OTC--25927--MS - SUEZ Water Technologies & Solutions Jul-16 ... The stress at which the sample...

Water Technologies & Solutions technical paper

Find a contact near you by visiting www.suezwatertechnologies.com and clicking on “Contact Us.” *Trademark of SUEZ; may be registered in one or more countries. ©2017 SUEZ. All rights reserved.

OTC-25927-MS_May2015.docx Jul-16

OTC--25927--MS evaluation of paraffin wax inhibitors: an experimental comparison of bench-top test results and small-scale deposition rigs for model waxy oilsPeter Perez, Eugene Boden, Kelly Chichak, A. Kate Gurnon, Lishun Hu, Julia Lee, John McDermott, John Osaheni, Wenqing Peng, William Richards and Xiaoan Xie, SUEZ Global Research

Copyright 2015, Offshore Technology Conference This paper was prepared for presentation at the Offshore Technology Conference held in Houston, Texas, USA, 4–7 May 2015. This paper was selected for presentation by an OTC program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect any position of the Offshore Technology Conference, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Offshore Technology Conference is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of OTC copyright.

abstract The performance of paraffin inhibitors from several chemical families was evaluated using two model waxy oils in bench-top tests such as cloud point and pour point, and in small-scale deposition rigs using a cold finger and a flow loop. In addition, rheological characterization of the inhibited systems was conducted to provide additional insights into the function of the paraffin inhibitors. Similarities and differences in the response of the model oils to the paraffin inhibitors are discussed along with the correlations between the results from bench-top tests and the observed wax deposit inhibition performance. The model waxy-oils were formulated using blends of commercially available waxes dissolved in a paraffinic solvent or a mixture of paraffinic and aromatic solvents. The wax blends were chosen to represent a broad range of carbon number distributions including one highly polydisperse distribution with an extended distribution of high carbon number waxes exceeding C50. Candidate paraffin inhibitors were initially evaluated for their impact on cloud point and pour point in these model oils. Subsequent testing of the inhibited systems included rheological characterization over a range of temperatures, cold finger deposition tests and flow loop deposition tests in a pipe-in-pipe geometry. All results are compared to the baseline results for the uninhibited model waxy oils. While changes in cloud point and pour point gave a good indication that a particular paraffin inhibitor was impacting the wax crystallization process, this did not guarantee a significant reduction in deposits of the inhibited system. Moreover, the experimental conditions used in the small-scale deposition tests also affected the observed performance of paraffin inhibitors, indicating that temperature gradients (i.e., oil temperature and coolant temperature) must be optimized to achieve the highest discriminating power. This study highlights the degree of correlation and the areas of departure between bench-top test- results and the main objectives of deposit reduction, and yield stress reduction in flowing systems. It also discusses the advantages and risks of using only the bench-top test results for the selection of paraffin inhibitors.

OTC-25927-MS_May2015.docx

introduction The precipitation of solid deposits during the production and transportation of crude oils, especially from offshore reservoir facilities, represents a serious operational and economic issue [1, 2, 3, 4]. Although the composition of these deposits is very complex, it is well known that normal paraffins are major constituents, which precipitate from the bulk liquid phase as the crude oil cools down in the pipeline during transportation [2, 3]. Preventive measures to minimize wax deposition problems involve heat insulation of subsea pipelines and the use of paraffin inhibitors [3, 5, 6]. When wax deposition occurs to an extent where crude oil production becomes compromised, mechanical removal of the deposits (pigging) becomes necessary, with the consequent halt or reduction in oil production and increase in operational costs [3, 7]. In the case of subsea tie-ins across long distances, pigging becomes technically and economically challenging. Thus, the use of paraffin inhibitors to help delay or prevent the accumulation of wax deposits becomes a cost-effective preventive measure [3, 6]. The effectiveness of paraffin inhibitors is typically assessed by the observed reduction in cloud point and pour point using standard methods [8, 9], and its wax deposition performance in small-scale rigs, such as the cold finger device [10]. These bench-top tests are relatively quick and easy to perform with a large number of candidate paraffin inhibitors for a particular crude oil and decisions are often made with those results [1, 6, 11]. However, the results from bench-top tests do not always translate to tests under flowing conditions, e.g., flow recirculation loops [1, 4, 12], which seek to reproduce temperature gradients, pressure, flow field, shear environment, etc., that the oil experiences in the pipeline [1, 10]. This study highlights the degree of correlation and the areas of departure between bench-top test results and the main objectives of deposit reduction, and yield stress reduction in flowing systems. Several paraffin inhibitors were evaluated for their impact on cloud point and pour point in model waxy oils formulated from refined paraffin waxes and commercial solvents. Subsequent testing of the inhibited systems included deposition tests on a cold finger and deposition tests in a pipe- in-pipe flow loop. The advantages and risks of using only the bench-top test results for the selection of paraffin inhibitors were discussed.

experimental materials model waxy oils. The model waxy oils were prepared by dissolving highly-refined paraffin waxes in either a paraffinic solvent or a mixture of the paraffinic solvent and an aromatic solvent. Paraffin waxes of different melting points were used to formulate model waxy oils with broad molecular weight distributions. As shown in Table 1, waxes with melting points from 44° C to 68 °C were dissolved in the solvents to achieve total wax contents of 5 % to 10 % by weight. The carbon number distribution of the waxes in the model oils was determined using high-temperature gas chromatography, and the composition of the wax was calculated according to the protocol established in the standard method ASTM D5442 [13]. Figure 1 shows that the carbon number distribution of the model oils is very broad, comprising hydrocarbons from C18 up to C50, where normal alkanes are major constituents.

wax inhibitors/pour point depressants Chemical additives used as wax inhibitors and/or pour point depressants were commercially sourced. Table 2 indicates the chemical type for each additive studied.

