Faced with increased workloads and time and budget constraints that often restrict external training support, many chemical process operators are forced to get the most out of their heat transfer system with less help. This article offers recommendations for how to carry out proactive maintenance on heat-transfer fluids, to maximize their useful life and minimize problems associated with fluid degradation, such as excessive downtime for unplanned maintenance when the heat transfer system has become unsafe or is no longer able to carry heat in a reliable manner. It is useful for anyone developing or refreshing asset-care-management programs related to heat-transfer fluids and systems.
Discussed below are the most common fluid-related problems encountered by heat-transfer systems and a variety of potential solutions. While individual system designs and variations in process and operating conditions make each application unique, all heat-transfer fluids share many common attributes, making these recommendations widely applicable.
Ultimately, our goal is to educate those involved with the operation and maintenance of liquid-phase heat-transfer systems, both large and small, that use an organic-based heat-transfer fluid. The organics include chemical aromatics, fluids based on petroleum derivatives, silicone or glycol, the polyalphaolefins (PAO; also referred to as API Group IV-based fluids) and more. A properly designed and operated heat-transfer system can be the biggest ally in maintaining (and even increasing) productivity while reducing overall maintenance and production costs.
It starts with smart selection
The selection of the heat-transfer fluid — whether at the system design phase, or on an ongoing basis after commissioning — should not be taken lightly. Fluid selection should not be dictated solely by the purchase price or any single physical characteristic. Rather, a variety of factors should be considered:
• The potential impact on workers of a given fluid, in terms of adequate training and protection that must be implemented to address hazards related to potential exposure to the fluid, in both its vapor form (inhalation risk and mist concentration) and liquid form (skin contact). In addition to direct exposure, the choice of the fluid could impact productivity engendering additional handling and paperwork protocols involving other internal resources within the company, such as the health and safety advisors, medical care personnel, personnel in the receiving department and so forth
• Freight charges related to delivery of fresh product
• Cost associated with the pickup, handling and disposal of the used oil and drums
• Proven fluid performance beyond fresh oil data (for instance, if vendor data is able to demonstrate the retention of fresh oil properties after some time in service, as demonstrated by extensive oxidation and thermal stability data)
• Can the current system accommodate the fluid being considered (in terms of compatibility with sealing materials, existence of a properly sized expansion reservoir, suitable match between the fluid properties and the existing hardware, such as the pump and safety-relief valve)
• Miscibility with current heat-transfer fluids if partial (rather than full) changeout is needed
• Documented success by the vendor in your type of application
• Level of liability coverage, service and expertise the fluid maker and distributor bring to the table
Further discussion of initial fluid selection is beyond the scope of this article, but is covered in Ref. [ 1–7 ].
Over time, the most common threats to the life of heat-transfer fluids (and sometimes the entire system) include the following: • Thermal degradation
• Oxidative degradation
• Process contamination
• Contamination by other materials
Each threat is discussed below, along with findings from real case studies, and practical recommendations for how to deal with these challenges.
Regardless of the chemistry of the heat-transfer media, thermal degradation can occur whenever the heat source provides more energy than the heat-transfer media can absorb and carry away at that particular time [ 8 ].
|FIGURE 1. In this example with heat-transfer fluid
n-hexacosane, thermal degradation occurs when
excess heat drives the cracking of a straight-chain
hydrocarbon (not shown is the formation of reactive
free radicals, which have been omitted for clarity)
Figure 1 shows a simple example of the thermal degradation of a typical petroleum-based heat-transfer fluid (n-hexacosane) with ISO viscosity grade 32. In this case, the fluid is a distribution of molecules of various lengths, averaging 26 carbons long.
As shown in Figure 1, when the energy submitted to the fluid exceeds the threshold necessary to start breaking the stable covalent carbon-carbon bonds, the result is the formation of shorter hydrocarbons. The example in Figure 1 shows the scission (cracking) of a perfect straight, long-chain alkane into shorter molecules, such as dodecane (C12) and tetradecane (C14), each having a lower boiling and flashpoint and viscosity compared to the starting C26 hydrocarbon.
The systematic result of thermal degradation is a reduction in the overall fluid viscosity and increased volatility, which increases the risk of leakage and loss through evaporation. Thermal cracking increases the vapor pressure, lowers the flashpoint and fire point, and sometimes, reduces autoignition temperature (AIT). As the name implies, the AIT is the temperature at which the fluid vapors are hot enough to ignite spontaneously in absence of an ignition source [ 9, 10 ].
|FIGURE 2. Excessive thermal stress often results in a breakdown
of the heat-transfer fluid, and the carbonaceous byproducts can
build up on the inside surfaces of pipes
As shown in Figure 2, the problem worsens if left unaddressed. Reynolds discovered in 1883 [ 12, p. 86 ], that low-viscosity fluids offer the best heat transfer behavior in a forced-convection situation such as a typical heat transfer system. Based on these findings, one may think thermal cracking is advantageous from a thermal conductivity point of view. However, the resulting drop in viscosity is not necessarily favorable.
