The Development and Field Application of New Surfactant Chemistries for Application to Heavy Oils

Heavy oil is characterised by having an API of less than 20° whereas extra heavy oil has an API lower than 10°. This means that this dense crude oil often has very high viscosities as well as having high molecular impurities such as asphaltenes. All this leads to the crude oil having a high boiling point. Heavy oil is becoming much more economical to oil producers due to the scarce nature of conventional oil sources and the current high oil prices. In the last ten years, oil prices have more than quadrupled and this current oil price of approximately $100 per barrel is expected to continue into the future. This is coupled with the technical advances in extraction methods which include the development of new novel demulsifiers.

It can be seen from the pie chart that 70% of the world's remaining recoverable crude oil reserves are not conventional with 40% being either heavy or extra heavy crude oil which pose a number of unique oil and water separation challenges in production, transportation and refining. There are still considerable oil reserves in North and South America, along with Russia, China and the Middle East. In these regions, especially the Americas, a lot of the available crude oil will be unconventional. (Slumberger, 2006)

Heavy crude oils pose many challenges focussing around demulsification in both production and refining applications. The first challenge is to demulsify the crude to separate the oil and the brine during production. Although the area of demulsification has had highly efficacious products developed for a number of scenarios, the application to heavy oils remains problematic due to the high viscosity of the oil and impurities present, therefore novel demulsifiers are required. In the refinery, demulsifiers are required to remove the salts from the crude oil/brine as desalters before refining as not to cause damage to the plant and also in dehazing the crude oil which ends up being used as fuel for its end use. In transportation, the high viscosity of the heavy oil can be problematic. Demulsifiers are therefore used as it is often easier to transport dehydrated oil without any brine present. As well as primary extraction, demulsification of challenging crudes is needed in a wide variety of secondary and tertiary enhanced oil recovery (EOR) applications and techniques. Two examples of such techniques are cyclic steam stimulation and steam assisted gravity drainage (SAGD) where separation of heavy oil and bitumen emulsions is required.

Theoretical models of how chemical demulsifiers break crude oil emulsions predict that the critical parameters determining the efficiency of the demulsifier are associated with the rheology of the oil-water interface. The work carried out here was to determine whether such correlations are justified in real systems by measuring the rheological parameters associated with the oil-water interface for various demulsifier species and correlating them with demulsifier performance. The principle of this model was then used to design more efficient demulsifier molecules. All demulsifiers were evaluated individually for comparsion and not as a blended demulsifier package.

The demulsifier requirements are such that it must be very interfacially active and needs to displace the surfactants in crude oil even at a use concentration of 10 ppm. A range of novel heavy oil demulsifiers have been developed with a wide demulsification chemistry portfolio including; resins, polymerics and esters to optimise and further develop more efficient molecules. Before a new heavy oil demulsifier is commercialised it is screened by a process which assesses its performance in a range of evaluations. Only if the new demulsifier molecule offers significant performance improvement on current products, will it be considered suitable for the marketplace.

Through customer relationships new demulsifiers undergo extensive field-testing in a range of different oilfields.

Interfacial tension is the term used to describe the surface free energy of a liquid, and indicates the strength of the forces within that liquid acting on the interface. The interfacial tension is the force which acts upon the liquid to reduce overall surface area, and is the same force which in a stable emulsion allows the individual droplets to remain dispersed and distinct from one another. In order to increase the surface area, such as in the destabilisation of an emulsion, the interfacial tension must be lowered to a value which allows the droplets to coalesce. In addition to reducing the surface tension, the elasticity of the interface (elasticity modulus) must also be reduced. An interface with a high modulus can be described like a rubber band; that is, it is flexible and can be deformed without breaking. In contrast, an interface with a low modulus is more like a strip of paper which easily breaks rather than stretching. From these analogies it can be appreciated that lowering the interfacial modulus and interfacial tension of an emulsion can cause destabilisation and eventual coalescence of the dispersed phase.

