Publications on FIHM Groups

A glass full of magnetism

Tue, 07 Apr 2026 08:00:00 +0000

Glasses are familiar materials, yet their atomic-scale organisation and physical properties remain difficult to fully understand. This challenge is particularly pronounced for magnetic glasses, where structural disorder strongly affects collective spin behaviour. Most conventional glasses are chemically simple inorganic systems, which limits their tunability and functionality. Metal–organic frameworks (MOFs) combine metal ions and organic linkers, and so by using organic chemistry to create different organic linker molecules, we have the ability to control the magnetic behaviour of MOFs.

Remixing 2D magnets

Sun, 17 Nov 2024 18:50:00 +0000

One of the key attractions of metal-organic materials is their variety: organic molecules can be tweaked and refashioned almost indefinitely. We have recently reported a new ‘crinkly magnet’, nickel dichloride 2,1,3-benzothiadizole (btd), with an unusual “noncollinear” magnetic structure. In this paper we explore what happens when the sulfur atom in btd is swapped for an oxygen atom.

Surfing the (mag)net

Thu, 12 Sep 2024 08:00:00 +0000

A net is a periodic (repeating) series of connected points. This concept is at the heart of solid state science because we often think of the chemical structures of materials as comprising atoms connected by bonds, which therefore form a net. The underlying net of a material allows for classification—is this structure related to that structure—but also prediction of its properties, e.g. is it likely to be stiff or soft. Although there are an infinite number of possible nets, the best estimate is that there are only twenty nets with only one kind of vertex and edge.¹ Most of these will be familiar, at least to chemists, like the simple cubic pcu net beloved of scaffolders or the face-centred cubic fcu net that emerges in pyramidal stacks of oranges. There’s a particular interest in the nets which are not bipartite (i.e. the points of the net can’t be coloured in red and blue such that no red neighbours another red and no blue a blue, normally containing a triangle), because these can very have unusual magnetic properties. In these non-bipartite nets, if a magnetic spin is put at every node and they interact such that the spins prefer to point in opposite directions, that is, they are antiferromagnetically coupled, there are very many states there are nearly equal in energy: they are frustrated.

All iNCSlusive

Tue, 09 Jul 2024 09:00:00 +0000

Figure: Periodic table of thiocyanates, showing the state of our knowledge

There are 118 elements and we are now confident that there will be no more stable elements discovered. Equally, we’re fairly confident that there won’t be many (if any) binary compounds made with stable elements. You might assume that this settled state of knowledge is correct for other simple compounds; however, this isn’t true.

Crinkle cut magnets

Tue, 02 Jul 2024 10:00:00 +0000

Figure 1. A schematic illustration of how tilting ferromagnetic chains with strong single ion anisotropy can produce a noncollinear structure with a net moment

In most magnetic materials, the underlying atomic spins line up either parallel (ferromagnetic correlation) or antiparallel (antiferromagnetic correlation): they are collinear. In noncollinear magnets spins can arrange themselves in any direction relative to each other (depending on the interactions and geometry of the material) and these noncollinear spin textures can produce magnetic properties that are hard to realise otherwise, including electrical control over magnetic spin or nanometric magnetic data storage. Creating materials with these noncollinear structures is challenging, however, because they typically require multiple different spin-spin interactions to be competitive with each other.

A molecular framework honeycomb magnet

Mon, 20 May 2024 10:00:00 +0000

In this paper we investigate a metal thiocyanate compound [Na(OH₂)₃]Mn(NCS)₃ which consists of rods of a honeycomb manganese thiocyanate network threaded by rods consisting of sodium cations and water. This structure has perfect three-fold symmetry, but unusually, the manganese atom do not lie on an inversion centre (i.e. each metal atom knows which way up it is), although the crystal as a whole does have inversion symmetry.

