Hydrate Management in Deadlegs: Detection of Hydrate Deposition

Jan 24, 2018 - Hydrate Management in Deadlegs: Detection of Hydrate Deposition. Using Permittivity Probe ... The vertical pipe system simulates a dead...
3 downloads 9 Views 3MB Size
Subscriber access provided by READING UNIV

Article

Hydrate Management in Deadlegs: Detection of Hydrate Deposition using Permittivity Probe Jeong-Hoon Sa, Bo Ram Lee, Xianwei Zhang, Kjetil Folgerø, Kjetil Haukalid, Jan Kocbach, Keijo J. Kinnari, Xiaoyun Li, Kjell Magne Askvik, and Amadeu K. Sum Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.7b03963 • Publication Date (Web): 24 Jan 2018 Downloaded from http://pubs.acs.org on January 29, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

!

Hydrate Management in Deadlegs: Detection of Hydrate Deposition using Permittivity Probe

Jeong-Hoon Sa,a Bo Ram Lee,a,b Xianwei Zhang,a Kjetil Folgerø,c Kjetil Haukalid,c Jan Kocbach,c Keijo J. Kinnari,d Xiaoyun Li,e Kjell Askvik,f* and Amadeu K. Suma*

a

Hydrates Energy Innovation Laboratory, Chemical & Biological Engineering Department, Colorado School of Mines, Golden, CO 80401, United States b

Department of Chemical Engineering, Pohang University of Science & Technology, 77 Cheongam-Ro, Nam-Gu, Pohang, Gyeongbuk, 37673, Korea c

Christian Michelsen Research AS, N-5892 Bergen, Norway d

e

Statoil ASA, N-4035 Stavanger, Norway

Statoil ASA, N-7005 Trondheim, Norway f

Statoil ASA, N-5020 Bergen, Norway

*Corresponding authors e-mail: [email protected], [email protected] 1 ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Formation of gas hydrates in oil/gas pipelines has to be properly managed as they can often lead to plugging, primarily by deposition, causing safety issues and significant expenses for repair and recovery. Early detection of hydrate deposition is thus critical for managing such risks and establishing stratagies for hydrate mitigation and remediation. Here, a permittivity probe is applied to an 1-inch vertical pipe system in order to detect hydrate deposition. The vertical pipe system simulates a deadleg, which is a pipe section used for intermittent services and maintenance in hydrate management. Hydrate deposition under water saturated gas environment is monitored by measuring the dielectric constant of the hydrate layer, which is considered as a three-component mixture of hydrates, gas, and water. The permittivity responses upon hydrate formation and dissociation are observed, and their physical interpretations are also provided. By applying appropriate models, thickness, wetness, and porosity of hydrate deposits are quantitatively estimated. Knowledge obtained from this work will be helpful in further developing a real-time monitoring of hydrate deposition by dielectric measurements.

2 ACS Paragon Plus Environment

Page 2 of 28

Page 3 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1. Introduction Gas hydrates are solid compounds in which gas and water form under low temperature and high pressure conditions.1 In oil/gas production and transportation systems, hydrates need to be properly managed as they can seriously affect the pipeline flow.2,3 Once the pipeline is plugged by hydrates, it needs a considerable amount of time and energy to recover. The pipe sections called deadlegs are occasionally used for special services in flow maintanance.4,5 Deadlegs can also be used to manage hydrate risk by monitoring the flow status, depressurizing, and injecting hydrate inhibitors such as methanol and ethylene glycol. However, deadlegs themselves pose a significant hydrate risk as well.6 As deadlegs have relatively stagnant flows, they become much colder than the main pipelines, causing water vapor to condense on the pipe wall and then to form hydrates. Deadlegs are consequently exposed to hydrate risk mainly by deposition on the wall. There have been several studies on hydrate deposition under water saturated gas systems as similar with deadlegs. The influence of several parameters like header/pipe wall temperatures and time duration on hydrate deposition in deadlegs were studied in the lab-scale experimental systems.6-8 The characteristic properties of hydrate deposits such as porosity and wetness were also quantified.8 The objective of this work is to study the applicability of dielectric measurements for characterizing hydrate deposition in a deadleg. Dielectric measurements have previously been demonstrated to be useful in studying hydrate formation. Jakobsen and co-workers published some papers9,10 in the 1990s where hydrate formation in emulsions was studied. Recently, Haukalid, Folgerø and co-workers published a series of papers, studying both hydrates in emulsions and hydrate layers.11-13 An important aspect in these papers is the application of dielectric mixing models to estimate the volume fractions of the different phases from the measured permittivity. The measurements were also used to estimate the layer thickness. 3 ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Paragon Plus Environment

