- Email: [email protected]
Effect of operating parameters and corrosion inhibitors on foaming behavior of aqueous methyldiethanolamine solutions
Accepted Manuscript Effect of operating parameters and corrosion inhibitors on foaming behavior of aqueous methyldiethanolamine solutions Mohammed Keewan, Fawzi Banat, Emad Alhseinat, Jerina Zain, Priyabrata Pal PII:
S0920-4105(18)30152-9
DOI:
10.1016/j.petrol.2018.02.046
Reference:
PETROL 4716
To appear in:
Journal of Petroleum Science and Engineering
Received Date: 8 March 2017 Revised Date:
12 February 2018
Accepted Date: 19 February 2018
Please cite this article as: Keewan, M., Banat, F., Alhseinat, E., Zain, J., Pal, P., Effect of operating parameters and corrosion inhibitors on foaming behavior of aqueous methyldiethanolamine solutions, Journal of Petroleum Science and Engineering (2018), doi: 10.1016/j.petrol.2018.02.046. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
3
Mohammed Keewan, Fawzi Banat*, Emad Alhseinat, Jerina Zain, Priyabrata Pal
4 5 6
Department of Chemical Engineering, Khalifa University of Science and Technology, SAN Campus, P.O. Box 2533, Abu Dhabi, United Arab Emirates *Email: [email protected]
RI PT
2
Effect of operating parameters and corrosion inhibitors on foaming behavior of aqueous methyldiethanolamine solutions
1
7
Abstract
9
Corrosion inhibitors are one of the main causes of amine foaming in gas sweetening units. A
10
detailed understanding of amine foaming behavior in the presence of corrosion inhibitors is of
11
great importance if the foaming is to be minimized. In this work, a comparative study was
12
carried out to investigate the foaming tendency and bubbles characteristics of aqueous
13
methyldiethanolamine
14
Hydroxyehyl)cocoalkylamine; BHCL) and hydrocarbon-based (HCB) corrosion inhibitors. The
15
effect of different operating parameters such as nitrogen flow rate, corrosion inhibitor
16
concentration, foaming time, solution temperature and pore size of the gas diffuser were studied
17
using foam scan instrument. The results showed that on increasing the foaming time, solution
18
temperature and corrosion inhibitor concentrations, the foaming tendency increased. For BHCL,
19
the foam height tends to decrease with a high flow rate and a small pore size of the gas diffuser.
20
Nevertheless, the opposite trend was observed in the presence of the HCB corrosion inhibitor.
21
Depending upon the type of corrosion inhibitors, careful optimization of the operating conditions
22
showed a high potential to minimize amine foaming.
24 25
M AN U
in
the
presence
of
fatty
acid-based
EP
TE D
(MDEA)
AC C
23
SC
8
26 27
Keywords: Foaming, corrosion inhibitors, gas diffuser, amine, gas sweetening
28
1
(Bis(2-
ACCEPTED MANUSCRIPT
1
1. Introduction Acid gases like CO2 and H2S are found in considerable quantity in crude natural gas, which
3
creates obstacles for the combustion and transportation processes, by reducing its combustion
4
value, fouling and corroding the pipelines. Thus, the presence of acid gases is a major source of
5
maintenance problems in the facilities. This further affects the financial outcome of the process
6
and hence must be reduced. Natural gas sweetening with methyldiethanolamine (MDEA) is the
7
commonly employed method to remove the common acid gases like H2S and CO2.
RI PT
2
Foaming is one of the major operational problems in the amine absorption process leading to
9
serious consequences like loss of absorption capacity, reduced mass transfer area and efficiency.
10
Foaming can be triggered by various contaminants like corrosion inhibitors used for pipelines
11
and condensed liquid hydrocarbons. Suspended fine solid particles like iron sulfide (FeS) also
12
increase the foam stability and contaminants from boiler feed water treating chemicals, amine
13
degradation products and excessive antifoam agent also cause foaming (Kohl and Nielsen, 1997;
14
Al-Dhafeeri, 2007; Alhseinat et al., 2015b; Abry and Dupart, 1995). There are very few studies
15
considering a systematic and quantitative investigation of foaming in amine solutions (Pauley,
16
1991; Chen et al., 2011).
M AN U
SC
8
In gas sweetening units, MDEA in the concentration range of 40-50 wt. % is usually used to
18
remove CO2 and H2S which may be present at up to 30% in the crude natural gases (Alhseinat et
19
al., 2014). Foaming can cause carry over of alkanolamine solutions. Thus, subsequent loss of
20
solvents reduce plant throughput and off-specification of produced gas.
TE D
17
Different contaminants like thermal degraded products, heat stable salts, total suspended
22
solids and dissolved oxygen in alkanolamine solutions are the main causes of corrosion in the
23
gas sweetening plant (Davoudi et al., 2014). Corrosion has been observed in different locations
24
of the gas sweetening plants and can be classified as (a) corrosion of carbon steel with wet acid
25
gas; by the reaction between iron with carbon dioxide and hydrogen sulfide to produce a thin
26
liquid film, which occurs mainly at the bottom of the absorber, condenser, and top of the
27
regenerator and (b) corrosion of carbon steel in aqueous alkanolamine solutions, which mostly
28
occurs at the bottom of the regenerator, reboiler and lean/rich heat exchanger (Nielsen et al.,
29
1995; Davoudi et al., 2014). Correct and effective corrosion monitoring procedures are usually
30
followed in plants as a proactive tool (Davoudi et al., 2014). Further, corrosion inhibitors are
31
usually used for preventing corrosion by solvents used in treating sour gas streams (Jones and
AC C
EP
21
2
ACCEPTED MANUSCRIPT
Alkire, 1985). Corrosion inhibitors are compounds such as alkyl or aryl amines (Garcia-Arriaga
2
et al., 2010; Ashassi-Sorkhabi and Nabavi-Amri, 2000). They are used to protect the metallic
3
surfaces from corrosion and several types of corrosion inhibitors are currently used
4
in amine systems for natural gas refining. It is believed that corrosion inhibitors play one of the
5
major roles in foaming in the gas plant. Liquid hydrocarbons, such as HCB, which is a
6
thoroughly investigated corrosion inhibitor, are used in gas sweetening plants due to their higher
7
solubility in amine solution (Pauley 1991). Even BHCL, which is a polar and soluble in MDEA,
8
is also used in alkanolamine gas sweetening plants. The compound contains oxygen and nitrogen
9
atoms which is adsorbed on the metal surfaces to block the active corrosion sites (El-Haddad,
10
2013). The BHCL builds up a protective hydrophobic layer on the metal surfaces, which
11
provides a barrier to the dissolution of the metal ions into the MDEA solutions (Yaro et al.,
12
2013).
M AN U
SC
RI PT
1
There is very little literature available that explains the effect of corrosion inhibitors on amine
14
foaming (Gilyazetdinov and Matishev, 1990; Pauley, 1991; Al-Dhafeeri, 2007). The main
15
objective of this study is to provide more insights into the effect of corrosion inhibitors on the
16
foaming of amine solutions. The effect of two different corrosion inhibitors, HCB and BHCL, on
17
the foaming of MDEA solvent, will be investigated. Foaming tendency and stability of amine
18
solution will be studied by varying liquid level, time of foaming, nitrogen flow rate, corrosion
19
inhibitor concentration, solution temperature and pore size of the gas diffuser. The bubble and
20
foam volume will also be characterized. Additionally, the present study aims to further enhance
21
the current knowledge regarding amine foaming through real-time investigation, utilizing a
22
world-class foam analyzer, and this should help in reducing the propensity of foaming due to the
23
addition of corrosion inhibitors.
25
EP
AC C
24
TE D
13
1.1. Foam Theory
26
Liquid foams consist of gas bubbles which are closely packed within a liquid carrier matrix,
27
by mechanical incorporation of a gas into a liquid generating bubble (Drenckhan and Saint-
28
Jalmes, 2015). Two main parameters used to study the foaming problem are foaming tendency
29
and foaming stability (Pauley and Hashemi, 1998, Alhseinat et al., 2014). The foaming tendency
30
is the ease with which a solution forms a bubble and produce foam. The foam stability is the
3
ACCEPTED MANUSCRIPT
1
resistance to break the foam into a liquid phase, measured on the basis of time to coalesce the
2
bubble. Once foams are formed, they may undergo a thinning process, which decides the foam
4
stability. Usually, foams are subjected to three main instabilities, such as liquid drainage, foam
5
coalescence and foam rupture.
RI PT
3
6 7
1.2. Foam Structure
Foams encompass an ensemble of different size bubbles (Pilon et al., 2002). The injection
9
system decides the bubble size distribution at the bottom of the foam layer (Narsimhan and
10
Ruckenstein, 1986). Bubbles can have different shapes (spherical and polyhedral), depending on
11
the gas and liquid fraction. Spherical foam is commonly known as Kugelschaum, where the
12
thickness of the liquid film is approximately equal to the diameter of the gas bubble, which turns
13
to polyhedral, usually called Polyederschaum, when the thickness of the liquid film decreases
14
due to drainage. Polyhedral bubbles are usually located at the top of the foam layer and are
15
always subjected to coalescence and rupture. Meanwhile, the spherical bubbles are located at the
16
bottom (Maiysa, 1992). Foam, once formed, always undergoes a thinning process caused by
17
drainage, coalescence, and rupture, according to Plateau’s laws (Bhakta and Ruckenstein, 1997);
18
when three bubbles link up, they create a plateau border (PB), concaving three liquid films at an
19
angle of 120o, further reduced to 109o in the case of four bubbles adjoining, thereby creating a
20
honeycomb structure, enhancing the liquid drainage through the interconnected PB structure.
