Document B562q41p1mNJXpQNVKkpKnGjL

Material Selection Criteria for Elastomeric Rotary Propeller Seals for Marine Applications J.W.M. Noordermeerl,* and M.A. Masen2 Professor (em.) ofElastomer Technology and Engineering, Noordermeer Rubber Consultancy, 6118 BV Nieuwstadt, the Netherlands 2 Associate Professor Tribology, Department ofMechanical Engineering, Imperial College London, Exhibition Road, London, SW7 2AZ, UK *Corresponding author: @gmail.corn Abstract The aim of this article is to search for technically feasible alternatives for the fluoro-elastomer material used for Propeller Shaft Seals in marine shipping transport. The shipping industry is vital to our economy and prosperity in the EU/EEA. The shipping industry today is responsible for transporting and delivering more than 80% of global trade, by volume roughly 11 billion tons annually. A backbone of global trade and the most efficient way of transportation of large volumes with high weights, compared to land and air routes. Elastomeric Propeller Shaft Seals (made out of fluoropolymer rubber) separate oily environments in ships from the (sea) water, in order to avoid environmental spillage of oil into the sea and vice versa. The propeller shaft is supported by white metal sliding bearings, fixed in the stern tube. These bearings function within an oil bath; the elastomeric rubber sealings are used to seal the oil bath. Since 2013, the environmentally allowed oils/lubricants have changed from mineral oils to primarily ester-based oils. Sealing is achieved by a series of 5 or more seals in line. Stationary circular lip seals fixed in the stern-tube, slip along the rotating screw shaft. Heat is evolved in the lip contact, which increases the temperature to 130 C -150 C. In the past the seals were manufactured from NBR (nitrile) rubber. Ships were commonly significantly smaller, approximately 1,500 TEU (Twenty foot Equivalent container Units) per vessel in 1970 compared to nowadays 24,000 TEU. Propeller speeds were also lower. Since the 1960's, after the development of FKM (fluor-elastomer) for aerospace and airplane applications, FKM came into use for sealing in the maritime sector as the one and only rubber-material. It was able to cope with the stringent requirements of temperature, oil- and water-resistance in combination with the required safe performance duration of minimum 5 years (the scheduled dockings for major maintenance overhauls of ship vessels) to avoid leakage of the seals, with consequential risk for the environment and to avoid the need to dock the vessels unplanned for seals replacement. The performance requirements for the use of these seals in marine vessels are detailed in this article, as the basis for the search for alternatives for FKM. FKM derives its high endurance from the fluor-carbon (F-C) bond that outperforms all other elastomeric materials. At the same time FKM has the required material features, such as mechanical and abrasion properties, and the resistance to acids and steam generated from the ester-type oils in combination with the high temperatures encountered by the seals. FKM is by far the highest temperature resistant elastomer, with excellent material properties and water/oil resistance with a duration of at least 5 years; there is no alternative rubber material available which can match its properties to meet the required high performance standards for the safe use of these seals in the marine transport sector. Experience over the past decades shows that no alternatives exist that match the combination of characteristics required to substitute FKM. Replacing FKM in propeller shaft seals is hardly realistic and will require at least many years of research. Key words: Maritime industry, Stern seals, Temperature resistance, oil resistance I. Introduction Aim of this article is to review whether there are technically feasible alternative materials for the fluoroelastomer material in Propeller Shaft Seals used in large container vessels. The shipping industry is vital in the world economy. It is responsible for transporting and delivering more than 80% of global trade1, by volume roughly amounting to 11 billion tons annually (year 2022). Around 77% of goods that are imported/exported to and from the European Union are transported by sea2, proving to be a backbone of global trade and the most efficient way of transportation of large volumes with high weight compared to land and air routes3. A vital element of the container vessel is the propulsion system (see figures 1a and 1b). Figure 1a: Propulsion system of a vessel. Courtesy of AEGIR Marine BV. Most enginepropelled ships derive thrust from a propeller or screw. The shaft driving the propeller and transmitting its thrust to the hull of the ship rotates within a metal tube, is called the propeller shaft. This propeller shaft, is supported by white metal sliding bearings, carried in the stern tube. Engine with propellor shaft 4 Figure 1b: Overview of the propulsion system of a ship. Courtesy of AEGIR Marine BV. 2 The stern tube extends to the sternoutside of the ship. Within this tube the propeller shaft while rotating is supported by white metal sliding bearings. These white metal sliding bearings function within an oil bath; the elastomeric rubber sealings are used to seal against leakage and spill of the oil bath to the environment and function also to guarantee the hull integrity. A major part of the tube is flooded with lubricating oil, except for the outer stern side which is in contact with (sea) water. To guarantee appropriate separation between the oil and water environments and to prevent leakage of oil into the water and the environment, a set of rotary lip seals is employed in combination with a series of compartments establishing a gradual transition from the underwater pressure to the oil pressure within the ship's stern tube4. This system of seals is schematically shown in figure 2. Figure 2: Concept of a ships propeller shaft rotating carried within the stern tube with rubber seals installed. Courtesy of Ref. [4]. Not on scale. The seals are typically made of an annularshaped elastomer material (i.e. rubber rings) in order to accommodate vibrations in the rotary shaft and any wear of the stationary seals in contact with the rotating propeller shaft. Depending on the construction of the shaft/tube combination, the transition from oilfilled at the engineside to the water at the sternside encompasses a set of compartments or chambers. On the engine side the chamber is fully oilfilled, and at the propeller side the chamber is waterfilled, with intermediate chambers filled with leaking water/oil combinations, from which the leaked oil/water combination can be drained if required on a regular basis: see figure 3. 