Printer Friendly

Estimation of long-term drag performance of fouling control coatings using an ocean-placed raft with multiple dynamic rotors.

Abstract An experimental setup was designed and built to estimate changes in the skin friction of fouling control coatings (FCC) over an extended period of time in conditions simulating the vast majority of ship profiles (regarding speed and activity) in the present market. The setup consisted of two separate parts: one aged FCCs directly in seawater in a dynamic manner by simulating the exposure condition of a ship's hull, and a second, laboratory part measured the torque (drag) of aged coatings in a rotary setup. From the spring to the autumn of 2013 and 2014, four commercial FCCs were exposed for 53 weeks in Roskilde Fjord, Denmark, which is characterized by relatively cold seawater and a salinity of approximately 1.2 wt%. The in situ immersion seawater conditions consisted of five-week cycles divided into 2 weeks of static immersion and 3 weeks of dynamic immersion, during which time the cylinders were rotated at a tangential velocity of 8.1 knots. The skin friction was found to generally increase more during the static period, compared to the dynamic ones. Over the course of the entire exposure period, the skin friction of the investigated FCCs decreased in the following order: fluorinated fouling release coating (FRC) (highest skin friction), hydrogel-based FRC without biocides, silylated acrylate self-polishing copolymer coating, and hydrogel-based FRC with biocides (lowest skin friction). However, the differences in skin friction between the latter three coatings were minor and often within the range of experimental uncertainty. The average surface rough ness of the FCCs in the newly applied and mechanically cleaned condition, determined as the Rt(50) and [R.sub.z] parameters, was evaluated as poor predictors of skin friction.

Keywords Skin friction, Fouling control coatings, Antifouling, Biofouling, Dynamic exposure

Introduction

Due to the high costs of fuel and environmental concerns (emissions of greenhouse gases, S[O.sub.2], and N[o.sub.x]), the fossil fuel efficiency of ships is becoming increasingly important. Commercial ships typically experience dry-docking intervals of 3-5 years, although extended dry-docking of up to 7.5 years can be granted for certain ship classes. During that time, substantial biofouling, defined as the accumulation of micro- and macro-organisms (e.g., bacteria, algae, slime, weed, or barnacles) on man-made structures can occur. This accumulation adversely affects ships through the loss of speed, decreased maneuverability, higher fuel consumption leading to a higher cost of operation, emission of harmful gases, frequency of dry-dockings, and translocation of invasive species. (1)

Frictional resistance nearly always represents a considerable portion of a ship's total resistance. (2) For example, frictional resistance constitutes 70-90% of the total resistance for slow trading ships (e.g., bulk carriers and tankers) and, occasionally, less than 40% for faster trading ships (e.g., cruise liners and container ships). (3) The choice of fouling control coating (FCC) thereby affects fuel consumption significantly because it directly influences the drag on the coated hull.

In this paper, the term FCCs cover all marine topcoats below the waterline of a ship with the purpose of minimizing friction arising from water flowing over the coated hull. The term "fouling release coatings" (FRCs) rely on a working mechanism with a dual mode of action, that is, nonstick properties and a biofouling release behavior. The general idea of FRCs is to minimize the adhesion between fouling organisms and the surface, so that the biofouling can be removed by hydrodynamic stress during navigation or by a simple mechanical cleaning. (4) Most of the present FRCs are based on cross-linked poly-dimethylsiloxanes (PDMS). Conventional biocidal antifoulings cover "self-polishing copolymer coatings" (SPC), "controlled depletion polymer" (CDP) coatings, and other topcoats which primary working mechanism consists of releasing biocide to minimize biofouling. SPC coatings are characterized by a controlled biocide release, which allows a fairly constant leaching rate over time at constant seawater conditions. (1)

It is well known that biofouling can cause severe drag increase; therefore, an FCC with low drag in newly applied and unexposed conditions can prove to be a poor choice if it easily biofouls. Thus, the value of the drag performance for a newly applied coating condition is limited unless the coating can stay completely clean. (2) Static exposure reproduces the biofouling pressure on an idle ship quite well, but the vast majority of larger commercial ships are primarily moving, experiencing only relatively short idle periods. Zargiel and Swain showed that biofilm adhesion, abundance, diversity, and community composition were impacted differently when exposed either statically or dynamically; thus, it was concluded that, rather than statically grown biofilms, dynamically grown biofilms should be used to address the influence of biofilms on drag and the associated costs. (5) The current drag performance estimation is often based on the newly applied coating drag performance and is typically followed by drag measurements after static immersion.

The literature generally reports that FRCs exhibit a lower drag, compared to conventional biocidal antifouling coatings in the newly applied condition [e.g., references (6-12)]. However, it has generally been reported that when FCCs are exposed statically, conventional biocidal antifouling coatings exhibit lower drag than FRCs [e.g., references (10, 11, 13)]. However, one study reported superior performance of FRCs compared to conventional biocidal antifoulings when examined after 3.5 months of static exposure. Since the results for exposure conditions resembling larger commercial ships are limited, it is yet uncertain whether FRCs or conventional biocidal antifouling coatings perform better under realistic conditions. Millett and Anderson reported that the use of an FRC led to a 12% lower fuel consumption in all weather conditions for an aluminum catamaran when compared to a low-copper TBT-free biocidal antifouling coating. (15)

According to the results in Swain et al., the relative performance between the four coatings tested, i.e., two FRCs, one copper ablative and one copper SPC, varied with the test condition, i.e., newly applied, after static immersion, and after dynamic immersion. (11) The test conditions clearly impacted which coating had the lowest friction. For instance, an FRC exhibited the lowest drag coefficients after dynamic exposure while the best performing coating under static conditions was a Cu-ablative coating. (11) It is, therefore, of utmost importance to determine the frictional performance in exposure conditions, which to the best possible extent simulate those a ship hull encounters during voyage to determine the optimal choice of coating with respect to friction. For this reason a new approach, which simulates exposure conditions of ships realistically, has been developed in this work to provide an accurate determination of an FCC's drag performance in ship-like conditions. The work presented in this paper includes experiments with FCCs during a long exposure period (approximately 53 weeks) with frequent drag measurements during the entire experimental period, a land-based and an in situ rotor setups in a combination not previously used and FCCs exposed to a high percentage of dynamic immersion (60%), as opposed to the more common experiments performed after primary static immersion.

Experimental setup

The experimental setup used in this study consisted of two parts. One part of the setup, a laboratory-placed rotor, was used to determine the drag performance (torque) of FCCs. The other part employed coated cylinders that could be rotated in the seawater from a raft. This latter setup was used to age coatings in conditions simulating those that FCCs typically encounter during voyages. For a period of 53 weeks, the drag performances of the coated cylinders were measured every 2 or 3 weeks in conditions resembling those of slow to fairly fast ocean-going ships. After 25- and 53-week exposures, the coatings were mechanically cleaned by hand with a soft wet cloth. The release of biofouling was removed with a light stroke over the FRCs, while the SPC coating needed a somewhat firm stroke to remove the biofouling. The coated cylinders used in this study were transferred from the raft to the laboratory rotor for drag performance measurements to be taken; afterward, they were returned to the raft for further aging. It should be noted that a typical dry-dock period is 5 years; therefore, the 53-week immersion period represents only some of the time between two dry-dockings.

Drag performance measurement setup

To measure the drag performance, a laboratory rotor setup in which an inner cylinder was rotated inside a static cylinder was used. This setup was originally developed to study the polishing and leaching rates of chemically active antifouling coatings. (16) It is common to use a rotary setup to measure the drag induced by FCC surfaces [e.g., references (6, 8, 9,12,17)]. Figure 1 displays the laboratory rotating cylinder system. The inner rotating cylinder, coated on the outside, had a diameter of 31 cm and a height of 30 cm. A smooth PVC disk was attached to the top and bottom before the torque measurements. Both disks had a height of 0.5 cm; thus, the total height of the rotating cylinder was 31 cm. The outer static cylinder shell had a diameter of 38 cm and a height of 31 cm. The volume of the seawater tank was approximately 600 L. The gap distance between the inner rotating and the outer static cylinder was 4 cm. The two bearings ensured the mechanical stability of the shaft to which the rotating cylinder was attached. The mechanical stability was needed due to vibrations from the rotary setup when operated at high rounds per minute (RPM).

