SPE 107335

Hydraflow: Avoiding Gas Hydrate Problems

H. Haghighi, SPE, R. Azarinezhad, SPE, A. Chapoy, R. Anderson, SPE, and B. Tohidi, SPE, Heriot-Watt U.

Copyright 2007, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Europec/EAGE Annual Conference and Results prove that the concept is viable (at least under
Exhibition held in London, United Kingdom, 11–14 June 2007.
laboratory conditions), and strongly suggest that
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
HYDRAFLOW technology could offer significant benefits over
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to existing flow assurance strategies, providing a novel low
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at CAPEX/OPEX solution for challenging (e.g. deepwater, long
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
tie-backs, mature fields) operations.
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than
300 words; illustrations may not be copied. The abstract must contain conspicuous Introduction
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, Texas 75083-3836 U.S.A., fax 01-972-952-9435.
Generally, most of the existing gas hydrate flow assurance
techniques are based on avoiding solid hydrate formation. This
Abstract is normally achieved by either reducing water content,
Flow Assurance continues to remain a major challenge in reducing heat loss to the environment (e.g. by insulation
offshore production operations. Current methods for avoiding and/or active heating) or the injection of thermodynamic
gas hydrate problems are generally based on one or a and/or kinetic hydrate inhibitors. However, all of these
combination of the following three techniques: (1) injection of approaches have limitations, particularly in the case of
thermodynamic inhibitors (e.g. methanol, ethylene glycol) to deepwater operations, high water cuts (e.g. mature fields)
prevent hydrate formation, (2) use of kinetic hydrate inhibitors and/or long tieback systems. As forecasted by ITI Energy,
(KHIs) to sufficiently delay hydrate nucleation/growth, and (3) global oil and gas production is increasingly being dominated
maintaining pipeline operating conditions outside the hydrate by exploitation of mature fields, meaning existing techniques
stability zone by insulation and/or active heating and/or by − which are more geared to new developments (lower water
controlling pressure.1 However, for many production cuts) − are becoming less practical/economic.2 Thus the
operations, particularly deepwater fields, those requiring long requirement for alternative strategies is expected to continue to
tiebacks, and mature reservoirs (where water cuts can be very increase in the future. HYDRAFLOW is a new, patented flow
high), the above techniques may not be economical and/or assurance technology, which aims to meet this need. The
logistically practical. Thus the industry needs novel and concept of HYDRAFLOW is to be able to transport oil/gas
improved techniques to tackle flow assurance problems for without any thermal (i.e. insulation or active heating) or high
such challenging conditions. The new approach presented in dosage chemical treatment, thus greatly reducing the costs of
this paper −HYDRAFLOW− aims to meet this need. multiphase hydrocarbon pipeline transport. The absence of
thermal and/or major chemical treatment means
HYDRAFLOW is a novel ‘cold flow’ concept, which breaks HYDRAFLOW falls under the umbrella of ‘cold flow’
from the tradition of straightforward hydrate prevention. technologies.
Instead, gas hydrate formation in pipelines is intentionally
induced and managed, eliminating the need for expensive SINTEF (Norway), in partnership with BP, are also
thermal/chemical inhibition while improving the economics developing a proposed cold flow technology.3-6 The basis of
and practicalities of multiphase fluid transport. this concept is the argument that ‘dry’ gas hydrates can be
formed under certain conditions (i.e. recycling some of the
Experimental analyses of low and high GOR systems for fluid containing gas hydrates to provide seeds for further
various different production scenarios with and without AA hydrate formation), and that these would be able to flow as
(Anti-agglomerant) have been undertaken as part of slurry in a continuous oil phase without agglomerating and
HYDRAFLOW concept development. The effect of simulated blockage. Therefore, the main objective is to convert all the
shut-in on fluid transportability has also been tested. A new water phase into hydrates. There seems to be limitations in this
experimental set-up for measuring the viscosity of the fluid technique with respect to high GOR and/or high water cut
system under high-pressure conditions has been designed and systems. Furthermore, from a practical viewpoint, there are
commissioned in order to evaluate the rheology of the systems several challenges with this approach, for example; subsea
under investigation. pump(s), adjusting the recycled fluid flow rate and mixing
pattern, degree of subcooling (i.e., fluid entry temperature to
2 SPE 107335

the loop), coping with changes in the produced fluid rates and • Examine the potential of the HYDRAFLOW loop
characteristics, application to saline produced water systems, concept
etc.
• Assess HYDRAFLOW for simulated shut-in

