano, Italy  c  Department of Physics and Astronomy, University College London, Gower Street, London WC1E 6BT, UK  A R T I C L E I N F O  Keywords: TOSCANeutron guideNeutron beam profileNeutron detectorScintillator detectorNeutron fluxVibrational spectrometer A B S T R A C T The primary flightpath of the TOSCA indirect geometry neutron spectrometer has been upgraded with a high- m 14.636 m (including 0.418 m of air gaps) neutron guide composed of ten sections in order to boost the neutronflux at the sample position. The upgraded incident neutron beam has been characterised with the help of thetime-of-flight neutron monitor; the beam profile and the gain in the neutron flux data are presented. At anaverage proton current-on-target of 160  μ A and proton energy of 800 MeV (ISIS Target Station 1; at the time of the measurements) we have found that the wavelength-integrated neutron flux (from 0.28 Å to 4.65 Å) at theposition of the TOSCA instrument sample (spatially averaged across a 3.0  ×  3.0 cm 2 surface centred around the(0,0) position) is approximately 2.11  ×  10 7 neutrons cm − 2 s − 1 while the gain in the neutron flux is as much as46-fold for neutrons with a wavelength of 2.5 Å. The instrument’s excellent spectral resolution and low spectralbackground have been preserved upon the upgrade. The much improved count rate allows faster measurementswhere useful data of hydrogen rich samples can be recorded within minutes, as well as experiments involvingsmaller samples that were not possible in the past. 1. Introduction TOSCA is an indirect-geometry inelastic neutron spectrometer opti-misedforhighresolutionvibrationalspectroscopyintheenergytransferregion between  − 24 and 4000 cm − 1 [1–3]. The instrument has been operational for almost two decades and during that time has set thestandard for broadband chemical spectroscopy with neutrons [4]. In autumn 2013 as part of the international beamline review [5] it was concluded that for TOSCA to be able to participate in strategic researchareas such as CO 2  capture and charge storage [6], an increase in the incident neutron flux  via  the provision of a neutron guide (asopposed to the simple collimation tubes present at the time) wouldbe highly beneficial. Such a development would allow detailed studiesof industrially relevant systems containing weak neutron scatters (SO 2 ,CO, NO) as well as faster parametric studies, particularly for hydrogencontaining molecules such as hydrocarbons. Additionally, as neutronscattering is an inherently intensity limited technique, studies of smallersamples, which are too expensive to produce in larger quantities, wouldbepossible.Sincethen,thismajorupgradehasbeenimplementedwhich * Corresponding author.  E-mail address:  svemir.rudic@stfc.ac.uk (S. Rudić). has involved extensive simulations together with the complete redesignof the TOSCA primary spectrometer to house a state-of-the-art, high- m neutron guide and associated chopper system to boost the incident fluxon the sample.Neutron guides are essential in order to boost the neutron fluxat a long distance from the neutron source. As neutrons fly nearlyparallel to the guide surface they are retained within the tube by aprocess of external reflection. Traditionally, neutron guides are squareor rectangular cross-section tubes made from optically flat materials,usually glass, that has been metal coated with alternating layers of metal with different scattering length densities. More recently, and inparticular for longer neutron guides, their shape can be rather complex(e.g. elliptical) in order to avoid the loss due to a large number of reflections. Over the years, progress in guide manufacture has led tosupermirror coated guides, having high reflectivity and high  m -valueswhere    =    c (supermirror)/   c (nickel)  i.e.  it is equal to the ratio of thecritical angle of reflection,    c , of the supermirror and nickel coatedoptically flat glass [7]. The neutron guides with greater  m  factor lead https://doi.org/10.1016/j.nima.2018.04.009Received 28 November 2017; Received in revised form 22 March 2018; Accepted 5 April 2018Available online 13 April 20180168-9002/ © 2018 Elsevier B.V. All rights reserved.   R.S. Pinna et al. Nuclear Inst. and Methods in Physics Research, A 896 (2018) 68–74 to increased divergence of the neutron beam, for which the maximumvalue is given by the critical angle    c [ ◦ ] = 0 . 