M AN U
TE D
EP
21
SC
8
1.3. Critical Micelle Concentration (CMC)
23
Corrosion inhibitors are surface-active compounds (Finšgar and Jackson 2014) and can be oil or
24
water-soluble surfactants (Hart, 2016). Surfactants are chemical compounds that lower the
25
surface tension of the liquid, and the interfacial tension between either two liquids or a liquid and
26
solid, and can act as foaming agents. Surfactants are amphiphilic molecules that contain an oil-
27
soluble hydrophobic hydrocarbon tail and a water-soluble hydrophilic head, where they
28
preferentially form micelles with a hydrophobic core and a hydrophilic shell in aqueous media.
29
The CMC is defined as the concentration above which micelles form. At low surfactant
30
concentrations, the surfactant molecules are arranged on the surface. The addition of more
31
surfactant will decrease the surface tension of the solution dramatically as more and more of
AC C
22
4
ACCEPTED MANUSCRIPT
1
surfactants exist on the surface, so the addition of surfactant molecules after the saturation point
2
will lead to the formation of micelles. The concentration of surfactant, above which micelles are
3
formed, is called the critical micelle concentration.
4
2. Materials and Methods
RI PT
5
The foaming tendency and stability for 50 wt. % MDEA (an equivalent amount of MDEA
7
was mixed with the equivalent amount of distilled water) were investigated. Next, the effects of
8
adding HCB and BHCL as corrosion inhibitors were explored. Different glass frits with different
9
pore sizes, i.e. P4 with pore size range of 10-16 µm, P3 glass frit with pore size range of 16-40
10
SC
6
µm and the P2 glass frit with pore size range of 40-100 µm, were tested.
12
M AN U
11
2.1. Materials
13
The fresh unused 50 wt. % MDEA was obtained from a gas sweetening unit and used for
14
these studies. HCB was obtained from Baker Hughes, USA while BHCL was supplied by
15
AkzoNobel, Sweden. The BHCL is a common corrosion inhibitor that is used by gas sweetening
16
plants (AkzoNobel 2008).
18
TE D
17
2.2. Experimental Setup and Procedures
Foaming experiments were carried out using the FoamScan instrument, which uses image
20
analysis and conductivity measurements to monitor foaming properties such as foamability, foam
21
stability as well as liquid content in the foam. Fig. 1 depicts the schematic diagram of the
22
FoamScan Apparatus. A fixed amount of the sample was injected into the glass tube (cuvette)
23
through the sampling pipe without mechanical shaking or stirring. Nitrogen gas was then
24
introduced to the cuvette at a specific flow rate through the glass frit at the bottom of the cuvette.
25
The test solution was vigorously bubbled by this nitrogen gas for a specific time (foaming time).
26
On foam production, the nitrogen injection was stopped, and the foam volume was recorded as a
27
function of time.
AC C
EP
19
28
5
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
1 2
Fig. 1. Schematic diagram of the FoamScan instrument.
3
The Bikerman Index characterized the stability of foam volume (mL) generated by a
5
given gas flow rate (mL/min), which is represented as BI (foam volume/volumetric gas flow).
6
The lower the BI, the lower is the foam volume, indicating a lower foam height.
TE D
4
At the end of each experiment, the foam images were analyzed using Cell Size Analysis
8
(CSA) software for bubble characterization. The film camera took several images at every 10 s
9
and then, by using CSA software mean number of bubbles, bubble radius and bubbles area were
10
EP
7
calculated. Table 1 shows the experimental matrix followed in this study.
12
AC C
11
Table 1. Experimental matrix. Experiments Liquid level (mL) Foaming time (s) Flow rate (mL/min) 1.
30
60 -120
100
2.
30
60 - 120
200
3.
30
60 - 120
300
4.
30
60 - 120
400
5.
45
60 -120
100
6
ACCEPTED MANUSCRIPT
45
60 -120
200
7.
45
60 -120
300
8.
45
60 -120
400
9.
60
60 -120
100
10.
60
60 -120
200
11.
60
60 -120
12.
60
60 -120
RI PT
6.
300 400
1
3. Results and Discussion
SC
2
Pure 50 wt. % MDEA was taken as a base line for all the studies. The foaming tendency of 50
4
wt. % MDEA solution was not remarkable in the range of glass frit pore sizes and the range of
5
temperature used.
M AN U
3
6
All the experiments were triplicated to get the standard deviation within 5%. Also, at a liquid
7
level of 30 mL, considerable foam volume was not obtained. Hence, most of the experiments
8
were carried out at 45 mL liquid level, unless specified.
9
3.1. Effect of HCB Corrosion Inhibitor
11
3.1.1. Effect of Nitrogen Flow Rate
TE D
10
Table 2 shows the effect of HCB addition on foam height of 50 wt. % MDEA solution at
13
different N2 gas flow rates. At low concentration, i.e. 0.05 wt. % (0.05 % from the total solution
14
weight) of HCB, no effect on foaming tendency was noticed, even at high N2 flow rates.
15
However, foaming was notable at 0.1 wt. % HCB concentration, mainly at high N2 flow rates.
16
This can be attributed to the change in surface tension of the MDEA solution in the presence of
17
HCB, as stated previously by Alhseinat et. al ( 2015a).
AC C
18
EP
12
19
Table 2. Effect of HCB addition on foam height for 50 wt. % MDEA solution at 20°C and 60 s
20
foaming time.
Sl. No.
1.
Concentration
Nitrogen flow rate
Foam height
(wt. %)
(mL/min)
(mL)
0.05
(100 – 400 )
0
7
ACCEPTED MANUSCRIPT
2.
0.1
100
0
3.
0.1
200
5
4.
0.1
300
5
5.
0.1
400
10
2
RI PT
1
3.1.2. Effect of Temperature
The effect of temperature on the foamability and stability of 50 wt. % MDEA solution in
4
the presence of 0.05 wt. % of HCB was studied with the medium pore size glass frit (P3; 16-40
5
µm) at 200 mL/min N2 flow rate, and the results are depicted in Fig. 2. The change in foam
6
volume and Bikerman index was small at the temperature range of 20-35 °C, but became more
7
visible at temperatures beyond 35 °C. Aguila-Hernandez et al. (2007) observed that surface
8
tension plays a major role on the foam behavior of aqueous MDEA, as a function of temperature.
9
As temperature increased, surface tension of the solution decreased, causing higher foam
10
volume. Henceforth, further studies on the effect of bubble size on foaming were carried out at
11
higher temperatures.
M AN U
SC
3
AC C
EP
TE D
12
13
8
ACCEPTED MANUSCRIPT
1
Fig. 2. Effect of temperature on foam volume and Bikerman Index for 0.05 wt. % HCB (foaming
2
time 60 s, gas flow 200 mL/min)
3
3.1.3. Effect of Glass Frits Fig. 3 shows the effect of glass frits of different pore sizes on foam volume for 0.05 wt.
5
% of HCB in 50 wt. % MDEA solution at 50 °C. Pore size did not have a linear relationship with
6
the foam volume and followed a parabolic path. It is noticed that at the small pore size, i.e. P4
7
(10-16) µm, there was no foaming, at the medium pore size P3 (16-40) µm, foaming was the
8
highest, and this dropped at the large pore size P2 (40-100) µm.
RI PT
4
This phenomena can be explained by ‘foam formation’ mechanism (Tse et al., 2003).
10
When a frit of smaller pore size is used, the bubbles formed are smaller in size, and the buoyancy
11
force cannot overcome the hydrostatic force and the surface force; hence the bubble is not
12
detached itself from the diffuser. In the case of the medium pore size frit, pore size is just
13
adequate to create a bubble, with a buoyancy force sufficient to overcome the forces on the
14
bubble itself (hydrostatic force and surface force) so that it is detached from the diffuser without
15
much turbulence, thus creating the foam. However, in the case of the larger pore size frit, the
16
bubbles formed will be larger, having a higher buoyancy force compared to inadequate
17
hydrostatic and surface forces. The larger bubbles are detached from the diffuser to follow a
18
coalescence-linked break-up mechanism (Tse et al., 1998).
M AN U
TE D
AC C
EP
19
SC
9
20 9
ACCEPTED MANUSCRIPT
1
Fig. 3. Effect of different glass frit on foam height for 0.05 wt % of HCB with 50 wt. % MDEA
2
solution at 60 s and 50 °C.
3
3.1.4. Bubble Characteristics The cell size analysis (CSA) software was used to analyze the number of bubbles, bubble
6
radius and distribution of bubbles, based on real-time photos of the foam. The software analyses
7
the bubble images produced in the Foamscan instrument. It detects the number of bubbles,
8
calculates its radius and shows the radius in ascending order. Thus, the mean number of bubbles,
9
the mean bubble radius (mm) and the mean bubbles area (mm2) can be obtained (Table 3). It is
10
observed that increasing the gas flow rate decreases the mean number of bubbles, but increases
11
the mean bubble radius and bubble area.
