Fluid Phase Equilibria 476 (2018) 112e117
Contents lists available at ScienceDirect
Fluid Phase Equilibria
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / fl u i d
Effects of ethanol on the performance of kinetic hydrate inhibitors
quio Vinicius Ribeiro Castro a, Adriana Teixeira b,
Bruno dos Santos Renato a, Eusta
a
Rayza Rosa Tavares Rodrigues , Natalia dos Santos Renato c, *
ria, ES, Brazil
Department of Chemistry, Federal University of Espírito Santo, Vito
CENPES, PETROBRAS, Rio de Janeiro, RJ, Brazil
c
Department of Agricultural Engineering, Federal University of Viçosa, Viçosa, MG, Brazil
a
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 3 May 2018
Received in revised form
30 July 2018
Accepted 31 July 2018
Available online 3 August 2018
The main types of hydrate inhibitors are thermodynamic (THI), kinetics (KHI) and anti-agglomerants
(AA) inhibitors. Kinetic inhibitors are used in low dosages but do not have satisfactory efficiency in
large subcoolings. On the other hand, thermodynamic inhibitors are widely used in the industry and
require high dosages. In order to study a combined inhibition, the behavior of three different commercial
kinetic inhibitors combined with ethanol (THI) was tested under subcooling conditions in cases which
unique kinetic inhibition did not yield satisfactory results. With 3% concentrations of kinetic inhibitors
and a 10% by mass ratio of ethanol, it was possible to identify a significant increase in induction time for
two of the inhibitors tested. Once the growth of the hydrate crystals began to appear, the presence of
inhibitors did not result in significant gain in delaying the crystallization process. Some physical characteristics of the formed hydrates were altered in the presence of kinetic inhibitors and ethanol,
evidencing a lower capacity of the crystals to adhere. With the same concentration of kinetic inhibitor
and a 30% by mass ratio of ethanol the KHI presented low effectiveness.
© 2018 Elsevier B.V. All rights reserved.
Keywords:
Gas hydrates
Kinetic hydrate inhibitors
Ethanol
Cristal growth inhibition
1. Introduction
Thermodynamic inhibitors such as methanol, ethanol, or
monoethylene glycol (MEG) have been effective in preventing the
occurrence of hydrates in the oil industry for the past 75 years.
These compounds interact with water, altering the thermodynamic
equilibrium of hydrate formation, in which case the water is no
longer available to form the structures that give rise to the cavities
and subsequently the crystals of hydrate [1]. The development of
the deepwater production of oil has made the pressure and temperature conditions more severe and consequently increased the
required inhibitor concentration for values that exceed 60% by mass
[2].
Kinetic hydrate inhibitors (KHI) are part of a group of inhibitors
known as low dosage hydrate inhibitor (LDHI). They are polymers
normally synthesized from the monomers vinylpyrrolidone (VP),
vinylcaprolactam (VCap), esteramides or alkylacrylamides. These
structures can adhere to the structures of the hydrate nuclei in
formation and thus impair nuclei growth to prevent them reaching
* Corresponding author.
E-mail address: natalia.renato@ufv.br (N.S. Renato).
https://doi.org/10.1016/j.fluid.2018.07.036
0378-3812/© 2018 Elsevier B.V. All rights reserved.
the critical nuclear size [3e6]. Laboratory and field tests showed the
efficiency of kinetic inhibitors, but presented limitations under high
conditions of subcooling [7].
Using kinetic inhibitors together with thermodynamic inhibitor
can reduce the volume required for inhibition and make it cost less.
Yang and Tohidi [8] studied the synergy between PVCap and
various types of glycols ethers, all used in dosages lower than 1%.
Their studies found remarkable synergy between the compounds.
The presence of glycols increased the nucleation and growth times
of the hydrates. It was also concluded that the effect is associated
with the size of the glycol molecule, being more pronounced in
larger molecules.
A methodology was developed by Anderson et al. [9] to evaluate
the performance of kinetic inhibitors in order to increase the
repeatability of the results, and further they tested the inhibitor
performances and their synergy with different compounds [10,11].