Page 4 of 28

Page 5 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 28

The measured permittivity depends on the layer thickness when the deposit layer is thinner than the sensitivity depth of the probe (i.e. thinner than 4-5 mm). It has been previously found that the following empirical model describes the relation between measured permittivity (!) ) and layer thickness12,14: .

.

!) $ !* %+ & , -/ ( 0 !1 , -/

(2)

where !* is the permittivity of the deposition layer, !1 is the permittivity of the backing fluid,

2 is the layer thickness, and 3 is the empirical probe constant. Thus, equation (2) can be used to estimate the layer thickness for layers thinner than the sensitivity depth if the layer permittivity is known. The dielectric constants of (the applied) gas, hydrates, and water are approximately 1.2, 3.15, and 85.5, respectively at 4 °C/100 bar. This indicates that even small changes in the water fraction will give significant changes in the measured permittivity. Contrary, changes in the hydrate/gas ratio in the layer will impact the measured permittivity much less. In this paper, a model considering the hydrate layer as a three-component mixture of dense hydrates, gas, and water is applied. Input parameters to the model are gas, water, and hydrate permittivity, porosity, and liquid water volume fraction. Permittivity values are given at operating pressure and temperature. To calculate the permittivity of dry hydrates, water is neglected and the hydrate layer is considered as a two-component mix of gas and dense hydrates. The porosity (45 ) is defined as the gas volume fraction in these dry hydrates. The permittivity of the dry hydrates (!6 ) is estimated using Looyenga’s model.15 !6 $

*: 7!8 9

0

*: 45 %!; 9

&

*: !8 9 (
?@4A 0 @B4A1

(4)

where 4A is the volume fraction of liquid water. This model is considered to be suitable for evaluating how changes in porosity and wetness affect permittivity, but since the model has not been tested on hydrates previously, calculated values should be interpreted with care.

3. Experimental Setup and Procedure 3.1. Materials The deionized water was used in all experiments. The pure N2 with industrial grade (99.0%) and the gas mixture of 74.2% CH4 and 25.8% C2H6 (±2%) were provided by General Air.

3.2. Dielectric Constants Dielectric constants at 1 GHz of the applied materials are presented in Table 1. The dielectric constant of gas is calculated from density and polarizability using Clausius-Mossotti equation.17 Polarizability values of CH4, C2H6, and N2 are taken from the literature.18 Liquid water dielectric constant at 1 GHz/1 bar is calculated using the data reported by Kaatze.19 The dielectric constant of water at low frequencies as a function of pressure was reported by Uematsu and Franck.20 By combining the pressure dependency20 with the frequency dependency,19 the dielectric constant of water at 1 GHz/100 bar is estimated. The dielectric constant of natural gas hydrates is assumed to be close to that of ice (!" " 3.15) within the frequency range studied in this work.21

7 ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 28

Table 1. Dielectric constants (at 1 GHz) for the applied materials Pressure

Temperature

CH4/C2H6

N2

Liquid water

Hydrates

1 bar

4 °C

1.0

1.0

85.45

3.15

1 bar

25 °C

1.0

1.0

Not applied

Not applied

100 bar

4 °C

1.18

Not applied

85.5

3.15

100 bar

25 °C

1.15

1.05

Not applied

Not applied

3.3. Experimental Apparatus The details on 1-inch vertical pipe experimental system are well described in our previous paper.8 Major difference is a permittivity probe which is assembled to the 1-inch pipe as indicated in the schematic of experimental apparatus (Fig. 3). The permittivity probe is installed at 5.6 in the distance from the header (C) / inner diameter of the pipe (ID). The permittivity probe is connected to a vector reflectometer (Copper Mountain R140) through a high-quality coaxial cable. Prior to installation in the apparatus, the probe is characterized by measurement on three fluids with known permittivity. Air, acetone, and cyclopentane are used as reference fluids. This characterization is done once and for all. The permittivity of an unknown sample can then be calculated from the measured reflection coefficient using the bilinear method.14 Repeated experiments have shown that the reproducibility of the dielectric constant measurement is within 0.05, whereas the precision (noise level) within each experiment is approximately 0.01.