3 Figure 3: Stern tube chambers. Courtesy of AEGIR Marine BV. It dictates the need for the rubber seals to be resistant to oilenvironments as well as (sea)waterenvironment, which in terms of properties are two conflicting requirements for rubbers. During the last decades, the requirements for seals have risen drastically. The rotational velocity of the shaft relative to the stationary seal varies from zero to approximately 8 m/s (linear speed) for the largest ships whilst operating at the highest rotation speed. Over time ships have increased significantly in size. To give an example. Could a container vessel transport in 1970 approximately 1,500 TEU (Twenty foot Equivalent Units), nowadays a container vessel transports 24,000 TEU. And also the diameters of the propeller shafts as well as the rotation speeds of the shafts increased the last decades. The corresponding linear rotational speed therefore has increased as well. The performance requirements of the seal materials have kept pace with these developments. The stationary rubber seal in contact with the rotating propeller shaft is subject to friction and consequent heat evolution in the seal/shaft contact5. The replacement of such seals is a time consuming major overhaul, requiring docking of the ship. It is therefore in the interest of the shipowner and the environment to reduce the need for replacement of the seals to a minimum. Typically, docking takes place once every five years, meaning that the safety and durability of the seals for at the least that period is required. Due to increasing and conflicting requirements such as high temperatures and oil and water resistance, the selection of the most appropriate material for the seals from the perspective of long perseverance is a point of major concern and a serious challenge, as detailed in the present review paper. 4 II. Rotary Propeller Seals for Marine Applications The design of rotary propeller seals is based on an elementary concept of elastomeric lip seals57, as shown in figure 4. The fact that the lip is hinged and with the circumferential Garter spring position the seal is in concept asymmetrical. In practice this results in the need to fit the seal correctly to avoid leakage, and/or to avoid unreliable performance results. The contact edge with the shaft is formed at the intersection of two conical surfaces and defining the respective sealing angles ( and in figure 4) is at the discretion of the designer. There are limits which are dictated by manufacturing and performance criteria but within these, variations will be found when examining seals from various supplies8,9. The contact between lip and shaft can be approximated from a simple geometric analysis which indicates that for a specific amount of radial wear, the cotangent of the angle in Figure 4 defines the wear. With a small angle the wear is increased by misalignment of the housing and any eccentricity of the propeller shaft. As a consequence of these factors, most modern designs have angles with the shaft of the order of 20 or more in the fitted state, and some may be up to 40. The other angle is determined by the need to have enough rubber at the contact to impart rigidity. Typically, the included angle forming the elastomer contact is between 90 and 120. Since the diameter at the lip contact is smaller than the propeller shaft diameter, the deformation of the elastomer, which includes both bending and tension, causes a sealing force, which is complemented and stabilized by the force from the circumferential Garter spring. Figure 4: Typical crosssection of an elastomeric rotary shaft lip seal with Garter spring10. Air Side/Back Side may also be the Water Side. Additionally, the pressure difference between the two sides of the seal translates into an extra radial component to the sealing force. The total value of the sealing force is known as the radial load. The contribution from the elastomer depends on its modulus or stiffness, the sectional shape of the lip, and the interference with the shaft, with each of these parameters being adjustable within limits. Garter springs are fitted not only to apply the load, but also to compensate for permanent deformation, creep and wear which elastomers undergo when subjected to permanent strain, and friction, augmented by heat. For a newly produced seal, the sealing edge is sharp. After mounting the rubber seal on the shaft the sealing load causes the contact to flatten and the contact patch has a width of about 5 0.5 to 2.0 mm11. Rotation of the shaft subsequently abrades the elastomer to form a wider contact band, which may vary considerably in size depending on the type of rubber and compound formulation used. The surface finish of the rotary shaft does have an influence on the amount of (potential) wear12. The size of the contact band stabilizes after a short period, which indicates that an intermediate coherent lubricant film is formed between the seal lip and the shaft13,14. Direct measurements and values calculated from frictional results indicate that this film is of the order of 1 m thick. It is difficult to perform highly accurate direct measurements, because the surface structure of the elastomer often has a similar roughness as the shaft. Various base elastomers or the same elastomer with alternative filler systems give different film thicknesses and consequently different frictional results, which highlights the complexity. There is a certain pumping action of the seal lip due to surface tension effects, which in steady state are counterbalanced by capillary forces determined by seal angles, film thickness and also the surface tension9. Under operational conditions, this results in hydrodynamic lift of the sealtip leading to the lubricating film. The buildup of a lubricant film is also due to the surface roughness asperities on the elastomer which are exposed or formed during beddingin and the macroscopic waviness of the sealing surface. The presence of a full lubricant film implies that the friction force is based on the viscosity of the fluid. As the shaftspeed increases, the heat generated in the contact causes a rise in the underlip temperature which for a hydrocarbonbased lubricant results in a reduction of the viscosity. A second influence is the variation in the film thickness, which is a function of lubricant viscosity and speed of the shaft, the governing principle being the need for the hydrodynamic lift to balance the sealing force. The very thin film gives rise to high shear rates and, hence the temperature under the lip can be substantially higher than that in the bulk fluid. Whilst there are disadvantages to the hydrodynamic formation of the thin lubricant films, they are also an integral part of the sealing mechanism. For example, if a lubricant film with an excessive thickness develops, it will usually lead to leakage. Where the foregoing basically applies to the rotary shaft seals in direct contact with the oil side of the construction, similar phenomena apply for the (sea)water side, where a thin waterfilm is present between the seal and the shaft. This highlights the need to consider the operation of the seals under study in the present document, in contact with both types of fluids. III. Typical Rotary Propeller Seal Configurations Figure 5 shows a typical configuration of a fivefold sealing system for rotary propeller seals, creating four chambers. From left to right: one chamber filled with a water/oil mixture and three oilfilled chambers. As all seals operate under the same mechanical conditions and the segregation of the oily and water environments are not clearly defined, all seals are made out of the same rubbery material. 