The concern is that the associated potential reduction of the AIT of the degraded fluid can make the operation of a closed system unsafe if the operating temperature nears or exceeds the AIT. Moreover, shortened molecules are not the only species formed during thermal degradation of the fluid.
On the other hand, an open system — that is, one in which the heated fluid is constantly in contact with the atmosphere — is even less forgiving. Any drop in the heat-transfer fluid’s flashpoint and fire point (defined as the temperature at which the fluid sustains a fire for five seconds in the ASTM-D92 Cleveland Open Cup, or COC flashpoint test apparatus) could jeopardize the entire operation, considering that the fluid was likely chosen, in part, based on its fresh oil, open-cup flashpoint rating (to which a safety margin was likely added).
Efforts to determine a definitive relationship between a drop in flashpoint and a drop in AIT have not proven successful. Fortunately for users, in many cases where a petroleum-based fluid exhibits a relatively low flashpoint, we have seen the AIT remained high, but this is not always the case.
The performance data shown in Table 1 demonstrate how progressive thermal degradation leads to steadily diminishing flashpoint and viscosity of the heat-transfer fluid. The gas chromatography distillation (GCD) test consists of a simulated distillation of the fluid in the laboratory. In the cited example, the initial distillation point (GCD 10%) drops over time, which again confirms the increased concentration of low-boiling components present in the fluid.
|Table 1. Analysis Data showing thermal degradation of the heat-transfer fluid at a meat-processing facility|
|Sample date, mm/dd/yy||Flashpoint, °C (COC)*||Water content, ppm (Karl Fisher)||Viscosity at 40 °C, (centistokes, cSt)||Gas chromatoraphy distillation (GDC)**|
|10% boiling, °C||90% boiling, °C||% boiling below 335°C|
|After startup and shutdown procedure modification of April 2003|
|New fluid properties||209||—||35.6||382||498||0.80|
|* COC represents analysis via the ASTM-D92 Cleveland Open Cup (COC) flashpoint test apparatus. ** GCD = gas chromatography distillation. The GCD test consists of a simulated distillation of the fluid in the laboratory. Comparison with the fresh-oil boiling curve allows for the detection of lighter and heavier molecules in the fluids.|
Another major consequence of thermal cracking is the formation of carbonaceous residues (Figure 2), which result from reactions of recombination. To a certain extent, these particles can be compared to soot that is produced during fuel combustion in a diesel engine, where it is documented that soot is harder than the metallic components of the engine [ 13 ].
Such unwanted carbon residues are not only abrasive toward the piping, but they also tend to stubbornly adhere and harden onto the hot surface points, forming an insulation layer inside the pipe. This occurrence often forces the user to increase the heater set temperature (increasing energy consumption) to maintain the desired operating fluid temperature.
As a general rule of thumb, Wheeler [ 14 ] reports that the widely used heat-transfer fluids based on polyalkylene glycols (PAGs) begin to experience thermal degradation near 250°C (482°F). Meanwhile, Wheeler also reports that the thermal degradation of uninhibited polyethylene glycol results in a mix of five organic acids [ 15 ]. The formation of these byproduct acids leads to increased corrosion over time in high-temperature systems.
Of similar importance is the fact that even systems running at temperatures that are considered to be relatively mild (for example, around 149–204°C or 300–400°F), are not exempt from the ravages that elevated temperatures can bring, in terms of the thermal cracking of the heat-transfer fluid. For example, consider a system in which the fluid experienced a change in physical properties, combined with oil-flow issues (for instance, from a defective pump, a fluid containing solids, or some piping restriction or pluggage) or a problem with the heater (for instance, the heater coil or electrical element has baked-on carbon that acts as an insulation layer forcing a higher energy demand to maintain the target fluid outlet temperature). Such factors can cause a rise in the skin-film temperature (the temperature of the fluid immediately touching the heated surface).
Any combination of the conditions mentioned above can cause the skin-film temperature to be significantly higher than the temperature of the fluid circulating in the center of the heated pipe (which is called the bulk oil temperature). The larger the gap between skin film and bulk oil temperature, the more energy the fluid tries to distribute within itself through turbulence. At some point, the fluid at the heated surface will receive more energy than it can absorb (its heat capacity), carry and release (its thermal conductivity), resulting in thermal degradation of the fluid.
Discussed below are ways to minimize the thermal degradation of a heat-transfer fluid in open systems.