The tangential velocity applied to the coated cylinders ranged from 3.2 to 15.8 knots (i.e., 100 RPM to 500 RPM), while measuring the torque. The measurements were carried out in natural seawater transported from Roskilde Fjord, Denmark, in order to maintain the biofouling in the best possible condition and to determine the drag performance of the FCCs in conditions that closely resembled those encountered during aging. Two 0.5-cm (unexposed) smooth PVC disks were mounted to the end parts, i.e., top and bottom, of the cylinders in order to ensure identical top and bottom contribution for every cylinder. In this way, a constant contribution from the top and bottom was achieved, and the differences in drag between the FCCs and their development over time should, therefore, only originate from the changes occurring on the sides of the cylinders (Fig. 2).

Aging setup

The purpose of the aging setup was to simulate the conditions encountered by the hull of an ocean-going ship during voyage, in order to ultimately predict the long-term drag performance of different FCCs on a ship. The raft with four rotor setups was located in Roskilde Fjord in Denmark at the DTU Riso campus (Lat: 55.692626 and Lon: 12.079442). Figures 3 and 4 show the raft with the main parts: motors, coated cylinder, shafts, bearings, protection against floating objects, and electronic control box.

Dick and Nowacki also used a rotor setup, as in this study. Advantages and disadvantages of their rotor compared to the one employed in this study are worth discussing. (18) Eighteen different coating systems with different substrates, pretreatments, anticorrosive coatings, and topcoats were investigated for 51 months by Dick and Nowacki. (18) Curved panels of 6 x 6 inches were applied on drums exposed to seawater near Daytona Beach, Florida (USA). The plastic drums were three ft in diameter. The coatings were exposed in monthly cycles: 1 month with dynamic immersion at a tangential velocity of 20 knots was followed by 1 month of static immersion. Complete inspections were made after the first, third, and fifth cycles, and afterward, one each year was made. Blank panels were also inspected both during the entire exposure period and each new season to determine the biofouling intensity of each season. The antifouling performance was rated with respect to total biofouling and in terms of the specific biofouling present (e.g., barnacles, mollusks, annelids, filamentous and encrusting bryozoan, hydroids, and algae). Their aging setup was, in many ways, similar to the one described here, with the main differences being that the setup of Dick and Nowacki used a smaller test area, a much longer exposure period, different exposure cycles, more coatings in the investigation, and much warmer seawater (i.e., a higher biofouling intensity was expected). (18) The main advantages of the setup used by Dick and Nowacki were that more coatings could be investigated and the coatings did not have to be moved to another setup. (18) Thus, it avoided the risk of impacting the biofouling that might have occurred when removing it from the seawater (discounting a presumably short inspection period on the site), and avoided the risk of mechanical damage during transportation. The main drawback was that the coatings could only be evaluated via visual inspection; a visual inspection is, at best, only an approximate indication of the drag performance, whereas a laboratory rotor setup of the present work provides accurate drag measurements. The visual inspection is, furthermore, likely to be subjective, making comparisons to other experimental data inaccurate. Another setup similar to the one employed by Dick and Nowacki was the one used by Swain and Zargiel. (5,18) A drum of 1 m rotating at 6 knots was used to investigate microfouling during static and dynamic immersion. The same advantages and disadvantages mentioned above are valid for this setup. Swain et al. used a static test site to expose panels of 10 x 12 inches and subsequently exposed them in a land-based tank with a stirrer to simulate dynamic exposure. (11) Finally, after a cycle of static and dynamic immersion the drag induced by the panels on a boat was measured. This setup exposed the coatings in a dynamic and static manner, which could be representative of certain ships; indeed, a useful approach toward realistic FCC exposure. However, the dynamic exposure would, ideally, occur in situ and not in a land-based tank with a once-through seawater supply.

Biofouling intensity at aging setup

The natural seawater present at the aging setup is relatively cold, i.e., ranging between approximately -5 and 25°C during the year. The salinity is approximately 1.2 wt% at the test site. (19) Figure 5 provides a close-up view of the bottom of the raft after 6 months of exposure. The uncoated PVC material at the bottom of the raft had received significant biofouling in the form of a thick slime layer, a vast number of mussels, long seagrass, and some barnacles.

One of the inherent challenges when comparing FCCs is that biofouling intensity is subject to changes over time and location. No apparent difference in biofouling intensity was found on the bottom of the raft, however; indeed, a similar biofouling intensity was noted around the raft and, therefore, also at the four rotor setups on the raft. However, to reduce any systematic uncertainty the position of the coated cylinders changed every week to minimize any differences in experienced biofouling intensity during exposure.

Measurement procedure

Prior to seawater immersion, drag measurements were carried out for the coated cylinders in the newly applied coating condition. Subsequently, 2 weeks of static immersion took place followed by additional drag measurements. After the static immersion, 3 weeks of dynamic immersion with subsequent drag measurements were conducted. This cycle of 2 and 3 weeks of static and dynamic immersion, respectively, followed by drag measurements, was applied from the spring of 2013 to the autumn of 2014. The exact periods were from May 21, 2013 to November 13, 2013 (25 weeks) and from April 1, 2014 to October 22, 2014 (28 weeks), resulting in a total exposure time of 53 weeks. After the end of each season, i.e., after November 13, 2013 and October 22, 2014, a soft wet cloth was gently applied to the FCCs to remove any attached biofouling. After November 13, 2013 the FCCs were stored indoors at room temperature in a dark room until re-immersion on April 1, 2014. Surface roughness measurements were taken for the newly applied and cleaned coating conditions. Figure 6 shows the exposure conditions for the FCCs.

The severity of biofouling was also measured by a visual inspection of the coated cylinder surfaces each week. The coated cylinders were transported to the laboratory rotor (close to the aging setup) and back to the aging setup in buckets with seawater, ensuring that any biofouling would be protected during transportation. The drag measurements carried out in the laboratory rotor began with 100 RPM (i.e., approximately a tangential velocity of 3.2 knots) and then subsequently increased in increments of 100 until reaching 500 RPM (i.e., approximately a tangential velocity 15.8 knots). Each torque measurement lasted 10 min, and the average torque value recorded during the last minute was applied to determine the skin friction coefficient. Only the last minute was used in order to ensure minimal and constant contribution from the friction in the bearings, which was found to remain stable after 10 min. The torque measurements were carried out in two sequential runs. The first run correlates to the drag performance when weakly adhered biofouling, if present, detaches; the second run correlates to the drag performance of well-attached biofouling.

Materials and sample preparation

The FFC technologies used in the experiment were:

* Silylated acrylate SPC coating.

* Fluorinated FRC.

* Hydrogel-based FRC (no biocides).

* Hydrogel-based FRC with biocides.

The four investigated commercial FCCs were applied according to their specifications from the producer, resulting in a dry film thickness of approximately 100 µm for the SPC coating and 150 µm for the three FRCs. The application was made with an airless spray gun. The same commercial poly(dimethylsiloxane)-based (PDMS) primer was applied below the three FRCs resulting in an approximate dry film thickness of 100 µm. A commercially cured, modified epoxy primer was used as the coating below the SPC coating with an approximate dry film thickness of 100 µm. The topcoats were left to cure at room temperature for 1 week after application to ensure a high mechanical strength. Table 1 provides the basic parameters describing the FCC systems used in the experiments.

The backbone of the three FRCs consist of PDMS. The main differences among the FRCs are the copolymer additives, typically present in a concentration from 1 to 10 wt%. (4) The relatively small difference arising from primary copolymer additives can, however, still be of high importance for the drag performance after exposure to natural seawater [e.g., reference (14)]. For a more thorough review of FRCs and their working mechanisms, see, e.g., reference (4). For information regarding detailed SPC coating working mechanisms, see, e.g., reference (1).

