Brazilian Journal of Chemical Engineering CHARACTERIZATION AND EVALUATION OF WAXY CRUDE OIL FLOW

Part of the oil found in the Brazilian subsoil has a high wax content, which makes its flow process difficult at low temperatures because of the increase in the viscosity of the fluid. This paper studied the flow behavior of waxy crude oil under variation in the temperature of the external environment of the flow, the volumetric flow rate of the oil and the emulsified water content of the oil. The results were compared with those obtained for a non-waxy crude oil that had similar rheological properties at temperatures above the wax appearance temperature (WAT). The proposed tests were based on the experimental design technique, and the behavior of the fluids was evaluated based on the pressure variation generated by the flow.


INTRODUCTION
Paraffinization is one of the main problems in oil production and causes considerable losses to the oil industry.The wax precipitation phenomenon associated with paraffin deposition can result in unscheduled production shutdowns and promote operational risk conditions.Moreover, it can cause production losses and irreparable damage to equipment (Pauly et al., 2004).
In the Bahian Recôncavo region, the produced crude oil exhibits a density of approximately 30° API, almost no sulfur and high concentrations of dissolved waxes.Although these properties are great for the manufacture of lubricant oils and yield high added value, the presence of wax adds many complications to production, transportation and storage by hindering the flow in pipes (Thomas, 2004;Novaes, 2009).
Paraffins are both linear (n-paraffins) and branched (iso-paraffins) chain alkanes, and they have low reactivity with most compounds.Their chains can have a high carbon number, which implies a higher wax appearance temperature.The low-molecular-weight paraffins are the main components of natural gas, and the medium-and high-molecular-weight ones are found in crude oil (Farayola et al., 2010;Gao, 2008;Jamaluddin et al., 2001).
Paraffins are in equilibrium with other crude oil components, and any change in pressure, temperature and even composition can affect the equilibrium, thereby influencing the formation of precipitate.According to Santos (1994), the greater the crude oil wax content, the greater the precipitation rate and, therefore, the amount of precipitated wax.The light oil components keep the waxes soluble.The high pressure of the reservoirs maintains the light compounds solubilized in the crude oil, which favors the solubili-  (Tinsley;Prud'Homme, 2010).This condition ensures low viscosity and Newtonian behavior of the crude oil (Azevedo, 2003).
Temperature also influences the solubility of waxes in crude oil.When approximately 5% of the waxes crystallize because of oil cooling, a crystal lattice appears and traps some of the oil inside; this process is called "gelling" and hinders the fluid flow.Thus, the crude oil flow rate also interferes with wax solubility.The lower the oil flow rate, the longer it stays inside the piping, which favors heat exchange with the external environment (Vieira, 2008).
Once production starts, the oil flows through the pipelines, losing heat to the external environment, with consequent temperature decreases and reduced soluble light oil fractions.Such production conditions cause the oil viscosity to increase, which leads to production problems due to the precipitation of waxes (Venkatesan et al., 2005;Gao, 2008).
According to Vieira (2008), the first paraffin crystals start to form at a specific temperature, which is called the wax appearance temperature (WAT) and varies depending on the origin of the crude oil.Crystallization occurs in three steps: Nucleation -formation of small particles of crystallized material from which the first paraffin crystals will grow.
Growth -mass transport of the solution towards the nuclei formed during the nucleation stage.
Agglomeration -when the growing crystals are joined together, thereby yielding larger crystals.
With the nuclei already formed, there is incorporation of new paraffin molecules at the growth sites, and additional molecules of other species are grouped at these sites and become part of the structure.The nuclei form an ordered lamellar-structure arrangement.
After crystallization starts in a medium that contains water as an emulsion, the crystal lattice formation phenomenon occurs in a different manner.When the emulsion is of the water-in-oil type, the oil is waxy and the fluid temperature is below the WAT, the precipitated waxes are deposited onto the surface of the water drops, thereby contributing to the growth of the formed precipitate (Oliveira et al., 2010).When a large crystal lattice is in the vicinity of the water drops, a structure is formed; this structure percolates the drops into the lattice and captures them.According to Visitin (2008), this structure also provides mechanical resistance to the flow, thereby resulting in an increase in the viscosity and pour point of the oil.
The present study aims to evaluate and compare the flow of two types of crude oil, waxy and nonwaxy, by measuring the pressure variation of the system under the influence of the flow rate, temperature and content of emulsified water.