equipment

cloud point/pour point apparatus The cloud point and pour point of the model waxy oils were determined using an automatic instrument, according to the standard methods ASTM D5773 and ASTM D5949, respectively [14, 15]. The automatic instrument utilizes Diffusive Light Scattering (DLS) to determine the cloud point and pour point in a single run with 150 μL of sample. During a test, a monochromatic light beam (660 nm) is directed at an acute angle onto a sample placed in a polished sample cup, and it is reflected toward the chamber lid and absorbed while the sample is homogeneous. As the sample is cooled down and crystals start to form at the cloud point, the light beam is scattered at the solid-liquid boundaries. The scattered light impinges on optical detectors placed directly above the sample cup. Sample temperature is precisely controlled using a Peltier device [14]. During a pour point test, the sample is first heated and then cooled down at 1 °C or 3 °C intervals (defined by analyst) while using the optical detectors to monitor the movement of the sample surface after a pulse of dried air is directed onto the sample. Light


scattered by the ripples on the sample surface created by the air puff is monitored (as an indication of movement) at sequentially lower temperatures until no surface movement is detected. The lowest temperature at which movement was observed is then deemed the pour point of the sample [15]. A detailed description of the equipment and principle of operation can be found in the respective methods [14, 15].

rheology Rheology experiments were performed on an stress controlled rheometer using a Couette geometry, which has an inner cylinder diameter of 28 mm and an outer cup diameter of 30 mm. The sample is conditioned at 65 °C for 15 minutes. Then the sample is cooled from 65 °C to 5 °C at a rate of 1.0 °C/min while a simultaneous torque of 25 μNm (1.99 Pa) torque is applied to the sample. Once the sample has reached 5 °C the sample is held at a constant temperature for 15 minutes with no shear. The yield stress (σy) is defined using a stress sweep during which the sample is exposed to an initial stress of 1 Pa and then increasingly larger shear stresses until the sample shears. The stress at which the sample begins shearing is defined as the yield stress of the wax-oil solution.

cold finger The cold finger is an attachment to a rotational rheometer. The cold finger is designed such that the temperature of the inner cylinder (Figure 2) is maintained by a circulating water bath at 10.0 ± 0.1 ° C and held fixed. While the outer cup (Figure 2) temperature is maintained by an oven at 30.00, 25.00, or 20.00 ± 0.01 ° C. The outer cup rotates at a constant rate of 50 s-1 during the 4 hour experiment. This contrast in temperature creates a temperature gradient across the gap (5.46 mm) between the inner cylinder and outer cup during the experiment, which promotes wax precipitation and deposition on the colder side. To determine the final amount of material deposited onto the cold finger the inner cylinder is initially weighed prior to the experiment and then again at the end of the experiment. The deposit is then scraped off from the cold finger cylinder using a razor blade and further compositional analysis on the deposit is made using high-temperature gas chromatography (HTGC). wax deposition flow loop A Flow Loop Apparatus was used to simulate wax deposition under more realistic flow conditions (Figure 3). The flow loop rig consists of a pipe-in-pipe heat exchanger through which the model waxy oil (at a bulk temperature close to its wax appearance temperature) was circulated and cooled down in the test section with cold water, forcing the precipitation of wax. The test section was built from stainless steel tubing, using an inner tube (hot side) with a nominal OD of 3/8” (9.52 mm), and an outer tube (cold side) with a nominal OD of 1” (25.4 mm). The total length of the test section is 1.52 m. The temperature of the oil and the coolant can be varied independently. Initial oil temperatures were adjusted according to the wax appearance temperature of the waxy oil to be tested, but were typically from 20 °C to 40 °C. Coolant temperatures were adjusted to have temperature gradients (ΔT = Toil – Tcoolant) of 10 °C to 15 °C. A 10-gallon reservoir, typically filled about 80% with the test oil, was used to minimize wax depletion during deposition tests. A rotary vane pump delivering flow rates of 0.7 L/min to 5.7 L/min was used to enable operation from the laminar flow regime (Re ~ 1,000) to the turbulent flow regime (Re ~ 8,000). Flow rates were measured using a Coriolis flowmeter installed downstream of the rotary vane pump. Wax deposition was monitored through the increase in pressure drop indicated by a differential pressure transmitter, which is correlated with the inner pipe diameter through the Darcy-Weisbach equation [4]. Calibration experiments using the paraffinic solvent were performed to determine experimentally the inner diameter of the clean pipe [4]. Additional details of this experimental rig are shown in Table 4. The deposition experiments in the flow loop were started once steady state conditions of flow rate, oil temperature and coolant temperature. The general protocol for the experiments consisted of: (1) oil conditioning, in order to erase the thermal history of the waxy oil and ensure that waxes were completely solubilized. This was done by heating the waxy oil above its cloud point for at least 1 hour. (2) Start of test, where experimental conditions were stabilized and data collection was started. (3) End of test, where data collection was stopped, the cold section was disconnected and the deposits accumulated in the inner tube were removed with a push rod for quantification (weighing) and analysis by high-temperature gas chromatography to determine the wax content (and entrained solvent content) of the deposits.