A sophisticated technique has been used to create profiles of interfacial tension and modulus for demulsifiers in model oils in order to predict the efficacy of the products in real-life situations. Use of these methods speeds up development of new materials and allows product recommendations to be made to suit specific conditions.

The IFT and elastic modulus measurements were run on a Teclis Tracker instrument. Image provided by Teclis, France.

Interfacial tension is obtained from the droplet shape using a series of equations (Young-Laplace equation and Equation for hydrostatic equilibrium at a plane passing through point M) that can be simplified to

where ∆ρ is the difference between densities of the two fluids, g is gravitation constant, b is 1/R, where R is radius of curvature for the drop at its apex. Note that the radii of curvature at the drop apex are equal, R= R', ω is the bond number, γ is the interfacial tension (mN/m). It is well known in the literature that the mechanism by which demulsifiers work is by initially displacing the natural "surfactants" in the oil which stabilize at the interface, before allowing coalescence of droplets because of the weak interfacial film formed. Hence, destabilization is achieved by reducing the interfacial tension at the emulsion interface. The elastic modulus (mN/m) was measured for each of the demulsifiers to enable the extent of emulsion coalescence stability to be assessed.

Demulsifiers that showed good performance in the IFT testing were then further evaluated in a range of crude oils with a range of differing API, water cuts, compositions (asphaltenic, napthenic, sweet, sour etc.) and from different oil producing regions.

The relative solubility number (RSN) of a demulsifier is a measure of its solubility properties. This is a key factor in demulsifier selection since solubility properties dictate whether the chemical will be capable of working effectively as a surface-active agent at the oil-water interface. The RSN is a measurement used to describe the hydrophile-lypophile balance of oilfield demulsifiers. It is a simple measure of water solubility and relates to an oil- water partition co-efficient. As such, it is ordinal rather than absolute and gives a relative measure of one product against another.

Solubility: visually test to see whether the solute is either; soluble, insoluble or dispersible in water Pour Point (°C) - ASTM D97

Viscosity (cP) and density (g/cm³): obtained using an Anton Paar SVM3000 viscometer

Appearance (visual)

pH (1% in 85/15 w/w in IPA/water)

Turbiscan Lab instrument was used to evaluate the demulsifiers which replicates on-field bottle testing. This accelerates and documents an ageing test for an in depth understanding of destabilisation mechanisms (creaming, sedimentation, flocculation, coalescence) of emulsions.

The measurement principle is that the dispersion is analysed in a cylindrical glass cell and the light source is an electro luminescent diode in the near infrared (880 nm). Two synchronous optical sensors receive respectively light transmitted through sample (180° from the incident light, transmission sensor), and light backscattered by the sample (45° from the incident radiation, backscattering detector). Work has been carried out in the lab to assess the feasibility of using this methodology as a substitute test for the bottle test in laboratory conditions. The analysis was carried out at 60°C and at a demulsifier concentration of 100 ppm. Good correlation was found between demulsification of an on-field crude oil and the analysis of the same oil on the Turbiscan in the laboratory.

A list of the initial chemistries and their physical properties are shown in Table 1.

To establish information regarding the equilibrium adsorption of the demulsifier materials at the oil-water interface, the IFT (mN/m) was measured over time (seconds) for these demulsifiers. Figures 3 and 4 show the IFT analysis for the demulsifiers listed in Table 1. The graphs show that the steeper the gradient, the faster acting the demulsifier and the lower the equilibrium IFT the more effective the demulsifier. Figure 5 shows the graph of the interfacial elastic modulus as a function of the interfacial tension. The interfacial elastic modulus is an important parameter because it is related to emulsion stability. High mechanical strength (high interfacial elasticity) equates to good stability.

The results show that an increased surfactant adsorption time results in a reduction in the value of interfacial tension and an increase in the value of the interfacial elastic modulus. However above an interfacial tension of about 15mN/m, the elastic modulus could not be accurately measured and these therefore have not been included in Figure 5.