Table 3. Effect of nitrogen flow rate on bubbles characteristics for 0.05 wt. % HCB at 65 °C. Nitrogen flow
Mean Number
Mean bubble
Mean bubbles
rate (mL/min)
of bubbles
radius (mm)
area (mm2)
100
44
0.401
0.571
200
36
0.422
0.707
34
0.458
0.803
25
0.463
0.968
300 400 14
TE D
13
M AN U
12
SC
RI PT
4 5
3.2. Effect of BHCL Corrosion Inhibitor
16
3.2.1. Effect of Foaming Time
EP
15
The effect of adding 0.05 wt. % BHCL on foaming behavior of 50 wt. % MDEA was
18
studied with time for four specific nitrogen flow rates (100-400 mL/min), as shown in Fig. 4a. It
19
is observed that the final foam volume increased with increasing foaming time at all flow rates.
20
At lower nitrogen flow rates (100 and 200 mL/min), there is a sharp increase in final foam
21
volume with time. While, at higher flow rates (300 and 400 mL/min), the increment is consistent
22
in the time period. This can be explained by the contribution of contact time between the gas
23
and liquid to create a bubble, on the foam volume (Kulkarni and Joshi, 2005).
AC C
17
24
The effect of having a higher concentration of BHCL (0.1 wt %) on amine foaming was
25
also investigated and shown in Fig. 4b. A trend similar to 0.05 wt % was observed: foam height 10
ACCEPTED MANUSCRIPT
increased with increasing the time of foaming. The final foam height at 120 s decreased with
2
increasing the gas flow rate as observed for 0.05 wt. % BHCL also (Fig. 4a). A higher gas flow
3
rate increases the turbulence in the solution that obstructs the foam formation. This might be due
4
to the fact that increased turbulence created by the increasing gas flow rate disrupts foam
5
formation and reduces foam stability (Thitakamol and Veawab, 2009).
RI PT
1
(a)
TE D
M AN U
SC
6
(b)
Fig. 4. Effect of foam height for 50 wt. % MDEA solution at 20 °C on addition of (a) 0.05 wt. %
8
BHCL and (b) 0.1 wt. % BHCL.
EP
7
9
3.2.2. Effect of Nitrogen Flow Rate Fig. 5 illustrates the effect of the gas flow rate on foam tendency and stability for 0.05 wt.
12
% BHCL solution for 120 s at 20 °C. Increasing the gas flow rate from 100 to 400 mL/min
13
reduced both the foam volume and Bikerman index by 6 and 14 fold, respectively. This could be
14
attributed to the micellar aggregation, as discussed by Alhseinat et al. (2015a). The concentration
15
of corrosion inhibitors in the foam increases with time (time of foaming) and will be adequate to
16
reach critical micellar concentration (CMC), leading to aggregation into micelles to lower foam
17
volume.
AC C
10 11
11
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
1
Fig. 5. Effect of nitrogen flow rate on foam volume and Bikerman Index for 0.05 wt. % BHCL at
3
20 °C.
4 5
TE D
2
3.2.3. Effect of Temperature
Fig. 6 shows the effect of temperature on the foamability and stability of 0.05 wt. %
7
BHCL at 60 s foaming time with the medium pore size glass frit (P3; 16-40 µm), and 200
8
mL/min N2 flow rate. Foam volume and Bikerman index increased significantly as a result of
9
increasing temperature. The foam volume increased from 6 mL at room temperature to 87 mL at
10
65 °C. A decrease in surface tension causes the foam volume to rise, as minimum work is
11
required to expand the interface to produce more foam at higher temperatures (Aguila-Hernandez
12
et al., 2007; Thitakamol and Veawab, 2009).
AC C
13
EP
6
12
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
1
Fig. 6. Effect of temperature on foam volume and Bikerman Index for 0.05 wt. % BHCL.
4 5
3.2.4. Effect of Glass Frit Fig. 7 illustrates the influence of using different glass frit pore sizes on the final foam
6
volume for 0.05 wt. % BHCL at 50 °C. Unlike the parabolic trend of HCB, the foam volume
7
increased with the increasing of the frit pore size in the presence of BHCL. The foam height
8
increased significantly by using the medium pore size glass frit P3 (16-40) µm, and the highest
9
foam tendency was observed at the largest pore size P2 (40-100) used. The trend observed with
10
BHCL is different from the foaming observed with HCB. The foam volume was found to be 101
11
mL with the medium pore size (P3), and increased to 128 mL by using the larger pore size (P4).
12
Foam stability is one of the reasons behind the increase in the number of bubbles produced,
13
which led to increased effective foam volume.
AC C
EP
TE D
2 3
14
13
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
1 2
Fig. 7. Effect of different glass frit pore sizes on final foam volume for 0.05 wt. % BHCL, with
3
50 wt. % MDEA solution at 60 seconds and 50 °C.
4
3.2.5. Effect of Foaming with Higher Liquid Levels
TE D
5
Figs. 8 a and b show the effect of BHCL addition on the final foam volume and the
7
Bikerman index at different flow rates with 60 mL liquid level. It was found that both the final
8
foam volume and Bikerman Index increased with increasing BHCL concentration at particular
9
N2 flow rate. With an increase in BHCL concentration, the surface tension of the solution
10
decreases, causing higher foam volume (Pilon et al., 2001; Alhseinat et al., 2014). However, with
11
the increase in N2 flow rate, both the parameters decreased simultaneously.
AC C
12
EP
6
14
(a)
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
(b)
1
Fig. 8. Effect of BHCL concentration at different N2 flow rates with 60 mL liquid level at 120 s
2
and 20 °C on (a) final foam volume and (b) Bikerman Index.
3 4
3.2.6. Bubble Characteristics
Table 4 presents the temperature effect on the average bubble radius, number of bubbles
6
and bubble area for 0.05 wt. % BHCL. It is observed that increasing the temperature increases
7
the mean number of bubbles. Fig. 11 (a, b and c) represents the CSA software images of foam at
8
different temperatures. The bubble size increases with time due to coalescence and the effect
9
becomes much more noticeable with higher temperatures. At a nitrogen flow rate of 200
10
mL/min, the mean bubble radius increased from 0.251 to 0.349 mm, when the temperature was
11
increased from 35 to 50 °C, but decreased to 0.232 mm at 65 °C. Thus, increasing the solution
12
temperature increases the foamability and stability of the foam in an aqueous MDEA solution.
13
Generally, an increase in temperature affects the physical properties like surface tension, density,
14
viscosity and solubility of the solution (Pilon et al., 2001; Alhseinatet al., 2014). The influence of
15
viscosity is more significant with temperature than density and surface tension of the solution
16
(Pilon et al., 2001). Higher temperature decreases the surface tension, density and viscosity of
17
the solution, while the solubility of corrosion inhibitor increases in 50 wt. % MDEA solution.
18
Thus, higher foamability and stability of the foam was observed.
AC C
EP
TE D
5
19 15
ACCEPTED MANUSCRIPT
1
Table 4. Temperature influence of 0.05 wt. % BHCL on bubbles characteristics using CSA
2
software. Temperature
Mean number of
Mean Bubbles
Mean Bubbles
(mL/min)
(°C)
bubbles
radius (mm)
area (mm2)
200
35
24
0.251
0.252
200
50
32
0.349
0.877
200
65
43
0.232
0.269
300
35
27
0.376
0.676
300
50
33
0.390
0.907
300
65
40
0.277
0.372
400
35
17
0.422
0.936
400
50
400
65
(a)
SC
M AN U 24
0.579
2.496
41
0.307
0.438
TE D
3
RI PT
Flow rate
(b)
(c)
Fig. 9. CSA software images for 0.05 wt. % BHCL at 400 mL/min and (a) 35 °C (b) 50 °C (c) 65
5
°C.
6
EP
4
3.2.7. Foam Characteristics To understand foam stability, the overall foaming time, with foam volume and liquid
9
fraction percentages, in foam were analyzed. The overall time represents the time of foaming,
10 11
AC C
7 8
including foam breaking time, and the foaming time starts when the gas supply is switched on. Fig. 10 demonstrates the influence of temperature for 0.05 wt. % BHCL on the foam
12
volume profile at 60 s and 400 mL/min. The foam volume was almost zero at room temperature
13
(20 °C). It is observed that foam volume increased with increasing temperature of the MDEA
14
solution; the foam volume became double when temperature was raised from 35 °C to 65 °C.
15 16
ACCEPTED MANUSCRIPT
120
20° C 35° C 50° C 65° C
80
RI PT
Foam Volume (mL)
100
60
40
0 0
50
100
150
200
M AN U
Time (s)
1
SC
20
2
Fig. 10. Effect of temperature on foam volume for 0.05 wt. % BHCL in 50 wt. % MDEA
3
solution at 60 s and 400 mL/min.
4
Figs. 11 a and b show the effect of the foaming time on foam volume and the liquid
6
fraction percentage profile for 0.1 wt. % BHCL, at a nitrogen flow rate of 100 mL/min. It was
7
observed that increasing foaming time from 60 to 120 s was sufficient to increase the foam
8
volume by 3 fold (Fig. 11a). Lowering the foaming time reduces the contact time between gas
9
and liquid to create a bubble, resulting in a fewer number of bubbles, which decreases foam
10
volume. A similar trend was found in Fig. 11b for the liquid fraction percentage in foam.