The results show good synergism between MEG and PVCap for both
low and high concentrations. The use of ethanol or methanol with
PVCAP was only effective at low concentrations.
Among the possibilities raised for reducing the efficiency of
ethanol at high concentrations is the fact that it can contribute in
part to the formation of hydrate crystals, being the host molecule
trapped in the large cages of the hydrate structure [12].
B.S. Renato et al. / Fluid Phase Equilibria 476 (2018) 112e117
In this work, three commercial kinetic inhibitors were evaluated
in concentrations higher than those usually found in the literature
(3%). The kinetics of hydrate formation was evaluated in the
absence and presence of ethanol in order to understand the influence of this thermodynamic inhibitor on the performance of kinetic
inhibitors when subjected to high subcooling.
Table 1
Composition of the gas used in the experiments.
2. Experimental section
2.1. Materials and apparatus
Distilled water from the Flow Assurance Laboratory of the Petrobras Research Center (CENPES) and the ethanol from Isofar LTDA
(99.5% C2H6O; 0.5% H2O; by volume) were used. The mixing was
performed with the ethanol cooled to a temperature of about 5 C
to avoid evaporation of the sample. The masses were determined
using a scale with an accuracy of 0.01 mg. For this procedure, the
weighing of the mass of ethanol was carried out and followed by the
addition of water to guarantee the proportion of 10% of ethanol by
mass. In the samples containing the kinetic inhibitor, the inhibitor
was weighed and then added to the water and ethanol mixture at
the mass ratio of 3% of the inhibitor (as provided by supplier) and
97% of the water and ethanol mixture. Stirring was carried out with
a glass stick until complete homogenization of the mixture.
A cylindrical reactor, manufactured by Top Industries, was used.
It is made of stainless steel, have 100 mL of capacity, and can
operate in pressures of up to 700 bar and temperatures in the range
of 40 C to 50 C. The reactor has no viewing windows. The
temperature in the reactor is maintained by a bath whose fluid
(water þ glycol) circulates through the jacket in which the reactor is
immersed. A thermocouple directly connected to a data acquisition
system was used to measure the temperature inside the reactor
with a ±0.1 C error, with information every 10 s. A differential
pressure measuring instrument connected to the same data
acquisition system determined the pressure inside the reactor. The
stirring system of the reactor is composed of a magnetic stirrer of
adjustable rotation having a stem with four fins. A cylinder containing the gas mixture is connected to a compressor, which sends
the gas at the test pressure to the reactor. A schematic presentation
of the experimental system is shown in Fig. 1.
The gas used in the experiment was provided by the company
White Martins, and its composition was defined from the average
values found in the gas streams produced in Brazil. The composition of the gas is described in Table 1.
113
Component
% (Mol)
Methane
Ethane
Propane
n-Butane
i-Butane
n-Pentane
i-Pentane
n-Hexane
Nitrogen
Carbon dioxide
86.85
10.07
2.02
0.05
0.04
0.03
0.02
0.02
0.50
0.40
2.2. Method
The main parameters measured in this work were the induction
time and the crystal growth time. These are, respectively, the time
required until the hydrate formation process can be perceived, and
the time until the crystallization develops. To determine them, the
experimental procedure using high pressure reactors was:
- Prepare liquid mixtures containing water and different combinations of ethanol and kinetic inhibitors;
- Place the calculated quantities of liquids in the reactor, filling
50% of its volume;
- Stabilize the system temperature at the desired test
temperature;
- Pressurize up to the test pressure and isolate the reactor;
- Initialize the test with the start of shaking.
The induction time was determined as the time elapsed from
the beginning of shaking to the time when the system pressure
drops 3 bar. The value of 3 bar was chosen after verification that the
pressure drops between 1 and 2 bar due to gas solubilization. The
crystallization time was already defined as the moment when the
pressure reduces from 3 to 20 bar.
For the experiments, the temperature of 4 C was chosen
because it represents the temperature of deepwater. Therefore, it is
a common operating temperature of the pipelines of offshore
production fields in Brazil. The test pressure chosen was 100 and
200 bar.