Surface roughness measurements

Micro and macro-roughness were measured on the coated cylinders before immersion and after 25 and 53 weeks of immersion, followed by mechanical cleaning. The microroughness consisted of measuring the [R.sub.z] roughness with a sampling length of 0.8 cm. The mathematical definition of [R.sub.z] is

[R.sub.z] = 1/n ([n.summation over (i=1)] [p.sub.i] - [n.summation over (i=1)] [l.sub.i]), (1)

where n is the number of measurements over the sampling length, p is the vertical distance from the center to the peak, l is the vertical distance from the center to the valley, and i represents individual values from the measurements. (20)

To measure the microroughness, a laboratory roughness measuring device, Handysurf E-350, was used. The number of measurements, n, was five when measuring the microroughness of the coated cylinders over the sampling length of 0.8 cm. The macro-roughness was measured as the Rt(50) roughness, which is often used when correlating roughness and drag performance of FCCs applied on ships. (2) Equation (1) is also valid when determining Rt(50) values that represent the maximum peak to valley (i.e., n = 1) over a sampling length of 50 mm. (21) The total quality control (TQC) hull roughness gage of the type DC9000 was used to determine the Rt(50) values. It has a reported accuracy of ± 5 µm or <2%, whichever is greater. (22) Sixty individual roughness measurements were made on every coated cylinder for both the micro and macro-roughness, and they were evenly distributed on the cylinder to obtain an accurate representation of the surface roughness for the entire surface. The Rt(50) roughness measurements for the FRCs, which required special treatment, were performed according to the methods provided by Anderson et al. (23) If the surface was dry, the stylus hopped over the rubber-like material, and if the surface was too wet, the gage skidded very easily; both practices would provide erroneous readings. (23) To prevent these errors and obtain meaningful measurements, the FRC surfaces were wetted slightly.

Conversion of torque into friction coefficients

This section describes the drag measured for the rotating cylinder wall in the laboratory rotor and the subsequent conversion into friction coefficients. Furthermore, friction coefficients for the rotating cylinder wall were converted into friction coefficients for flat plates, which are assumed to resemble the skin friction of a ship's hull surface. The shear stress applied to the coated cylinder wall while rotating in the aging setup and its conversion to a flat plate are also described. The investigated shear stress range (10-100 Pa) applied to the rotating cylinder wall is similar to that experienced by larger commercial ocean-going ships.

Conversion of laboratory torque values into friction coefficients for cylinders and flat plates

From laboratory rotor torque values to cylinder friction coefficients

The torque measured by the torque sensor ([M.sub.T]) consists of the contribution from the drag acting on the rotating coated cylinder wall ([M.sub.C]), on the end surfaces (top and bottom) of the cylinder ([M.sub.E]), on the outer shaft surface area ([M.sub.S]), and on the bearings ([M.sub.B]). In quantitative terms,

[M.sub.T] = [M.sub.C] + [M.sub.E] + [M.sub.S] + [M.sub.B]. (2)

The relevant drag is that of the FCC surface ([M.sub.C]), i.e., the rotating coated cylinder wall, excluding all other effects. The contributions from the end surfaces, the shaft, and the bearings are eliminated by a correction factor, which can be expressed as

[M.sub.Cor] = [M.sub.E] + [M.sub.B] + [M.sub.S]. (3)

Thus, subtracting the correction factor from the torque values provides the torque contribution from the rotating cylinder wall at the investigated tangential velocities

[M.sub.C] = [M.sub.T] - [M.sub.Cor]. (4)

The torque values for the rotating cylinder wall can be converted to skin friction coefficients, [C.sub.f], assuming that this value can be directly related to the wall shear stress, [t.sub.w]

[M.sub.C] = [t.sub.w] Ar, (5)

where A is the wetted rotating cylinder wall (excluding the end surfaces) and r is the cylinder radius. The wetted surface area can be expressed as

A = 2prh, (6)

where h is the cylinder height.

The wall shear stress can be expressed as a function of the friction coefficient, [C.sub.f], the water density, ?, and the tangential velocity of the cylinder, U. (24)

[t.sub.w] = 1/2 [C.sub.f] ?[U.sup.2] (7)

By combining equations (5), (6), and (7), an expression for the skin friction coefficient for the rotating cylinder wall can be derived

[C.sub.f] = [M.sub.C]/?[U.sup.2][r.sup.2]hp. (8)

Correlation between rotor shear stress and ship speed

The shear stress induced by the rotation of a cylinder is different from the shear stress experienced by the hull surface of a ship, even when the free-stream velocities are identical, because of the different geometries. The free-stream velocity is taken as the moving velocity of the flat plate and in the case of a cylinder as the tangential velocity when rotated. These differences must be accounted for when converting the laboratory rotor measurements into the full-scale FCC drag performance for a ship (i.e., a large flat plate). The aim of the chosen velocities in this study was to determine the drag performance of FCCs for speeds that are representative of the vast majority of larger commercial ships in the present market.

The skin friction coefficient on the cylinder wall of a smooth surface is described by the following relation for open sea rotors, i.e., with no outer static cylinder present (25):

1/[square root of [C.sub.f]] = 4.07 x log ([Re.sub.r] x [square root of [C.sub.f]]) - 0.6. (9)

In equation (9) [Re.sub.r] represents the Reynolds number of the cylinder, given as

[Re.sub.r] = Ur/[??], (10)

where r is the radius of the cylinder, U is the tangential velocity of the rotating cylinder, and [??] is the kinematic viscosity of the fluid.

On the other hand, the theoretical skin friction coefficient on the cylinder wall of a smooth surface in a Coutte type of flow was described by Arpaci and Larsen (26) as

1/[square root of [C.sub.f]] = 1/k [square root of 2] Ln ([Re.sub.g] [square root of [C.sub.f]/2] + 5.5 (11)

with k being the von Karman constant with a value of 0.41. [Re.sub.g] represents the Reynolds number based on the channel gap between the two cylinders, described as

[Re.sub.g] = [Ul.sub.g]/[??], (12)

where [l.sub.g] is the gap distance between the outer static and inner rotating coated cylinder, U is the tangential velocity of the rotating cylinder, and [??] is the kinematic viscosity of the fluid.

In this study, it is assumed that the approximate shear stress on a ship's hull can be determined by that of a smooth, flat plate. According to Schlichting (24) the average skin friction coefficient representative for an entire smooth, flat plate in the turbulent regime for a Reynolds number up to [10.sup.9] can be found as

[[bar.C].sub.f] = 0.455/(log[([Re.sub.p])).sup.2.58]. (13)

In equation (13) [Re.sub.P] is the Reynolds number based on the length of the plate, given as

(14) [Re.sub.p] = [Ul.sub.p]/[??], (14)

where U is the velocity of the plate, [l.sub.p] is the length of the plate, and [??] is the kinematic viscosity of the fluid.

The average shear stress on a flat, smooth plate can, therefore, be found by combining equations (7) and (13) as

[[bar.t].sub.w] = 0.2275 ?[U.sup.2]/((log[([Re.sub.p])).sup.2.58]. (15)

Figure 7 displays the average shear stress correlations presented above with the dimensions for the applied setup. The open sea rotor used a tangential velocity of 8.1 knots. A kinematic viscosity of 1.37 x [10.sup.-6] [m.sup.2]/s, i.e., an approximate value of seawater at 10°C, was applied when determining the average shear stress. (27) By using the free-stream velocity and shear stress relationship for the open sea rotor and the laboratory rotor, as shown by equations (9) and (11), respectively, conversion into the full-scale shear stress of a ship can be determined, as exemplified in Fig. 7. It is seen that the shear stress is only slightly higher for a flat, smooth plate of 100 m, compared to one of 400 m. However, it must be mentioned that the velocity and shear stress correlation is only valid up to a Reynolds number of [10.sup.9], which is exceeded for the 400 m plate at around 6.6 knots, making the actual shear stress and speed relationship uncertain beyond this velocity. However, for the shorter 100 m plate, the speed and shear stress relation is valid up to a speed of approximately 27 knots. Figure 7 shows that a tangential velocity just above 15 knots in the laboratory rotor corresponds to a ship velocity of approximately 19 and 20.5 knots for a ship of 100 and 400 m in length, respectively.

Table 2 provides the tangential velocities applied during the dynamic aging and the laboratory drag measurements, as well as the corresponding ship velocities for both a 100 m and a 400 m flat, smooth plate, resulting in identical average wall shear stresses. The velocity conversion is based on the assumption that the average shear stress is identical at the converted velocities.

Once the average skin friction coefficient has been determined with the laboratory rotor and the tangential rotor velocity has been converted into the corresponding velocity of a ship with a given length, the theoretical effective power (towing power), [P.sub.f] (in W), necessary to overcome the skin friction resistance of the hull can be determined as

[P.sub.f] = [F.sub.f] U = 1/2 [[bar.C].sub.f] [U.sup.3] A, (16)

where [F.sub.f] is the skin friction resistance, U is the speed of the ship, [[bar.C].sub.f] is the average skin friction coefficient, A is the wetted hull area, and ? is the seawater density.