1 ×    ×    [Å]. In ourpreliminary feasibility analysis of the TOSCA neutron guide we haveexcludedtheuseofellipticalormorecomplexgeometries(thataremoreexpensive to manufacture and may considerably increase the neutrondivergence at the sample position) in order to concentrate our resourceson a guide with an advanced supermirror coating. Thus we have used atapered guide in order to focus the beam at the sample position whilemaking sure to preserve the homogeneity of the beam.Based on extensive neutron-transport simulations and baseline stud-ies of the guide neutronic response [8–10], in this article we present the careful mechanical engineering design of the new TOSCA primary spec-trometerandprovideareviewoftheactualperformancegainsfollowingthis upgrade. A comparison between experimental observations andMonte Carlo simulations is provided, and the effect of the upgrade onthe instrument spectral resolution and background have been assessedas well. 2. The primary spectrometer In our design of the neutron guide, we have followed two principles:position the supermirrors to include as much of the neutron flightpathas possible, and to increase the area viewed of the water moderatorby increasing the size of the shutter entrance (from its srcinal value of 84 mm  ×  84 mm) while preserving the beam size at the sample position.Aschematicdrawingofthenew14.636mlongneutronguide(including0.418 m of air gaps) which is dedicated to the TOSCA instrument isshown in Fig. 1. The first section of the guide, G1, is installed in the new 1.937 m long shutter that is positioned at a distance of 1.626 m fromthe moderator centre. The shutter contains an    = 5  straight squareguide with an aperture of 100 mm  ×  100 mm. The remaining ninesections of the guide are tapered, starting from the 100 mm  ×  100 mmentrance of section G2, all the way towards the end of section G10 with40 mm  ×  40 mm aperture (positioned at a distance of 16.262 m fromthe moderator centre, i.e. 0.748 m from the sample position). The angleof taper,  ∼  0.136494 ◦ , has been kept equal in each tapered section,while the  m -factor was increased in steps from    = 5  for sections2 to 6,    = 6  for sections 7 to 9, and    = 7  for section 10. Thisgradualincreaseensuredoptimalgainandenabledthateventhoseshortwavelength (high energy) neutrons whose divergence was sufficientlysmall can be retained. Geometrical parameters of each section and theneutron guide as a whole can be found in Table S1 of the supplementaryinformation. We have chosen mostly tapered over mostly straight guideas neutron simulations pointed towards increased flux gain in the caseof the former [8–10]. As part of the upgrade, the current single disc chopper [3] has been replaced by a double disc chopper positioned at a distance of 9.455 mfrom the moderator centre, and between sections G5 and G6 of theguide. The air gap between the two sections required to fit the discs waskepttoaminimumof6.6cm,thusreducingtheneutronfluxattenuationdue to scattering by air. The new double disc chopper allows utilisationof all the neutron pulses arriving at TOSCA, even when Target Station1 operates in 50 Hz mode (albeit without access to the elastic line; seereference 3 for further details about the extension of the spectral regionof the instrument).In order to achieve the highest neutron flux at the sample posi-tion, the flightpath through the neutron guide is in vacuum (p =2.5  ×  10 − 2 mbar). Since all the sections of the guide could not be joinedtogether in a single housing, various housings are sealed with the helpof 0.5 mm thick aluminium windows. Overall, there are a total of ninealuminium windows along the beamline flightpath ( i.e . between themoderator and the sample position) as indicated by the orange verticallines in Fig. 1.Two neutron beam monitors are positioned along the flightpath; thefirst before the chopper, at a distance of 8.900 m from the moderatorcentre and the second after the chopper at a distance of 15.871m fromthe moderator centre. The latter monitor is used for the normalisation of the neutron flux. The monitors allow measurements of the time-of-flightspectra of neutrons; and were made from GS20 cubes (cerium-activatedlithium aluminosilicate glass, 0.25 mm in size) that were distributed7 mm apart and across a 7  ×  6 array [11,12]. TOSCA has four  3 He detector tubes (two on each side of the beam)inback scatteringthat areusedfor diffractionmeasurements. Asa resultof the beam upgrade and in order to accommodate the last section of the guide they needed to be slightly moved away from the centre of theflightpath i.e .theyarenowpositionedatanangleof175 ◦ and176 ◦ inthebackward direction. Although the tubes are stationed in air, virtually alltheflightpathbetweenthesamplepositionandthediffractiontubesisinvacuum, with the vanadium window acting as the boundary. The finalsection of the neutron guide, G10, is connected to the instrument by thetapered flight tube and the whole volume is kept at cryogenic vacuum( <  10 − 6 mbar).The instrument was not operational between the end of May 2016and mid-February 2017. By the end of November 2016, all sections of the guide were in place apart from the new TOSCA