In the literature, hydrate formation experiments usually use
rotations that vary between 400 and 800 revolutions per minute
[13e16]. In this work, the samples were shaken during all
Fig. 1. Scheme of experimental system.
114
B.S. Renato et al. / Fluid Phase Equilibria 476 (2018) 112e117
Table 2
Description of kinetic hydrate inhibitors (KHI) used.
KHI
Name
Supplier
Chemistry
Solvent
KHI-A
KHI-B
KHI-C
Hytreat 9873 K
Inhibex 501
Luvicap EG HM
Clariant
Ashland (ISP)
Basf
PVP/PVCap
PVCap
Butyl glycol ether
Ethylene glycol
experiments at 500 revolutions per minute. The electric current
used by the magnetic stirrer was measured. The value of the electric
current is proportional to the torque required to perform the stirring and varies as solids become part of the mixture and agglomerate. The analysis of this variable helps to explain the physical
characteristics of the formed solid.
The water at 100 bar of pressure and in the presence of the gas
used in these experiments forms hydrate at temperatures below
19.6 C. At a temperature of 4 C, the mixture will undergo a cooling
of 15.6 C. For a mass ratio of 10% ethanol, the temperature at which
the hydrate starts to form is 16.8 C. Therefore, the subcooling is
12.8 C in this scenario. For the pressure of 200 bar the temperature
at which the hydrate starts to form is 23 C, with a mass ratio of 30%
ethanol, this temperature decrease to 14 C. The equilibrium temperatures for hydrate formation were calculated using the CALSEP
software PVTSim NOVA version 1.1.
The three different commercial kinetic inhibitors that were
chosen for this work are known as KHI-A, KHI-B and KHI-C
(Table 2). Kinetic inhibitors are used at low concentrations, usually in the order of 1% by mass relative to water. A concentration of
3% was chosen, starting from the premise of being excess inhibitor.
Each liquid mixture was used only once, and the repetition of
the test condition was performed with a new sample. This action
was taken to avoid the influence of an earlier crystallization in the
Fig. 2. Pressure behavior during KHI-A tests at 100 bar.
following experiment. This phenomenon is known as memory effect and could reduce induction time if the same sample was used
[15]. Due to the stochastic nature of the crystallization of hydrates,
three tests were performed for each test condition.
3. Results and discussion
During the experiments the behaviors of three parameters were
monitored: pressure, rate of pressure drop, and electric current of
the agitator.
3.1. Pressure
Fig. 2 shows the pressure variation over time in the first tests
performed at 100 bar (initial pressure) and 4 C. As the reactor is
isolated, the pressure drop indicates that the process of hydrate
formation is occurring, since there is consumption of gas for the
formation of crystals. In tests made with water, it is possible to
notice that there are two distinct velocities of pressure drop: in the
beginning it falls slower and later the speed increases as the growth
of the hydrate occurs. The ethanol tests enabled the observation of
its kinetic character as a hydrate inhibitor, and the induction times
found, as well as the crystal growth times, were slightly higher
(Table 3). The pressure curve shows a behavior similar to that of
water but with lower slopes. One reason why this increase in crystal
induction and growth times occurs is the decrease in subcooling
from 15.6 to 12.8 C.
In the assays performed with the KHI-A inhibitor at 3% concentration and without the presence of ethanol at 10%, the induction
and crystal growth times showed little difference. It is possible to
clearly notice the presence of two very distinct slopes for the
pressure curves during the onset of the crystal growth process,
being slower than the tests without the inhibitor. For the tests
performed with the combination of KHI-A and ethanol, the induction time increased considerably while the crystal growth time was
close to that of pure water. The slope of the pressure curve remains
practically constant during the whole growing time of the crystals.
Assays with the KHI-B inhibitor resulted in crystal induction and
growth times close to those encountered with pure water, and their
combination with ethanol at 10% by mass resulted in crystal induction and growth times near the tests with only water and
ethanol, showing an absence of efficiency in these tested concentration and subcooling conditions (Fig. 3).