8 ACS Paragon Plus Environment

Page 9 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

To dissociate hydrates under pressure, a stepwise cooling and heating is applied to the header and the pipe wall. The header is cooled down to 25 oC first to reduce the further water supply to the pipe. As the pipe wall is then heated up from 4 oC to 25 oC, hydrates start to dissociate since the hydrate phase equilibrium temperature for CH4/C2H6 gas mixture at 100 bar is 18.9 oC (estimated by CSMGem). These processes are also visually monitored through the top window. In several tests, the header and pipe wall is cooled down to 25 oC and -10 oC, respectively. After the temperatures become stabilized, the system is depressurized to atmospheric pressure to prevent hydrates from melting. A visual inspection to measure the hydrate deposit thickness is then carried out.

4. Results and Discussion 4.1. Baseline Tests for Permittivity Probe System To better interpret the measurements in the deadleg experimental set-up, permittivity responses upon gas density change are obtained (Fig. 4). Under atmospheric pressure and room temperature (Figs. 4a and 4b), the measured dielectric constant of N2 is approximately 1.03, which is slightly higher than the dielectric constant of 1.00 for dry N2, but within the estimated reproducibility of 0.05. This is believed to be due to systematic errors caused by remaining uncertainty after calibrating the reflectometer. The dielectric constant increases up to 1.13 when pressurizing N2 to 100 bar, which is slightly higher than the estimated reference value. A pressure swing is then applied, and the reliable permittivity responses are confirmed. The test is repeated with CH4/C2H6 mixture (Figs. 4c and 4d). The general trend observed in the permittivity responses is similar to the N2 test. The dielectric constant at atmospheric conditions is slightly higher than that of dry gas, but within the estimated reproducibility. A shift in dielectric constant levels of around 0.03-0.05 between the first pressure swing and the next two 10 ACS Paragon Plus Environment

Page 10 of 28

Page 11 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Permittivity responses upon liquid water condensation are also tested (Fig. 5). In order to avoid hydrate formation, the system is pressurized with N2 to 100 bar, as N2 forms hydrates at -14.3 oC (estimated by CSMGem). To induce water condensation, the pipe wall is cooled down to 4 oC, and the header is set to be 30, 60, or 80 oC to vary the rate of water evaporation and the saturated amount of water vapor. Under 30/4 oC (Fig. 5a), the dielectric constant gradually increases with a constant rate for initial 77 h, but then it suddenly drops and resumes to increase again. Once the water droplets condense on the probe, they grow in size over time with a continuous water supply from vapor. As the water droplets become larger and thus heavier, they would drip down, resulting in a sudden drop in dielectric constant. Similar permittivity responses are observed in both 60/4 oC (Fig. 5b) and 80/4 oC (Fig. 5c) tests, while their increasing rates of dielectric constant are much faster. Such behaviors repeat again, indicating that the system would reach a steady state. According to the center temperature profiles (Fig. 5d), a higher header temperature leads to a higher center temperature at a given location, thus having a larger temperature gradient along the radial direction. In addition, the amount of water vapor also increases with temperature due to a higher water vapor pressure. These result in a faster water condensation on the pipe wall. As the dielectric constant changes with the amount of water condensed on the probe, the slope of permittivity curves would be related to the water condensation rate, and it is confirmed that these increase with header temperature (Fig. 5e). Such characteristic permittivity responses observed here can be used to understand hydrate deposition on the pipe wall.

