SPE 130159

Efficiency in Cement Jobs for Fresh Water Formations Isolation During Plug
and Abandonment in Canadon Seco Field, South Argentina
H.J. Graterol, SPE, YPF

Copyright 2010, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE/ICoTA Coiled Tubing and Well Intervention Conference and Exhibition held in The Woodlands, Texas, USA, 23–24 March 2010.

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 have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of 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 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 SPE copyright.

Abstract
In the San Jorge Basin, located in South Argentina, Well Plug and Abandonment (P&A) regulations include the
guarantee of fresh water formations isolation from once productive formations. Cement Bond and Variable Density
Logging (CBL/VDL) are recorded after a squeeze job to confirm cement integrity in the annulus, allowing protecting
the most important fresh water formation in the basin: Patagoniano.
Historically, to isolate the Patagoniano multiples of costly squeeze jobs were done to achieve an appropriate result
in the CBL/VDL. The main problem from this shallow formation is its high degree of fluid losses, most slurry is lost
in these unconsolidated and high-permeability shallow sands (while pumping and during cure). Under this condition
costs in the Canadon Seco Field for P&A were exceeded.
To overcome this problem, a review of previous results was done and many flaws were identified in the slurry
physical characteristics and the used placement techniques. A new slurry and procedure was selected (after being
laboratory tested) to be applied in the field; then a campaign of 14 wells was done using the selected slurry and
following a general technique; cost/time reduction was drastic in comparison with previous P&A operations.
This paper presents the whole procedure for achieving these results and also includes details in each phase of
testing/design and execution. Geological and mechanical features of the wells are shown, as well as a description
of the new technique / slurry (including concepts) and the obtained field results.

Introduction
Based on the needs of changes during the fresh water formations isolation in Plug & Abandonment (P&A) jobs in
the Canadon Seco field (South Argentina Business Unit), the P&A engineering group began to review some
cementing concepts to analyze and evaluate previous results in this area. Some of the concepts reviewed were:
Static Gel Strength (SGS), filtrate loss, thickening time, acoustic impedance, among others. In order to have higher
success probabilities a whole procedure was proposed, evaluated in laboratories and executed on field.
In 2008 13 wells under similar conditions (fresh water aquifer exposed) were P&A in this area, in 2009 44 wells
were P&A with Workover rigs and a trend of time reduction was observed mainly because of operational
optimization but not due to a reduction in effective cementing operations. Then, the new methodology / slurry (after
being tested in the laboratory) was used on the field during the P&A campaign with 2 Coiled Tubing and 1
Workover rig, convincing positive results were obtained in 14 wells.

Wells: Mechanical / Geological Features

a.- Geological Characteristics:
The base of the Patagoniano formation in Canadon Seco field is located at an average of 320 mts (vertical depth);
figure 1 show an open hole logging of a well in the area.
Different sand bodies are present in this formation, including a sandstone body containing formation water with
high salinity at an average of 200 mts, a second sandstone body containing the fresh water at an average of 140
mts (base) and a third group of sandstone layers containing salty water (a an average of 394 mts), all of them
separated with layers of shales. All of these sandstone bodies contribute to the high level of admissions found in
this formation.
2 SPE 130159

Fresh Water Aquifer

(Depth Average: 140
mts)

Salt Water Aquifer

(Depth Average:
200 mts)

Formation Base
Patagoniano. Depth
Average: 320 mts
Figure 1. Open Well Log from shallow formation in Canadon Seco

b. - Degree of fluid losses severity

Due to shallowness of the Patagoniano formation, the unconsolidated (and extremely high permeability) sands
show high level of fluid losses. Historically this problem has been confronted (without proven or continuous
success) using sodium metasilicate (Na2SiO3) pads, glass fiber, two stages cementing squeeze jobs, among
others additives and procedures

In the following lines a summary of this fluid losses severity is shown; sandstones bodies (in the Patagoniano)
feature high permeability (unconsolidated sands), where traditional formation plugging fluids are not enough
(sodium metasilicate, glass fiber) to contain the fluid losses. Then, most of the placed cement is lost into the
formation and no proper isolation is seen on the CBL in one single operation under previous techniques / slurries.