shutter with theinitial section of the guide within it, G1, which remained the same asbefore the upgrade. Since the overwhelming majority of the guide wasinstalled we tested the setup for enhanced neutron flux, in order to havebetter idea about the influence of the guide inside the shutter (installedsubsequently) on the neutron flux, beam profile, spectral resolution andbackground(seeSI).WewillrefertothisinterimconfigurationastheC1configuration, while the configuration before the upgrade (i.e. withoutthe neutron guide) will be denoted as C0 [13]. The last section in theshutter was installed in January 2017 to give the final (so called C2)configuration. After the two weeks of commissioning measurements theinstrument was returned to the user programme. 3. Experimental setup The experimental setup used to measure the neutron flux at thesample position was described in Ref. [13]. The neutron sensitive component was a cuboid of cerium-doped glass scintillator, measuring0.96  ×  0.95  ×  0.53 mm 3 . The TOSCA closed cycle refrigerator (CCR)was removed from its position in order to accommodate the assemblyframeontotheflange,andthusthemeasurementsatthesamplepositionwere performed in open air and at room temperature. The positionof this point-sampling detector was controlled  via  a computer scriptwhich moved it automatically after the accumulation of 10000 framesat each spatial point (1 frame = 100 ms), each frame containingfour consecutive neutron pulses, without the need to interrupt thebeam between different runs. 169 points around the beam centre (from − 3.0 cm to  + 3.0 cm in the X (horizontal) and Y (vertical) directions,when looking downstream) were measured, sampling the time-of-flightspectrum every 5 mm, see Fig. S1 in supplementary information (SI).Subsequently the data were calibrated to give the neutron flux atthe sample position in units of neutron cm − 2 s − 1 Å − 1 and eventuallyintegrated in the wavelength range of interest (see reference 13 forfurther details). At the time of this study, the first nine sections of theguide starting from the shutter towards the sample position were undervacuum,whilesectiontenoftheguideandthesampleenvironmentareawere in air. Computational details.  The McStas software package [14] was used inorder to perform Monte Carlo simulations of the TOSCA beamline. Thegeometrical parameters of the upgraded instrument primary beamline(see Table S1 in SI) were implemented in the virtual instrument, whilethe water moderator file [14] was provided by the ISIS Neutronics GroupandwasbuiltusingMCNP-XcalculationsoftheactualTS1target-reflector-moderator assembly. In the simulation the angle between theTOSCA beamline axis and the moderator face was kept at  90 ◦ i.e . themoderator face and the shutter face were perfectly aligned/parallel. Inreality, the beamline axis is tilted by  ∼ 13 . 2 ◦ from the line perpendicularto the moderator face [15] and this precise information has been taken 69   R.S. Pinna et al. Nuclear Inst. and Methods in Physics Research, A 896 (2018) 68–74 Fig. 1.  Schematic representation of the side view of the TOSCA neutron guide as installed on the beamline. The guide sections are numbered in the order in whichthey appear along the flightpath. The starting position of each section, in relation to the moderator centre, as well as its length are provided (in mm units). Aluminiumwindows (W) are indicated by the orange vertical lines. into account when generating the moderator file [16]. Thus for thepurposes of McStas calculations the difference between the real and thesimulatedangleisirrelevant,astherealitybasedmoderatorfilecontainsall the necessary information about the neutrons travelling towardsthe instrument. The ‘pre-guide’ C0 configuration of the TOSCA wassimulated in order to have the baseline performance of the instrumentas a reference for the subsequent simulations regarding the gain inneutron flux due to the neutron guide. To calculate the performanceof the supermirror neutron guide we used experimentally determinedreflectivity profiles (provided by the manufacturer, Swiss Neutronics)for each section of the guide. The Mantid software package [17–20] was usedinordertoanalysetheexperimentaldata,anddetailsoftheappliedmethodology of data analysis are the same as described in Ref. [13]. 4. Results and discussion  Final (C2) configuration:.  