The KHI-C inhibitor assays showed an atypical behavior when
compared to the others, since the two different slopes of pressure
drop were separated by a period when the pressure remained
practically constant (Fig. 4). However, the combination of KHI-C and
Table 3
Induction and crystal growth times for assays with water, ethanol, Hytreat 9873 K (KHI-A), Inhibex 501 (KHI-B) and Luvicap EG HM (KHI-C).
Sample
Subcooling
Induction Time (h)
Crystal Growth (h)
( C)
Exp1
Exp2
Exp3
Exp1
Exp2
Exp3
Water
Water þ KHI-A
Water þ KHI-B
Water þ KHI-C
15.6
615.6
15.6
15.5
0.23
0.38
0.46
1.13
0.11
4.86
1.29
0.15
0.12
0.34
0.30
0.32
0.55
1.20
0.62
8.05
0.62
0.77
0.54
0.84
0.84
0.97
0.79
1.29
Water þ Ethanol 10%
Water þ Ethanol 10% þ KHI-A
Water þ Ethanol 10% þ KHI-B
Water þ Ethanol 10%þ KHI-C
12.8
12.8
12.8
12.8
2.28
9.43
1.25
>100
0.15
10.14
1.54
36.30
1.81
e
0.25
>100
0.90
0.62
0.55
1.44
1.36
0.66
1.05
1.48
1.23
e
0.56
e
Water þ KHI-C
Water þ Ethanol 30%
Water þ Ethanol 30%þKHI-C
10.0
10.0
10.0
1.34
0.08
0.3
23.76
0.09
1.84
>210
e
10.04
2.63
1.91
0.59
2.56
2.76
0.72
e
e
0.73
Note: Uncertainty of induction and crystals growth times: 0.01 h.
B.S. Renato et al. / Fluid Phase Equilibria 476 (2018) 112e117
115
Fig. 3. Pressure behavior during KHI-B tests at 100 bar.
Fig. 5. Pressure behavior during KHI-C tests at 200 bar.
ethanol showed to be efficient in altering the induction time even
when subjected to high subcooling. Once the formation of hydrates
was observed, the crystallization time was similar to the times
found in the other experiments.
With these experiments, it is possible to note that the presence
of ethanol at 10% by mass increases the induction time. This increase was very clear in the combinations of ethanol with both KHIA and KHI-C, since there was an increase in induction time. However, the combination of ethanol with KHI-B may be considered
irrelevant.
Although ethanol increases the crystallization time when
compared to water alone, the combination of kinetic inhibitors
with ethanol did not bring about significant changes in the crystallization times.
Due to good performance of the KHI-C, it was chosen to be
tested at 200 bar. The test without ethanol was made at 13 C and
the others two ones were made with 30% of ethanol by mass and at
4 C. Therefore, the subcooling was 10 C at the three scenarios. The
tests at 200 bar shows that at 10 C of subcooling the KHI can increase the induction time when it is used just with water (Fig. 5),
but when the same subcooling is achieved by displacing the equilibrium temperature with a high concentration of ethanol the KHI is
not effective.
3.2. Crystal growth rate
Fig. 6 shows the rate of pressure variation during the hydrate
growth process. Time is counted from the moment the pressure
drop begins. In the experiment carried out only with water, it is
possible to observe a hydrate formation at rates of pressure drop
close to 20 bar/h, followed by a period where the rate falls below
10 bar/h and then rises to reach a peak over 120 bar/h. The ethanol
in the proportion of 10% resulted in the peak crystallization velocity
occurring later and with less intensity, not exceeding 60 bar/h.
Fig. 4. Pressure behavior during KHI-C tests at 100 bar.
In the KHI-B experiments, although the crystallization initiated
with lower rate of depressurizing, peaks were observed earlier and
with values between 60 and 90 bar/h, both with and without
ethanol. KHI-A was the only inhibitor that enabled hydrate growth
at a rate higher than that achieved by pure water, but the onset of
crystallization occurred at rates below 10 bar/h. In their combination with 10% ethanol, the hydrate crystal growth occurred almost
uniformly at rates of about 30 bar/h over most of the crystallization.