11
Decreasing the foaming time to half reduce the liquid fraction from 6.5 to almost 2.5% in the
12
foam. The higher liquid fraction in the foam implies that thickness of the bubble wall is high and
13
will remain intact over a period of time and will not coalesce quickly; thus more stable foam will
14
be produced. Thus, 60 s foaming time was enough to enhance the drainage effects from bubbles
15
resulting in drier foam (a low liquid fraction percentage) (Hill and Eastoe, 2017).
EP
AC C
16
TE D
5
17
ACCEPTED MANUSCRIPT
100 7
60 s 90 s 120 s
60 s 90 s 120 s
6
60
40
20
5
4
3
2
1
0
0
50
100
150
200
250
300
350
400
0
50
100
150
200
SC
0
RI PT
Liquid Fraction in Foam (%)
Foam Volume (mL)
80
250
300
350
400
Time (s)
Time (s)
2
for 0.1 wt. % BHCL in 50 wt. % MDEA solution at 100 mL/min and 20 °C.
M AN U
1
(a) (b) Fig. 11. Effect of (a) foam volume and (b) liquid fraction percentage in foam with foaming time
3 4
4. Conclusions
The FoamScan instrument was used to study the effect of HCB and BHCL corrosion
6
inhibitors on the foaming tendency of an aqueous MDEA solution. It was found that increasing
7
the corrosion inhibitor concentration increased both foamability and stability.
8
solution temperature played a major role in increasing the foamability and stability of the MDEA
9
solution regardless of the used corrosion inhibitor. Increasing the nitrogen flow rate increased the
10
foam height in the presence of HCB, but the behavior was almost opposite in the presence of
11
BHCL. Temperature had a pronounced effect for both the inhibitors. The use of the large pore
12
size of the gas diffuser minimized the foam height for HCB, while the medium pore size was
13
recommended for BHCL. Thus, the process operating conditions related to these physical
14
parameters need to be examined carefully in the column design to minimize amine foaming.
16
EP
In addition,
AC C
15
TE D
5
Acknowledgments
17
The authors are grateful to the Gas Research Centre, The Petroleum Institute in Abu Dhabi for
18
funding the project (GRC006). Sincere thanks to the GASCO, Habshan unit for their co-
19
operation and support. We would like to thank Dr. Mark Wyatt, for checking English language
20
of the manuscript. 18
ACCEPTED MANUSCRIPT
1
References
3
Abry, R., Dupart, R., 1995. Amine Plant troubleshooting and optimization. Hydro. Proc. 4, 41-
4
50.
5
Aguila-Hernandez, J., Trejo, A., Garcia-Flores, B.E., 2007. Surface tension and foam behaviour
6
of aqueous solutions of blends of three alkanolamines, as a function of temperature. Coll. Surf.
7
A: Phy. Eng. Asp. 308. 33-46.
8
AkzoNobel (2008) "AkzoNobel Surface Chemistry in the oil industry."
9
Al-Dhafeeri, M.A., 2007. Identifying sources key to detailed troubleshooting of amine foaming.
SC
RI PT
2
Oil Gas J. 1, 1-12.
11
Alhseinat, E., Amr, M., Jumah R., Banat, F., 2015a. Removal of MDEA foam creators using
12
foam fractionation: Parametric study coupled with foam characterization. J. Nat. Gas Sci. Eng.
13
26, 502-509.
14
Alhseinat, E., Pal, P., Ganesan, A., Banat, F., 2015b. Effect of MDEA degradation products on
15
foaming behavior and physical properties of aqueous MDEA solutions. Int. J. Green. Gas Cont.
16
37, 280-286.
17
Alhseinat, E., Pal, P., Keewan M., Banat, F., 2014. Foaming study combined with physical
18
characterization of aqueous MDEA gas sweetening solutions. J. Nat. Gas Sci. Eng. 17, 49-57.
19
Ashassi-Sorkhabi, H., Nabavi-Amri, S.A., 2000. Corrosion inhibition of carbon steel in
20
petroleum /water mixtures by n-containing compounds. Acta Chimi. Slov. 47, 507-517.
21
Bhakta, A., Ruckenstein, E., 1997. Decay of standing foams: drainage, coalescence and collapse.
22
Adv. Coll. Inter. Sci. 70, 1-124.
23
Chen, X., Freeman S.A., Rochelle G.T., 2011. Foaming of aqueous piperazine and
24
monoethanolamine for CO2 capture. Int. J. Green. Gas Cont. 5(2), 381-386.
25
Davoudi, M., Safadoust, A.R., Mansoori, S.A., Mottaghi, H.R., 2014. The impurities effect on
26
thermal degradation and corrosivity of amine solution in South Pars gas sweetening plants. J.
27
Nat. Gas Sci. Eng., 19, 116-124.
28
Drenckhan, W., Saint-Jalmes, A., 2015. The science of foaming. Adv. Coll. Inter. Sci. 222, 228-
29
259.
30
El-Haddad, M.N., 2013. Chitosan as a green inhibitor for copper corrosion in acidic medium.
31
Inter. J. Bio. Macro. 55, 142-149.
AC C
EP
TE D
M AN U
10
19
ACCEPTED MANUSCRIPT
Finšgar, M., Jackson, J., 2014. Application of corrosion inhibitors for steels in acidic media for
2
the oil and gas industry: A review. Corro. Sci. 86, 17-41.
3
Garcia-Arriaga, V., Alvarez-Ramirez, J., Amaya, M., Sosa, E., 2010. H2S and O2 influence on
4
the corrosion of carbon steel immersed in a solution containing 3 M diethanolamine. Corro. Sci.
5
52, 2268-2279.
6
Gilyazetdinov, L.P., Matishev, V.A., 1990. Prevention of foaming in amine treatment of natural
7
gas. Chem. Tech. Fuels Oils. 26(5), 250-258.
8
Hart, E., 2016. Corrosion Inhibitors : Principles, Mechanisms and Applications. Hauppauge,
9
New York, Nova Science Publishers, Inc.
SC
RI PT
1
Hill, C., Eastoe, J., 2017. Foams: From nature to industry. Adv. Coll. Inter. Sci. 247, 496-513.
11
Jones, L.W., Alkire, J.D., 1985. Corrosion inhibitor for amine gas sweetening systems, United
12
States Patent, Number: 4541946.
13
Kohl, A., Nielsen, R., 1997. Gas Puri. Houston, Gulf Professional Publishing.
14
Kulkarni, A.A., Joshi, J.B., 2005. Bubble formation and bubble rise velocity in gas-liquid
15
systems: A Review. Ind. Eng. Chem. Res. 44 (16), 5873-5931.
16
Maiysa, K., 1992. Wet foams: Formation, properties and mechanism of stability. Adv. Coll.
17
Inter. Sci. 40, 37-83.
18
Narsimhan, G., Ruckenstein, E., 1986. Hydrodynamics, enrichment, and collapse in foams.
19
Lang. 2, 230-238.
20
Nielsen, R.B., Lewis, K.R., McCullough, J.G., Hansen, D.A., 1995. Controlling corrosion in
21
amine treating plants. Proceedings of the Laurance Reid Gas Conditioning Conference.
22
Oklahoma, USA.
23
Pauley, C.R., 1991. Face the facts about amine foaming. Chem. Eng. Prog. 87, 33-38.
24
Pauley, C.R., Hashemi, R., Caothein, S., 1998. Analysis of foaming mechanisms in amine plants.
25
Colorado, American Institute of Chemical Engineer’s Summer Meeting, Aug. 22-24.
26
Pilon, L., Fedorov, A.G., Viskanta, R., 2001. Steady-state thickness of liquid-gas foams. J. Coll.
27
Inter. Sci. 242, 425-436.
28
Pilon, L., Fedorov, A.G., Viskanta, R., 2002. Analysis of transient thickness of pneumatic
29
foams. Chem. Eng. Sci. 57, 977-990.
30
Thitakamol, B., Veawab, A., 2009. Foaming model for CO2 absorption process using aqueous
31
monoethanolamine solutions. Coll. Surf. A: Physicochem. Eng. Asp. 349, 125-136.
AC C
EP
TE D
M AN U
10
20
ACCEPTED MANUSCRIPT
Tse, K.L., Martin, T., McFarlane, C.M., Nienow, A.W., 2003. Small bubble formation via a
2
coalescence dependent break-up mechanism. Chem. Eng. Sci. 58, 275-286.
3
Tse, K.L., Martin, T.M., McFarlane, C.M., Nienow, A.W., 1998. Visualisation of bubble
4
coalescence in a coalescence cell a stirred tank and a bubble column. Chem. Eng. Sci. 53, 4031-
5
4036.
6
Yaro, A.S., Khadom, A.A., Wael, R.K., 2013. Apricot juice as green corrosion inhibitor of mild
7
steel in phosphoric acid. Alex. Eng. J. 52, 129-135.
AC C
EP
TE D
M AN U
SC
8
RI PT
1
21
ACCEPTED MANUSCRIPT
Highlights •
Foaming of aqueous MDEA solution in presence of BHCL/HCB corrosion inhibitors was
•
RI PT
studied.
Foaming tendency increases with increasing foaming time, temperature and inhibitor concentration.
