IADC/SPE-178840-MS
Cement Placement with Tubing Left in Hole during Plug and Abandonment
Operations
Bjarne Aas, Jostein Sørbø, and Sigmund Stokka, IRIS/DrillWell; Arild Saasen, Det Norske Oljeselskap;
Rune Godøy Statoil, Øyvind Lunde, Conoco Phillips, and Torbjørn Vrålstad, SINTEF Petroleum
Research/DrillWell
Copyright 2016, IADC/SPE Drilling Conference and Exhibition
This paper was prepared for presentation at the IADC/SPE Drilling Conference and Exhibition held in Fort Worth, Texas, USA, 1–3 March 2016.
This paper was selected for presentation by an IADC/SPE program committee following review of information contained in an abstract submitted by the author(s).
Contents of the paper have not been reviewed by the International Association of Drilling Contractors or the Society of Petroleum Engineers and are subject to
correction by the author(s). The material does not necessarily reflect any position of the International Association of Drilling Contractors or the Society of Petroleum
Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the International Association
of Drilling Contractors or 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 acknowledgment of IADC/SPE copyright.
Abstract
Well abandonment operations can be very time-consuming and costly, and thousands of wells need to be
permanently plugged and abandoned offshore Norway during the upcoming years. One possible way to
reduce costs during P&A operations is to leave most of the production tubing in the well, as this would
save significant rig time. A major concern with such an approach is, however, whether the cement will
properly displace the original fluid, due to lack of tubing centralization and possible unfavorable flow
dynamics in the annulus. In this paper, we demonstrate by full-scale experimental tests that it is possible
to obtain good cement placement when the tubing is left in the hole, with and without control lines.
Full-scale tests have been performed with both conventional and expandable cement to determine the
sealing ability of annulus cement when tubing is left in hole. The quality of the cement placement was
evaluated by pressure tests with water; where leakage rates and pressure drops over the test sections were
recorded, and by visual inspection after cutting the test assemblies at different places. It is seen from the
experiments that cement is well placed in the annulus when tubing is left in hole, but some microannuli
are detected.
Introduction
Plug and abandonment (P&A) of offshore wells represents a significant part of the drilling affiliated
activity both for exploration and production wells. These well abandonment operations can be very
time-consuming. Offshore Norway for example, a large number of wells need to be permanently plugged
and abandoned during the upcoming years (Straume 2014). In direct costs, P&A costs from production
wells can be huge. Sometimes, a field’s total P&A costs can be in the range 40% to 60% of the total
decommissioning costs for that field.
The fundamental of a P&A operation is to restore cap rock functionality for maintaining the
well-integrity permanently. To succeed, an appropriate permanent barrier shall be placed across suitable
formation(s) by utilization of relevant equipment. Regardless of P&A type, operators must leave a
wellbore behind when it is secured according to local regulations. Different regulatory bodies have their
2 IADC/SPE-178840-MS
own requirements and operators must strictly adhere to local well-abandonment regulations. Local
regulations are the minimum requirements and have changed considerably over the years to facilitate P&A
operations in a safe and proper way. Besides, some operators have their own internal standards and tend
to follow them where the regulatory authorities do not provide minimum standards.
Several attempts have been performed to minimise the costs of P&A operations. Although well
cements are the most common, several other material types have been suggested, and some are used.
Application of concentrated sand as alternative to cement has shown to reduce the operational time
(Saasen et al., 2011). A summary of such materials was presented by Khalifeh (2013). Others have studied
cost reduction of subsea well P&A by use of boats instead of drilling rigs to reduce cost (Moeinikia et al.,
2014, 2015), and techniques have been implemented to make section milling more efficient (Scanlon et
al., 2011). Furthermore, other techniques have been developed to perforate, wash and cement the annulus
in a remedial operation within the P&A operation, as described by Ferg et al. (2011), without using section
milling at all.
