Molecular modelling prediction of physico-chemical properties of products – PREDIMOL
The PREDIMOL project has focused in particular on amines and organic peroxides using both types of predictive methods: Quantitative Structure-Property Relationships (QSPR) for the prediction of hazardous physico-chemical properties and molecular simulation (molecular dynamics and method of Monte Carlo) for equilibrium properties (e.g. vapour pressure) and transport properties (e.g. viscosity). From the gathering and critical analysis of existing predictive models dedicated to physico-chemical properties and inventory of available data for both studied families of substances, a new experimental database was built from tests performed by INERIS and Arkema considering 38 organic peroxides and 5 properties. Based on these data and quantum chemical descriptors allowing characterizing the reactivity of organic peroxides, two QSPR models were developed for the prediction of thermal stability and a specific force field for organic peroxides was optimized for the calculation of thermophysical properties. The MedeA® simulation platform designed by Materials Design was extended with new functionalities. The automated calculation allows now to perform thousands of property predictions in a shorter time than that of typical experimental approaches.
The project has demonstrated that molecular modeling is a pertinent alternative to experimental measures to provide reliable and fast predictions of the physico-chemical properties of chemicals using only the molecular structure as input. Based on the collection and critical analysis of existing models for physico-chemical properties and on a new experimental database (for organic peroxides), predictive models were developed to determine properties of organic peroxides and amines as required by REACH.
-
The project has produced many international communications dedicated to the QSPR models developed for the hazardous properties of organic peroxides or for the development of high-throughput simulations. One thesis was defended, 3 scientific papers have already been published and others have been submitted. In particular, the evaluation of all existing models for the calculation of physico-chemical properties required by REACH, was performed from the analysis of 700 references and is planned for publication in Chemical Reviews.
The new EU regulation REACH (for “Registration, Evaluation, Authorization and Restriction of Chemicals”) has entered in its product-registration phase after December 2008. It requires the evaluation of the physico-chemical, toxicological and ecotoxicological properties of more than 143,000 substances which have been pre-registered by 65,000 companies in order to allow their use before 2018. More precisely, the reduction of the environmental and economic impact of the chemical products used in industry is based on a systematic and reliable knowledge of their physico-chemical properties. Considering the huge number of chemical compounds used nowadays in industry, their modifications and the introduction of new products on the one hand, and the reduced number of laboratories (properly equipped and certified) for the determination of the physico-chemical properties on the other hand, a lack of knowledge of these properties is foreseeable. It is of great urgency that private and public research develops fast and reliable methods for property-screening, which would benefit from the transfer of methods from the private sector (automation and production of large volumes of data). Moreover, all new substances will require an early examination to identify possible hazards. It is necessary to anticipate the risks and to recognize the dangerous substances as soon as possible in their development. In this economic, social and regulatory context, it is necessary to find reliable solutions to this problem.
Taking into account the number of properties, the number of substances, the timing, the economic costs, the feasibility at the R&D level, the ethics (tests on animals), the risks for the manipulator, in particular for the characterization of the dangerous physico-chemical properties (explosibility, flammability), the experimental measurement of all the data is not realistic. Thus, the development of alternative predictive methods for the evaluation of the properties of chemical substances is recommended in the framework of REACH and the associated system of classification, the CLP (for “Classification, Labelling and Packaging off chemical substances and mixtures”). Molecular Modelling can play a critical role in this context. Indeed, for several decades, modelling and molecular simulation in all forms (in particular quantum chemistry, molecular dynamics and Monte Carlo methods with empirical force-fields, and empirical correlations) have progressed enormously. In fact, this development is fuelled by the revolution in the power and availability of computers. During the last 25 years, compute power in all its dimensions has enormously increased by several orders of magnitude: speed of operations, random access memory, disc storage capacity, connectivity, parallelism, and power consumption. Similarly, the price for the same functionality decreased from several million Euros to a few hundred Euros, therefore by a factor of 10,000! It is imperative to seize this opportunity in the context of the systematic, effective, and reliable prediction of the physico-chemical properties of chemical compounds. This is the context and scope of the PREDIMOL project, where molecular modelling represents an interesting way of development, which is not concurrent but complementary to the experimental characterization, in order to predict in a reliable and fast manner, the whole range of physico-chemical properties of substances required by EU-REACH's regulation (annexes VII and IX). Thus, PREDIMOL project is an industrial research project aiming at answering the research effort and innovation related to the implementation of REACH.
Project coordination
Patricia ROTUREAU (INSTITUT NATIONAL DE L'ENVIRONNEMENT INDUSTRIEL ET DES RISQUES)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
Partner
Arkema ARKEMA France
ENSCP CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE - DELEGATION REGIONALE ILE-DE-FRANCE SECTEUR EST
LCP CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE - DELEGATION REGIONALE ILE-DE-FRANCE SECTEUR SUD
INERIS INSTITUT NATIONAL DE L'ENVIRONNEMENT INDUSTRIEL ET DES RISQUES
IFP INSTITUT FRANCAIS DU PETROLE
Help of the ANR 952,386 euros
Beginning and duration of the scientific project:
- 36 Months