MECHANISMS, MEASUREMENT AND MITIGATION OF BARITE SAG

Mario Zamora, M-I SWACO

This paper was presented at the Offshore Mediterranean Conference and Exhibition in Ravenna, Italy, March 25-27, 2009. It
was selected for presentation by the OMC 2009 Programme Committee following review of information contained in the abstract
submitted by the authors. The Paper as presented at OMC 2009 has not been reviewed by the Programme Committee.

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

ABSTRACT
Recent industry papers on barite sag suggest consensus on its causes and best practices.
However, this potentially critical drilling concern is not yet fully understood and serious
problems still arise from time to time. Directional wells with narrow drilling windows, hole-
cleaning difficulties, and drilled with oil-based mud under HTHP conditions continue to be the
most problematic. This paper looks at barite sag from three perspectives - mechanisms,
measurement, and mitigation - with the goal of reducing frequency, severity and concern.
Barite sag is a deceptively complex phenomenon. A qualitative review of the kinetics of barite
sag is presented to provide insights into their practical impact on drilling-operation and mud-
engineering strategies. Also discussed is the correlation among these mechanisms and
common wellsite and laboratory measurements. The tests are used to help quantify severity,
to assist with the development of new mud products or systems, and to evaluate the
combined effects of mud properties and drilling practices. The latter can be assisted by an
improved relationship introduced in this paper for correlating one of the wellsite tests with
field conditions.
Best practices are of consequence only if they are well understood and properly applied
during planning and in the field. On critical wells with multiple issues, sometimes actions to
improve one are detrimental to others. Lack of attention to barite sag can lead to and
exacerbate other serious drilling problems. To assist in this regard, key barite-sag best
practices are aptly grouped in this paper according to prevention, detection, mitigation, and
remediation.

INTRODUCTION
There is a short list of drilling problems that have plagued the industry since rotary drilling
was first introduced. Lost circulation, hole cleaning, stuck pipe, loss of well control, and
wellbore instability are among the most problematic. Barite sag was unknowingly added to
the list with the very first directional well drilled with a weighted mud. It was years later that
the industry recognized the classic signature of sag and how it could cause and exacerbate
other critical drilling problems. In fact, the cementers were the first to report the detrimental
effects of barite beds (called “mud channels”) that formed while cementing casing in
directional wells.
The term “barite sag” is commonly used (and will be used in this paper) because barite has
long been the traditional weight material for drilling fluids. However, sag can occur with any
solid, inert weighting agent, including hematite, illmenite, and manganese tetroxide.
Barite sag continues to be a real conundrum for the drilling industry and universal solutions
for its prevention are not readily at hand. Problems persist despite the general agreement on
the causes of barite sag and best practices for its mitigation. On one hand, it is more of a
precursor to other drilling problems than a problem by itself. It must be purposely measured
for proper identification and diagnosis so that its effects can be properly managed. Most

1
importantly, actions to mitigate barite sag often are bounded by well conditions. It is because
of these boundaries that critical directional wells with narrow drilling windows, hole-cleaning
difficulties, and drilled with oil-based mud under HTHP or deepwater conditions continue to
be the most problematic with regard to barite sag. It is not uncommon for actions to improve
one problem to be detrimental to others.
Numerous drilling papers have been published on barite sag in the past two decades.
Initially, there were some differences of opinion, especially as to whether barite sag was
predominately a static or dynamic settling problem. This was significant because the primary
settling mode was critical to whether gel strength or low-shear viscosity was the key
rheological parameter. Over time, barite-sag concepts, causes, and best practices seemingly
converged, to the point where many of today’s efforts are focused on less-intuitive areas in
hopes that one will be the elusive “magic bullet”. Much of this effort has been directed to mud

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

characteristics, especially rheology, but it is clear that barite sag is not just a mud problem.
Rather, drilling practices also play key roles, both positive and negative.
This paper is concerned with the general mechanisms, measurements, and mitigation of
barite sag. While these issues have been discussed previously, both individually and
collectively, they are addressed here from different perspectives. For example, the qualitative
kinetics of particle settling in inclined tubes are revisited to help clarify the unique settling and
slumping mechanisms of barite sag and why the problem is so difficult to prevent or mitigate
in directional wells.
Regarding measurements, a major obstacle has been the lack of formal or even defacto
industry testing standards that would foster equitable comparisons (Scott et al. 2004). These
tests are used to evaluate the combined effects of mud properties and drilling practices. In
addition, they are required to screen and develop new mud products or systems and to
improve drilling techniques. The need for standardization should be resolved in the near
future as a task group commissioned by API Subcommittee 13 completes its work to
standardize wellsite test methods. Various laboratory methods and some of the practical
wellsite tests being considered for inclusion in the API recommended practice are reviewed
in this paper. Also included is a proposed relationship for correlating one of these wellsite
tests with field conditions.
Despite the inherent value of standardized tests, barite sag is basically a field problem that
can be managed effectively if the causes are anticipated and properly addressed. Lack of
attention easily can exacerbate and lead to more serious drilling problems. Training is
paramount, so a new PDMR scheme is proposed to logically categorize key barite-sag best
practices according to Prevention, Detection, Mitigation, and Remediation.