Normally, these P&A operations are conducted by removing completion equipment followed by
placement of a series of cement plugs. One possible way to reduce costs during P&A operations is to leave
most of the production tubing in the well, as this would save significant rig time. A major concern with
such an approach is, however, whether the cement has properly displaced the original fluid in the annulus
outside the tubing where the P&A plug is planned to be placed. Poor cement quality here may be due to
designing the original cement job with too low cement slurry volume, lack of tubing centralization or
possible unfavorable flow dynamic conditions in that annulus. Independent on the choice of equipment
used for performing the P&A operation, the most economical way is to leave as much of the tubing as
possible in the well during the operation. Hence, if tubing can be left in hole in a way satisfying the long
term abandonment criteria, a significant restriction preventing cost efficient P&A has been overcome.
In this paper, it is demonstrated by full-scale experimental tests that it is possible to obtain good cement
placement when the tubing is left in the hole. This is the case both with and without control lines attached
to the tubing. This is demonstrated by full scale tests, where several assemblies with 7⬙ tubings in 9 5/8⬙
casings were cemented. In one test series conventional cement was used, and in another test series
expandable cement.
Experimental equipment and methods
Full-scale tests, with several assemblies of 7⬙ tubings cemented in 9 5/8⬙ casings, were performed to
determine the sealing ability of annulus cement when tubing is left in hole. Tests were performed with
both conventional cement and expandable cement, and with and without control lines present. An
overview of the different test assemblies is given in Table 1.
Table 1—Overview of different test assemblies in large-scale
cementing tests. All tests performed with 7ⴖ tubings cemented
in 9 5/8ⴖ casings.
Assembly # Cement type Control lines
Conv-A Conventional No
Conv-B Conventional Yes
Exp-A Expandable No
Exp-B Expandable Yes
Exp-C Expandable Yes
The quality of the cement placement was determined by pressure tests with water; where leakage rates
and pressure drops over the test sections were recorded, and by visual inspection after cutting the test
assemblies at different places.
IADC/SPE-178840-MS 3
Tests with conventional cement
For the tests with conventional cement, two assemblies were used (A and B). Both assemblies consisted
of 3 casing lengths; i.e. they were 36 m long in total. The assemblies were inclined to 85° off vertical. Fig.
1 shows the tubings and casings used, where assembly B had a two lead control cable clamped to the
tubing. Cable clamps on tubing collars created an offset of 26 mm, corresponding to an upward
eccentricity of 6.0%. Assembly A had no control cable or cable clamps. The tubing rested on the casing
wall. The tubing collars created an offset of 11 mm, corresponding to a downward eccentricity of 7.8%.
Figure 1—Pictures of 7ⴖ tubings and 9 5/8ⴖ casings used in the tests, inluding control line at one assembly.
In the cementing process, a 1.92 s.g. conventional Dyckerhoff class G cement slurry displaced a 1.2
s.g. brine. Cement was pumped down the tubing, following a wiper ball, out into the annulus through
perforations at the lower end of the tubing, and up the annulus. The pumping proceeded relatively slowly,
at ca. 300 L/min, in order to have gravity help with the displacement of a brine, simulating the completion
fluid, as presented by Jakobsen et al. (1991) and Kroken et al. (1996) for conventional primary cementing
jobs.
The cement was provided, mixed and pumped by a major service company. Fig. 2 shows the cementing
equipment and full-scale test set-up during placement operations. The assemblies were insulated by
Rockwool, in order to keep the cement warm during curing. Fig. 3 shows the temperature development
during the curing period. The temperature reached a maximum of ca. 75°C, after about one day.
Figure 2—Pictures of cementing equipment and full-scale test set-up during cement displacement.
4 IADC/SPE-178840-MS
Figure 3—Temperature development in assemblies during curing with conventional cement. Sensor on Assy. B malfunctioned during
second day.
The assemblies were provided with pressure sensors, which measured the internal pressure at different
places along the casing annulus and on the tubing stub. Fig. 4 shows a schematic illustraction of Assembly
B with placement of sensors. The sensor ports also served as channels for water injection or leakage
measurement.
Figure 4 —Illustration of Assembly B with conventional cement: Green line is control cable, blue rectangles are cable clamps, PT1– 6
are pressure sensors, and TT1 is temperature sensor.