Use the right fluid for the job. By choosing a fluid with a high thermal stability, Guyer and Brownell [ 16 ] suggest that most problems associated with localized or temporary temperature excursion can be prevented. Ashman [ 17 ] also emphasizes the importance of using a heat-transfer fluid with a suitable thermal stability for the application. Hudson, Sahasranaman [ 6, 7 ] and many others acknowledge that petroleum-based fluids of pharmaceutical quality produced by a severe hydrogenation and hydrocracking process (also referred to as “white mineral oils”) tend to have greater thermal stability compared to petroleum base oils that are produced from other refining methods [ 6, 7 ].
Use appropriate venting. Venting involves the periodic release (from the fluid and the system) of the light, more highly volatile hydrocarbons that form during thermal cracking. Venting is typically carried out by circulating some of the hot fluid to the expansion reservoir, so that those molecules with a relatively high vapor pressure can naturally migrate into the gas phase above the fluid. Then, depending on the system design, the vapors are released directly into the atmosphere or sent to a collection drum or tank, although laws governing volatile organic compounds (VOCs) and other environmental trends cause most users to collect the condensed low-boilers and properly dispose of them.
Fresh fluid needs to be added periodically, to maintain the desired fluid level (to prevent pump starvation and cavitation when the system charge contracts after a shutdown). As a precautionary note, users should remember that fresh fluid must never be added directly into the hot oil stream; rather it should be added into the expansion tank or other cool reservoirs connected to the system.
Venting continuously or for extended periods is not advised, because the resulting rise in the bulk fluid temperature in the expansion tank will accelerate oxidation (discussed below).
We recommend the use of an oil-analysis program to determine the rate of generation of low-boilers during any operation. With proper venting and analysis, users can establish how often, and for how long, the fluid must be periodically vented, in order to safely operate a high temperature system with a fluid that stays in good condition (maintaining characteristics that are similar to the fresh oil for as long as possible).
Adopt proper startup and shutdown procedures. The successful startup of any heat-transfer system is important, since the faster the heat-transfer fluid reaches its desired operating temperature, the faster the facility can produce its products and begin to fulfill orders. This becomes even more important for systems that stop and start up regularly.
One may say that running the pump and the heater for a few extra hours to accommodate a slower, more-gentle startup is not cost-effective, but for many applications, such an approach pays its own dividends. For instance, by maintaining a more-gradual heating profile at startup, the fluid will be able to effectively remove heat and reduce the risk of thermal degradation, and minimizing the formation and buildup of baked-on residues. The net result will be extended planned-maintenance intervals and greater component life expectancy.
Shutdown procedures also impact system efficiency and fluid life. For instance, Stone [ 19 ] and others recommend maintaining oil circulation after the heater is turned off until it’s been cooled to 65°C (150°F). The refractory material in a furnace is designed to retain heat for as long as possible, so stopping the oil flow immediately after the heat source has been turned off provides an opportunity for the stagnant fluid to crack, forming low-boiling fractions and carbon residues. This negatively impacts the life of the fluid and the overall heater efficiency.
With regard to smaller systems such as temperature-controlled units (TCUs) or extruders, many designs have improved greatly in recent years and now maintain fluid circulation for some period of time following shutdown as a common approach.
An insufficient shutdown interval was the overall problem at the facility whose degraded fluid was shown in Table 1. After a service call, it was determined that the 249°C (480°F) system was shut down on Friday evening with only a short circulation period following shutdown. It was fired up again at 7:00 a.m. on Monday, to allow for production to start at 9:00 a.m.
A full system shutdown and cleaning was deemed impossible by the user at that time, so the fluid was left untouched, but better future practices were implemented. The last set of results in Table 1 shows that two months after the initial analysis, the rate of generation of low-boilers had diminished (as seen in the percentage boiled below 335°C). As a direct result, the facility did not add any new oil. The increase in kinematic viscosity and flashpoint, and the fact that the strainer no longer collected carbon residues in any appreciable amount, provided evidence of improvement.
Consult your suppliers about proposed design or operational changes.Business is booming, more production is expected from the plant, more parts must be produced, and lines need to be added. Do you need to increase the operating temperature? What about the flowrate, is it adequate? What does your heater manufacturer think of the proposed addition? Operators should get as much input as possible from their system designer. manufacturers, and parts and fluid suppliers before any major changes are implemented. Stone  recommends that operators should maintain an updated list of contacts and keep it handy for questions or troubleshooting help.
It is relevant to document the skin-film temperature in the current system and in the proposed operating conditions. Make certain your fluid supplier confirms your current heat-transfer fluid’s ability to handle any new operating parameters.