HYDRAFLOW Concept7 Experimental Methods and Equipment

The cold flow concept proposed here − termed HYDRAFLOW Experimental Equipment
to distinguish it from ‘dry’ hydrate concepts − is based on All experiments were conducted in a high-pressure (400 bar)
allowing gas hydrate formation in the presence of excess kinetic rig (Figure 2) designed and constructed in-house. As
water, but preventing solids agglomeration using low doses part of the HYDRAFLOW study, the rig has been modified
(few % by weight) of chemical anti-agglomerates where and calibrated for viscosity measurements up to 4000
necessary. The objective is to minimise or completely centipoises (cp).
eliminate the gas phase by converting it into hydrates through
reaction with produced or added water (e.g. seawater). Where Materials
possible, it is proposed to recycle all or part of the Materials used were distilled water, a standard natural gas
liquid/aqueous phase as part of a ‘Loop’ concept, wherein the (composition given in Table 1), a North Sea crude
recycled fluid plays the role of carrier fluid, collecting (composition given in Table 2), and 2 different commercial
produced fluids from various wells, reacting with the gas AAs (termed AA1, AA2).
phase to form hydrates and transporting them as hydrate slurry
in oil and/or water to production facilities (Figure 1). Experimental Procedures
The standard experimental procedure is as follows. Desired
The proposed HYDRAFLOW approach should volumes of water and/or oil with or without AA are pre-loaded
reduce/eliminate the risks of hydrate blockage; low doses of into the cell, with a headspace left for gas (or live oil). Initially
AA chemicals (if necessary) working to prevent the temperature of the cell is set to 20ºC. Prior to gas (or oil)
agglomeration and reduce slurry viscosity. By converting the injection, the viscosity of the system is measured. Gas (or oil,
gas phase into hydrates, the density difference between phases or water, depending on the experiment) is then injected into
(e.g., water, oil, hydrates) will be greatly reduced, which the system at fixed temperature until stable pressure is
should help to alleviate pipeline slugging problems, thus achieved. When the system reaches equilibrium, the viscosity
lowering OPEX/CAPEX. HYDRAFLOW could also reduce of the system is measured at various rpm. Temperature is then
pipeline costs by eliminating the need for heating/insulation, reduced to 4ºC, simulating seabed temperature. This cooling
while reducing operating pressures (or increasing capacity). induces hydrate formation. Once conditions have stabilised
Furthermore, it could have the extra benefit of reducing following hydrate growth, the slurry viscosity is then
chemical and disposed water treatment costs (this could measured again at various rpm.
include other chemicals, e.g., corrosion, scale, wax, asphaltene
inhibitors) through aqueous phase recycling; this also helping At each step, depending on the specific experiment, additional
to reduce environmental impact. gas, saturated oil or water is injected into the cell. Following
associated gas dissolution and hydrate formation, the viscosity
An additional benefit of the HYDRAFLOW concept is that it of the system is then measured again at various rpm and the
could also potentially reduce wax deposition problems by volume of hydrates calculated from PVTX relations.
maintaining the fluid temperature for a longer time through
exothermic hydrate formation reaction, by providing solid Oil systems in the absence of AA
seeds for wax nucleation in the flowing liquid phase rather Previous studies conducted in this laboratory and elsewhere
than on pipeline walls, and by abrasion of deposited wax by have demonstrated that some oils can form transportable
fast moving hydrate particles. Finally, for companies hydrate/water/oil slurries without the need for chemical AA
considering gas transportation in the form of hydrates, additives; this favourable property being attributed to natural
HYDRAFLOW could eliminate the need for a hydrate reactor; surfactants present in the oil. To examine the potential benefits
the subsea pipeline acting as the reactor, producing hydrate in such behaviour could offer for HYDRAFLOW, and to act as a
oil and/or water slurries. reference for studies on systems with AA present, initial
experiments were carried out on simple oil-distilled water
Aims and objectives systems without any additives. In the course of experiments,
To determine the feasibility of transportation of hydrate the effect of water cut and mass fraction of hydrate present on
slurries, the aim of work to date was to investigate the the viscosity of slurries was measured for various initial
rheological behaviour of hydrate slurries for various oil and oil/gas/water volumetric ratios, as illustrated in Figure 3.
gas systems in the presence or absence of AAs. Specific Measured viscosity and hydrate fraction data for each run at
objectives were to: each oil injection stage are reported in Table 3.
• Assess HYDRAFLOW for low and high GOR oil
systems The relative viscosities obtained for systems were found to be
exponentially related to the mass fractions of hydrate (and
• Investigate the effect of anti-agglomerates on the water) present in the system, as shown in Figure 4.
above
SPE 107335 3