12 ACS Paragon Plus Environment

Page 12 of 28

Page 13 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

an inflection point is observed around 3 to 4 h. The dielectric constant then finally becomes stabilized at around 2.5. This increase would follow a build-up of hydrate deposits over time. The trend observed here is repeatable under 60/4 oC. Under 30/4 oC, permittivity response is similar with 60/4 oC, but it takes longer to present an inflection point, and then the dielectric constant increase much slowly (Fig. 6b). The result from 40 h test under 60/4 oC (Fig. 6c) is similar with repeat tests shown in Fig. 6a, but it is interesting to see a slight decrease in dielectric constant after 20 h, which will be discussed later. On the other hand, permittivity response under 80/4 oC is completely different from the others (Fig. 6d). The dielectric constant rapidly increases up to 5.5 at the initial stage. When considering the dielectric constant of gas (approximately 1.2) obtained from the baseline test and theoretical dielectric constant of 100% hydrates (approximately 3.15), there is some amount of liquid water condensed on the probe. The dielectric constant then suddenly drops, but not below 3, implying that there are still some of hydrate deposits covering the probe. The dielectric constant increases and drops again, where this trend is similar with the responses observed upon water condensation. It can be interpreted that liquid water droplets condense on the existing hydrate deposits due to the high temperatures and drip down as their size becomes larger. Such excess water supplied from the header would result in wet hydrate deposits as described in our previous paper.8 Another interesting thing to point out here is a decreasing tendency of the dielectric constant over time under 80/4 oC. While the dielectric constant increases and drops again, the lowest and highest dielectric constant values of each cycle tend to decrease. As seen from 60/4 oC tests (Fig. 6c), such decrease in dielectric constant is believed to be related with the characteristic properties of hydrate deposits such as porosity and wetness.

14 ACS Paragon Plus Environment

Page 14 of 28

Page 15 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Paragon Plus Environment

Page 16 of 28

Page 17 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

4.4. Qualitative Interpretation of Permittivity Responses An interesting feature in the permittivity plots is the inflection point upon hydrate formation, which has typically been observed within a couple of hours after the initiation of tests under 30/4 oC and 60/4 oC. Additional tests are performed to investigate what happens at this particular time. The system is first pressurized with CH4/C2H6 mixture to 100 bar under room temperature, so the dielectric constant increases to 1.3 within 0.5 h (Fig. 9a). To induce hydrate formation, the header and pipe wall temperatures are set to be 60 oC and 4 oC, respectively. This test is then intentionally stopped at 2.9 h, and there is no inflection point shown by that time. A stepwise heating and cooling is applied to dissociate hydrates if they formed (Fig. 9b), but the resulting permittivity responses are pretty much different from the permittivity changes observed during hydrate dissociation (Fig. 7). The slight fluctuation of the dielectric constant is attributed to the temperature changes applied to dissociate hydrates. The test is repeated again for a bit longer until an inflection point is shown, and continues for 1.5 h more (Fig. 9c). As the pipe wall is heated up, the dielectric constant dramatically increases to 3.3, and is then suddenly dropped to 1.4 (Fig. 9d), as typically observed in hydrate dissociation. The subsequent dielectric constant is slightly higher than that of dry gas due to the probable liquid water droplets condensed on the probe. It can be implied from these experimental results that a linear increase of the dielectric constant before showing an inflection point is due to the liquid water condensation on the probe. In addition, the inflection point shown in the permittivity plots is an indication of the onset of hydrate formation on the probe. This does not necessarily indicate that there are no hydrates formed anywhere in the pipe before that since they could not be detected. However, the formation of hydrates most likely occurs across the entire pipe.

18 ACS Paragon Plus Environment

Page 18 of 28

Page 19 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Paragon Plus Environment

Page 20 of 28

Page 21 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 28

for uncertainty in the measured permittivity is estimated by a Monte-Carlo analysis assuming a normal distributed uncertainty of ±0.05 (95% confidence level) in the measured permittivity. The resulting changes are ±3% (absolute, 95% confidence level) in gas porosity and ±0.5% (absolute, 95% confidence level) in water porosity. Note that uncertainty in reference parameters (i.e., layer thickness and permittivities of gas, water and dry hydrates) are not considered in this sensitivity analysis. Although this analysis does not consider the representativeness of the model, it gives an estimate of the model sensitivity for permittivity changes.

