Frequently Asked Questions

CrystoGen is a simulation tool that predicts how crystals grow at the molecular level using kinetic Monte Carlo methods. It analyses the free energy of each surface site, based on the coordination environment of a growth unit in the crystal lattice, to determine whether the site is favourable for growth or dissolution.

CrystoGen allows researchers to connect structure with morphology, enabling the rational design of crystallisation conditions and optimisation of solid forms. The software can simulate the morphology of a wide range of materials, including ionic, organic, and inorganic crystals, metal organic frameworks (MOFs), and co-crystals.

By adjusting simulation parameters to reflect real-world experimental conditions, such as supersaturation (the driving force for crystal growth), temperature, and the inclusion of crystal defects like screw dislocations, intergrowths, modifiers, or seed engineering. 

The free energies of crystallisation used to determine whether a surface site grows, or dissolves are obtained differently depending on your material type:

· For organic crystals, CrystoGen integrates Open Computational Chemistry (OCC), which uses DFT-based calculations to analyse intermolecular interactions and calculate the energy associated with each possible surface site. This gives a physically realistic input for modelling.

· For inorganic crystals or tiles, energies can be derived using third-party tools such as ToposPro, which helps define the structure and generate a network file from your CIF. Alternatively, users can input energy values obtained from their own DFT or force-field-based calculations, or from experimental data and published literature.

This flexibility allows you to either compute the energies from first principles or import known values to set up your simulation. 

  CrystoGen can simulate any single crystal for which you have a CIF (Crystallographic Information File). This includes:

· Ionic compounds (e.g., salts like NaCl)

· Inorganic crystals (e.g., metals, minerals)

· Organic molecules (e.g., pharmaceuticals)

· Co-crystals (with two or more components in the lattice)

· Metal-organic frameworks (MOFs)

· Tile structure (e.g. Zeolites)

Co-crystals can be modeled as long as the CIF correctly captures the multi-component structure. While polymer systems and amorphous solids aren't currently supported, single crystal growth of complex materials is possible within the limitations of your CIF.

 CrystoGen uses Monte Carlo methods to simulate crystal growth, treating it as a stochastic process where the probability of adding or removing a growth unit at a site is determined by the free energies of crystallisation. The core concept is that molecules or ions preferentially attach to low-energy sites that make the most favourable interactions with their neighbours.

For organic crystals, CrystoGen obtains the necessary pair-wise interaction energies from the Open Computational Chemistry (OCC) programme developed by Peter Spackman. OCC, which uses density functional theory (DFT), calculates these local pair-wise interactions between molecules in the crystal. CrystoGen then uses this information to compute the site energies for all possible coordination environments, which in turn drive the Monte Carlo simulation.

This approach makes CrystoGen particularly powerful for capturing the anisotropic nature of crystal growth and understanding how molecular packing, and intermolecular interactions influence the final crystal shape.

 The only essential input to get started is a CIF file or a tiles file of your crystal structure. Generate a structure and network file (usually done via OCC for organics or ToposPro for inorganics). Once configured, the simulation will output the predicted morphology and topography information.

You can obtain a CIF file (Crystallographic Information File) from several sources:

· Experimental methods: Using X-ray diffraction.

Databases: For known structures, CIFs can be downloaded from:

· CSD (Cambridge Structural Database)

· ICSD (Inorganic Crystal Structure Database)

· COD (Crystallography Open Database)

· CrystoGen Zeolite Structure Database. 

A standard CIF should contain accurate atomic coordinates, unit cell parameters, space group information and operations, which are used to inform the simulation. 

 This feature is currently under development. If you’re interested in projects relating to interface growth, please contact us at team@CrystoGen.org

At present, CrystoGen does not support polymer systems. The tool is designed for simulating single crystal growth. It can also handle co-crystals, provided the multi-component structure is well-defined in the CIF file.

These capabilities may be considered for future development depending on demand and feasibility.