Following viscosity measurements on stabilised hydrate HYDRAFLOW Loop Concept for High GOR Systems
slurries, the mixer was generally stopped and the system left to For high GOR loop concept assessment, a high GOR oil was
settle for more than 24 hours at static conditions in order to prepared in the laboratory (GOR = 750 vol/vol). As with the
simulate the effect of pipeline shut-in. At the end of the shut-in low GOR studies, for simplicity, produced fluids injected at
period, the stirrer was restarted (simulating start-up) and various stages were identical, and all had the same flow rates.
viscosity measured again. For the oil systems studied (no AA Similarly, tests were conducted both with and without AA
present), no hydrate blockage occurred, with similar pre-shut present.
in viscosity values being observed on mixer restart, as shown
in Figure 5. For the system with no AA present to start the experiment, as
illustrated in Figure 10, water was first loaded into the cell (50
Effect of AA on Hydrate Slurry Viscosities vol% of reactor). For the first step, 10 vol% of the high GOR
To examine the effect of AAs on rheological properties, oil was injected. Cell temperature was then set to 4 °C.
viscosity measurements were carried out on systems with Massive hydrate formation occurred upon cooling, with this
identical initial gas/oil ratios − as illustrated in Figure 6 − in finally resulting in system blockage, meaning the test had to
the presence and absence of AA for various hydrate fractions. be abandoned. In light of the above, in all further tests on the
Increasing hydrate concentrations were achieved by stepped high GOR system, AA was added to the aqueous phase.
addition of gas.
With 1% AA present, stirrer blockage again occurred but this
Measured viscosity data are reported in Table 3 and presented time following the third injection at 49 mass% hydrates. The
in Figures 7 and 8. Results demonstrate that small doses (1-4 experiment was subsequently repeated with an alternative AA
wt. % of the aqueous phase) of AA can reduce the viscosity of at a higher concentration (4 wt% AA2). Again, stirrer
hydrate slurries significantly (up to one order of magnitude). blockage occurred when the hydrate fraction reached around
50% (Table 5).
An interesting finding for both systems with and without AA
present was that the initial formation of small fractions of Thus, as might be expected, it seems that there is an upper
hydrate resulted in a reduction in the viscosity, demonstrating limit to the mass of solid hydrate which can be formed
that hydrate/water/oil slurries can have lower viscosities than following which fluids become non-transportable, even with
parent water/oil emulsions. Further increases in the mass AAs present. However, studies do demonstrate that significant
fraction of hydrates in the system resulted in an increase in the volumes of hydrate can be mobilised in oil/water/hydrate
viscosity of slurries, as would be expected. slurries with viscosities suitable for pipeline transport, and
slurry viscosities can even be lower than those of oil-water
HYDRAFLOW Loop Concept for Low GOR systems emulsions.
As noted, a specific objective of development work was to
assess the HYDRAFLOW ‘loop’ concept for both low and high Conclusions
GOR systems. In the loop concept (Figure 1), water acts as the
• Gas can be readily converted to hydrates in the
primary circulating carrier fluid to which reservoir fluids are
presence of excess water, yielding transportable
added from wellheads at various points. To simulate this, a
oil/hydrate/water slurries with viscosities suitable for
low GOR oil was prepared in the laboratory (GOR = 60
pipeline transport.
vol/vol) and this was added in stages (simulating wellheads) to
the autoclave cell which contained the aqueous ‘carrier’ phase, • For some systems, the viscosity of hydrate/oil/water
as illustrated in Figure 9. For simplicity, produced fluids slurries can be actually lower than that of oil/water
injected at various stages were identical, were all above the emulsions.
bubble point and all had the same flow rates. The experimental
• AAs can significantly reduce the viscosity of slurries,
procedure was run twice; once with and once without AA
improving fluid transportability.
present. Measured viscosity and hydrate fraction data for both
runs at each oil injection stage are reported in Table 4. • Simulated shut-in conditions had no detrimental
effect on the transportability of slurries, with pre-
For the low GOR systems without AA, average viscosities in shut-in and post-start-up viscosities being
the Newtonian region were measured as 175, 155, 170, 193 comparable.
and 225 cP for steps 1, 2, 3, 4, and 5, respectively. Again the
initial formation of hydrates at the first injection stage reduced
the fluid viscosity relative to the parent water/oil emulsion The above results prove the concept is viable (at least under
(175 to 155 cP), confirming previous findings as discussed. laboratory conditions) and strongly suggest that
HYDRAFLOW technology could offer significant benefits over
For the systems with AA, average viscosities in the Newtonian existing flow assurance strategies, providing a novel low
region were measured as 66 and 78 for the second and third CAPEX/OPEX solution to challenging (e.g. deepwater, long
injections. Non-Newtonian behaviour was observed after the 4th tie-back, mature field) operations.
injection, hence no viscosity value is reported (Table 4). The
fifth injection stage was unsuccessful – where very large
volume of hydrate formed resulting in blockage of the impeller.
4 SPE 107335

