OTC-29886-MS

ScaleProTect – Scale Deposition Modeling in Pre-Salt Reservoir

Reda Bouamra, Giovanna Carneiro, and Patrick Machado, Schlumberger; Manoel Feliciano da Silva, Jr.,
Petrobras; Gabriela Franquiz, Hua Guan, and Thomas Lindvig, Schlumberger

Copyright 2019, Offshore Technology Conference

This paper was prepared for presentation at the Offshore Technology Conference Brasil held in Rio de Janeiro, Brazil, 29–31 October 2019.

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 complexity of carbonate reservoir in pre-salt plays dictates the use of advanced inflow devices to enable
optimized reservoir recovery. Intelligent Completions are considered one of the most effective systems
for flow control for a production interval. Depending on the actual design and geometry of the intelligent
completion tool, inorganic scale deposition can be very difficult to prevent and/or to mitigate. The failure
of such equipment tools can cause severe production impairment and affect the entire field economics.
In order to provide a high accuracy assessment of the scale deposition risk, a novel methodology
involving scale deposition experiments and model generation has been developed. The proposed workflow
combines 1D mechanistic methods to define scaling conditions, thermodynamic models to define the scaling
tendencies at these conditions and 3D models to predict particles accumulation in complex geometries using
numerical techniques describing the multiphysics interactions between solid particles and live fluids. The
basis of the models will be presented in this article along with numerical results of the precipitation and
deposition phenomena at downhole conditions.
This work represents the first holistic approach to characterize scale build-up in the vicinity of inflow
equipment walls. Numerical results are presented in this paper.

Introduction
With the advancement of drilling and production technologies, oil and gas have been produced from deeper
formations than ever before. High pressure and high temperature (HPHT e.g. 200°C, 20,000 psi), and
excessive Total Dissolved Solids (TDS) in excess of 300,000mg/L are often encountered at these depths,
particularly for subsea development fields. This brings significant challenges in scale control, especially for
calcium carbonate scaling which is a scale type that is most vulnerable to pressure drop [1]. The complexity
of such reservoirs and the comingling of waters from different streams can significantly prompt scale
formation, it also makes monitoring and performance control difficult. The delivery of scale inhibitors is a
challenge as well due to the limited umbilical and access to wells, and the possible degradation of chemicals
under such harsh conditions further complicates the effectiveness of scale prevention measures.
Due to the above-mentioned challenges, severe calcium carbonate scaling risk is expected for pre-salt
reservoir. For best production control and surveillance, smart completions are considered as the most
2 OTC-29886-MS

favorable method, since it allows shutting off or reducing water production or commingling of incompatible
water chemistries. However, depending on the actual design, size and geometry of the smart completion
tool, the scaling deposition can be very difficult to prevent and/or to mitigate. The failure of a functional
smart completion tool can cause severe production impairment and impact the entire field economics.
To provide a high accuracy assessment of the scale deposition risk, a methodology guiding scale
deposition experiments and model generation was developed and introduced here as ScaleProtect [2].
The methodology consists of evaluating the scaling severity associated with smart completion tools under
specific field environment. Computational Fluid Dynamic (CFD) is intensively used at all stages to develop
a unique workflow for scale characterization. Advanced multi-physics models are used to describe the
complex fluid configuration in the Inflow Control Valve (ICV) and the lab apparatus describe the different
turbulent regimes and particles trajectories required to understand and predict the deposition process.
This work describes in detail the ScaleProtect methodology to assess scale deposition in pre-salt wells
completions.