High pressure behaviour of the magnetic van der Waals molecular framework Ni(NCS)₂

Tue, 31 Oct 2023 09:00:00 +0000

Summary by Madeleine Geers

Selected as Editor’s Suggestion

We investigated what would happen to Ni(NCS)₂ when subjected to the force of an elephant over a square centimetre. Elephant by Dmitry Abramov

Detecting pressure changes plays an essential role in our everyday lives, from biomedical applications such as assessing adhesion of prosthetics, to entering a code on a PIN pad or monitoring gas flow on planes. Often we cannot measure pressure changes directly, so identifying materials whose physical properties, such as magnetism, change when a pressure is applied can be very useful. Two-dimensional, or “van der Waals”, materials are particularly exciting for this kind of application. The atoms in 2D materials are strongly bonded in two directions, whilst having weak “dispersion” or van der Waals forces along the third axis: with graphite being the most famous example. If a material is magnetic, the temperature at which it magnetically orders can be increased quite a lot by applying pressure, for example the 2D magnetic material NiI₂ increases its magnetic ordering temperature by 235 K when compressed to 190 kbar, an increase of +1.6 % kbar^–1. 190 kbar is a considerable pressure: James Cameron in his submarine reached pressures of 1 kbar at the bottom of the Mariana trench and a typical volcano erupts with between 0.2–3 kbar of pressure. It would be useful to have materials which show big changes at more everyday pressures.

An open source webtool for calculating compressibilities & thermal expansion

Sun, 22 Oct 2023 18:44:00 +0000

Figure: PASCal Logo shows an indicatrix, which plots the anisotropy of response of a crystal.

This paper describes the newest version of PASCal, a simple webtool designed to calculate the principal thermal expansivities, compressibilities or response to electrochemical strain of a system from lattice parameters. For low symmetry crystals, the arbitrary choice of unit cell means that if you only look at the cell lengths, it is possible to miss interesting features in the (thermo)mechanical behaviour of a crystal, for example negative thermal expansion or negative compressibilities. PASCal is designed to be an easy method for calculating the principal strain values that give the complete description of these systems. The new version is completely open source, available offline and includes electrochemical strain.

Nature inspired complex magnetic order

Fri, 10 Mar 2023 12:00:00 +0000

Summary by Maddie Geers

Nature is a deep-rooted source of inspiration for innovations, a notion continually woven into our everyday lives. Magnesium silicate (MgSiO₃) is one of the most abundant minerals on Earth and about 20 years ago it was discovered to undergo a structural transition. In most of the Earth’s mantle, magnesium silicate adopts the perovskite structure, however, plunging to the lower depths of the Earth’s mantle it evolves to have the post-perovskite structure. The high temperatures and pressures needed for this transition to occur (1200 K and 120 GPa) make studying this compound very challenging, so identifying compounds that exhibit the post-perovskite structure at atmospheric pressure is a useful task. One method to achieve this is by replacing the atomic oxygen anions with molecular anions, for example dicyanimide (NCNCN^–) or thiocyanate (NCS^–) ligands.

Designing quantum phases in metal-organic magnets

Mon, 09 Jan 2023 00:00:00 +0000

Summary by Jem Pitcairn

Figure 1. Left: the structure of CrCl₂(Pym), showing the alternating spin chains weakly coupled by the organic molecules. Right: Inelastic neutron scattering (INS) measurements and calculations which allowed us to measure the strength of the magnetic interactions.

At the end of the tether

Sat, 31 Dec 2022 00:00:00 +0000

Figure 1: Schematic diagram of the behaviour of oligoMOFs

A summary of this paper can be found in Nature Reviews Chemistry.

This work was led by Prof Seth Cohen at the University of California, San Diego.

Long-distance (magnetic) relationship

Sun, 26 Jun 2022 12:00:00 +0000

Figure. 1 Magnetic order in CrBi(SCN)₆. Left: the magnetic superexchange pathways. Right: Neutron diffraction data showing magnetic order.

Ions with unpaired electrons are magnetic and when those magnetic ions are assembled into a crystalline solid, their magnetic spins interact. At low enough temperatures, these spins therefore typically lock into an ordered state. The temperature at which this ordering happens depends, however, on the strength of the magnetic ’exchange’ interactions between the spins, which is strong for ions next to each other (e.g. in magnetite, Fe₃O₄) but gets rapidly weaker as the magnetic ions are separated by more intervening atoms (e.g. chrome alum, KCr(SO₄)₂.12H₂O). In crystals where the magnetic ions are connected by multiple multi-atomic molecules this means that the temperature at which magnetic ordering occurs can very often be too cold to practically measure (as low as 0.05K, i.e. within a twentieth of a degree of absolute zero).

Tracking the formation of MOFs in solution

Wed, 17 Nov 2021 00:00:00 +0000

Summary by Francesca Firth

Figure 1: Schematic of stages in the formation of hcp UiO-66(Hf) in solvothermal conditions, through different cluster intermediates.