high-temperature gas chromatography (HTGC) Solid deposits collected from tests in the cold finger and flow loop were analyzed by high-temperature gas chromatography to determine the carbon number distribution and content of paraffins (waxes) using a protocol based on the standard method for the analysis of petroleum waxes, ASTM D5442 [13], but extended to 615 °C, the boiling temperature of C72. The analyses were performed in a gas chromatograph equipped with an autoinjector, a cool-on-column (COC) injection port and a flame ionization detector (FID). The samples were diluted in carbon disulfide (CS2) at a concentration of 2% by weight, using a sample injection size of 1 μL. A high-temperature column, 100% dimethyl polysiloxane, 30 m long, 0.53 mm inner diameter, 0.25 μm thick stationary phase was used for the full resolution of normal paraffins C5 – C72. The initial oven temperature was 40 °C and it was heated to 410 °C using a rate of 10 °C/min. Helium was used as the carrier gas at a flow rate of 5 mL/min. The calibration of the retention times was performed by injecting 1 μL of a mixture of C5 – C72 normal paraffins, while the response factor of the FID was determined using n-hexadecane (0.05 % by weight in carbon disulfide). Standard wax analysis calculations to determine the total content of waxes in the deposits (paraffins > C18) were performed using a simulated distillation software.

results and discussions cloud point and pour point of model waxy oils. The effectiveness of an additive to inhibit wax precipitation and deposition is often assessed by the reduction of the pour point [5, 6] and, in some cases, the reduction of cloud point [3, 16, 17]. Table 5 presents a summary of the data on cloud point (CP) and pour point (PP) using the automatic instrument, and the wax appearance temperature (WAT) and no-flow point at 5 °C determined from rheological measurements. Figure 4 shows graphically the reduction effect of the wax inhibitors on the cloud point of Model Oil 1 as a function of concentration of active polymer. In general, the effect of the additives on cloud point was sensitive to dosage, decreasing the cloud point of the model oils as concentration increased. This was the case for the alpha olefin maleic anhydride esters, the polyacrylate, the EVA copolymer. An exception was observed with Additive A (alkylphenol), which was less effective at 200 ppm than it was at 100 ppm. Furthermore, the highest concentration dose of Additive A at 400 ppm is only equally as effective at reducing the cloud point as the other additives tested at concentrations of 100 or 200 ppm. These results are consistent with those reported by Tinsley et al., in which maleic anhydride copolymers also reduced the cloud point of a model waxy oil by 2 °C – 3 °C [17]. They explained the cloud point depression by a mechanism of thermodynamic suppression of nucleation which was associated with a reduction in wax deposition on a cold surface [17]. Paso et al. also proposed a thermodynamic solubilizing effect of polymeric additives to explain the cloud point reduction of a model oil system [18]. Figure 5 shows the pour point reduction of Model Oil 1 by the same wax inhibitors at the same additive concentrations as those presented in Figure 4. In the case of alpha olefin maleic anhydride esters (Additives B, C, and D) the maximum pour point reduction does not correspond to the maximum additive concentration, while Additive A and Additive F were the most sensitive to additive dosage. Generally, the cloud point and pour point are correlated with one another and the two together are used to predict the effectiveness of an additive at reducing the wax deposited in a pipeline. Comparing Figure 4 and Figure 5 we observe that Additive A and Additive E have the largest effect on pour point, but they are less effective at reducing the cloud point of Model Oil 1. Additive C and Additive F were very effective to reduce pour point at the lowest dosage (50 ppm), but pour point does not decrease further at higher dosages as does the cloud point. Additive C and Additive D also reduce cloud point as dosage increases, but the pour point shows maximum reduction around 100 ppm. Previous work has proposed that the differences in dosage effectiveness of the additives to reduce cloud point and pour point probably result from the predominant action mechanism of the active polymers as paraffin solubilizers, wax crystal modifiers, and interfacially-active agents [18, 19]. The correlation between pour point reduction and cloud point reduction can be more easily interpreted from Figure 6. Overall, the data show that for a reduction in cloud point of less than ~ 1 °C there is a systematic increase in pour point reduction. However, when the cloud point reduction is greater than ~ 1 °C the pour point reduction does not exceed ~ 30-40 °C. It will be shown in the following that this limited correlation between the cloud point and pour point makes it difficult to predict an additive’s effectiveness at reducing wax deposition from pour point results determined by the ASTM D5949 method.