Fluid Loss Tests Averages:

- Test Rate: 1,66 bpm
- Surface Test Pressure: 195 psi
- Perforations Depth: 289 mts
- Bottom Test Pressure (at the perforations): 599 psi
- Fluid Loss Index (rate in kilo barrels per minute by unit of pressure applied): 2,78 kbpm/psi
Low temperatures are also found on this shallows depths (29°C), therefore cure time and other physical
characteristics of the slurries are affected.

c. - Well Mechanical Features

Most wells to be abandoned in the area are considerably aged (14 wells P&A with the new methodology has an
average of 45 years old). The aged wells bring significant issues to a satisfactory result (especially in cement
bonding quality), for instance: casing corrosion, unknown annular fluids, low casing resistance, and excessive
caliper, among others. Casing varies from 5 ½” (6) to 7” (4) and 6 5/8” (4).

Cementing Concepts
Some cementing / slurries concepts are reviewed in the next few lines; many of them have been historically
forgotten in the pumped slurries, and by analyzing each one of them individually some of the poor results in the
past can be explained, as well as contributing to the new proposal and basis for laboratory analysis.
SPE 130159 3

a. - Thickening Time (pumpability)

Impact in our operation: When squeeze jobs are done in this high fluid-loss formation, one of the key for a
successful operation is to achieve elevated closure pressures. If cement slurry has an extensive thickening time,
this closure pressure is hard to achieve.
An additional technique is used in the proposed and applied new methodology is the “Hesitation” pumping, where
pumping is stopped at regular intervals allowing the slurry to accelerate its gelling and also a reduction in
operational time is achieved by reaching closure pressures sooner.
b. - Slurry Rheology
Impact in our operation: Tixotropy effect is a key to achieve earlier closure pressures. By evaluating a cement
slurry from the rheological point of view, a particular slurry can be neglected (or approved) based on its Gel
formation properties (10 seconds / 1 minute / 10 minutes) and Power Law coefficients (k´ and n´).
c. - Static Gel Strength - SGS
Generic Concept: When cement changes from liquid (slurry) to solid (during cure time), its matrix develops a
behavior between these two stages. SGS represents the slurry / cement resistance to flow under any given shear
stress. Gelation characteristics allow us to know how the slurry will behave during cure time (providing information
of cement loss in that time); units Lbs/100ft^2. SGS can be measured using an UCA (Ultrasonic Cement Analyzer)
where a relationship between transit time and SGS generates the changes in time. It is considered that a slurry
under the 100 lbs/100ft^2 behaves like liquids and when SGS reaches 500 lbs/100ft^2 and above it is considered a
solid (during transition time between 100 and 500 lbs/100ft^2 slurry behaves like a polymer).

Impact in our operation: During cure time and before it becomes solid, the slurry column generates hydrostatic
pressure; fluid (slurry) loss may occur in lower deconsolidated (and extremely high permeability) sands. By taking
into account this property in the slurry / cement design, a major improvement has been made and real field results.
d. - Filtrate Loss
Impact in our operation: If water is lost from the cement slurry, the matrix will be dehydrated and setting process
will be affected, therefore final result (evaluated with CBL) will not be successful (less attenuation properties).
e. - Compressive Strength
Impact in our operation: By knowing the compressive strength evolution with time, Wait on Cement (WOC) time
can be optimized, it will allow us to determine when to rotate the cement (optimizing mill selection).
f. - Acoustic Impedance
Impact in our operation: By knowing the acoustic impedance evolution with time, waiting time for CBL running can
be optimized. Also, with ultrasound logging devices (commonly used in the basin), materials bonded to the casing
can be determined (this logging allows to measure this property) and comparison with laboratory tested cements
will tell us if the material is the pumped cement.