Metal-organic frameworks (MOFs) are of great interest for applications such as energy storage and carbon capture, particularly due to their outstanding chemical tunability. In particular, zirconium (Zr) and hafnium (Hf) MOFs are promising for real-world applications because they are very stable. Our previous work discovered that changing the synthesis conditions, including which solvent was used what temperature the reaction was carried out, changed the structure of the MOF that formed for the important UiO family of MOFs. In particular we found that we could introduce missing clusters or make MOFs with ‘double’ Hf₁₂ clusters, rather than classic UiO Hf₆ clusters. These new MOF structures have very different chemical and physical properties.

Combining complex orders in perovskites

Fri, 04 Dec 2020 12:00:00 +0000

Many of the most useful properties of materials arise from the collective ordering of many individual nanoscale components across an entire crystal: for example, ferromagnetism is the result of aligning the magnetic spins within a material and ferroelectricity is the ordering of polar electric distortions. One way to create new properties in materials is therefore to order simultaneously multiple different features at once (e.g. combining ferromagnetic order with ferroelectric order to create magnetic and electric ‘multiferroics’). Making use of this strategy, however, means confronting the general tendency for materials towards disorder.

Layered magnetic molecular-frameworks

Mon, 20 Jul 2020 09:00:00 +0000

Figure: Structure and magnetic properties of the magnetic molecular frameworks M(NCS)₂ (M = Mn, Fe, Co, Ni). Left: the chemical structure of the M(NCS)2 frameworks; centre: the net magnetic interaction strength for each framework and right: the magnetic structures of M(NCS)₂, with arrows denoting the relative orientation of the bar magnets. The magnetic structures of Mn(NCS)₂, Fe(NCS)₂ and Co(NCS)₂ are in orange, and in blue for Ni(NCS)₂.

Directly detecting defect domains

Fri, 17 Jul 2020 00:00:00 +0000

Figure: Scanning electron diffraction images of UiO-66. Top: a single crystal of UiO-66 visualised in (purple). Bottom: the defect nanodomains (green).

Defects are key to the properties of all kinds of materials, from silicon chips to turbine blades. Metal-organic frameworks (MOFs) are no exception. Previously, we have shown that nanodomains consisting of a more open structure can be produced within crystals of the most important MOFs, so-called UiO-66, by controlling the synthesis. In this paper, we collaborated with teams based at the Universities of Cambridge and Leeds who are experts in an emerging technique, scanning electron diffraction (SED), which allowed us to image directly entire defect nanodomains in a MOF for the first time.

Diverse defects in metal thiocyanate frameworks

Fri, 07 Feb 2020 09:00:00 +0000

“Your imperfections are what make you perfect” is not just a twee word-art slogan on the wall of a chain coffee shop but also a guide for materials chemistry. Defects, the deviations from perfect crystalline order, are key to producing the useful properties of many solids, from solar cells to gemstones. The challenge facing chemists, however, is that defects are hard to see: there are not many of them and they are usually randomly distributed throughout the material. This means that standard crystallographic techniques, which rely on long-range order, are much less useful. As a result, it is very challenging to know what defects are present in a material, even if we can see their consequences.

SquidLab

Fri, 07 Feb 2020 09:00:00 +0000

A team at the Universities of Warwick, Nottingham and Cambridge have developed open-source software for analysing magnetic data and allowing more sensitive measurements

Summary by Dr Matt Coak

Measurements of magnetic properties in the lab are commonplace and essential across a range of physics, chemistry, materials science and even biology. There are many cases however where the magnetic signal of the sample holder or environment swamps that of the material to measure. Our new software, SquidLab, provides a powerful and flexible toolbox to remove magnetic backgrounds to recover the sample’s signal and allow measurements that were previously impossible.

Ionic and Electronic Conduction in TiNb₂O₇

Mon, 09 Dec 2019 09:00:00 +0000

Figure: TiNb₂O₇ is a battery anode material, which allows lithium ions to rapidly pass through it and is a good electrical conductor when lithium is inserted into its structure.

This work was carried out in collaboration with the groups of Prof Clare Grey and Dr Siân Dutton at the University of Cambridge, Dr Andrew Morris at the University of Birmingham and Prof Graham Henkelman at the University of Texas.

