Sample environments are an essential and integral inclusion to all neutron scattering experiments. While it is possible that measurements at ambient temperature and pressure, and under the influence of the Earth's magnetic field are sufficient for a scientific study, this is rarely the case. Consequently, additional equipment to apply specific conditions and to induce the sample into a phase or state of particular interest is desirable. A non-exhaustive list includes subjecting the sample to high or low temperatures, high pressure or under vacuum, varying magnetic field, controlled shear stress or strain, exposure to vapour, chemical manipulation or a simultaneous combination of several such conditions. Note that while some of the above options have the primary objective of keeping a certain parameter fixed (e.g., pressure), others may be used to deliberately manipulate the structure or dynamics of the sample during measurement (e.g., following a temperature quench or observing the relaxation of a complex fluid following the cessation of shear).

‘Standard environments’ could include a humidity generator, a temperature-ramping cell, tumbler cell (to prevent sedimentation), stop-flow [1] or continuous flow cell. However, despite their critical role, conducting a literature search in Scopus using the search term “neutron” AND “sample AND environment*” AND “scattering OR radiograph* OR tomograph* OR reflect*” and limiting the search to articles that only had "sample environment" as a keyword yielded only 113 articles from the time period 2005 to October 2023. Analysing the search result in VOS viewer [2] using a co-occurrence analysis shows how the articles are related to one another according to keyword terms. In Figure 1 each ball represents a keyword term, and the size of the ball is the weight of that term. As can been seen the term “sample environment” is the largest item as every article in the analysis used this as a keyword. The lines represent the link between each term and the colour represents a cluster of items (e.g., red is one cluster), and a cluster is the strength of the relatedness of the terms. In Fig. 1 there are nine clusters with the largest cluster being the red cluster with 29 terms and the smallest cluster only having one term (“pulsed magnet”). The most distinct and second largest cluster is the green cluster which has terms mainly relating to cryogenics and low temperature effects. Terms that could be conceived as being particularly related to colloid and interface science such as “soft matter”, “rheology”, “lipid bilayer”, etc., are generally spread out and not concentrated in a particular cluster. What we interpret from this is that colloid and interface science makes the most demand for novel or bespoke sample environments whereas other areas of science that utilise neutron scattering tend to (but not exclusively) use more ‘standard’ or commercially available equipment.

Neutron scattering centres will readily advertise their range of sample environments [3,4] that can be selected when a proposal is prepared. Many will be common across facilities e.g., cryostats, superconducting magnets, etc. while others will be unique. A number of reviews have described sample environments for neutron scattering. Lindner et al. [1] discuss sample environments for small-angle neutron scattering (SANS) and neutron reflectometry (NR). The reader is directed to the excellent 2021 review of sample environments specifically for SANS with a focus on time-resolved neutron scattering experiments [5]. More recently Martel and Gebel reviewed this area more broadly for structural biology [6]. Recent reviews of developments in NR have also discussed aspects of sample environments that can be used for surface and interface analysis [7,8]. It is also relevant to consult several previous works dedicated to X-ray facilities [[9], [10], [11]] despite some of the described devices not being currently available for neutrons, since an appreciation of what is possible can both inform the technical aspects required to develop neutron-suitable equipment and highlight potential opportunities. An example is differential scanning calorimetry – a technique widely used simultaneously at X-ray facilities – which, only relatively recently has been implemented for SANS [12]. It is indisputable that the development of sample environments for X-ray scattering is more advanced than for neutron scattering. Reasons for this relate to both beam intensity, and thus the ability to achieve much greater time resolution, as well as (and indeed related) the opportunity to use implicitly smaller beams. However, the advantages of contrast, beam penetration as well the avoidance of beam-mediated sample damage position neutrons as a probe important in their own right. These and other advantages of neutron scattering are briefly outlined below.

Sample environments which are non-standard, that may be bespoke or not generally commercially available or simply have been used in a non-standard way, and that find application to colloid and interface research, represent the focus of this article. As a consequence, extreme environments including high magnetic fields and very low and very high temperature devices are not considered here. A recurring theme concerns the ability to collect neutron scattering data while, at the same time, measuring with a complementary technique. This is a particular advantage for in situ or in operando studies of complex nanoscale systems which irreversibly change their structure with time, e.g., upon a reaction, crystallisation, aggregation, gelation, particle formation, growth, or aging process. The potential gain in information, compared to conducting two independent measurements (i.e., one for each technique) should be balanced with the effort required to implement such a ‘simultaneous’ system. As a rule of thumb, for studies in which structural changes occur over a period of, say, greater than two minutes, independent measurements generally have the possibility to be correlated without a significant error in time [12]. The review therefore also extends itself to the concept of tandem techniques as opposed to individual sample environments per se where combined methods are used to reveal additional information on a sample e.g., combined SANS and SAXS. Where appropriate, the focus is also on the last few years since the publication of the most recent reviews in this field.