Foaming behavior depends on the type of corrosion inhibitor.
•
Operating conditions need to be optimized to reduce amine foaming.
AC C
EP
TE D
M AN U
SC
•
S0920-4105(18)30152-9
DOI:
10.1016/j.petrol.2018.02.046
Reference:
PETROL 4716
To appear in:
Journal of Petroleum Science and Engineering
Received Date: 8 March 2017 Revised Date:
12 February 2018
Accepted Date: 19 February 2018
Please cite this article as: Keewan, M., Banat, F., Alhseinat, E., Zain, J., Pal, P., Effect of operating parameters and corrosion inhibitors on foaming behavior of aqueous methyldiethanolamine solutions, Journal of Petroleum Science and Engineering (2018), doi: 10.1016/j.petrol.2018.02.046. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
3
Mohammed Keewan, Fawzi Banat*, Emad Alhseinat, Jerina Zain, Priyabrata Pal
4 5 6
Department of Chemical Engineering, Khalifa University of Science and Technology, SAN Campus, P.O. Box 2533, Abu Dhabi, United Arab Emirates *Email: [email protected]
RI PT
2
Effect of operating parameters and corrosion inhibitors on foaming behavior of aqueous methyldiethanolamine solutions
1
7
Abstract
9
Corrosion inhibitors are one of the main causes of amine foaming in gas sweetening units. A
10
detailed understanding of amine foaming behavior in the presence of corrosion inhibitors is of
11
great importance if the foaming is to be minimized. In this work, a comparative study was
12
carried out to investigate the foaming tendency and bubbles characteristics of aqueous
13
methyldiethanolamine
14
Hydroxyehyl)cocoalkylamine; BHCL) and hydrocarbon-based (HCB) corrosion inhibitors. The
15
effect of different operating parameters such as nitrogen flow rate, corrosion inhibitor
16
concentration, foaming time, solution temperature and pore size of the gas diffuser were studied
17
using foam scan instrument. The results showed that on increasing the foaming time, solution
18
temperature and corrosion inhibitor concentrations, the foaming tendency increased. For BHCL,
19
the foam height tends to decrease with a high flow rate and a small pore size of the gas diffuser.
20
Nevertheless, the opposite trend was observed in the presence of the HCB corrosion inhibitor.
21
Depending upon the type of corrosion inhibitors, careful optimization of the operating conditions
22
showed a high potential to minimize amine foaming.
24 25
M AN U
in
the
presence
of
fatty
acid-based
EP
TE D
(MDEA)
AC C
23
SC
8
26 27
Keywords: Foaming, corrosion inhibitors, gas diffuser, amine, gas sweetening
28
1
(Bis(2-
ACCEPTED MANUSCRIPT
1
1. Introduction Acid gases like CO2 and H2S are found in considerable quantity in crude natural gas, which
3
creates obstacles for the combustion and transportation processes, by reducing its combustion
4
value, fouling and corroding the pipelines. Thus, the presence of acid gases is a major source of
5
maintenance problems in the facilities. This further affects the financial outcome of the process
6
and hence must be reduced. Natural gas sweetening with methyldiethanolamine (MDEA) is the
7
commonly employed method to remove the common acid gases like H2S and CO2.
RI PT
2
Foaming is one of the major operational problems in the amine absorption process leading to
9
serious consequences like loss of absorption capacity, reduced mass transfer area and efficiency.
10
Foaming can be triggered by various contaminants like corrosion inhibitors used for pipelines
11
and condensed liquid hydrocarbons. Suspended fine solid particles like iron sulfide (FeS) also
12
increase the foam stability and contaminants from boiler feed water treating chemicals, amine
13
degradation products and excessive antifoam agent also cause foaming (Kohl and Nielsen, 1997;
14
Al-Dhafeeri, 2007; Alhseinat et al., 2015b; Abry and Dupart, 1995). There are very few studies
15
considering a systematic and quantitative investigation of foaming in amine solutions (Pauley,
16
1991; Chen et al., 2011).
M AN U
SC
8
In gas sweetening units, MDEA in the concentration range of 40-50 wt. % is usually used to
18
remove CO2 and H2S which may be present at up to 30% in the crude natural gases (Alhseinat et
19
al., 2014). Foaming can cause carry over of alkanolamine solutions. Thus, subsequent loss of
20
solvents reduce plant throughput and off-specification of produced gas.
TE D
17
Different contaminants like thermal degraded products, heat stable salts, total suspended
22
solids and dissolved oxygen in alkanolamine solutions are the main causes of corrosion in the
23
gas sweetening plant (Davoudi et al., 2014). Corrosion has been observed in different locations
24
of the gas sweetening plants and can be classified as (a) corrosion of carbon steel with wet acid
25
gas; by the reaction between iron with carbon dioxide and hydrogen sulfide to produce a thin
26
liquid film, which occurs mainly at the bottom of the absorber, condenser, and top of the
27
regenerator and (b) corrosion of carbon steel in aqueous alkanolamine solutions, which mostly
28
occurs at the bottom of the regenerator, reboiler and lean/rich heat exchanger (Nielsen et al.,
29
1995; Davoudi et al., 2014). Correct and effective corrosion monitoring procedures are usually
30
followed in plants as a proactive tool (Davoudi et al., 2014). Further, corrosion inhibitors are
31
usually used for preventing corrosion by solvents used in treating sour gas streams (Jones and
AC C
EP
21
2
ACCEPTED MANUSCRIPT
Alkire, 1985). Corrosion inhibitors are compounds such as alkyl or aryl amines (Garcia-Arriaga
2
et al., 2010; Ashassi-Sorkhabi and Nabavi-Amri, 2000). They are used to protect the metallic
3
surfaces from corrosion and several types of corrosion inhibitors are currently used
4
in amine systems for natural gas refining. It is believed that corrosion inhibitors play one of the
5
major roles in foaming in the gas plant. Liquid hydrocarbons, such as HCB, which is a
6
thoroughly investigated corrosion inhibitor, are used in gas sweetening plants due to their higher
7
solubility in amine solution (Pauley 1991). Even BHCL, which is a polar and soluble in MDEA,
8
is also used in alkanolamine gas sweetening plants. The compound contains oxygen and nitrogen
9
atoms which is adsorbed on the metal surfaces to block the active corrosion sites (El-Haddad,
10
2013). The BHCL builds up a protective hydrophobic layer on the metal surfaces, which
11
provides a barrier to the dissolution of the metal ions into the MDEA solutions (Yaro et al.,
12
2013).
M AN U
SC
RI PT
1
There is very little literature available that explains the effect of corrosion inhibitors on amine
14
foaming (Gilyazetdinov and Matishev, 1990; Pauley, 1991; Al-Dhafeeri, 2007). The main
15
objective of this study is to provide more insights into the effect of corrosion inhibitors on the
16
foaming of amine solutions. The effect of two different corrosion inhibitors, HCB and BHCL, on
17
the foaming of MDEA solvent, will be investigated. Foaming tendency and stability of amine
18
solution will be studied by varying liquid level, time of foaming, nitrogen flow rate, corrosion
19
inhibitor concentration, solution temperature and pore size of the gas diffuser. The bubble and
20
foam volume will also be characterized. Additionally, the present study aims to further enhance
21
the current knowledge regarding amine foaming through real-time investigation, utilizing a
22
world-class foam analyzer, and this should help in reducing the propensity of foaming due to the
23
addition of corrosion inhibitors.
25
EP
AC C
24
TE D
13
1.1. Foam Theory
26
Liquid foams consist of gas bubbles which are closely packed within a liquid carrier matrix,
27
by mechanical incorporation of a gas into a liquid generating bubble (Drenckhan and Saint-
28
Jalmes, 2015). Two main parameters used to study the foaming problem are foaming tendency
29
and foaming stability (Pauley and Hashemi, 1998, Alhseinat et al., 2014). The foaming tendency
30
is the ease with which a solution forms a bubble and produce foam. The foam stability is the
3
ACCEPTED MANUSCRIPT
1
resistance to break the foam into a liquid phase, measured on the basis of time to coalesce the
2
bubble. Once foams are formed, they may undergo a thinning process, which decides the foam
4
stability. Usually, foams are subjected to three main instabilities, such as liquid drainage, foam
5
coalescence and foam rupture.
RI PT
3
6 7
1.2. Foam Structure
Foams encompass an ensemble of different size bubbles (Pilon et al., 2002). The injection
9
system decides the bubble size distribution at the bottom of the foam layer (Narsimhan and
10
Ruckenstein, 1986). Bubbles can have different shapes (spherical and polyhedral), depending on
11
the gas and liquid fraction. Spherical foam is commonly known as Kugelschaum, where the
12
thickness of the liquid film is approximately equal to the diameter of the gas bubble, which turns
13
to polyhedral, usually called Polyederschaum, when the thickness of the liquid film decreases
14
due to drainage. Polyhedral bubbles are usually located at the top of the foam layer and are
15
always subjected to coalescence and rupture. Meanwhile, the spherical bubbles are located at the
16
bottom (Maiysa, 1992). Foam, once formed, always undergoes a thinning process caused by
17
drainage, coalescence, and rupture, according to Plateau’s laws (Bhakta and Ruckenstein, 1997);
18
when three bubbles link up, they create a plateau border (PB), concaving three liquid films at an
19
angle of 120o, further reduced to 109o in the case of four bubbles adjoining, thereby creating a
20
honeycomb structure, enhancing the liquid drainage through the interconnected PB structure.