Maintain, inspect and perform preventative maintenance on system components. Even though liquid-phase systems commonly operate above the flashpoint of the fluid (but below its auto-ignition temperature), the risk of fire should be very low in a normal, well-designed system, especially one that is kept oil-tight, leak-free and subject to regular inspection and maintenance [ 19 ].
For any system where heat is purposely generated to raise the fluid temperature, ensuring proper operation of the heat source is critical to achieve optimum performance. Daily inspections, using a consistent checklist of items to monitor are recommended [ 18, 20 ]. For instance, fired heaters should be inspected for flame impingement, especially if the burner is oversized or cycles frequently. In the case of flame impingement, the flame (whose temperature is typically on the order of 1,093–1,650°C, or 2,000–3,000°F) subjects the oil tubes to excessive localized heat flux, which can cause tube deformation and coking (resulting from thermal degradation, as seen in Figure 2), and leakage with increased risk of fire [ 21 ].
In the case of systems equipped with immersed electrical heaters, excessive watt density, lack of fluid turbulence around the hot tubes, or insufficient flowrate often causes premature degradation of the fluid. Such degradation can be offset in part by proper fluid selection and maintenance practices [ 22, 23 ].
In any system, the oil-circulation pump can be compared to the heart, moving the fluid around. The pump should be well-maintained. Specifically, drive bearings on the electric motor and pump seals should receive proper attention. Centrifugal pumps should ideally operate at or near their best efficiency point (BEP), with bearings well-maintained and seals working properly. Finally, the expansion reservoir, piping, connections and valves should be selected and maintained appropriately, as part of a world-class maintenance program.
Meanwhile, the life blood of the operation — the fluid itself — should be tested regularly. While further discussion of the types of tests, their significance and data interpretation is beyond the scope of this article, the American Society for Testing and Materials (ASTM) Method D5372 should be followed to properly monitor the condition of heat-transfer fluids [ 24 ].
For the purpose of this article, we define fluid oxidation as the reaction of the heat-transfer fluid with oxygen from air. The oxidative degradation of organic compounds is extremely complex, as it involves a series of chemical reactions that result in the formation of high energy, unstable and reactive free-radicals and peroxides. One initial free-radical allows for the possibility of forming two radical species, which results in the formation of a variety of oxygen-containing species, mainly organic acids. These long-chained organic acids may be weak on their own, but as their concentration grows in the fluid, the oil eventually becomes more corrosive [ 25 ].
These acids also polymerize — often to a level that is sufficient to modify the fluid properties, causing an increase in viscosity, discoloration and eventually, precipitation as lacquer, varnish and sludge [ 26 ] such as that shown in Figure 3. The varnish formation is seldom a concern in heat transfer applications because of relatively large pipe diameters and valves with high tolerances. However, further oxidation will lead to the formation of heavier acids and sludge. Oxidation-related sludge is not very soluble in heat-transfer fluids, so it tends to adhere to metallic surfaces or settle in areas of low flow and low turbulence.
|FIGURE 3. These illustrations shows the type of varnish (left)
and sludge (center, right) that can result from oxidation-related
degradation of a petroleum-based, chemical aromatic and
polyalkylene glycol (PAG) fluid
Such sludge also tends to settle at the bottom and the sides of the expansion tank, and can also circulate throughout the system and make its way into control valves.
Fluids for a specific project are generally chosen based on their properties in a fresh state. Any alteration of the fluid physical properties (resulting from degradation or contamination) could negatively impact the heat absorption and dissipation capabilities of the heat-transfer media.
Table 2 provides oil-analysis data for an uninhibited, chemical aromatic (synthetic), heat-transfer fluid that experienced oxidation in a large 27,000-L (7,132-gal) system in Europe. (In this context, the term “uninhibited” refers to the fact that the fluid does not contain additives such as anti-oxidants and rust-corrosion inhibitors to prevent degradation.) The acid number (AN) — as determined by ASTM D664 Method and used to quantify the level of acids in an oil sample — was increasing over time.
|Table 2. OIL-ANALYSIS DATA DESCRIBING A FLUID THAT HAS EXPERIENCED oxidation (source: Petro-Canada LUBRICANTS, A SUNCOR ENERGY COMPANY)|
|Sample date, mm/dd/yy||Flash point (COC), °C||Water content, ppm (Karl Fisher)||Viscosity at 40°C, cSt||Acid number (AN*), mg/KOH/g||Solids (insolubles), wt.%||GCD|
|10% boiling, °C||90% boiling, °C||% boiling below 335°C|
|*Acid Number (AN) is obtained using ASTM D664 titration method, which is used to quantify the levels of acid in an oil sample.|
The distillation of the fluid, represented by the GCD 10% boiling point, shows the initial boiling is at the same temperature as fresh oil, so thermal degradation does not seem to be an issue in this example. We notice the viscosity has risen by 30% over time and the end of the distillation curve (GCD 90%) is shifting toward higher temperatures, indicating the increasing presence of heavy compounds not found in the fresh oil.