Estimation of the experimental uncertainty

The uncertainty of the measured torque is a combination of the uncertainties of the torque sensor, the reproducibility of the reading, the reproducibility of the sample preparation, and the uncertainty of the correction factor ([M.sub.Cor]). One of the inherent challenges when estimating the uncertainty during the period of FCCs exposure is the fact that biofouling, if attached, is expected to be released during the measurements. This was indeed observed, in particular for the fluorinated FRC, which was the FCC that exhibited the most biofouling. Drag measurements carried out with a biofouled surface, therefore, cannot be repeated; thus, the uncertainty estimations are unknown for biofouled surfaces. However, in the newly applied condition and mechanically cleaned condition after seawater exposure, the measurement uncertainty can be estimated, because it is assumed that the FCC surface will not change significantly during the drag measurements, as there is no biofouling present and the mechanical roughness is not expected to change significantly due to the shear stresses induced while the cylinder rotates. The drag measurement uncertainty in the newly applied and mechanically cleaned conditions was based on three replicate measurements in each condition at the investigated tangential velocities. In future experiments, the drag uncertainty during the exposure period can be estimated by using replicate cylinders. Table 3 provides the standard deviation obtained via measurements in the newly applied condition and mechanically cleaned condition after 25 weeks of seawater exposure, along with the average of the newly applied and mechanically cleaned condition. The average standard deviation obtained from the newly applied condition and from the mechanically cleaned condition after 25 weeks of immersion is assumed to represent the measurement uncertainty for cylinders with and without biofouling. In both the newly applied and mechanically cleaned condition the uncertainty decreased with increasing tangential velocities; this pattern was also observed by Weinell et al. (12)

Results and discussion

This section presents the skin friction coefficients at the newly applied coating condition and the development of the coefficients over 53 weeks, with mechanical cleaning after 25 and 53 weeks. Measured surface roughness parameters and their correlation to skin friction are discussed. Furthermore, the rating of the biofouling attachment observed on the coated cylinders during the exposure time is presented. The objective of measuring the roughness parameters was to obtain an estimate of the drag of an FCC in clean condition, without biofouling present, which often is the case for a short period after either a new coating has been applied or a soft hull cleaning (i.e., no severe changes in surface roughness). The purpose of determining the biofouling level attached to an FCC from visual measurements was to determine whether the visual observations could deliver accurate estimations of the actual drag values.

Smooth cylinder

To determine the correction factor, [M.sub.Cor] the drag of three smooth cylinders made of PVC with heights of 15,22.5, and 30 cm was measured. Figure 8 reveals that a linear relationship between the total torque and the cylinder height exists. Extrapolation of these lines to the interception with the y-axis (zero cylinder height) represents the correction factor, [M.sub.Cor]- Note that [M.sub.Cor] is a function of the tangential velocity. It is assumed that the correction factor is independent of the biofouling presence on the sides of the cylinders. This is assumed because the flow over the sides of the cylinders with the fairly low biofouling levels encountered during the test period would not significantly alter the flow compared to a clean side. Drag changes in the bearings are not expected, despite increasing shear stress on the sides of the cylinders due to biofouling, because the RPM remained identical and, therefore, the resistance in the bearings would not have been impacted by the presence of biofouling.

Material properties

Impact of the coating's surface roughness on drag

The micro and macro-roughness were measured by [R.sub.z] and Rt(50), respectively, in the newly applied coating condition and after 25 and 53 weeks of immersion with subsequent mechanical cleaning (see Table 4). The surface roughness of the silylated acrylate SPC coating is expected to have been influenced by the mechanical cleaning because brown (reddish) material from the coating remained on the soft cloth during the mechanical cleaning. The surface roughness measured after the mechanical cleaning is, therefore, not potentially a valid indicator of the surface roughness during the last part of the exposure periods for the silylated acrylate SPC coating. However, it could still represent a gentle hull cleaning in which minor impact to the surface is likely to occur. The biofouling attached to the FRCs, if present, was easily released from the surface, and significant changes to the surface roughness are not expected to result from the mechanical cleaning.

Table 4 shows that the silylated acrylate SPC coating had significantly higher [R.sub.z] roughness values than the three FRCs. Furthermore, the [R.sub.z] roughness values remained fairly constant for the FRCs when comparing the measurements before and after exposure. The increase in [R.sub.z] for the silylated acrylate SPC coating is assumed to be explained primarily by the mechanical cleaning process, as some material from the coating was removed. Compared to the FRCs, the much higher [R.sub.z] roughness values of the silylated acrylate SPC coating, ranging approximately from a factor 8-18 did not translate into a significantly higher drag in the newly applied or mechanically cleaned coating condition. The drag performance of FCCs indicated by their [R.sub.z] roughness values is, therefore, not an accurate predictor of their drag.

It was found that the average macro-roughness typically increased after the 25- and 53-week immersion period and subsequent mechanical cleaning. However, notice the large standard deviations leaving it doubtful whether the macro-roughness did increase significantly. The silylated acrylate SPC coating was much rougher than all of the FRCs, in the newly applied condition and after 25 and 53 weeks of seawater immersion followed by mechanical cleaning. There were some differences among the measurements of the FRCs. Candries and Atlar questioned the validity of using the Rt(50) roughness as a means of estimating the drag performance of FCCs because their rotor and large-plate towing tank experiments did not correlate well with the measured drag and Rt(50) measurements. (28) Candries et al. found that a poorly applied FRC exhibited lower drag than an SPC coating, even though the FRC had a higher Rt(50) roughness. (29) Thus, Candries et al. concluded that Rt(50) may not be a suitable parameter to adequately describe the roughness of FRCs. (29) However, Weinell et al. found a relationship between drag and the roughness parameters [R.sub.z] and Rt(50) by using a rotor setup, although only significant at tangential velocities above 25 knots. (12) Flack et al. stated that drag due to surface roughness depends on many surface parameters, including roughness height, shape, and density. (7) The present study revealed a strong indication that neither of the two investigated surface height roughness parameters, i.e., Rt(50) and [R.sub.z], would be a valid parameter to describe the drag performance of FCCs. Even in the case of the much larger [R.sub.z] roughness for the silylated acrylate SPC coating, there was no significant impact on the skin friction when compared to the FRCs, which had a much lower [R.sub.z] roughness and a similar skin friction in the newly applied condition as well as after 25- and 53-week immersion periods followed by mechanical cleaning.

Contact angle

The contact angle was determined for the four FCCs with demineralized water and a plate beneath the coatings, which was heated to 22°C. The static contact angle was measured after 60 s, which allowed the single drop (sessile) of 6 µm to settle. Five measurements were carried out on two replicate samples of each FCC. A dry film thickness of approximately 150 µm was obtained on a glass substrate. Table 1 shows the average contact angle and the standard deviation. It is clear that the fluorinated FRC, hydrogel-based FRC with biocides, and silylated acrylate SPC coating have similar contact angles, while the hydrogel-based FRC without biocides had a much lower contact angle. The different contact angles reveal the existence of differences in the surface properties. The variations between the FRCs are believed to be due to primarily differences in the copolymer and biocide content. It should, however, be mentioned that the contact angles changed with time and, therefore, also the surface properties of the coating. Ideally, the contact angle should be determined from FCCs that have been exposed to seawater; it is important to classify the surface properties from coatings that have been immersed in water, since seawater exposed coatings are the surfaces biofouling encounters (besides the first short period right after out-docking). Obviously, this is complicated to measure, as exposed FCCs would dry to some extent while measuring the contact angles, providing uncertainties from this parameter.

Tensile strength

The tensile strength was measured in a Zwick tensile strength machine of the type Roel. It was only possible to measure the tensile strength of the FRCs due to their flexible nature, whereas the SPC coating did not have sufficient flexibility to be determined. The FRCs were left to cure for 2 days on a plastic film and afterward cut to fit the grips in the Zwick tensile strength machine. The width was three cm and the distance between the grips, which held the coatings, was 2.5 cm. The stretch velocity was 100 mm/min in a vertical direction. The maximum forces the FRCs exhibited were measured for 10 samples and are seen in Table 1. The forces exhibited by the coatings varied, as they were stretched with the highest force usually obtained just prior to the FRCs breaking. The difference between the coatings was within one standard deviation; thus, distinction of the tensile strength between the FRCs was not possible. However, the FRCs should preferably be tested after exposure to seawater after sufficient time has passed to absorb seawater to saturation. In this way, the tensile strength of the FRCs during realistic exposure conditions can be determined.