M AN U
TE D
EP
21
SC
8
1.3. Critical Micelle Concentration (CMC)
23
Corrosion inhibitors are surface-active compounds (Finšgar and Jackson 2014) and can be oil or
24
water-soluble surfactants (Hart, 2016). Surfactants are chemical compounds that lower the
25
surface tension of the liquid, and the interfacial tension between either two liquids or a liquid and
26
solid, and can act as foaming agents. Surfactants are amphiphilic molecules that contain an oil-
27
soluble hydrophobic hydrocarbon tail and a water-soluble hydrophilic head, where they
28
preferentially form micelles with a hydrophobic core and a hydrophilic shell in aqueous media.
29
The CMC is defined as the concentration above which micelles form. At low surfactant
30
concentrations, the surfactant molecules are arranged on the surface. The addition of more
31
surfactant will decrease the surface tension of the solution dramatically as more and more of
AC C
22
4
ACCEPTED MANUSCRIPT
1
surfactants exist on the surface, so the addition of surfactant molecules after the saturation point
2
will lead to the formation of micelles. The concentration of surfactant, above which micelles are
3
formed, is called the critical micelle concentration.
4
2. Materials and Methods
RI PT
5
The foaming tendency and stability for 50 wt. % MDEA (an equivalent amount of MDEA
7
was mixed with the equivalent amount of distilled water) were investigated. Next, the effects of
8
adding HCB and BHCL as corrosion inhibitors were explored. Different glass frits with different
9
pore sizes, i.e. P4 with pore size range of 10-16 µm, P3 glass frit with pore size range of 16-40
10
SC
6
µm and the P2 glass frit with pore size range of 40-100 µm, were tested.
12
M AN U
11
2.1. Materials
13
The fresh unused 50 wt. % MDEA was obtained from a gas sweetening unit and used for
14
these studies. HCB was obtained from Baker Hughes, USA while BHCL was supplied by
15
AkzoNobel, Sweden. The BHCL is a common corrosion inhibitor that is used by gas sweetening
16
plants (AkzoNobel 2008).
18
TE D
17
2.2. Experimental Setup and Procedures
Foaming experiments were carried out using the FoamScan instrument, which uses image
20
analysis and conductivity measurements to monitor foaming properties such as foamability, foam
21
stability as well as liquid content in the foam. Fig. 1 depicts the schematic diagram of the
22
FoamScan Apparatus. A fixed amount of the sample was injected into the glass tube (cuvette)
23
through the sampling pipe without mechanical shaking or stirring. Nitrogen gas was then
24
introduced to the cuvette at a specific flow rate through the glass frit at the bottom of the cuvette.
25
The test solution was vigorously bubbled by this nitrogen gas for a specific time (foaming time).
26
On foam production, the nitrogen injection was stopped, and the foam volume was recorded as a
27
function of time.
AC C
EP
19
28
5
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
1 2
Fig. 1. Schematic diagram of the FoamScan instrument.
3
The Bikerman Index characterized the stability of foam volume (mL) generated by a
5
given gas flow rate (mL/min), which is represented as BI (foam volume/volumetric gas flow).
6
The lower the BI, the lower is the foam volume, indicating a lower foam height.
TE D
4
At the end of each experiment, the foam images were analyzed using Cell Size Analysis
8
(CSA) software for bubble characterization. The film camera took several images at every 10 s
9
and then, by using CSA software mean number of bubbles, bubble radius and bubbles area were
10
EP
7
calculated. Table 1 shows the experimental matrix followed in this study.
12
AC C
11
Table 1. Experimental matrix. Experiments Liquid level (mL) Foaming time (s) Flow rate (mL/min) 1.
30
60 -120
100
2.
30
60 - 120
200
3.
30
60 - 120
300
4.
30
60 - 120
400
5.
45
60 -120
100
6
ACCEPTED MANUSCRIPT
45
60 -120
200
7.
45
60 -120
300
8.
45
60 -120
400
9.
60
60 -120
100
10.
60
60 -120
200
11.
60
60 -120
12.
60
60 -120
RI PT
6.
300 400
1
3. Results and Discussion
SC
2
Pure 50 wt. % MDEA was taken as a base line for all the studies. The foaming tendency of 50
4
wt. % MDEA solution was not remarkable in the range of glass frit pore sizes and the range of
5
temperature used.
M AN U
3
6
All the experiments were triplicated to get the standard deviation within 5%. Also, at a liquid
7
level of 30 mL, considerable foam volume was not obtained. Hence, most of the experiments
8
were carried out at 45 mL liquid level, unless specified.
9
3.1. Effect of HCB Corrosion Inhibitor
11
3.1.1. Effect of Nitrogen Flow Rate
TE D
10
Table 2 shows the effect of HCB addition on foam height of 50 wt. % MDEA solution at
13
different N2 gas flow rates. At low concentration, i.e. 0.05 wt. % (0.05 % from the total solution
14
weight) of HCB, no effect on foaming tendency was noticed, even at high N2 flow rates.
15
However, foaming was notable at 0.1 wt. % HCB concentration, mainly at high N2 flow rates.
16
This can be attributed to the change in surface tension of the MDEA solution in the presence of
17
HCB, as stated previously by Alhseinat et. al ( 2015a).
AC C
18
EP
12
19
Table 2. Effect of HCB addition on foam height for 50 wt. % MDEA solution at 20°C and 60 s
20
foaming time.
Sl. No.
1.
Concentration
Nitrogen flow rate
Foam height
(wt. %)
(mL/min)
(mL)
0.05
(100 – 400 )
0
7
ACCEPTED MANUSCRIPT
2.
0.1
100
0
3.
0.1
200
5
4.
0.1
300
5
5.
0.1
400
10
2
RI PT
1
3.1.2. Effect of Temperature
The effect of temperature on the foamability and stability of 50 wt. % MDEA solution in
4
the presence of 0.05 wt. % of HCB was studied with the medium pore size glass frit (P3; 16-40
5
µm) at 200 mL/min N2 flow rate, and the results are depicted in Fig. 2. The change in foam
6
volume and Bikerman index was small at the temperature range of 20-35 °C, but became more
7
visible at temperatures beyond 35 °C. Aguila-Hernandez et al. (2007) observed that surface
8
tension plays a major role on the foam behavior of aqueous MDEA, as a function of temperature.
9
As temperature increased, surface tension of the solution decreased, causing higher foam
10
volume. Henceforth, further studies on the effect of bubble size on foaming were carried out at
11
higher temperatures.
M AN U
SC
3
AC C
EP
TE D
12
13
8
ACCEPTED MANUSCRIPT
1
Fig. 2. Effect of temperature on foam volume and Bikerman Index for 0.05 wt. % HCB (foaming
2
time 60 s, gas flow 200 mL/min)
3
3.1.3. Effect of Glass Frits Fig. 3 shows the effect of glass frits of different pore sizes on foam volume for 0.05 wt.
5
% of HCB in 50 wt. % MDEA solution at 50 °C. Pore size did not have a linear relationship with
6
the foam volume and followed a parabolic path. It is noticed that at the small pore size, i.e. P4
7
(10-16) µm, there was no foaming, at the medium pore size P3 (16-40) µm, foaming was the
8
highest, and this dropped at the large pore size P2 (40-100) µm.
RI PT
4
This phenomena can be explained by ‘foam formation’ mechanism (Tse et al., 2003).
10
When a frit of smaller pore size is used, the bubbles formed are smaller in size, and the buoyancy
11
force cannot overcome the hydrostatic force and the surface force; hence the bubble is not
12
detached itself from the diffuser. In the case of the medium pore size frit, pore size is just
13
adequate to create a bubble, with a buoyancy force sufficient to overcome the forces on the
14
bubble itself (hydrostatic force and surface force) so that it is detached from the diffuser without
15
much turbulence, thus creating the foam. However, in the case of the larger pore size frit, the
16
bubbles formed will be larger, having a higher buoyancy force compared to inadequate
17
hydrostatic and surface forces. The larger bubbles are detached from the diffuser to follow a
18
coalescence-linked break-up mechanism (Tse et al., 1998).
M AN U
TE D
AC C
EP
19
SC
9
20 9
ACCEPTED MANUSCRIPT
1
Fig. 3. Effect of different glass frit on foam height for 0.05 wt % of HCB with 50 wt. % MDEA
2
solution at 60 s and 50 °C.
3
3.1.4. Bubble Characteristics The cell size analysis (CSA) software was used to analyze the number of bubbles, bubble
6
radius and distribution of bubbles, based on real-time photos of the foam. The software analyses
7
the bubble images produced in the Foamscan instrument. It detects the number of bubbles,
8
calculates its radius and shows the radius in ascending order. Thus, the mean number of bubbles,
9
the mean bubble radius (mm) and the mean bubbles area (mm2) can be obtained (Table 3). It is
10
observed that increasing the gas flow rate decreases the mean number of bubbles, but increases
11
the mean bubble radius and bubble area.