After about one week of curing, flow tests were performed on the assemblies, where water was injected
through PT1 and drained through PT3 (see Fig. 4). Flow rate out of PT3 was then measured, together with
the pressures at PT2, PT4, PT5 and PT6. Due to the cement shrinkage there was relatively good pressure
communication along the annulus. PT2 and PT4 read essentially the same pressure, and the pressure on
PT5 is representative for the internal pressure at that end of the assembly. Pressure differences between
PT1, PT2 or PT4, and PT5 gave the frictional pressure drop from the flow along the annulus.
The flow tests in the casing annuli were performed with ca. 50 and 100 bar internal pressures in the
casing. All tests were performed at outside temperatures. After finishing the pressure tests, both
assemblies were cut open for visual inspection of cement placement quality.
Tests with expandable cement
For the tests with expandable cement, three assemblies (A, B and C) were used, and all these were 12 m
long, i.e. one standard casing section. As for the previous tests with conventional cement, the assemblies
were inclined to 85°.
This time the control lines were strapped to the high side of the tubing, and cable clamps were not used.
The eccentricity of the tubing in the casing was the same as if the tubing had rested on the bottom of the
casing. I.e., the 11 mm stand-off, corresponding to 7.8% downward eccentricity, was defined by the
diameter of the tubing collars. The assemblies B and C had broad, flat control cables stretched along the
IADC/SPE-178840-MS 5
top side of the annulus, while assembly A had no cable. Fig. 5 shows pieces of the control cables in
Assembly B and C. On the outside of the outer casing in each assembly, the assemblies were insulated
by a layer of Rockwool. Inside the Rockwool layer, heating cables were wrapped around the casings and
used for regulating the temperature. Each cable could deliver up to 3 kW heating power.
Figure 5—Control cables used in tests with expandable cement; at Assembly B (left) and cable at Assembly C (right).
As for the previous tests, the cement was pumped relatively slowly through the assemblies, by ca. 300
L/min, where 1.92 s.g. cement displaced 1.0 s.g. water. Again, the amount of cement pumped corre-
sponded to setting a 100 m long plug. The cement was provided, mixed and pumped by a major service
company.
After the cement had been placed, the heating was turned on and kept on until temperatures in the
assemblies had reached 90° C. It was then turned off, and chemical processes in the cement increased the
temperature further to ca. 110° C, before it, after 7 to 8 hours, again started to fall. When it had sunk to
90° C, the thermostats turned the heating back on and kept the temperatures at 90° C for the rest of the
testing period. Fig. 6 shows the temperature developments at the middle of the assemblies for the first two
days. Furthermore, the pressure in the assemblies was kept at ca. 20 bar, where the elevated pressure and
temperature kept the conditions inside the assemblies somewhat more similar to downhole conditions.
Figure 6 —Temperature development in assemblies during curing with expandable cement.
6 IADC/SPE-178840-MS
Pressure tests were performed after waiting three weeks for cement to cure, thereby allowing the
expandable cement time to expand. During flow tests, the injected fluid was heated to a temperature
between 90° and 100° C. This was done to prevent cold fluid from cracking the hot cement behind the
injection ports. As an example of the test set-up, Fig. 7 shows a sketch of Assembly B, with placement
of sensor ports, which also served as channels for water injection or leakage.
Figure 7—Illustration of Assembly B with expandable cement: Green line is control cable, PT1–7 are pressure sensors, and TT1 is
temperature sensor.
The flow tests were performed with injection pump pressures at ca. 70 and ca. 120 bar. For leakage
tests in the casing annulus, normally the port at the lower end of the assemblies (PT1) was used. Usually,
leakage rates were measured from the upper end (PT5 or PT6), but sometimes from a middle port (PT3
or PT4). Two to four leakage measurements were taken at each position, and the calculated values for the
effective micro annuli averaged.