determination of additive effectiveness with rheometry methods The wax appearance temperature measured by a rheometry experiment is analogous to the cloud point defined by the diffuse light scattering (DLS) measurement using the automatic instrument and are both definitions for the onset of wax precipitation. Though reasonable agreement should be expected, each of the measurements is done with different sample volumes and wax precipitation is detected at different length scales. Therefore, achieving exactly the same temperature of wax precipitation is difficult, if not impossible. Similarly, the pour point measured using diffuse light scattering is analogous to the no flow point defined from a rheometry experiment. Both of these define the temperature at which a transition between a liquid waxy-oil transitions to a solid waxy-oil state. A comparison of the two methods clearly demonstrates how the two experiments potentially result in different transition temperatures. For instance, in the diffuse light scattering method a pulse of air of undetermined strength is directed at the surface of the waxy-oil solution (at the sample-air interface) and the pour point is defined as the temperature at which no change in the light scattered due to ripples at the sample-air interface is detected. For the no flow point the rheometer applies a pre-determined stress of 25 μNm to the bulk sample during cooling. As temperature decreases, the no flow point is defined as the temperature at which there is no change in the step displacement; i.e. no flow. A no flow point is only defined for the neat Model Oil 1 at 28.4 °C for the conditions defined for the rheometry experiment. For all of the samples containing an additive there is no measurable no flow point using the rheometry method. This result is presented in Table 5. Figure 7 is an example of the Arrhenius plot used to define the wax appearance temperature (WAT) for the model waxy-oil during a rheometry experiment. During steady shear the viscosity is measured as a function of decreasing temperature. The WAT is defined as the temperature at which the viscosity increases by more than 0.01 mPa s. The complete set of WAT and no flow point (NFP) is reported in Table 5. Interestingly, for Additives C, D and E the samples having 50 and 100 ppm have the same WAT as measured by rheometry where the CPs are different as they are defined using DLS. This is a simple but telling comparison of the two measurements given our discussion about the different length scales and the sensitivity of each measurement. Clearly, the CPs are capable of differentiating between the two samples that have 50 and 100 ppm of additive whereas the rheometry measurement cannot. For the samples reported in Figure 7 the WAT of Model Oil 1 with no additive is the highest at 33.70 °C and with increasing additive concentration the WAT decreases to 32.26 °C with the addition of 200 ppm Additive C. Clearly there are differences in the viscosity measured at temperatures less than the WAT for all samples. Specifically, over a 3 °C decrease in temperature the viscosity of the neat Model Oil 1 increases by an order of magnitude while those samples with Additive C never reach those same viscosities during the 25 °C temperature decrease. The stark difference between the neat Model Oil 1 and those samples with an additive at any concentration indicates that the additive affects the nature of wax precipitation and the resulting wax/oil slurry. During the temperature cooling ramp experiment the NFP is defined as the temperature after which there is no additional deformation measured as defined by the Displacement Step (radians). We measure a clear NFP (28.40 °C) in Figure 7 and yield stress (17.10 Pa) for Model Oil 1. However, under the experimental conditions measured, no NFP or yield stress is measured for the Model Oil 1 with any concentration of Additive C. The rate at which the viscosity rises as the temperature decreases is different among all of the samples in Figure 7. The step change in the viscosity for the neat surrogate fluid clearly dictates the WAT of the sample. However, for those samples with an additive, the steady increase in viscosity makes defining the WAT particularly difficult. Furthermore, the true WAT may be much higher than the temperature measured during the rheometry experiment given that the rheometry experiment defines the WAT based on a bulk sample measurement (20 mL) rather than one made at considerably smaller length scales and sample size, as those experiments which are performed using the automatic instrument for cloud point (ASTM D5773). Other techniques even define a WAT 20 °C or more, higher than that measured by either of these measurements. The discrepancy is attributed to the much smaller sample volume and detectable length scales measured using DSC [20, 21] or cross-polar microscopy. Figure 8 presents the correlation between the WAT measured by rheometry experiments and the Cloud Point (CP) measured by DLS ASTM D5773. The rheometry experiment measures the bulk mechanical properties of the solution (22 mL) to define the WAT - measuring the shear rate of the sample defines the viscosity as a function of temperature. In contrast, the DLS uses light to interrogate smaller length-scales and fluid volumes (150 μL) to


detect changes in the light scattered from the solution during cooling thereby defining the CP. Furthermore, the crystal morphology may play a significant role in the rheological response of the waxy-oil slurry however; a change in the aspect ratio of the crystals may not affect the amount of scattered light measured using DLS. The correlation presented in Figure 8 between the two experiments is acceptable, with two notable exceptions, considering the two experiments define the onset of wax precipitation by two different methods, sample volumes and at different microstructure length scales.

wax deposition experiments in the cold finger Figure 9 presents the reduction in total deposit weight on the cold finger as a function of additive and additive concentration. A clear trend is maintained for all additives interrogated –the higher the additive concentration, the larger the reduction in the amount of deposits on the cold finger. At the highest concentration we recorded more than an 80 % reduction in deposit weight for all additives. The most significant differences between additives appear at the lower additive concentrations, for instance at 50 ppm additive concentration, the effectiveness of the additive at reducing the deposit weight ranges from 30 % to almost 70 % reduction. Additive B was the least dose-sensitive when reducing the amount of deposits on the cold finger experiments, mainly because its performance was very good even at low dosage. In contrast, cloud point measurements reported in Figure 4 showed that increasing Additive B concentration proportionally increased the cloud point reduction. Clearly, the responsiveness to dosage of the bench-top test to determine the cloud point was not observed in the more dynamic conditions of the cold finger. Figure 10 presents a weak correlation between the pour point reduction and deposit reduction in a cold finger experiment. We find that a measure of the reduction in pour point for the additives measured is not an adequate predictor for how well the additive, at a particular concentration, will perform in inhibiting wax deposition on the cold finger. The method by which the pour point is defined for the ASTM D5949 standard method uses diffuse light scattering. Figure 11A presents a strong correlation between the cloud point reduction measured by the standard diffuse light scattering method defined by ASTM D5773 and the deposit reduction measured by the cold finger experiment. The results suggest that a significant reduction in cloud point is a good indicator of a subsequent reduction in waxy-oil deposited onto the cold finger. Interestingly, a 0.5 to 3.0 °C decrease in the cloud point corresponds to the entire range of additives that can reduce the cold finger deposit from 35 % to greater than 80 %. A compositional analysis performed with HTGC shows that an increase in the concentration of Additive B decreases the amount of wax deposited (g) onto the cold finger however, the mass percent (%) of wax increases.

wax deposition experiments in the flow loop The deposition experiments in the flow loop produced data on the performance of the additives at different conditions for Model Oil 1 and Model Oil 2. The oil and coolant temperatures were set according to the cloud point of the model oil to be tested, targeting temperature gradients (Toil – Tcoolant) of 10 – 20 °C. Table 7 and Table 8 summarize the data in terms of total deposit weight and deposit weight reduction observed with the additives with respect to the untreated oil. The dataset also includes the compositional analysis to determine the wax content (C18+ paraffins) of the deposits, which was used to calculate the amount of wax accumulated during the test. Figure 11B summarizes results from cold finger and flow loop experiments. There is an apparent trend between the reduction of cloud point and the amount of deposit observed to accumulate on the cold finger. A weaker trend with cloud point reduction is observed for those experiments performed in the flow loop. This is attributed to the difference in the temperature gradient between the two experiments as well as the different flow conditions [10]. The cold finger operates under laminar flow conditions while the flow loop operates in the turbulent flow regime. Figure 12 compares two deposition conditions with Model Oil 1 at a constant temperature gradient of 10 °C, but using different oil and coolant temperatures. For the condition Toil = 35 °C and Tcoolant = 25 °C, both Additive B and Additive C were effective to reduce the amount of total deposits (wax + solvent) accumulated, but the amount of wax actually increased compared to the baseline. This was because the wax content of the deposits in the presence of the additives increased 2-3 times with respect to that of the deposits from the untreated oil, which also led to hard deposits that were harder to remove. Similar results have been reported in the literature [3, 4, 11], but the fundamental reason for the reduction in the amount of entrained oil in the presence of paraffins inhibitors has yet to be established. It has been argued that some additives affect the porosity of the deposit [6],