Visual biofouling grading

The visual biofouling grading was ranked according to the classification system used by the U.S. Navy, i.e., Naval Ships' Technical Manual (NSTM). (30) The NSTM rating and the impact on the total resistance for an Oliver Hazard Perry class frigate (FFG-7) are described by Schultz. (31) The biofouling evaluation can be considered subjective, as it is based on individual judgment. Obvious difficulties exist in estimating the degree of biofouling with the naked eye via thickness, coverage, and type, for example. However, the NSTM rating is considered capable of providing an approximate estimate of the FCCs' drag performance. (31) Table 5 shows predictions of the change in total resistance (?[C.sub.T]) for an Oliver Hazard Perry class frigate (FFG-7) with a range of representative coating and biofouling conditions at speeds of 15 and 30 knots, as cited by Schultz. (31) It can be seen that the total resistance increases much when the biofouling rating increases.

Figure 9 provides the NSTM rating during the 53 weeks of seawater exposure. Note, however, that after November 13, 2013, the coatings received a mechanical cleaning with a soft wet cloth leaving the biofouling rating at zero when the FCCs were reimmersed on April 1, 2014. It was found that the fluorinated FRC quickly biofouled and remained higher than the other FCCs during the vast majority of the immersion period. The NSTM rating ranged between 10 and 20 during 2013, and it increased to 50 during 2014 for the fluorinated FRC. The silylated acrylate SPC coating quickly obtained a very thin slime layer, which remained practically constant throughout the remaining exposure period. The hydrogel-based FRC without biocides experienced some or limited biofouling. In 2013, the majority of the observations corresponded to zero, although there were a few observations of 10 and 20 NSTM ratings; in 2014, frequent NSTM ratings of 10 and 20 were detected, as well as a few ratings of zero. The hydrogel-based FRC with biocides experienced little or no visible biofouling until the middle part of the exposure period in 2014, at which time it did experience a minor coverage of thick slime (i.e., 2-10% coverage), corresponding to an NSTM rating of 10.

Cushing (32) stated that the algae activity for temperate waters has a major peak in the spring and a minor peak in the late summer. The trend of fast biofouling attachments primarily observed for the fluorinated FRC could, therefore, be due to a high biofouling upon the initial immersion (May 21, 2013), and a decrease in biofouling attachments by the end of 2013 could be due to the lower biofouling intensity during this period. The biofouling, however, remained more firmly attached at the end of 2014, which could be due to a slightly higher level of biofouling, making it more difficult to self-clean. Another cause may have been the loss of biofouling release properties, due to changes in the surface properties of the FRCs.

Drag performance of fouling control coatings

Newly applied coating condition

The drag performance of newly applied and unexposed FCCs only represents, at best, a very short period immediately after dry-docking. Furthermore, no indication of an FCC's ability to prevent or limit biofouling is provided when measuring the drag of newly applied coatings. It is well-known that biofouling can cause severe drag increase and, therefore, an FCC with low drag in newly applied and unexposed conditions can prove to be a poor choice if the coating easily biofouls. Nonetheless, the newly applied drag performance serves as a viable tool to establish the expected best possible drag performance because the drag of FCCs typically increases with time, although this is not always the case and minor reductions can occur over time [e.g., reference (12)].

The drag performances of the FCCs were measured prior to immersion in order to obtain the newly applied performance. The four FCCs exhibited similar drag performances, particularly at increasing speeds, when the uncertainty decreased. At a tangential velocity of 15.8 knots in the laboratory rotor, which corresponds to a ship speed of approximately 19.7 knots for a 100 m ship (Table 2) the skin friction coefficient was 5.7% higher for the fluorinated FRC, 4.4% higher for the silylated acrylate-based SPC coating, and 1.3% higher for the hydrogel-based FRC without biocides, when compared to the hydrogel-based FRC containing biocides. Lindholdt et al. measured a smaller and insignificant difference in drag between three of the coatings investigated in this study, i.e., a fluorinated FRC, a hydrogel-based FRC without biocides, and a hydrogel-based FRC with biocides, as opposed to the significant difference found in this study. (14) Figure 10 illustrates the skin friction coefficients for tangential velocities ranging from 6.3 to 15.8 knots.

Mechanically cleaned coating condition

Figures 11 and 12 display the skin friction coefficients of the mechanically cleaned coatings after 25 and 53 weeks, respectively. The drag measurements in the mechanically cleaned condition revealed that the differences in drag were smaller than the experimental uncertainty at the higher tangential velocities. It is well known that, when immersed in seawater, FCCs typically absorb water, which is particularly true for hydrogel-based coatings. Therefore, the mechanically cleaned condition is likely to offer a more accurate representation of the FCC condition relevant for a ship free of biofouling better, as compared to the newly applied condition, where the coating has not been immersed in seawater. The average roughness parameters [R.sub.z] and Rt(50) were slightly higher for the mechanically cleaned condition compared to the newly applied coating condition, although in most cases within the standard deviation. This increase in roughness would have been expected to result in a slightly higher drag, although a minor decrease was found for the mechanically cleaned coating condition. In short, the differences in the mechanically cleaned conditions were smaller than the experimental uncertainty when the FCCs were tested after 25 weeks; thus, no significant distinction between the FCC drag performances could be made in the clean condition.

Figure 12 shows the skin friction coefficients after 53 weeks of seawater exposure and mechanical cleaning, confirming the similarity in performance of the FCCs in this condition. Notice that the [C.sub.f] values are slightly lower in Fig. 12 compared to the values in Fig. 11. This could be due to uptake of seawater by the coatings, which apparently causes some leveling of the surfaces over time. This was also partly found by Lindholdt et al. (14)

Long-term drag performance evaluation

One of the major challenges regarding the determination of the drag performance of FCCs is creating a method to predict and accurately measure the long-term performance in conditions that simulate those experienced by ocean-going ships. Figure 13 shows the skin average skin friction coefficients for tangential laboratory rotor velocities of 9.5, 12.6, and 15.8 knots from the second run, i.e., those corresponding to well-attached biofouling, over the course of 53 weeks. The latter mentioned rotor velocities were chosen as they represent a velocity range widely used by a large part of the commercial fleet. The measurements from the second run are shown because they are more applicable to moving ships, since most of the loosely attached biofouling, if present, is assumed to have been released during the first run. The data from May 21, 2013 represent the newly applied condition prior to immersion, while the subsequent measurements were carried out after 2 weeks of static immersion and then after 3 weeks of dynamic immersion for the remaining exposure period. The drag was measured after mechanical cleaning after November 13, 2013 and again on October 22, 2014. Below the experimental values in Fig. 13, a letter indicates the exposure condition which had taken place prior to the drag measurements: N represents the newly applied condition, S represents the static immersion condition, D represents the dynamic exposure condition, and C the mechanically cleaned condition. It is seen that the hydrogel-based FRC without biocides, the silylated acrylate SPC coating, and the hydrogel-based FRC containing biocides had similar skin friction throughout the entire exposure period, with values often within the experimental uncertainty. The fluorinated FRC, on the other hand, had a much higher skin friction when measured after the static immersion periods in 2013 and after the initial part of the exposure periods in 2014 it was significantly higher, irrespective of the exposure condition being static or dynamic. This reveals that the fluorinated FRC had sufficient biofouling release properties during 2013 to return to a skin friction similar to the newly applied condition, when dynamically exposed. However, it showed inferior biofouling prevention properties, especially pro nounced during static immersion, which led to increased skin friction. When comparing the skin friction of the four FCC systems after dynamic aging had taken place during 2013, the difference was small and often within experimental uncertainty. Evaluation of the four FCCs during 2014 revealed small differences in drag, besides for the fluorinated FRC. The results reveal the performance of the coatings applied to a ship traveling 3 out of 5 weeks and being idle for the remaining two weeks in waters similar to those at Roskilde Fjord. Furthermore, the results show that the immersion condition impacted the resulting skin friction significantly, revealing the need for immersion conditions representative of traveling ship patterns, rather than the typical approach of static immersion followed by drag measurements. The biofouling rating of the four FCCs provided in Fig. 9 revealed varying ratings throughout the season with values ranging from zero to 50. The biofouling rating and increased total resistance noted in Table 5 reports an increased total resistance due to biofouling ranging from 2% to 34% in the case of an Oliver Hazard Perry class frigate moving at 15 knots. When comparing the skin friction coefficients and the biofouling rating, a qualitative correlation was found, although minor changes in the skin friction were not consistently reported along with a higher biofouling rating. The visual biofouling rating is, therefore, useful when aiming at an evaluation of an approximate increase in resistance, but providing an accurate estimate of the drag based solely on visual inspection is, indeed, complicated.

