Skip to main content

Icephobic Coatings - Development of test methods

Final Report Summary - ICECODE (Icephobic Coatings - Development of test methods)

Executive Summary:
This report summarises work performed under project 323456, Icephobic Coatings – Development of Test Methods (acronym: IceCoDe).
Icephobic coatings are intended to support wing ice protection systems by reducing the formation or adhesion of ice and consequently reducing energy requirements of the system (and thus fuel costs). Development of such coatings can be complex and costly, in part due to the expense associated with icing wind tunnel testing – the only method available for characterising anti-ice behaviour of a surface which is representative of the aircraft operational environment. The goal of this project was to develop a less expensive but suitable test alternative, which could predict anti-ice coating performance in an icing wind tunnel, and thus the operational environment. Two approaches were taken towards achieving this goal. The first approach attempted to identify chemical and physical characteristics of a surface which contribute to ice adhesion resulting from icing in a wind tunnel. This was done by characterising the surface of the coated test plate using a variety of analytical tests, and comparing these results to ice adhesion test results obtained in an icing wind tunnel by icing the test plate and recording the electro-expulsive energy required to remove the ice, in order to determine trends. The ideal target was a predictive model, whereby, based upon surface characteristics measured in a laboratory, performance of a coating in an icing wind tunnel (energy required to de-ice) could be predicted. This objective was considered a major challenge to meet, and indeed it was only met in part, in that a validated predictive model was not produced. However, characteristics of a surface which may contribute to ice accretion rate and ice adhesion were identified and/or tested. Some of these characteristics have not been found to have been reported elsewhere, and their identification may lead to further research, which in turn may contribute to an increased understanding of ice accretion and adhesion to surfaces. The second approach towards achieving the goal of a test method alternative to the icing wind tunnel was development of a laboratory-based icing and ice adhesion test device, which provides a good degree of capability to predict anti-ice performance of coatings in an icing wind tunnel, and thus the natural environment. To this end, a laboratory-based, fully-enclosed icing and ice adhesion test device, which has conditions controllable to enable a high level of natural environmental representation, has been designed, developed, manufactured and tested for fitness for function. The equipment has indeed been proven fit for function, and has been found capable of producing environmentally-representative conditions, for example in terms of pressure, humidity, freezing rain droplet size, and ice type, and once the test specimen is iced, a shear ice adhesion test is performed in situ without need to remove the sample or open the test chamber. The tests performed with the equipment on various coated surfaces during the project suggest good correlation with icing wind tunnel test data and thus achievement of the project goal. Due to complexity in the design and manufacture, and the large number of parameter settings it is possible to vary, some further work is required to fully determine capabilities of the equipment. However, at the point of closure of this project, the test equipment developed shows great promise for use as a method to increase the understanding of ice adhesion properties of surfaces, to reduce cost and duration of programs targeted at development of aircraft surfaces with reduced ice adhesion, and to batch test anti-ice coatings for quality assurance purposes once developments have been productionised.

Project Context and Objectives:
Icing conditions pose major risk for aircraft. In icing conditions, accumulation of ice on front facing surfaces have a dramatic impact on the aerodynamics of the aircraft which could led to aircraft handling and even loss of control of the aircraft ( Federal Aviation Administration, 2007). There are numerous readily available technologies that enable removal of ice accretion, such as electrothermal ice protection systems, the majority of which reduce the efficiency of the aircraft. Development has focussed on icephobic coatings in recent years in the hope that ice formation/adhesion can be reduced thus improving the efficiency of aircraft.
The objective of this work was to develop reliable laboratory-based test methods to characterise or predict anti-ice performance of surfaces and coatings in a real-world environment. The most reliable offline method of assessing improvements in anti-icing performance is the icing wind tunnel test. This test however has drawbacks in that it requires a large facility, is costly and time-consuming. This can result in testing being performed quite late in product development, on limited replicates of only one or two candidates, selected based on predictive tests or modelling. The likely consequence is development and implementation of non-optimal products. What is required is a more cost and time efficient test method which enables more rapid screening, development and optimisation work to be performed in a laboratory environment, by providing predictability of performance of developed coating candidates and other surfaces under icing conditions (and thus predict order of results in a wind tunnel test).

This project intended to take two approaches to the development of a laboratory-based test method.
• The creation of an anti-ice performance prediction model
• The development of a bench top icing test method

Prior to proceeding with test method development, wind tunnel testing was performed on a number of the coating candidates to determine the order of results that the laboratory-based tests should predict.