which would directly affect the process that lead to deposit aging [10]. Notice that the condition Toil = 40 °C, Tcoolant = 30 °C, resulted in much lower deposit accumulations. Table 5 shows that the cloud points of the untreated and treated oils are 32 - 33 °C, still slightly above the coolant temperature, so it might be the seems that longer deposition times would have been required to increase deposition to levels that could allow to detect the inhibition performance of the additives. These results suggest that the actual inner wall temperature (Twall) is closer to the cloud point (CP) of the oils, which decreases the thermal gradient drive for wax accumulation CP – Twall [4, 12]. Oil temperature in this condition was also higher, which has the indirect effect of increasing the inner wall temperature leading to lower deposition [12]. The results obtained from the deposition studies with Model Oil 2 also highlight the importance of choosing appropriate conditions (i.e., temperature gradient), and the appropriate system (i.e., model waxy oil) to compare the performance of chemical additives. Notice in Figure 13 that the condition with the larger temperature gradient (Toil = 30 °C, Tcoolant = 5 °C) resulted in almost similar or higher deposit accumulation in the presence of Additive A and Additive C, leading to the conclusion that the additives were not very effective wax inhibitors. At Toil = 20 °C and Tcoolant = 10 °C, both additives were effective, with Additive A showing superior performance. Interestingly, Additive A not only decreased the amount of total deposits (wax + entrained solvent), but also reduced the amount of wax with respect to the untreated oil. In this case, the deposits were also hard due to the higher wax content of the deposits, but the reduction of total deposits was 71%, leading to a lower amount of wax. Future studies are focused on correlating the chemical structure of the additive to its performance. Figure 14A explores the correlation between the performance of the additives in the flow loop tests (in terms of deposit reduction %) and the cloud point reduction determined in the bench-top apparatus. As opposed to the relatively good correlation observed between performance and cloud point depression for the tests in the cold finger (Figure 11), this correlation was found to be very weak in the flow loop. This results from the dependency of wax deposition on the shear environment [1, 22]. A simplified explanation could be that the quiescent environment of the bench-top test for cloud point allows for the long-chain paraffin molecules to align as temperature decreases at a low cooling rate, which facilitates the nucleation processes [1]. At the flowing conditions in the flow loop this nucleation process is affected by higher shear rates and higher cooling rates, which probably govern over the marginal effects that paraffin inhibitors might have on the nucleation process. Figure 14B shows that pour point reduction was not correlated to the deposition reduction observed in the flow loop experiments, as it was the case in the cold finger experiments, Figure 10. The comparison of the particular subset of the performance data of the additives in the cold finger and the flow loop shown in Figure 11 warns about the risks of extrapolating performance (deposit reduction %) and dosage sensitivity from cold finger data to turbulent flow conditions such as the flow loop. For the experimental condition with similar temperature gradient (Toil = 35 °C; Tcoolant = 25 °C), Additive B reduced deposit formation in the cold finger by 64 % at 50 ppm, and it was even better (85 %) at 200 ppm. However, the performance observed in the flow loop was less encouraging. The deposit reduction was 35 % at 50 ppm and it only was marginally improved to 42 % when concentration was increased to 200 ppm. Thus, not only Additive B showed better performance in the cold finger, but also the cold finger test was more responsive to additive concentration than the deposition test in the flow loop. The takeaway message is that experimental conditions in cold finger and flow loop deposition tests must be carefully selected to ensure the proper screening of chemical inhibitors. These results also create the need to develop consistent guidelines about pass/fail criteria in the cold finger that would translate to flow loop tests and—further down—to oilfield tests.


conclusions The performance of several paraffin inhibitors was evaluated in terms of the reduction in the cloud point and pour point of two model waxy oils, and inhibition of deposit accumulation in cold finger and flow loop experiments. The degree of correlation between bench-top tests and deposition tests were discussed. The main conclusions from these studies follow: • The cloud point determined by a bench-top apparatus (ASTM D5773) and the wax appearance temperature determined by

rheology showed reasonable agreement, considering the differences in the experimental methods used to define the appearance of the first wax crystals.

• The cloud point reduction and deposition reduction in a cold finger correlated well, suggesting that the cloud point is a good predictor for the effectiveness of chemical additives to reduce wax deposition.

• The cloud point (ASTM D5773) was found to be a weak predictor of deposition reduction of chemical additives in the flow loop.