Future work
In addition to further studies of hydrate slurry rheology for
different oil/gas/water systems and AA performance, it is
planned that future work will include:
• Further experimental assessment HYDRAFLOW ring
pipeline ‘loop’ concept, in particular for high GOR
systems.
• Upscaling of HYDRAFLOW concept studies on a
flow loop to be constructed at the Centre for Gas
Hydrate Research, Heriot-Watt in 2007.
• Modelling the pipeline pressure.
• Separating the individual effects of water-cut and
fraction of hydrates on the measured viscosities.

Acknowledgment
The authors would like to acknowledge the Scottish Enterprise
Proof of Concept Programme for financial support. The
authors also wish to thank Mr Jim Davis, Mr Kourosh Bassiti,
Dr Mike Cox, and Mr David Liddle for their support,
comments and suggestion during the course of this project.

References
1. Sloan, E.D.: “Clathrate Hydrates of Natural Gases”,
second edition, Marcel Dekker Inc., New York, USA,
(1998).
2. “Foresighting Report – Maximing Value of Mature
Oil and Gas Assets”, ITI Energy, Scotland (October
2004).
3. Oskarsson, H., Lund, A., Hjarbo, K.W., Uneback, I.,
Navarrete, R.C. and Hellsten, M.: “New Technique
for Evaluating Anti agglomerate Gas-Hydrate
Inhibitors in Oilfield Applications”, paper SPE 93075
presented at the 2005 SPE International Symposium
on Oilfield Chemistry, Houston, Feb. 2-4.
4. Larsen, R., Lund, A., Andersson, V. and Hjarbo,
K.W.: “Conversion of Water to Hydrate Particles”,
paper SPE 71550 peresented at the 2001 SPE Annual
Technical Conference and Exhibition, New Orleans,
30Sep.-3 Oct.
5. Lund, A. and Larsen, R.: “Conversion of Water to
Hydrate Particles- Theory and Application”, paper
presented at 14th (2000) Symposium on
Thermophysical Properties, Boulder, June 25-30.
6. Wolden, M., Lund, A., Oza, N., Makogen, T., Argo,
C. B. and Larsen, R.: “Cold flow black oil slurry
transport of suspended hydrate and wax solids”,
paper presented at Fifth (2005) International
Conference on Gas Hydrates, Trondheim, June.
7. Tohidi, B.: “Novel Hydrate Based Systems”,
International Patent No.WO 2006/048666 A2 (2006).
SPE 107335 5

H yd ra te
S e p a ra tio n
W E L L -1 W a te r R e c yclin g
U n it
EXPORT

W E L L -2

W E L L -4
W E L L -3

Figure 1. Illustration of the HYDRAFLOW pipeline ‘loop’ concept.

