A shorter route

Partnership between Unicamp and Rhodia finds new solutions for the company's industrial processes

eduardo cesarAt the Rhodia laboratory in Paulínia, the standard list of solvents for testing fell from ninety to six or seven typeseduardo cesar

A new production process in the chemical industry almost always begins with the choice of the most appropriate solvent for the formulation of a new medicine, a quicker drying paint or even a shampoo that helps to untangle hair. By selecting the appropriate solvent, which is nothing more than a liquid such as acetone, benzene, ethyl alcohol and many others, it is possible to separate one ingredient from a complex mixture of substances. Even if the final product should be a solid, the industrial manufacturing processes in their initial stages generally start off with liquid bases. In traditional laboratory tests, the experimenting phase to identify the solvents best suited for a new product, or even to modify the product recipe that can already be found in the market, can take various days or even weeks. Now a software program developed in a partnership between Rhodia Brasil and the Chemical Engineering Faculty (FEQ) of the State University of Campinas (Unicamp) allows for making this process quicker, with greater economy of time.

Rapid reply
“The tool has brought a huge benefit to us, which is the speed of reply to our clients’ problems and agility in the development of new productivity processes”, says  Richard Macret, the company’s Latin America research and development director. The company is a subsidiary of the Rhodia Group, based in France and that has industrial units in almost one hundred countries, and which is focused on fine chemicals, fibers and polymers. Previously, Rhodia had already developed other software in order to look for the best mixture of solvents for polymers. But this could not forecast the solubility – the maximum quantity of solid that can be dissolved in a solvent at a particular temperature – of a large class of products, which are the crystalline solids. In this chemical category, so called because of their structural arrangement, which always follows a regular standard of distribution, one can find, for example, the main active ingredients of medicines.

Before beginning to make use of the new software, the result of a research project funded by FAPESP through the Partnership for Technological Innovation (PITE) program, Rhodia Brasil had been working with a standard list of around ninety solvents, both to produce their products and to attend to clients’ requests. “They carried out experiments using each one of them, a slow and expensive process and not very appropriate”, says professor Martín Aznar, the project’s coordinator. “We developed a tool that allowed for restricting this list to only six or seven solvents, but which even at that is representative”, he adds. In order to create the software, the researchers took as a base a theoretical model, called Hansen’s Methodology, used in the chemical industry since the decade of the 70’s for formulating products with polymers, which are made from plastics, rubbers and/or silicon, and organic solvents, especially paints. But this method had still not been used for crystalline solids.

Equilibrium phase
The model used to determine the solubility values uses three parameters, called the dispersion force, polar force and hydrogen bonding. These three variables act during the interaction of the reagent molecules with those of the solid. By making use of this methodology it is possible to identify the solvents or mixtures of solvents that are most appropriate for dissolving polymers. The researchers extended the application to crystalline solids. And then went beyond. “Before our studies, the parameters of solubility were determined starting from quantitative information”, says professor Aznar. This type of information is obtained by the evaluation of the equilibrium phases between solid and liquid. To understand how this functions, one only has to resort to a simple image: the dissolving point of sugar in water. For a determined volume and temperature, the water is capable of dissolving a certain amount of sugar. But when the sugar begins to be deposited at the bottom of the receptacle, the solution is saturated, or that is to say, reaches a point of equilibrium. In order to attain the desired result using this method, it is necessary to make use of a large quantity of reagents.

One of the changes proposed by the project is the use of qualitative data. For this method, it is enough to take a simple test tube, to place in it a standard volume of pure solvent, stipulated at 0.9 grams per 0.1 grams of solid, and let it be shaken for twenty four (24) hours by a machine that keeps the mixture in the vertical position. In this manner, both the solid and the solvent are in contact all of the time. In order to make the test, forty seven flasks with different solvents are used. “It’s a faster and cheaper method than the quantitative method, because you only have to look to see if the solid has dissolved or not”, says Marlus Pinheiro Rolemberg, who also participated in the research through a post-doctorate grant.