Anti-ice performance prediction model
The first and ideal (targeted) approach was to determine, through qualitative and quantitative tests, surface properties which in sum determine the icephobicity of the material under test. Currently, GKN can find no evidence that another organisation has fully determined the composite overriding parameters of a surface which determine ice adhesion to that surface. There is much research literature in the field, but experiments performed tend to focus on correlation of ice adhesion with a singular or very small number of surface attributes.

Not only were ‘initial’ coating surface parameters be considered, but also any surface properties changes which could be induced in the flight environment. These effects may be temporary, such as surface temperature or cleanliness, or permanent and/or progressive, such as micro-cracking caused by UV degradation. The effect of environment was believed by GKN to be of great importance, particularly as the laboratory-assessed ice adhesion performance of commercially-available purported icephobic coatings appears rarely to be (at least in full) realised in ‘real-world’ flight conditions. Generation and validation of a good predictive model for icing performance, which takes into account both ‘natural’ and ‘induced’ properties of a surface, therefore would truly be progress beyond state of the art.

Bench top icing test
The second approach, which serves as a risk mitigator for one of the set of key properties being missed in the first approach, is to develop a fully-contained bench-top test which has the best possible representation of a wind tunnel test. This test does not need to be an identical ‘mini-replication’ of the wind tunnel, but needs to ensure that the mode of icing and adhesion testing are similar enough to provide the same order of results as the wind tunnel. The test must however be less costly and more rapid, whilst providing consistency. GKN was not aware of such a bench-top test at the time of starting this project (and based on recent searching, equipment of this type is still not known to exist elsewhere) and marked benefits would be realised if one could be developed.
2.1 WP1 Wind Tunnel Characterisation of Coatings
Characterise anti-icing performance of coatings using standard icing wind tunnel test. Performing this test up-front enables order of performance to be understood ahead of development of test methods.
The criteria of this work package will be met when:
• The specifics of the test setup have been agreed by the topic manager
• An appropriate number and variety have been sourced or formulated, and tested to a degree that they demonstrate differing characteristics in the icing wind tunnel
• The effect of aging on performance of the coatings is understood
• The deliverables have been submitted to, and approved by, the topic manager
2.2 WP2 Test Method Development
Develop predictive model and lab-based icing test. The former incorporates assessment of coating parameters to determine properties key to anti-ice performance, and subsequent generation of a predictive model. The latter incorporates construction of a bench-top icing and adhesion test which produces results in an order similar to that of the icing wind tunnel.
The criteria of this work package will be met when:
• Sufficient effort has been exercised to enable understanding of the overriding parameters contributing to ice adhesion
• Sufficient testing of various coating standards has been performed against these parameters to maximise chances of successful predictive model development
• A physical lab-based test method has been designed and constructed
• The deliverables have been submitted to, and approved by, the topic manager
2.3 WP3 Test Method Validation
Validate the developed test methods by using results of tests using these methods to predict performance of candidate coatings in an icing wind tunnel.
The criteria of this work package will be met when:
• Degree of correlation (with appropriate statistical data) between results predicted by the model and results obtained in the icing wind tunnel is determined
• Degree of correlation (with appropriate statistical data) between results predicted by the model and results obtained in the icing wind tunnel is determined
• The deliverables have been submitted to, and approved by, the topic manager
2.4 WP4 Final Report
Complete all encompassing report of work performed on project and any additional findings and/or recommendations.
The criteria of this work package will be met when the final report has been approved internally to GKN and externally by the topic manager and project officer.
2.5 WP10 Literature Review
Perform a literature review of currently-existing icing test methods and factors which are thought or known to affect icephobicity.
The criteria of this work package will be met when:
• A sufficient assessment of currently-existing ice adhesion test methods has been performed to determine benefits and inadequacies of each, to enable direction in test method development, or confirmation of suitability and feasibility of current ideas.
• A sufficient assessment of research papers in the topic area of parameters responsible for ice adhesion has been performed, to increase likelihood of successfully developing a predictive model for ice adhesion