After finishing the pressure tests, assemblies A and B were cut open for visual inspection of cement
placement quality. Assembly C was left uncut, in order to enable further tests. After the water flow tests
on assembly C had been finished, flow tests were continued with light oil (iso-paraffin), to determine the
effect of fluid on sealing ability. For the duration of that test, the fluid coming out of the leakage ports
was a mixture of water and oil.
Calculation of ⴖeffectiveⴖ microannuli
During the pressure tests, the fluid flow through the assemblies was measured at different pressures, and
resulting microannuli were calculated from these flow measurements. This calculated, ⬙effective⬙ mi-
croannulus is defined as the concentric, uniform microannulus which would give the measured pressure
drop at the measured flow rate. It translates pressure drops and leakage rate measurements into a coarse
geometric measure for the space available for leakage between cement and steel wall. Although it is
assumed in the calculations that the microannuli are uniform and homogeneous, they are in reality
eccentric and irregular in shape, which has been experimentally shown recently by Vrålstad et al. (2015).
But, since the term microannulus is used in the industry in releation to fluid flow, the flow measurements
have been converted quantitatively into ⬙uniform⬙ microannuli for comparative reasons.
At sufficient high pressures during pressure tests, the casing expanded and the cement contracted due
to the pressurization, which resulted in the creation of an induced microannulus between cement and steel
wall, and these induced microannuli also influenced the measured flow. Appendix A gives the formulas
used for calculating effective microannuli; both permanent and induced.
Results
Tests with conventional cement
As an example from a test on Assembly A, Table 2 shows measured pressures and flow rates, as well as
the calculated microannuli; both permanent and induced. Fig. 8 shows a measurement series in assembly
A. The cement presses against the casing wall at the PT3 port, and chokes the flow out of the port.
IADC/SPE-178840-MS 7
Therefore, the pressure drop along the casing annulus is relatively small. Similar results were also
observed for Assembly B.
Table 2—Pressures and flow rates in Assembly A with calculated microannuli gaps, as well as calculated increase in radius due to
casing expansion and cement deformation during pressurization.
Pressure PT1 Pressure PT1-PT4 Leakage point Leakage Calculated microannulus ␦R induced casing ␦R induced cement
(bar) (bar) (ml/min) (m) (m) (m)
54 0.5 PT3: 14 65 28 8
93 2.5 PT3: 95 48 14
56 3.5 PT3: 16 56 29 8
PT2: 40
94 4.5 PT3: 43 48 14
PT2: 93
Figure 8 —Flow measurement in Assembly A with conventional cement. Water inlet through PT1 and outlet through PT3.
The cement had cured for only 8 to 9 days when the flow tests were performed. After a couple of more
weeks, the cement could have shrunk more (Reddy et al., 2007) and larger microannulus gaps could have
been seen.
If the measured leakage rates should be explained by flow through permeable cement only, an
⬙equivalent⬙ cement permeability of ca 3 Darcy for the lowest annular pressure points would be needed.
After finishing the flow tests, the assemblies were cut crosswise through at different places at the ends
and middle of the assemblies. Fig. 9 shows that there seems to be perfect cement displacement, also
around the control lines.
8 IADC/SPE-178840-MS
Figure 9 —Cut test assemblies with conventional cement. Without control lines (left) and with control lines (right).
Tests with expandable cement
As an example, Table 3 shows measured pressures and flow rates from Assembly C, as well as the
calculated microannuli; both permanent and induced. Furthermore, Fig. 10 shows the flow measurement
through Assembly A. Injection port is PT1, at lower end of the assembly. Between 11:02 and 12:30, the
leakage port was PT6, at the upper end of the assembly. The pressure drop along the upper half of the
assembly is given by the readings at PT3 or PT4, minus the reading at PT5. The pressure drops are large
enough relative to the pump noise to give meaningful measurements. Between 10:05 and 11:02, the
leakage port was PT4, at the middle of the assembly. I.e., leakage was measured along the lower half of
the assembly. The pressure drop is now given by pressure readings at PT2 minus PT5. Here, the pressure
drops are too low relative to the pump noise to be measurable. Similar results were observed for all three
assemblies.
