- Email: [email protected]
Inhibition of methane hydrate re-formation in offshore pipelines with a kinetic hydrate inhibitor
Journal of Petroleum Science and Engineering 88–89 (2012) 61–66
Contents lists available at SciVerse ScienceDirect
Journal of Petroleum Science and Engineering journal homepage: www.elsevier.com/locate/petrol
Inhibition of methane hydrate re-formation in offshore pipelines with a kinetic hydrate inhibitor Yutaek Seo a, Seong-Pil Kang b,⁎ a b
CSIRO Petroleum Resources, 26 Dick Perry Avenue, Kensington, Perth, WA 6151, Australia Climate Change Research Division, Korea Institute of Energy Research, 71-2 Jang-dong, Yuseong-gu, Daejeon 305-343, Republic of Korea
a r t i c l e
i n f o
Article history: Received 31 May 2011 Revised 6 October 2011 Accepted 17 November 2011 Available online 1 December 2011 Keywords: gas hydrate methane production kinetic hydrate inhibitor PVCap flow assurance
a b s t r a c t Methane gas from marine hydrate deposits can be produced by one or a combination of three methods; depressurization, thermal stimulation, and the injection of hydrate inhibitors. Because residual hydrate structures known as hydrate precursors will exist in the liquid water phase after dissociation, the risk of methane hydrate re-formation has to be evaluated during the production and transportation of methane gas through offshore pipelines. New experimental procedures composed of three stages are designed to simulate the dissociation of marine hydrates and the transportation of produced fluids. The obtained experimental results have shown that methane hydrates are rapidly re-formed when the temperature of dissociated water falls into the hydrate formation region during the cooling down process. The subcooling for three different dissociation pressures of 80, 70, and 60 bar were 1.3, 1.2 and 1.6 °C, respectively. One viable option to avoid hydrate re-formation is injecting hydrate inhibitors before transporting the fluids through pipelines. Among various hydrate inhibitors, Poly(N-vinylcaprolactam) (PVCap) was selected as a possible candidate for a Kinetic Hydrate Inhibitor (KHI) and injected into dissociated water before cooling down the fluids. The concentration of PVCap was varied from 0.5 to 3.0 wt.%. With an increase in the PVCap concentration, the subcooling increased to 7.8 °C at a dissociation pressure of 80 bar, which suggests that the risk of hydrate re-formation can be reduced by injecting PVCap. Moreover it is observed that the subcooling increased to 8.8 °C at the PVCap concentration of 3.0 wt.% in the presence of NaCl in the water phase. Although the use of KHI in conventional gas production has become common, its applicability to methane hydrate production has not yet been studied thoroughly, especially in the presence of residual hydrate structures. In this work, the application of KHI to methane hydrate production is discussed. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Gas hydrates are crystalline compounds that form when guest molecules are incorporated in host cages formed by water molecules through hydrogen bonding (Sloan and Koh, 2008). Low molecular weight gas molecules such as methane and carbon dioxide are captured in these cages, and each hydrate lattice consists of at least two types of polyhedral cages. Three distinct structural families, termed structures I, II, and H, are known, and they show distinct structural characteristics in terms of their cage types, distribution of guest molecules, and thermodynamic equilibrium conditions. Recently, gas hydrates have attracted a great deal of attention because of their significant potential as a source of methane gas. There are massive hydrate deposits both under the permafrost and in sediment on the continental margins. One issue that has been discussed within the hydrate research community is possible technologies to
⁎ Corresponding author. Tel.: + 82 42 860 3475; fax: + 82 42 860 3134. E-mail address: [email protected] (S.-P. Kang). 0920-4105/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.petrol.2011.11.001
develop gas hydrate deposits. Concepts for methane production from hydrate deposits proposed to date involve one or a combination of the following three methods: depressurization, thermal stimulation, and the use of hydrate inhibitor injections. Among them, depressurization is considered to be the most promising method because the highest energy profit ratio can be achieved (Kurihara et al., 2008a). Following the successful production from hydrate deposits in the Mackenzie Delta (Kurihara et al., 2008b; Yamamoto, 2007), there have been considerable efforts to determine the optimum production conditions for marine hydrate deposits as well (Konno et al., 2010; Oyama et al., 2009). However, most of these works focused on the characteristics of hydrate dissociation under various production scenarios, and there has been little effort to understand the transportation of dissociated methane gas and water via pipelines from hydrate wells to processing facilities. When the produced fluids are transported through offshore pipelines, they cool down due to a heat exchange with sea water. Therefore the operating conditions of pipelines may fall into a hydrate stability region. We note that there would be an abundance of hydrate precursors after the dissociation of methane hydrate in the
Y. Seo, S.-P. Kang / Journal of Petroleum Science and Engineering 88–89 (2012) 61–66
2. Materials and experimental procedures 2.1. Materials Methane gas with a stated purity of 99.9 mol% and deionized water with a purity of 99.99 mol% were used for hydrate formation. Among numerous KHI candidates, the KHI used in this work was PVCap (poly(N-vinylcaprolactam)) with a purity of 98.0 mol% and a molecular weight of 5,000. Sodium chloride (NaCl) with ≥99.0% was purchased from Aldrich. 2.2. Apparatus The apparatus for hydrate formation was designed to monitor the status of the fluids via a high-resolution video camera (Canon, VIXIA HF10) while measuring the pressure and temperature. A highpressure cell made of 316 stainless steel was equipped with two thermally reinforced sight glasses for visual observation. It has an internal volume of 300 cm3 and a working pressure of 12 MPa. The fluids inside the cell were agitated by a magnetic spin bar at 700 rpm that was coupled with a magnet placed under the cell. The cell was located inside an air bath (ETAC, FL414P) where the temperature was controlled by a refrigerator/heater with an accuracy of ±0.1 K. A K-type thermocouple probe with a digital thermometer was inserted into the cell to measure the temperature of the fluids within an uncertainty of ±0.1 K. The pressure of the system was measured by a pressure transducer with an uncertainty of ±0.01 bar in a range of 0 to 12 MPa. 2.3. Procedures Fig. 1 shows the three main stages of a typical experiment. Methane hydrate was formed in the first hydrate formation stage. The high-pressure cell containing liquid water or the PVCap aqueous solution was placed inside the air bath and the formation temperature was maintained at 2 °C. When the cell temperature became stable, methane gas was supplied from a high-pressure cylinder into the cell until the pressure reached 95 bar while stirring the liquid water. Hydrate formation was then induced in the liquid water which resulted in a decrease of the pressure. Make-up gas was supplied to
1st hydrate formation
Hydrate dissociation
2nd hydrate formation
100 Pdisso
Pressure (bar)
water phase, which increases the risk of hydrate re-formation due to memory effect of hydrate precursors (Sloan and Koh, 2008), thus requiring careful management to avoid the formation of hydrate plug in pipelines. One viable option to avoid hydrate re-formation is the injection of a hydrate inhibitor. Although a MEG injection is considered to be the standard method in an offshore gas production system, Kinetic Hydrate Inhibitor (KHI) is also becoming popular as its dosage rate is expected in the range of 0.5–3 wt.%, which is much lower than the 30–60 wt.% dosage of MEG. A KHI is a water-soluble polymer that acts by delaying the initial hydrate nucleation. However, the applicability of KHI to methane hydrate development, especially in the presence of hydrate precursors in the water phase, has not been studied thoroughly. There have been attempts to develop a KHI evaluation method using hydrate precursors. (Duchateau et al., 2009, 2010) However, the hydrate-forming gas was a binary mixture of methane and propane that forms structure II, while methane gas from marine hydrate deposits forms structure I. Most KHIs were developed to inhibit the formation of structure II and it is necessary to investigate the applicability of KHI to inhibit the formation of structure I before applying them to the production of methane gas from marine hydrate deposits. In this work, we conduct experiments to investigate methane hydrate re-formation in the presence of hydrate precursors and the effects of KHI on its inhibition. The risk of hydrate re-formation in offshore pipelines while transporting methane and dissociated water is also discussed.