In the FEQ laboratory, at Unicamp, because of the size of the shaker, the flasks used are traditional standard test tubes. But at Rhodia Brasil the glass receptacles are minute and do not reach 5 cm in length. In this manner, there is a decrease in reagent costs for carrying out the solubility tests and in the response time. With the qualitative information and the Hansen Methodology, the program developed at Unicamp shows which of the solvents is best indicated for the work with crystalline solids and also with polymers. The most appropriate appear on the computer’s screen.

The research that led to the development of the software was done in partnership with the Paulínia Research Center, one of five belonging to the French Group located throughout the world, responsible for the development of applications and of new products and processes. A further two are installed at Lyon and Aubervilliers in France, there is one at Cranbury in the United States and the final one is located at Shanghai in China. They work in an interlinked manner, as a network. “When there’s a solvent problem in France, in the United States or in China, they check us out to know what’s the recommended solvent for a determined application”, says Macret.

At the end of last year, the software was used to provide a solution for a problem detected in a product developed at one of the French research centers. By reason of industrial secrecy, the researchers only said that they had been dealing with a catalyst (a substance that alters the speed of a chemical reaction), which, when tested on a pilot scale in the laboratory, had been reacting with a solvent. “They wanted to have options of other solvents”, related Dr. Rolemberg. “We carried out tests and pointed out which were the most appropriate for the compound that they had been working with.”

Besides manufacturing solvents, the company also attends to clients’ requests that necessitate changing something in the formulae in order to attend to environmental problems, legislation or even because the solvent used reacts with the product. “The demand for a change of solvent that is more efficient has grown both in Europe and the United States and is coming to Brazil”, says director Macret. “Our software gives an indication of which solvents could be used in an exchange. Instead of testing six, we go directly to two or three.” The tool is treated by Rhodia Brasil as an important assistant because it helps with projects developed in the laboratory during the initial stages of the industrial process. “It allows for entering at an angle that other software programs do not manage”, adds Macret. For the researchers, the path to arriving at the software and the application of the Hansen Methodology for crystalline solids has resulted in scientific articles that are in their preparatory stage.

Ether spray
The Rhodia Brasil research center at Paulínia, a town close to the city of Campinas, is located on an old farm purchased by the company in 1944 in order to plant sugarcane and to produce alcohol. It was at this center that a new line of products for treating leather was developed, as well as studies in the use of polymers in shampoos that  have the function of straightening out very curly hair. Of French origin, the company set itself up in Brazil in December of 1919 to manufacture ethyl chloride, the main component of lança-perfume (an ether spray), a mild stimulant that was a sales success for the company in the country many years ago during Carnival. The first factory was installed at the town of Santo André, a suburb of São Paulo, producing chemical and pharmaceutical products. As time went by, it extended its product line and began to work on the development of synthetic textile fibers. The first one was a polyamide, used in the production of socks and swimsuits, which started to be used for other purposes, such as the production of car tires.

The Brazilian company employs around 3,000 people and during 2003 had a gross income of around R$ 1.8 billion. Throughout the world, the Rhodia Group has 23,000 employees and had a gross income of about 5.4 billion Euros during 2003. The annual investments in research and development that include the research centers and laboratories spread through out the world, run at about € 180 million. In Brazil this spending is around R$ 30 million per year. The investments in this sector also include direct partnerships with universities. For professor Aznar, this interaction is important and useful to both sides. “The company manages to solve a practical problem by making use of academic knowledge and the university can apply its knowledge and generate a valuable result for society.”

The Project
Prediction on solubility of polar polymers and crystalline solids in solvent mixtures (nº 01/11123-9); Modality Partnership for Technological  Innovation (PITE) Program; Coordinator Martín Aznar – Chemical Engineering College of Unicamp; Investment R$ 50,505.00 (Rhodia) and R$ 15,997.00 (FAPESP)