Figure 2. Schematic illustration of the high-pressure autoclave cell used for slurry
rheology studies. Various impellers, including a helical tube impeller designed in-
house, can be used for viscosity measurements.

STEP 1 STEP 2 STEP 3 STEP 4

45 %
45 %
60 %
60 %
54 %
30 % 40 %
20 %

GAS OIL WATER

Figure 3. Illustration of initial phase volumes for distilled water/oil/gas systems
investigated in turn. No AA present. Water cut for Steps 1 to 4 are 25, 33, 47, and 55%
respectively.
6 SPE 107335

300

250

Viscosity /cP
200

150

100

50

0
10 20 30 40 50 60
Water Cut, vol%

Figure 4. Viscosity of slurries as function of water cut and hydrate mass fraction.

6
Torque/mN.M

0
0 1000 2000 3000 4000 5000
T im e /m in

Figure 5. Mixer torque required to maintain constant rpm before simulated shut-in and after start-up
conditions. Zero torque periods represents the shut-in time when the system was static.

GAS

45 % O IL
W ATER

50 %

Figure 6. Schematic illustration of initial cell phase volumes used to assess effect of AA on slurry
viscosities. At constant water/oil volumetric ratios, gas was subsequently added in steps to achieve
increasingly higher hydrate contents.
SPE 107335 7

10000
P=15psia vis=157cp W/O (watercut 50%)
P=600psi vis=326cp H%=4.4
P=605psi vis=500cp H%=7.2
P=610psi vis=330cp H%=13.2
P=1100psi vis=715cp H%=15
P=1700psi vis=865cp H%=17
P=1900psi vis=810cp H%=18
1000
Viscosity / cp

100

10
0 50 100 150 200 250 300 350 400 450 500
RPM

Figure 7. Slurry viscosity as a function of mixer rpm for varying hydrate mass fractions without AA
present. Full symbols = increasing shear rate, open symbols = decreasing shear rate. H% = mass% of
hydrate, vis = average viscosity in Newtonian area.

10000
p=595psia vis=79cp H%=10.1
p=617psia vis=146cp H%=14
p=637psia vis=186cp H%=19.1
p=890psia vis=297cp H%=24
p=1549psia vis=487cp H%=30.2
1000 p=2400psia vis=686cp H%=35
Viscosity /cP

p=2800psia vis=947cp H%=38

100

10
0 10 20 30 40 50 60 70 80 90
RPM
Figure 8. Slurry viscosity as a function of mixer rpm for varying hydrate mass fractions with AA present. Full
symbols = increasing shear rate, open symbols = decreasing shear rate. H% = mass% of hydrate, vis = average
viscosity in Newtonian area.
8 SPE 107335

Figure 9. Schematic illustration of phase volumes at each oil injection stage for low GOR Loop Concept assessment
experiments.