Figure 14 shows the average skin friction coefficients for the entire 53-weeks exposure (i.e., 25 data points for each FCC), whereas Fig. 13 showed the skin friction coefficients at the various investigated dates. Clearly, the difference between the best performing FCCs is small. The average difference, relative to the best performing FCC (i.e., the hydrogel-based FRC with biocides) based on the three highest tangential velocities, where the measurement uncertainty is lowest, was 17.7% for the fluorinated FRC, 1.2% for the hydrogel-based FRC, and 0.6% for the silylated acrylate SPC coating. The two hydrogel-based FRCs and silylated acrylate SPC coating performed within the experimental uncertainty.

The fluorinated FRC was by far the coating most prone to biofouling attachment. The impact of skin friction (i.e., biofouling on the FCC), exposure conditions (i.e., static or dynamic), and measurement conditions (i.e., 1. or 2. run) are, therefore, more relevant to study for this coating, as opposed to the other fairly clean FCCs. Figure 15 provides the average skin friction coefficients of the fluorinated FRC for the tangential velocity range of 9.5-15.8 knots for the first and second runs. The skin friction coefficients were often lower for the second run, which is explained by the release of biofouling. In most cases, the difference was significant when static immersion occurred prior to the drag measurements, while mainly insignificant when dynamic immersion occurred prior to the drag measurements. This reveals that even though the dynamic aging was operated at 8.1 knots and the laboratory rotor operated up to 15.8 knots, a significant release of biofouling did not occur at the higher tangential velocities if dynamic aging had occurred beforehand.

Table 6 shows the differences in the average skin friction coefficients at tangential laboratory rotor velocities of 9.5, 12.6, and 15.8 knots for the four investigated FCCs during the entire exposure period, including the values for the mechanically cleaned conditions. The average differences in the skin friction coefficients between the first and second runs of drag measurement are also shown; this data indicate that some biofouling was released, as the skin friction coefficients decreased between the first and second run. The results from the first run indicate the capability of an FCC to resist and release biofouling, while results of the second run primarily provide the drag of well-attached biofouling, which could be expected on a moving ship. The small difference in the skin friction coefficient for the hydrogel-based FRC containing biocides is explained by the fact that almost no biofouling is attached during the 53 weeks of exposure.

As a potential way to improve the long-term setup, it was considered to install a torque sensor on the aging setup to avoid the regular detachment of cylinders and obtain in situ measurements. In this case, the main gain would be continuous measurements and, therefore, potentially a more detailed knowledge of the development of the FCCs' drag over time. However, extra effects from changes in bearing resistance and biofouling attachment on the top and bottom of the cylinder and shaft would be difficult to extract from the measurements, causing uncertainties that are minimized with the land-based rotor setup. The change in friction of the bearings is expected to vary significantly over time due to wear from a continuous use and the presence in a highly corrosive seawater environment (including seawater splashes), compared to the well-protected and less used land-based laboratory rotor setup. Furthermore, although the aging setup was located in a fairly well-protected location, waves and currents are expected to impact the measurements. The much higher economic cost of installing four torque sensors, which can be expected to have a limited life-time in the highly corrosive marine environment, should also be mentioned.

Fuel and power predictions for a tanker

In this section, an illustrative example of fuel and power predictions for a medium-sized tanker, i.e., Tarantella, (33) is presented. Based on the skin friction coefficients estimated from the laboratory rotor, the power consumption from Tarantella at various velocities and the fuel consumption based on its normal operating speed are estimated. Tarantella has a length between each perpendicular of 176 m, a draught design of 11 m, and a breadth of 32 m, i.e., a wetted surface area of approximately 9540 [m.sup.2]. Tarantella has a typical operating speed of 14 knots, which results in a fuel consumption of 30 tons per day when the hull is free of biofouling. (33) Figure 16 displays the power consumption due to skin friction in the mechanically cleaned condition after 25 weeks (i.e., average of the four FCCs in the mechanically cleaned condition). Furthermore is the power consumption due to the skin friction seen for the four investigated FCCs based on their average skin friction values during the 53 weeks of exposure (skin friction coefficients are seen in Fig. 14). Equation (16) was used to determine the power consumption due to skin friction based on a seawater density of 1025 kg/[m.sup.3] and the laboratory rotor skin friction coefficients. Tangential rotor velocities of 6.3, 9.5, 12.8, and 15.8 knots correspond to ship velocities for the investigated ship of 8.30, 12.35, 16.38, and 20.25 knots, with a kinematic seawater viscosity at 10°C (i.e., 1.37 [m.sup.2]/s).

Figure 16 reveals that there are very small differences in power consumption for the investigated FCCs, besides the fluorinated FRC. The increase in fuel consumption for the fluorinated FRC resulted in an increase of 5.1 tons per day compared to the clean condition when operated at 14 knots. This is based on the assumption that the skin friction is responsible for 70% of the total resistance [e.g., reference (3)] and, therefore, also fuel consumption. With an assumed operating activity of 80% at typical speed (14 knots) and an oil price of $500 USD per ton of bunker oil, an additional yearly fuel expense for Tarantella would be $0.75 million USD.

The visual biofouling rating for the fluorinated FRC during the 53-weeks exposure period ranged between 0 and 50; the majority of the ratings being either 10 or 20, which according to Schultz (31) would result in increase of 11% and 20% in total resistance for an Oliver Hazard Perry class frigate. The increase in skin friction of the fluorinated FCC compared to newly applied condition when evaluated over the entire 53-weeks exposure period was approximately 18%. Assuming that the frictional contribution to the total resistance of the investigated ship is 70%, a total increase in resistance will then be approximately 26%. With an increase of this order for the frigate, the predictions by Schultz (31) would be a biofouling rating of 20-30, which matches well with biofouling observations on the fluorinated FRC cylinder. This shows that the biofouling rating can be used to obtain a first approximate estimate of the biofouling impact on the total resistance and, therefore, also on the fuel consumption. However, the laboratory drag measurements provided a much higher accuracy and less uncertainty than the visual biofouling evaluation.

Conclusions

This study employed a setup to systematically study the drag performance of FCCs over time in the presence of biofouling species (i.e., natural seawater) at typical speeds and activity conditions for larger commercial ships. No significant differences in the drag performance between a hydrogel-based FRC containing biocides, a hydrogel-based FRC without biocides, and a silylated acrylate SPC coating were found over a period of 53 weeks with seawater exposure. The fluorinated FRC showed a drag performance similar to the other FCCs when evaluated in the newly applied condition and in the mechanically cleaned condition. However, it became biofouled after approximately 2 weeks, and remained so with varying degrees throughout the entire exposure periods, which resulted in a significantly higher drag during the majority of the exposure period. The three other FCCs only had limited biofouling attachment, and the newly applied drag performance was, therefore, similar to that found during the 53-weeks exposure period. As the three best performing FCCs hardly changed over the period of 1 year it is suggested that future drag measurements can be carried out with less frequent intervals, e.g., every 2 or 3 months. If, however, significant changes in the level of biofouling are observed via regular inspection, e.g., every 2 or 3 weeks, then drag measurements are recommended. For a coating with substantial changes in drag over shorter time, which was the case for the fluorinated FRC, frequent drag measurements (e.g., every 2 or 3 weeks) are recommended. The best performing coating was the FRC with biocides incorporated. It is believed that it primarily was the biocide release of this hydrogel-based FRC which outperformed the other hydrogel-based FRC without biocides as their copolymer content and type were similar. The fluorinated FRC primarily differed in copolymer additives, which is believed to be the primary reason for the much worse performance compared to the hydrogel-based FRCs without biocides. The SPC coating operates with a very different biofouling prevention method to the FRCs making a direct comparison between the biofouling prevention methods complicated, but the SPC coating was found to perform at a similar level of the two hydrogel-based FRCs.