Table 3—Pressures and flow rates in Assy. C with calculated microannuli gaps, as well as calculated increase in radius due to
casing expansion and cement shrinkage during pressurization.
Pressure PT3 Pressure PT3 - PT6 Leakage ␦R induced casing ␦R induced cement Calculated microannulus
(bar) (bar) (ml/min) (m) (m) (m)
125 48 98 44 15 22
126 48 94 44 15 22
127 49 92 45 15 22
96 50 49 32 11 18
95 51 48 31 10 17
66 47 23 19 6 14
65 44 23 19 6 14
42 30 13 9 3 13
42 30 13 9 3 13
IADC/SPE-178840-MS 9
Figure 10 —Flow measurement in Assembly A with expandable cement. Water inlet through PT1 and outlet through PT4 and PT6.
If the measured leakage rates should be explained by flow through permeable cement only, an
⬙equivalent⬙ cement permeability of 11 to 13 mDarcy for the lowest annular pressure points would be
needed.
After finishing the flow tests on Assemblies A and B, the assemblies were cut crosswise through at
different places at the ends and middle of the assemblies. Fig. 11 shows that there seems to be perfect
cement displacement, also around the control lines.
Figure 11—Cut test assemblies with expandable cement without control lines (left) and with control lines (right).
Assembly C was also tested with light oil as leak fluid, where the oil was pumped in through PT1 at
the lower end of the assembly, and let out through PT5 at the upper end. Fig. 12 shows the measured flow
rates of both water and oil as function of the pressure at the middle point of the casing annulus, PT3. It
is seen that the flow is substantially restricted relative to the water tests, probably due to capillary effects.
The flow rate was practically zero for PT1 below ca. 80 bar.
10 IADC/SPE-178840-MS
Figure 12—Flow tests with water and oil in Assembly C with expandable cement.
Fig. 13a shows flow measurements on the upper half of the casing annulus of Assembly C with
expandable cement. Water is injected through PT1, at the lower end of the assembly. The pressures at PT6
(blue points), at the upper end of the assembly, close to the leakage port, are quite proportional to flow
rates. The pressure at PT3 (red points), at the middle of the assembly, are proportional to flow rates only
up to about 80 bar. From then on, the pressure increases less than linear with flow rate. In Fig. 13b, the
pressure difference between PT3 and PT6, i.e. over the upper half of the assembly, is included (green
points). It can be seen that the pressure difference is fairly proportional to flow rate up to about 50 bar.
From then on, it decreases slightly with increasing flow rate. The crossing point of the two green lines can
be determined to be at PT3 ⫽ 80 ⫾ 2 bar. Up to this annular pressure there is only a residual (i.e.
permanent) microannulus in the casing annulus, which is fairly independent of the pressure. Above this
pressure, there is a ⬙real⬙ (i.e. induced) microannulus between cement surface and steel wall. As the
annular pressure is increased, this induced microannulus will extend further up the annulus. By coinci-
dence, this proceeds in a way so that the pressure at PT3 remains almost constant with increasing flow
rate.
Figure 13—Flow tests in Assembly C with expandable cement. Green points: pressure drop along casing annulus.
Discussion
It is seen from the pictures of the cut assemblies in Figs. 9 and 11 that the lack of tubing centralization
does not prevent cement placement in the annulus. It is also seen that the cement is well placed around
IADC/SPE-178840-MS 11
the control lines as well. In other words, the experiments show that it is possible to obtain good cement
placement outside tubing when the tubing is left in hole. Although the results for measured flow and
calculated microannuli in Tables 2 and 3 show that there are leakages through the assemblies, the
measured flow rates are relatively low and should not have significant practical consequenses in a real
well with several hundred meters axial length of cemented annulus.
However, it should be noted that the pumping proceeded relatively slowly during the cement placement
in these experiments, in order to have gravity help with the displacement as described by Jakobsen et al.