Prior Art – Scale Deposition Bibliographic Review

Inorganic scale management has been extensively studied recently and the petroleum literature exhibits a
number of interesting articles related to this topic ([3], [4], [5]). In the context of the Brazilian carbonate pre-
salt play, calcium carbonate scaling is the primary concern. The later CaCO3 forms at downhole following
the below chemical reactions:
(1)
(2)
Precipitation conditions are met downhole where formation water is oversaturated with bicarbonate
and calcium ions. Oversaturation is a factor of several variables, including temperature, pressure and pH.
Pressure reduction, for example, results in carbon dioxide leaving the brine phase as gas, increasing solution
pH and, with that, carbonate solubility is reduced leading to calcium carbonate precipitation (as per Eq. 2).
Several variables must be considered to predict if the precipitated scale will stay in the bulk solution or
will deposit at the surface walls. Parameters such as temperature, pressure, flow regime, surface roughness,
scale saturation ratios, residence time and others are all important in this prediction, turning it into a
non-trivial research topic. To better understand deposition and adhesion, numerical models and laboratory
experiments have been developed over the past years. The remainder of this document will focus on
reviewing the state of the art of scale prediction focusing on scale deposition.
According to Guha [6], the particle transport to the wall is the most important part of the deposition
phenomenon. The main mechanisms of particle transport to be taken in account at downhole environments
are, particle brownian motion, turbulent diffusion and turbophoresis. The latter is characterized by a
tendency for the particles to migrate in the direction of the wall due to the negative turbulent energy gradient
from the turbulent flow regime to the laminar layer. After the particle reaches the wall, a sticking probability
must be considered to investigate if the particle interaction with the wall will result in sticking of the particle
or not. Most of the deposition models consider that the sticking probability is one [7] while the particle
concentration is zero at the distance of half of the particle size from the wall [6] [8].
Usually, a parameter called deposition velocity, which is described as the volume deposition rate per
unit surface, is used to quantify the amount of scale deposited in experimental tests. The later defines the
deposition velocity as a function of the so-called relaxation time (τ).
Eskin et al. [7] stated that the deposition process is divided into three different regions as a function of
particle size: in the first scenario (small particles), flow regime is controlled by Brownian and turbulent
particle diffusion, as the particle size increases, the deposition velocity decreases since Brownian diffusion
has an inversely relationship with particle size. The second scenario presents a decrease in deposition rate,
OTC-29886-MS 3

which happens because turbulent diffusion starts to dominate upon the Brownian diffusion and deposition
rate increases around three to four orders of magnitude during the second regime. This fast increase in the
deposition rate is due to the so called turbophoresis effect. In the third scenario, the turbophoresis effect
loses efficiency and the deposition velocity is slightly reduced with the increase in particle size.
Most of recent research work focuses on modeling the particle transport to the wall as it is the main driver
in the overall process. One of the most suitable methods nowadays is the advection-diffusion [7]. In 1970,
Beal [9] suggested a calculation model for deposition velocity based on the "stopping distance" which turns
to be one of the few models that considers the sticking probability, as can be seen in the following equation.
(3)
Where p is the sticking probability, v’ is the mean particle fluctuation velocity, Cst is the volume particle
concentration within the stopping distance and Vdep is the deposition velocity.
More recently, Eskin et al. [7] studied the deposition process of a turbulent steady-state particulate flow in
a vertical pipe. Considering dilute flow, the authors used the advection-diffusion approach and formulated
the mass and momentum conservation equation in Cartesian coordinates.
From an experimental perspective, the literature reports different experiments with the aim of assessing
the parameters cited above. Vazirian et al. [10] used rotating cylinder electrodes with different types of
coating to investigate separately each phenomenon. A similar approach is used to determine experimentally
the influence of surface roughness and surface energy in the scale deposition process in Lokmane et al. [5].
A reliable thermodynamic and kinetic model for prediction of scale rate is required [11]. In order to
understand the relationship between the calcium carbonate precipitation and surface deposition, Sanni et al.
[11] developed an experimental process based on a flow rig system which is a non-recirculating tool. The
use of this tool allows the study of precipitation and deposition with a constant saturation ratio throughout
the experiment. The setup allowed the collection of images in real time to access the turbidity and deposition
levels during the experiment.
The relationship between induction time and saturation ratio in terms of the excess free energy (ΔG)
according to the size of a nucleus was studied by Sohnel and Mulling [10]. In addition to the established
relationship of induction time, Euvrard et al. [4] developed a method to quantify the extended surface area
based on the type of nucleation process during the deposition phenomena. The extended surface area is
the covered area plus the effect of nuclei overlapping which gives a more robust model that considers
superposition of different nuclei on the surface.
Considering Brazilian pre-salt scenarios, Guan et al. [2] have run experiments similar to the ones
described above, with the difference that a pre-selection of experimental setup was defined based on results
of Computational Fluid Dynamics (CFD) simulations. Hydrodanymic simulations were run to predict where
in the ICV scale was more likely to occur. Simulation results indicated that particle accumulation is more
prone to happen before the choked area where there is a high turbulence kinect energy.
Similarly, CFD simulations were done for coupons in reactors for the testing of different coating types
to choose velocity profiles to be applied in the reactor test. The study aimed to best reproduce turbulence
intensity that was observed on ICV simulations and better replicate well conditions. Simulations were
also done on treated and non-treated coupons, to predict the likelihood of scale adhesion as a function of
surface properties. Experimental results were highly correlated to CFD simulations both in ICV (in flow
loop tests) and coupon tests [2]. Experiments also indicated that coated coupons reduce or eliminate surface
accumulation and adhesion of scale particles. However, Guan et al. [2] still suggests further investigations
on coated surfaces.
4 OTC-29886-MS