Since biofouling will likely occur at some time during the service life of an FCC, the newly applied drag performance and any related roughness parameters are poor indicators of the long-term drag performance. However, if an FCC remains free of biofouling, the drag in the newly applied condition can be a valid indicator of the drag performance, even long after coating application. The need for simulating exposure conditions similar to those of ocean-going ships was shown. Static immersion, which often does not represent larger commercial ships, but is widely used to characterize the performance of FCCs, often resulted in a significantly higher drag, compared to the dynamic immersion periods.

DOI 10.1007/s11998-015-9713-0

A. Lindholdt, K. Dam-Johansen, S. Kiil ([mail])

Department of Chemical Engineering, Technical University of Denmark, DTU, Building 229, 2800 Kgs. Lyngby, Denmark

e-mail: sk@kt.dtu.dk

D. M. Yebra, S. M. Olsen

Hempel A/S, Lundtoftegardsvej 91, 2800 Kgs. Lyngby, Denmark

List of Symbols

A Surface area ([m.sup.2])
C Mechanically cleaned
[C.sub.f] Skin friction coefficient
[[bar.C].sub.f] Average skin friction coefficient
D Dynamic
?[C.sub.T] Change in total resistance
?[C.sub.f] Difference in skin friction coefficient
FCC Fouling control coating
FRC Fouling release coating
[F.sub.f] Skin friction resistance (N)
h Height of cylinder (m)
k Von Karmen constant (0.41)
[l.sub.g] Gap distance between inner (rotating)
 cylinder and outer (static) cylinder (m)
[l.sub.p] Length of plate (m)
l Vertical distance from center line to peak (m)
L Length (m)
[M.sub.B] Torque from the bearings (Nm)
[M.sub.C] Torque on the side of the rotating cylinder (Nm)
[M.sub.Cor] Correction factor (Nm)
[M.sub.E] Torque from the top and bottom of the
 cylinder (Nm)
[M.sub.S] Torque from the shaft (Nm)
[M.sub.T] Torque measured by torque sensor (Nm)
n Number of measurements
N Newly applied
NSTM Naval Ships' Technical Manual
p Vertical distance from center line to peak (m)
? Density (g/[m.sup.3])
Pa Pascal (N/[m.sup.2])
r Radius (m)
Re Reynolds number
[Re.sub.g] Reynolds number for laboratory rotor setup
 based on its gap distance between the (inner)
 rotating and outer (static) cylinder
[Re.sub.r] Reynolds number for a cylinder based on its
 radius
[Re.sub.p] Reynolds number for a flat plate based on its
 length
RPM Rounds per minute ([min.sup.-1])
Rt(50) Roughness parameter (µm)
[R.sub.z] Roughness parameter (µm)
S Static
SPC Self-polishing copolymer
TQC Total quality control
[t.sub.w] Wall shear stress (Pa)
[[bar.t].sub.w] Average wall shear stress (Pa)
U Velocity (m/s)
[??] Kinematic viscosity ([m.sup.2] [s.sup.-1])


Acknowledgments Financial support provided by The Hempel Foundation and the Technical University of Denmark is gratefully acknowledged.

References

(1.) Yebra, DM, Kiil, S, Dam-Johansen, K, "Antifouling Technology-Past, Present and Future Steps Towards Efficient and Environmentally Friendly Antifouling Coatings." Prog. Org. Coat., 50 75-104 (2004)

(2.) Lindholdt, A, Dam-Johansen, K, Olsen, SM, Yebra, DM, Kiil, S, "Effects of Biofouling Development on Drag Forces of Hull Coatings for Ocean-Going Ships: A Review." J. Coat. Technol. Res., 12 415-444 (2014)

(3.) MAN Diesel & Turbo, Basic Principles of Ship Propulsion. http://www.mandieselturbo.com/files/news/filesof5405/ 5510_004_02%201ow.pdf. (Accessed on February 12, 2014) (2011)

(4.) Lejars, M, Margaillan, A, Bressy, C, "Fouling Release Coatings: A Nontoxic Alternative to Biocidal Antifouling Coatings." Chem. Rev., 112 4347A390 (2012)

(5.) Zargiel, KA, Swain, GW, "Static vs Dynamic Settlement and Adhesion of Diatoms to Ship Hull Coatings." Biofouling, 30 1-15 (2013)

(6.) Candries, M, Atlar, M, Mesbahi, E, Pazouki, K, "The Measurement of the Drag Characteristics of Tin-Free Self-Polishing Co-polymers and Fouling Release Coatings Using a Rotor Apparatus." Biofouling, 19 27-36 (2003)

(7.) Flack, KA, Schultz, MP, Rose, WB, "The Onset of Roughness Effects in the Transitionally Rough Regime." Int. J. Heat Fluid Flow, 35 160-167 (2012)

(8.) Ghani, MPA, Karim, K, Milani, K, "Experimental Investigation of the Drag Characteristics of Different Ship Hull Coating with Using Rotor Apparatus." Jurnal Mekanikal, 31 92-102 (2010)

(9.) Mirabedini, S, Pazoki, S, Esfandeh, M, Mohseni, M, Akbari, Z, "Comparison of Drag Characteristics of Self-Polishing Co-polymers and Silicone Foul Release Coatings: A Study of Wettability and Surface Roughness." Prog. Org. Coat., 57 421-429 (2006)

(10.) Schultz, MP, "Frictional Resistance of Antifouling Coating Systems." J. Fluids Eng., 126 1039-1047 (2004)

(11.) Swain, GW, Kovach, B, Touzot, A, Casse, F, Kavanagh, CJ, "Measuring the Performance of Today's Antifouling Coatings." J. Ship Prod., 23 164-170 (2007)

(12.) Weinell, CE, Olsen, KN, Christoffersen, MW, Kill, S, "Experimental Study of Drag Resistance Using a Laboratory Scale Rotary Set-Up." Biofouling, 19 45-51 (2003)

(13.) Holm, E, Schultz, M, Haslbeck, E, Talbott, W, Field, A, "Evaluation of Hydrodynamic Drag on Experimental Fouling-Release Surfaces, Using Rotating Disks." Biofouling, 20 219-226 (2004)

(14.) Lindholdt, A, Dam-Johansen, K, Olsen, SM, Yebra, DM, Klijnstra, JW, Kiil, S, "Measurements of Drag Performance of Statically Exposed Fouling Control Coatings Using a Spinning Disk." (2015)

(15.) Millett, J, Anderson, C, "Fighting Fast Ferry Fouling." Fast'97, Conference Papers, 1 493-495 (1997)

(16.) Kiil, S, Weinell, CE, Pedersen, MS, Dam-Johansen, K, "Analysis of Self-Polishing Antifouling Paints Using Rotary Experiments and Mathematical Modelling." Ind. Eng. Chem. Res., 40 3906-3920 (2001)

(17.) Kawaguchi, Y, Ashida, T, Ando, H, Senda, T, "Development of drag reducing antifouling paint and experimental investigation on mass transfer phenomenon near the painted wall." 1-10

(18.) Dick, JD, Nowacki, LJ, "Accelerated, Simulated-Service Exposures of Antifouling Coating Systems for Ships and Navigational Buoys." Organic Coatings and Plastic Chemistry, 39 (1978)

(19.) Pedersen, MI, "Effektvurdering af aleudstetninger i Roskilde Fjord." DTU Aqua-report 230 (2010)

(20.) Gadelmawla, E, Koura, M, Maksoud, T, Elewa, I, Soliman, H, "Roughness Parameters." J. Mater. Proc. Technol., 123 133-145 (2002)

(21.) Molland, AF, The Maritime Engineering Reference Book: A Guide to Ship Design, Construction and Operation. Butterworth-Heinemann, Oxford, 2008

(22.) TQC (Total Quality Control). http://www.hull-roughness. com/product/hull-roughness-analyser-dc9000 (Accessed on November 25, 2014) (2012)

(23.) Anderson, CD, Townsin, RL, Candries, M, "Idiosyncracies of Low Surface Energy (Foul-Release) Antifoulings: Drag, Roughness and Maintenance." Presented by Anderson, CD, at the 10th International Congress on Marine Corrosion and Fouling, Melbourne (1999).