6 This construction aims to minimize the migration of oil into the sea water and environment and of water into the oil. The intermediate chambers are operating under a slight overpressure to prevent excessive leakage of water and or oil underneath the seal lips via the hydrodynamic films, as discussed in the previous paragraph. Figure 5: Typical example of a rotary seal configuration with 5 seals and an oil/water filled chamber4. Other configurations with more seals are also employed, as well as configurations where air chambers are present and where leaked water and oil are collected in one of the chambers. Figure 6 shows a closeup of a seal. Figure 6: 3Dpicture of an FKMrubber rotary seal element running on a stainless steel propeller shaft4. 7 IV. Performance Requirements On basis of the 5 years standard maintenance overhauls schedule for ships and corresponding expected high performance of the seals, the following rubberrelated requirements may be formulated for the use of these seals in the marine transport sector: 1. A high continuous use temperature (minimum of 130 oC), well above the maximum temperatures which may occur during operation of the seals, during at least 5 years; 2. Longterm chemical resistance towards oil: limited or preferably no swell; 3. Longterm chemical resistance towards (salt)water; 4. Sufficient mechanical properties: tensile and tear strengths, fatigue strength, abrasion resistance; 5. A static modulus (or alternatively hardness), sufficient to carry the axial force exerted by the Garter spring; 6. Low compression set (permanent deformation under compression or creep) at room temperature, at 0 oC and at operating temperatures for various durations; 7. Brittleness temperature in marine applications 5 oC. In order to specify these requirements in more detail: Continuous use temperature Firstly, the maximum permitted temperature for continuous use of a rubber (e.g. an elastomer) cannot be taken in isolation, without a link to its service conditions. The load and duration of loading, whether the temperature is continuous or cycles intermittently and the action of surrounding media (oil and (sea)water in the present case), in air or in anaerobic conditions, all play a decisive role. But most importantly, the duration at which the heat acts plays a key role when it comes to the upper temperature limit required in practice. Various industries employ different criteria to define the capacity to withstand the highest achievable temperatures, from e.g. 22,000 hours continuous temperature load for the cable industry, to 1000 hours frequently used in the automotive industry. Hence there is no clearly defined concept of continuous use temperature and consequently different figures are found in literature. Clearly, the longer the exposure to high temperature, the lower the corresponding continuous use temperature. For the rotary stern tube seals to last for 5 years (minimum), equally 44,000 hours and assuming approximately 50% operational time of the propeller shaft, it is most appropriate to adhere to a duration of 22,000 hours active use and use the 22,000 hours continuous use temperature data. Figure 7 shows the measured temperature development versus the linear shaft speed (m/s) during constant operation of both seal #1, i.e. the one closest to the (sea)water, and seal #2 which operates in the water/oil transition stage as shown in Figure 5. All data were collected in fourfold with thermocouples mounted maximum 1 mm from the contact between the seal lip and rotating shaft. Measurements were done at three speeds, maximum 6.1 (m/s) due to equipment limitations. On the other hand, for large vessels sailing at full speed the linear speed may reach 8 (m/s), well above the limitation of the experimental setup. However, all data points develop in moreorless a linear manner vs. the rotational shaft speed. For that reason it seems reasonable to expect that the temperatures at 8 m/s corresponding to the largest vessels can be obtained by extrapolation to result in avg. 130oC for seal #2. And 8 because of the fact that the temperature sensors are away from the actual contact by +/ 1 mm, it is reasonable to expect that the actual temperature is still a bit higher due to heat leakage to the rubbermass around the lipcontact. Seal #1 operates at a lower temperature because of its direct contact with the waterside, whilst the operational temperatures for seals # 3, 4 and 5 are comparable to those of seal #2. To conclude: the data do demonstrate the high temperature requirement (130 oC to 150 oC) for the rubber seals, to be seen in relation to the need to perform well for a long duration of 5 years. Figure 7: Seal lip temperatures (oC) vs. linear shaft speed (m/s) measured for seals #1 and #2 (ref. Figure 5) in the lips at max. 1 mm from the contact with the rotating shaft. Each measurement is an average of 4 measurement points. By courtesy of AEGIRMarine B.V., the Netherlands. Swelling The second main performance requirement for the seal pertains to the tendency of rubbers in general to substantially swell when in contact with liquid media with a close or similar polarity or solubility parameter15,16. As rubbers are basically crosslinked liquids, they want to 9 enter in solution, which is prevented by the crosslinks. A very significant amount of swelling may take place, depending on the degree of crosslinking and the viscosity of the liquid medium. The resistance of rubbers to swell in contact with oil is commonly measured by immersing the rubber at elevated temperature, depending on the rubber type e.g. for a duration of 70 hrs as given in Figure 8. In addition to the maximum permitted temperature, also the duration of contact with the liquid medium is an important factor in the present context, i.e. 44.000 hrs for 5 years, or 22.000 hrs at 50% operation of the vessels. The heat resistances or max. continuous use temperatures for the various elastomers are given in Figure 8 for 1000hrs continuous use, as commonly specified by the automotive industry. As an empirical rule, each increase in exposure time by a factor of 2 (two) lowers the continuous use temperature by 10oC. To reach 44.000 or 22.000hrs continuous use, 4 5 steps of factors of 2 are needed relative to 1000 hrs. This corresponds to a decrease of appr. 45oC continuous use temperature relative to the values presented in Figure 8, for all rubbers given. To predict continuous use temperatures for 22.000 hrs, appr. 45oC have to be subtracted from the temperatures given for 1000 hrs in the plot Figure 8: Continuous use/service temperature vs. Volume swell in ASTMoil type 3, for various rubber types for automotive applications17,18: 1000 hr continuous use. To predict continuous use temperatures for 22.000 hrs, appr. 45oC have to be subtracted from the temperatures given for 1000 hrs in the figure. Oilswell test for low volume increase rubber was done at 150 oC, others at 70 oC. For clarification of