80 60
Supply make-up gas
Ponset
Discharging dissociated gas
40 20 0 20
Temperature (oC)
62
15 Teq
10 Tonset
5 0
Time
tind
Pressure Temperature Pressure Temperature increase increase decrease decrease Fig. 1. Experimental procedures.
maintain the pressure at around 95 bar. After the hydrate formation process was completed, the temperature was increased to the dissociation temperature of 12 °C at a rate of 0.5 K·h − 1. When NaCl solution is used to investigate the effect of salt ions on the hydrate reformation, the formation temperature was 0.5 °C to take account the shift of hydrate equilibrium conditions in the presence of NaCl and then the temperature was increased to the dissociation temperature of 10.5 °C at a rate of 0.5 K·h − 1. When the cell temperature reached the dissociation temperature and there was no indication of pressure fluctuation, hydrate dissociation was initiated by depressurization while maintaining the temperature. The first dissociation pressure was set to 80 bar. As the dissociation of the hydrate induced a pressure increase, it had to be controlled very carefully in order to maintain the driving force for hydrate dissociation. Fig. 2 shows the obtained images during the experimental stages. As shown in Fig. 2 (a), it was considered that the dissociation of the hydrate phase to be completed when no hydrate crystals could be observed and the pressure remained at the hydrate dissociation pressure, Pdisso. Subsequently, the liquid phase was agitated and the cell temperature was decreased from the dissociation to formation temperature within 40 min, as shown in Fig. 2 (b), to simulate the cool-down of the dissociated water and methane gas in pipelines. When the cooling down process started, monitoring of the fluids inside the high-pressure cell with a video camera commenced. The hydrate onset temperature could be determined by the observation of white hydrate crystals in the liquid phase, as shown in Fig. 2 (c). The hydrate onset temperature, Tonset, was lower than the hydrate equilibrium temperature, Teq, and the subcooling defined by ΔT = Teq − Tonset, was calculated to evaluate the risk of hydrate formation under the corresponding operation conditions. As shown in Fig. 2(d) the liquid phase started to be converted to hydrate slurry followed by solid hydrate chunks after Tonset. The induction time was defined as the difference between the hydrate onset and the time when the temperature became lower than the hydrate equilibrium temperature. The dissociation pressures were varied from 80 to 60 bar to simulate various drawdown scenarios. Two different cooling rates of 0.25 and 0.08 °C/min were used in order to observe the effect
Y. Seo, S.-P. Kang / Journal of Petroleum Science and Engineering 88–89 (2012) 61–66
(a)
(b)
(c)
(d)
63
Hydrate onset
Fig. 2. Images of the hydrate onset conditions. (a) The complete dissociation of hydrate from the first formation stage, (b) stirring of liquid phase during the cool down, (c) hydrate onset condition, (d) hydrate slurry followed by solid hydrate chunks.
of the cooling rate on the hydrate onset conditions. It was noted that the PVCap was injected after dissociation of the hydrate and before the second hydrate formation stage using a high pressure pump to simulate the injection of the KHI into the production choke and mixing with the production fluids of methane and the dissociated water. 3. Results and discussion Fig. 3 shows a pressure and temperature (P–T) trace plot used to determine the hydrate onset conditions for fresh water at three different dissociation pressures of 80, 70, and 60 bar. The purpose of this experiment was to represent the hydrate formation characteristics in a system containing fresh water, where fresh water indicates water that has never experienced hydrate formation. While decreasing the temperature at a rate of 0.25 °C/min, a slight decrease of the
100 90
Pressure (bar)
Hydrate onset 80 70 60
pressure was observed due to increased gas solubility and density at lower temperatures. The pressure decreased precipitously upon the hydrate onset and growth, which was observed by the appearance of white hydrate particles floating in the liquid phase. As shown in Fig. 3, the hydrate onset temperature, Tonset, for fresh water was determined to be 4.2 °C at 76.1 bar when the dissociation pressure, Pdisso, was 80 bar. As the hydrate equilibrium temperature, Teq, is predicted to be 10.5 °C at 76.1 bar from the thermodynamic software CSMGEM, the subcooling, ΔT, is 6.3 °C according to the definition of ΔT = Teq − Tonset. The induction time, tind, for hydrate onset at 80 bar was 25.0 min. The decrease of the dissociation pressure to 70 bar induces a decrease of both the subcooling and the induction time. They were further decreased when the dissociation pressure was set to 60 bar. The resulting hydrate onset conditions, subcooling temperatures and induction times for fresh water are summarized in Table 1. It was suggested that hydrate crystallization in system containing fresh water leads to scattered results (Duchateau et al., 2009), which is due to the stochastic nature of hydrate nucleation. The hydrate onset conditions shown in Fig. 3 represent these phenomena. Each hydrate onset condition at different dissociation pressure averages at least three measurements. By reducing the dissociation pressure from 80 to 60 bar, the subcooling changes from 6.3 to 1.9 °C, resulting a relative dispersion of about 60% with a mean value of 3.9 °C. Therefore it makes difficult to evaluate the hydrate formation
Table 1 The hydrate onset conditions for fresh water at different initial pressures and cooling rates.
50 40 0
2
4
6
8
Temperature
10
12
14
(oC)
Fig. 3. The hydrate onset conditions for fresh water at dissociation pressures of (a) 80 bar (▼), (b) 70 bar (■), and (c) 60 bar (●). Filled symbol indicates the hydrate onset condition for each dissociation pressure.
Cooling rate
Pdisso
Hydrate onset conditions
Subcooling
Induction time
(°C/min)
(bar)
Tonset (°C)
Ponset (bar)
(°C)
(min)
0.25
80 70 60 80
4.2 5.7 5.8 5.9
76.1 67.0 57.1 77.4
6.3 3.6 1.9 4.7
25.0 14.3 7.9 59.0
0.08
64
Y. Seo, S.-P. Kang / Journal of Petroleum Science and Engineering 88–89 (2012) 61–66
risk for a system containing fresh water and leads to complete avoidance of hydrate in the operation of conventional gas fields. However, the hydrate precursors or residual structures stemming from the dissociation of the hydrate phase enhance the hydrate formation, which results in a decrease of the subcooling. This is described as a memory effect among hydrate researchers, as hydrates are assumed to retain a memory of their structure – the hydrate precursors – when experiencing dissociation. Hydrate forms more easily from residual structures obtained from the dissociation of hydrate plugs than it does from fresh water with no previous hydrate history (Kumar et al., 2008; Sloan and Koh, 2008). It has been suggested that the memory effect can be avoided at temperatures greater than 28 °C or after several hours of heating (Sloan and Koh, 2008). However, such actions to avoid the memory effect are difficult to be carried out during the production of methane gas from marine hydrate deposits. It is considered that, after the dissociation of the hydrate, fluids will cool rapidly due to a heat exchange with the environment, especially when transported via offshore pipelines. In order to assure the fluids flow inside the pipelines, it is necessary to assess the risk of methane hydrate re-formation for the dissociated water where dissociated water indicates that the water that has experienced hydrate formation and dissociation. Fig. 4 shows a P–T trace plot during the cooling process of the dissociated water; the filled symbol for each trace indicates the hydrate onset condition at the given dissociation pressure. This clearly shows that the onset of hydrate formation from dissociated water occurs more rapidly. As shown in Fig. 4, when the dissociation pressure was 80 bar the hydrate onset was detected at 9.6 °C and 78.2 bar, which results in subcooling of 1.3 °C from the prediction of Teq = 10.8 °C. Rapid reformation of methane hydrate was observed in each experiment, and the obtained subcooling, as shown in Table 2, ranged from 1.2 and 1.6 °C for the dissociation pressures. The relative dispersion was about 14% from a mean value of 1.4 °C. This indicates that the hydrate would re-form just 1.4 °C below the hydrate equilibrium conditions during the cooling down of the dissociated water. The stochastic nature of hydrate nucleation was not observed in the dissociated water. Compared to the subcooling of 6.3 °C for fresh water at the dissociation pressure of 80 bar, the subcooling of 1.2 °C for dissociated water clearly indicates a high risk of hydrate plug formation during the transportation of methane from marine hydrate deposits. There are two hypotheses related to hydrate nucleation. One is the formation of a labile cluster, presuming that liquid water molecules are arranged around a dissolved solute molecule in a prehydrate
100
Pressure (bar)
90 80 70 60 50 40 0
2
4
6
8
10
12
14
Temperature (oC) Fig. 4. The hydrate onset conditions for dissociated water at dissociation pressures of (a) 80 bar (▼), (b) 70 bar (■), and (c) 60 bar (●). Filled symbol indicates the hydrate onset condition for each dissociation pressure.