Short-range ordering in a battery electrode, the ‘cation-disordered’ rocksalt Li_1.25Nb_0.25Mn_0.5O₂

Mon, 01 Jul 2019 10:00:00 +0000

Powder diffraction data, left, X-ray crystal structure, centre, and battery performance, right, the the more ordered and less ordered forms of LiNbMnO.

This work was carried out in collaboration with the groups of Prof Clare Grey and Dr Siân Dutton at the University of Cambridge.

Faulty Goods? the exciting consequences of introducing defects in MOF frameworks with water

Wed, 20 Feb 2019 09:30:00 +0000

Summary by Francesca Firth

Reading the abstracts from conferences on fields ranging from crystallography through energy storage to the many branches of materials chemistry, it is a challenge to avoid spotting one on the subject of metal-organic frameworks (MOFs). With their incredible tunability and versatility, arising from the wide range of metal nodes and organic linkers available from which to build a multi-dimensional porous network, MOFs are the source of great excitement in the search for materials for all kinds of applications. However, changing the identity of the metal or linker is not the only method of altering the properties of the framework – the formation of defects, such as missing linkers, missing nodes, or stacking faults, in a structure can also significantly change the properties of a material (often with less disastrous consequences than a misplaced screw in a build-at-home chair…). UiO family MOFs are ideal for studying defects, as their strong metal-linker coordination and high connectivity allow them to incorporate linker and metal-cluster vacancies to a high degree.

Thiocyanate 'Prussian Blues'

Mon, 29 Oct 2018 11:00:00 +0000

Left: the structure of Fe[Bi(SCN)₆] determined using single crystal X-ray diffraction. Right: the optical absorption spectrum of M[Bi(SCN)₆], showing which wavelengths of light each compound absorbs.

Prussian Blue is an iconic material. Discovered around 1706, ostensibly by adding iron sulfate to a concotion of boiled animal blood and potassium carbonate, it was the first modern purely synthetic pigment. It made blue paint that wouldn’t fade available to artists at a low cost and perhaps most famously dyed the uniforms of the Royal Prussian Army from the early 18th century through to the 20th. Since then Prussian Blue and related compounds have been found to be useful for all kinds of applications, including electrochromics (materials that change colour when a voltage is applied to them), batteries, and even as magnets. Prussian Blue is also one of the most simple examples of a coordination polymer: a material where metal atoms are connected by multi-atom linkers into infinite chains or frameworks. In this case, iron atoms are connected by cyanide (CN^–) into a three dimensional cubic network, with the chemical formula (Fe³⁺₄[Fe²⁺(CN)₆]₃)· x H₂O.

Low-dimensional quantum magnetism in CuNCS₂

Thu, 26 Apr 2018 09:00:00 +0000

Structure and magnetic properties of copper(II) thiocyanate. (a) The structure of Cu(NCS)₂. The polymeric chains go into the screen. (b) The low temperature magnetic structure of Cu(NCS)₂.

In most magnetic materials the magnetic spins stop moving and freeze into an ordered array when cooled down. For some materials, for example iron, this happens at hundreds of degrees Celsius; for others, it can be a fraction of a degree above absolute zero (-273 ºC). However, in spin liquids, the spins never freeze due to the combination of their geometrical arrangement and interactions. These spin liquid phases have many strange properties, for example ‘fractionalisation’, where electrons in the crystal behave as if their fundamental properties can be separated into distinct quasiparticles, like ‘holons’, which have charge but no spin, and ‘spinons’, magnetic but uncharged quasiparticles.

Keep it simple, silicon

Wed, 14 Jun 2017 09:00:00 +0000

Left: structural model of amorphous silicon. Right: The electronic energy levels of amorphous silicon

Amorphous materials, that is, materials which don’t have long-range order, are critical for both windows (silica glass) and Windows (amorphous silicon for TFT screens). Despite their technological importance, we still do not fully understand how the structures of amorphous materials relate to their properties. The relative paucity of experimental information available from X-ray and neutron scattering experiments on amorphous materials (in comparison to crystals) makes determination of their structures challenging. Even once we know how the atoms are connected, it is not trivial to explain their resultant properties.