abbreviations, see text. In practice, the balance between continuous service temperature19 and the oilswell resistance is the first selection criterium for rubbertypes to be used in specific applications. Figure 8 provides such a comprehensive overview of practically all rubber types, where the used ASTMoil type 3, also known as ASTM Industry Reference Material, IRM 3 is a reasonable 10 representative for most mineralbased oil types20 For further details on the oil types used in stern tubes in contact with seals, see the next paragraph. To conclude: in relation to rotary marine propeller shaft seals in permanent contact with oil, the types of rubbers located at the very low volume increase side (the right hand side of Figure 8) are clearly most relevant. Their abbreviations stand for: NBR acrylonitrile/butadiene rubber with high acrylonitrile content, shortly "nitrile rubber"; XNBR - carboxylated nitrile rubber; CO and ECO - epichlorohydrine rubber; and FKM - fluor rubber. Mechanical properties Key functional mechanical properties for rubbers in seal applications include the tensile strength, tear strength, fatigue strength and abrasion resistance. These parameters should be related to the mechanical load during operation, as well as compared to alternative rubbers. The static modulus (or, alternatively, and directly related to the modulus, the hardness of the rubber) should be high enough to carry the load of the Garter spring and the effects of counterpressure of the oil without major deformation. The compression set of a rubber, defined as the change in thickness under a specific load and duration of loading, at low, room and high temperatures is a sort of creeptest. A sample of predefined dimensions is commonly held under 25% compression during a predefined time period and at set temperature. After release of the load on the sample, the amount of deformation to which the sample recovers compared to its original thickness, within 30 minutes is registered. The remaining deformation relative to the imposed deformation of 25%, expressed in percentage, is called the Compression Set21. The lower the remaining deformation, the better. For proper understanding: if the sample returns to its original thickness with zero deformation remaining, the Compression Set is 0%: the best achievable. Worst is 100%, when the sample does not return at all. It is important to perform this test at various temperatures, especially at low and high temperatures. For instance, if the rubber tends to crystallize at low temperature, it will have a very negative effect on compression set. At elevated temperature the compression set parameter provides an indication for ageing phenomena taking place during the compression exposure. The brittleness temperature marks the point where the rubber, coming from deepcooling to - 70 oC whilst heating up slowly, recovers its elastic rubbery properties, i.e. returning from the glassy or brittle state. It is an important performance criterion for rotary shaft seals22. To conclude: for proper functioning, rubbery behavior is required along the temperature range, from low, freezing temperatures to the elevated temperatures occurring at the seal tip in contact with the propeller shaft (as previously discussed, over 130 oC). V. Oil types employed in Marine Applications Governmental regulations limit the use of oils and lubricants in applications where there is a risk of environmental damage. In December 2013, the use of Environmentally Acceptable Lubricants (EALs) became mandatory in large ships sailing within the coastal waters of the USA by the Environmental Protection Agency (EPA)23. The German Blue Angel, the European Eco label24, and the American Vessel General Permit (VGP) are the most wellknown labelling programs for EALs. A lubricant can only have the label of an EAL if it meets the requirements 11 of the VGP: when it is biodegradable, nonbio accumulative and minimally toxic. As a consequence, a large range of lubricants tailored to comply with these criteria have been introduced in the marine lubricants world. The American VGP 2013 allows four kinds of base oils for the formulation of EALs: Hydraulic oil Environmental Triglycerides (HETG) types; Hydraulic oil Environmental Ester Synthetic (HEES); Hydraulic oil Environmental Polyglycol (HEPG) and Polyolefins PAO or HPER)25. It is the responsibility of the oil product manufacturers to meet the EPA's EAL definition. The HETG are lubricants obtained from plants and animal fats. Their quick aging when exposed to water and heat makes them unsuitable for hydraulic systems. The strict European EcoLabel program restricts the content of a high fraction of natural esters or synthetic from the renewable resources in the formulation of marine lubricants25. The second and third classes HEES and HEPG are produced by esterification of carboxylic acids and alcohols, mainly glycerol. These base oils can be specially tailored to the application by selecting the proper acids and alcohols and therefore are the obvious choice for most ship owners. The fourth class, PAOs obtained from polymerization of olefins, are basically nonpolar in nature and are often mixed with esters, acting as carriers of polar additives to increase the additive solubility. There is an ongoing discussion as to whether or not PAO's are actually EALs, since they do not meet any renewable source standards and only the low viscosity types are somewhat biodegradable. However, the synthetic esterbased oils (HEES and HEPG) are susceptible to hydrolysis and regenerate acids at higher temperatures. These acids may attack the sealing ring, especially when mixtures or emulsions of oil and water build up in any of the chambers in the sealing system or in the stern tube. Consequently, the lifetime of the liptype sealing system may decrease due to this aggressive mixture of oil and hot water or steam. VI. Material selection for Marine Rotary Propeller Seals on Basis of Performance Requirements Basically only four elastomers need further consideration for application in marine rotary stern seals, as detailed above and shown in Figure 8, based on the balance between continuous use and oil resistance requirements: NBR acrylonitrile/butadiene rubber with high acrylonitrile content, shortly "nitrile rubber"; XNBR - carboxylated nitrile rubber, a special grade of NBR, vulcanizable with zinc oxide, not relevant in the present context; CO and ECO - epichlorohydrine rubber are propertywise largely comparable with NBR, but distinctly higher priced; and FKM - fluor rubber. FFKM, perfluorinated fluor rubber, is extremely highly priced and only used for the highest achievable temperatures: a step too far in the context of the 12 present propeller shaft seals. To conclude: because of the beforementioned, the following discussion focusses on nitrile rubber NBR and fluoroelastomer FKM. Table 1 provides an overview of typical properties obtained for NBR with low and high ACN content, and FKM.17, 26, 27 NBR is a random copolymer of butadiene (CH2=CHCH=CH2) and acrylonitrile (ACN) (CH2=CH CN), wherein the ACN is a highly polar monomer which provides the