(24.) Schlichting, H, Boundary Layer Theory. McGraw-Hill, New York, 1968

(25.) Theodorsen, T, Regier, A, "Experiments on Drag of Revolving Disks, Cylinders, and Streamline Rods at High Speeds." Natl Advis. Comm. Aeronaut., 793 367-384 (1944)

(26.) Arpaci, VS, Larsen, PS, Convection Heat Transfer. Prentice-Hall, Englewood Cliffs, 1984

(27.) ITTC (International Towing Tank Conference), "ITTC-Recommended Procedures Fresh Water and Seawater Properties." Revision 02, 1-45 (2011). http:// ittc.sname.org/CD%202011/pdf%20Procedures%202011/7.502-01-03.pdf. (Accessed on February 12, 2014) (2011)

(28.) Candries, M, Atlar, M, "Estimating the Impact of New-Generation Antifoulings on Ship Performance: The Presence of Slime." Proc. IMarEST-Part A, J. Marine Eng. Technol., 13-22 (2003a)

(29.) Candries, M, "Drag, Boundary-Layer and Roughness Characteristics of Marine Surfaces Coated with Antifoulings." PhD Thesis. Department of Marine Technology UK: University of Newcastle-Upon-Tyne. (2001)

(30.) Naval Ships' Technical Manual "Waterborne Underwater Hull Cleaning of Navy." http://www.hnsa.org/wp-content/ uploads/2014/07/ch081.pdf (Accessed on December 2, 2014) (2006)

(31.) Schultz, MP, "Effects of Coating Roughness and Biofouling on Ship Resistance and Powering." Biofouling, 23 331-341 (2007)

(32.) Cushing, DH, Marine Ecology and Fisheries. Cambridge University Press, Oxford, 1975

(33.) Shipyard Trogir. http://www.hb.hr/LinkClick.aspx?fileticket= RetQFnntemc=. Accessed November 25, 2014. (2014)

Table 1: Basic parameters describing the FCCs used in
the experiments. All the coatings are solvent-based

Parameter Fouling control coating systems

 Fluorinated FRC Hydrogel-based
 FRC (no biocide)

Topcoat dry film 150 150
thickness (µm)

Application procedure Airless spray Airless spray

Primer Commercial PDMS-based
 tie-coat used for all
 FRCs

Primer dry film 100 100
thickness (µm)

Backbone structure Polysiloxane Polysiloxane

Active biocide None None

Volume solids (%) 74 ± 2 71 ± 1

Color Red Red

Number of components 3 (curing agent, base 2 (curing agent and
 and, accelerator) base)

Maximum force when 7.2 ± 2.4 6.4 ± 1.3
stretched (N)

Contact angle 79.7° 28.1°
(after 60 s) ± 2.1° ± 2.9°

Parameter Fouling control coating systems

 Hydrogel-based SPC coating
 FRC with biocide

Topcoat dry film 150 100
thickness (µm)

Application procedure Airless spray Airless spray

Primer Commercial cured,
 modified epoxy

Primer dry film 100 100
thickness (µm)

Backbone structure Polysiloxane Silyl acrylate

Active biocide Copper pyrithione Copper pyrithione

Volume solids (%) 70 ± 1 58 ± 1

Color Red Brown

Number of components 2 (curing agent and 1
 base)

Maximum force when 6.3 ± 1.8 N/A
stretched (N)

Contact angle 92.9° 88.3°
(after 60 s) ± 1.9° ± 6.4°

Table 2: Tangential velocities of dynamic aging and laboratory
rotor setups and the corresponding ship velocity, which results
in identical average shear stress experienced by the hull for a
smooth surface at 10°C

Setups Tangential Corresponding ship velocity
 velocities (i.e., flat smooth plate) (knots)
 (knots)
 Ship length Ship length
 of 100 m of 400 m

Dynamic aging 8.1 10.4 11.4
Laboratory rotor 3.2 3.9 4.7
 6.3 7.9 8.8
 9.5 11.9 13.0 (a)
 12.6 15.8 17.3 (a)
 15.8 19.7 21,5 (a)

(a) Correlation is beyond the valid
relationship described in equation (13)

Table 3: The table provides the uncertainty of the measured torque

 Newly Cleaned Average of
Tangential applied condition newly applied
velocity condition after 25 and cleaned
(knots) (Nm) weeks (Nm) condition (Nm)

3.2 0.071 0.041 0.056
6.3 0.071 0.037 0.054
9.5 0.064 0.031 0.048
12.6 0.064 0.033 0.049
15.8 0.056 0.035 0.045

Standard deviations from the torque measurements in the newly
applied condition, the mechanically cleaned condition after 25
weeks of seawater exposure, and the average of the newly applied
and mechanically cleaned conditions are shown. The average
standard deviation of the newly applied and mechanically cleaned
conditions after 25 weeks of seawater exposure is assumed to
represent the measurement uncertainty from the laboratory rotor

Table 4: Mean [R.sub.z] and Rt(50) roughness values with one
standard deviation at the newly applied condition, after 25
weeks' immersion with subsequent mechanical cleaning, and after
53 weeks' immersion with subsequent mechanical cleaning

 Roughness
 parameter Hydrogel-based
Coating condition (µm) FRC (no biocides) Fluorinated FRC

Newly applied [R.sub.z] 1.0 ± 0.6 0.8 ± 0.3
25 weeks (cleaned) [R.sub.z] 1.0 ± 0.3 0.8 ± 0.4
53 weeks (cleaned) [R.sub.z] 0.9 ± 0.4 1.0 ± 0.6
Newly applied Rt(50) 46 ± 23 52 ± 25
25 weeks (cleaned) Rt(50) 44 ± 20 74 ± 44
53 weeks (cleaned) Rt(50) 67 ± 43 81 ± 57

 Hydrogel-based FRC Silylated acrylate
Coating condition with biocides SPC coating

Newly applied 1.1 ± 0.3 8 ± 1
25 weeks (cleaned) 1.0 ± 0.4 12 ± 5
53 weeks (cleaned) 1.2 ± 0.4 16 ± 4
Newly applied 39 ± 16 63 ± 21
25 weeks (cleaned) 62 ± 28 81 ± 39
53 weeks (cleaned) 59 ± 31 105 ± 79

Table 5: Predictions of the change in total resistance
(?[C.sub.T]) for an Oliver Hazard Perry class frigate
(FFG-7) with a range of representative coating and biofouling
conditions at speeds of 15 and 30 knots (31)

 % ?[C.sub.T]
Description of condition NSTM rating @ U = 15 knots

Hydraulically smooth surface 0 --
Typical as applied AF coating 0 2%
Deteriorated coating or light slime 10-20 11%
Heavy slime 30 20%
Small calcareous fouling or weed 40-60 34%
Medium calcareous fouling 70-80 52%
Heavy calcareous fouling 90-100 80%

 % ?[C.sub.T]
Description of condition @ U = 30 knots

Hydraulically smooth surface --
Typical as applied AF coating 4%
Deteriorated coating or light slime 10%
Heavy slime 16%
Small calcareous fouling or weed 25%
Medium calcareous fouling 36%
Heavy calcareous fouling 55%

Table 6: Average skin frictions coefficients for tangential
rotor velocities of 9.5, 12.6, and 15.8 knots during the 53-week
immersion period, and the differences between measurements
of the first and second run

FCC system Average skin friction coefficient for 9.5-15.8
 knots during the 53-week immersion period

 ?
 [C.sub.f] [C.sub.f] [C.sub.f]
 (x [10.sup.3]) (x [10.sup.3]) between 1.
 1. run 2. run and 2. run
 (%)

Fluorinated FRC 3.73 3.55 4.95%
Hydrogel-based FRC 3.12 3.05 2.32%
Silylated acrylate SPC 3.08 3.03 1.74%
Hydrogel-based FRC 3.02 3.01 0.13%
 with biocide


----------

Please note: Some tables or figures were omitted from this article.
COPYRIGHT 2015 American Coatings Association, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2021 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lindholdt, A.; Dam-Johansen, K.; Yebra, D.M.; Olsen, S.M.; Kiil, S.
Publication:Journal of Coatings Technology and Research
Geographic Code:4EUDE
Date:Nov 1, 2015
Words:11377
Previous Article:Iron as an alternative drier for curing of high-solid alkyd coatings.
Next Article:Influence of alcohols on stability and mechanical properties of polyacrylate coating material.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |