Understanding Turbine Oil Demulsibility

Why oils and water don’t mix

Water and oil are immiscible because of the chemical nature of water and oil molecules. The old adage in chemistry used to describe this is “like dissolves like”. Water molecules are polar possessing a electronegative oxygen atom and the electropositive hydrogen atoms. Oil on the other hand is non-polar in nature without any strongly electronegative or electropositive atoms. When these two fluids are mixed, the water atoms will agglomerate together rather than mix with the organic hydrocarbon molecules. This action rejects water from the oil when attempting to mix. Oil floats on water because its molecules are larger and significantly less dense than water molecules.

Small amounts of water are miscible in turbine oil. This moisture is referred to as dissolved water. Turbine oils can hold about 100-150ppm of water in solution at room temperature. Small amounts of polar-heteroatom hydrocarbon species in the oil will greatly impact water separability characteristics.

Reasons why Demulsibility Properties fail

In-service turbine oils fail demulsibility because of the presence of polar constituents in the fluid, allowing water to become miscible in the oil. An extremely small amount of polar molecules mixed into the oil can have a devastating impact on demulsibility depending on the polarity and chemistry of these compounds. A study on demulsibility by ExxonMobil indicated that Ca levels of 3ppm as calcium alkylbenzene sulfonate can impact demulsibility values.  There are some field tests that have indicated values even smaller concentrations than this were shown to be an issue. This is why it is very challenging to identify the root cause of turbine oil demulsibility failure. Occasionally, there is evidence found through the oil analysis that can point to a root cause – like when a calcium alkylbenzene sulfonate detergent is  found to be present. In most cases however, there is too little material present to identify through conventional analysis (such as FTIR or ICP spectroscopy).

Without having a testing protocol that consistently and accurately identifies demulsibility failures, one cannot determine the root cause of the failure and what subsequent correction to may be.

Polar constituents get into turbine oil for one of two reasons: fluid contamination or fluid degradation.

Fluid Contamination

Contaminants may enter the turbine oil system through several means:

•        Oil top-offs which can introduce contaminants and new additive chemistries to the fluid
•        Maintenance activities such as flushing which introduce detergents or cleaners
•        Air ingression which often carries  contaminants to the fluid
•        Water and steam leaks which can pollute the oil with water and its treatment chemistries.

Alkylbenzene sulfonate detergents are a common contaminant that may remain in the turbine oil system following a flush. This chemistry is known to adhere to the metal parts as part of its cleaning action, making it difficult to remove.  The detergent is known to destroy the demulsibility at very low concentrations. This is because the chemical structure of this molecule has a very strongly-polar head group (the sulfonate) and a non-polar hydrocarbon tail (alkyl-benzene). The polar head sticks into the water phase and the hydrocarbon tail sticks in the oil phase to allow combinations of these two liquid phases. Other additives like carboxylic acids, oxidation products, esters, amines and alcohols all have this same functional behavior, but to a lesser extent. The higher the component concentration, the more bi-functional the molecule present and the poorer of the demulsibility result.

Compatibility of a fluid is measured by the interaction of the components in the fluid. If the interaction is antagonistic you will first observe effects to the interfaces – air/fluid (foam), solid/fluid (MPC or haze) and water/fluid (demulsibility). Some of the things that affect compatibility are mixing of fluids, improper flushing and formulation changes. All of these add chemistry to the turbine oil that changes the polarity of the fluid and the polarity at the surface of the fluid. Increasing the fluid’s polarity means the polar water molecules interact better with the polarity of the fluid to affect the demulsibility property. These ingress chemistries can have polar-head/hydrocarbon-tail functionality that allows behavior similar to the detergents but they can also have internal polarity to simple increase the overall fluid’s polarity.

Fluid Degradation

As turbine oils are thermally and oxidatively stressed, degradation products referred to as soft contaminants are formed. These degradation products are polar in nature, which is why they have solubility issues and can be measured by the MPC test (ASTM D7843). A large amount of oil degradation products can adversely impact a fluid’s demulsibility characteristics. This is not always the case however many oils with high varnish potential also exhibit poor demulsibility characteristic.

Demulsibility Retention

All new turbine oils have good water separation characteristics. The absence of polar bodies in the oil allows the fluid to readily separate from water. These characteristics have improved even further with the introduction of GII and GIII turbine oil base stocks. One could surmise that the use of GII and III turbine oils has resulted in lower demulsibility problems in the field. Ironically, there appears to be more demulsibility problems today with the newer generation of oils.

Understanding the differences between API Group I and Group II & III baseoils provides insight into why this may be. It is well known that Group I fluids contain a multitude of chemistry including aromatics, hetero-molecular components and a variety of aliphatic hydrocarbons. These aromatics and hetero-compounds have increased polarity over the bulk aliphatic hydrocarbons of the fluid. That means when a polar material is introduced into the fluid there isn’t as large an influence as might be if these base oil polar materials were not present. Group II & III fluids are produced from a hydrotreating process that “cleans-up” the fluid chemistries to get rid of the aromatics and hetero-compounds. This makes the Group II & III more “pure” in nature. Group IV (PAO) is even cleaner. The purer the fluid is the less ability it has to compensate for ingress of something new.  Thus, Group I would be less sensitive to ingress material de-rating its demulsibility than Group II & III, which is less than Group IV. In other words, GI fluids have higher demulsibility retention property than GII and III based fluids.

Potential solutions for In-Service Turbine Oils with failing demulsibility

Since there is currently not an established way of consistently determining why a fluid is failing demulsibility, it should be noted that there is not a definitive solution. However several things have been done in the field with in-service turbine oils that suggest remedies are possible. One strategy that is discouraged is the introduction of demulsibility additives. These often don’t solve the problem and in many cases can increase water emulsions.  Many of these additives are polar in nature by design.  They therefore can easily interact with other chemistry in the fluid in either positive or negative directions.

The two strategies that show some promise are to increase the solubility of the oil or to remove the soft contaminants from the fluid.

Potential Solution 1: Increasing solubility

Increasing the solubility of in-service turbine oils appears to have a positive impact on improving the solubility in several cases. Increasing the solubility of polar molecules by a fluid is achieved by increasing the overall polarity of the fluid, not just a portion or component in the fluid.  This has been demonstrated using a solubility enhancing varnish removal cleaner (Fluitec Boost VR).

The following table demonstrates the impact on the fluid’s demulsibility in 18 cases when this solubility enhancer has been added to in-service turbine oil.

Table 1: Impact on Demulsibility on 18 In-Service Turbine Oils after the Addition of a Solubility Enhancer (Fluitec Boost)

The treat rate of the solubility enhancer used in all examples was 5%. The data suggests that improving a GII turbine oil’s solubility may correct water separation characteristics as can be seen in samples 6,7,13 and 14. It also suggests that a solubility enhancer may not have an impact on a GI oil with failing demulsibility, however more data is needed on these fluids.

There is also evidence to support that a bleed and feed with a GI turbine into a GII turbine oil with failing demulsibility may also restore demulsibility characteristics. Table 2 below shows the impact of a 30% bleed and feed of a GI turbine oil into a GII turbine oil. The demulsibility characteristics were improved and recent testing indicates that demulsibility has been maintained at 40/40/0 suggesting that this could provide more than a short-term fix.  This may be another example of a solubility enhancement of the fluid by a large addition of a weaker solubility enhancer.

Table 2: Impact on demulsibility after a 30% bleed in feed with a GI turbine oil in a steam turbine

Potential Solution 2: Removing Polar Contaminants

Polar contaminants may also be removed, which in some cases can restore water separability characteristics. Varnish mitigation systems like Fluitec’s ESP technology typically have a positive impact on water separability, however the results are not consistent enough to claim that this technology is a solution to all demulsibility problems.

Following is an example of removing polar contaminants to restore water separation characteristics. A steam turbine at an incineration plant in Germany had failing demulsibility on a Siemens steam turbine after a new charge of turbine oil. The results as tested by IP19 were >1200 mins. The corresponding ASTM D1401 results were 5/10/65(60) on the sample. Analysis revealed no clear evidence identifying the cause of the demulsibility failure. All other oil properties were within specification, including the MPC value.

The plant installed an ESP varnish mitigation system to try to correct the problem since the technology does have the potential to remove some types of polar contaminants.

Table 3: The Impact on Demulsibility after installing an ESP varnish mitigation system

Conclusion

A turbine oil’s ability to rapidly separate from water is one of its primary performance characteristics. If a fluid fails demulsibility and water contaminates the system, significant equipment damage can occur. A power plant may also have a false sense of security because their water removal technology may be ineffective if their turbine oils demulsibility properties have been destroyed.

The reasons that turbine oils fail demulsibility can be varied, yet finding a definitive root cause can be elusive. Without knowing the root cause of fluid failure, a consistent solution cannot be applied. However, through field testing two possible solutions appear to show promise: Improving a fluid’s solubility and removing polar contaminants.