Project Results:
3.2 WP1 Wind Tunnel Test Description Interim Report and Coating Results Report
3.2.1 Objective
The objective of this work package was to characterise anti-icing performance of coatings using standard icing wind tunnel test. Performing this testing up-front enabled the order of performance to be understood ahead of development of test methods.
3.2.2 Summary of Work Conducted
The formation of ice in an icing wind tunnel occurs through the bombardment of a leading edge structure with super-cooled water droplets travelling at speed. The process by which ice is formed in the wind tunnel and the type of ice created, is similar to the ice accretion seen on forward facing aircraft parts.
The use of the GKN Luton icing wind tunnel, in conjunction with a leading edge assembly and an electro-expulsive actuator system is the method by which coatings were tested for ice adhesion. The wind tunnel conditions are outlined below:
• All tests were conducted with the model set with a 30 degrees angle of attack.
• The large working section of the Icing Wind Tunnel was set with a free-stream speed of 100 knots.
• The tunnel static temperature was set to -15°C.
• The droplet flow rate was set to 80 cc per minute, through each nozzle with the use of the four middle nozzles.
• The actuator system was set up at the front.
The electro-expulsive actuator test procedure consisted of two main stages; ice accretion and actuator activation. Spray atomiser bars were used to expel droplets of water into the wind tunnel aircraft. The supercooled droplets were deposited onto the test surface over a 10 minute period. After the ice accretion stage, the actuator is activated. The activation of the actuator involves the completion of a parallel circuit. The two copper strips which have a positive current running through them are situated close together. The repulsive force between the strips causes the panel to distort and ice is expelled. Appendix A details the composition of the actuator system used to control the electroexpulsive icing wind tunnel method. The strength of the repulsion was controlled by the amount of current that was run through the parallel copper strips. Within this report, this has been described as the power level, a value between 1 and 10, with 10 being the strongest.
A significant number of coated substrates (>25) were manufactured within this work package. The coatings were selected based on their disparity of surface properties. Five of the coatings were assessed for their performance in icing wind tunnel conditions. The coatings were selected based on their surface properties. The initial properties are summarised in the table below. The performance of each coating was assessed before and after 353 hours of UVA exposure.
Coating Name Coating Characteristics
Nano-structured polyurethane Superhydrophobic, nano-roughened
Titanium dioxide fortified polyurethane Smooth, erosion resistant, soft
Conductive polyurethane Smooth, electrically conductive
UV cured acrylate based coating Glassy, hardcoat, smooth
Nano-structured UV cured acrylate based coating Hydrophoboc, glassy, hardcoat, nano-roughened
Table 1 – Coatings Assessed in Icing Wind Tunnel Tests
Appendix B details the initial icing wind tunnel test results obtained using the electroexpulsive icing test method developed within this work package.
Whilst, the UV cured coatings appeared to perform better once they had been aged for a significant portion of time. This was not really the case as the coatings had degraded significantly with the UV A exposure, leading to a semi-sacrificial coating. Under the UV A exposure, the UV hard coat may have cured further and then cracked due to stress within the coating.
Using the test method developed within this work package, the anti-ice performance of coatings could be evaluated. The best performing coating candidate was the titanium dioxide fortified polyurethane even though it had none of the characteristics identified in the literature as important for icephobicity, such as superhydrophobicity. This provides further evidence to suggest that surface roughness and hydrophobicity are not the only influencing factors on ice adhesion.
3.2.3 Conclusions
The main focus of the work within this project was the development of a bench top ice adhesion test method, whether that be through the creation of a predictive model or development of a novel piece of test equipment. To support this development, an in-icing wind tunnel test method using the GKN icing wind tunnel was required. To this end, a leading edge construct and electro-expulsive ice protection device was sourced, and an icing test method devised. The electro-expulsive test method worked through the application of a large positive current into two opposing copper strips situated behind the test substrate. The repulsion of the two positive currents caused the panel to distort and expel ice from the surface. The strength of ice adhesion to a surface could be quantified by the amount of force required to remove it from the coated substrate. Reliability testing of the method was conducted and the method was found to be repeatable and enabled the ice adhesion performance of different coated substrates to be quantified. The procedure was adopted for all future icing wind tunnel test campaigns.
