"Formulators can get rid of trial and error"

A well-made coating system is time-consuming and requires many repetitive tasks. Because of predictive science, a lot of this work can be abandoned. In our interview, VLCI CEO Sander van Loon explained how this works and the role that high throughput technology plays in this

.
Van Loon is CEO of VLCI, a formula designer and research and development services company. He will explain the cost and process of optimizing coating formulations at the European Coatings Technology Forum.
can you explain the use of predictive science in the coatings industry?
van Loon: Predictive science is used to provide parameters for ingredients in order to find compatible ingredients with their composition. These parameters work in a way that matches the ingredients in the same area. This means that they can form stable things and improve the efficiency of all types of formulations; Solution, dispersion or emulsion.
if the ingredients are not in the same area of the parameters and they are not compatible, you will experience problems with stability, film forming, and other performance. For example, when you have ten ingredients and want to make sure that these values overlap at least some, you can use it so that each ingredient is "satisfactory" with the other. This provides the best performance. Formula, the lowest number of ingredients, the lowest concentration.
parameters do you use to achieve this?
Loon: Hansen Solubility Parameter (HSP) is the best, especially for coating formulations. These parameters can obtain all relevant ingredients such as fillers, resins, additives, etc. It is not limited to solubility, but also to dispersion, or, more generally, to the discovery of similarities.
HSP consists of three parameters: dispersion force, polarity and hydrogen bond. When actually determined, these center values carry spheres, where other components are compatible or incompatible when outside the sphere.
do you get these parameters?
Van Loon: HSP has two ways. You can calculate components based on their structure (SMILES). But it's not very precise, and it's not representative of polymers. The second method is to identify them in a practical way. Here you have a set of solvents, you know HSP, add unknown ingredients, and then you evaluate all these samples. Typically, you must evaluate 30 to 40 samples, from good to bad solubility, and then enter the rating into the HSPiP software for calculation. In this way, you can obtain HSPs and spheres of unknown components.
this is what you do in your company, so I hope the manufacturer or supplier of raw materials does not provide their data sheet to the HSP?
Loon: No, most of them don't at the moment. But I hope it changes soon! Our VLCI is a research and development services company that identifies and implements these parameters for the coatings industry, as well as the personal care, home or polymer industries. We do this for raw material suppliers, so they know more about their products and formula designers, so they can find matching ingredients to improve the stability of their formulations.
, you can predict the stability of the formula, but you don't make any predictions about the nature of the final product?
: Yes, stability, and efficacy, so in many ways attributes. For example, if you use anticorrosive pigment preparation without selecting the best dispersant, the maximum coating performance will not work. Therefore, matching ingredients with HSP can improve performance because you can better disperse or resolve your ingredients.
in addition to HSP, you also use the so-called HHLD-NAC. What is it?
: Yes, it represents hydrophoretic differences - net mean curvature, used for, for example, emulsion polymerization. You can characterize monomers and, depending on the parameters of the monomers, you can find suitable surfactants to emulsify them and manufacture your polymers. Basically, it works like an HSP. You can use parameters to calculate the correct composition of a particular lotion.
you mentioned earlier, you need to test a large number of samples to get data on an ingredient. What are you going to do with this?
Loon: This is where high-volume devices come in. In order to actually determine the parameters, each ingredient requires about 40 samples, sometimes 50 different samples. So, as you know, there are thousands or even hundreds of thousands of ingredients on the market. To get all these parameters, you need high throughput. Even in a paint formulation, you have about 10 different ingredients, which means you have to create hundreds of samples for this situation. But once you have an ingredient HSP, you don't have to do it anymore.
I think you now have a pretty large database of these values?
: There is a database, especially HSP, but we do all the measurements for our customers. This means that most of the work we do is confidential. Sometimes our customers publish values, but it's their decision, not ours.
sometimes we work for ourselves, and then we can post something on our website. However, there is a database available for the calculation of HSP values. This is a program we have worked with Special Chem and Professor Steven Abbott. We created an ingredient selector on the SpecialChem website that contains approximately tens of thousands of calculated HSP values. These are not as good as the actual HSP, but they can serve as a guide and teach people what they can do.
it is to use this data? Do you need special skills?
Loon: a bit complicated but feasible. Most of the time, it's just a different mindset and it takes a little time. But once you understand how it works, the formulater can start using it. In addition, we provide training and teach people how to use it.
through HSP, formulat designers can get rid of trial and error. Many formulaters still mix ingredients together to see what happens and then plan further. In some cases, they include experimental design methods that limit the difficulty of trial and error. But it's still just shooting in the dark.
this is very different from predictive science. Once you have HSP, you can predict matching ingredients and limit the number of samples required to develop a formula. It was the beginning of an effort, but then the fun began!
high-volume laboratories can be very expensive. Is this something small companies can afford?
Loon: You can make it as crazy as you want it to be, and it can bring in millions of euros in investment. However, if the budget is small, say 100,000 euros, you can still get a smaller high-volume, ready sample. But in most cases, outsourcing to a service company like ours makes more sense for smaller companies, so you only need to use it when you need it. Because for them, maybe after a week or a month of HT screening you'll have a lot of data available, HT will be useless for a long time, but we also work for big companies because they want to buy resources on a project basis.