461251 Application of Systematic Methodology for Design of Tailor-Made Blended Products: Lubricant Design

Tuesday, November 15, 2016: 9:10 AM
Union Square 3 & 4 (Hilton San Francisco Union Square)
Marina Fedorova, Department of Chemical and Biochemical Engineering, Technical University of Denmark, Kgs. Lyngby, Denmark, Laurent Grosset, TOTAL MARKETING SERVICES, Solaize, France, Zhou Fang, Department of Chemical Engineering, Tsinghua University, Beijing, China and Rafiqul Gani, Department of Chemical and Biochemical Engineering, Technical University of Denmark, DK-2800 Lyngby, Denmark

Application of systematic methodology for design of tailor-made blended products: Lubricant design.

Marina Fedorova1, Laurent Grosset2, Fang Zhou3, Rafiqul Gani1

1 Department of Chemical and Biochemical Engineering, Technical University of Denmark, Lyngby, Denmark

2 TOTAL Marketing&Services, Solaize, France

3 Tsinghua University, Beijung, China

Over past decades, chemical product engineering has received much attention among the chemical engineering community. This is due to the transformation of industries in manufacturing and selling chemical products based on the product performance rather than compositional specifications.

Traditional methods used in new product design/development combine a broad knowledge of existing products with scientific experimentation.  The chemical product is developed based on scientific hypothesis, intuition or simple trial-and-error. However, this approach is costly, very time consuming and only a limited number of experiments can be performed. In order to efficiently design chemical products, a systematic methodology followed by focused experimental verification is a better approach. It can be implemented at the initial stage of the design, where it could minimize the required number of experiments, as well as to identify possible solutions that normally would lie outside the scope of traditional approaches, thus increasing the chances of finding better solutions with less resources.

The systematic methodology for solving mixture/blend design problem consists of four main tasks: 1) problem definition – identification of the final product needs and translations of them into target properties; 2) property models identification; 3) mixture/blend design; 4) experimental or model-based verification. Additional tools and methods needed to perform a specific task in the methodology, which are the property model library, the chemicals database and the mixture/blend design algorithm, also need to be available.

Application of the methodology is highlighted with several case studies related to lubricant design. Lubricants are required in almost all modern machines. Lubricating oil is a substance introduced between two moving surfaces to reduce the friction between them, improving efficiency and reducing wear. The functions of the lubricant depending on application may also include cooling, cleaning and suspending, protection and transfer power.

The application of lubricant is classified into two types – engine lubricants and non-engine lubricants. In this work, the lubricants are designed as engine oils, which is the largest application of lubricants. Typically lubricants are a mixture of base oil (65-98%) and performance additives to achieve required performance and end-user requirements. The target properties for the lubricant blend can be divided into performance properties, long life properties and environmental properties.

Focus of this work is performance properties, as they are central for any lubricant application. These properties include, for example, viscosity, thermal properties, low and high temperature properties, friction coefficient and film thickness that lubricant provides. The first part of the case study is to design the base oil, which is the main component in lubricant. The second part focuses on the development of the specific lubricant additive - its properties are thoroughly studied and the database of the chemicals is created to be used in the mixture design. Currently this database includes 1202 items. Novel blends with besigned functional properties are found in all cases. The method is generic and also be applied for other blended products.

Extended Abstract: File Not Uploaded
See more of this Session: Physical Properties for Chemical Process and Product Design
See more of this Group/Topical: Process Development Division