Whilst the development of an icing wind tunnel test method was a main objective of this work package, a second objective was the characterisation of a variety of coated substrates in the wind tunnel and through laboratory based characterisation methods. As the test method required more definition than what was thought at the project outset, the decision was made to conduct the laboratory assessments in WP2 Test Method Development. The need to have a repeatable in-wind tunnel test method to support future technical development within this project was identified as critical to project success.
3.3 WP2 Test Method Development
3.3.1.1 Objective
The objective of this work package was to develop a predictive model and lab-based icing test. The former incorporates assessment of coating parameters to determine properties key to anti-ice performance, and subsequent generation of a predictive model. The latter incorporates construction of a bench-top icing and adhesion test which produces results in an order similar to that of the icing wind tunnel.
3.3.2 Summary of Work Conducted
WP2.1 Predictive Model Method
The coated substrates that were manufactured in WP1 Wind Tunnel Test Description Interim Report were evaluated using the following analytical techniques:
• Surface Energy and Contact Angle (with and without static)
• Surface Roughness
• Coefficient of Friction
• Pendulum Hardness
• Visual Observations
• Gloss
• X-Ray Photoelectron Spectroscopy (elemental composition)
• Dynamic Mechanical Thermal Analysis
The test methods utilised for the characterisation of the coated surfaces were selected as in combination, they provide a comprehensive assessment of the properties of a coating. The selection of test methods covers surface characteristics, chemical composition and the mechanical properties of a coating/ surface.
The icing wind tunnel results obtained during the work conducted in WP1 were plotted against the data collected from the laboratory based methods. By analyzing the resultant scatterplots, a number of properties stood out as potential influencing factors. These were:
• Polarity - indications suggest that the more polar a surface is, the more power is required to expel ice.
• Coefficient of friction – static and kinetic – the greater the force required to initially move a stationary load across the coating (static) the lower the power required to expel ice, whilst the kinetic shows that the lower the value the lower the power required.
• Pendulum hardness/ mechanical properties – as currently it suggests a softer coating has lower ice adhesion strength
• The effect of tribolectric charging. This was explored through the effect of static charge on the surface energies of the coated substrates.
• Coating thickness – currently the results show the thicker the coating the lower the ice adhesion.
• Elemental composition – oxygen and silicon – an increase in oxygen and silicon on the coating surface the lower the power required to expel ice.
A factor that was expected to show a relationship with ice adhesion, due impart to focus within the literature was surface roughness. A link was not established using the icing wind tunnel results obtained. However, the range of roughness may have been too small to show any relationship.
Due to the fact that only a small number of coatings had been tested to date in the icing wind tunnel, the relationships between the icing wind tunnel and laboratory test results only provided indications of the critical factors influencing icephobicity. To conclusively state that a relationship between the coating characteristics and icephobicity has been found, the coating test pool must be opened up to include a statistically significant number of candidates, this could any number over 30.
The above suggests that there are potentially a large number of influencing factors on ice adhesion, the majority of which have not been previously discussed in the literature. This could be the reason that icephobic coatings have not been developed. Hydrophocity and roughness do not have any results to indicate that they are factors that influence adhesion of ice. Although at this stage in the investigation, it could also be stated that the relationship has not been found as the coating pool is too small to show the differences detailed in the literature.
Improvements to the electroexpulsive test method were suggested during the evaluation of the wind tunnel test data. Whilst all of the coatings tested accreted ice across the entirety of their surfaces, some coatings appeared to accumulate more than others. It was unknown at this stage whether ice accretion rate and the ice adhesion strength of a coating were related. To this end, a weight of ice test was devised to explore this relationship. The test idea was proposed in WP2.1 Predictive Model Method but was trialled with success in WP3.1 Predictive Model Validation.
WP2.2 Lab-Based Icing Method
To enable a lab-based icing method to be developed that could incorporate the representative conditions that an aircraft is exposed to, against a basic concept sketch a number of specification documents were composed which detailed the requirements of icing conditions ( Federal Aviation Administration, 2007). The documents detailed what the individual sections of the adhesion device concept were to achieve. Figure 1 is a basic diagram of the lab-based ice adhesion test device.