Slurry Tests in Laboratory / Analysis

With the objective of selecting the technical-economical best-fit slurry, many mixtures were tested (up to 7 different
slurries: A, B, C, D, E, F and G), and full analysis of the properties mentioned above was done. Details of the
slurries and its additives are shown in Table 1, and laboratory results are shown in Table 2.
4 SPE 130159

Add. 1 Add. 2 Add. 3 Add. 4 Add. 5 Add. 6 Comments

A Calcium Sodium Dispersant - - - - Typically (historical) used slurry.
Chloride Methasilicate @ 0.3% - High pumpability and elevated
@ 2% @ 0.3% BWOC filtrate loss.
BWOC BWOC - Rejected
B Calcium Sodium Dispersant Expansive - - - High pumpability and elevated
Chloride Methasilicate @ 0.3% @ 1% filtrate loss.
@ 2% @ 0.3% BWOC BWOC - Volume did not increase (0.0036
BWOC BWOC %) with the expansive additive.
- Rejected.
C Calcium Sodium Dispersant Expansive - - - High pumpability and elevated
Chloride Methasilicate @ 0.3% @ 5% filtrate loss.
@ 3% @ 0.3% BWOC BWOC - Volume did not increase (0.05%)
BWOC BWOC with the expansive additive.
- Rejected.
D Calcium Sodium Dispersant Expansive Filtrate - - High pumpability and good filtrate
Chloride Methasilicate @ 0.3% @ 5% loss control.
@ 3% @ 0,3% BWOC BWOC reducer - Volume did not increase (0.05%)
BWOC BWOC @ 0.5% with the expansive additive.
BWOC - Rejected.
E Calcium Sodium Dispersant Expansive Filtrate Microsilicium - Moderate pumpability and good
Chloride Methasilicate @ 0.3% @ 5% loss @ 5% filtrate control.
@ 3% @ 0.3% BWOC BWOC reducer BWOC - Volume did not increase (0.05%)
BWOC BWOC @ 0.5% with the expansive additive.
BWOC - Rejected by pumpability,
operational difficulties during mixing
and costs.
F Calcium Sodium Dispersant Expansive Filtrate Microsilicium - Moderate pumpability and good
Chloride Methasilicate @ 0.3% @ 5% loss @ 5% filtrate control.
@ 3% @ 0.3% BWOC BWOC reducer BWOC + - SGS Test:
BWOC BWOC @ 0.5% Latex @ 5% SGS @ 100 lbs/100ft^2: 11
BWOC BWOW minutes
SGS @ 500 lbs/100ft^2: 47
minutes
- Rejected by pumpability,
operational difficulties during mixing
and costs.
G Calcium - Dispersant - Filtrate Microsilicium - Short pumpability and good filtrate
Chloride @ 0.3% loss @ 3% control.
@ 3% BWOC reducer BWOC - Pumpability under hesitation is
BWOC @ 0.5% measured (55 minutes).
BWOC - SGS Test:
SGS @ 100 lbs/100ft^2: 35
minutes
SGS @ 500 lbs/100ft^2: 66
minutes
- Approved for Field Tests.
Table 1. Tested Slurries in Laboratory

Test Conditions:
- Bottom Hole Static Temperature (BHST): 36°C
- Bottom Hole Circulating Temperature (BHCT): 29°C
- Depth / Temperature average: 300 mts @ 1.7 °F/100 ft
SPE 130159 5

Thickening Time
Rheology @ 100 Bc UCA
Comp. Acoust.
Yield Filt. Strength Imp.
Plast. Point Loss Hesita @ 12 @ 12
Dens. Gel @ Gel @ Visc. lbf/100ft (ml/30 Conv. tion hours hours
Sl. (ppg) n´ K´ 10 seg 10 min (cP) 2 min) (min.) (min. (psi) (MRayl)
A 15,72 0,52 0,03 0 2,7 51 27 741 210 1450 5,04
B 15,78 0,12 0,64 49,7 116,3 22,5 101,5 668 147 1250 4,90
C 15,97 0,26 0,35 35,9 44,7 63,75 106,25 608 153 1225 4,89
D 15,67 0,42 0,14 51,2 68,7 97,5 78,5 13 138 1200 4,85
E 15,65 0,34 0,25 62,5 71,3 96 106 14 122 1050 4,63
F 15,63 0,43 0,14 88,3 140,6 105 81 13 110 1400 4,93
G 15,69 0,15 0,76 102,2 113,1 41,25 139,75 29 90 55 1250 4,88
Table 2. Summary of Test Results in Laboratories