• The pour point (ASTM D5949) was a poor predictor of deposition reduction in both cold finger and flow loop experiments. • The experimental conditions chosen in the flow loop experiments impacted the apparent efficacy of the additive.

references [1] C. Lira-Galeana and A. Hammami, "Wax precipitation from petroleum fluids: A review," Developments in Petroleum Science, pp. 557-608, 2000. [2] D. Merino-Garcia and S. Correra, "Cold flow: A review of a technology to avoid wax deposition," Petroleum Science and Technology, vol. 26, no. 4, pp. 446-459, 2008. [3] A. Aiyejina, D. P. Chakrabarti, A. Pilgrim and M. Sastry, "Wax formation in oil pipelines: a critical review," International Journal of Multiphase Flow, vol. 37, no. 7, pp. 671-694, 2011. [4] R. Hoffmann and L. Amundsen, "Single-Phase Wax Deposition Experiments," Energy & Fuels, vol. 24, p. 1069 – 1080, 2010. [5] K. S. Pedersen and H. P. Ronningsen, "Influence of wax inhibitors on wax appearance temperature, pour point, and viscosity of waxy crude oils," Energy \& Fuels, vol. 17, no. 2, pp. 321-328, 2003. [6] R. Hoffmann and L. Amundsen, "Influence of wax inhibitor on fluid and deposit properties," Journal of Petroleum Science and Engineering, vol. 107, pp. 12-17, 2013. [7] T. S. Golczynski and E. C. Kempton, "Understanding wax problems leads to deepwater flow assurance solutions," World Oil, vol. 227, no. 3, 2006. [8] ASTM D2500-11, Standard Test Method for Cloud Point of Petroleum Products, West Conshohocken, PA: ASTM International, 2011. [9] ASTM D97 - 12, Standard Test Method for Pour Point of Petroleum Products, West Conshohocken, PA: ASTM International, 2012. [10] R. Venkatesan, V. Sampath and L. A. Washington, "Study of Wax Inhibition in Different Geometries," in Offshore Technology Conference, OTC-23624-MS, Houston, TX, 30 April-3 May, 2012. [11] D. W. Jennings and K. Weispfennig, "Effect of shear on the performance of paraffin inhibitors: Coldfinger investigation with Gulf of Mexico crude oils," Energy & Fuels, vol. 20, no. 6, pp. 2457-2464, 2006. [12] H. O. Bidmus and A. K. Mehrotra, "Solids deposition during “cold flow” of wax- solvent mixtures in a flow-loop apparatus with heat transfer," Energy \& Fuels, vol. 23, no. 6, pp. 3184-3194, 2009. [13] ASTM D5442-93(2013), Standard Test Method for Analysis of Petroleum Waxes by Gas Chromatography, West Conshohocken, PA: ASTM International, 2013. [14] ASTM D5773-10, Standard Test Method for Cloud Point of Petroleum Products (Constant Cooling Rate Method), West Conshohocken, PA, 2010: ASTM International, 2010. [15] ASTM D5949-14, Standard Test Method for Pour Point of Petroleum Products (Automatic Pressure Pulsing Method, West Conshohocken, PA: ASTM International, 2014. [16] M. d. C. Garcia, L. Carbognani, M. Orea and A. Urbina, "The influence of alkane class-types on crude oil wax crystallization and inhibitors efficiency," Journal of Petroleum Science and Engineering, vol. 25, no. 3, pp. 99-105, 2000. [17] J. F. Tinsley, R. K. Prud'homme, X. Guo, D. H. Adamson, S. Callahan, D. Amin, S. Shao, R. M. Kriegel and R. Saini, "Novel laboratory cell for fundamental studies of the effect of polymer additives on wax deposition from model crude oils," Energy \& fuels, vol. 21, no. 3, pp. 1301-1308, 2007. [18] K. G. Paso, K. K. Kruckert, H.-J. Oschmann, H. Ali and J. Sjoblom, "PPD architecture development via polymer--crystal interaction assessment," Journal of Petroleum Science and Engineering, vol. 115, pp. 38-49, 2014. [19] F. Yang, Y. Zhao, J. Sjoblom, C. Li and K. G. Paso, "Polymeric Wax Inhibitors and Pour Point Depressants for Waxy Crude Oils: A Critical Review," Journal of Dispersion Science and Technology, 2014. [20] D. Erickson, V. Niesen and T. Brown, "Measurement and Prediction of the Kinetics of Paraffin Deposition," in SPE 68th Annual Technical Conference and Exhibition, SPE 26548, Houston, TX, 1993. [21] D. Erickson, V. Niesen and T. Brown, "Thermodynamic Measurement and Prediction of Paraffin Precipitation in Crude Oil.," in SPE 68th Annual Technical Conference and Exhibition, SPE 26604, Houston, TX, 1993. [22] R. Venkatesan, N. Nagarajan, K. Paso, Y.-B. Yi, A. Sastry and H. Fogler, "The strength of paraffin gels formed under static and flow conditions," Chemical Engineering Science, vol. 60, no. 13, pp. 3587-3598, 2005.


table 1: composition of model waxy oils

Component Weight % in solvent Model Oil No. 1 Model Oil No. 2

Wax, m.p. 44 °C 5.0 % -- Wax, m.p. 53 °C -- 5.0 % Wax, m.p. 68 °C 5.0 % --

Solvent (Balance) Paraffinic/Aromatic Blend (7:3) 100 % Paraffinic Total Wax (C18 +) 10.0 % 5.0 %

n-Alkanes 7.0 % 4.5 % Unknown 3.0 % 0.5 %

table 2: chemical additives used as wax inhibitors/pour point depressant.

Inhibitor Chemistry Additive A Alkylphenol Additive B Alpha olefin maleic anhydride ester. Additive C Alpha olefin maleic anhydride ester Additive D Alpha olefin maleic anhydride ester Additive E Alpha olefin maleic anhydride ester Additive F Polyacrylate, ethylene-vinyl acetate copolymer

table 3: specifications of the automatic instrument for the determination of cloud point and pour point.

Property Method Temperature Range Cooling Rate Repeatability Reproducibility Cloud Point ASTM D5773 −60 °C to +49 °C 1.5 ± 0.1 °C/min 1.3 °C 2.5 °C Pour Point ASTM D5949 −57 °C to +51 °C 1.5 ± 0.1 °C/min 1.6 °C @ 1 °C 3.2 °C @ 1 °C

table 3: specifications of the automatic instrument for the determination of cloud point and pour point.