The demulsifiers were then further evaluated in the Turbiscan, in some crude oils with a range of differing API and water cuts. A typical graph obtained from the Turbiscan is shown in Figure 6. (Actual image taken from TLab Thermo software). Three heavy crude oils with different APIs and basic sediment and water (BS&W) were used for screening of the demulsifiers and the details of the crude oils are given in Table 2.

Tables 3 – 5 show the Turbiscan results from the evaluation with the crude oils.

The relationship with regards to the characteristics affecting demulsifier performance show that molecular weight, relative solubility (RSN) and functional groups are key to providing good separation of the water and oil in heavy oils. The IFT test gives a bias towards high RSN demulsifiers, which are stronger water droppers and this is mirrored by the Turbiscan tests in the crude oils, as all demulsifiers < 8 RSN did not exhibit any separation. However, when the RSN was increased slightly (~11) the performance improved significantly regardless of the chemical backbone of the molecule. This is related to the extent of alkoxylation and those products that have high mixed alkoxylation levels, have better demulsification properties than single alkoxylated products. In general the more water soluble demulsifiers show rapid interfacial adsorption and high values of interfacial elastic modulus. The value of interfacial elastic modulus is related to the strength of interaction between molecules at the oil-water interface. Bulky hydrophobes based on resins, which cannot pack well at the interface, have lower elastic modulus and promote coalescence. With higher molecular weight chemistries, the interfacial adsorption rate decreases, the interfacial activity increases and the interfacial elastic modulus decreases. Therefore, for higher molecular weight demulsifiers the initial effect is slower but more effective overall. The other physical properties evaluated in this study show no correlation to performance but are required for the user to understand how to handle and use the demulsifiers in field conditions.

The chemistries chosen were varied to assess surfactants other than those based on nonyl phenol resins. Although the nonyl phenol based demulsifiers are still the best performers, other chemistries have been shown to be suitable as demulsifiers. In particular an ethoxylated polysorbate adipate (patent pending) gives good performance in heavy oil whilst classified as yellow and inherently biodegradable for use in the North Sea.

Internal development work was also carried out evaluating resins with longer alkyl chains as resin alkoxylate demulsifiers. These were phenolic resins with a C18 alkyl chain and a C24-28 alkyl chain. Each resin was then alkoxylated using mixed alkoxylation to low, medium and high levels. The samples were evaluated for RSN and IFT. From the results, a higher proportion of alkoxylation leads to a stronger performing product with a higher RSN. The C24-28 resin alkoxylates performed better than the C18 products, but the difference was much smaller than that between different alkoxylation levels of the same resin. Therefore the EO:PO mix and resin:alkoxylate ratio are more important for demulsification performance than the type of resin used.

Customer trials with demulsifiers were carried out and the leading products correlated with the performance in the lab, however a formulation of several demulsifiers is commonly used to obtain optimum performance in the field. Work is ongoing to evaluate the demulsifiers in heavy oils supplied from around the world.

Leading edge techniques have been used which measure demulsifier interfacial properties and hence how these interfacial properties relate, along with evaluation in heavy crude oils, to demulsifier performance. The elements of molecular structure which determine adsorption properties have been discussed and this understanding has guided the development of new demulsifier types with improved performance. Although the nonyl phenol resin alkoxylates are the best performing products, a variety of chemistries can be used in heavy oil including ethoxylated polysorbate adipates. In field situations, the most effective demulsifiers are likely to be a formulation including high and low molecular weight materials. The high molecular weight species occupies a large area at the interface which reduces the surfactant concentration, whereas the lower molecular weight materials will aid rapid relaxation of the interfacial tension. The difficulty in designing better performing demulsifiers lies in balancing these various parameters, some of which are in opposition.

The authors would like to thank the help and support of all Croda employees involved in the production of this paper. The authors would also like to thank Recherche Exploitation Produits (REP) and Star Energy for the supply of crude oil along with Teclis and Fullbrook Systems for technical support using the Tracker and Turbiscan.