The Sika Surface & Material Science Lab in Zurich, Switzerland, is dedicated to the following activities:

  • Material characterization using surface investigation techniques
  • Powder characterization
  • Electrochemical (corrosion related) investigations
  • Industrial forensics

Please allow us to guide you through the course of a typical surface related topic

Sika Solutions for Wind Blades
yes
The Case: Wind Turbine Blades

Wind Turbine Blades are complex composite structures which in many cases consist of glass fiber reinforced epoxy resin (GRE).

Within the production of wind blades, two profiled shells are joined by typically  using two component (2C) Polyurethane- or Epoxy-based blade bonding adhesive. Sika develops such products. During development preliminary testing and evaluation of the adhesive formulations is performed on standard GRE substrates with defined build-up and geometry, called “Lap Sheer Test Specimens”.

For application in the field the most promising candidates will also be tested on substrate used and provided by the customer. Tests on “Lap Sheer Test Specimens” provide not only figures regarding the force required for “breaking” the adhesion. In addition, the position and the morphology of the fracture contains valuable information regarding adhesion quality. The different imaging techniques employed in the Surface & Material Surface Lab combined with the surface related elemental and molecular information accessible with our instrumentation make this information available.

Sika Team: Surface and Material Science Lab
yes
Step 1: Discussion of the Investigation Topic

Face-to-face discussions with the specimens at hand facilitate the definition of an investigation strategy.

In the present case, the pretreatment of the standard GRE test specimens was modified from manual to automatic grinding. Corresponding lap shear specimens were tested using Sikadur® WTG-1280 LD being a 2 component epoxy based adhesive for blade bonding.

While the force determined in lap sheer testing is similar, the optical aspect of the fracture surfaces appears to be more homogeneous with the new pretreatment.

To figure out the cause for such a difference in appearance, the aim of this analysis should be, “Monitoring and Discrimination of Fracture Mechanisms on Lap Sheer Specimens after Different Mechanical Pretreatment of the Substrates.”

IR Microscope Measurement
yes
Step 2: Characterization of the Substrates

For substrate characterization infrared spectroscopy (IR), electron microscopy (SEM) and topography (WLI) were employed.

IR investigation revealed that the substrate ´s surfaces are contaminated with traces of organic compounds. Part of these originate from contact with unprotected hands (sweat) and from improper packing material.  

On the other hand, it's shown that while both the substrate´s organic binder and the adhesive binder consist of epoxy resin, differentiation by IR is possible, the
corresponding IR spectra being different.

Using SEM it can be concluded that the upper polymer layer of the original GRE sheet is removed by grinding independent from the procedure used. Both by manual and by automated grinding the resulting surface mainly represents a deranged glass fiber network with minor binder areas in between.

Topographic measurements indicate that the original substrate is very smooth, the grinded substrate surfaces being significantly rougher. The roughness of the manually grinded sample seems somewhat lower than that of the automated grinded one. However, for the automated grinded sample the reproducibility is higher, indicating a more homogeneous surface structure.

Original Substrate

Original Substrate

Manually Grinded

Manually Grinded

Automatically Grinded

Automatically Grinded

Overview of tested Lap Sheer Specimens
yes
Step 3: Characterization of the Lap Sheer Test Specimen Fractures

First screening of the fractures was performed by using optical microscopy (OM). On behalf of the GRE side fracture it is exemplarily shown below that the fracture morphology of the manually grinded specimen is less homogeneous, clearly affecting several GRE layers. The fracture of the automatically grinded specimen initially seems homogeneous and even. It takes higher magnifications to recognize morphologic differences within the fracture surface.

Image: Overview Investigated Specimens

Fracture of Manually Grinded GRE

Fracture of Manually Grinded GRE

Fracture of Automatically Grinded GRE

Fracture of Automatically Grinded GRE

By using SEM, material contrast not only between GRE glass fibers and epoxy but also between the GRE epoxy binder and the adhesive epoxy could be visualized.

Electron Microscopy in the S&MSL

Electron Microscopy in the S&MSL

Detail SEM Image of Fracture Surface

Detail SEM Image of Fracture Surface

In the more detailed SEM image of the fracture surface as seens above the different components, such as glass fibers (very bright), GRE epoxy (bright gray) and adhesive epoxy (dark gray) are visible. The material contrast between the two epoxy matrices is owed to different fillers and additives, as proven by EDS analysis (see spectra comparison from below).

EDS Spectra Comparison of the 2 Epoxy Types

EDS Spectra Comparison of the 2 Epoxy Types

By image mapping the fracture surfaces by SEM the fracture mechanism can be assessed over the whole fracture surfaces. The SEM images shown below correspond to the OM images presented further above.

Fracture of Manually Grinded GRE

Fracture of Manually Grinded GRE

Fracture of Automated Grinded GRE

Fracture of Automated Grinded GRE

The Sika Team
yes
Step 4: Discussion of the Results

The investigation clarifies multiple aspects regarding the fracture mechanism of lap sheer specimens, such as:

  • A considerable area of the bonding is performed on glass fiber strands exposed during grinding.
  • For all samples a 100 % cohesive fracture can be observed.
  • In most cases a combination of interface near adhesive and GRE fracture is present.
  • The GRE fracture is dependent on pretreatment.
  • Manual grinding, being less uniform, induces flaws in several GRE layers. Thus, the fracture plane can “jump” from layer to layer.
  • Automated grinding is more uniform but somewhat rougher. The amount of torn out fibers depends on the grinding plane.
  • There are indications that the joining of the adhesive on the rougher automated grinded surface leads to enhanced entrapment of air bubbles in the interface near the adhesive layer.

These and many more findings are comprehensively presented in an investigation report.