3D MOFs to 2D sheets in 1 easy step

Fri, 31 Mar 2017 09:00:00 +0000

Figure: Conversion of hcp UiO-67 into nanosheets of hxl UiO-67

In the past decade or so, people have rediscovered the joys of two dimensions – particularly because of remarkable electronic properties of graphene and other related materials. Very thin sheets of nanoporous materials (like zeolites, for example) are extremely interesting because of their potential as catalysts or membranes for chemical separations. Their thinness lets molecules diffuse rapidly through them, but because they retain the small and uniform pores, they can still discriminate between different kinds of guest molecule.

Cyanide and magnetism

Sun, 21 Feb 2016 18:50:00 +0000

Both chemists and physicists try to control interactions in materials to produce unusual and useful states of matter, but usually they make use of very different techniques and for very different kinds of materials. In this paper we show that chemists can learn a lot from physicists and hopefully vice versa!

We look at the surprisingly complicated structures of some simple metal cyanide compounds, and show that these structures can be explained by using an analogy to the physics of intrinsically disordered ‘frustrated’ magnets. By changing the chemistry in the material, we were therefore able to explore the physics of these frustrated magnets, including quite unusual magnetic states!

A breathing MOF with nano-correlations

Fri, 12 Feb 2016 09:00:00 +0000

This work was was result of a collaboration with a number of universities and facilities but particularly the group of Prof. Dirk De Vos at KU Leuven university and with Bart Bueken and Frederik Vermoortele.

One of the remarkable features of MOFs is their flexibility - they can show all kinds of unusual behaviour, from negative linear compressibilities to The most distinctive behaviour is perhaps breathing - guest triggered transitions between very two structures with very different porosities.

Disorder, design and vibrations

Mon, 08 Feb 2016 18:50:00 +0000

Disorder and defects are often thought to be a problem to be solved in materials science - for example, the presence defects and disorder greatly reduces electrical conductivity in crystals. In this paper we examine a particular kind of disordered crystals we termed ‘procrystals’ and showed that these procrystals have properties impossible to replicate in perfectly ordered crystals.

Structure of a MOF-derived molecular glass

Wed, 06 Jan 2016 18:50:00 +0000

This work was was result of a collaboration with Kyoto university, particularly with Dr. Daiki Umeyama and Prof. Satoshi Horike.

Most of the interest in metal-organic frameworks (MOFs) has focussed on those that are ordered and crystalline. Moreover, MOFs typically don’t melt when heated, instead decomposing, which means that they can be difficult to form into the shapes needed for useful applications. Glass, in contrast, is is incredibly useful because we can form it into very elaborate shapes without introducing any internal boundaries.

Defect-controlled MOF Mechanics

Fri, 06 Mar 2015 10:00:00 +0000

In our previous work we discovered that UiO-66, one of the canonical metal-organic frameworks (MOFs), can contain correlated defects, and showed that we have a degree of synthetic control over the presence and distribution of defects.

In this paper we use this control over defects to tune the physical properties of UiO-66. In particular, we look at the thermomechanical properties of UiO-66 - its response to changing temperature - examine how they change with different concentrations of defects. We also show that the behaviour of UiO-66 on heating is pretty unusual. First, it rapidly densifies when heated to about 300 ^oC (as loosely bound molecules are driven off). Second, this densified material exhibits ‘colossal’ isotropic negative thermal expansion - i.e. when heated it shrinks in every direction.

Flexibility in a ferroelastic MOF

Sat, 31 Jan 2015 18:50:00 +0000

One of most exciting features of metal-organic frameworks (MOFs) is their potential to be incredibly flexible. In this paper we report a new MOF, copper(I) tricyanomethanide, Cu(tcm) that illustrates this very well. Cu(tcm) undergoes a ferroelastic phase transition at T_f = 240 K and below this temperature shows thermal expansion an order-of-magnitude larger than above the transition. In fact, the low-temperature phase α-Cu(tcm) shows ‘colossal’ positive and negative thermal expansion that is the strongest ever reported for a framework material. We also found that Cu(tcm) can be converted just by exposure to gaseous acetonitrile to another new phase (acetonitrilo-copper(I) tricyanomethanide), and that this reaction can be reversed by putting copper tricyanomethanide under vacuum Infrared spectroscopy measurements are sensitive to the phase change, suggesting that Cu(tcm) might perhaps find application in solid-phase acetonitrile sensing.