Figure 1 – Basic Diagram of the Lab-Based Ice Adhesion Device Concept

The device that was designed and constructed within this work package comprises of a water ejection system, housing unit and a shear adhesion test device. Water droplets of a specified diameter (15μm to 50μm) are ejected at a velocity faster than gravity across an area of 100mm, the droplets fall through the interior of a column cooled by liquid nitrogen and become supercooled during their descent hitting the target substrate at the base of the chamber and freezing instantly. Once an ice layer of a defined size has been accreted the water ejection unit is turned off. The adhesion of the ice layer to the target substrate is assessed through the completion of a shear adhesion test. The ice accretion stage and adhesion test can all be viewed through a viewport attached to the side of the test chamber.
To achieve the environmental conditions required for representative ice forming as detailed in ‘Pilot Guide: Flight in Icing Conditions ( Federal Aviation Administration, 2007) the following variables can be altered as required:
• Substrate to nozzle distance
• Temperature at the base of the housing unit in the location of the substrate
• Temperature gradient between the top and base of the housing unit
• Motor speed for the shear adhesion test
• Substrate test area
• Droplet diameters (altered by pump pressure and nozzle diameters)
• Water droplet temperature at ejection
Tests were completed on aluminium samples with the following parameters set:
• Base temperature: -16 to -18°C
• Dry nitrogen purged the interior
• Water droplet spray duration: 40 seconds
• Water droplet temperature at ejection: 30°C
During the initial test trials, it was established that the test method was able to create rime, glaze and ice crystals. The image below (figure 2) shows the view down the viewing port of rime ice accreted onto a test sample.