oilresistance vs. the butadiene phase, which has no oil resistance. NBR is commercially available in a variety of ACNcontents ranging from typically 18 to 51 wt%. The lower the ACNcontent, the higher the butadienelevel and consequently the lower the oil resistance, which is accompanied by a lower glass transition temperature, which does enhance the dynamic rubbery properties at low temperatures. The range of oilresistance for NBR marked in Figure 8 covers both types, where the utmost left point corresponds to the high ACN variants with the poorest dynamic properties. The glass transition temperature by itself cannot be taken as a measure for dynamic properties in comparison with other rubbers. In particular the fact that the glass Table 1: Comparison of typical elastomeric properties between NBR and FKM. Property Standard NBR NBR FKM Ref. Low ACN High ACN Low temp. glass trans. ISO 2921 28 10 13 17, 26 TR10 (oC) Tensile strength *) ISO 37 M/H M/H M/H 17 Resistance to tear *) ISO 34A M M M 17 Resistance to abrasion *) DIN 53516 M/H M/H H 17 Compression set at ISO 815 40 45 50 17 20oC / 70 hr (%) Compression set at ISO 815 8 8 18 17 Room temp. / 168 hr (%) Compression set at ISO 815 50 55 20 17 + 120oC / 70 hr (%) Compression set at ISO 815 17 27 + 200oC /22 hr (%) Continuous use working 80 110 85 120 180 250 17, 27 temp. (oC)**) Max. Working temp., 125 125 250 17 short duration Swelling after 70 h in ISO 1817 5 (100 oC) 25 (100 oC) 2 (150 oC) 17 ASTM oil #3 (%) *) M: medium; H: high, acc. to common rubber standards. **) Highest temperatures typically apply to 1000 h continuous use; lowest temperatures typically apply to 10.000 h continuous use. transition temperature for the high ACN NBR is practically the same as typically for FKM, does not have predictive value for the dynamic properties of FKM. It basically means that both types of rubber become glasslike brittle at about the same subzero temperature of 10 oC. 13 FKM was developed by the DuPont Company in 1957, in response to the extreme performance sealing needs in the aerospace industry. The type of FKM under consideration here for marine lip seals is a random copolymer of vinylidene fluoride (CH2=CF2) and tetra fluor ethylene (CF2=CF2) which is vulcanized or crosslinked with a peroxide curing system. Most conspicuous for FKM in comparison to NBR is the continuous service temperature, which is approximately 100 oC higher, compared under same testconditions. It highlights the unique range of temperatures for FKM (180 oC) relative to NBR (85 oC). Another important performance criterium for FKM vs. NBR is its very high abrasion resistance, particularly in relation to its expeted performance of min. 5 years dictated by the scheduled regular maintenance overhaul dockings for ship vessels. In ECHA Annex XV Restriction Report, Proposal for a Restriction for Per and polyfluoroalkyl substances (PFASs)28, page 351, Table E.114, the statement is made that NBR and CR (Polychloroprene or shortly Neoprene) are generally suitable for waterlubricated bearings in stern tube seals for marine vessels, though at the cost of inferior friction and wear characteristics compared to PTFE (PolyTetraFluoro Ethylene or shortly Teflon). The use of CR is highly unlikely because CR is positioned on the scale of Oil Resistance in Figure 8 right in between oil and waterresistant: basically not compatible with either medium, at least not for sufficiently long time. The high temperature range for both rubbers quoted to be + 150 oC does apply for short durations of just a few hours at most, to be compared with the highest continuous use working temperature of 120 oC for NBR for automotive continuous use of 1000 h. In the past, NBR was widely employed for stern tube seals, however with the advent of FKM it was essentially fully replaced in response to the ever increasing requirements by larger ships. In terms of time temperature application range, FKM clearly stands out as the best performing of all available elastomer/rubber types, as demonstrated in Figure 8, and confirmed in ECHA Table E.114 by a lifetime of NBR of approx. 10% of fluorocarbon and above 100 oC even lower. For a comprehensive overview of the eventual other alternatives for FKM with reference to the performance criteria mentioned: see Table 2. To conclude: comparing the mechanical properties for highACN NBR and FKM, they seem roughly comparable, where FKM positively stands out on the compression set at +120 oC, again on basis of its much higher temperature resistance in combination with durability and thus safety (see in this context also the issue on blisterformation, mentioned below). 14 Substance FKMFluoroelastomers A potential alternative substance for FKM for use as seals and/or in sealing applications in marine shipping; must fulfil - at least the following cumulative requirements: a) continuous high temperature as from 130 oC and much higher degrees; b) longterm chemical resistance to oil; c) longterm chemical resistance to (salt)water; d) sufficient mechanical properties: such as tensile and tear strengths, fatigue strength, abrasion resistance; e) sufficient static modulus or hardness to carry axial force exerted by the Garter spring; f) low compression set at different temperatures between 0 oC and operating temperatures for various durations; g) brittleness temperature in marine applications 5 oC; and h) need to perform safely for a duration of 5 years (drydocking) Potential Alternative substance Suitable alternative? HNBR No, because HNBR does not fulfil requirement(s) a), b), c) and h). FVMQ No, because it fails severely on c) and d), and so on h). ACM No, because it fails on a), d), g) and h). High CAN NBR No, because it fails on a), d), g) and h). Table 2: Overview potential alternatives for FKM for use as seals and/or in sealing applications in marine shipping. VII. Blisterformation A common failure mechanism in seals is the formation of small blisters on one or both sides of the contact of the seal lip with the shaft, Figure 9. These blisters are most commonly found on seals that are in contact with sea water. The phenomenon has been extensively studied and occurs on both elastomers used for rotary propeller seals, however it is more common on NBR than on FKM29,30. For NBR the formation of a blister requires a temperature at the lip contact of 130 oC for a longer duration. 