The KHI-C inhibitor showed the lowest growth rates of hydrates
having peaks close to 20 bar/h. Fig. 5 shows that the process
occurred with two moments of hydrate growth divided by a
moment of stable pressure. In its combination with ethanol, a slow
onset of hydrate formation was observed as well as a peak of about
40 bar/h.
In experiments without the presence of kinetic inhibitor, the
high concentration of gas dissolved in water in the interface region
between water and gas explains the high initial rate of crystal
growth. After crystallization begins, its speed is reduced, however,
as the surface of hydrates increases, the speed tends to increase
until the mass of liquid water becomes the limiting factor for the
continued growth of the hydrates.
Hydrate inhibitors adhere to the surface of the first formed
hydrate crystals, disrupting the onset of crystal growth. After some
growth, the crystal size allows for the increase in the crystallization
speed, even in the presence of the inhibitors.
3.3. Physical appearance of formed crystals
After conducting the experiments and opening the reactor, a
visual check of its physical characteristics was performed. At the
time of the check, the hydrates were in an unstable region and their
dissociation had already begun. The hydrate formed only with
water, presented long and well-adhered crystals forming a solid
and rigid mass. In the hydrate formed with KHI-A and KHI-B, it was
possible to notice the existence of small crystals, which also joined
together to form a solid and unique mass but with less rigidity and
being more easily broken when touched. In the experiments carried
out with KHI-B or KHI-C in the presence of ethanol, the formation of
small crystals that do not adhere to each other or adhere with very
small force is observed. In this case, the hydrate was not removed
by the agitator when opening the reactor.
Fig. 7 shows a strong variation in current intensity in the tests
only with water showing a probable breakage of the forming
crystals by the agitator. After a period, the hydrate adheres to the
agitator and starts to rotate along with it. The time is counted from
the beginning of the formation of the hydrate crystals.
The KHI-B inhibitor, when combined with ethanol, significantly
reduced the current required to maintain the rotation of the stirrer
which can be explained by an anti-agglomeration effect. The KHI-A
116
B.S. Renato et al. / Fluid Phase Equilibria 476 (2018) 112e117
Fig. 6. Rate of pressure drop during nucleation.
Fig. 7. Electric current used to maintain rotation at 500 rpm.
and KHI-C inhibitors also showed a reduction in actuator torque in
both the pure dosed experiments and the experiments in which
they were combined with ethanol. In the experiments performed
without ethanol, the current signal noise was higher than in the
experiments with ethanol.
The formation of small crystals indicates the action mechanism
of PVCap impairing the process of crystal growth and causing
several crystal nuclei to form separately. This behavior was presented by Lee et al. [16].
4. Conclusions
Kinetic and thermodynamic inhibitors influence the nucleation
and growth process of hydrate crystals, and the combination of these
may allow the reduction of THI flow rates commonly used in petroleum production, being of economic and environmental concern.
The results of the experiments showed that ethanol has a kinetic
effect and a good synergy with KHIs when used at a concentration
of 10% by mass. The results showed a considerable increase in
B.S. Renato et al. / Fluid Phase Equilibria 476 (2018) 112e117
induction time for the combination of two of the inhibitors tested
with ethanol, KHI-A and KHI-C. The KHI-B presented the time of
induction almost equal to those found without the addition of kinetic inhibitors. The KHI-B with ethanol can help to avoid pipeline
blockage because it disrupts the agglomeration of the crystals, but
its use is not indicated for the predominant gas flow. Despite the
good synergy at concentration of 10% by mass of ethanol, the
synergy between ethanol and the KHI tested was impaired at high
concentration of ethanol (30%).
The samples were tested without the memory effect, which
approximates the reality tests, but decreases their repeatability
since the phenomena of nucleation and hydrate growth are stochastic. The behavior of the electric current and pressure curves
presented very similar results in all the repetitions performed, even
if the induction times showed high dispersion.