Table 2. Volume fractions of gas/water and porosity estimated from the measured dielectric constant before and after freezing using equations (3) and (4). Measured data Estimated volume fraction Porosity (45 ) Duration Final !D Experiment !D after (hour) thickness Gas hydrate before freezing Gas (4; ) Water (4A ) (mm) freezing A B C D E

13 20 40 145 235

2.0 2.5 4.2 7.5 9.0

2.60 2.63 2.76 2.75

1.74 2.02 2.13 2.19

44.5% 53.0% 57.0% 59.5%

48.0% 42.0% 38.0% 36.0%

7.5% 5.0% 5.0% 4.5%

In stage IV, a decrease in dielectric constant is observed. As described above, this is attributed to the changes in porosity and wetness in combination with increasing layer thickness. In experiment C, the dielectric constant changes from 2.69 to 2.63 during stage IV from 30 to 40 hours, and layer thicknesses are approximately 2.5 mm after 20 hours and 4.2 mm after 40 hours (Table 2). Assuming a linear layer growth between 20 and 40 hours and the same growth rate in experiment B and C, the thickness at the beginning of stage IV in experiment C (i.e. after 30 hours) is estimated to be approximately 3.3 mm. It is presumed that the decrease in 22 ACS Paragon Plus Environment

Page 23 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

dielectric constant is caused by conversion of free water into hydrates, and then equation (4) in combination with equation (2) can be used to estimate the changes in wetness. The change in dielectric constant corresponds to the change in water volume fraction from 5.7% to 5.0% within this 10 hour interval when increasing thickness from 3.3 mm to 4.2 mm. Similarly, the dielectric constant changes from 2.9 to 2.76 in stage IV in experiment D. Using the same methodology, this is found to correspond to the change in water volume fraction from 5.9% to 4.9% for approximately 120 hours. It is assumed for all these calculations that the gas volume fraction is constant, and that the porosity and wetness are uniform through the deposit layer. Equation (2) can then be used to estimate layer thickness with known permittivities of backing fluid (gas) and layer. As discussed above, however, the wetness and porosity change over time. It is therefore necessary to make assumptions regarding the variations in gas and water content over time to estimate the growth rate. In the following analysis of experiment C, the gas volume fraction is assumed to be 42% as estimated above. Several different wetness profiles are assumed (i.e. the change in water volume fraction with time), and the corresponding estimated layer thicknesses are calculated (Fig. 12). As described above, the water volume fraction changed from 5.7% to 5.0% over 10 h interval in stage IV (i.e. a change of 0.07% per hour). A linear decrease in water fraction is assuemed in Case 1 (Fig. 12a), and the estimated thickness at the onset of hydrate formation is approximately 1 mm (Fig. 12b), which is very unlikely. It is thus more reasonable to assume that the change in wetness is steeper during the initial stage of hydrate formation. Cases 2-4 show three wetness profiles with an exponential time dependency eventually converging into the linear assumption model. The resulting thickness estimation are significantly different from each other. This illustrates the challenge of deposit thickness estimation when the water fraction is unknown. Future improvements of

23 ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Paragon Plus Environment

Page 24 of 28

Page 25 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

5. Conclusion In this Article, we introduced a permittivity probe system applied to a 1-inch vertical pipe deadleg to study hydrate deposition under water saturated gas environment. Capabilities of the permittivity probe are tested for detection of hydrate deposits consisting of hydrates, gas, and water. From the dielectric measurements, liquid water condensation on the pipe wall and onset of hydrate formation are identified. The changes in wetness and porosity of hydrate deposits are also observed. By applying an appropriate model, quantitative estimates of hydrate deposit growth rate are obtained with assumed wetness profiles, and these are compared with the measured deposit thickness. Physical interpretations of characteristic permittivity responses upon hydrate formation and dissociation are also provided. Experimental results presented here show the potential of the permittivity probe system for detection of hydrate deposition. To better achieve an accurate real-time monitoring of hydrate deposition, several challenges are still remained. Modification of the experimental system is suggested in order to obtain both wetness and thickness information of hydrate deposits.

Author Information Corresponding Authors *E-mail: [email protected] (KMA), [email protected] (AKS) Notes The authors declare no competing financial interest.

Acknowledgements The authors wish to express their appreciation to Statoil for funding this project and granting permission to publish this paper.

25 ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

References 1.!

Sloan, E. D.; Koh, C. A. Clathrate Hydrates of Natural Gases, 3rd ed.; CRC Press: Boca Raton, FL, 2008.

2.!

Sloan, E. D.; Koh C. A.; Sum A. K. Natural Gas Hydrates in Flow Assurance. Gulf Professional Publishing, 2010.

3.!

Kinnari, K.; Hundseid, J.; Li, X.; Askvik, K. M. Hydrate Management in Practice. J. Chem. Eng. Data 2015, 60, 437-446.