Table 2 The hydrate onset conditions for dissociated water at different initial pressures and cooling rates. Cooling rate
Pdisso
Hydrate onset conditions
Subcooling
Induction time
(°C/min)
(bar)
Tonset (°C)
Ponset (bar)
(°C)
(min)
0.25
80 70 60 80
9.6 8.3 6.4 9.9
78.2 68.5 58.3 79.2
1.3 1.2 1.6 0.9
5.1 4.9 6.4 11.6
0.08
structure, with essentially the correct coordination number. The other is local structuring, where the prehydrate structure consists of a locally ordered water-guest structure rather than individual hydrate cavities. In both hypotheses, hydrate nucleation starts from the formation of the prehydrate structure. It is envisaged that the residual hydrate structures in the water phase act as templates to form these prehydrate structure and induce faster hydrate reformation, as shown in Fig. 4. When methane hydrate deposits are dissociated by depressurization, it is important to establish an optimized dissociation pressure. Although it is possible to obtain a fast dissociation rate by applying a large pressure difference between the dissociation pressure and the hydrate equilibrium pressure, this may cause a large temperature drop in the hydrate core due to endothermic hydrate dissociation, resulting in hydrate reformation in the hydrate well casing. Accordingly, it is suggested that the dissociation pressure should be close to the hydrate equilibrium condition, where it is almost impossible to avoid the presence of residual hydrate structures after the dissociation process. As confirmed in above experiments, the hydrate precursors will result in rapid methane hydrate reformation during the transportation of methane and dissociated water through offshore pipelines once the temperature of fluids falls into hydrate stability zone. An active hydrate mitigation strategy must be employed in order to avoid hydrate blockage in offshore pipelines, which could potentially result in a costly production stoppage. The pipelines can be insulated to ensure that the fluid temperature remains above the hydrate equilibrium temperature. Otherwise, the continuous injection of a chemical, such as mono ethylene glycol (MEG) or methanol can be employed to shift the hydrate equilibrium curve outside the operation conditions. However, all of these strategies could have an enormous impact on the production economics. Among the hydrate mitigation strategies, the use of a kinetic hydrate inhibitor requires a lower injection rate than the use of MEG or methanol. Moreover, it has shown reliable performance under subcooling conditions below 12 °C (Clark and Anderson, 2007). The application of a KHI as a possible hydrate mitigation strategy was investigated for dissociated water. PVCap was selected as the KHI in this work. Fig. 5 shows the effect of the addition of PVCap on the hydrate onset conditions for dissociated water. Table 3 presents the hydrate onset conditions along with the calculation of the subcooling and induction time with different PVCap concentrations. The concentration of PVCap was varied from 0.5 to 3.0 wt.% while the cooling process was started from 12 °C at 80 bar for all concentrations at a rate of 0.25 °C/min. The hydrate onset temperature was 5.9 °C for the 0.5 wt.% PVCap and became 4.5 °C for the 1.0 wt.% PVCap. It reached 2.5 °C with the addition of 3.0 wt.% PVCap. This indicates that the methane and dissociated water can be transported without hydrate formation for 31 min until the fluid temperature reaches 2.5 °C by adding 3.0 wt.% of PVCap. While the subcooling was 1.2 °C for the dissociated water without PVCap, it increased to 4.7, 6.0, and 7.8 °C at PVCap concentrations of 0.5, 1.0, and 3.0 wt.%, respectively. These results clearly represent the inhibition effect of PVCap on the hydrate formation.
Y. Seo, S.-P. Kang / Journal of Petroleum Science and Engineering 88–89 (2012) 61–66
Table 4 The hydrate onset conditions in the presence of sodium chloride at different PVCap concentrations at Pdisso = 80 bar and a cooling rate of 0.25 °C/min.
86 84
NaCl
PVCap
Hydrate onset conditions
Subcooling
Induction time
(wt.%)
(wt.%)
Tonset (°C)
Ponset (bar)
(°C)
(min)
3.5
0.0 0.0 0.5 1.0 3.0
2.9 7.5 2.1 1.5 1.1
76.9 78.7 76.7 75.9 74.9
6.2 1.8 6.9 7.4 8.8
24.8 7.3 27.8 29.7 35.3
82 No PVCap
Pressure (bar)
65
80
PVCap 0.5wt% PVCap 1.0wt%
78
PVCap 3.0wt%
76 74
2
4
6
8
10
12
14
Temperature (oC) Fig. 5. Effect of the addition of PVCap on the hydrate onset conditions at dissociation pressure of 80 bar for dissociated water. The concentration of PVCap was 0.0 (▼), 0.5 (■), 1.0 (●), and 3.0 (♦) wt.%.
The dissociated water from marine hydrate deposits may contain salts. In order to investigate the effect of salt ions on the hydrate reformation as well as the performance of KHI, synthetic brine with 3.5 wt.% NaCl was used to form and dissociate hydrate. Methane hydrate was formed at 0.5 °C and 95 bar, then the dissociation was carried out at 10.5 °C and 80 bar to take account the shift of hydrate equilibrium conditions in the presence of NaCl. Table 4 summarizes the hydrate onset conditions with and without PVCap. The hydrate dissociation pressure was fixed at 80 bar and the cooling rate from 10.5 to 0.5 °C was 0.25 °C/min. When fresh 3.5 wt.% NaCl solution with no hydrate history was used for the hydrate onset experiment, the hydrate onset was detected at 2.9 °C and 76.9 bar, which resulted the similar subcooling of 6.2 °C to that of fresh water, 6.3 °C, at the dissociated pressure of 80 bar in Table 1. Interestingly, when the NaCl solution experienced hydrate formation and dissociation is used for the hydrate onset experiment, the subcooling becomes 1.8 °C, which is similar to 1.3 °C for dissociated water at the dissociation pressure of 80 bar in Table 2. These results indicate the presence of NaCl in water phase doesn't affect seriously the hydrate onset conditions whether the water phase experienced hydrate formation or not. It is also noted that the reliable data for hydrate onset conditions can be obtained once the same cooling down process is applied to measure the hydrate onset conditions as studied in this study. Further experiments were carried out to investigate the hydrate onset conditions of dissociated NaCl solution after injecting PVCap. As shown in Table 4, the subcooling increased substantially by injecting PVCap into dissociated NaCl solution. At the concentration of PVCap of 0.5 wt.%, the subcooling was 4.7 °C in dissociated water without NaCl, but increased to 6.9 °C in dissociated NaCl solution. Considering the increase of subcooling in the presence of NaCl for both 1.0 and 3.0 wt.% of PVCap concentrations, there might be a synergistic inhibition effect of NaCl ions on the performance of PVCap.
Table 3 The hydrate onset conditions for dissociated water at different PVCap concentrations at Pdisso = 80 bar. Cooling rate
PVCap
Hydrate onset conditions
Subcooling
Induction time
Low Risk
0
40
Hydrate onset time (min)
70
No Yes Yes Yes Yes
Fig. 6 represents the risk of hydrate formation depicted in subcooling vs. hydrate onset time diagram. As indicated in the axis, low subcooling and short hydrate onset time represent the high risk of hydrate formation for a given fluids, whereas high subcooling and long hydrate onset time indicate the low risk of hydrate formation. For fresh water and NaCl solution without hydrate history, the risk of hydrate formation would be considered to be medium. However, once the water experiences the formation and dissociation of hydrates, the risk moves into high risk region as shown in the subcooling range of 1.2 and 1.8 °C. The addition of PVCap into the dissociated water reduces the risk of hydrate formation by moving subcooling range into 4.7 and 7.8 °C. The presence of NaCl ions further reduces the risk into subcooling range of 6.9 and 8.8 °C clearly showing the benefit of injecting PVCap. The injection of a KHI such as PVCap is a possible candidate to avoid hydrate reformation while transporting the fluids to the processing facilities. The inhibition mechanism of KHI remains not fully understood. One theory holds that the KHI polymer has functional groups that operate by distorting the hydrate structure and increasing the energy barrier needed to form a hydrate building unit that would grow into hydrate crystals (Sloan and Koh, 2008). If this theory is applicable to the formation of hydrate from residual structures, the obtained results indicate that PVCap is able to suppress the formation of hydrate building units from dissociated water containing residual hydrate structures. The residual hydrate structures may be adsorbed onto the functional groups of PVCap and prevented from further growth into the critical hydrate nuclei. Further studies are necessary to explore the inhibition mechanism of KHI in the presence of residual hydrate structures and different concentrations of salt ions.