15 Under these conditions, NBR shows excessive hardening and subsequent blister formation at either side of the contact. Blister formation was originally one of the major factors in the switch from NBR to FKM once the latter became available as an alternative material. Figure 9: Example of blisters alongside the seal lip of a NBRseal. The root cause of these blisters remains a point of discussion in the scientific literature in spite of all research devoted to this phenomenon. The most comprehensive study of the phenomena was reported by Fr. Schultz31, who refers to a combination of factors, which each individually or jointly may cause the blisters to develop: Dynamic load and, as a consequence, fatigue by which small cracks are formed. These are filled with oil or water by capillary forces due to the pressure under the lip. These subsequently grow to a visual size whilst volatile enclosures in the rubber evaporate when operating at a persistent high temperature. Additives included in the oil for reasons of viscosity control and thermal stabilisation, in particular aminecontaining compounds. Apparently, oils without additives lead to very little or no blister formation. Also by this author29 the pivotal role of misalignment and the resulting variable loads. Contact of at least one side of the seal with air, in combination with high temperatures. This points obviously to oxidative ageing. Starvation of the lubricant film under the lip as a result of insufficient transport. To this listing has to be added the much more aggressive nature of the present EAL oils, which were still not in focus at the time of the Schultz study31. While all these failure mechanisms occur with NBR and to some extent with FKM as well, it is well understood that FKM outperforms NBR on these critical aspects because of its far better thermal stability. FKM seals are overall chemically far more resistant than NBR seals. Table 3 provides detailed test results for compatibility of of a selection of EALs with NBR and FKM. Table 3: Compatibilities of various oiltypes with NBR and FKM, with further classification as VGP EAL compliancy. By courtesy of Lagersmit32. 16 Company name Castrol Chevron ExxonMobil Fuchs Gulf Oil Marine Klber Shell Product name Biostat 68 Biostat 100 Biostat 150 Biostat 220 Clarity synth. EA Gear Oil 100 Clarity synth. EA hydraulic Oil 46 Clarity synth. EA Hydraulic Oil 68 SHC Aware ST 100 SHC Aware ST 220 Plantosyn 68 HVI Plantogear 100 S Plantogear 150 S GulfSea BD Sterntube Oil 68 GulfSea BD Sterntube Oil 100 GulfSea BD Sterntube Oil 220 Klberbio RM2150 Klberbio EG268 Klberbio EG2100 Shell Naturelle S4 Gear Fluid 68 Shell Naturelle S4 Gear Fluid 100 Viscosity [cSt @ 40C] 70 NBR * compatible * FKM compatible VGP EAL compliant *** 103 * 144 * 207 * 100 X 46 X 68 X 100 * 220 * 68 X 100 X 150 * 68 * 100 * 220 * 150 X 68 X 100 X 68 X 100 X 17 Terresolve Envirologic 200 68 X Envirologic 210 100 X Envirologic 3046 46 X Envirologic 3068 68 X Total Biohydran TMP 100 100 * Bioneptan 100 100 ** Bioneptan 150 150 ** ** Bioneptan 220 220 ** ** Vickers Hydrox Bio 68 68 * Hydrox Bio 100 100 * Hydrox Bio 220 220 * Biogear XP 68 68 X Notes: EALs can chemically affect the sealing rings by hydrolysis. Especially when emulsions are built up in the oil chamber of the sealing system or in the stern tube, these biooils interact with the water present and tend to break down. The lifetime of any liptype sealing system can decrease due to this aggressive mixture. FKM seals are chemically more resistant than NBR seals * NBR compatibility was tested with clean oil. Hydrolysis and higher operating temperatures than 40 C may limit the lifetime of NBR lip seals. VIII. Endoflife disposal In the context of the recent proposal of ECHA to impose a restriction on the production and use of PFAS (Per and Polyfluoralkyl substances), published on 07.02.2023, the question of disposal of end oflife articles is a major point of concern because of their extremely high persistence and accumulation in nature28. During docking and overhaul of ship vessels, the stern tube seals are commonly onebyone replaced by new ones, recovering as many used seals as installed new ones. Apart from the seal lips which have been in continuous contact with the propellor shaft during extended use, most part of the seals have only been in contact with oil, (seawater) or a mixture of both. The lipcontact represents only a minor part of the seal. The FKM by nature not being compatible with oil nor with water will not have absorbed relevant quantities of either one. In fact, the 18 seals represent a rather pure material, apart from contaminations which may have adhered to the surface, but can be removed easily by proper cleaning. Ladder of Lansink Wornout seals into new Material reuse Back to feedstock Energy recovery Figure 10: Ladder of Lansink.33 The various ways to discard waste can best be portrayed in terms of the socalled ladder of Lansink33, Figure 10, listing the hierarchy in waste management for endoflife seals from high to low: Prevention Reuse of articles into new Recycling by material reuse Back to feedstock Energy recovery by burning/incineration Disposal Disposal on a landfill: is not considered as a feasible option to consider by virtue of the primary objective of the ECHA restriction proposal: to prevent the extremely high persistence and accumulation in nature of PFASs and so of FKM. Incineration with Energy recovery: At the present moment in time, most wornout seals are burned /incinerated: the lowest step on the ladder above. There have been various studies as to the conditions necessary in waste incinerators, demonstrating that at sufficiently high >850oC oven temperature the combustion products are practically only HF (hydrogen fluoride) and CO2 (Carbondioxide)34,35. Minor to trace amounts of low molecular PFASs may still be traced in the CO2 recovery. According to Bakker et al.34 all waste incineration plants in the Netherlands fulfill on average the requirement of an incineration temperature > 850 oC. No statement is made about the feasibility of energy recovery, although the amount of energy to be recovered in municipal and industrial waste ovens by burning the relatively low amounts of FKM is negligeably small. The company AEGIR Marine in the Netherlands has recently implemented a circular collection system for wasted seals to be collected and returned to their homebase in the Netherlands. 19 From thereon they will be kept separate to be burned under controlled conditions as depicted above. Under implementation. Back to feedstock: On basis of their compound composition the rotary propeller seals consist for appr. 70 - 80 wt% out of FKM elastomer and for appr. 30 - 20 wt% of N990 carbon black. N990 carbon black has no practical value as reinforcing filler because of its largest primary particle size in the whole range of reinforcing and nonreinforcing carbon blacks17. Its main role in the compound is for cost savings and as a pigment to stabilize the material against UV attack (not an issue in the present context). Where the FKM cannot be recovered without the rupture of the crosslinks (see later under devulcanization) the only back to feedstock option is pyrolysis of the rubber. Pyrolysis at temperatures around 600700oC to result in a vapour phase, oilphase and a solid phase called carbon black, is a back to feedstock option catching much attention these days. Particularly for tirewaste. The carbon black phase consists of reinforcing carbon black to be reused in the production phase of the tire compounds. It also includes pyrolyzed remains of the elastomers, with additionally mineral fillers and remnants of the vulcanization ingredients36. As FKM compounds contain practically no carbon black, except some quantities of N990 black, any pyrolysis remnants are per definition lowmolecular weight PFASs and as such this technology for back to feedstock does not