Table 3. Effect of nitrogen flow rate on bubbles characteristics for 0.05 wt. % HCB at 65 °C. Nitrogen flow
Mean Number
Mean bubble
Mean bubbles
rate (mL/min)
of bubbles
radius (mm)
area (mm2)
100
44
0.401
0.571
200
36
0.422
0.707
34
0.458
0.803
25
0.463
0.968
300 400 14
TE D
13
M AN U
12
SC
RI PT
4 5
3.2. Effect of BHCL Corrosion Inhibitor
16
3.2.1. Effect of Foaming Time
EP
15
The effect of adding 0.05 wt. % BHCL on foaming behavior of 50 wt. % MDEA was
18
studied with time for four specific nitrogen flow rates (100-400 mL/min), as shown in Fig. 4a. It
19
is observed that the final foam volume increased with increasing foaming time at all flow rates.
20
At lower nitrogen flow rates (100 and 200 mL/min), there is a sharp increase in final foam
21
volume with time. While, at higher flow rates (300 and 400 mL/min), the increment is consistent
22
in the time period. This can be explained by the contribution of contact time between the gas
23
and liquid to create a bubble, on the foam volume (Kulkarni and Joshi, 2005).
AC C
17
24
The effect of having a higher concentration of BHCL (0.1 wt %) on amine foaming was
25
also investigated and shown in Fig. 4b. A trend similar to 0.05 wt % was observed: foam height 10
ACCEPTED MANUSCRIPT
increased with increasing the time of foaming. The final foam height at 120 s decreased with
2
increasing the gas flow rate as observed for 0.05 wt. % BHCL also (Fig. 4a). A higher gas flow
3
rate increases the turbulence in the solution that obstructs the foam formation. This might be due
4
to the fact that increased turbulence created by the increasing gas flow rate disrupts foam
5
formation and reduces foam stability (Thitakamol and Veawab, 2009).
RI PT
1
(a)
TE D
M AN U
SC
6
(b)
Fig. 4. Effect of foam height for 50 wt. % MDEA solution at 20 °C on addition of (a) 0.05 wt. %
8
BHCL and (b) 0.1 wt. % BHCL.
EP
7
9
3.2.2. Effect of Nitrogen Flow Rate Fig. 5 illustrates the effect of the gas flow rate on foam tendency and stability for 0.05 wt.
12
% BHCL solution for 120 s at 20 °C. Increasing the gas flow rate from 100 to 400 mL/min
13
reduced both the foam volume and Bikerman index by 6 and 14 fold, respectively. This could be
14
attributed to the micellar aggregation, as discussed by Alhseinat et al. (2015a). The concentration
15
of corrosion inhibitors in the foam increases with time (time of foaming) and will be adequate to
16
reach critical micellar concentration (CMC), leading to aggregation into micelles to lower foam
17
volume.
AC C
10 11
11
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
1
Fig. 5. Effect of nitrogen flow rate on foam volume and Bikerman Index for 0.05 wt. % BHCL at
3
20 °C.
4 5
TE D
2
3.2.3. Effect of Temperature
Fig. 6 shows the effect of temperature on the foamability and stability of 0.05 wt. %
7
BHCL at 60 s foaming time with the medium pore size glass frit (P3; 16-40 µm), and 200
8
mL/min N2 flow rate. Foam volume and Bikerman index increased significantly as a result of
9
increasing temperature. The foam volume increased from 6 mL at room temperature to 87 mL at
10
65 °C. A decrease in surface tension causes the foam volume to rise, as minimum work is
11
required to expand the interface to produce more foam at higher temperatures (Aguila-Hernandez
12
et al., 2007; Thitakamol and Veawab, 2009).
AC C
13
EP
6
12
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
1
Fig. 6. Effect of temperature on foam volume and Bikerman Index for 0.05 wt. % BHCL.
4 5
3.2.4. Effect of Glass Frit Fig. 7 illustrates the influence of using different glass frit pore sizes on the final foam
6
volume for 0.05 wt. % BHCL at 50 °C. Unlike the parabolic trend of HCB, the foam volume
7
increased with the increasing of the frit pore size in the presence of BHCL. The foam height
8
increased significantly by using the medium pore size glass frit P3 (16-40) µm, and the highest
9
foam tendency was observed at the largest pore size P2 (40-100) used. The trend observed with
10
BHCL is different from the foaming observed with HCB. The foam volume was found to be 101
11
mL with the medium pore size (P3), and increased to 128 mL by using the larger pore size (P4).
12
Foam stability is one of the reasons behind the increase in the number of bubbles produced,
13
which led to increased effective foam volume.
AC C
EP
TE D
2 3
14
13
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
1 2
Fig. 7. Effect of different glass frit pore sizes on final foam volume for 0.05 wt. % BHCL, with
3
50 wt. % MDEA solution at 60 seconds and 50 °C.
4
3.2.5. Effect of Foaming with Higher Liquid Levels
TE D
5
Figs. 8 a and b show the effect of BHCL addition on the final foam volume and the
7
Bikerman index at different flow rates with 60 mL liquid level. It was found that both the final
8
foam volume and Bikerman Index increased with increasing BHCL concentration at particular
9
N2 flow rate. With an increase in BHCL concentration, the surface tension of the solution
10
decreases, causing higher foam volume (Pilon et al., 2001; Alhseinat et al., 2014). However, with
11
the increase in N2 flow rate, both the parameters decreased simultaneously.
AC C
12
EP
6
14
(a)
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
(b)
1
Fig. 8. Effect of BHCL concentration at different N2 flow rates with 60 mL liquid level at 120 s
2
and 20 °C on (a) final foam volume and (b) Bikerman Index.
3 4
3.2.6. Bubble Characteristics
Table 4 presents the temperature effect on the average bubble radius, number of bubbles
6
and bubble area for 0.05 wt. % BHCL. It is observed that increasing the temperature increases
7
the mean number of bubbles. Fig. 11 (a, b and c) represents the CSA software images of foam at
8
different temperatures. The bubble size increases with time due to coalescence and the effect
9
becomes much more noticeable with higher temperatures. At a nitrogen flow rate of 200
10
mL/min, the mean bubble radius increased from 0.251 to 0.349 mm, when the temperature was
11
increased from 35 to 50 °C, but decreased to 0.232 mm at 65 °C. Thus, increasing the solution
12
temperature increases the foamability and stability of the foam in an aqueous MDEA solution.
13
Generally, an increase in temperature affects the physical properties like surface tension, density,
14
viscosity and solubility of the solution (Pilon et al., 2001; Alhseinatet al., 2014). The influence of
15
viscosity is more significant with temperature than density and surface tension of the solution
16
(Pilon et al., 2001). Higher temperature decreases the surface tension, density and viscosity of
17
the solution, while the solubility of corrosion inhibitor increases in 50 wt. % MDEA solution.
18
Thus, higher foamability and stability of the foam was observed.
AC C
EP
TE D
5
19 15
ACCEPTED MANUSCRIPT
1
Table 4. Temperature influence of 0.05 wt. % BHCL on bubbles characteristics using CSA
2
software. Temperature
Mean number of
Mean Bubbles
Mean Bubbles
(mL/min)
(°C)
bubbles
radius (mm)
area (mm2)
200
35
24
0.251
0.252
200
50
32
0.349
0.877
200
65
43
0.232
0.269
300
35
27
0.376
0.676
300
50
33
0.390
0.907
300
65
40
0.277
0.372
400
35
17
0.422
0.936
400
50
400
65
(a)
SC
M AN U 24
0.579
2.496
41
0.307
0.438
TE D
3
RI PT
Flow rate
(b)
(c)
Fig. 9. CSA software images for 0.05 wt. % BHCL at 400 mL/min and (a) 35 °C (b) 50 °C (c) 65
5
°C.
6
EP
4
3.2.7. Foam Characteristics To understand foam stability, the overall foaming time, with foam volume and liquid
9
fraction percentages, in foam were analyzed. The overall time represents the time of foaming,
10 11
AC C
7 8
including foam breaking time, and the foaming time starts when the gas supply is switched on. Fig. 10 demonstrates the influence of temperature for 0.05 wt. % BHCL on the foam
12
volume profile at 60 s and 400 mL/min. The foam volume was almost zero at room temperature
13
(20 °C). It is observed that foam volume increased with increasing temperature of the MDEA
14
solution; the foam volume became double when temperature was raised from 35 °C to 65 °C.
15 16
ACCEPTED MANUSCRIPT
120
20° C 35° C 50° C 65° C
80
RI PT
Foam Volume (mL)
100
60
40
0 0
50
100
150
200
M AN U
Time (s)
1
SC
20
2
Fig. 10. Effect of temperature on foam volume for 0.05 wt. % BHCL in 50 wt. % MDEA
3
solution at 60 s and 400 mL/min.
4
Figs. 11 a and b show the effect of the foaming time on foam volume and the liquid
6
fraction percentage profile for 0.1 wt. % BHCL, at a nitrogen flow rate of 100 mL/min. It was
7
observed that increasing foaming time from 60 to 120 s was sufficient to increase the foam
8
volume by 3 fold (Fig. 11a). Lowering the foaming time reduces the contact time between gas
9
and liquid to create a bubble, resulting in a fewer number of bubbles, which decreases foam
10
volume. A similar trend was found in Fig. 11b for the liquid fraction percentage in foam.
