QMT Features: October 2009
FT-IR for coatings analysis
A fast and accurate analysis of coated materials using infrared spectroscopy eliminates user variation.  By Joe Dorsheimer, applications scientist, and Mike Bradley, FT-IR product manager, Thermo Scientific Corporation

From food and beverage containers to linoleum floor covering, countless products across almost every industry must use materials with the correct composition and characteristics to successfully perform their required role. In order to achieve this, special coatings are often applied to the material during the production process.

The effects of these coatings vary considerably depending on the intended use of the material in question. For example, the inside of drinks cans are often coated to protect and preserve the taste of the contents, while linoleum floor coverings may be coated in order to protect the material from environmental damage such as scuffing, scratching and staining.

To fit their intended use, the formulation of these coatings must be carefully tailored during the development process. The coated material is put through rigorous testing throughout the formulation process in order to observe how the material wears, fades and ages. Once the coating has been optimized for its intended use, production can begin in a pilot plant before moving to full-scale manufacturing. However, a fast and accurate means of monitoring the coating must be developed for quality control and quality assurance purposes. Not only is the chemical composition of the coating vital to its success, the homogeneity and viscosity are further parameters which serve an important role in the protection and longevity of the coated material.

Traditional methods of analyzing the coatings typically require the coating to be stripped from an area of a sample to calculate the amount of coating on the surface. However, this approach presents a number of limitations including the issue of user variation during the sample stripping process and the fact that homogeneity cannot be assessed. Additionally, the physical stripping of a coating from a material means that some of the final product must be sacrificed.

Infrared spectroscopy provides a non-destructive analysis tool which can be used for many coated samples. Infrared spectroscopy has many unique advantages including simple sample handling, non-destructive analysis and considerable macro-building power to enable the development of Standard Operating Procedure (SOP). Using the examples of coated vinyl materials in the marine and automotive industries, this article examines the use of infrared spectroscopy as a tool for analyzing coatings.
Case study: marine and automotive upholstery
Vinyl upholstery is widely used in both the marine and automotive industries and is required to be hardwearing in order to withstand the effects of abrasive environments. Excessive UV light exposure is also an important factor to consider as both marine and automotive vehicles often spend long periods in open sunlight. Additionally, marine upholstery must also contend with highly corrosive saltwater, mildew and mold.

As a consequence, the vinyl materials used in these industries require a coating which can meet these demands. Therefore the coating must be exposed to various simulated environments during the development process by varying factors such as temperature, humidity and exposure to sunlight.

In order to overcome the limitations of traditional approaches, mid-infrared spectroscopy can be used successfully where a material is highly reflective. However, vinyl is not a reflective material and absorbs much of the mid-IR light. Instead, near-infrared spectroscopy can be employed which is more effective with non-reflective materials as it has greater spectral range coverage from 10,000 to 4,000cm-1, compared to mid-IR, which has a range of 4,000 to 400cm-1.

In order to test the ability of infrared spectroscopy for the analysis of coated vinyl samples, an experiment was set up using a Thermo Scientific Smart NIR integrating sphere accessory mounted into a Thermo Scientific Nicolet iS10 FTIR spectrometer equipped with NIR optics. The sphere accessory contains an internal gold flag for use as the background reference. In order to give a more representative average spectrum for the sample, a sample spinner was used to analyze a large piece of vinyl.

Spectra data was collected in approximately 30 seconds using the Thermo Scientific OMNIC spectroscopy software. A partial least squares (PLS) quantitative model was developed using the Thermo Scientific TQ Analyst chemometrics software package. A series of samples with various coat weights were analyzed and a calibration curve was constructed. Replicates of each coat weight were collected to show the repeatability and robustness of the method. The same vinyl substrate was used for each coat weight thus removing any variation from the vinyl material. 

The vinyl samples were prepared with coat weights of 5.1, 9.6, 11.9, and 17.3 g/m2, as determined from a primary (weight-based) method. Multiple spectra were collected from each sample to observe and account for measurement variation. Spectral differences were correlated via TQ Analyst to the variation in film thickness.

A spectrum from a coated vinyl sample is shown in Figure 1. The PLS analysis of the data focused on the C-H combination bands seen in the 4600-4200cm-1 region of the spectrum, highlighted in the figure. Because the same vinyl substrate was used, the differences in spectra can be attributed to the properties of the coating. In this case, the same coating was used, so the spectral changes relate directly to the coating thickness applied to the vinyl. Figure 2 shows the data processed using a second derivative algorithm to enhance the differences between the samples and remove any baseline effects.

Performing the PLS calibration resulted in the correlation curve seen in figure 3. The correlation coefficient is quite high (0.998) and the root-mean-square error of calibration (an approximate error bar for the calibration) was 0.25 over the thickness range of 5.1-17.3 g/m2. A larger set of calibration standards could improve the robustness and accuracy of the method. PLS could also model different vinyl substrates or different coatings, given a more extensive calibration set. However, the current data set clearly demonstrates the feasibility of such a measurement.

The Nicolet iS10 is primarily a laboratory instrument. The calibration can now be shifted to an industrial spectrometer, such as the Thermo Scientific Antaris II. The combination of laboratory-based instrumentation specific to the QC lab and manufacturing based instrumentation provides the flexibility required to develop many such analysis methods.

The development, formulation, quality control and quality assessment of coatings are critical to the overall performance of vinyl products – especially those used in harsh and demanding environments. The use of infrared spectroscopy delivers a fast, reliable and non-destructive means of analytical testing of coatings, which overcomes many of the limitations of traditional methods.
Infrared spectroscopy eliminates the issue of user variation by enabling SOP development. Costs associated with the use and disposal of solvents typically used with traditional approaches is also eliminated, while the increased speed and reliability of analysis adds substantial value and provides excellent return on investment. The ability to roll-out a laboratory-generated solution to the manufacturing floor provides enhanced flexibility and future capabilities to the facility.

Email: Mike.bradley@thermofisher.com
Email: Joe.dorsheimer@thermofisher.com


You can now view all QMT Magazine issues on your favourite tablet or smart phone.
Download the free Quality Manufacturing Today App from the Apple iTunes App Store or from QMT Magazine on Google Play.

Rob Tremain Photographer
Click above to see full page display and links to QMT articles.
TCT Inspex 2016 logo
Creaform logo
Aberlink logo
Vision 2016 ad
Control logo
Hexagon Logo 2
Mitutoyo logo
Bowers logo