(1991) and Kroken et al. (1996). The displacement process was therefore idealized in the current set-up,
and in a field application the cement slurry could be vulnerable to effects like U-tubing (Nelson and
Guillot 2006). Also, these displacement results should thus not necessarily be extrapolated directly to
primary cementing situations in the field, since the annulus may be filled with a drilling fluid having a
more complex rheological performance. It has however been reported that such a slow pumping rate
during cement placement has been used successfully in the field (Kroken et al. 1996).
It is also seen from the results in Tables 2 and 3 that the test assemblies with expandable cement
provide better sealing than the test assemblies with conventional cement. This is probably caused by the
expanding nature of the expandable cement, as opposed to conventional cement, and it can be concluded
that expandable cement provides better sealing than conventional cement.
It should however be noted that the use of expanding agent is not the only difference between the test
set-ups for these two cases: The assemblies with expandable cement were heated externally and kept at
a temperature of 90 °C, whereas the assemblies with conventional cement where not heated and thus
exposed to the ambient temperature of about 10 °C during curing. Therefore, by comparing the
temperature development in the samples (Figs. 3 and 6), it is seen that the assemblies with conventional
cement experienced an internal temperature difference of about 60 °C during cement curing (i.e. between
about 10 °C and 70 °C, as seen in Fig. 3), whereas the assemblies with expandable cement experienced
a temperature difference of only about 20 °C (i.e. between about 90 °C and 110 °C, as seen in Fig. 6).
It is possible that this large temperature difference for the conventional cement assemblies resulted in
casing expansion during cement curing, thereby contributing to microannuli formation at the cement-
casing interface. The observed differences between the assemblies should therefore not be attributed
solely to the use of expandable cement.
Furthermore, with respect to measured microannuli, it is important to separate between the two
⬙different⬙ types of microannuli; i.e. ⬙permanent⬙ microannuli and the microannuli that are induced during
pressure testing. In reality, these induced microannuli are also likely to be non-uniform. In the tests with
conventional cement, which shrunk slightly during curing, permanent microannuli were observed, also
without any internal pressures in the assemblies. With expanding cement however, the cement was
pressing against the steel walls, and a significant internal pressure was needed to create a microannulus
with sufficient flow. During the calculation of microannuli thickness, it was necessary to include both of
these types in order to explain the measured leakage rates (Tables 2 and 3, and Fig. 13). It is likely that
the formation process and resulting geometries of microannuli will be different in a real well situation,
where there is external confinement around the casing.
Conclusions
The experiments show that it is possible to have cement well placed in the annulus also when tubing is
left in hole. Some microannuli are detected in the two series of full-scale cement placement experiments.
The calculated microannuli are relatively small and also probably non-uniform, and should not open for
large leakage rates in real P&A operations because of the axial length of the cemented annulus.
Furthermore, in the tests, the presence of control lines did not represent any additional leakage paths.
12 IADC/SPE-178840-MS
Acknowledgements
This work has been performed by the DrillWell research centre - Drilling and Well Centre for Improved
Recovery, which is a research cooperation between IRIS, NTNU, SINTEF and UiS, and was funded by
the Research Council of Norway, ConocoPhillips, Det norske oljeselskap, Lundin Norway, Statoil,
Talisman, Total and Wintershall.
Statoil is greatfully acknowledged for providing the 7⬙ tubings and 9 5/8⬙ casings used in the
experiments, Roxar is acknowledged for providing cable clamps, and the authors want to thank Gunnar
Lende from Halliburton for technical input and fruitful discussions during cement selection and displace-
ment preparations.
References
Ferg, T. E. Lund, H. J. Mueller, D., Myhre, M., Larsen, A., Andersen, P., Lende, G., Hudson, C., Prestegaard, C. and Field,
D. 2011. Novel Approach to more Effective Plug and Abandonment Cementing Techniques. Paper SPE 148640
presented at the SPE Arctic and Extreme Environments Conference & Exhibition held in Moscow, Russia, 18 –20
October
Jakobsen, J., Sterri, N., Saasen, A., Aas, B., Kjøsnes, I. and Vigen, A. 1991. Displacements in Eccentric Annuli During
Primary Cementing in Deviated Wells, paper SPE 21686 presented at the Production Operations Symposium,
Oklahoma City, April 7–9, 1991.