The experimentally derived TOF spectrumas a function of wavelength of the TOSCA neutron flux after theguide installation (spatially averaged across the 3.0  ×  3.0 cm 2 surfacecentred around the (0,0) position) is shown in Fig. 2 (green trace). Itsfeatures are in line with our expectations [21] since the peak in the ‘moderated hump’ appears at around 1.1 Å which is characteristic of the room temperature water moderator. Its shape is characterised bythe epithermal component at short wavelengths and the Maxwelliancomponent that follows. Equally, in terms of its overall profile theexperimental spectrum is in line with the results of the Monte Carlosimulations (blue empty diamond symbols) performed with the help of the McStas software, although the latter needed to be scaled down bya factor of 2.53 in order to make the simulated and the experimentalintegrated neutron flux in the region between 0.28 Å and 4.65 Åequal. Similar discrepancies have been observed before [13] and will be discussed in more detail later in the text. The TOSCA beam profile atthe sample position is shown in Fig. 3.Its central region, 2.5  ×  2.5 cm 2 in area, is relatively homogeneous,while the overall beam has a roughly 4.5 (H)  ×  4.5 (V) cm 2 squareshape (taking into account the region with the neutron flux higherthan 50% of the maximum intensity) as a consequence of the neutronguide positioned along the beamline. The beam centre appears to bedisplaced by  ∼  2 mm (up and to the right when looking downstream,  i.e .from the moderator towards the instrument) from the nominal primaryflightpath. Such displacement (2 mm at a distance of 17010 mm) issmaller than before the guide upgrade [13], partly as a result of the instrumentrepositioning,by2mmtowardsright i.e. thebeamflightpathis now better overlapped with the sample position. The beam shift mayindicate a minor shutter misalignment or possible unwanted reflections.The evaluation of the beam spatial profile as a function of wavelengthis shown in Fig. S3 (see SI). It appears that the beam homogeneity is Fig. 2.  Neutron flux at the TOSCA sample position as a function of wavelength.Experimentally derived values (spatially averaged across the 3.0  ×  3.0 cm 2 surface centred around the (0,0) position) are shown by the green line, whilethose obtained from the Monte Carlo simulations (spatially averaged across the4.0  ×  4.0 cm 2 surface centred around the (0,0) position) are shown in blue asempty diamond symbols. Monte Carlo values (not corrected for attenuation) of the neutron flux were scaled down by a factor of 2.53 in order to make thesimulated and the experimental integrated neutron flux in the region between0.28 Å and 4.65 Å equal. For comparison, the neutron flux (experimental— black line and simulated — red empty triangle symbols) at the TOSCAsample position before the guide upgrade are also shown (configuration C0)(see reference 13 for further details). preserved across neutrons with various energies. In particular, the beamprofile for neutrons with wavelengths between 0.8 Å and 1.2 Å and thebeam profile for neutrons with wavelengths between 2.5 Å and 2.9 Åhave very similar distribution, although the latter flux is one order of magnitude smaller.The McStas simulation of the TOSCA beam profile at the sampleposition with the guide upgrade is depicted in Fig. 3. The simulated beam has a 4.0 (H)  ×  4.0 (V) cm 2 (taking into account the region witha neutron flux greater than 50% of the maximum intensity) Gaussianprofile (due to the collimation). The flux is roughly constant across the2.0 (H)  ×  2.0 (V) cm 2 surface as indicated by the horizontal cut, seeFig. 4. Due to the symmetrical geometry of the square cross sectionguide, there are no privileged directions of reflections and, as a con-sequence, the results show a perfect spatial symmetry about the centralposition. As emphasised before, the simulated neutron flux is 2.53 timeshigher than the experimentally observed values. This is not surprising 70   R.S. Pinna et al. Nuclear Inst. and Methods in Physics Research, A 896 (2018) 68–74 Fig. 3.  Measured (left) and simulated (right) neutron beam profile at the TOSCA sample position with the guide installed. The measured and calculated valuesof time averaged neutron flux, integrated across 0.28 Å to 4.65 Å wavelength range of interest to TOSCA were obtained with i.e. scaled to the average protoncurrent-on-target of 160  μ A, respectively. since the simulation was performed in vacuum while in reality, at thetime of this experiment out of 17.01 m long flightpath between themoderator centre and the TOSCA sample position 14.1465 m wereunder vacuum, while 1.3030 m were in air, 1.5560 m were in helium,and 0.0045 m were in aluminium. Thus in order to compare with theexperiment, the simulated values should be corrected by the factor of 0.8765 which corresponds to the transmitted neutron flux upon travelalong the above described flightpath. The corrected simulated valuesare shown in Fig. 4 as well. They are still 2.22 (2.53  ×  0.8765) timeslarger than the experimental values, and the difference can be ascribedto various factors. Firstly, the muon producing target (located upstreamof the TS-1 target) is not in the MCNPX model so the calculated