ScaleProtect
The overall goal is to develop a novel methodology to test scale deposition and better predict and mitigate
the risk of losing the operability of downhole/surface equipment. The overall workflow is presented in the
below figure.

Figure 1—ScaleProtec overall methodology

The new methodology consists of combining advanced modeling techniques with lab testing in order to
design replicable lab equipment, to define a representative testing protocol and to assess scale deposition
on metal surfaces. The purpose of the procedure is to build a robust lab testing apparatus and representative
laboratory testing protocol and to develop the framework for a scale deposition model. The methodology
can be divided into 1. precipitation modelling, 2. downhole valve modelling and 3. mockup valve modelling.
This work describes the initial deployment of the methodology, consisting on building the scale
deposition model using advanced numerical methods and lab experiments. It consists of first optimizing
the brine used in the experiments. The brines showing a high scaling tendency are selected and numerically
tested for scale precipitation using inorganic solids modeling software. The aim of this first step is to define
the worst-case conditions with respect to scaling and perform scale precipitation experiments under these
conditions.
Once the chemistery is defined, the lab scale valve has to be designed. This is done based on CFD models
for different Mockup Equipment (ME) configurations and assessing the scale deposition risk based on the
solid deposition theory. The testing protocol is then updated with the most representative ME design and
flow test conditions. The MEs are then manufactured and tested for scale deposition in the lab. A scale
deposition model is then derived in order to predict and monitor the scale deposition risk along the life of
the field. This is achieved using the lab deposition results, deposition theory and advanced CFD models.
OTC-29886-MS 5

Precipitation modeling
To predict the amount of precipitated scale in the completion, state of the art of coupled thermal-hydraulic
and thermodynamic models are used as Hua et al. [2]. Fluid property and scaling properties are generated
and passed to the thermal-hydraulic model to generate the scaling conditions and assess scale precipitation.
The overall workflow is presented below.

Figure 2—Methodology to asses scale precipitation

Scale formation tendency is represented by the thermodynamic-based saturation index, SI, and is defined
as the relation between ion activity product and thermodynamic solubility product. Its expression is given
in the following equation for calcite:

(4)

If the SI is negative, there is no potential to scale, and ions Ca2+ and CO3- will not be combined into
CaCO3. If the SI is close to zero, the brine is at a borderline scale potential. Precipitation may occur if SI is
positive. In the case of a SI>3, scaling risk is moderate and it turns to severe scaling risk in case of a SI>5.
A typical pre-salt field acrhetiture, e.g. single well tie-back, is modeled in the thermal-hydraulic software
and field production conditions are used. The resulting pre-scaling tendency of all the solids in the system
are shown as a function of the different locations in Figure 3.
6 OTC-29886-MS

Figure 3—Pre-scaling tendency of all the solids in the system

The amount of CaCO3 at bottomhole is around 0.106 (kg/m3). CaCO3 amount at wellhead and topside
is respectively 0.1526 (kg/m3) and 0.178 (kg/m3). CaCO3 (calcite) scale amount in each location of the
system is presented in Figure 4.

Figure 4—CaCO3 scale amount in the system

The pre-scaling tendency of CaCO3 in all the range of temperatures and pressures is presented in Figure 5.
OTC-29886-MS 7

Figure 5—Pre-scaling tendency of CaCO3 (calcite)

The SI for calcite at HPLT (P > 80 bar and T < 78 °C) and LPHT (P < 6.5 bar and T > 114 °C) is close
to zero. This means borderline scale potential (water quality or changes in temperature could change the
index). The yellow zone means that scale can form and CaCO3 precipitation may occur. In the orange and
red zones CaCO3 scale precipitates instantaneously.
To reproduce the same conditions at a laboratory scale, a synthetic brine was designed using the same
methodology. The composition is given in the table below.