4.!

Habib, M. A.; Badr, H. M.; Said, S. A. M.; Hussaini, I.; Al-Bagawi, J. J. On the Development of Deadleg Criterion. J. Fluids Eng. 2005, 127, 124-135.

5.!

Lunde, K.; Kinnari, K.; Labes-Carrier, C. Deadleg. US Patent, 9297237 B2, 2016.

6.!

Kinnari, K. J.; Askvik, K. M.; Li, X.; Austvik, T.; Zhang, X.; Sa, J.-H.; Lee, B. R.; Sum, A. K. Hydrate Management of Deadlegs in Oil and Gas Production Systems – Background and Development of Experimental Systems. Energy Fuels 2017, 31, 11783-11792.

#$!

Zhang, X.; Lee, B. R.; Sa, J.-H.; Kinnari, K. J.; Askvik, K. M.; Li, X.; Sum, A. K. Hydrate Management in Deadlegs: Effect of Header Temperature on Hydrate Deposition. Energy Fuels 2017, 31, 11802-11810.!

8.!

Sa, J.-H.; Lee, B. R.; Zhang, X.; Kinnari, K. J.; Li, X.; Askvik, K. M.; Sum, A. K. Hydrate Management in Deadlegs: Hydrate Deposition Characterization in a 1-in. Vertical Pipe System. Energy Fuels 2017, 31, 13536-13544.

9.!

Jakobsen, T.; Sjöblom, J.; Ruoff, P. Kinetics of Gas Hydrate Formation in W/OEmulsions the Model System Trichlorofluoromethane/Water/Non-ionic Surfactant Studied by Means of Dielectric Spectroscopy. Colloids Surf. A 1996, 112, 73-84.

26 ACS Paragon Plus Environment

Page 26 of 28

Page 27 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

10.!

Jakobsen, T.; Folgerø, K. Dielectric Measurements of Gas Hydrate Formation in Water-in-Oil Emulsions using Open-Ended Coaxial Probes. Meas. Sci. Technol. 1997, 8, 1006-1015.

11.!

Haukalid, K.; Askvik, K. M.; Folgerø, K.; Thomas, P. J. Impact of Gas Hydrate Formation/Dissociation on Water-in-Crude Oil Emulsion Properties Studied by Dielectric Measurements. Energy Fuels 2015, 29, 43-51.

12.!

Haukalid, K.; Folgerø, K. Broad-Band Permittivity Measurements of Formation of Gas Hydrate Layers using Open-Ended Coaxial Probes. Energy Fuels 2016, 30, 7196-7205.

13.!

Haukalid, K.; Folgerø, K.; Barth, T.; Fjermestad, S. L. Hydrate Formation in Waterin-Crude Oil Emulsions Studied by Broad-Band Permittivity Measurements. Energy Fuels 2017, 31, 3793-3803.

14.!

Folgerø, K.; Tjomsland, T. Permittivity Measurement of Thin Liquid Layers using Open-Ended Coaxial Probes. Meas. Sci. Technol. 1996, 7, 1164-1173.

15.!

Looyenga, H. Dielectric Constants of Heterogeneous Mixtures. Physica 1965, 31, 401406.

16.!

Tiuri, M.; Sihvola, A. H.; Nyfors, E. G.; Hallikaiken, M. T. The Complex Dielectric Constant of Snow at Microwave Frequencies. IEEE J. Ocean. Eng. 1984, 9, 377-382.

17.!

Böttcher, C. J. F. Theory of Electric Polarization, Elsevier Publishing Company, 1952.

18.!

Bates, D.R.; Bederson, B. Advances in Atomic and Molecular Physics. Vol 20, Academic Press, 1985.

19.!

Kaatze, U. Reference Liquids for the Calibration of Dielectric Sensors and Measurement Instruments. Meas. Sci. Technol. 2007, 18, 967-976.

20.!

Uematsu, M.; Franck EU. Static Dielectric Constant of Water and Steam. J. Phys. Chem. Ref. Data 1980, 9, 1291-1306.

27 ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

21.!

Kliner, J. R.; Grozic, J. L. Determination of Synthetic Hydrate Content in Sand Specimens using Dielectrics. Can. Geotech. J. 2006, 43, 551-562.

28 ACS Paragon Plus Environment

Page 28 of 28