fluxshould be corrected by a factor of 0.95. Secondly, the physics modelchoice for MCNPX simulations can affect the flux at the level of 10%–15%. For example, if the CEM03 physics model (used in this set of simulations) is replaced by INCL4-ABLA model — the correction factoris 0.86. Thirdly, the building of TS-1 MCNPX model [22] is still awork in progress. Recent findings [15] showed that the TS-1 target cooling water is not pure heavy water (as defined in the model) buta mixture of 80% heavy water and 20% light water, and that the6 mm thick boral plate is positioned between the target and the watermoderator. As a result, the calculated flux should be corrected by anadditional factor of 0.87. After applying all of these corrections, theagreement between experimental and calculated flux is well below afactor of 2. Please note that further uncertainties exist which appearto be more difficult to estimate quantitatively such as the details of material composition of the decouplers around the TS-1 moderators,precise position of the poisoning foil, the (real) vertical position of theTOSCAshutterinsert,andpossibleoperationaleffects[23]etc.,butmayexplain the discrepancy.As can be seen from Fig. 3, the measured and the calculated neutron flux at the centre of the sample, the (0,0) position (see SI),are 2.54  ×  10 7 neutron cm − 2 s − 1 and 7.32  ×  10 7 neutron cm − 2 s − 1 , respectively. The corresponding values spatially averaged across3.0  ×  3.0 cm 2 and 4.0  ×  4.0 cm 2 surface centred around (0,0) positionare 2.11  ×  10 7 neutron cm − 2 s − 1 and 5.35  ×  10 7 neutron cm − 2 s − 1 ,respectively. Between the latter two values a normalisation factor of 2.53 must be applied to match the absolute values. The neutron fluxvalues associated with the plateau shown in Fig. 4, and averaged along the vertical axis (between  − 2.0 cm and  + 2.0 cm) are slightly smaller. Inparticular the simulated values were derived with the help of the linearflux monitor, not the position sensitive detector used in the case of thebeam profile shown in Fig. 3.Fig. 5 shows the TOSCA simulated divergence profile at the sampleposition before and after the guide upgrade. In line with expectations,after the upgrade the neutrons arriving at the sample position havea larger divergence distribution: the full width at half maximum is Fig. 4.  The experimental (green filled squares), the simulated (black emptytriangles)andthecorrectedsimulated(forattenuationinthehelium,aluminiumand air; blue empty diamonds) neutron beam profile projected along thehorizontal axis at the TOSCA sample position. The beam flux has been averagedalong the vertical axis (between  − 2.0 cm and  + 2.0 cm, i.e. within the beamheight) and includes only neutrons within the wavelength range of interest toTOSCA (from 0.28 Å to 4.65 Å). approximately 1.3 ◦ , as opposed to 0.3 ◦ before the upgrade. As alreadydescribed, the neutron guide is able to retain neutrons (with an incidentangle  <   c ) within the guide tube, rather than acting only as a simplecollimator and thus by the time they reach the sample surface they cancrossitatananglelargerthanallowedbythedirectlineofsightbetweenthe moderator and the sample.As a result of the guide being installed on TOSCA, the neutron fluxat the sample position has been significantly increased, see Fig. 6. The green trace shows the experimentally determined gain as a function of wavelength and the black trace corresponds to the gain derived fromthe Monte Carlo simulations. Since measurements of the neutron fluxat the sample position for the final C2, and before the upgrade C0,configurations were performed with the proton energy on target of 800 MeV and 700 MeV, respectively, the former C2 neutron flux hasbeen scaled down by a factor of 1.17 [see Fig. S5 in SI] before takingthe ratio to calculate the experimental gain due to the neutron guide.Additionally, as theoretical calculations were performed in vacuum,these values of the neutron flux gain need to be corrected for theattenuation as neutrons travel through helium, aluminium windows andair (blue trace) before they are compared with the experimental data.The gain in neutron flux is 6 times for neutrons with a wavelength of  71   R.S. Pinna et al. Nuclear Inst. and Methods in Physics Research, A 896 (2018) 68–74 Fig. 5.  The TOSCA simulated divergence profile at the sample position before (left) and after (right) the guide upgrade. Fig. 6.  Gain in the neutron flux at the TOSCA sample position as a functionof wavelength. Green trace shows experimental values derived with the beaddetector, while black and blue traces show simulated values derived from theMonte Carlo calculations without, and with, the correction for attenuation inthe helium, aluminium and air, respectively. Please see the SI for further details.