Selected Slurry (“G”)

The combination of additives (Calcium Chloride @ 3% BWOC, Filtrate Loss Reducer @ 0.5 % BWOC,
Microsilicium @ 3% BWOC and Dispersant @ 0.3% BWOC), has provided the slurry to develop the main
properties identified as critical:
- Short time Pumpability (<90 minutes and 55 minutes under hesitation).
- Filtrate Loss Control (<30 ml/30 min).
- Reduced SGS Times (<70 min for 500 lbs/ft^2; 66 minutes for the selected slurry, see Figure 2).
- Excellent tixotropic characteristics.
Additional Notes:
- “F” Slurry presented a slightly better characteristics than the selected one (for instance SGS for 500
lbs/ft^2 of 47 minutes), but its economical evaluation was not positive (included latex, expansive additive,
micro-silicium).
- The reduction of Microsilicium from 5% to 3% responds to difficulties during mixing at surface.
With the selection critical objectives are achieved:
- Reduce slurry loss into the formation during pumping and cure time.
- Reduce premature dehydration of the slurry.
- Achieve elevated closing pressures during squeeze job.
- Obtain good results in the CBL/VDL by working directly on the cement attenuation properties (with
microsilicium and fluid loss control).

Figure 2. SGS Test Graph for Slurry “G”

6 SPE 130159

Cement Pumping Program

A generic pumping program for the new methodology is shown in the following lines:
a. Perforate with a 4” Gun, 4 spf, and 22 grs. Hydraulic testing is done to measure severity of fluid
losses into the formation (from wellhead).
b. Slugs: Pumping of solid removal slug (mechanical), 20 bbls (viscous activated fracture gel).
Pumping of Second Slug (Chemical) Sodium Methasilicate + Calcium Chloride + Fresh Water. 20
bbls (for fluid loss control and also to improve cement adherence in the casing).
c. Mix 200 saks of Cement “G” + Additives squeeze and displace from wellhead. Displacement is
done @ 0.5 bbls/min. (Hesitate during pumping and displacement, stop pumping every 10
minutes).
d. Displace up to 10 mts over the top of the perforation interval and closure pressures of around 700
psi. Total operational time should be around the limits of the pumpability (+ 10%).
Additional comments about the pumping program:
- Viscous Gel Slug (Mechanical) works as a solid removal (a lot of corrosion and solids from formations can
be removed). It is also an added feature in comparison with old methodology.
- Chemical Slug (Sodium Methasilicate + Calcium Chloride + Fresh Water), besides of helping with the loss
control, the Sodium Methasilicate reacts with the Calcium and the outside casing wall becomes vitrified to
allow a better cement adherence.

Field Results
Using the proposed methodology 14 wells have been intervened, 13 of them have been approved in just one
squeeze job (after CBL), in just one of them a second squeeze job was required to reinforce an initial good result in
order to exceed expectations of isolation quality (8 meters were found on the first job). Table 3 shows a summary
of the results in the 14 wells.

Well Saks Initial Closure Depth of Length with >80% Observations

Pressure Pressure Top of cement (mts)
(psi) (psi) Cement
(mts)
CL-213 200 77 225 230 12 Elevated Closure Pressures
difficult to find.
CL-447 250 25 195 280 28 Normal Operation. Difficult to
find Closure Pressures.
CL-583 200 254 417 200 8 A second cementing operation
is required to reinforce the first
one.
CL-669 200 143 1500 200 65 Normal Operation.
CL-695 200 80 477 125 36 Normal Operation.
CL-748 250 306 1177 187 45 Normal Operation.
CL-757 300 0 5 100 47 Fluid loss excess, an elevated
closure pressure could not be
found.
CL- 200 255 626 90 150 With Retainer Plug (WO).
1156
CL- 200 40 908 155 12 With Retainer Plug (WO).
1520
CL- 200 120 432 115 113 With Retainer Plug (WO).
1946
EC-31 250 117 413 50 105 With Retainer Plug (WO). The
whole length up to surface with
cement.
EC-405 310 20 500 115 59 Normal Operation.
ME- 300 96 618 175 73 Normal Operation.
101
ME- 200 154 459 60 240 Normal Operation. The whole
268 length up to surface with
cement.
Table 3. Field Results Summary using new methodology