Table 1—Initial Synthetic Brine Composition

Volume: 1000 litters

Product Quantity (kg)

Sodium bicarbonate (NaHCO3) 2

Calcium chloride
31
dihydrate (CaCl2.2H2O)

Sodium chloride (NaCl) 182

Considering that the objective is to evaluate the scaling risk of the provided brine at downhole reservoir
condition, it is important to have representative calcium carbonate scaling tendencies in the lab similar to
the field condition. The calcium carbonate scaling tendency of the above brine composition was calculated
at the reservoir condition (136 °C and 472 bar) which gives a scaling index of 2.31.
In the laboratory, the testing is carried out by pumping two brines (one contains calcium ions, the other
contains bicarbonate ions) individually at the same flow rate while mixing them at constant temperature of
60 °C and atmosphere pressure. Therefore, the above brine is to be split into two brines, which contains the
representative amounts of calcium and bicarbonate respectively. When the two brines mix at lab condition
(60 °C and 1 bar), the calcium carbonate scaling tendency should be similar to what it is for the produced
brine at reservoir condition. A number of simulations using different calcium and bicarbonate concentrations
were carried out in order to derive a similar calcium carbonate scaling tendency in the laboratory.
To have similar calcium carbonate scaling tendency in the lab, the concentration of bicarbonate has to
be increased from 2 to 3 g/l in the bicarbonate brine which gives a similar scaling index of 2.24 in the lab
in comparison to the reservoir condition.
As the two brines will be mixed in the lab, ion concentrations of each split brine have to be doubled.
Formulations of the two brines to be made in the lab are the presented in the table below.
8 OTC-29886-MS

Table 2—Laboratory Brines Formulation

NaCl [g/L] CaCl2.2H2O [g/L] NaHCO3 [g/L]

Brine 1 182 62 -

Brine 2 182 - 6

It is worth to note, however, that this test brine provided has greater scaling tendency than that of the
formation water. The scaling index for formation water at downhole reservoir condition is around 1.15 while
the scaling index for the synthetic brine is 2.31. It is common to use a conservative brine composition in the
laboratory which has increased scaling tendencies in order to obtain the required scale amount within the lab.

Downhole Valve Modeling

The aim of the 3D model is defining, with a high accuracy, the hydrodynamic fields in the downhole ICV
and design a mock-up valve that will mimic these fields to produce the same deposition patterns. The
figure below presents the overall workflow to reproduce the most accurate deposition behavior at laboratory
conditions.

Figure 6—3D Modeling Workflow

The first step consists on defining the flowing parameters at downhole conditions, this is performed by
modeling the full-scale valve and inserting particles to track their trajectories in the geometry. The latter
will be related to their deposition after the acquisition of experimental data. The second step consists of
designing a mockup valve that reproduces the same flow and particulate flow behavior based on the design
criteria listed on the figure below.
OTC-29886-MS 9

Figure 7—Mockup Valve Design Criteria

Velocity and turbulence are the main parameters to be matched in the design. Near-wall behavior and
particles dynamics will follow.
The valve geometry was uploaded to the 3D modeling software and meshed to perform CFD simulations.
The model uses a high resolution polyhedrical mesh with a minimum of ten cells on the main hydraulic
diameter and ten prism layers to capture the flow laminar sub-layer. The main areas of interest, e.g. chokes
and sleeve, have been refined with ten times the definition of the main mesh size. Figure 8Error! Reference
source not found. below shows the main mesh cells used in the model.

Figure 8—Downhole ICV Geometry and Mesh

The physics used in the model represents a single-phase water flow at downhole conditions. Mass
particles were introduced in a second model to assess particle trajectories. The downhole ICV simulations
were run under steady conditions. The flow is modeled only through the annulus and the contribution from
the main tubing is not considered (modeling a bottom zone valve). The figure below shows the resulting
velocity profile on a vertical cross-section of the geometry. Velocity profile and Re number at several
locations are presented on Figure 9.
10 OTC-29886-MS