 CrystoGen models layer-by-layer growth mechanism based on a stepwise Monte Carlo approach:

· Growth in CrystoGen starts from an existing nucleus or seed. Within the simulation, new layers form through 2D nucleation at the lowest-energy surface sites.

· Crystal surfaces emerge naturally as growth proceeds one unit at a time, with each addition or removal event determined by the site’s free energy.

· Over time, the simulation captures the anisotropic expansion of different crystal faces, allowing you to understand which surfaces dominate and why.

This mimics real crystallisation behaviour and can be tailored with different conditions to explore alternative morphologies. 

  CrystoGen can incorporate several types of defects and modifiers into your simulation:

· Screw dislocations

· Growth modifiers (e.g., additives that may preferentially adsorb onto certain faces)

· Diffusion-limited growth

· Nucleation from seed crystals

· Seed engineering (e.g., exposing a non-natural surface)

Some of these features, particularly complex defect types, are currently in beta testing and are available internally. Once validated across a broader range of materials, they will be released publicly in future software updates.

CrystoGen was originally developed by Professor Mike Anderson, a leading expert in materials chemistry at the University of Manchester, in collaboration with Dr Adam Hill, a technical specialist in computational chemistry at the same institution, and their research team. The software was first created to simulate the growth of zeolite crystals but has since evolved into a powerful, versatile tool capable of modelling the growth of many crystal types.

Dr Peter Spackman contributed to CrystoGen by developing the Open Computational Chemistry (OCC) and CrystalClear methods, which are used to calculate the interaction energies that drive the simulations. He is also the developer of CrystalExplorer (CE), a long-established tool for analysing crystal structures, which now supports complementary visualisation of CrystoGen outputs.

The software’s development has also been supported by numerous PhD and postdoctoral researchers. Notably, Mollie Truman lead the intergrowth feature and created the original graphical user interface (GUI), Alvin J. Walisinghe developed CG Aspects, CrystoGen’s 3D crystal visualisation and data analysis tool, and Alex Shield is currently enhancing the software by implementing diffusion capabilities. CrystoGen continues to grow and improve through the contributions of this dedicated and collaborative team.

CrystoGen supports a broad range of industries and research areas where crystal morphology and solid-form design are critical. It is widely used in pharmaceutical and agrochemical industries. In materials science, technology, and energy, it aids in the design of advanced materials, such as semiconductors, battery components, and perovskites. The food industry benefits from insights into crystallisation processes in products like sugars and fats, while the mining sector uses it to understand mineral growth and optimise extraction processes. CrystoGen is also a valuable tool in academic research, supporting crystallography, chemistry, and solid-state physics across universities and institutions worldwide.

There is no difference between the academic and industry versions of CrystoGen — we firmly believe that researchers should have free access to the same powerful tools to advance science. Supporting education and research is one of our core principles, and it remains at the heart of everything we do. Industry users, however, benefit from dedicated support and early access to new features as they are released.

CrystoGen can simulate a wide range of crystal morphologies, including needle-like, laths, plate-like, blocky and dendritic. It accurately predicts how morphology changes in response to different solvents, additives, and growth conditions, making it highly versatile for both academic studies and industrial formulation.

Yes, CrystoGen can simulate crystal growth in a wide range of solvents. It uses SMD (Solvation Model Density) solvent parameters to calculate the effect of implicit solvation on the intermolecular interactions (using OCC) and predict how different solvents influence growth rates and resulting crystal morphologies. The software currently includes over 180 solvents, and we are actively expanding the database to include ACS green solvents, supporting more sustainable solvent screening and selection.

 CrystoGen allows you to model the effect of impurities or growth modifiers by selecting highly populated sites or any other preferred site. Information about the different sites can be extracted from a “count” file generated during any simulation. You can choose specific sites to block based on facets, local coordination environments, or particular intermolecular interactions of interest. This means you can target sites that are more likely to interact with a specific additive or that are critical for the growth of certain crystal faces. The Monte Carlo model then simulates the presence of the growth modifier by blocking those sites according to the strength and frequency of attachment you specify. This enables detailed exploration of how additives influence crystal morphology and the relative growth rates of different crystal surfaces.