High Risk
72
Hydrate history
30
Injecting PVCap to dissociated water and NaCl solution
20
Fresh water
10
Dissociated water
0 0
2
4
6
8
10
Subcooling (oC)
(°C/min)
(wt.%)
Tonset (°C)
Ponset (bar)
(°C)
(min)
0.25
0.5 1.0 3.0
5.9 4.5 2.5
76.9 76.1 74.9
4.7 6.0 7.8
18.8 24.0 31.2
High Risk
Low Risk
Fig. 6. Comparison of hydrate onset conditions in the presence of PVCap and sodium chloride. (○) Dissociated water, (●) Dissociated 3.5 wt.% NaCl solution, (∇) Fresh water, (▼) Fresh 3.5 wt.% NaCl solution, (□) PVCap injected dissociated water without NaCl, (■) PVCap injected dissociated 3.5 wt.% NaCl solution.
66
Y. Seo, S.-P. Kang / Journal of Petroleum Science and Engineering 88–89 (2012) 61–66
Methane production from marine hydrates will be different from conventional offshore gas fields in terms of water production and related flow assurance issues. Water production in gas fields can be described via the water:gas ratios, which are typically less than 10 bbl/ MMscf including condensed water and formation water production (Walsh et al., 2009). However the water production from marine hydrates can vary greatly depending on the hydrate reservoir type. Moreover, it has been suggested that water:gas ratios in excess of 1000 bbl/MMscf are possible (Walsh et al., 2009). It was noted that gas hydrates are composed of 15 wt.% of methane and 85 wt.% of water from the stoichiometry of structure I, 8CH4 · 46H2O. Moreover the presence of residual hydrate structures in dissociated water will increase the risk of methane hydrate reformation resulting hydrate plug when transporting produced fluids via offshore pipelines. If we consider the concentration of MEG to avoid hydrate formation when operating offshore pipelines at 80 bar and minimum temperature of 4 °C, the required MEG concentration will be 0.09 bbl MEG/bbl water from thermodynamic prediction, which demands the MEG distribution system and regeneration unit to handle 90 bbl MEG/MMscf based on a water:gas ratio of 1000 bbl/MMscf. From the scenario of producing methane from class 2 deposits involving a hydrate layer over a mobile water zone (Walsh et al., 2009), the maximum gas production rate would reach 70 MMscf/day, thus, the maximum MEG injection rate would be 6300 bbl/day. Considering that MEG regeneration unit of the Asgard B field for Statoil producing 840 MMscf/day gas and 220,000 bbl/day oil is handling about 6800 bbl of rich MEG flow (Brustad et al., 2005), MEG handling for marine hydrate is expected to be more costly compared to that in conventional offshore gas fields. Accordingly the associated operational expenditure and the handling of large quantities of MEG can make the KHI cost-effective hydrate mitigation option for marine hydrate development. From the above calculation of the water production of 70,000 bbl/day, the daily required PVCap would be around 4900 kg if the PVCap concentration of 3 wt.% was applied, which is much less than 6300 bbl of MEG. Moreover once it is confirmed that the PVCap has synergistic effect in the presence of salt ions, the dosage rate of PVCap would be able to be further reduced to save operational expenditure. The performance of KHI has to be evaluated carefully before its application in terms of the lowest operation temperature of offshore pipelines and the residence time of the produced fluids in the hydrate formation region. Although more comprehensive studies have to be designed based on a field development plan to apply KHI to methane hydrate production, the results obtained in this work suggest that a cost-effective hydrate prevention strategy will be vital for the success of methane hydrate production, as the risk of methane hydrate re-formation in dissociated water is much higher than in fresh water. 4. Conclusions Hydrate formation in fresh water leads to a subcooling of 6.3 °C at a dissociation pressure of 80 bar, while hydrate reformation in dissociated water, which has previously experienced hydrate formation and dissociation was outstandingly fast, resulting in a subcooling of 1.3 °C. The dissociation pressures after the first hydrate formation
stage varied from 80 to 60 bar; however, under each dissociation pressure, rapid hydrate reformation was observed during the cooling of dissociated water. Although the presence of residual hydrate structures enhances the nucleation of hydrate building units, the polymer structure of PVCap can suppress the nucleation process. An increase of the PVCap concentration from 0.5 to 3.0 wt.% resulted in an increase of subcooling to 7.8 °C. This is even higher than the subcooling of fresh water and suggests that the addition of PVCap will be an efficient hydrate mitigation strategy to avoid hydrate reformation during the transportation of methane and dissociated water through offshore pipelines. Moreover it is observed that the subcooling increased to 8.8 °C at the PVCap concentration of 3.0 wt.% in the presence of NaCl in the water phase suggesting that there might be synergistic effect of salt ions on the performance of PVCap as a KHI. Acknowledgements This work was supported by the Energy Efficiency & Resources Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Knowledge Economy (No. 2010201030001A), and partially supported by the Australian Academy of Science-NRF Exchange Program. References Brustad, S., Loken, K.P., Waalmann, J.G., 2005. Hydrate prevention using MEG instead of MeOH: Impact of experience from major Norwegian developments on technology selection for injection and recovery of MEG. Offshore Technology Conference, 2–5 May, 2005. Houston, Texas, U.S.A. Clark, L.W., Anderson, J., 2007. Low Dosage Hydrate Inhibitors (LDHI): Further advances and developments in flow assurance technology and applications concerning oil and gas production systems. International Petroleum Technology Conference, 4–6 December, 2007. Dubai, U.A.E. Duchateau, C., Peytavy, J., Glenat, P., Pou, T., Hidalgo, M., Dicharry, C., 2009. Laboratory evaluation of kinetic hydrate inhibitors: a procedure for enhancing the repeatability of test results. Energy Fuel 23, 962–966. Duchateau, C., Glenat, P., Pou, T., Hidalgo, M., Dicharry, C., 2010. Hydrate precursor test method for the laboratory evaluation of kinetic hydrate inhibitors. Energy Fuel 24, 616–623. Konno, Y., Masuda, Y., Hariguchi, Y., Kurihara, M., Ouchi, H., 2010. Key factors for depressurization-induced gas production from oceanic methane hydrates. Energy Fuel 24, 1736–1744. Kumar, R., Lee, J., Song, M., Englezos, P., 2008. Kinetic inhibitor effects on methane/propane clathrate hydrate-crystal growth at the gas/water and water/n-heptane interfaces. J. Cryst. Growth 310, 1154–1166. Kurihara, M., Sato, A., Ouchi, H., Narita, H., Masuda, Y., Saeki, T., Fujii, T., 2008a. Prediction of gas productivity from Eastern Nankai Trough methane-hydrate reservoirs. Offshore Technology Conference, Houston, Texas, U.S.A., OTC 19382. Kurihara, M., Funatsu, K., Ouchi, H., Masuda, Y., Yamamoto, K., Narita, H., Dallimore, S.R., Collett, T., Hancock, S.H., 2008b. Analysis of production tests and MDT tests conducted in Mallik and Alaska methane hydrate reservoirs: what can we learn from these well tests? Proceedings from the 6th International Conference on Gas Hydrates, July 6–10, 2008. Vancouver, British Columbia, Canada. Oyama, H., Konno, Y., Masuda, Y., Narita, H., 2009. Dependence of depressurizationinduced dissociation of methane hydrate bearing laboratory cores on heat transfer. Energy Fuel 23, 4995–5002. Sloan, E.D., Koh, C.A., 2008. Clathrate hydrates of natural gases, Third edition. CRC Press, 6000 Broken Sound Parkway NW, suite 300, Boca Raton, FL, U.S.A. Walsh, M.R., Hancock, S.H., Wilson, S.J., Patil, S.L., Moridis, G.J., Boswell, R., Collett, T.S., Koh, C.A., Sloan, E.D., 2009. Preliminary report on the commercial viability of gas production from natural gas hydrates. Energy Econ. 31, 815–823. Yamamoto, K., 2007. Introduction of the 2007–2008 JOGMEC/NRCan/Aurora Mallik gas hydrate production research program, NWT, Canada. AGU Fall Meeting.
