Materials

Materials Characterisation

Gnosys Global are experienced in utilising a number of characterisation techniques and applying them over a diverse range of projects. Characterisation methods can be used to identify materials and potential impurities, examine aging and the chemical changes that may lead to variation in performance, examine the stability of materials in different environments and many other applications.

Some of the methods utilised include:

Spectroscopy – in-house developed methods (such as TranSpec) and conventional spectroscopy (FTIR and Raman) have been used to assess drug stability under storage conditions, and to predict the failure of polymeric materials due to aging. Spectroscopic techniques can be combined with statistical methods to give powerful prediction and data mining tools (link to MVSA and whatever Henryk is providing?)

Microscopy – examination of materials using microscopic techniques gives valuable insight into the composition of materials which can in turn reveal information on the materials performance.

Thermal methods – the thermal and mechanical properties of materials can be examined in isolation or combination using a variety of methods, including TGA, DSC, DMA and TMA.

Experimental Design

Experimental design is the process of planning a study to meet specified objectives.  Planning an experiment properly is very important in order to ensure that the right type of data and a sufficient sample size are available to answer the research questions of interest as clearly and efficiently as possible.

In an experiment, we deliberately change one or more process variables (or factors) in order to observe the effect the changes have on one or more response variables. The statistical design of experiments (DOE) is an efficient procedure for planning experiments so that the data obtained can be analyzed to yield valid and objective conclusions.

DOE is intimately linked with Multivariate statistical analysis  - aka Chemometrics.  This is the science of extracting information from chemical systems by data-driven means. It is a discipline that links many areas of scientific endeavour, using methods frequently employed in core data-analytic areas such as multivariate statistics, applied mathematics, and computer-science, but used to investigate and address problems in chemistry, biochemistry and chemical engineering. It

1. connects analytical data to useful information, such as infrared spectroscopy to voltage breakdown in cable insulation
2. Explores scientific hypotheses
3. Predicts things that are hard to measure using data that is easier to obtain, such as replacing a tedious and error-prone engine test for octane number with a very quick measure of the near infrared spectrum of the fluid.

Nanomaterials

The incorporation of nanomaterials into a polymer matrix is not a new area of science, however, new applications for nanomaterials are constantly being imagined, based on the enhancements addition of nanomaterials can lend to a composite. An important aspect of nanotechnology is the vastly increased ratio of surface area to volume present in nanoscale materials. This can lead nanocomposite materials to have vastly different properties from the virgin matrix material alone.

The uniformity and dispersion of nanomaterials in the host matrix can also have an impact on the properties of a the composite. In some cases poor dispersion may lead to a detrimental effect on properties and generate inconsistent results.

Gnosys Global have experience in generating new nanocomposites and optimising them to meet the requirements of various clients. This includes sourcing and quality assuring nanomaterials, i.e. ensuring that batch to batch variations, which can be an issue with nanomaterials, are minimised. Gnosys Global are also experienced in dispersing nanomaterials within matrices using various techniques and in ensuring good interfacial chemistry, first by understanding and choosing the appropriate materials and also through surface modification were required.

Characterisation of nanocomposites is also an important factor and Gnosys’ methods have been successfully used to examine the purity of the nanomaterials, the interfaces between nanomaterial and polymer, assess dispersion and uniformity and to quantify the changes in materials performance.

Analytical Data Mining

Many types of data mining methods can be used for various applications, including the analysis of materials: classification methods including decision trees, k-nearest neighbour (k-NN) and artificial neural networks (ANN); clustering methods including principal component analysis (PCA) and extensions of this such as soft independent modelling of class analogies (SIMCA), and regression methods including principal component regression (PCR) and partial least squares (PLS).

For spectroscopic data, multivariate statistical analysis methods are used. Specifically, for qualitative discrimination of materials and their properties a SIMCA approach is used, which allows us to quickly and efficiently identify a material or one of its components by comparing with all or part of a calibration database. For quantitative calibration such as the estimation of additive concentrations and other measureable physical properties, regression methods such as PCR or PLS are used; the specific method applied depends on the quality of the calibration data available.

Our TransChem software application provides a user-friendly interface that facilitates the input of calibration or validation data, suitable data processing, development of calibration models and fast on-line prediction of materials and their properties by applying these models to spectroscopic measurements.