Exclusive Interview with Dr. Charles Hansen – Inventor of HSP

In a world where digitalization is rapidly penetrating every field and everything is data-driven, it is no surprise that R&D is transforming too. Even in a field like formulation, which is known for relying heavily on empirical methods and the good old “trial and error”, formulators & engineers are starting to adopt science-based modeling, predictive tools & apps.

Dr. Charles Hansen

When it comes to issues of solubility, diffusion, dispersion, compatibility…, Hansen Solubility Parameters (HSP) is one such powerful tool. Though HSP has been around for several years, there has been an explosion of interest in recent years for two reasons:

1. With growing interdisciplinary work, formulators are looking for tools that deliver rapid progress.
2. Having the theory packaged as software (HSPiP), eBook and datasets has made it a convenient tool for today’s formulator to reach optimal formulations.

Having worked closely on this topic and being an advocate for science-based formulation, I was only too thrilled to get the opportunity to interview the inventor of HSP himself – Dr. Charles Medom Hansen. It’s not every day that one gets to interview someone who has created a theory that revolutionized a sector and has a field that bears his name!

Dr Hansen, now in "active semi-retirement", is still pushing the boundaries of what is possible with HSP, and still trying out new ideas! One of the co-creators of HSPiP, Prof. Steven Abbott says, “He is still happy to wrestle with a difficult technical challenge and use his decades of experience to come up with ideas which, more often than not, work out well in practice.”

As he completes 80 years, it was a pleasure to learn more about his journey…

If you look back at your long, illustrious career, what has been your biggest achievement? Any surprises along the way?

I felt that the ability to reliably predict how mixtures of two non-solvents could dissolve a solute was a clue to something that required a greater mind than mine to fully explain and exploit. I sensed that something here was perhaps universally right. There was a potential for wide usage, but others would also have to take up the challenge and enhance understanding. Only three names come immediately to mind as those who did indeed do so.

Dr. Alan Beerbower (Esso) said he was just waiting for this approach to come along and proceeded to expand on it as a true pioneer just after my first relevant publications in 1967.

Then there was a long period of doubt whether or not it would even survive. Someone had to research it and teach it to those who should be using it. I had expected some academic would have immediately taken it to a higher level, but that did not happen.

Then along came Prof. Steven Abbott and Dr. Hiroshi Yamamoto, and the whole thing got pushed to a still higher level of capability and distribution with the HSPiP software, eBook and datasets that first appeared in 2008.



The growth of interest since then leads me to believe that the methodology will survive. A full understanding of just what is going on will ultimately evolve. I believe, with the above in mind, that my biggest achievement has been somehow to help establish a basis for the future development of the HSP concept. Recognizing that so much can be done with the HSP concept, the biggest surprise in all of this has been that it has taken so long to get this far.

How did the idea of Hansen Solubility Parameters (HSP) originate?

The development of HSP was triggered by plots of 12 polymers in 32 solvents using the two solubility parameters developed by Blanks and Prausnitz as an extension of the single Hildebrand solubility parameter. They used a dispersion parameter and a “polar” parameter but did not include alcohols since these were hydrogen bonded and hydrogen bonding was to be avoided. I calculated these for a number of alcohols. It turned out that a set of solvents that dissolved given polymers did not dissolve a second set of polymers, while those dissolving the second set did not dissolve the first polymer set.

HSP Sphere

I thought a third dimension might resolve the issue. Steel bars were mounted at regular intervals using acrylic sides so that magnets representing the solvents could be moved systematically in a trial and error procedure. I realized that the cohesion energy (density) involved breaking all the bonds upon evaporation and was quantified by the latent heat.

The Blanks and Prausnitz “polar” parameter was quantitatively split into:

• The HSP polar parameter (that correlates with the dipole moment), and
• The HSP hydrogen bonding parameter (found by subtracting the dispersion and polar energies from the total energy).

The trial and error procedure of moving the magnets showed that the good solvents clumped together for each polymer. Confirmation of the methodology was accomplished by predictably mixing two non-solvents to give a dissolving blend for an expanded number of solvents and polymers.

I would also like to add that the Hansen solubility parameter terminology originated with Dr. Alan Beerbower (Esso), who immediately recognized the usability of the three-dimensional solubility parameters and renamed them. His pioneering work in the area was instrumental in what I did later, such as tabulation of the HSP of about 1200 chemicals using his tables as reported in the CRC Hansen Solubility Parameter Handbook (1999, 2007).

There was very little academic interest in the concept until the appearance of the handbooks, and especially after the appearance of the HSPiP software.

How can today’s coatings formulator benefit from using the Hansen Solubility Parameters?