Contents lists available at SciVerse ScienceDirect
Journal of Petroleum Science and Engineering journal homepage: www.elsevier.com/locate/petrol
Inhibition of methane hydrate re-formation in offshore pipelines with a kinetic hydrate inhibitor Yutaek Seo a, Seong-Pil Kang b,⁎ a b
CSIRO Petroleum Resources, 26 Dick Perry Avenue, Kensington, Perth, WA 6151, Australia Climate Change Research Division, Korea Institute of Energy Research, 71-2 Jang-dong, Yuseong-gu, Daejeon 305-343, Republic of Korea
a r t i c l e
i n f o
Article history: Received 31 May 2011 Revised 6 October 2011 Accepted 17 November 2011 Available online 1 December 2011 Keywords: gas hydrate methane production kinetic hydrate inhibitor PVCap flow assurance
a b s t r a c t Methane gas from marine hydrate deposits can be produced by one or a combination of three methods; depressurization, thermal stimulation, and the injection of hydrate inhibitors. Because residual hydrate structures known as hydrate precursors will exist in the liquid water phase after dissociation, the risk of methane hydrate re-formation has to be evaluated during the production and transportation of methane gas through offshore pipelines. New experimental procedures composed of three stages are designed to simulate the dissociation of marine hydrates and the transportation of produced fluids. The obtained experimental results have shown that methane hydrates are rapidly re-formed when the temperature of dissociated water falls into the hydrate formation region during the cooling down process. The subcooling for three different dissociation pressures of 80, 70, and 60 bar were 1.3, 1.2 and 1.6 °C, respectively. One viable option to avoid hydrate re-formation is injecting hydrate inhibitors before transporting the fluids through pipelines. Among various hydrate inhibitors, Poly(N-vinylcaprolactam) (PVCap) was selected as a possible candidate for a Kinetic Hydrate Inhibitor (KHI) and injected into dissociated water before cooling down the fluids. The concentration of PVCap was varied from 0.5 to 3.0 wt.%. With an increase in the PVCap concentration, the subcooling increased to 7.8 °C at a dissociation pressure of 80 bar, which suggests that the risk of hydrate re-formation can be reduced by injecting PVCap. Moreover it is observed that the subcooling increased to 8.8 °C at the PVCap concentration of 3.0 wt.% in the presence of NaCl in the water phase. Although the use of KHI in conventional gas production has become common, its applicability to methane hydrate production has not yet been studied thoroughly, especially in the presence of residual hydrate structures. In this work, the application of KHI to methane hydrate production is discussed. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Gas hydrates are crystalline compounds that form when guest molecules are incorporated in host cages formed by water molecules through hydrogen bonding (Sloan and Koh, 2008). Low molecular weight gas molecules such as methane and carbon dioxide are captured in these cages, and each hydrate lattice consists of at least two types of polyhedral cages. Three distinct structural families, termed structures I, II, and H, are known, and they show distinct structural characteristics in terms of their cage types, distribution of guest molecules, and thermodynamic equilibrium conditions. Recently, gas hydrates have attracted a great deal of attention because of their significant potential as a source of methane gas. There are massive hydrate deposits both under the permafrost and in sediment on the continental margins. One issue that has been discussed within the hydrate research community is possible technologies to
⁎ Corresponding author. Tel.: + 82 42 860 3475; fax: + 82 42 860 3134. E-mail address: [email protected] (S.-P. Kang). 0920-4105/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.petrol.2011.11.001
develop gas hydrate deposits. Concepts for methane production from hydrate deposits proposed to date involve one or a combination of the following three methods: depressurization, thermal stimulation, and the use of hydrate inhibitor injections. Among them, depressurization is considered to be the most promising method because the highest energy profit ratio can be achieved (Kurihara et al., 2008a). Following the successful production from hydrate deposits in the Mackenzie Delta (Kurihara et al., 2008b; Yamamoto, 2007), there have been considerable efforts to determine the optimum production conditions for marine hydrate deposits as well (Konno et al., 2010; Oyama et al., 2009). However, most of these works focused on the characteristics of hydrate dissociation under various production scenarios, and there has been little effort to understand the transportation of dissociated methane gas and water via pipelines from hydrate wells to processing facilities. When the produced fluids are transported through offshore pipelines, they cool down due to a heat exchange with sea water. Therefore the operating conditions of pipelines may fall into a hydrate stability region. We note that there would be an abundance of hydrate precursors after the dissociation of methane hydrate in the
Y. Seo, S.-P. Kang / Journal of Petroleum Science and Engineering 88–89 (2012) 61–66
2. Materials and experimental procedures 2.1. Materials Methane gas with a stated purity of 99.9 mol% and deionized water with a purity of 99.99 mol% were used for hydrate formation. Among numerous KHI candidates, the KHI used in this work was PVCap (poly(N-vinylcaprolactam)) with a purity of 98.0 mol% and a molecular weight of 5,000. Sodium chloride (NaCl) with ≥99.0% was purchased from Aldrich. 2.2. Apparatus The apparatus for hydrate formation was designed to monitor the status of the fluids via a high-resolution video camera (Canon, VIXIA HF10) while measuring the pressure and temperature. A highpressure cell made of 316 stainless steel was equipped with two thermally reinforced sight glasses for visual observation. It has an internal volume of 300 cm3 and a working pressure of 12 MPa. The fluids inside the cell were agitated by a magnetic spin bar at 700 rpm that was coupled with a magnet placed under the cell. The cell was located inside an air bath (ETAC, FL414P) where the temperature was controlled by a refrigerator/heater with an accuracy of ±0.1 K. A K-type thermocouple probe with a digital thermometer was inserted into the cell to measure the temperature of the fluids within an uncertainty of ±0.1 K. The pressure of the system was measured by a pressure transducer with an uncertainty of ±0.01 bar in a range of 0 to 12 MPa. 2.3. Procedures Fig. 1 shows the three main stages of a typical experiment. Methane hydrate was formed in the first hydrate formation stage. The high-pressure cell containing liquid water or the PVCap aqueous solution was placed inside the air bath and the formation temperature was maintained at 2 °C. When the cell temperature became stable, methane gas was supplied from a high-pressure cylinder into the cell until the pressure reached 95 bar while stirring the liquid water. Hydrate formation was then induced in the liquid water which resulted in a decrease of the pressure. Make-up gas was supplied to
1st hydrate formation
Hydrate dissociation
2nd hydrate formation
100 Pdisso
Pressure (bar)
water phase, which increases the risk of hydrate re-formation due to memory effect of hydrate precursors (Sloan and Koh, 2008), thus requiring careful management to avoid the formation of hydrate plug in pipelines. One viable option to avoid hydrate re-formation is the injection of a hydrate inhibitor. Although a MEG injection is considered to be the standard method in an offshore gas production system, Kinetic Hydrate Inhibitor (KHI) is also becoming popular as its dosage rate is expected in the range of 0.5–3 wt.%, which is much lower than the 30–60 wt.% dosage of MEG. A KHI is a water-soluble polymer that acts by delaying the initial hydrate nucleation. However, the applicability of KHI to methane hydrate development, especially in the presence of hydrate precursors in the water phase, has not been studied thoroughly. There have been attempts to develop a KHI evaluation method using hydrate precursors. (Duchateau et al., 2009, 2010) However, the hydrate-forming gas was a binary mixture of methane and propane that forms structure II, while methane gas from marine hydrate deposits forms structure I. Most KHIs were developed to inhibit the formation of structure II and it is necessary to investigate the applicability of KHI to inhibit the formation of structure I before applying them to the production of methane gas from marine hydrate deposits. In this work, we conduct experiments to investigate methane hydrate re-formation in the presence of hydrate precursors and the effects of KHI on its inhibition. The risk of hydrate re-formation in offshore pipelines while transporting methane and dissociated water is also discussed.