lend itself for small size fluoroelastomer consumers, but rather for a large scale operation at e.g. fluoropolymer producers, where the low molecular weight PFASs could immediately be reused for renewed polymer synthesis; if at all feasible. Material reuse: Grinding towards submillimeter powder and mixing it into virgin rubber compounds is often employed for FKM. The smaller the ground particles, the less damaging for the mechanical properties. Commonly appr. 5% powder can be accommodated in virgin material with limited loss in properties. Some people quote amounts till 15%. It is most commonly employed for FKM with production waste. As seals are replaced on a onetoone basis much higher quantities of regrind should be feasible than 5% to accommodate this route for recycling of used seals. Furthermore, on 30 August 2022, the European Commission published a proposal to restrict the placing on the market of microplastics, including where they are added in mixtures. Microplastics are defined here as < 5mm. The restriction will be adopted under the REACH Regulation, which establishes the EU chemicals framework. According to this recent proposal from ECHA there will be a ban on microplastics production and use. Consequently, grinding used rubber to particles smaller than 1 mm and adding them to virgin rubber may soon come under scrutiny37. It does not seem a feasible solution for the long term. Recycling as an option for worn out seals into new: Reclaim/devulcanization. Apart from reclaiming which is a devulcanization technology developed in the second world war for Natural Rubber and still employed in large quantities, reclaiming or rather devulcanization of synthetic rubbers, to include FKM is still in its infancy. For a statusreport see ref.38. The devulcanization technology has proven itself for Natural and EPDMrubbers and is in the development stage for the other rubbers employed in tyres: SBR, BR and IIR (Butyl) rubbers, under supervision of Assoc. Prof. Dr. W.K. Dierkes of the University of Twente, Enschede, the 20 Netherlands38. To extent this technology to FKM requires a totally new approach, because of the special unique nature of fluoropolymers. At present discussions are underway between a consortium of companies under leadership of AEGIRMarine in the Netherlands, to join forces with the University of Twente to screen the feasibility of devulcanization for FKM. This will aim at the highest step of the ladder of Lansink with substantially increased percentages of reuse than the 515% for grinding, preferably approximately 50%. Special attention has to be paid to avoid the risk of formation of lowmolecular PFASs as byproducts! To conclude: there remain basically two feasible options on the long term to dispose of used endoflife seals, which are incineration at sufficiently high temperature and reclaim/devulcanization, the latter still in its infancy needing extensive research and development efforts. IX. Conclusions: no suitable alternative for FKM, extensive R&D needed Elastomeric Rotary Propeller Seal Systems for Marine Applications typically consist of a series of circular seals in a row, allowing for a gradual transition from oil in the inside of the stern tube to (sea)water outside the hull of the ship. Material selection of the appropriate elastomer for such seals is a potential choice between NBR (nitrilerubber) and FKM (fluoro rubber). Over the years, the increase in size and tonnage of ships has resulted in a growth of thrust, and lead to larger and faster rotating screw shafts within the stationary seals. Due to the inevitable and functionally necessary contact force and resultant friction between the seal lip and the propeller shaft, the temperature of the seal contact steadily increased and can be up to a minimum of approximately 130oC. This temperature greatly surpasses the maximum allowable temperature for NBR of 85oC to enable proper and safe operation for (at least) 5 years before requiring replacement during a ship's scheduled overhaul. This has gradually resulted in a switch away from NBR, and at present FKM is used almost exclusively as this elastomer permits a substantially elevated continuous use temperature, of 100oC extra. Furthermore, FKM is better resistant and inert to oils, surfactants and (sea)water. This is increasingly important as according to recent American and European legislation, the use of Environmentally Acceptable Lubricants based on esters of carboxylic acids is mandated for stern tube purposes. These EALs are more aggressive to the rubber seals than purely mineral oils, as these tend to decompose by saponification under the generation of organic acids in contact with hot water or steam at the high temperatures occurring under the lips. Also in this respect FKM cannot be substituted by NBR. The root cause of the much better properties of FKM over NBR is based on the fluorocarbon bond prevailing in FKM: Table 4. At this moment in time incineration of endoflife seals at high enough temperatures >850 oC is practically the only way to deal with proper disposal. Efforts are being undertaken to see whether reclaim/devulcanization can replace this way of disposal in order to achieve an acceptably higher proportion of reuse of the still valuable FKM. 21 Experiences over the past decades show that no alternatives exists that match the combination of characteristics required to substitute FKM. Replacing FKM in propeller shaft seals is practically not realistic and will require at least many years of research. Table 4: Typical Bonddissociation temperatures and bond energies of some typical chemical bonds in relevant vulcanised elastomers15. Chemical Bond 1 CF2 CF2 2 H3C F 3 CH2 CH2 4 H3C H 5 CH 2 CH2 CH = CH 6 CH2 = CH CH2 H 7 C S S C 8 C Sx C X 3 Tdiss in oC 500 400 390 320 ~ 160 Ediss in kJ/mol 400 445 320 420 300 320 270 120 22 References 1 R. Grynspan, "Review of the maritime transport 2022", UNCTAD, 2022. https://unctad.org/system/files/officialdocument/rmt2022_en.pdf. 2 European Commission, The EU Blue Economy Report, 2022, Publications Office of the European Union, Luxembourg, p. 93, available at https://oceansandfisheries.ec.europa.eu/system/files/202205/2022blue economyreport_en.pdf. 3 https://www.chilternmaritime.com/whyisthemaritimeindustrysoimportant/. 4 T. Briggs, Ph. Cann, M. Masen, "Understanding Degradation in Stern Tube Seals", Imperial College, London, UK. 5 "Rotary Seals" in: Flitney, R. (2014). "Seals and Sealing Handbook", 6th ed., Elsevier Science. (2014). ISBN/EAN 9780080994161. 6 D.E. Johnston, "Rotary shaft seals", Tribology International 19, 170174 (1986). https://doi.org/10.1016/0301 679X(86)900514. 7 P.G.M. van Bavel, "The leakagefree operation of radial lip seals", Ph.D. Dissertation, Eindhoven University of Technology, the Netherlands (1997). 8 Y. Kawahara, H. Hirabayashi, "A study of sealing phenomena on oil seals", ASLE Transaction 22, 46 (1979). 9 Y. Kawahara, M. Abe, H. Hirabayashi, "An analysis of sealing characteristics of oil seals", ASLE Transactions 23, 93 (1980). 10 M.J.L. Stakenborg, "On the Sealing Mechanism of Radial Lip Seals", Tribology International 29, 335 (1988). 