Parameter Units Range Inner pipe actual ID (oil side) mm (in) 7.75 (0.305) Inner pipe nominal OD (oil side) mm (in) 9.52 (3/8) Outer pipe nominal OD (coolant side) mm (in) 25.4 (1.0) Pipe length m (ft) 1.52 (5.0) Flow rate L/min 3.8 Linear flow velocity m/s 1.4

Deposition time h 4 (Model Oil 1) 16 (Model Oil 2)

Reynold’s number (oil side) -- 5,000 Reynold’s number (coolant side) -- 16,000 Wall shear stress Pa 7.5


table 5: cloud point (CP), pour point (PP), wax appearance temperature (WAT) and no-flow point (NFP) of model oils dosed with paraffin inhibitors.

Starting Material Automatic Instrument ASTM D5773/ASTM D5949 Rheometer Base Fuel Additive Actives

(ppm) CP (°C) ΔCP†

(°C) PP

(°C) ΔPP† (°C)

WAT (°C)

NFP (°C)

Model Oil 1 (High

wax)

Blank 0 33.20 0 29.9 0 33.70 28.40

Additive A

50 32.97 0.23 17.00 12.90 NA NA 100 32.79 0.41 8.00 21.90 NA NA 200 33.09 0.11 -4.00 33.90 NA NA 400 31.66 1.54 -10.5 40.4 30.75 < 5

Additive B

50 32.10 1.1 9.0 20.9 32.26 < 5 100 31.60 1.6 -3.0 32.9 22.25* < 5 200 31.00 2.2 -11.0 40.9 31.76 < 5 400 30.40 2.8 0.0 29.9 31.00 < 5

Additive C

50 32.50 0.7 -3.0 32.9 32.50 < 5 100 31.90 1.3 -5.0 34.9 32.50 < 5 200 31.50 1.7 -5.0 34.9 32.26 < 5 400 31.29 1.91 -2.0 31.9 26.30* < 5

Additive D

50 32.61 0.59 10.0 19.9 32.76 < 5 100 32.10 1.1 -3.0 32.9 32.76 < 5 200 31.60 1.6 -3.0 32.9 32.50 < 5 400 30.98 2.22 0.0 29.9 31.76 < 5

Additive E

50 32.90 0.3 23.0 6.9 33.26 < 5 100 32.50 0.7 4.0 25.9 33.26 < 5 200 32.20 1 -2.0 31.9 32.76 < 5 400 31.50 1.7 -7.0 36.9 32.24 < 5

Additive F

50 31.70 1.5 1.0 28.9 32.25 < 5 100 31.30 1.9 -2.0 31.9 31.75 < 5 200 30.77 2.43 -1.0 30.9 31.25 < 5 400 30.00 3.2 -4.0 33.9 31.01 < 5

Model Oil 2 (Low wax)

Blank 0 15.0 0 11.0 0 NA NA Additive A 400 12.3 2.7 -5.0 16 11.51 < 5

Additive C 400 13.1 1.9 -34.0 45 NA NA

† Δ CP and Δ PP calculated as the difference between the CP/PP of the untreated oil (blank) and that of the treated oil. * WAT outliers because of WAT defined by rheometry experiment for which the criteria of the WAT is when the viscosity increases by

more than 0.1 Pa·s. See Figure 7.


table 6: results from wax deposition experiments in the cold finger Starting Material Experiment Conditions Cold Finger Result

Base Fuel

Additive

Dosage (ppm)

Coolant Temperature

(°C)

Oil Temperature

(°C)

Deposit Weight

(g)

Deposit Reduction

(%)

Wax in deposit from HTGC (%)

Wax Mass in Deposit (g)

Model Oil 1 (High wax)

Blank 0

25

35

5.22 NA 8.8 0.46

Additive B

50 1.87 64.2 17.6 0.33 100 1.31 74.9 24.5 0.32 200 0.77 85.3 43.4 0.33 400 0.61 88.3 45.0 0.28

400* 0.51 90.3 43.4 0.22

Additive C

50 2.90 44.5 9.01 2.63 100 1.56 70.2 14.83 3.58 200 1.06 79.6 18.54 4.75 400 0.32 94.0 17.21 15.52

Additive D

50 3.63 30.5 8.66 3.01 100 2.52 51.7 9.72 3.15 200 1.55 70.4 15.98 4.24

400 0.74 85.9 21.07 5.43

Additive E

50 3.07 41.1 8.51 3.09 100 2.78 46.7 8.45 3.14 200 1.43 72.7 NA NA 400 0.62 88.2 NA NA

Blank 0 10

30

8.39 NA 7.7 0.64 Additive A 400 1.47 82.5 26.5 0.39 Additive C 400 2.01 76.1 22.7 0.45

Blank 0

10

30

1.13 NA 6.2 0.07 Model Oil 2 (Low wax)

Additive A 400 0.36 68.0 32.3 0.12 Blank 0

20 3.69 NA 5.5 0.20

Additive A 400 0.33 91.1 28.8 0.09

* Repeat experiment for Additive C at 400 ppm in Model Oil 1.


table 7: wax deposition experiments in the flow loop for Model Oil 1. deposition time: 4 hours.