In order to perform field tests, a hydration control approach
based on risk management should be considered. Gas export systems with low volumes of water and efficient pressure monitoring
are best suited for field tests since hydrate growth can be perceived
without duct blocking and, therefore, without financial and security losses.
Acknowledgements
We would like to thank PETROBRAS for funding this work and
Michael J. Stablein of the University of Illinois Urbana-Champaign
for translating this paper.
References
[1] E.D. Sloan, Natural Gas Hydrates in Flow Assurance, Oxford, 2011.
[2] C.A. Koh, R.E. Westacott, W. Zhang, K. Hirachand, J.L. Creeck, A.K. Soper,
Mechanisms of gas hydrate formation and inhibition, Fluid Phase Equil. 194
(2002) 143e151.
[3] X. Zhao, Z. Qiu, W. Huang, Characterization of kinetics of hydrate formation in
117
the presence of kinetic hydrate inhibitors during deepwater drilling, J. Nat.
Gas Sci. Eng. 22 (2015) 270e278.
[4] H.J. Huo, R.H. Wang, H.J. Ni, Y.L. Liu, An experimental study on the synergetic
effects of kinetic and thermodynamic gas hydrate inhibitors, Petrol. Sci.
Technol. 32 (2014) 1940e1947.
[5] H.K. Abay, T.M. Svartaas, Effect of ultralow concentration of methanol on
methane hydrate formation, Energy Fuels 24 (2010) 752e757.
[6] A.A. Sulaimon, M.Z.M. Tajuddin, Application of COSMO-RS for pre-screening
ionic liquids as thermodynamic gas hydrate inhibitor, Fluid Phase Equil. 450
(2017) 194e199.
[7] C.W. Foo, L. Ruan, X. Lou, The inhibition performance in relation to the
adsorption of a polymeric kinetic inhibitor towards THF hydrates in the
presence of methanol, ethanol and monoethylene glycol, J. Nat. Gas Sci. Eng.
35 (2016) 1587e1593.
[8] J. Yang, B. Tohidi, Characterization of inhibition mechanisms of kinetic hydrate
inhibitors using ultrasonic test technique, Chem. Eng. Sci. 66 (2011) 278e283.
[9] R. Anderson, H. Mozaffar, B. Tohidi, Development of a crystal growth inhibition based method for the evaluation of kinetic hydrate inhibitors, in: Proceedings of the 7th International Conference on Gas Hydrates, Edinburgh,
2011.
[10] R. Anderson, H. Mozaffar, B. Tohidi, Application of a new crystal growth inhibition based KHI evaluation method to commercial formulation assessment,
in: Proceedings of the 7th International Conference on Gas Hydrates, Edinburgh, 2011.
[11] R. Anderson, H. Mozaffar, B. Tohidi, Effect of alcohols and diols on PVCapinduced hydrate crystal growth patterns in methane systems, Fluid Phase
Equil. 425 (2016) 1e8.
[12] K. Yasuda, S. Takeya, M. Sakashita, H. Yamawaki, R. Ohmura, Binary ethanolmethane clathrate hydrate formation in the system CH4-C2H5OH-H2O:
confirmation of structure II hydrate formation, J. Phys. Chem. C 113 (2009)
12598e12601.
[13] M.A. Kelland, History of the development of low dosage hydrate inhibitors,
Energy Fuels 20 (2006) 825e847.
[14] X. Zhao, Z. Qiu, W. Huang, Characterization of kinetics of hydrate formation in
the presence of kinetic hydrate inhibitors during deepwater drilling, J. Nat.
Gas Sci. Eng. 22 (2015) 270e278.
[15] E.R. May, R. Wu, M.A. Kelland, Z.M. Aman, K.A. Kozielski, P.G. Hartley,
N. Maeda, Quantitative kinetic inhibitor comparisons and memory effect
measurements from hydrate formation probability distributions, Chem. Eng.
Sci. 107 (2014) 1e12.
[16] S.Y. Lee, H.C. Kim, J.D. Lee, Morphology study of methaneepropane clathrate
hydrates on the bubble surface in the presence of SDS or PVCap, J. Cryst.
Growth 402 (2014) 249e259.