80 60
Supply make-up gas
Ponset
Discharging dissociated gas
40 20 0 20
Temperature (oC)
62
15 Teq
10 Tonset
5 0
Time
tind
Pressure Temperature Pressure Temperature increase increase decrease decrease Fig. 1. Experimental procedures.
maintain the pressure at around 95 bar. After the hydrate formation process was completed, the temperature was increased to the dissociation temperature of 12 °C at a rate of 0.5 K·h − 1. When NaCl solution is used to investigate the effect of salt ions on the hydrate reformation, the formation temperature was 0.5 °C to take account the shift of hydrate equilibrium conditions in the presence of NaCl and then the temperature was increased to the dissociation temperature of 10.5 °C at a rate of 0.5 K·h − 1. When the cell temperature reached the dissociation temperature and there was no indication of pressure fluctuation, hydrate dissociation was initiated by depressurization while maintaining the temperature. The first dissociation pressure was set to 80 bar. As the dissociation of the hydrate induced a pressure increase, it had to be controlled very carefully in order to maintain the driving force for hydrate dissociation. Fig. 2 shows the obtained images during the experimental stages. As shown in Fig. 2 (a), it was considered that the dissociation of the hydrate phase to be completed when no hydrate crystals could be observed and the pressure remained at the hydrate dissociation pressure, Pdisso. Subsequently, the liquid phase was agitated and the cell temperature was decreased from the dissociation to formation temperature within 40 min, as shown in Fig. 2 (b), to simulate the cool-down of the dissociated water and methane gas in pipelines. When the cooling down process started, monitoring of the fluids inside the high-pressure cell with a video camera commenced. The hydrate onset temperature could be determined by the observation of white hydrate crystals in the liquid phase, as shown in Fig. 2 (c). The hydrate onset temperature, Tonset, was lower than the hydrate equilibrium temperature, Teq, and the subcooling defined by ΔT = Teq − Tonset, was calculated to evaluate the risk of hydrate formation under the corresponding operation conditions. As shown in Fig. 2(d) the liquid phase started to be converted to hydrate slurry followed by solid hydrate chunks after Tonset. The induction time was defined as the difference between the hydrate onset and the time when the temperature became lower than the hydrate equilibrium temperature. The dissociation pressures were varied from 80 to 60 bar to simulate various drawdown scenarios. Two different cooling rates of 0.25 and 0.08 °C/min were used in order to observe the effect
Y. Seo, S.-P. Kang / Journal of Petroleum Science and Engineering 88–89 (2012) 61–66
(a)
(b)
(c)
(d)
63
Hydrate onset
Fig. 2. Images of the hydrate onset conditions. (a) The complete dissociation of hydrate from the first formation stage, (b) stirring of liquid phase during the cool down, (c) hydrate onset condition, (d) hydrate slurry followed by solid hydrate chunks.
of the cooling rate on the hydrate onset conditions. It was noted that the PVCap was injected after dissociation of the hydrate and before the second hydrate formation stage using a high pressure pump to simulate the injection of the KHI into the production choke and mixing with the production fluids of methane and the dissociated water. 3. Results and discussion Fig. 3 shows a pressure and temperature (P–T) trace plot used to determine the hydrate onset conditions for fresh water at three different dissociation pressures of 80, 70, and 60 bar. The purpose of this experiment was to represent the hydrate formation characteristics in a system containing fresh water, where fresh water indicates water that has never experienced hydrate formation. While decreasing the temperature at a rate of 0.25 °C/min, a slight decrease of the
100 90
Pressure (bar)
Hydrate onset 80 70 60
pressure was observed due to increased gas solubility and density at lower temperatures. The pressure decreased precipitously upon the hydrate onset and growth, which was observed by the appearance of white hydrate particles floating in the liquid phase. As shown in Fig. 3, the hydrate onset temperature, Tonset, for fresh water was determined to be 4.2 °C at 76.1 bar when the dissociation pressure, Pdisso, was 80 bar. As the hydrate equilibrium temperature, Teq, is predicted to be 10.5 °C at 76.1 bar from the thermodynamic software CSMGEM, the subcooling, ΔT, is 6.3 °C according to the definition of ΔT = Teq − Tonset. The induction time, tind, for hydrate onset at 80 bar was 25.0 min. The decrease of the dissociation pressure to 70 bar induces a decrease of both the subcooling and the induction time. They were further decreased when the dissociation pressure was set to 60 bar. The resulting hydrate onset conditions, subcooling temperatures and induction times for fresh water are summarized in Table 1. It was suggested that hydrate crystallization in system containing fresh water leads to scattered results (Duchateau et al., 2009), which is due to the stochastic nature of hydrate nucleation. The hydrate onset conditions shown in Fig. 3 represent these phenomena. Each hydrate onset condition at different dissociation pressure averages at least three measurements. By reducing the dissociation pressure from 80 to 60 bar, the subcooling changes from 6.3 to 1.9 °C, resulting a relative dispersion of about 60% with a mean value of 3.9 °C. Therefore it makes difficult to evaluate the hydrate formation
Table 1 The hydrate onset conditions for fresh water at different initial pressures and cooling rates.
50 40 0
2
4
6
8
Temperature
10
12
14
(oC)
Fig. 3. The hydrate onset conditions for fresh water at dissociation pressures of (a) 80 bar (▼), (b) 70 bar (■), and (c) 60 bar (●). Filled symbol indicates the hydrate onset condition for each dissociation pressure.
Cooling rate
Pdisso
Hydrate onset conditions
Subcooling
Induction time
(°C/min)
(bar)
Tonset (°C)
Ponset (bar)
(°C)
(min)
0.25
80 70 60 80
4.2 5.7 5.8 5.9
76.1 67.0 57.1 77.4
6.3 3.6 1.9 4.7
25.0 14.3 7.9 59.0
0.08
64
Y. Seo, S.-P. Kang / Journal of Petroleum Science and Engineering 88–89 (2012) 61–66
risk for a system containing fresh water and leads to complete avoidance of hydrate in the operation of conventional gas fields. However, the hydrate precursors or residual structures stemming from the dissociation of the hydrate phase enhance the hydrate formation, which results in a decrease of the subcooling. This is described as a memory effect among hydrate researchers, as hydrates are assumed to retain a memory of their structure – the hydrate precursors – when experiencing dissociation. Hydrate forms more easily from residual structures obtained from the dissociation of hydrate plugs than it does from fresh water with no previous hydrate history (Kumar et al., 2008; Sloan and Koh, 2008). It has been suggested that the memory effect can be avoided at temperatures greater than 28 °C or after several hours of heating (Sloan and Koh, 2008). However, such actions to avoid the memory effect are difficult to be carried out during the production of methane gas from marine hydrate deposits. It is considered that, after the dissociation of the hydrate, fluids will cool rapidly due to a heat exchange with the environment, especially when transported via offshore pipelines. In order to assure the fluids flow inside the pipelines, it is necessary to assess the risk of methane hydrate re-formation for the dissociated water where dissociated water indicates that the water that has experienced hydrate formation and dissociation. Fig. 4 shows a P–T trace plot during the cooling process of the dissociated water; the filled symbol for each trace indicates the hydrate onset condition at the given dissociation pressure. This clearly shows that the onset of hydrate formation from dissociated water occurs more rapidly. As shown in Fig. 4, when the dissociation pressure was 80 bar the hydrate onset was detected at 9.6 °C and 78.2 bar, which results in subcooling of 1.3 °C from the prediction of Teq = 10.8 °C. Rapid reformation of methane hydrate was observed in each experiment, and the obtained subcooling, as shown in Table 2, ranged from 1.2 and 1.6 °C for the dissociation pressures. The relative dispersion was about 14% from a mean value of 1.4 °C. This indicates that the hydrate would re-form just 1.4 °C below the hydrate equilibrium conditions during the cooling down of the dissociated water. The stochastic nature of hydrate nucleation was not observed in the dissociated water. Compared to the subcooling of 6.3 °C for fresh water at the dissociation pressure of 80 bar, the subcooling of 1.2 °C for dissociated water clearly indicates a high risk of hydrate plug formation during the transportation of methane from marine hydrate deposits. There are two hypotheses related to hydrate nucleation. One is the formation of a labile cluster, presuming that liquid water molecules are arranged around a dissolved solute molecule in a prehydrate
100
Pressure (bar)
90 80 70 60 50 40 0
2
4
6
8
10
12
14
Temperature (oC) Fig. 4. The hydrate onset conditions for dissociated water at dissociation pressures of (a) 80 bar (▼), (b) 70 bar (■), and (c) 60 bar (●). Filled symbol indicates the hydrate onset condition for each dissociation pressure.
