Experimental characterisation and numerical stress modelling of fatigue performance and durability of marine epoxy ballast tank coatings

Oriaifo, Aigbokhaebho M. (2016) Experimental characterisation and numerical stress modelling of fatigue performance and durability of marine epoxy ballast tank coatings. Doctoral thesis, Northumbria University.

Text (Doctoral Thesis)
oriaifo.aigbokhaebho_phd_13042708.pdf - Submitted Version

Download (10MB) | Preview


Water ballast tanks (WBTs) are the most challenging areas of a merchant ship to protect and maintain against corrosion due to their size, structural complexity and operational conditions.

Prior to commercial use, WBT coatings must be ‘type approved’ by meeting criteria described in the International Maritime Organisation (IMO), Performance Standard for Protective Coatings for Dedicated Seawater Ballast Tanks (PSPC). Despite being ‘type approved’, early cracking failure of some WBT coatings has been observed in-service resulting in increased efforts from industry and scientific community to address this issue by developing a fundamental understanding of the factors influencing cracking. Typically, the cracks appear on welds where the coating is applied thickest, surface preparation is different, geometry is involved and after exposure to hygrothermal cycling conditions typically found in WBT. In addition, so called ‘pure epoxies’ are positioned as having superior performance in WBTs although there is no documented evidence to confirm this belief.

This study is aimed at characterising the optimum fatigue performance and durability of WBT coatings against cracking.

Test protocols commonly utilised in evaluating the performance of WBT coatings were reviewed. Factors affecting cracking of WBT coatings were prioritised for investigation. WBT coating formulations were grouped and 4 typical variants: solvent containing pure epoxy (SPE), solvent free pure epoxy (SFPE), solvent containing modified epoxy (SME) and solvent free modified epoxy (SFME) were manufactured. Each formulation was spray applied on T specimens (which had two types of surface preparation) and polytetrafluoroethylene (PTFE) panels. Free films peeled from the PTFE panels were shaped into rectangular and dog bone specimens respectively. In addition, two types of dry film thicknesses (DFT) for dog bone and T specimens were investigated.

Under laboratory conditions, a proposed fatigue test was designed to characterise and establish relationship between resistance to cracking and coating properties. In addition, thermal analysis techniques, differential scanning calorimetry (DSC), thermo-mechanical analysis (TMA) and dynamic mechanical analysis (DMA) were used to measure coating properties of free films including glass transition temperature (Tg), coefficient of thermal expansion (CTE) and storage modulus (E’) of specimens that were unaged and aged. DFTs were measured for unaged and aged T specimens. Tensile fracture strength and elongation to fracture were measured. A reanalysis with duplicate set of specimens was performed to validate the fatigue test and thermal analytical techniques. Finite element analysis (FEA) was used to simulate stress distribution as a function of coating film thickness. Statistical analysis (ANOVA test) was used to compare the mean effect of volume loss of DFTs measured before and after fatigue test.

The results of the fatigue testing produced a specific ranking twice and demonstrated the effect of hygrothermal stress on WBT formulations. SFPE was most crack resistant with fewer T specimens failing at the intersection weld and SPE was the worst. FEA modelling confirmed the increase of stress on the welded corner with increase in coating film thickness.

Thermal analysis measurement were not invariable exhibiting a mixture of increase and decrease in magnitude with subsequent exposure conditions. Tg and E’ values increased significantly for SPE and SME while the reverse was for SFPE and SFME. CTE values was significantly higher for SFPE and SFME but lower for SPE and SME.

Mechanical tensile testing showed that the fracture strength of the aged specimen increased for the four formulations with SFPE showing the least increase by 2 times the unaged and SPE with the most increase by 9 times the unaged. The elongation to fracture decreased significantly with SFPE decreasing to 1.87% while other formulations including SPE decreased to approximately 1 %.

ANOVA statistics found that the amount of volume loss from DFT was significant and depended on formulation. SFPE had significantly less volume loss than others including SPE that showed the highest amount of volume loss.

The implication from this finding is that it is unsafe to generalise the performance behaviour of WBT coatings from one constituent of the formulation. Rather, comments on performance should be on the formulation as whole.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: Hygrothermal cycling, corrosion, test protocols, loss of flexibility and embrittlement, marine asset integrity management
Subjects: H300 Mechanical Engineering
Department: Faculties > Engineering and Environment > Mechanical and Construction Engineering
University Services > Graduate School > Doctor of Philosophy
Depositing User: John Coen
Date Deposited: 18 Mar 2022 09:47
Last Modified: 18 Mar 2022 10:00
URI: http://nrl.northumbria.ac.uk/id/eprint/48695

Actions (login required)

View Item View Item


Downloads per month over past year

View more statistics