Figure 2 – Ice accretion on a sample viewed down the viewing port
A runback simulator was also developed; it can be affixed on top of the motor unit. There are two independently controlled heater elements separated by an insulator. The simulator works by accreting ice on the test sample and observing the behaviour of runback ice across the sample. The angle of incline of the simulator can be changed electronically by the use of the motor.
3.3.3 Conclusions
The objective of this work package was the development of two lab-based ice adhesion test methods, a predictive model based on the understanding and impact of the physical characteristics of coatings and a lab-based ice adhesion test device. Both test methods were to provide performance results of coatings that correlate to icing wind tunnel tests.
The work conducted on the creation of a predictive model identified factors that indicate a relationship with icing wind tunnel test performance. Whilst the literature review identified superhydrophobicity and surface roughness as critical, the factors that actually stood out were polarity (a measure of surface energy), the elemental composition of the surface, coeffient of friction (a measure of surface slip) and the mechanical properties of the coating. The identification of these factors shows a significant step forward in the understanding of ice adhesion. The objective to create a predictive model was challenging and highly ambitious, it was not completed during this work package. However, further development/ investigation into the true impact of the factors identified within this work package could lead to significant advancements in icephobic coating development. It would serve as a guide to development programmes of where to focus time and effort.
The development of a lab-based ice adhesion test device was initially proposed within this project to serve as a risk mitigator to the predictive model. The predictive model was deemed a high risk strategy and the lab-based ice adhesion test method appeared to have a higher probability of succeeding. Using the information gained from the literature review as well as industry documents detailing typical icing conditions experienced by aircraft, a basic concept sketch and requirements list for the adhesion device were created. The requirements list focussed on the ability of the equipment to simulate representative conditions.
The device that was designed and constructed as part of this project comprises of a water ejection system, housing unit and shear adhesion test device. Water droplets of a specified size are ejected at a velocity faster than gravity. The droplets fall through the interior of a cooled column, becoming supercooled during the descent and freeze on impact with the target substrate at the base. Once an ice layer of a defined size has been accreted, the adhesion of the ice layer to the target substrate is assessed through the completion of a shear adhesion test. The ice accretion stage and adhesion test can all be viewed through a viewport attached to the side of the test chamber. To achieve the environmental conditions required for representative ice formation, parameters can be adjusted such as substrate to nozzle distance, test area diameter, test temperature and droplet size.
Along with the predictive model, the objective of this work package was the development and test of a lab-based ice adhesion test method. A significant amount of work was conducted to design the equipment and the initial trials conducted were extremely promising. The fact that the equipment has been used to demonstrate the creation of rime, glaze and ice crystals has meant that the equipment can be used to simulate representative icing conditions that are experienced by aircraft. This was a significant milestone in the development of this test method and at the time of writing this report, it is unknown if it has previously been achieved.
The correlation of the ice adhesion performance of coated substrates was initially scheduled to be conducted within this work package. However, the decision was made to move the icing wind tunnel testing into WP3 Test Method Validation due to the delivery of equipment late into the programme. This meant that the main milestones of the project could still be met.
3.4 WP3 Test Method Validation
3.4.1 Objective
The objective of this work package was to validate the developed test methods by using results of the tests to predict performance of candidate coatings in an icing wind tunnel. This will be met when a degree of correlation (with appropriate statistical data) between results predicted by the model and results obtained in the icing wind tunnel is determined and a degree of correlation (with appropriate statistical data) between results predicted by the lab-based ice adhesion method and results obtained in the icing wind tunnel is determined.
3.4.2 Summary of Work Conducted
WP3.1 – Predictive Model Validation
Through characterisation of the coating surface, coatings were identified in WP2.1 to test under icing wind tunnel conditions. Commercially available coatings perceived to be hydrophobic and icephobic, as well as standard wing paints, were sourced for testing. Coatings were assessed using the electro-expulsive test method developed within WP1 and also included the new method measuring the amount of ice accreted on a surface.
In excess of 30 coatings were assessed for their ice adhesion performance and through the statistical analysis of the data, multiple relationships between wind tunnel performance and lab-based results were identified. The understanding of the factors previously identified within WP2.1 was achieved, and the following relationships were also indicated:
• Polarity
• Oxygen
• Sodium
With the inclusion of the weight of ice accreted test, the following factors were deemed as influencing:
• Coefficient of Friction
• Calcium
• Oxygen
Mechanical properties of the coating have not been characterized in depth and lab testing methods should be sought for this.
To explore the effect of the identified factors a design of experiment should be completed; this should further strengthen the understanding and add confidence to the results by allowing prediction of results on manipulated coatings through purposely controlling the variables and the interaction effects.
The current wind tunnel methods have a number of issues that should looked to be improved to allow increased control of the results collected through testing. Some of the issues are as follows:
• Ice created is a mixture of rime and glaze
• Unknown temperature of the leading edge heater
• The sample has to be detached from the leading edge to complete the weight of ice test.
These issues could be interfering with the collated data, and may be having an influence on creating a predictive model by lowering the R² values. The following improvements could improve the method:
• Changing the angle of incidence of the leading edge to enable more runback (rime) ice and limit the amount of glaze ice.
• Changing the heater in the leading edge so that the temperature can be monitored and controlled.
• Able to measure the amount of ice build-up, and amount of ice expelled from the surface by 3D scanning of the process.
Increased knowledge and understanding has been gained through development of the test methods for icing, with factors as being more significant than others. Although a predictive model has unable to be developed, it is clear that a considerate improvement has been made within understanding of the factors that affect ice adhesion. This provides a good platform for further work, and will focus a design of experiment to gain additional understanding to potentially generate a predictive model in the future.
WP3.2 Lab-Based Icing Method Validation
Coatings were identified to test within this work package from the pool that were tested within WP3.1. The coatings were ranked in order of performance for the two wind tunnel test methods; the best and worst coatings were chosen to be remanufactured for the lab-based icing and wind tunnel icing, as well as an aluminium standard.
The ice created within the ice adhesion equipment was glaze ice; the temperature gradient between the top and the bottom of the chamber was approximately 35°C which resulted in droplets merging on the chamber walls and falling on to the test samples.
On analysis of the output data from Fx, Fy, Fvec it became apparent that Fx was the information of interest. For in depth detail on the analysis refer to DEV/R/8871/306. Appendix C records the data from both test methods and ranks the results in order of their performance. The results were plotted to see if there was any correlation between the two test methods.
The peak height was plotted against the power required to expel ice and the R² indicates a relationship with a value of 0.50. Also the performance rankings were investigated to see any potential relationship and the R² value is 0.76 which is significant and extremely encouraging (figure 2).