An increasing amount of insoluble particles are forming, and the AN values are rising. By connecting the dots, we conclude that oil oxidation is causing an increased formation of heavy acidic polymers that will foul the low-flow areas of the system. This degraded oil, with its higher viscosity, cannot deliver the same performance capabilities as fresh oil, and in today’s context of high energy costs, any loss of efficiency is costly.
In the example discussed above, the company could not afford a shutdown to clean its system this year. Instead, operators opted for a partial fluid replacement of 50% of the entire charge this year (incurring an expenditure of roughly $175,000, excluding waste oil disposal and labor) and are planning a full drain, clean, flush and refill next year. In general, fluid oxidation imposes great cost penalties on any system; the selection of a fluid with better oxidation stability could have avoided this massive spending and offered many more years of useful life.
Discussed below are several options that are available to avoid or minimize potential oxidative degradation.
Inert gas blanketing. In closed systems, the most effective way to eliminate the potential for oxidation is to install an inert gas blanket in the expansion tank headspace. Hudson  provides details and recommendations on how to install such systems. The basic principle relies on substituting air (which contains oxygen) with an inert gas (most often nitrogen, although carbon dioxide and argon may also be considered) in the only location where warm oil can come into contact with oxygen from air — the expansion tank headspace. Displacing oxygen that might react with the fluids virtually eliminates oxidation.
The pressure of the inert gas is maintained slightly above atmospheric pressure. Gas-blanketing systems, including the safety-relief valve, require ongoing inspection and maintenance to prevent inert gas leaks and limit unnecessary, costly gas consumption.
Choose a fluid formulated for the job. Oxidation-inhibitor additives are also available to enhance the performance of heat-transfer fluids. Most chemical aromatics sold today contain one or a few varieties of molecules and do not contain any performance enhancing additives such as antioxidants or rust and corrosion inhibitors.
The additives that are used in heat-transfer fluids are different from the ones found in other industrial lubricants that are not subjected to such elevated temperatures. Specifically, in the case of antioxidants, some technologies combat oxidation by reacting with free radicals before they can lead to acid formation, while others attack intermediate peroxides [ 25 ].
Fluid selection is complicated by the fact that it is extremely difficult to determine the oxidation stability of a heat-transfer fluid by its technical data sheet. Even though many of the heat-transfer fluids on the market today are unadditized, their respective marketing materials often praise their fouling resistance and promote their outstanding oxidation stability. Thus, users should assess all product claims with a critical eye.
In general, systems with an enormous amount of oil tend to be more forgiving because it takes a longer time to oxidize a larger volume of fluid to a point where it raises concerns in terms of oil analysis results. In these cases, user experience, references, testimonials and competitive benchmarking studies should be evaluated in conjunction with vendor data, to assess the likely longevity of a fluid for the application at hand and avoid costly changeouts in large systems.
Compared to closed or blanketed systems, open systems allow the hot fluid to come in direct contact with air, making oxidative damage a harsh reality rather than a possibility. In these cases, the importance of choosing a robust product to maintain high productivity standards becomes even more important.
For example, an electronic company operating an open system at 175°C (350°F) was replacing its heat-transfer fluid every six months, after which time the fluid had become viscous and dark with a burnt odor. Switching to a fluid with better resistance to oxidation enabled longer service life. In fact, judging by the oil-analysis results, the oil properties still look like new after more than 24 months of service in these harsh conditions. This obviously saves the facility money in terms of time, labor and fluid purchases.
In closed systems with no inert gas blanketing, the key is to maintain the fluid temperature in the expansion tank below 65°C (150°F), if possible. The main reason is because there is a direct relationship between the temperature and the rate of oxidation. For instance, the rate of reaction between a petroleum-derived oil with oxygen (doubles for every 10°C (15°F) increase above 80°C (175°F) (with slight variations depending on the author) , so the higher the temperature the more severe the degradation. And this does not take into account the fact that the oxidation reaction is exponential and is accelerated by contaminants such as copper or iron particles, water and other catalysts.
Oxidation could occur in systems with a design that allows the oil to circulate through the expansion reservoir with full flow, either directly after the heater or on the return from the heat users. Such design exposes the hot fluid directly to oxygen from air, thereby acceleraing oxidation and greatly reducing fluid life.
Using the oil-analysis results, fluid oxidation can be monitored by paying close attention to acid number (AN) and gas chromatographic distillation (GCD) results.