Table 2 The hydrate onset conditions for dissociated water at different initial pressures and cooling rates. Cooling rate
Pdisso
Hydrate onset conditions
Subcooling
Induction time
(°C/min)
(bar)
Tonset (°C)
Ponset (bar)
(°C)
(min)
0.25
80 70 60 80
9.6 8.3 6.4 9.9
78.2 68.5 58.3 79.2
1.3 1.2 1.6 0.9
5.1 4.9 6.4 11.6
0.08
structure, with essentially the correct coordination number. The other is local structuring, where the prehydrate structure consists of a locally ordered water-guest structure rather than individual hydrate cavities. In both hypotheses, hydrate nucleation starts from the formation of the prehydrate structure. It is envisaged that the residual hydrate structures in the water phase act as templates to form these prehydrate structure and induce faster hydrate reformation, as shown in Fig. 4. When methane hydrate deposits are dissociated by depressurization, it is important to establish an optimized dissociation pressure. Although it is possible to obtain a fast dissociation rate by applying a large pressure difference between the dissociation pressure and the hydrate equilibrium pressure, this may cause a large temperature drop in the hydrate core due to endothermic hydrate dissociation, resulting in hydrate reformation in the hydrate well casing. Accordingly, it is suggested that the dissociation pressure should be close to the hydrate equilibrium condition, where it is almost impossible to avoid the presence of residual hydrate structures after the dissociation process. As confirmed in above experiments, the hydrate precursors will result in rapid methane hydrate reformation during the transportation of methane and dissociated water through offshore pipelines once the temperature of fluids falls into hydrate stability zone. An active hydrate mitigation strategy must be employed in order to avoid hydrate blockage in offshore pipelines, which could potentially result in a costly production stoppage. The pipelines can be insulated to ensure that the fluid temperature remains above the hydrate equilibrium temperature. Otherwise, the continuous injection of a chemical, such as mono ethylene glycol (MEG) or methanol can be employed to shift the hydrate equilibrium curve outside the operation conditions. However, all of these strategies could have an enormous impact on the production economics. Among the hydrate mitigation strategies, the use of a kinetic hydrate inhibitor requires a lower injection rate than the use of MEG or methanol. Moreover, it has shown reliable performance under subcooling conditions below 12 °C (Clark and Anderson, 2007). The application of a KHI as a possible hydrate mitigation strategy was investigated for dissociated water. PVCap was selected as the KHI in this work. Fig. 5 shows the effect of the addition of PVCap on the hydrate onset conditions for dissociated water. Table 3 presents the hydrate onset conditions along with the calculation of the subcooling and induction time with different PVCap concentrations. The concentration of PVCap was varied from 0.5 to 3.0 wt.% while the cooling process was started from 12 °C at 80 bar for all concentrations at a rate of 0.25 °C/min. The hydrate onset temperature was 5.9 °C for the 0.5 wt.% PVCap and became 4.5 °C for the 1.0 wt.% PVCap. It reached 2.5 °C with the addition of 3.0 wt.% PVCap. This indicates that the methane and dissociated water can be transported without hydrate formation for 31 min until the fluid temperature reaches 2.5 °C by adding 3.0 wt.% of PVCap. While the subcooling was 1.2 °C for the dissociated water without PVCap, it increased to 4.7, 6.0, and 7.8 °C at PVCap concentrations of 0.5, 1.0, and 3.0 wt.%, respectively. These results clearly represent the inhibition effect of PVCap on the hydrate formation.
Y. Seo, S.-P. Kang / Journal of Petroleum Science and Engineering 88–89 (2012) 61–66
Table 4 The hydrate onset conditions in the presence of sodium chloride at different PVCap concentrations at Pdisso = 80 bar and a cooling rate of 0.25 °C/min.
86 84
NaCl
PVCap
Hydrate onset conditions
Subcooling
Induction time
(wt.%)
(wt.%)
Tonset (°C)
Ponset (bar)
(°C)
(min)
3.5
0.0 0.0 0.5 1.0 3.0
2.9 7.5 2.1 1.5 1.1
76.9 78.7 76.7 75.9 74.9
6.2 1.8 6.9 7.4 8.8
24.8 7.3 27.8 29.7 35.3
82 No PVCap
Pressure (bar)
65
80
PVCap 0.5wt% PVCap 1.0wt%
78
PVCap 3.0wt%
76 74
2
4
6
8
10
12
14
Temperature (oC) Fig. 5. Effect of the addition of PVCap on the hydrate onset conditions at dissociation pressure of 80 bar for dissociated water. The concentration of PVCap was 0.0 (▼), 0.5 (■), 1.0 (●), and 3.0 (♦) wt.%.
The dissociated water from marine hydrate deposits may contain salts. In order to investigate the effect of salt ions on the hydrate reformation as well as the performance of KHI, synthetic brine with 3.5 wt.% NaCl was used to form and dissociate hydrate. Methane hydrate was formed at 0.5 °C and 95 bar, then the dissociation was carried out at 10.5 °C and 80 bar to take account the shift of hydrate equilibrium conditions in the presence of NaCl. Table 4 summarizes the hydrate onset conditions with and without PVCap. The hydrate dissociation pressure was fixed at 80 bar and the cooling rate from 10.5 to 0.5 °C was 0.25 °C/min. When fresh 3.5 wt.% NaCl solution with no hydrate history was used for the hydrate onset experiment, the hydrate onset was detected at 2.9 °C and 76.9 bar, which resulted the similar subcooling of 6.2 °C to that of fresh water, 6.3 °C, at the dissociated pressure of 80 bar in Table 1. Interestingly, when the NaCl solution experienced hydrate formation and dissociation is used for the hydrate onset experiment, the subcooling becomes 1.8 °C, which is similar to 1.3 °C for dissociated water at the dissociation pressure of 80 bar in Table 2. These results indicate the presence of NaCl in water phase doesn't affect seriously the hydrate onset conditions whether the water phase experienced hydrate formation or not. It is also noted that the reliable data for hydrate onset conditions can be obtained once the same cooling down process is applied to measure the hydrate onset conditions as studied in this study. Further experiments were carried out to investigate the hydrate onset conditions of dissociated NaCl solution after injecting PVCap. As shown in Table 4, the subcooling increased substantially by injecting PVCap into dissociated NaCl solution. At the concentration of PVCap of 0.5 wt.%, the subcooling was 4.7 °C in dissociated water without NaCl, but increased to 6.9 °C in dissociated NaCl solution. Considering the increase of subcooling in the presence of NaCl for both 1.0 and 3.0 wt.% of PVCap concentrations, there might be a synergistic inhibition effect of NaCl ions on the performance of PVCap.
Table 3 The hydrate onset conditions for dissociated water at different PVCap concentrations at Pdisso = 80 bar. Cooling rate
PVCap
Hydrate onset conditions
Subcooling
Induction time
Low Risk
0
40
Hydrate onset time (min)
70
No Yes Yes Yes Yes
Fig. 6 represents the risk of hydrate formation depicted in subcooling vs. hydrate onset time diagram. As indicated in the axis, low subcooling and short hydrate onset time represent the high risk of hydrate formation for a given fluids, whereas high subcooling and long hydrate onset time indicate the low risk of hydrate formation. For fresh water and NaCl solution without hydrate history, the risk of hydrate formation would be considered to be medium. However, once the water experiences the formation and dissociation of hydrates, the risk moves into high risk region as shown in the subcooling range of 1.2 and 1.8 °C. The addition of PVCap into the dissociated water reduces the risk of hydrate formation by moving subcooling range into 4.7 and 7.8 °C. The presence of NaCl ions further reduces the risk into subcooling range of 6.9 and 8.8 °C clearly showing the benefit of injecting PVCap. The injection of a KHI such as PVCap is a possible candidate to avoid hydrate reformation while transporting the fluids to the processing facilities. The inhibition mechanism of KHI remains not fully understood. One theory holds that the KHI polymer has functional groups that operate by distorting the hydrate structure and increasing the energy barrier needed to form a hydrate building unit that would grow into hydrate crystals (Sloan and Koh, 2008). If this theory is applicable to the formation of hydrate from residual structures, the obtained results indicate that PVCap is able to suppress the formation of hydrate building units from dissociated water containing residual hydrate structures. The residual hydrate structures may be adsorbed onto the functional groups of PVCap and prevented from further growth into the critical hydrate nuclei. Further studies are necessary to explore the inhibition mechanism of KHI in the presence of residual hydrate structures and different concentrations of salt ions.