11
Decreasing the foaming time to half reduce the liquid fraction from 6.5 to almost 2.5% in the
12
foam. The higher liquid fraction in the foam implies that thickness of the bubble wall is high and
13
will remain intact over a period of time and will not coalesce quickly; thus more stable foam will
14
be produced. Thus, 60 s foaming time was enough to enhance the drainage effects from bubbles
15
resulting in drier foam (a low liquid fraction percentage) (Hill and Eastoe, 2017).
EP
AC C
16
TE D
5
17
ACCEPTED MANUSCRIPT
100 7
60 s 90 s 120 s
60 s 90 s 120 s
6
60
40
20
5
4
3
2
1
0
0
50
100
150
200
250
300
350
400
0
50
100
150
200
SC
0
RI PT
Liquid Fraction in Foam (%)
Foam Volume (mL)
80
250
300
350
400
Time (s)
Time (s)
2
for 0.1 wt. % BHCL in 50 wt. % MDEA solution at 100 mL/min and 20 °C.
M AN U
1
(a) (b) Fig. 11. Effect of (a) foam volume and (b) liquid fraction percentage in foam with foaming time
3 4
4. Conclusions
The FoamScan instrument was used to study the effect of HCB and BHCL corrosion
6
inhibitors on the foaming tendency of an aqueous MDEA solution. It was found that increasing
7
the corrosion inhibitor concentration increased both foamability and stability.
8
solution temperature played a major role in increasing the foamability and stability of the MDEA
9
solution regardless of the used corrosion inhibitor. Increasing the nitrogen flow rate increased the
10
foam height in the presence of HCB, but the behavior was almost opposite in the presence of
11
BHCL. Temperature had a pronounced effect for both the inhibitors. The use of the large pore
12
size of the gas diffuser minimized the foam height for HCB, while the medium pore size was
13
recommended for BHCL. Thus, the process operating conditions related to these physical
14
parameters need to be examined carefully in the column design to minimize amine foaming.
16
EP
In addition,
AC C
15
TE D
5
Acknowledgments
17
The authors are grateful to the Gas Research Centre, The Petroleum Institute in Abu Dhabi for
18
funding the project (GRC006). Sincere thanks to the GASCO, Habshan unit for their co-
19
operation and support. We would like to thank Dr. Mark Wyatt, for checking English language
20
of the manuscript. 18
ACCEPTED MANUSCRIPT
1
References
3
Abry, R., Dupart, R., 1995. Amine Plant troubleshooting and optimization. Hydro. Proc. 4, 41-
4
50.
5
Aguila-Hernandez, J., Trejo, A., Garcia-Flores, B.E., 2007. Surface tension and foam behaviour
6
of aqueous solutions of blends of three alkanolamines, as a function of temperature. Coll. Surf.
7
A: Phy. Eng. Asp. 308. 33-46.
8
AkzoNobel (2008) "AkzoNobel Surface Chemistry in the oil industry."
9
Al-Dhafeeri, M.A., 2007. Identifying sources key to detailed troubleshooting of amine foaming.
SC
RI PT
2
Oil Gas J. 1, 1-12.
11
Alhseinat, E., Amr, M., Jumah R., Banat, F., 2015a. Removal of MDEA foam creators using
12
foam fractionation: Parametric study coupled with foam characterization. J. Nat. Gas Sci. Eng.
13
26, 502-509.
14
Alhseinat, E., Pal, P., Ganesan, A., Banat, F., 2015b. Effect of MDEA degradation products on
15
foaming behavior and physical properties of aqueous MDEA solutions. Int. J. Green. Gas Cont.
16
37, 280-286.
17
Alhseinat, E., Pal, P., Keewan M., Banat, F., 2014. Foaming study combined with physical
18
characterization of aqueous MDEA gas sweetening solutions. J. Nat. Gas Sci. Eng. 17, 49-57.
19
Ashassi-Sorkhabi, H., Nabavi-Amri, S.A., 2000. Corrosion inhibition of carbon steel in
20
petroleum /water mixtures by n-containing compounds. Acta Chimi. Slov. 47, 507-517.
21
Bhakta, A., Ruckenstein, E., 1997. Decay of standing foams: drainage, coalescence and collapse.
22
Adv. Coll. Inter. Sci. 70, 1-124.
23
Chen, X., Freeman S.A., Rochelle G.T., 2011. Foaming of aqueous piperazine and
24
monoethanolamine for CO2 capture. Int. J. Green. Gas Cont. 5(2), 381-386.
25
Davoudi, M., Safadoust, A.R., Mansoori, S.A., Mottaghi, H.R., 2014. The impurities effect on
26
thermal degradation and corrosivity of amine solution in South Pars gas sweetening plants. J.
27
Nat. Gas Sci. Eng., 19, 116-124.
28
Drenckhan, W., Saint-Jalmes, A., 2015. The science of foaming. Adv. Coll. Inter. Sci. 222, 228-
29
259.
30
El-Haddad, M.N., 2013. Chitosan as a green inhibitor for copper corrosion in acidic medium.
31
Inter. J. Bio. Macro. 55, 142-149.
AC C
EP
TE D
M AN U
10
19
ACCEPTED MANUSCRIPT
Finšgar, M., Jackson, J., 2014. Application of corrosion inhibitors for steels in acidic media for
2
the oil and gas industry: A review. Corro. Sci. 86, 17-41.
3
Garcia-Arriaga, V., Alvarez-Ramirez, J., Amaya, M., Sosa, E., 2010. H2S and O2 influence on
4
the corrosion of carbon steel immersed in a solution containing 3 M diethanolamine. Corro. Sci.
5
52, 2268-2279.
6
Gilyazetdinov, L.P., Matishev, V.A., 1990. Prevention of foaming in amine treatment of natural
7
gas. Chem. Tech. Fuels Oils. 26(5), 250-258.
8
Hart, E., 2016. Corrosion Inhibitors : Principles, Mechanisms and Applications. Hauppauge,
9
New York, Nova Science Publishers, Inc.
SC
RI PT
1
Hill, C., Eastoe, J., 2017. Foams: From nature to industry. Adv. Coll. Inter. Sci. 247, 496-513.
11
Jones, L.W., Alkire, J.D., 1985. Corrosion inhibitor for amine gas sweetening systems, United
12
States Patent, Number: 4541946.
13
Kohl, A., Nielsen, R., 1997. Gas Puri. Houston, Gulf Professional Publishing.
14
Kulkarni, A.A., Joshi, J.B., 2005. Bubble formation and bubble rise velocity in gas-liquid
15
systems: A Review. Ind. Eng. Chem. Res. 44 (16), 5873-5931.
16
Maiysa, K., 1992. Wet foams: Formation, properties and mechanism of stability. Adv. Coll.
17
Inter. Sci. 40, 37-83.
18
Narsimhan, G., Ruckenstein, E., 1986. Hydrodynamics, enrichment, and collapse in foams.
19
Lang. 2, 230-238.
20
Nielsen, R.B., Lewis, K.R., McCullough, J.G., Hansen, D.A., 1995. Controlling corrosion in
21
amine treating plants. Proceedings of the Laurance Reid Gas Conditioning Conference.
22
Oklahoma, USA.
23
Pauley, C.R., 1991. Face the facts about amine foaming. Chem. Eng. Prog. 87, 33-38.
24
Pauley, C.R., Hashemi, R., Caothein, S., 1998. Analysis of foaming mechanisms in amine plants.
25
Colorado, American Institute of Chemical Engineer’s Summer Meeting, Aug. 22-24.
26
Pilon, L., Fedorov, A.G., Viskanta, R., 2001. Steady-state thickness of liquid-gas foams. J. Coll.
27
Inter. Sci. 242, 425-436.
28
Pilon, L., Fedorov, A.G., Viskanta, R., 2002. Analysis of transient thickness of pneumatic
29
foams. Chem. Eng. Sci. 57, 977-990.
30
Thitakamol, B., Veawab, A., 2009. Foaming model for CO2 absorption process using aqueous
31
monoethanolamine solutions. Coll. Surf. A: Physicochem. Eng. Asp. 349, 125-136.
AC C
EP
TE D
M AN U
10
20
ACCEPTED MANUSCRIPT
Tse, K.L., Martin, T., McFarlane, C.M., Nienow, A.W., 2003. Small bubble formation via a
2
coalescence dependent break-up mechanism. Chem. Eng. Sci. 58, 275-286.
3
Tse, K.L., Martin, T.M., McFarlane, C.M., Nienow, A.W., 1998. Visualisation of bubble
4
coalescence in a coalescence cell a stirred tank and a bubble column. Chem. Eng. Sci. 53, 4031-
5
4036.
6
Yaro, A.S., Khadom, A.A., Wael, R.K., 2013. Apricot juice as green corrosion inhibitor of mild
7
steel in phosphoric acid. Alex. Eng. J. 52, 129-135.
AC C
EP
TE D
M AN U
SC
8
RI PT
1
21
ACCEPTED MANUSCRIPT
Highlights •
Foaming of aqueous MDEA solution in presence of BHCL/HCB corrosion inhibitors was
•
RI PT
studied.
Foaming tendency increases with increasing foaming time, temperature and inhibitor concentration.
Foaming behavior depends on the type of corrosion inhibitor.
•
Operating conditions need to be optimized to reduce amine foaming.
AC C
EP
TE D
M AN U
SC
•