MOF Defect Nanoregions

Sat, 31 Jan 2015 18:49:00 +0000

The properties of many conventional materials are intimately connected to the presence of correlated defects, including both the solid electrolytes in batteries and high temperature oxide superconductors. Although defects have been recently found to exist in MOFs, the prevailing understanding is that these defects are randomly distributed. In this paper we show for the first time that defects can be correlated in MOFs and that we can control this defects synthetically. This suggests that we might be able to create MOFs that show the same diversity of useful properties as in conventional materials.

Negative Area Compressibility

Sat, 31 Jan 2015 18:48:00 +0000

In this paper we report the first known example of a material that shows negative area compressibility , that is a material that expands in two perpendicular directions when compressed hydrostatically (i.e. equally in all directions).

This counterintutive property is the result of the layered structure of silver (I) tricyanomethanide, Ag(tcm). Due to the weak forces between the layers it is easy to compress the layers together. This in turn also flattens out the ripples within the layers, causing the crystal to expand in the other two directions.

Parameterising the INVERT approach

Sat, 31 Jan 2015 18:47:00 +0000

This paper gives more detail about the ‘INVERT’ approach and outlines both the challenges faced in trying to use data-driven methods for amorphous structural solution and some of the possible solutions to this problem.

Figure: Ordinarily, the problem solving the structure of amorphous materials from diffraction data is both underconstrained (there are too many solutions) and taxing to reach (the fitting landscape is poorly shaped). This is shown on the left and is compared to an idealised ’target’ landscape on the right.

Measuring Simplicity

Sat, 31 Jan 2015 18:46:00 +0000

Quantifying and understanding the structures of amorphous materials can be difficult, because they don’t have the long-rage periodicity we rely on to understand the structures of crystals. In this paper, we extend the ideas of our previous work by introducing the concept of ‘structural simplicity’ in the context of amorphous materials. We also show that structural simplicity can be quantified, and give two easily calculated geometric measures of simplicity. These methods could prove useful for both understanding and determining the structures of amorphous materials.

Accelerated aging for making porous MOFs

Sat, 31 Jan 2015 18:45:00 +0000

In our previous work, we showed that accelerated aging is a practical method for making metal-organic frameworks (MOFs) without needing bulk solvent or high temperatures from cheap and safe starting materials. In this paper we demonstrate that it is possible to directly make porous MOFs using accelerated aging. We also extend the chemistry to additional metals.

Figure: An example of a typical aging reaction on 10g scale with Canadian penny for scale.

A webtool for calculating compressibilities & thermal expansion

Sat, 31 Jan 2015 18:44:00 +0000

This paper describes PASCal, a simple webtool designed to calculate the principal thermal expansivities or compressibilities of a system from lattice parameters. For low symmetry crystals, the arbitrary choice of unit cell means that if you only look at the cell lengths, it is possible to miss interesting features in the (thermo)mechanical behaviour of a crystal, for example negative thermal expansion or negative compressibilities. PASCal is designed to be an easy method for calculating the principal strain values that give the complete description of these systems. The paper includes three case-studies where re-analysis using PASCal revealed some extra information about the response of the crystals to changing pressure or temperature.

Accelerated aging for making MOFs

Sat, 31 Jan 2015 18:43:00 +0000

Metal-organic frameworks have the potential to be materials that form a part of a greener economy. Laboratory syntheses of MOFs however use high temperatures, large amounts toxic solvents and relatively expensive soluble metal sources. If we are ever going to use these materials, therefore, we will need to develop new ways to synthesise them In this paper we report a new low-energy method of making MOFs, which we call ‘accelerated aging’. We show that just by mixing a metal oxide, the organic ligand, and small amount of an ammonium salt catalyst, and leaving the reactants in a warm humid environment for a few days, it is possible to make phase pure MOF. The success of these aging reactions also demonstrates the surprising reactivity of MOFs even without bulk solvent.

Amorphous structures from diffraction data?

Sat, 31 Jan 2015 18:42:00 +0000

Despite the importance of disordered materials, we still find it very difficult to determine their structures, because the lack of long-range prevents the use of conventional crystallographic techniques. This ’nanostructure problem’ as it has been termed, is one of the biggest challenges in the solid state sciences. In this paper we show that by making use of the idea of ‘invariant’ local environments in structure determination - in particular by minimising the variance in local data - it is possible to make significantly better models of amorphous and nanostructured materials.