11 F.X. Borras, M. Bazrafshan, M.B. de Rooij, D.J. Schipper, "Stern tube seals under static condition: a multi scale contact modelling approach", Pr, oc. IMech. E., Part J: J. Engineering Tribology, 235, 181 (2021). https://doi.org/10.1177/1350650120925583. 12 D.E. Johnston, R. Vogt, "Rotary shaft seal friction, the influence of design, material, oil and shaft surface", SAE Transactions, 104(6): J. Pass. Cars, Part 1, 14531466 (1995). https://www.istor.org/stable/44612305. 13 F. Hirano, H. Ishiwata, "The lubricating condition of a lip seal", Proc. Inst. Mech. Eng., vol 180, pt. 3B, 187196 (1965). D.E. Johnston, R. Vogt, "Rotary shaft seal friction, the influence of design, material, oil and shaft surface", SAE Transactions, 104(6): J. PASS. Cars: Part 1, 14531466 (1995). https://www.jstor.org/stable/44612305. 14 T. Engelke, "Einfluss der ElastomerSchmierstoffKombination auf dass Betriebsverhalten von Radialwellendichtringen", Ph.D. Dissertation, Leibnitz Universitt Hannover, Germany (2011). 15 "Hansen Solubility Parameters: A User's Handbook", CRC Press, Inc., Boca Raton FL, 1999. ISBN: 08493 15255. 16 D.W. van Krevelen, "Properties of Polymers", 3rd ed., Elsevier, Amsterdam, 1997. ISBN: 9780444596123. 17 W. Hoffmann, "Rubber Technology Handbook", Hanser Publishers, Munich, Vienna, New York, 1996. 18 ASTM Standard D20018: "Standard Classification System for Rubber Products in Automotive Applications". 19 ASTM Standard D573 04: "Standard Test Method for Deterioration in an Oven". 20 ASTM standard D47110: "Standard Test Method for Rubber Property - Effect of Liquids". 21 ASTM standard D39518: "Standard Test Method for Rubber Property - Compression Set". 23 22 ASTM standard D132916: "Standard Test Method for Evaluating Rubber Property - Retraction at Lower Temperatures (TRTest)". 23 United States Environmental Protection Agency: "Environmentally Acceptable lubricants", Washington USA, 2011. 24 European Union Application Pack for Lubricants; European Commission : den Haag, the Netherlands, 2014. 25 J.V. Sherman, "Water soluble, environmentally acceptable lubricants for stern tube applications" In: Proceedings of the Society of Naval Architects and Marine Engineers (SNAME), 14th Propeller and Shafting Symposium, Norfolk, VA, USA, 1516 September 2015. 26 Th. Timm, "Die physikalischen Leistungsgrenzen von Elastomeren", Kautschuk Gummi Kunststoffe 39, 15 (1986). 27 R.J. Dunn and H.A. Pflisterer, "Using simulated enduse conditions in assessing the hightemperature performanc of oil resistant vulcanizates", J. Elast. Plast. 9, 193 (1977). 28 ECHA, European Chemicals Agency, Annex XV Restriction Report, Proposal for a restriction of Per and polyfluoroalkyl substances (PFASs); 0 7.02.2023. 29 Seiji Yamajo and Nobuhiro Hayami, "Study on Stern Tube Sealing System (Blister of Sealing Ring)", Nihon Hakuyo Kikan Gakkaishi, 17, 453 (1982). 30 Seiji Yamajo, Tadashi Yokoyama and Hitoshi Ishikawa, "Wear of rotary seals for shipscrews", J. Soc. Rubb. Sci. Techn. Japan, 58, 4 (1985). 31 Fr. Schulz, "Untersuchungen zur Blschenbildung bei Radialwellendichtringen aus FluorElastomer bei der Abdichtung von l", Mass Market Paperback. Shaker Verlag GmbH, Germany - 24 august 2000. ISBN13: 978 3826577550. 32 Lagersmit Information Letter: "Heading towards VGP compliance". Lagersmit, P.O. Box 176, 2950 AD Alblasserdam, the Netherlands. 33 https://www.recycling.com/downloads/wastehierarchylansinksladder/. 34 J. Bakker, B. Bokkers, M. Broekman, "Per and polyfluorinated substances in waste incinerator flue gases", RIVM report 20210143, Netherlands National Institute for Public Health and the Environment, Ministry of Health, Welfare and Sport. 35 K. Aleksandrov, HJ. Gehrmann, M. Hauser, H. Mtzing, D. Pigeon, D. Stapf, M. Wexler, "Waste incineration of Polytetrafluoroethylene (PTFE) to evaluate potential formation of per and Polyfluorinated Alkyl Substances (PFAS) in flue gas", Chemosphere 226, 898906 (2019). 36 Arqam Anjum, "Recovered Carbon Black from Waste Tire Pyrolysis: characteristics, performance and valorisation"; PhDthesis University of Twente, December 8, 2021. 37 "COMMISSION REGULATION (EU) .../... of XXX amending Annex XVII to Regulation (EC) No 1907/2006 of the European Parliament and of the Council concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) as regards synthetic polymer microparticles". 38 https://www.recybem.nl/sites/recybem.nl/files/user/position_paper__devulcanisatie_vs._reclaim.pdf 24 Resumes Resume prof. dr. ir. J.W.M. Noordermeer Jacques Noordermeer is professor (em) in Elastomer Technology and Engineering at the University of Twente, Enschede, the Netherlands. After completing his study biochemistry (cum laude) he obtained his PhD at the Delft University of Technology. He worked as a postdoctoral research associate at the Rheology Research Center of the University of Wisconsin, Madison, Wis. USA. After working several years as Director R&D in the industry in the field of rubber development, he specialized in elastomer technology applications. As of 1995 he worked as professor in Rubber Technology at the University of Twente, Enschede, the Netherlands. In 1999 he was recipient of the Technical Award of the IISRP: Institute of Synthetic Rubber Producers. In 2000 he was awarded with the Original Contribution Award of the 157th meeting of the Rubber Division of the American Chemical Society, Dallas, Texas. In 2005/2006 he received the Gold medal of the International Rubber Conference Organization (IRCO). In 2010 he was recipient of the George Staffort Whitby Award of the American Chemical Society, Rubber Division. In 2011 he was awarded as Dutch Master of Materials. In 2014 he received the Original Contribution Award of the 184th meeting of the Rubber Division of the American Chemical Society, Cleveland. In 2019 he was honored as Officer in the Royal Dutch Order of Oranje Nassau for professional and societal merits. In 2019 he received a Honorary Doctorate Degree in Polymer Technology of the Prince of Songkla University, Hat Yai, Thailand. Currently Jacques Noordermeer has several professional memberships: - 1975 - present: Member of the Sigma Xi, The Scientific Research Society, USA; - 1982 - present: Member of the American Chemical Society, Rubber Division; - 1982 - present: Member of the Dutch Association of Rubber and Plastics Technologists VKRT; - 1986 - present: Member of the Board of the International Rubber Conference Organizing Committee, on behalf of the Netherlands; - 2000 - present: Board Member of the Dutch Natural Rubber Foundation. 25 Resumes Resume prof. dr. ir. M. Masen Marc Masen is associate professor in Tribology and Mechanical Engineering Design at Imperial College London. After obtaining his PhD he spent several years in industry, after which he returned to academia. Within the field of tribology, his research ranges from wear mechanisms and design for sustainability to biotribology and soft materials, which includes elastomers and biological tissue. Marc is the recipient of the Imperial College President's Award for Excellence in Education and the 2022 Institution of Mechanical Engineers Donald Julius Groen Prize. 26