Figure 9—Downhole ICV Velocity Profile

The resulting Re number values show a turbulent flow regime in the main areas of interest. In the vicinity
of the flow inserts, a high velocity gradient is observed in the two main insert and a large recirculation zone
in vicinity of the sliding sleeve. This zone has been identified as prone to scale deposition in the previous
phases of the project [13]. The rest of the inflow points also exhibits a significant velocity gradient with
low velocities around the inflow points. These areas also were identified as high deposition risk zone. For
these reasons, the same flow regime will be reproduced in the mockup valve to create an analogue flow
at laboratory conditions.
A Lagrangian phase is introduced to model the particles behavior in the fluid. Particles are modeled as
solids with the same density as carbonate particles and with size of 50μm. Fluid and solid particles are
coupled using a two-way coupling method (flow impacting the particles and the particles influencing the
flow). The particles concentration calculated in the previous section are used at the inlet of the annulus space.
The deposition theory described in [3] is used to model the particle interaction in the different flow regions.
The approach is however simplified here as experimental inputs are required to enhance the resolution of
the model. In this work, the two-way coupling is represented by a fully coupled momentum between fluid
particles. And absorption model is set at the walls and the sticking probability is set to 1. Figure 10 shows
the particles distribution in the valve and their corresponding Re numbers.
OTC-29886-MS 11

Figure 10—Particles Reynolds Number

It is noticed that the particles are concentrating in the vicinity of the insert in the near wall region. This
behavior is expected as these zones exhibit a high turbulent energy gradient inducing a local turbopherisis
and leading to potential solids accumulation. This charteristics are used to define the mockup valve
geometry.

Figure 11—Turbulent Energy and Particles Distribution

Mockup Valve Modeling

As described in the previous section, the flow velocity in the near-wall region downstream the insert is very
low leading to an accumulation zone in the vicinity of the valve sliding sleeve. This area has been identified
12 OTC-29886-MS

in the previous work [13] as a high risk accumulation zone and the fluid patterns need to be reproduced
with highest accuracy in this zone.
For the above reasons it is proposed to keep the ICV geometry and not to scale down the trim (Figure
12). However, the below modifications are made:

• Based on the velocity profile in the downhole ICV (Figure 9) the flow is fairly developed upstream
and downstream the valve trim. Therefore, it has been decided to cut the valve geometry and keep
1 m upstream and downstream the valve.
• In order to ensure a homogeneous brine mixture upstream the mixing point, a pipe with two inlet
points has been added to the geometry. Two different brines will be injected into each of the inlets
and mix in the main pipe bore. The flow will then enter the annulus space of the mockup ICV.
• Another pipe downstream the valve is also added to discharge the brine at atmospheric pressure.

The mockup is designed to be a flow through system to avoid any scale depletion as in a looped system.
Detailed drawings are presented on Figure 12.

Figure 12—Mockup Valve Geometry

Following the same methodology, the mockup valve geometry was meshed and fluid flow was analyzed
under steady state conditions.
One can observe from Figure 13 that the flow develops uniformly from the pipe inlets into the ICV.
Brines are also mixing uniformly along the inlet pipe with a mixing ratio of 50% at the annulus inlet and
along the mockup valve.

Figure 13—Velocity Streamlines - Mockup Valve

OTC-29886-MS 13

Figure 14—Brine Mixing in Mockup Valve

The overall fluid flow behaviour is similar to what is observed in the downhole ICV. Velocities (Figure),
turbulent energy and Re numbers are similar (Table 3).

Table 3—Re numbers comparison in downhole ICV and Mockup Valve

Location Characteristic Length [m] Average Re Average Re – Mockup

Small Insert 0.004 3.10e+04 1.31e+04

Large Insert 0.031462 3.47e+05 1.46e+05

Tubing 0.1 4.00e+05 1.81 e+05

Annulus 0.02 9.97e+04 3.111e+04

Particles dynamics is also similar and accumulations occur in the same locations as in the downhole ICV.
Based on these results, the proposed mockup valve design fulfills the requirements for lab testing.