(For interpretation of the references to colour in this figure legend, the readeris referred to the web version of this article.) 0.5 Å (2640 cm − 1 ), 46 times for neutrons with a wavelength of 2.5 Å(105 cm − 1 ), and 82 times for neutrons with a wavelength of 4.6 Å(31 cm − 1 ),  i.e.  the most significant gain is for the neutrons used tostudy low energy molecular and lattice vibrations. The slight dip in theexperimental neutron gain between 3.7 Å and 4.6 Å is a result of thedifferent starting times (wavelengths) at which the double disc chopperstarted to block the slow neutrons (in order to prevent the 20 ms sub-frame overlap) at the time of C2 (3.7 Å) and C0 (4.25 Å) configurationsmeasurements. Figure S7 in the SI shows the experimental (left) andsimulated (right) gain in the neutron flux as a function of the positionwithin the TOSCA sample area. Since after the upgrade the neutronbeam is more divergent, and some neutrons can be detected at theposition where there were none before, the gain appears larger aroundthe periphery than in the centre of the sample position.In order to understand the effect of possible shutter (i.e. initialsection of the guide G1) misalignment on the neutron beam properties,we have performed additional simulations in which the shutter wasmisaligned by 0.5 ◦ (to the left when looking downstream,  i.e.  fromthe moderator towards the instrument) from the intended instrumentflightpathandpivotedaroundtheshutterentrancepositionatadistanceof 1.626 m from the moderator. As a consequence, the wavelength-integrated neutron flux (from 0.28 Å to 4.65 Å) at the TOSCA sampleposition was reduced by 20% (see Fig. S8 in the SI). Furthermore,it appears (see Figs. S9 and S10) that the misalignment does notsignificantly alter the beam profile, although the beam is slightly shifted(at the sample position) in the same direction (towards left) as themisalignment. Fig. S11 shows the consequence of misalignment on thebeam divergence at the sample position. This suggests that as a resultof the misalignment, patterns in the divergence profile can be observedwithout alteration of the overall average divergence.In comparison to the interim C1 configuration (which had a simplecollimation tube within the shutter and an 84 mm  ×  84 mm entranceopening) the final C2 configuration (which has 1.937 metre long    = 5 neutron guide inside the shutter and an enlarged 100 mm  ×  100 mmshutterentrance)givessignificantlyenhancedneutronfluxatthesampleposition; compare experimental values of the neutron flux in Fig. 2 andFig. S13. For further information about the properties of the neutronbeam at the TOSCA sample position in relation to C1 configurationplease see SI.  Spectral resolution and background.  While the increased neutron fluxat the sample position has been achieved as part of this project, itwas equally important to preserve the excellent spectral resolution andlow background afforded by TOSCA [3]. In order to check the effectof the installed neutron guide on the instrument spectral performancewe have performed neutron scattering measurements of the standardcalibrant, 2,5-diiodothiophene [24]. Fig. 7 shows the elastic line of  the 2,5-diiodothiophene sample recorded before the TOSCA upgrade(black trace, C0 configuration) and after the upgrade (blue trace, C2configuration). The same sample was used in both cases and the spectrawere accumulated for the same period of time. The spectra suggest thatthe spectral resolution at the elastic line is essentially unchanged andbelow 2 cm − 1 .Equally, examination of the inelastic neutron scattering spectra of the same compound before and after the upgrade, see Fig. 8, highlights thatthespectralbackgroundhasremainedlowacrossthespectralrange, i.e.  it has been maintained close to the pre-upgrade levels. Note thatthe total background (black and blue trace) is not very much differentfrom the spectral background as the combined contribution of thewhole instrument and the empty standard TOSCA flat aluminium cellto the recorded signal (red trace) is almost negligible, and has not beensubtracted from the black and blue trace. Furthermore, one can observethat upon the installation of the neutron guide the spectral resolutionat high energy transfers is ever so slightly diminished, but as Fig. S12shows for the band associated with the CH stretches (positioned around3085 cm − 1 ) the change does not significantly alter the quality orunderstandingofthedata.Intheupperrightcornerof Fig.8,thespectral region in between 0 and 100 cm − 1 is shown on an expanded scale. 5. Conclusions As part of the TOSCA primary spectrometer upgrade the simplecollimator tube between the moderator and the instrument has been 72
Search
Similar documents
View more...
Tags
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