Oil Temperature

(°C)

Coolant Temperature

(°C) Additive Dosage

(ppm) Deposit

Weight (g) Deposit

Reduction (%) Wax in deposit from HTGC (%)


35 25

Blank Additive C Additive B Additive B

0 50 50

200

19.5 13.2 13.2 12.7

0.0 32.2 34.6 41.6

16.0 51.9 44.3 64.1

3.1 6.9 5.6 7.3

30 Blank Additive B

0 50

15.9 8.6

0.0 45.7

16.4 69.2

2.6 6.0

40

25 Blank 0 8.0 0.0 39.9 3.2

30

Blank Additive C Additive B Additive B

0 50 50

200

3.3 3.0 3.4 2.8

0.0 10.5 -3.0 15.9

34.9 46.2 60.7 40.1

1.2 1.4 2.1 1.1

table 8: wax deposition experiments in the flow loop with model oil 2. deposition time: 16 hours.

Oil Temperature

(°C)

Coolant Temperature

(°C) Additive Dosage

(ppm) Deposit

Weight (g) Deposit

Reduction (%) Wax in deposit from HTGC (%)


20

5 Blank Additive A

0 400

16.98 7.47

-- 56.01

19.51 67.02

3.1 5.01

10 Blank

Additive C Additive A

0 400 400

11.41 6.34 3.31

-- 44.43 70.99

24.5 73.78 55.21

2.78 4.68 1.83

25

5 Blank Additive A

0 400

7.8 6.4

-- 17.95

53.54 66.31

4.18 4.24

10 Blank Additive A

0 400

6.65 3.42

-- 48.57

35.58 56.51

2.37 1.93

30 5

Blank Additive C Additive A

0 400 400

5.2 7.21 4.5

-- -38.65 13.46

56.19 70.8

56.81

2.92 5.1

2.56

10 Blank Additive A

0 400

5.81 1.38

-- 76.25

25.25 49.495

1.47 46.965


figure 1: Carbon number distribution of paraffin waxes in model oils. Compositional analysis (normal paraffins, unknown) performed by high-temperature gas chromatography (HTGC).

figure 2: Cold-finger experimental set-up. Cold-finger inner cylinder is attached to the head of the rheometer and the cup outer cylinder is attached to the rheometer motor so that a constant shear is applied during the experiment.


figure 3: Schematic diagram of the flow loop apparatus for wax deposition studies.


figure 4: Reduction in the cloud point of Model Oil 1 by wax inhibitors as a function of concentration of active additive for 50

ppm (■), 100 ppm (■), 200 ppm (■) and 400 ppm (■). Data from Table 5.

figure 5: Reduction in the pour point of Model Oil 1 by wax inhibitors as a function of concentration of active additive for 50 ppm (■), 100 ppm (■), 200 ppm (■) and 400 ppm (■). Data from Table 5.


figure 6: Reduction in the Pour Point (PP) as a function of the reduction in the Cloud Point (CP) for additives Additive A (♦),

Additive B (■), Additive C (●), Additive D (▲), Additive E (▼) and Additive F (◄) at different concentrations in Model Oil 1. Data from Table 5.

figure 7: Arrhenius plot of the viscosity versus temperature for Model Oil 1 with no additive (■) and with Additive C additive at concentrations of 50 ppm (●), 100 ppm (▲), 200 ppm (♦), 400 ppm (▼). * Example of WAT outlier for Additive C at 400 ppm in

Table 5 because of criteria for WAT from rheology experiment where viscosity must increase by more than 0.1 Pa·s.


figure 8: Reduction in the Wax Appearance Temperature (WAT) from rheometry experiments and the Cloud Point (CP) from the automatic instrument for Model Oil 1 with the addition of different concentrations of additives (Table 5): Additive A (♦), Additive B (■), Additive C (●), Additive D (▲), Additive E (▼), and Additive F (◄). *Outliers defined in Figure 7 and Table 5 because of criteria for WAT from rheometry experiment where viscosity must increase by more than 0.1 Pa·s.

figure 9: Percent reduction in cold finger deposit weight as a function of additive concentration: 50 ppm (■), 100 ppm (■), 200 ppm (■) and 400 ppm (■) for Additives A, B, C, D, and E in Model Oil 1 (Table 6). All experiments are performed with cold

Finger temperature = 25 °C and an oven temperature = 35 °C for all additives except for Additive A where the cold finger temperature = 10 °C and the oven temperature = 30 °C.


figure 10: Percent reduction in cold finger deposit versus pour point (PP) difference between the neat Model Oil 1 and the

Model Oil 1 with different concentrations of additives Additive A (♦) , Additive B (■), Additive C (●), Additive D (▲), Additive E (▼), (Table 6).

figure 11: A) Cold Finger experiment results for percent reduction in deposit versus cloud point (CP) difference between the neat Model Oil 1 and the Model Oil 1 with different concentrations of additives: Additive A (♦), Additive B (■), Additive C (●),

Additive D (▲) and Additive E (▼) (Table 6). B) Flow Loop Experimental results for deposit reduction at experiment conditions for oil temperature at 35 °C and coolant temperature at 25 °C (closed stars) and oil temperature at 40 °C and coolant

temperature at 30 °C (open stars). Colors coincide with different additives as in the cold finger experiments.


figure 12: Total deposit weight and composition from experiments in the flow loop with Model Oil 1. Concentration of additives

(active) was 50 ppm . Deposition time was 4 hours.

figure 13: Total deposit weight and composition from experiments in the flow loop with Model Oil 2. Concentration of additives (active) was 400 ppm. Deposition time: 16 hours.


figure 14: Deposit reduction % in the flow loop versus (A) cloud point reduction and (B) pour point reduction induced by additives in bench-top tests. Shape of symbols indicate model oil: Model Oil 1 (■) and Model Oil 2 ). Color indicates

additive: Blank ), Additive A ), Additive B ), and Additive C ).

OTC--25927--MS - SUEZ Water Technologies & Solutions Jul-16 ... The stress at which the sample...

Documents

Transcript of OTC--25927--MS - SUEZ Water Technologies & Solutions Jul-16 ... The stress at which the sample...