Polymer Testing, Polymer Analysis, Plastic Testing, and Plastic Analysis by ExcelPlas
Call Us: 03 9532 2207

40-50

Reinforced polypropylene – Water pipes

Title

Analysis of PP-R Green Pipe

Date

2013

Objective

To analyse a sample of PP-R green pipe to determine the cause of cracking failure

Photo

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image034.png

Figure 1. Position of leak on outside of pipe

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image036.png

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image038.png

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image040.png

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image042.png

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image044.png

Figure 2a, 2b, 2c, 2d, 2e. Photomicroscopy of inside of pipe

Testing Undertaken

OIT as per ASTM D3895

Digital microscopy

FTIR as per ASTM F2102

NMR spectroscopy

Failure Analysis

Antioxidants in the pipe were almost totally depleted (only Irganox 1010 and Irfagos168 were present), thus the pipe was not fit for service. The severe oxidative cracking of the pipe was due to a combination of thermal oxidation and stresses associated with thermal expansion and contraction

 

Rubber – Bellows

Title

FAILURE ANALYSIS OF RUBBER BELLOWS

Date

2012

Objective

To investigate the cause of cracking of rubber bellows.

Photo

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image096.png

 

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image098.png

Figure 1a and 1b. Photomicroscopy showing crack in fold of bellows

Testing Undertaken

Photomicroscopy

IR-Spec

Failure Analysis

The rubber bellows failed by cracking due to ingress of lithium ions into the nitrile rubber molecular structure.

 

Rubber – Floorings

Title

DISCOLORATION OF RUBBER FLOORING

Date

2013

Objective

To investigate the discoloration of rubber flooring and in particular to determine the mechanism of yellowing and the root cause of the yellowing. The flooring was installed in various shops.

Photo

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image002.png

Figure 1. Sample comparison after UV exposure to non-discoloured sample.

Testing Undertaken

IR Spec analysis (ASTM E1252 – 98) on the exposed surface of the rubber flooring

Accelerated UV exposure testing (QUV) (ASTM D7238 – 06)

Failure Analysis

Yellowing is due to photodegradation of the urethane, which containedchromophores (most likely from the isocyanate precursor) which interact with the UV light leading to poor light stability and photoyellowing.

Using chromophore-free isocyanates and/or including UV absorbers would prevent yellowing.

 

Rubber – Mouldings

Title

IDENTIFICATION OF A BLACK MOULDED ITEM

Date

2013

Objective

To identify the material from which black item was moulded

Testing Undertaken

Fourier Transform Infrared (FT-IR) according to ASTM E 573, using an Alpha Measurement Module

Digital Scanning Calorimetry (DSC) according to ASTM D 3418, using a DSC Q20 by TA instruments

Conclusion

Material was most likely a thermoplastic elastomer such as Santoprene (co-polymer of polypropylene and EPDM), as the IR spectrum indicated a polypropylene, EPDM and talc mixture, and the DSC scan showed no crystalline melting point behaviour.

IR also showed some unidentifiable absorption peaks.

 

Rubber – O-rings

Title

IDENTIFICATION OF RUBBER O-RINGS AND RELEASE AGENTS

Date

2013

Objective

To identify the material in Rubber O-Rings and their release agents.

Testing Undertaken

The O-rings were washed in chloroform to dissolve the release agents from the O-ring surface.

An ATR FTIR spectrometer was then used to measure the IR spectrum of the chloroform wash residue and the washed O-rings.

Conclusion

The IR spectra of the release agents matched Vaseline.

The IR spectra of three of the four O-rings matched butyl rubber. The fourth sample (analysed by Pyrolysis IR spectroscopy) matched an inorganic silicate filled ethylene propylene rubber.

 

Silicone – Electrical cables

Title

FAILURE ANALYSIS OF SILICONE
INSULATED ELECTRICAL CABLE

Date

2014

Objective

To conduct a failure analysis on the cracked silicone insulated cable to determine the root causes for the herniating of the cables from a materials perspective. The cables were part of a 75kW electric motor that drives a jet fan in a highway tunnel.

Photo

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image218.png

Figure 1. Faulty silicone sample

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image220.png

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image222.png

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image224.png

Figures 2a, 2b, 2c. BSE images of failed silicone samples

Testing Undertaken

Ashing at 600°C

Elemental Analysis by EDAX

SEM

Thermogravimetric analysis from RT to 700 °C

Failure Mode

Silicone insulation failed by embrittlement due to thermal degradation. The cable is apparently rated for 300 °C but turned brittle at lower temperatures (‘good’ silicone sample turned brittle after just 97 hours at 240 °C).