High Risk
72
Hydrate history
30
Injecting PVCap to dissociated water and NaCl solution
20
Fresh water
10
Dissociated water
0 0
2
4
6
8
10
Subcooling (oC)
(°C/min)
(wt.%)
Tonset (°C)
Ponset (bar)
(°C)
(min)
0.25
0.5 1.0 3.0
5.9 4.5 2.5
76.9 76.1 74.9
4.7 6.0 7.8
18.8 24.0 31.2
High Risk
Low Risk
Fig. 6. Comparison of hydrate onset conditions in the presence of PVCap and sodium chloride. (○) Dissociated water, (●) Dissociated 3.5 wt.% NaCl solution, (∇) Fresh water, (▼) Fresh 3.5 wt.% NaCl solution, (□) PVCap injected dissociated water without NaCl, (■) PVCap injected dissociated 3.5 wt.% NaCl solution.
66
Y. Seo, S.-P. Kang / Journal of Petroleum Science and Engineering 88–89 (2012) 61–66
Methane production from marine hydrates will be different from conventional offshore gas fields in terms of water production and related flow assurance issues. Water production in gas fields can be described via the water:gas ratios, which are typically less than 10 bbl/ MMscf including condensed water and formation water production (Walsh et al., 2009). However the water production from marine hydrates can vary greatly depending on the hydrate reservoir type. Moreover, it has been suggested that water:gas ratios in excess of 1000 bbl/MMscf are possible (Walsh et al., 2009). It was noted that gas hydrates are composed of 15 wt.% of methane and 85 wt.% of water from the stoichiometry of structure I, 8CH4 · 46H2O. Moreover the presence of residual hydrate structures in dissociated water will increase the risk of methane hydrate reformation resulting hydrate plug when transporting produced fluids via offshore pipelines. If we consider the concentration of MEG to avoid hydrate formation when operating offshore pipelines at 80 bar and minimum temperature of 4 °C, the required MEG concentration will be 0.09 bbl MEG/bbl water from thermodynamic prediction, which demands the MEG distribution system and regeneration unit to handle 90 bbl MEG/MMscf based on a water:gas ratio of 1000 bbl/MMscf. From the scenario of producing methane from class 2 deposits involving a hydrate layer over a mobile water zone (Walsh et al., 2009), the maximum gas production rate would reach 70 MMscf/day, thus, the maximum MEG injection rate would be 6300 bbl/day. Considering that MEG regeneration unit of the Asgard B field for Statoil producing 840 MMscf/day gas and 220,000 bbl/day oil is handling about 6800 bbl of rich MEG flow (Brustad et al., 2005), MEG handling for marine hydrate is expected to be more costly compared to that in conventional offshore gas fields. Accordingly the associated operational expenditure and the handling of large quantities of MEG can make the KHI cost-effective hydrate mitigation option for marine hydrate development. From the above calculation of the water production of 70,000 bbl/day, the daily required PVCap would be around 4900 kg if the PVCap concentration of 3 wt.% was applied, which is much less than 6300 bbl of MEG. Moreover once it is confirmed that the PVCap has synergistic effect in the presence of salt ions, the dosage rate of PVCap would be able to be further reduced to save operational expenditure. The performance of KHI has to be evaluated carefully before its application in terms of the lowest operation temperature of offshore pipelines and the residence time of the produced fluids in the hydrate formation region. Although more comprehensive studies have to be designed based on a field development plan to apply KHI to methane hydrate production, the results obtained in this work suggest that a cost-effective hydrate prevention strategy will be vital for the success of methane hydrate production, as the risk of methane hydrate re-formation in dissociated water is much higher than in fresh water. 4. Conclusions Hydrate formation in fresh water leads to a subcooling of 6.3 °C at a dissociation pressure of 80 bar, while hydrate reformation in dissociated water, which has previously experienced hydrate formation and dissociation was outstandingly fast, resulting in a subcooling of 1.3 °C. The dissociation pressures after the first hydrate formation
stage varied from 80 to 60 bar; however, under each dissociation pressure, rapid hydrate reformation was observed during the cooling of dissociated water. Although the presence of residual hydrate structures enhances the nucleation of hydrate building units, the polymer structure of PVCap can suppress the nucleation process. An increase of the PVCap concentration from 0.5 to 3.0 wt.% resulted in an increase of subcooling to 7.8 °C. This is even higher than the subcooling of fresh water and suggests that the addition of PVCap will be an efficient hydrate mitigation strategy to avoid hydrate reformation during the transportation of methane and dissociated water through offshore pipelines. Moreover it is observed that the subcooling increased to 8.8 °C at the PVCap concentration of 3.0 wt.% in the presence of NaCl in the water phase suggesting that there might be synergistic effect of salt ions on the performance of PVCap as a KHI. Acknowledgements This work was supported by the Energy Efficiency & Resources Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Knowledge Economy (No. 2010201030001A), and partially supported by the Australian Academy of Science-NRF Exchange Program. References Brustad, S., Loken, K.P., Waalmann, J.G., 2005. Hydrate prevention using MEG instead of MeOH: Impact of experience from major Norwegian developments on technology selection for injection and recovery of MEG. Offshore Technology Conference, 2–5 May, 2005. Houston, Texas, U.S.A. Clark, L.W., Anderson, J., 2007. Low Dosage Hydrate Inhibitors (LDHI): Further advances and developments in flow assurance technology and applications concerning oil and gas production systems. International Petroleum Technology Conference, 4–6 December, 2007. Dubai, U.A.E. Duchateau, C., Peytavy, J., Glenat, P., Pou, T., Hidalgo, M., Dicharry, C., 2009. Laboratory evaluation of kinetic hydrate inhibitors: a procedure for enhancing the repeatability of test results. Energy Fuel 23, 962–966. Duchateau, C., Glenat, P., Pou, T., Hidalgo, M., Dicharry, C., 2010. Hydrate precursor test method for the laboratory evaluation of kinetic hydrate inhibitors. Energy Fuel 24, 616–623. Konno, Y., Masuda, Y., Hariguchi, Y., Kurihara, M., Ouchi, H., 2010. Key factors for depressurization-induced gas production from oceanic methane hydrates. Energy Fuel 24, 1736–1744. Kumar, R., Lee, J., Song, M., Englezos, P., 2008. Kinetic inhibitor effects on methane/propane clathrate hydrate-crystal growth at the gas/water and water/n-heptane interfaces. J. Cryst. Growth 310, 1154–1166. Kurihara, M., Sato, A., Ouchi, H., Narita, H., Masuda, Y., Saeki, T., Fujii, T., 2008a. Prediction of gas productivity from Eastern Nankai Trough methane-hydrate reservoirs. Offshore Technology Conference, Houston, Texas, U.S.A., OTC 19382. Kurihara, M., Funatsu, K., Ouchi, H., Masuda, Y., Yamamoto, K., Narita, H., Dallimore, S.R., Collett, T., Hancock, S.H., 2008b. Analysis of production tests and MDT tests conducted in Mallik and Alaska methane hydrate reservoirs: what can we learn from these well tests? Proceedings from the 6th International Conference on Gas Hydrates, July 6–10, 2008. Vancouver, British Columbia, Canada. Oyama, H., Konno, Y., Masuda, Y., Narita, H., 2009. Dependence of depressurizationinduced dissociation of methane hydrate bearing laboratory cores on heat transfer. Energy Fuel 23, 4995–5002. Sloan, E.D., Koh, C.A., 2008. Clathrate hydrates of natural gases, Third edition. CRC Press, 6000 Broken Sound Parkway NW, suite 300, Boca Raton, FL, U.S.A. Walsh, M.R., Hancock, S.H., Wilson, S.J., Patil, S.L., Moridis, G.J., Boswell, R., Collett, T.S., Koh, C.A., Sloan, E.D., 2009. Preliminary report on the commercial viability of gas production from natural gas hydrates. Energy Econ. 31, 815–823. Yamamoto, K., 2007. Introduction of the 2007–2008 JOGMEC/NRCan/Aurora Mallik gas hydrate production research program, NWT, Canada. AGU Fall Meeting.