Crude Oil
The characteristics that influenced the choice of oils used in this study were obtained from the rheological behavior of the samples.Although the available oils had different wax contents, WATs and compositions, for a comparative study of the influence of the content of emulsified water, temperature and oil flow rate in the context of loss of flow, it was necessary for the oils to be rheologically similar such that any differences originated exclusively from phenomena that characterize the increase in fluid viscosity and its implications for the flow.
After a series of comparative tests to search for a non-waxy oil with rheological behavior similar to that of the waxy oil above the WAT, a non-waxy oil was defined as the reference for comparison in the study.
The WAT of each oil was determined through differential scanning microcalorimetry, µDSC.The analysis was made using a DSC-VII microcalorimeter, Setaram, with a 500 µL stainless steel pressure cell and the data acquisition and analysis was made through the Setsoft 2000 software.The procedure realized in the tests consists of heating the sample to 80 °C during one hour and a sample of known weight is placed in the cell and then in the equipment.The analysis is made by cooling the sample from 80 °C to 0 °C at a rate of 0.8 °C/min.Microcalorimetry measures any release or absorption of heat by the sample while it cools.The evaluated temperature range consisted in the cooling of the oils from 80 °C to 0 °C.The only possible exothermic event in this temperature range is the release of heat related to the crystallization of waxy species present in the sample.The greater the crystallization peak area, the greater the amount of wax present in the sample, and the higher the temperature at which crystallization occurs, the greater the length of the carbon chain of the crystallized paraffins.Figure 1 shows the microcalorimetry curve of the waxy oil, which has a WAT of 310.5 K.
Figure 2 shows the microcalorimetry curve of the low-paraffin oil.This sample exhibits two points at which the line tangent to the crystallization curve crosses the abscissa, thereby generating a WAT at 316.4 K and a second crystallization event at 290.7 K.This effect occurs because the long-chain n-alkanes,   Initially, 1 ent is heated ntee the sol Then, the oil emperature i when the sam eeping its st ath unit at th During the 0 °C, so tha 5 °C.After s reheated to hen returns to The tempe nd output of pot meters, nce between ver time, wa Due to th ressure, the ure of 9 bar, ity.This lim alues with a ests relating o 8 bar, whe ific operatio een reached, f flow.The il were acco ue to a small Figure 4 sh  The highest levels of the variable "water cut" were set to be 35% because, when this level was greater than 40%, the viscosity increased, which yielded difficulties in pumping the sample.
The influence of the temperature on the crystallization of paraffins dissolved in crude oil, and consequently its viscosity, is of paramount importance to this work.Preliminary tests demonstrated that, below 291.15 K, the waxy crude oil used as a sample did not flow properly because of its high viscosity.For this reason, 293.15K was set as the lowest temperature level used in this study.
The highest temperature level was defined based on the need to expose the waxy crude oil to conditions in which crystallization of solubilized paraffins occurs, i.e., at temperatures close to its WAT.Therefore, the highest level was set to 298.15 K.
The residence time in the test tube is also a factor that influences the crude oil flow.The greater the residence time, the greater the heat exchange with the external environment, which results in a wider variation of fluid viscosity.Based on this information, it was assumed that 150 mL/min would be adequate for the minimum flow rate level.
The maximum flow rate level was set to 200 mL/min, which is close to its maximum operating condition, 275 mL/min.

Experimental Methodology
One liter of the sample with a determined water cut was initially heated to 333.15 K in a storage vessel such that the solubility of all paraffins in the oil was ensured.At the same time, the oil was circulated through the system's bypass until its temperature was stabilized.At the end of this step, the sample flowed through the cooling system.
During the test, the oil that fed the unit was cooled to 328.15 K.After the cooling that occurred throughout the test tube, the oil returned to the beginning of the process to be reheated to make the precipitated wax soluble.
The temperature and pressure values at the inlet and outlet of the test tube were recorded during the tests every 10 seconds.The pressure difference between the inlet and outlet of the test tube measured over time was the response obtained for each test.In most of the tests, the maximum pressure variation was achieved in less than two hours of operation.The results indicated that the best strategy to compose the experimental design was acquisition of pressure change data at a fixed time.The shortest time at which the maximum pressure was attained was used as the time to measure the differential pressure value for all tests.

Waxy Crude Oil
The tests were conducted according to the distribution in Table 2 and the pressure differential occurred due to the wax precipitation.The pressure difference was calculated from the pressure variation between two manometers, the first one located at the beginning of the flow and the second one at the end of the flow; these manometers present 99 percent accuracy.In most cases, the maximum pressure variation of the system occurred at different times.Because the goal of the research is to study the pressure variation based on the influence of the variables (the emulsified water in the crude oil (A), temperature (T) and flow rate (Q)), the differential pressure values considered as test responses were those attributed to the lowest operating time that reached the maximum pressure variation of the system.The worst flow condition was achieved in test 1, with a temperature of 293.15 K, flow rate of 150 mL/min and water cut of 5%; in this test a pressure difference of 8.03 bar was achieved in 36 minutes.Thus, the reference operating time was 36 minutes.The pressure variation responses are found in Table 3.The data in Table 3 were processed with the aid of parametric statistics.The empirical models presented in Figures 6 and 7 were evaluated for significance through analysis of variance based on the Pareto diagram shown in Figure 5.
It can be observed from Figure 5 that the variables that significantly influenced the flow process were the temperature, oil flow rate, the interaction between the temperature and oil flow rate and the interaction between the temperature and water cut.The significance of the interactions between the two variables leads to the conclusion that it is not possible to analyze the behavior of the system based on only one variable while keeping the other variables fixed.The experimental data were properly fitted to a linear plane model with a regression coefficient (R²)     uenced by th n of these tw of wax prese e that was th which was pr ater the amou stem viscosi plets was sim was significa influence w he greater th nsequently, th n.
G. B. Tarantino, L. C. Vieira, S. B. Pinheiro, S. Mattedi e Silva, L. C. L. Santos, C. A. M. Pires, L. M. N. Góes and P. C. S. Santos Brazilian Journal of Chemical Engineering zation of waxes in the fluid

Figure
Figure 1: M rude oil.

Figure
Figure 2: Mi rude oil.
Figure 6 s erature and ressure varia ressure drop nt temperatu ange.The hig st ambient t ressure drop when a higher

Figure
Figure 11: T

Table 4 p
sed in the sta