Minimizing process contamination
Process contamination can be extremely damaging to the heat-transfer fluid and the system components. As is often the case, logic suggests that contamination is unlikely since the pressure is greater on the fluid side, but real life experience has shown on many occasions that process material can enter the heat-transfer fluid stream. The urgency required to fix a process leak really depends on the severity, the type of contaminant (chemistry), and the heat transfer media it comes in contact with. The case of contamination by water is discussed in the next section, although water is sometimes part of the process.
For example, in the oil-and-gas industry, a natural-gas-extraction facility may experience an unintended leak of the process hydrocarbons into the heat-transfer fluid system. Being hydrocarbon-based, the heated gaseous molecules will mix very well with heat-transfer fluids of a similar chemistry, such as petroleum-based fluids, chemical aromatics and PAO Group IV synthetic fluids ([ 4 ] provides details on competing fluid types). Within a short time, the viscosity of the entire fluid charge will be greatly reduced and its overall volatility increased.
In a situation such as this one, emphasis must be put into venting the heat-transfer fluid to release those light hydrocarbons into the proper collection device in order to maintain a safe operation, and if at all safely possible, to keep the unit running until the next shutdown opportunity to repair the leak.
Another example of process contamination in the petroleum industry occurs frequently at asphalt terminals. Similar to the example discussed above, any unintended ingress of asphalt in the heat-transfer fluid circuit will mix very well with most of the fluids, since the majority are based on long hydrocarbon chains. However, the highly viscous hydrocarbon asphalt will quickly thicken the fluid.
We have seen heat-transfer fluids increase to several hundred centistokes or even become too thick to measure at 40°C (104°F), thereby ruining the fluid’s ability to transfer heat effectively. The heavy asphalt components will also coat the system internals and plug small lines, meaning a full system cleaning and flushing will eventually become necessary to restore the system to efficient operation.
In some cases, the contaminant itself may be inert to the fluid but it may still react with traces of moisture to form acidic or insoluble compounds. These byproduct contaminants can accelerate rust and cause corrosion and fluid degradation.
Depending on the process contaminants that are inadvertently leaking into the fluid system, it might be possible to detect them (qualitatively) via oil analysis, using the common elemental analysis method like Inductively Coupled Plasma – Atomic Emission Spectrometry (ICP-AES). Sometimes the contaminant can be detected indirectly after it has reacted with another compound in the fluid. In some cases regular oil analysis will not detect the process contaminant and specialized methodology and instruments are needed, such as those found in specialized research-and-development facilities.
A quantitative evaluation to determine the type and extent of the contamination generally requires sophisticated equipment (such as an electronic microscope, or gas chromatography coupled with mass spectrometry), as well as well-trained analysts who are knowledgeable about the product being tested and informed on what contaminant types to look for.
Whenever a process leak is suspected, it is advisable to reach out to your fluid supplier’s technical contact immediately and explain the situation. A sample of the fluid should be analyzed right away.
Other sources of contamination
In addition to contamination that can arise from process materials (discussed above), heat-tranfer fluids may also become contaminated by the environment (rain or snow), condensation, foreign liquids (such as the wrong fluid put in the system), or the ingress of air. For systems where the expansion reservoir is outside and vented to the atmosphere, it is critical to have — at a minimum — an enclosed tank with a 180-deg, goose-neck pipe on the top.
This may sound very basic, but we were once called to investigate unusual noise coming from the hot oil piping at a saw mill. After assessing the noise, we climbed up to the top of the burner building to examine the expansion tank. The 12-by-12-in. steel cover normally bolted to the side of the 250-gal expansion tank was laying on the catwalk, covered by a foot of dirt, wood dust and snow and no one could remember who had been up there last. Rainwater and snow falling directly into the expansion tank from the open hole was responsible for the high water content we later measured in the fluid and the knocking noise in the piping below.
New construction or recently cleaned systems or heat exchangers are not typically flushed before commissioning. However, in systems where a full or partial cleaning was performed, traces of aggressive cleaning fluids or water-based solutions that are not removed could accelerate corrosion, fouling or create their own polymerization and insoluble residues . In newly commissioned systems, aside from the typical wood debris, welding rods and rags, residual water from pressure testing is most often the culprit for startup problems. Unlike many industrial applications, water in the heat-transfer fluid is more easily detectable by operators and unforgiving because it is heated above its boiling point during service in most applications.
Entrained water will affect various fluid chemistries in different ways. In lubricating and circulation fluids based on mineral and synthetic Group IV PAO oils, prolonged exposure to water may cause the following :
• Hydrolysis or precipitation of oil additives (for those oils that have them)
• Accelerated rust and corrosion of system internals
• Accelerate degradation (oxidation)
• Cause pump cavitation and wear
• Create a gargling noise in the expansion tank and knocking in the hot oil piping
Based on years of examining real-life oil-analysis results, we can say that in general, water does not appear to pose immediate productivity concerns at concentrations below 500 ppm (0.05 wt.%), although we have encountered certain, more-sensitive systems where lower concentrations did have a noticeable impact. However, residual water at concentrations above 1,000 ppm (0.1 wt.%) becomes alarming and calls for investigation and removal.
In the case of mineral oils, the best practical way to remove the water from a heat-transfer fluid while the system is running involves more of a two-step process. First, vent the fluid, allowing the water vapor to migrate into the expansion tank. Once inside the expansion tank, some of the steam will have sufficient vapor pressure to leave through the vent pipe or safety-relief valve when it opens.
In the case of PAG-based fluids, the numerous oxygen atoms in their structure produces strong hygroscopic behavior that is directly proportional to relative humidity in the environment. Wheeler [ 15 ] reports that at 50% relative humidity, pure ethylene glycol absorbs 20% water at equilibrium. This can cause serious concerns.
Lastly, operators must take steps to guard against potential contamination by airborne vapors or particles that could affect the fluid. Just think of a saw mill example, where entrained cellulose dust from the wood dust may not degrade the fluid itself, but will affect the fluid’s ability to flow, which will reduce the thermal efficiency and accelerate fouling in the system [ 29 ]. Such an occurrence is more likely to happen if the expansion tank is located in a very dusty environment.
Discussed below are a variety of techniques for minimizing contamination that can threaten heat- transfer fluids.
Investigate and fix. All cases of contamination should be investigated and fixed, and such incidents should also be reported to your fluid supplier, for advice on the potential impact on the fluid. As mentioned earlier, sometimes the contaminant can be evacuated, boiled off or it could ruin the fluid and foul the system in a short time.
Flush new constructions or recently cleaned systems before startup.Operating companies and builders seldom factor in the cost of a system flush, since they often assume the blowing of the water will be done correctly and the contractors will not leave debris in the piping. Unfortunately, discovering such contaminants after the system is running can prove to be costly down the road. While nobody needs the extra costs of flushing a new system (especially when the fluid of choice is relatively expensive, like PAGs or silicone-based fluids), it is nonetheless a good practice. With systems filled with mineral oils, circulating a virgin base oil of the same viscosity as the heat-transfer fluid of choice is a cost-effective way to remove any potential contaminants.
Keep an eye on filters and strainers.Solids collection in the oil filters or strainers should be noted in a log book and monitored closely, preferably with photos taken. The size, texture and color of the deposits all tell a story, and such residues can be sent periodically to a laboratory with sophisticated analytical equipment for accurate identification.
Keep in mind that different solids may come from more than one source, and may become discolored, so don’t jump to conclusions.
Similarly solids from the previous fluids may reside in the system for a long time before an event such as pipe work or partial fluid replacement creates enough disturbance to loosen them. We see this in cases where a used furnace is bought and commissioned without cleaning and flushing prior to the connection to the main system.
Often solids may have a familiar smell or texture that suggests an origin, but could well be something else. For example, a plant was using a heat-transfer fluid that caused valve malfunction because of deposits accumulating inside the valve spools. The black, abrasive deposits looked and felt like carbon particles (abrasive, gritty between the fingers). However, lab analysis identified the material as copper sulfide, formed by the localized chemical attack of sulfur present in the fluid’s base stock onto the copper from the brass valves.
The facility could have spent several thousands of dollars in parts and labor to upgrade all the valves to more expensive stainless steel. Instead it switched to a properly formulated fluid based on highly refined API Group II base oils containing only traces of sulfur. This replacement fluid has proven to be harmless to copper components after years of service, and has had the added benefit of extending oil changes considerably, based on oil-analysis results.
Edited by Suzanne Shelley
Gaston Arseneault is a senior technical advisor with Petro-Canada Lubricants, a Suncor Energy business (1310 Lakeshore Road West, Mississauga, Ontario, Canada L5J 1K2; Phone: 973-673-3164; E-mail: email@example.com), located in the Newark, N.J., area. With the company for more than ten years, Arseneault holds an M.S. in analytical chemistry from the Université de Montréal in Canada and is a member of Society of Tribologists and Lubrication Engineers, from which he has obtained the Certified Lubrication Specialist (CLS) and Oil Monitoring Analyst (OMA I) certifications. He also holds the Machinery Lubrication Technician I certification from the International Council for Machinery Lubrication.
1. Guffey II, G.E., Sizing up heat transfer fluids and heaters”, Chem. Eng., Oct. 1997, pp. 126–131.
2. Shanley, A., and Kamiya, T., Heat transfer fluids: A buyers’ market, Chem. Eng., Sept. 1998, pp. 63–66.
3. Arseneault, G., Seven criteria for selecting heat transfer fluid, Process Heating, January 2008, pp. 2–3.
4. Arseneault, G., Safe handling of heat transfer fluids, Chem. Eng. Prog., April 2008, pp. 42–47.
5. Crabb, C., A fluid market for heat transfer, Chem. Eng., April 2001, pp. 73–76.
6. Hudson, J., Choosing the heat transfer fluid, Process Heating, January 2007.
7. Sahasranaman, K., Get the most from high-temperature heat-transfer-fluid systems, Chem. Eng., March 2005, pp. 46–50.
8. Guyer E.C. and Brownell D.L., “Handbook of Applied Thermal Design,” McGraw-Hill, ISBN 0070253536, 1988, pp. 5–46.
9. Singh, J., “Heat Transfer Fluids and Systems for Process and Energy Applications,” CRC, ISBN 0824771915, 1985, p. 214.
10. Petro-Canada, “TechData: Handbook of Petroleum Product Terms,” Revised Edition 89.06, 1989.
11. Kay, J.M., and Nederman, R.M., “Fluid Mechanics and Transfer Processes”, ISBN 0521303036, 1985, p. 18.
12. Peters, M.S., “Elementary Chemical Engineering,” McGraw-Hill, 1984, p. 85.
13. Petro-Canada, “TechBulletin: The Truth About Soot,” Edition 89.06, 1989.
14. Society of Tribologists and Lubrication Engineers (STLE), “Basic Handbook of Lubrication,” 2nd ed., 2003, Section 3, p. 8.
15. Wheeler, K., Technical Insights into Unihibited Ethylene Glycol, P rocess Cooling & Equipment, July/August 2002.
16. Guyer, E.C., and Brownell, D.L., “Handbook of Applied Thermal Design,” McGraw-Hill, ISBN 0070253536, 1988, pp. 5–47.
17. Ashman, L.A., Troubleshooting problems in heat transfer systems, Process Heating, October 1988, www.process-heating.com.
18. Stone, C.D., “Eight tips to extend your thermal fluid system’s service life,” Process Heating, April 2003.
19. BP (British Petroleum), “Transcal Heat Transfer Fluids”, BP Printing, England, Document # MP59, 1979.
20. Arseneault, G., Preventative maintenance for heat transfer systems, World Pumps, April 2008, pp. 40–43.
21. Vinagayam, K., Minimizing flame impingement in fired heaters, Chem. Eng., pp. May 2007, pp. 70–73.
22. Klein, R., Immersion heaters: Selection and implementation, Chem. Eng., January 2006, pp. 44–50.
23. Arseneault, G., Preventive maintenance for heat transfer systems using electrical immersion heaters, Process Heating, November 2006, p. 12.
24. ASTM International, ASTM D5372-04: Standard Guide for Evaluation of Hydrocarbon Heat Transfer Fluid, 2004, p. 3.
25. The Lubrizol Corporation, “Ready Reference for Grease,” The Lubrizol Corp., Wickliffe, Ohio, Version 2.00, May 2007, pp. 36–37.
26. Singh, J., “Heat Transfer Fluids and Systems for Process and Energy Applications”, CRC, ISBN 0824771915, 1986, p. 179.
27. Hudson, J., The Expansion Tank, Process Heating, September 2007, accessed online at www.process-heating.com.
28. Society of Tribologists and Lubrication Engineers, Alberta Section, “Basic Handbook of Lubrication,” 2nd ed., Section 26, 2003, p. 8.
29. Guyer, E.C., and Brownell, D.L., “Handbook of Applied Thermal Design,” McGraw-Hill, New York, 1988, pp. 5–50.
30. Bloch, H.P., “Practical Lubrication for Industrial Facilities,” The Fairmount Press, 2000, pp. 464–465.
Researchers from the Fraunhofer Institute for Environmental, Safety and Energy Technology (Umsicht; Sulzbach-Rosenberg; www.umsicht.fraunhofer.de) and RWTH Aachen University (both Germany;…
Collecting representative samples of heat-transfer fluid for routine quality checks falls short of the goal if safety or the environment…
Industrial microbiology continues to hold the promise of producing an increasingly large array of commercially important chemicals because an ever-wider…
The management of process plants requires understanding and applying the fundamental tenets of safe design, construction and operation of facilities.…
Changing out used heat-transfer fluid is critical to the efficiency of a heat-transfer fluid system. A well-operated heat-transfer-fluid system can…
Detect and correct anomalies early in your batch processes
ABB at ACHEMA 2018
The Future of Project Execution
Digitalization: What does it deliver today… and tomorrow?