Figure 15—Mockup Valve Fluid Flow Modeling

14 OTC-29886-MS

Figure 16—Particles Dynamic in ICV vs. Mockup Valve

Conclusion
The ScaleProtec methodology and workflow has been succesfully deployed in this work to study the scale
deposition process in downhole completions.
This work started by stating the precipitation and potential deposition risk at downhole conditions. Firstly,
the coupled thermodynamic and thermal hydraulic modeling approach allowed the determination of the
precipitation conditions and assessed the severity of scaling of the formation water. A synthetic brine was
then defined based on those results and tested in a 3D numerical model. The latter showed a scale deposition
risk under downhole conditions.
The second phase of the work consisted of defining the best lab apparatus able to reproduce downhole
scale deposition behavior. To this end a mockup valve was designed and tested with paramaters defined in
the first part of the work. The selected geometry showed a very good correlation with the downhole valve
and a similar scale particle behavior.
CFD approach was intensively used during the course of lab apparatus design where velocity and
turbulence as the two main paratemers were successfully matched in the mockup valve in comparison to
the downhole ICV.
The foundation for a deposition model has been set in this numerical work, laboratory work will be
carried out using the verified lab apparatus in the near future to validate the approach and to further define
the complexity of the model. With that it is expected to increase the reliability of the model.

References
1. O. J. Vetter, "Oilfield Scale--Can We Handle It?," SPE, vol. 2118, no. 5879-PA, 1976.
2. H. Guan, R. Bouamra, T. Lindvig and J.-C. Vernus, "Scale Risk Assessment and Novel Coating
for Smart Completion: Scale Simulation and CFD Modelling Approach," SPE, Vols. 114103-MS,
2018.
OTC-29886-MS 15

3. O. Sanni, T. Charpentier, N. Kapur and A. Neville, "CaCO3 Scale Surface Nucleation and
Growth Kinetics in a Once-Through in situ Flow Rig as a Function of Saturation Ratio (SR),"
NACE International, 2016.
4. M. Euvrard, F. Membrey, C. Filiatre, C. Pignolet and A. Foissy, "Kinetic study of the
electrocrystallization of calcium carbonate on metallic substrates," Journal of Crystal Growth,
vol. 291, no. 2, pp. 428–435, 2006.
5. S. B. Lokmane, T. V. J. Charpentier, A. Neville, C. Hurtevent, J. R. O. Varela, F. M. Nielsen, V.
Eroini, J. H. Olsen, J. A. Ellingsen and O. Bache, "Comparison of characteristic of anti-scaling
coating for subsurface safety valve for use in oil and gas industry," International Petroleum
Technology Conference, 2014.
6. A. Guha, "Transport and Deposition of Particles in Turbulent and Laminar Flow," Annual
Review of Fluid Mechanics, vol. 40, no. 1, 2008.
7. D. Eskin, J. Ratulowski and K. Akbarzadeh, "Modeling of particle deposition in a vertical
turbulent pipe flow at a reduced probability of particle sticking to the wall," Chemical
Engineering Science, vol. 66, pp. 4561–4572, 2011.
8. S. K. Friedlander and H. F. Johnstone, "Deposition of Suspended Particles from Turbulent Gas
Streams," Industrial & Engineering Chemistry, vol. 49, no. 7, pp. 1151–1156, 1957.
9. S. K. Beal, "Deposition of Particles in Turbulent Flow on Channel or Pipe Walls," Nuclear
science and engineering: the journal of the American Nuclear Society, vol. 40, no. 1, pp. 1–11,
1970.
10. M. M. Vazirian, T. V. J. Charpentier, M. O. Penna and A. Neville, "Surface inorganic scale
formation in oil and gas industry: As adhesion and deposition processes," Journal of Petroleum
Science and Engineering, vol. 137, pp. 22–32, 2016.
11. O. Sanni, T. Charpentier, N. Kapur and A. Neville, "Study of Surface Deposition and Bulk
Scaling Kinetics in Oilfield Conditions Using an In-Situ Flow Rig," NACE international, 2015.
12. O. Sohnel and J. W. Mullin, "Interpretation of crystallization induction periods," Journal of
Colloid and Interface Science, vol. 123, no. 1, pp. 43–50, 1988.
13. R. Bouamra, "Mock Up Valve Design –CFD Modeling Results – 20100162-REP-003 A,"
Schlumberger Flow Assurance Consulting, 2012.
14. M. Scheck and G. Ross, "Improvement of Scale Management using Analytical and Statistical
Tools.," SPE, Vols. 114103-MS, 2008.
15. B. Y. H. Liu and J. K. Agarwal, "Experimental observation of aerosol deposition in turbulent
flow," Journal of Aerosol Science, vol. 5, no. 2, pp. 145–148, 1974.