I can look back at many examples from my own formulation life. Things that worked for me 50+ years ago will still work well in 2018. For coatings, a key issue has been the rational optimization of solvent blends among other things to get good adhesion to a substrate (by ensuring solvent-substrate HSP similarity) and getting high-gloss coatings by ensuring that the more volatile part of the blend is a less good HSP match, leaving the better matched solvent towards the end, allowing the polymer to be as relaxed as possible and to flow as long as possible.



There were other formulating issues like the retained solvent that could be reduced by making sure the last solvent out was essentially linear in structure thus having relatively higher diffusion coefficients.

An additional key issue was getting the most out of the pigments. The HSP of the pigment relative to the HSP of the polymers and solvents in the coating can be critical. Many have difficulty thinking that a pigment can have a Solubility Parameter so recently we’ve started calling them “Similarity Parameters” if that makes people feel more comfortable.
As an example, I formulated a number of special conductive inks that contained mixtures of different kinds of carbon blacks. Success was obtained with a binder with good HSP overlap with the pigments that was supplied in a solvent that had no HSP similarity to the pigments whatever.

For adhesion, the trick is often getting some “bite” between the adhesive or coating and a polymer. Too much bite and the surface can be ruined, too little and adhesion is poor.

One can make a trial blend with two components that are of different HSP distance from the surface then rationally alter the ratio to be:

• A bit closer if the bite is insufficient, and
• A bit further if the bite is too much

Without HSP, it is not possible to do this science-based, systematic formulation work so efficiently.

What’s your view on the current regulatory environment? And how can HSP help attain the goals of the EU’s Chemical Directive (REACH)?

My own country, Denmark, was the first to recognize the harm that white spirits were doing to ordinary painters who were just going about an apparently safe job. The introduction of tough regulation was an opportunity to create a range of much safer and, incidentally, more effective solvents. And I helped found a very successful company producing safer cleaners based on HSP principles. Solvent usage remained significant since it was not always possible to find a satisfactory solution with a strictly water reducible product.

My disappointment with more recent regulations is that everything is getting labelled as bad. So increased regulation is tending to allow backsliding; if an objectively safer formulation is still labelled as bad, and happens to be more expensive, and then users will revert to a cheaper, less safe alternative.

When I created the HSPiP software package with Prof Steven Abbott, we included the ability to obtain “read across” of HSP values along with other key physical values. We had expected this to be a much-used capability but the obvious point that solubility properties in three dimensions (dispersion, polar, hydrogen-bonding) should be a key aspect of safety assessment continues to be largely ignored, in favor of logP, a far inferior mono-dimensional parameter.

But we remain confident that HSP’s time will come. For example, the recent bio-based solvent standard insists on there being an HSP value because what is the good of a bio-based solvent if it isn’t going to dissolve the solute of interest? The standard makes it clear that via HSP the prospective user can get a good idea of potential suitability. A systematic approach to substitution would seem to require including an HSP analysis.

Do HSP values also help predict compatibility between formulation ingredients? Can you give us an example?

The answer is a very obvious “Yes” and is inherent in the examples above. So, let’s take it up a level. I want a pigment and a polymer to be stable in a formulation so I must use a dispersant.

It might seem obvious to choose a dispersant that is a great match to the solvent. But this can be a disaster if the solvent stripped the dispersant from the pigment. So, the system has to be tuned delicately so that:

• The dispersant is on the edge of solubility (in HSP terms, near the radius of the solubility sphere), while
• The polymer is comfortably within its solubility zone

I was especially delighted to see some recent work from Toyota on sophisticated TaC ceramics where they explicitly followed this “borderline solubility” principle for their dispersant.

The differences in final ceramic properties between a solvent blend that followed the correct principles and the alternative blends were very clear in the published images and measured properties.

If you can optimize a complex ceramic blend via HSP, you can optimize just about any complex formulation!



Still another example is the requirement that an SBS polymer be only partially compatible in the asphalt used in road building. This improves performance at both higher and lower temperatures. The HSP for asphalt sources around the world was determined to allow formulation to this required partial compatibility in the final asphalt product. An analysis of which affinities are required will many times lead to an HSP solution although other factors may be important in whatever compromise may be required.

How do you see the future of formulation? How can formulators move towards science-based formulation vs. empirical trial and error?

Artificial intelligence is the way forward in this area as well as in many others. One needs reliable algorithms and larger amounts of reliable data on which to base them.

The demand for such data will increase. Formulators: you should be requesting the required data from your suppliers. Suppliers: you will reach a much larger customer base by expanding your data.

It would also be helpful if the procedures required for the use of data were easy to handle for a person reasonably skilled with the area.

What are your thoughts on the new HSP-based tool being developed jointly by SpecialChem, VLCI and Professor Abbott (The Science-based Formulation Group)?