Figure 3 - Fx Analysis Graph showing the Icing Wind Tunnel Performance Rating against Ice Adhesion Test Rating
A greater understanding of what is required to perform ice adhesion tests has been gained and could act as a starting point for further development. To assess the areas for improvement, the equipment was evaluated in light of the following categories:
• Ease of Test Set-Up
• Duration of Test
• Measurement/ Monitoring Systems
• Test Parameters
It was suggest that the liquid nitrogen feed changed to an automated system, with a lower temperature gradient between the top and the bottom of the chamber. With incorporation of further temperature gauges and moisture sensor in to the chamber would allow improved control over the internal conditions. For the in depth evaluations on the equipment refer to DEV/R/8871/306.
The speed of the motor unit has been arbitrarily set based on the success of the initial adhesion test investigations detailed in DEV/R/8863/306. To further understand the relationship outlined above it would be extremely worthwhile to explore slower turn speeds, perhaps this would show more clearly the difference between an icephobic and standard coating.
To further support the validation of the lab-based ice adhesion device more coated specimens need to be tested. The test parameters used to date could continue to be used going forward, although investigations into the effects of temperature and/ or pressure could prove extremely useful/ critical in the development of understanding of both the test method and factors that influence ice adhesion so should not be overlooked. Moreover, the objective of predicting icing wind tunnel performance using the lab-based equipment should be conducted once a statistically significant number of samples has been assessed using the method (over 30). If the results obtained using the ice adhesion test device can be used to predict the ‘real-world’ anti-ice performance of substrates, this would be a turning point in the development and, hopefully, the achievement of icephobic coatings.
3.4.3 Conclusions
The objective of this work package was the validation of both the predictive model and the lab-based ice adhesion test device. The validation process centred around creation of coating variants that were to be assessed using the laboratory techniques. The icing wind tunnel performance of the coatings was to be predicted from the data collected and wind tunnel testing used to confirm the reliability of the predictions.
In excess of 30 coatings were assessed for their ice adhesion performance and through statistical analysis of the data, multiple relationships between wind tunnel performance and lab-based results were identified. The coatings assessed comprised a mixture of standard aircraft paints, commercially available icephobic coatings and GKN Luton modified coatings. The modified coatings were manipulated specifically to provide a wide variety of surface characteristics. The aim was not to develop an icephobic coating but to create a foundation of results to use for analysis. The relationships that were identified from analysis of this data were:
• The polarity of the surface
• Elemental composition of the surface
• Coefficient of friction (a measure of surface slip)
Whilst the surface characteristics on their own do have an influence on the ice adhesion strength to a surface, it is the properties in combination that have the most impact. Using the knowledge gained during the factor identification stage, the project team were able to formulate a coating that had superior icephobic performance in comparison to other candidates in the coating pool.
The ambitious objective of the validation of the predictive model was not achieved in the course of this project. However, by the analysis of over 30 coating variants using the wind tunnel and laboratory based techniques, the results of the statistical analysis have reached good level of confidence. This means that the relationships that have been identified as part of this project are true indications of critical factors affecting the icephobicity of a surface. In future development programmes, the knowledge gained from this project could be invaluable to the programme’s success. The creation of development targets and focus of technical research could be directed at the project outset leading to faster development with improved quality of the resultant product.
The lab-based ice adhesion test method objective for this work package was also the validation of its results against icing wind tunnel tests. Initial adhesion test trials were conducted on coatings that had undergone icing wind tunnel testing. The order of performance of both the initial lab-based adhesion test trials and icing wind tunnel results were compared to understand whether a correlation between the two methods existed. It was established that a correlation between the two methods did exist and was, based on this sample set, extremely strong. As the test sample pool was smaller than desired, validation of the equipment was not able to be achieved during the course of this project. However, as with the predictive model work, significant progress towards the highly challenging objectives has been made in the short project duration.
Due to the large number of potential settings on the lab-based ice adhesion device, more work is required to understand the optimum settings for specific test conditions as well as how to accurately interpret the performance data going forwards. Once this has been achieved, along with the factors determined in the predictive model work conducted within this programme, it could have significant benefits for the development of icephobic coatings through reductions in cost/ time and an increase in the quality of development programmes. An icephobic coating if successfully developed would serve to reduce the overall power output requirements of an aircraft, therefore, improving efficiency but it could also help to improve the safety of future aircraft.
3.5 Final Report
3.5.1 Objectives
Complete an all encompassing report of work performed within project and any additional findings and/or recommendations. The criteria of this work package will be met when the final report has been approved internally to GKN and externally by the topic manager and project officer.
3.5.2 Summary of Work Conducted
The final report was submitted as the final deliverable at the end of the project. It details in summary all of the work conducted within the report and any recommendations for further work that would enhance/ build upon that should be convinced if a future development programme were to continue technical learning.
3.5.3 Conclusions
The report was completed and submitted. The technical progress of the project was presented to the topic manager at the end of the project. The topic manager expressed satisfaction over the summary of work provided and technical progress completed within the project time frame.
3.6 WP10 - A Review of the Properties of Ice Adhesion and Developmental Ice Adhesion Measurement Techniques
3.6.1 Objective:
The objective of this work was to conduct a literature review of all the currently existing test methods and the factors perceived to influence icephobicity.
3.6.2 Summary of Work Conducted
Hydrophobicity or super-hydrophobicity effect on ice adhesion is a widely investigated factor where research has reported that by measuring the water contact angle of a surface the icephobicity can be predicted. Other research suggests that whilst there is a benefit to using hydrophobic materials their effect is greatly reduced when the temperature is taken below -8°C. This suggests that hydrophobicity is not a singular influencing factor on the understanding of ice adhesion.
It is professed that an increase in the surface roughness results in an increase in the ice adhesion strength; it gives a larger surface area for water to adhere to, although this needs to be explored in greater depth. It is also thought that having a nano-structured surface aids hydrophobic properties and reduces the contact area a droplet has with a surface. This has been seen to reduce the energy required to remove accreted ice. It is apparent that there are other factors that influence icephobicity than just roughness and hydrophobicity as there are conflicting areas of research.
There are a number of different test methods designed to measure the adhesive strength of ice on a variety of surfaces.
• The Centrifugal Adhesion method uses non-representative ice formations. A coated cantilever with ice formed upon is affixed to a centrifuge to measure the shear stress necessary to remove the ice from the surface.
• Icing shear test requires glaze ice to be formed at -10°C between to metallic cylinders. The force required to rotate the inner cylinder is the measured adhesive strength.
• Icing double lap shear test is ice that is created within a conventional freezer and pulled in differing directions until the adhesive material (the ice) breaks.
• The vibrational cantilever method uses an icing wind tunnel to create rime ice, which is representative of the ice accreted on aircraft. Although this forms the relevant ice it defeats the purpose of having a lab based test method by using a wind tunnel.
3.6.3 Conclusions
Whilst, it was established that a significant amount of research had been completed on superhydrophobicity and surface roughness and their impact on icephobicity, only a small amount of literature was found exploring the effects of surface characteristics such as surface energy and the chemical composition of the surface. Also, the impact of interations between influencing factors have not been explored. The evaluation of the current ice adhesion test methods aside from the icing wind tunnel identified one possible reason why the correlation between icing wind tunnel tests and lab-based ice adhesion methods had not been achieved. The majority of ‘out of wind tunnel’ test methods rely on the creation of ice through the use of conventional freezers which lead to the creation of ice not representive of the ice accreted during icing conditions.
The learning taken from this literature review was used as a starting point for the development of both the predictive model and lab-based ice adhesion test method.