STEP 1 STEP 2 STEP 3

5 % 10 % 15 %

50 % 50 % 50 %

GAS O IL W ATER
Figure 10. Schematic illustration of phase volumes at each oil injection stage for high GOR
Loop Concept assessment experiments.
SPE 107335 9

Table 1. Natural Gas

Composition Table 2. Oil Composition
Component Mole % Component MW Mole%
nC4 58 0.11
N2 1.70 iC5 72 0.56
nC5 72 1.24
C1 89.00
C6s 89 4.03
CO2 1.65 C7s 92 9.77
C8s 103 12.18
C2 5.50 C9s 117 9.00
C10s 132 7.70
C3 1.50 C11s 147 5.90
C12s 161 4.30
iC4 0.16
C13s 174 5.54
nC4 0.31 C14s 189 3.89
C15s 199 3.93
iC5 0.07 C16s 218 3.06
C17s 231 2.27
(nC5) + C6+ 0.11 C18s 149 2.58
C19s 157 2.71
Total 100.00 C20+ 460 21.23

Table 3. Experimental hydrate slurry viscosity data for distilled water/natural

gas/North Sea crude oil systems with and without AA present.

Water/Oil System Increasing Mass of Hydrate (without AA)

o
System T/ C P/ psia Viscosity /cP Additives

W/O 4 15 157 NA

4.4 wt% Hydrate in W/O 4 600 326 NA

7.2 wt% Hydrate in W/O 4 605 500 NA

13.2 wt% Hydrate in W/O 4 610 330 NA

15.0 wt% Hydrate in W/O 4 1100 715 NA

17.0 wt% Hydrate in W/O 4 1700 865 NA

18.0 wt% Hydrate in W/O 4 1900 810 NA

Water/Oil System with Increasing Mass of Hydrate (with AA)

10.1 wt% Hydrate in W/O 4 595 79 1 wt% AA1

14.0 wt% Hydrate in W/O 4 617 146 1 wt% AA1

19.1 wt% Hydrate in W/O 4 637 186 1 wt% AA1

24.0 wt% Hydrate in W/O 4 890 297 1 wt% AA1

30.2 wt% Hydrate in W/O 4 1650 487 1 wt% AA1

35.0 wt% Hydrate in W/O 4 2400 686 1 wt% AA1

38.0 wt% Hydrate in W/O 4 2800 947 1 wt% AA1

10 SPE 107335

Table 4. Experimental hydrate slurry viscosity data for low GOR system
simulating HYDRAFLOW Loop Concept with and without AA present.

Low GOR Oil Injection (GOR = 60 vol/vol)

o
System T/ C P / psia Viscosity /cP Additives

Step 1: Water / Oil Emulsion 4 280 175 NA

Step 2: 25-wt% Hydrate in W/O 4 600 155 NA

Step 3: 33-wt% Hydrate in W/O 4 650 170 NA

Step 4: 41-wt% Hydrate in W/O 4 1700 193 NA

Step 5: 49-wt% Hydrate in W/O 4 3600 225 NA

Step 2: 29-wt% Hydrate in W/O 4 450 66 1 wt% AA1

Step 3: 40-wt% Hydrate in W/O 4 580 76 1 wt% AA1

Step 4: 49-wt% Hydrate in W/O 4 500 NA 1 wt% AA1

Table 5. Experimental hydrate slurry viscosity data for high GOR system
simulating HYDRAFLOW Loop Concept with and without AA present.

High GOR Oil Injection (GOR = 750 vol/vol)

System T / °C P / psia Viscosity / cP Additives

Step 1: 14-wt% Hydrate in W/O 4 552 3.3 1 wt% AA1

Step 2: 49-wt% Hydrate in W/O 4 628 510 1 wt% AA1

Step 1: 17-wt% Hydrate in W/O 4 624 2 4 wt% AA2

Step 2: 54-wt% Hydrate in W/O 4 565 91 4 wt% AA2