Khalifeh, M., Saasen, A., Hodne, H. and Vrålstad, T., 2013. Techniques and Materials for North Sea Plug and
Abandonment Operation, paper OTC 23915 presented at the Offshore Technology Conference held in Houston, Texas,
USA, 6 –9 May
Kroken, W., Sjaholm, A.J., Olsen, A. S., 1996. Tide Flow: A Low Rate Density Driven Cementing Technique for Highly
Deviated Wells. Paper SPE-35082-MS presented at the SPE/IADC Drilling Conference held in New Orleans,
Louisiana, USA, 12–15 March 1996
Moeinikia, F., Fjelde, K.K., Sørbø, J., Saasen, A. and Vrålstad, T., 2015. A Study of Possible Solutions for Cost Efficient
Subsea Well Abandonment, paper OMAE2015– 41261 presented at the ASME 2015 34th International Conference on
Ocean, Offshore and Arctic Engineering, May 31 - June 5, St. John’s, Newfoundland, Canada.
Moeinikia, F., Fjelde, K. K., Saasen, A., Vrålstad, T., & Arild, Ø. 2014. A Probabilistic Methodology to Evaluate the Cost
Efficiency of Rigless Technology for Subsea Multiwell Abandonment. SPE Production & Operations, SPE-167923-
PA, Doi: 10.2118/167923-PA
Nelson, E.B. and Guillot, D., 2006. Well Cementing. Schlumberger, 2nd edition.
Reddy, B.R., Xu, Y., Ravi, K., Gray, D. and Pattillo, P. D., 2007. ⬙Cement Shrinkage Measurements in Oilwell Cementing
– A comparative Study of Laboratory Methods and Procedures.⬙ Paper SPE-103610 presented at the SPE Rocky
Mountain Oil & Gas Technology Symposium, Denver, Colorado, USA, 16 –18 April, 2007.
Saasen, A, Wold, S, Ribesen, B. T., Tran, T.N., Huse, A., Rygg, V., Grannes, I. and Svindland, A., ⬙Permanent
Abandonment of a North Sea Well Using Unconsolidated Well Plugging Material⬙, SPE Drilling and Completions,
vol. 26, pp. 371–375, 2011. doi: 10.2118/133446-PA
Scanlon, E. Garfield, G. Brobak, S. 2011. New Technologies to Enhance Performance of Section Milling Operations that
reduces Rig time for P&A Campaign in Norway. Paper SPE/IADC-140277 presented at the SPE/IADC Drilling
Conference and Exhibition held in Amsterdam, The netherland, 1–3 March 2011
Straume, M., 2014. Need for new and cost effective P&A technology. Presentation given at Norwegian Plug &
Abandonment Forum (PAF) Workshop arranged by the Norwegian Oil and Gas Association, Oct 2014
Vrålstad, T., Skorpa, R., Opedal, N., De Andrade, J., 2015. Effect of Thermal Cycling On Cement Sheath Integrity:
Realistic Experimental Tests and Simulation of Resulting Leakages. Paper SPE-178467 presented at the SPE Thermal
Well Integrity and Design Symposium held in Banff, Alberta, Canada, 23–25 November 2015
IADC/SPE-178840-MS 13
Appendix A: Formula for microannulus gaps
When a casing with radius RC and wall thickness hC is put under internal pressure P, the radius will
expand by
where Est is Yong’s modulus of steel.
A cement sheet in the casing annulus of thickness ⌬Rcem will shrink by
where Ecem is Yong’s modulus of cement.
The radius of a cement cylinder within the tubing of radius RT, will shrink with pressure as
An effective microannulus gap, ␦R, can be determined from a flow test with the formula
where Q is fluid flow rate, is the fluid viscosity, and ⌬P is the pressure drop along the assembly
length L.
The above formulas will be modified by Poisson’s number effects, depending on geometry. These
effects are included in the calculations given in the article.