The thermogravimetric scans showed that both silicone samples began to degrade at 250 °C, thus the silicone is not properly heat stable and was likely exposed to high conductor temperatures.

The observed degradation of the film also suggested the cables were exposed to excessive temperatures around 300 °C

 

Silicone rubber – Seals

Title

FAILURE ANALYSIS OF A SEAL ON A GLASS CONTAINER

Date

2012

Objective

To investigate the cause of a silicone rubber seal sticking to glass and undergoing damage on removal.

Photo

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: IMG_0882 2.JPG

Figure 1. Failed seal showing loss of material at failure site

 

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: IMG_8527.jpg

Figure 2. Seal showing areas of cohesive failure

 

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: IMG_8614.jpg

Figure 3. Difference in gloss levels between good (dull) and bad  (glossy) samples

Testing Undertaken

IR-Spec

Microscopy

Failure Analysis

The glossy surface of the ‘bad’ silicone rubber has led to ‘auto-adhesion’ on the glass and this in turn has led to cohesive failure of the silicone.

Use of a less glossy silicone, or one containing an anti-block additive (such as silica), or a grade of silicone with a higher tear strength value is recommended.

 

Teflon – Flakes

Title

IDENTIFICATION OF WHITE FLAKES

Date

2013

Objective

To identify the composition of white flakes

Testing Undertaken

IR spectroscopy

DSC

Failure Mode

Both tests indicated the white flakes consist of Teflon

 

Thermoplastic – Mouldings

Title

ANALYSIS OF BLOOMING ON A THERMOPLASTIC DASHBOARD MOULDING

Date

2013

Objective

To investigate the chemical nature of the blooming species on the surface of thermoplastic dashboard mouldings (which are allegedly causing respiratory issues to users of the vehicles)

Testing Undertaken

Surface ATR-FTIR Spectroscopy

Surface solvent swap followed by NMR

ToF-SIMS

Conclusion

The blooming residue on the surface of the mouldings was composed of glycerolmonostearate (GMS) and related glycerol monooleate (GMO). Both are safe additives which pose no toxicity issues or chemical respiratory symptoms.

 

Thermoplastic polyurethane
– Layflat hoses

Title

IMMERSION TESTING OF TPU LAYFLAT HOSE SAMPLES

Date

2012

Objective

To investigate the cause of observed cracking of layflat hose carrying potable water in the field and to investigate the susceptibility to cracking of various layflat hose samples in aqueous chlorine solutions whilst under stress.

Photo

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image070.png

Figure 1. Used 1.0% free chlorine. Note extensive microcracking

Testing Undertaken

Chlorine Exposure Testing (under bending stress)

Digital Microscopy (50x)

Failure Analysis

The hose failed due to oxidative stress cracking in part caused by chlorine exposure and in part by stress.

 

Toner and acrylic paint – Recyclings

Title

CHEMICAL ANALYSIS OF TONER MIXED WITH ACRYLIC PAINT

Date

2013

Objective

To characterise the composition of waste toner mixed with acrylic paint.

Photo

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image160.png

Figure 1. EM photo showing good adhesion by Toner particles (red colour)

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image162.png

Figure 2. EM photo showing poor adhesion by ferrite particle

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: D:\Work\Failure Analysis Reports # 2300 - 3799\resized_files\image164.png

Figure 3. Flexural Testing of Compression Moulded Sample (4 mm displacement & break)

Testing Undertaken

Thermal Analysis by DSC

Electron Microscopy

Compression Moulding followed by flexural testing

Re-emulsification testing

Conclusion

Good adhesion between toner phase and acrylic binder is evidence of mutual compatibility and miscibility. Mechanical testing indicated the toner is capable of moderate flexing, however it exhibits brittle behaviour. It is recommended a rubber modifier be added to reduce brittleness.

 

Accreditation

plastics

Gallery

Gallery

Contact Us

Polymer Testing (Division of ExcelPlas)

473 Warrigal Road (Rear Factory)
Moorabbin VIC 3189

Postal Address
PO Box 147
Moorabbin VIC 3189

e. info@polymertesting.com.au
p. 03 9532 2207

www.polymertesting.com.au