Close This Window

Please download official ILL logos here


For using on the web or on a screenFor printing in high resolutionWhite version, for dark backgrounds

Download PNG

Download AI

Download white PNG

Download JPG


Download white AI

Data treatment

The Computing for Science (CS) group supports ILL scientists, students and visitors in a number of activities including data analysis, instrument simulation and sample simulation.

Back to ILL Homepage
English French Deutsch 

Data treatment

Treatment of ILL Experimental Results

Version 1.2 - R. Ghosh, January 2000


Many programs have been under development for a considerable time and are consequently written in Fortran. The open source compilers g77, g95 and gfortran have been extensively used, and are available from linux to PCs and OSX, allowing a single source to be recompiled for each platform. To demonstrate that data analysis programming is quite simple, since most reading and plotting activities can be performed by standard libraries, there is a short guide to programming display and fitting!


Some experiments use well established procedures to probe new samples, others use the available instrumentation in novel modes. Keeping up with the diverse fields of study, and cataloguing programs in use will always be incomplete. Only the most used programs are mentioned here, with some indications of other available tools and libraries.


In general there are three phases to experiments. During the experiment the incoming data are monitored for self-consistency. The second phase of data reduction involves removing some of the instrumental characterisics, to yield a generic data set. (In practice, at ILL, as a consequence of history, this may have several inter-convertible forms!) This is then analysed in combination with other measurements in the final stage, and the results compared with simulations and other calculations.


Data treament will be subdivided, crudely, by the nature of the problems solved by each Instrument Group.

To top

Nuclear and fundamental physics

The CERN program package PAW++ has become one of the principal tools for reducing and analysing data, and a number of macros have been customised to simplify use for the NFP instruments.

Lohengrin - PN1 fission fragment spectrometer
PN1 measures data of the type: energy, energy-loss, intensity(1-4).

The spectra comprise many peaks: regions of interest are selected, backgrounds identified, then peaks marked and fitted.

polarisation and calculation of spin-flip ratios


To top

Three Axis spectrometers

These instruments are used for measurement of dispersion relations in single crystals and are programmed to measure specific trajectories in (q,E) space yielding small volumes of quality data, but requiring good knowledge of inherent resolution effects. Polarisation analysis options on certain instruments help distinguish between excitations with magnetic or nuclear origins. The TAS Program Suite is used to treat data from these instruments.


This comprises the largest group of instruments, with the longest history of precision data treatment. Four principal classes are identified here.

Data from multi-detectors, or banks of detectors are typically analysed to refine model structural parameters, or specific peaks are monitored as a function of external constraint or temperature to elicit information on phase and composition changes, or assess residual stress effects.

Oriented single crystals are scanned with either a single detector or, more commonly today, a multidetector, and the Bragg peak intensities obtained are used to refine a large set of structural paramters. Obtaining reliable intensities from the data is the primary function of the initial reduction programs.

Measurement of flipping ratios of magnetic reflections allows detailed interpretation of the distribution of unpaired electrons in magnetic materials. Results are used to confirm models of delicately balanced long range interactions.

Data are measured to very high precision permitting isotope substitution to be used to solve partial structure factors of binary and ternary components and derive pair correlations by Fourier inversion.

To top

Time of Flight and High Resolution Spectroscopy

The first category includes conventional instruements, IN4, IN5 and IN6 and also the Back-scattering spectrometers IN10 and IN16, which have a more restricted energy scan window. Cross-sections are small in samples where multiple scattering is to be minimised. Typically results require combinations of measurements of sample, background and detector calibration. In the case of D7 the use of polarised neutrons in addition allows distinction between coherent and spin-flip incoherent scattering.

The second group of instruments, IN11, IN15, spin-echo spectrometers, are evolving rapidly still, and each experiment leads to modified data treatment schemes for measuring the loss of echo signal of the neutron beams after scattering by condensed matter undergoing very slow relaxation motions.

Large Scale Structures

Long-wavelength neutrons are used to probe long distance correlations in ordered and disordered materials.

  • Structural studies on Membranes

Diffraction patterns, especially exploiting H/D exchange can be used to analyse membrane layers. Analogous techniques are used for thick films in multi-component systems.

  • Small-Angle Neutron Scattering

    This has been used to examine a wide range of problems in materials science, biology and physics looking at long range scattering-density correlations, using all available methods of contrast variation. The low resolution data must be carefully corrected for background and detector effects. Typically many samples prepared under a variety of conditions are compared.

The long wavelength neutrons available on this group of instruments can also be used to probe surface scattering, agin using the full strengths of H/D isotope substitution and very long wavelengths (up to 30A on D17). It is again crucial to achieve a low background, and to subtract such effects.

To top