I’m not a theoretician or a pure scientist. I’m a “what works” type of person with a chemical engineering background. 50 years ago, HSP seemed to me to work surprisingly well for a simple theory and I was delighted that so many others agreed with my view.

But fashions changed and HSP seemed a bit old hat and its use tended to tail off, especially in the academic community. What frustrated me was not that HSP was being used less, but that it wasn’t being replaced by something better – inevitably it was replaced by something worse such as LogP or “hydrophilic/hydrophobic”.

So when Prof Abbott, Dr. Yamamoto and I created the HSPiP package we confidently predicted that we would fairly rapidly saturate the market for HSP tools. I am happy to say that we were completely wrong. HSPiP has led to a big resurgence of HSP usage in both industry and academia, and all around the world.

It is this interest that has inspired companies like VLCI to promote science-based formulation, using HSP, with their customers. I’m truly delighted that SpecialChem, with its wide influence across the formulation world, has joined up with Prof Abbott and VLCI to promote the HSP approach, to provide the basic tools that any formulator can use, and an enormous number of HSP values to help the formulators choose wisely.



Thanks to the Science-based Formulation Group (SBFG), significant values will become available in much larger numbers than before. If a potential ingredient is not in the data source, it will not be found. It is therefore of special importance for those who wish their products to be used in the relevant contexts of coatings, cosmetics, adhesives and plastics to determine and make available HSP as well as other relevant data for their products and offer it to sources such as the SBFG group.

HSP at this scale has never been done before and only someone as big as SpecialChem has the resources to do this. Many of those values are estimates (based on the Y-MB engine in HSPiP) and we all look forward to the day when suppliers, encouraged by SpecialChem, provide accurate values for their customers.

For an 80-year-old to see his work take off in this way, at SpecialChem’s large scale, is truly gratifying, and I wish you great success. But what pleases me most is that there will be many formulators, like me in my younger days, who will try the “what works” approach, and find that they can reach their formulation goals much faster. It is a good feeling knowing that something you’ve helped to create makes a formulator’s life just that little bit easier and more productive.

Finally, I would like to add that knowing HSP methodology has enhanced my general understanding of the world around me, and I sincerely hope this will also be the case for all of those who have not yet had contact with the HSP way of thinking.

About Dr. Charles Hansen

Dr. Charles Medom Hansen is an American-born scientist with Danish ancestry who is now a Danish citizen living in the Copenhagen area.

After a degree (1961, Louisville) and Masters (1962, Wisconsin) in chemical engineering, Hansen worked for his Ph.D. at the Technical University of Denmark, initially on the problems of solvent retention in polymers, starting his lifetime interest in diffusion science.

However, the problems of predicting the compatibility of solvents with polymers took over, leading him to overcome the problems of the then-popular Hildebrand solubility parameter by dividing Hildebrand's single parameter into three components: dispersion, polar and hydrogen bonding. His 1967 doctoral thesis (Dr. techn.) provided the basis for what came to be called Hansen solubility parameters, or HSPs.

In addition to providing a theoretical framework, the thesis provided the first working set of HSP values for common solvents and for a range of commercial polymers, making it of immediate practical use to academia and industry.

His theories were put to immediate use during his 8 years (1968–76) at the R/D Center, PPG Industries in Pittsburgh where the universality of HSP became apparent as they were shown to work equally well for problems such as optimizing the compatibility of pigments within paint and ink formulations.

He went on to hold the following positions:

• Director of the Scandinavian Paint & Printing Ink Research Institute in Hørsholm, Denmark from 1976 to 1985
• Senior Scientist at Hempel Group from 1985 to 1987
• Senior Scientist at FORCE Technology, Copenhagen from 1988 to 2004

Upon retirement, he became, and remains, an independent consultant.

In addition to over 130 published papers and 8 patents (h-index 31), he authored Hansen Solubility Parameters – A User's Handbook in 1999 followed by an expanded 2nd Edition in 2007. With Abbott and Yamamoto, he authored the package of software, eBook, and datasets called Hansen Solubility Parameters in Practice, in 2008 which is currently in its 5th Edition.

About Science-based Formulation Group

SpecialChem, Van Loon Chemical Innovations (VLCI) and Professor Steven Abbott decided to join hands to promote the shift from empirical trial & error method to science-based predictive tools. And so, the Science-based Formulation Group (SBFG) was created, with the aim to accelerate innovation & product development. As SBFG’s first initiative, the focus currently is on the predictive power of Hansen Solubility Parameters (HSP).

Interested in contributing ideas?

Reach out to me at sreeparna.das@specialchem.com.

You are also welcome to join the Science-based Formulation Community to actively participate & share your ideas.