JEC COMPOSITES MAGAZINE - Issue #87 - March 2014 Special JEC Europe - 126

analysis

as well as to the nozzle height. The
adhesive peel strength G1c values
were measured on selected sections of
the coupons that had been exposed to
the different nozzle heights. For the
temperature flux evaluation, a number
of nozzle position temperatures were
independently measured as a function
of the distance from the nozzle tip to
the surface of the CRFP using a thermal camera with recording sensors
measuring time and temperature.

Results and discussion
The spectra of release agent-impregnated CFRP, recorded when the
plasma nozzle was at minimum and
maximum distance from the coupon
surface, exhibit spectral changes
(figure 3). For both the the PDMS
and Wax treated composite, there are
indications of thermal damage when
the nozzle is closest to the surface
(highest temperature condition).
This is demonstrated by the increased
intensity of the carbonyl peak at ca.
1720 cm-1, which is highly indicative
of thermal oxidative damage. Also,
there is a noticeable drop in the O-H
stretching vibration at 3300 cm-1
in both the PDMS and Wax based
treated samples. Though the PDMS
based release agent exhibits some differences in the Si-O band at approxi-

Fig. 4: PLS multivariate analysis of FTIR spectra as a function of position (i.e., nozzle gap/
height, and therefore temperature) for PDMS
release agent on CFRP coupon

mately 1100cm-1, the position and
shape of this band makes univariate
quantitation difficult.
Multivariate partial least squares
was used to analyze the relationship
between measurement position on the
coupon (which in turn relates directly
to the nozzle gap and also to the
temperature flux to which the surface
is exposed) and changes in the spectra. To develop this model, 560 data
points per spectra for the regions 7801850 cm-1 and the 2715-3700 cm-1
were used together with the appropriate pre-processing conditions and
transformations.
The final cross-validated model used

Fig. 3: Infrared spectra of PDMS (top) and Wax (bottom) release agent on
CFRP. Prior to treatment, the PDMS and Wax formulations have similar
spectra, which indicate that they contain similar functional groups. The
most significant difference is the presence of the Si-O group evident as the
raised shoulder near 1100cm-1.

126 jec composites magazine / No87 March 2014

only three factors and the actual vs.
predicted plot (figure 4) for PDMS
release agent, as an example, demonstrates very good agreement and the
model is able to predict within +/- 1
cm of the nozzle position (and temperature flux) on the coupon.
The results from the independent
adhesive peel strength measurements
clearly demonstrate where plasma
under-treatment and over-treatment
of the release agent occurs. The
predicted values derived from the
PLS model can then be related to the
peel strength results (figure 5). Using
the conditional reporting feature of
the 4300 Handheld FTIR MicroLab
software, areas of improper plasma
treatment, as determined by the midinfrared spectra, can be reported to
the user in terms of visual color and
text alerts.
The various temperatures to which
the CFRP is exposed results in surface
chemistry changes that are directly
related to the duration and nozzle
position. The results of FTIR analysis
can detect:
- under-treatment (large nozzle
height), which results in adhesive fai-

Fig. 5: FTIR spectra indicate effectiveness of plasma treatment of CFRP.
The Actual vs. Predicted values for FTIR spectra are zoned by the adhesive
peel test G1c values. Optimal treatment range (based on G1c values) is
shown in green; under-treatment and over-treatment are shown in red;
and a marginal region (either over or under optimal) is shown in yellow

JEC COMPOSITES MAGAZINE - Issue #87 - March 2014 Special JEC Europe

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #87 - March 2014 Special JEC Europe