Home / Engineering
Industrial crystallisation — from doctorate to plant floor.
Crystallisation is one of the few unit operations where product quality is decided at the micrometre scale, and where one wrong parameter costs you the yield, the filtration time and the whole batch. This is what we know properly.
Who you are talking to
A doctorate in industrial crystallisation, followed by working experience in the oil and gas industry. That is an unusual pairing: crystallisation is taught as an academic subject, but few people who know the theory have stood next to a crystalliser fouling shut at two in the morning.
The field is narrow in name and broad in practice. Crystallisation, precipitation and solution-phase transitions turn up everywhere — fertiliser production, battery materials, pharmaceuticals, water treatment, mineral processing, and every form of flow assurance. We cover most of that spectrum.
Recognise any of these?
These are the problems that make people call us. All of them are symptoms of a crystallisation step that is not under control.
- Crystals too small. Filtration takes three times as long as it should and drying falls behind.
- Product cakes. It clumps in the bag or silo after a few weeks and the customer complains.
- Wrong polymorph, or several at once. Batches behave differently and nobody can point to why.
- The crystalliser fouls. Encrustation on walls and cooling surfaces forces washouts and downtime.
- Yield too low. Too much product stays in the mother liquor and recycling builds up impurities.
- Batch-to-batch variation. Same recipe, different result. No one knows which variable is driving it.
- Scale-up failed. What worked in the lab behaves completely differently at production scale.
- Deposits in the process train. Scale, wax or salt precipitating where it should not.
The common thread is almost always the same: supersaturation is not controlled, it just happens. Nucleation and growth compete for the same material, and which one wins determines the entire particle size distribution downstream.
Areas
The core: crystallisation and precipitation
- Solubility curves, metastable zone width and supersaturation control
- Nucleation and crystal growth — kinetics, competition, control
- Cooling, evaporative, antisolvent and reactive crystallisation
- Seeding strategy: when, how much, what size
- Polymorphism, solvates and hydrates — control and characterisation
- Agglomeration, breakage and particle size distribution (PSD/CSD)
- Population balance modelling and MSMPR analysis
- Batch versus continuous operation — and when switching actually pays
- Downstream: filtration, washing, drying, caking and storage stability
- PAT and in-line measurement — FBRM, Raman, ATR-FTIR, turbidity
- Scale-up from lab to pilot to plant
Oil and gas — flow assurance
Most of flow assurance is crystallisation under a different name. Something precipitates somewhere you do not want it.
- Wax precipitation and pipeline deposition
- Asphaltene precipitation and stability assessment
- Gas hydrates — formation, inhibition, MEG and methanol systems
- Scaling — carbonate and sulphate deposits, inhibitor selection and dosing
- MEG regeneration and salt precipitation in the reclamation loop
- Produced water — precipitation, deposition and treatment
Adjacent industries
- Fertiliser and inorganic chemistry — nitrate and phosphate salts, urea, granulation
- Battery materials — cathode precursor co-precipitation, lithium carbonate and hydroxide
- Pharma and fine chemicals — API crystallisation, polymorph control, purity
- Water treatment — evaporation, zero liquid discharge, salt recovery
- Mineral and metals processing — leaching, precipitation, purification stages
- CO₂ capture — amine systems, salt precipitation and degradation products
What you actually get
| Deliverable | Content |
|---|---|
| Diagnosis | What is actually driving the problem, from operating data, samples and physics — not guesswork |
| Phase basis | Solubility curve, metastable zone and operating window for your system |
| Operating recipe | Cooling profile, seeding strategy, agitation and residence time — quantified |
| Model | Population balance or mass balance where the problem calls for it |
| Experimental programme | Which lab experiments will actually answer the question, and which are wasted |
| Scale-up basis | What scales, what does not, and where the risk sits |
| Second opinion | Independent review of a vendor proposal before you sign |
And the tool, if you need it
Consulting is the main service. But we are also developers, which gives us an option most consultancies do not have: when a review reveals the calculation lives in a spreadsheet inherited through three generations of staff, we can build the tool that does it properly.
- Calculation tools for solubility, mass and energy balance — with traceability and version control
- Simulation of cooling and dosing profiles
- Dashboards consolidating PAT and process data into one view
- Automated report generation from measurement data
- Small internal applications replacing repetitive manual work
This is an add-on, not a package you have to buy. Most engagements are consulting only.
Engagement and pricing
Bounded assignments — a diagnosis, a second opinion, an experimental programme — are fixed price agreed up front. Longer engagements run on an agreed day rate. We work in Norwegian and English.
Send a short description of the problem and we will tell you honestly whether we are the right fit. If we are not, we will say so immediately.
Questions
We do not have crystallisation in our process. Is this relevant?
Possibly more than you think. If you have deposits, precipitation, caking, unexpected solids in a liquid stream or filters blinding, you have crystallisation — you just call it something else.
Do you need to visit the plant?
Often we get far with operating data, samples and a couple of video calls. For problems involving geometry, agitation or encrustation, a visit is usually worth it.
Do you have your own laboratory?
No. We design the experimental programme and interpret the results, working with laboratories or using yours when experiments are needed.
What about confidentiality?
We sign an NDA before seeing anything. That is standard and we expect it.
Tell us what is not behaving.
Two paragraphs is enough. We reply with what we think is driving it and what it would take to find out for certain — usually within two business days.