Potential Impact:
This project has delivered two notable advances in the field, from which it is believed significant benefit can be derived. Firstly, characteristics of a surface which may contribute to ice accretion rate and ice adhesion have been identified and/or tested. Some of these characteristics have not been found to have been reported elsewhere, and their identification may lead to further research, which in turn may contribute to an increased understanding of ice accretion and adhesion to surfaces. Secondly, a laboratory-based, fully-enclosed icing and ice adhesion test device, which has conditions controllable to enable a high level of natural environmental representation, has been designed, developed, manufactured and tested for fitness for function. The reason that this project originally came about was that a less expensive but suitable alternative (or at least an addition) to the current industry standard for assessing a surface’s behaviour in response to icing of the nature experienced by an aircraft was needed. At the time the topic area was defined, the only method which could ice surfaces under representative conditions was the icing wind tunnel, and this presents a number of detrimental issues which impact upon time, cost and quality of delivery of coatings to the industry which vastly reduce the adhesion of ice, reducing power output requirements of ice protection systems (and thus fuel costs) and enhance flight safety. Firstly, icing wind tunnel setup, costs are significant, with even the cost of building a small facility running into millions of Euros, before consideration of running, maintenance and upgrade costs. Secondly, due to this high entry barrier, there are not many icing wind tunnels available. Thirdly, the physical footprint of icing wind tunnels is significant. Fourthly, expertise is required to run the tunnel due to the complexity of settings and parameters and necessity to perform ad hoc maintenance works. Finally, and a result of the four previously mentioned issues, testing is difficult to schedule and very expensive. Costs run into thousands of Euros per day. The knock-on effect of this last issue is that: representative icing testing in development programs is limited, being performed usually late in a product development program; and development projects on a budget or undertaken by smaller organisations will exclude icing wind tunnel testing, and confine tests to cheaper, non-representative versions. The result is often little or at the very least sub-optimal anti-ice performance benefit of the surface being developed, and likely further added time and cost in redevelopment efforts. The two approaches taken under this project to develop laboratory-based test methods which would be able to predict anti-ice performance of coatings in the icing wind tunnel and thus the natural environment, both targeted the issues described above and it is believed that the resulting methods, particularly when used in combination, can go a long way to eliminating these issues, reducing the duration and cost of development programs and increasing likelihood of product development success. The cost of the test equipment and its operation are a fraction of those of an icing wind tunnel and the equipment can be housed within a small laboratory. This makes it accessible to smaller organisations as well as large, who wish to enhance their understanding of, or develop, anti-ice surfaces. Additionally, cost of development and project timescales can be reduced and product performance quality increased, by: (1) use of knowledge of factors which effect ice adhesion to drive chemistry and structure targets of the anti-ice surface, and thus drive a more efficient development approach; (2) testing of coatings from early development stages and throughout the program using the laboratory-based ice adhesion tester; then (3) performance confirmation (if required) later in the program in the icing wind tunnel. A laboratory-based ice adhesion tester which could be correlated to icing wind tunnel performance sufficiently to be considered as a standard test method could however obviate step 3 in some regards.
The potential benefit then, is in summary, improved aircraft fuel economy and flight safety achieved in a shorter period of time and lower cost, through more rapid and directed product development and less expensive testing provided by the methods developed, and increased competition as more organisations are likely to be able to afford the costs associated with use of these methods by comparison to the icing wind tunnel.
Prior to commencement of dissemination activities, some additional validation work will be performed on the test equipment to enable its full capabilities to be understood and to more accurately determine test validity and repeatability. This will take a period of around six months after the end date of the project.
Once this work has completed, dissemination will primarily be by conference presentation. It is intended that publications and presentations be provided at a Clean Sky forum, and at relevant aerospace icing or coatings conferences, such as: SAE International Conference on Icing of Aircraft, Engines, and Structures; IntAirCoat; European Coatings Congress; and European Functional Coatings Conference.
The equipment developed will be used for anti-ice coating development programs and other programs where surface characterisation for ice accretion and adhesion is required, with access intended to be internal, by project partners and potentially by other parties by agreement. The option regarding the manufacturing replicate items of the test equipment will be broached at such time that the pre-dissemination activities noted above have been completed and any required improvements have been implemented.

List of Websites:
N/A