Specific example of operation and result (Well CL-748)

Figure 3 shows the operational chart of the squeeze job using the new methodology / slurry. Also CBL/VDL results
are shown in Figure 4.
SPE 130159 7

Hesitation

Pc=1177 psi

Pi=308 psi
Slugs

Figure 3. Squeeze Job Chart in Well CL-748

Section between 200 y 250 mts. Section between 250 y 200 mts.
Figure 4. CBL / VDL Section between 300 y 200 mts in Well CL-748

Conclusions

1) Implementation of the new methodology / slurry has allowed a cost reduction of 41 % in comparison with 2008;
operational days are also reduced in a considerable manner (39 %).

Year Conditions Squeeze jobs / Well. For Wells Cost Days / Well
Patagoniano Isolation (KUS$/Well)
2008 Previous methodology / 2.15 13 131 8.5
supervision / engineering
2009 * General changes in 2.12 44 95 6.9
supervision (optimization of
operations)
2009 New Methodology full 1.1 14 78 5.2
** implementation
Table 1. Summary of results (comparison of 2008 and 2009)
Note: By December 31st, 2009) up to 45 wells have been already P&A using this methodology (in different
areas outside Canadon Seco).
Effective results continue with 1.1 Squeeze Jobs/well with 4.9 days/well and 78 KUS$/Well.
8 SPE 130159

2) In the Canadon Seco Field (where high fluid loss severity is found, depth of perforations around 300 mts and
Calipers of >10”) an average of at least 200 saks (under current slurry) per squeeze job are required to expect
good results.

3) Regarding the slurry characteristics: Controlled SGS times is a must to find cement column in the aquifers zone,
good results are expected with SGS times under 70 minutes (for 500 lbs/ft^2). Filtrate control of at least 30 ml/30
min is required. Pumpability under 90 minutes is a plus to expect success (<60 minutes under hesitation).
Microsilcium promotes SGS time reduction, it is also an attenuator helper by establish itself within the slurry matrix,
allowing a better CBL result.

4) Regarding the operation: Displacement rate should be kept as low as possible (around 0.5 bpm) in order not to
promote fluid loss into the formation. Mechanical Slugs is effective for the chemical slug to perform.

XI.- Acknowledgements
Recognitions are sent to the whole YPF Abandonment Group in South Argentina, including Company Men and
Supervisors, which operational help was a key factor to achieve to expected results. Also to YPF Direction and
Management for allowing the presentation of this paper. Recognition is also sent to service cementing companies
(San Antonio and Geopatagonia), where tests were performed and which experiences and equipments were also
main factors to achieve the results, also an acknowledgement is sent for the crew members of the Coiled Tubing
and Workover rigs.

XII.- References
1) J. Moon, S. Wang. “Acoustic Method for determining the Static Gel Strength of Slurries”. SPE-55650.
2) V. Gonzalo, B. Aiskely, C. Alicia. “A Methodology to evaluate the Gas Migration in Cement Slurries”. SPE-
94901.
3) A. Carreira. “Fundamentos de Cementación”. Curso de Dirección Técnica YPF.
4) F. Sabbins, D. Sutton. “The relationship of Thickening time, Gel Strength and Compressive Strengths of Oil
Cements. SPE-11205.
5) F. Sabbins, D. Sutton, J. Tinsley. “Transition Time of Cement Slurries between the Fluid and Set States”. SPE-
9285.
6) H.K.J. Ladva, B. Craster, T.G.J. Jones, G. Goldsmith, D. Scott. “The Cement-to-Formation Interface in Zonal
Isolation”. SPE-88016.
7) D.D. Fitzgerald, B.F. McGhee, J.A. McGuire. “Guidelines for 90% Accuracy in Zone-Isolation Decisions”. SPE-
12141.
8) H. Gai, C.F. Lockyear. “Cement Bond Logging- A new analysis to improve Reliability”. SPE-23729.
9) J. Jutten, I. Toma, Y. Morel, B. Ferreol. “Integration of Cement Job Data for Better Bond Index Interpretation”.
SPE-21690.