A Zingg diagram is a classification tool used to describe the shape of crystals based on their dimensions (length, width, and thickness). It categorises morphologies into four types: needle, plate, block, and laths. In CrystoGen, Zingg diagrams are used to automatically classify simulated crystal shapes, helping users quickly understand how different conditions, such as solvent choice or the presence of additives, affect overall crystal shape.

CrystoGen simulations have been validated against experimental data and show strong agreement with observed crystal morphologies. Validation is presented in three peer-reviewed papers available on our website, along with additional publications co-authored with industrial partners. While absolute growth rates can vary due to real-world complexities, the software consistently captures relative trends in shape and facet development, making it a reliable predictive tool for both academic and industrial use.

To obtain a licence for CrystoGen, simply reach out to us at team@CrystoGen.org. We offer flexible licencing options for industrial licensingusers. Our team will guide you through the licencing process, provide a quotation, and ensure that you have access to the appropriate features and support for your needs.

A CrystoGen licence includes the full Monte Carlo crystal growth software with features like variable supersaturation, temperature, screw dislocations, and growth modifiers. It comes with additional tools such as a visualisation package, data analysis software (CG Aspects), and ToposPro. You'll also receive sample input files, video tutorials, and access to ongoing software updates. The licence can be used on all operating systems, and we provide guidance and support to help you get started. We're always here to help; feel free to reach out to us by email at any time.

We continuously strive to improve CrystoGen to make it as adaptable and accurate as possible for all crystal types. New features go through thorough beta testing before being released to ensure reliability. Over the coming years, we are heavily investing in software development to expand capabilities and user experience. Upcoming features include nucleation, seed engineering, diffusion, and major enhancements to our visualisation and data analysis tool.

CrystoGen is unique because it is an emergent, unit-by-unit simulation tool, meaning crystals develop naturally during the simulation without pre-defining facets, growth mechanisms, or relying on experimental growth data.

It captures fine structural details such as screw dislocations, surface roughening, curved surfaces, intergrowths, co-crystals, solid solutions, and site-specific additives, features that most other models cannot handle. Its molecule-by-molecule growth approach also naturally models surface topography and crystal habit, allowing integration of experimental observations, such as AFM data, to refine predictions.

Because CrystoGen does not require pre-existing growth rate data, it is particularly powerful for screening new structures, solvents, or experimental conditions, while remaining versatile enough to explore the effects of additives, defects, or growth conditions in detail.

 Just reach out to us at team@CrystoGen.org, we are more than happy to help. We also have resources on our website and YouTube channel to help you!

 You don’t need any prior knowledge of crystallographic faces or face-indexing to use CrystoGen. The software automatically calculates growth rates on all symmetry-independent faces using the structural information you provide, so there’s no need to manually identify or label crystal planes.

A basic familiarity with Miller indices (hkl) can be useful for interpreting outputs and colouring your crystal, but it isn’t required. If you do want to explore faces and indexing in more detail, you can use complementary software such as CrystalExplorer or VESTA to model and visualise them alongside your CrystoGen results.

CrystoGen produces several output files in .XYZ, .vis, .txt, and .csv formats. These include:

· Visualisation files (e.g. .XYZ, .vis) contain Cartesian coordinates and structural data, allowing simulated crystals to be viewed in the CrystoGen Visualiser, CGAspects or third-party programs like Ovito.

· Data files (.txt) capture key results such as crystal size versus time, supersaturation levels, facet and site information, and all simulation parameters. They also include checkpoint files for restarting simulations.

· Log files (.csv) provide a running record of simulations and summaries of parameter variations.

From these outputs, users can create higher-level analyses using CG Aspects such as Zingg plots, correlation matrices, and solvent effect plots to study morphology and solvent interactions in detail.