MECHANISMS OF BARITE SAG

Barite sag originally was thought to be a static settling problem because sections with density
stratifications were observed in directional wells after circulation was stopped for extended
periods. Hanson et al. (1990) were among the first to identify sag primarily as a dynamic
settling phenomenon. Based on extensive laboratory studies, they proposed that barite beds
formed while circulating, thickened when flow was static, and subsequently slumped (slid or
oozed downwards) to create density variations in the fluid column. Axial and cross-sectional
density variations were measured in the laboratory using an inclined flow loop with nine
sampling ports along the top, middle and bottom of the test section. A similar design was
used later by Saasen et al. (1995) to help predict sag from rheological measurements. A
shaft positioned along the axis inside the angled pipe was used to simulate movement of
small logging tools which aggravated slumping.
Hindered and Boycott Settling Kinetics. Studies with slant-tube flow loops validated the
curious settling phenomenon first reported by Boycott (1920), who observed that blood
corpuscles in narrow tubes settled a good deal faster if the tubes were inclined rather than
vertical. The kinetics are illustrated and compared in Fig. 1. In both examples, suspended
particles substantially denser than the suspending fluid settle vertically due gravitational
effects at a rate v0 as indicated schematically by the bold, downward arrows.

2
The drawing on the left depicts the three regimes of particle sedimentation in a vertical tube,
commonly referred to as “hindered” settling. As illustrated, the particle concentration
systematically increases with time from top to bottom until virtual equilibrium is reached. The
regimes are defined as follows:
• Clarification Regime - the few remaining particles settle mostly individually (“free”
settling) with only minor interference from the tube walls and nearby particles. Stokes’
law applies.
• Hindered Settling Regime - the concentration is sufficiently high that surrounding
particles crowd and interfere with the settling of individual particles, thereby slowing
their settling rate below that of free settling. If the particles aggregate to form clusters
or flocs, the settling rate can be somewhat greater due to the increased size.

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

• Compaction Regime - particles that have reached the tube bottom support each other
mechanically. Excess fluid is expelled very slowly upwards as the bed compacts.

Fig. 1 – Hindered (left) and Boycott (right) settling kinetics under static conditions.

There are many factors that affect the hindered settling rate, but of particular interest here is
the efficiency by which displaced fluid can travel upwards and permit particles to settle. For
most weighted drilling fluids, the solids loading is such that the settling rate would be very low
even after discounting any gelation. This would be the case for a mud left undisturbed in a
container at room temperature.
Inclination of the tube as illustrated in the right-hand drawing in Fig. 1 dramatically changes
the kinetics. Particles still settle vertically, although the maximum travel distance is greatly
reduced. However, it is the behavior of the clarified layer that significantly increases overall
settling rate. The clear-fluid layer forms quickly on the entire high side on the tube and flows
upward along this boundary due to buoyancy. In effect, the clarified layer provides an
efficient and orderly means for displaced fluid to collect and move out of the way.
Particles subsequently accumulate on the low side and form a sediment bed that slumps
downwards and concentrates at the bottom of the tube. As expected, maximum clarified-
layer velocity and settling rate occur when the inclination is around 45°. At higher angles, the
buoyancy effect on the clear layer decreases and the settling rate is proportionately lower.
Hindered settling applies when the tube is placed horizontally. Due to the increased surface
area available to the displaced fluid; however, overall particle settling efficiency should be
higher than that experienced when the tube is vertical.
Boycott Modeling. The kinetic model for Boycott settling developed by Ponder (1925) and
independently by Nakamura and Kuroda (1937) is appropriately known as the PNK Model.
Validated by numerous experimental studies and referenced from Acrivos and
Herbolzheimer (1979), the following expressions are for the case of particle settling between
inclined parallel plates:

3
 1 sin .1
cos

1 sin .2

In Eq. 1, S(t) is the effective settling rate (the volumetric rate at which clarified fluid is
accumulated per unit depth); v0 is the particle settling velocity in the suspension; α is the
inclination; H is the suspension height; and b is the distance between plates. In Eq. 2, H(t) is
the instantaneous suspension height. Acrivos and Herbolzheimer (1979) and Pasaly et al.
(2007) are among those that have presented more inclusive models using the principles of
continuum mechanics. However, discussion of their complex mathematical treatments is

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

beyond the scope of this paper. The latter model is based on a phenomenological analysis of
dynamic sag in drilling muds that considers particle settling behaviour in Bingham plastic
fluids.
There are two parameters of particular interest in Eqs. 1-2. First is the particle settling
velocity v0, because of its equivalency to the “slip” velocity commonly used in drilling. More
important, however, is the aspect ratio H/b which is a very large value for directional wells.
While there are no known studies of Boycott settling in very long, narrow tubes, the
implication is that static settling could be much greater than generally expected if gel
strengths are not properly implemented. Undoubtedly, several limiting factors (perhaps a
“terminal settling velocity” in an inclined wellbore) must come into play if the kinetic model
described by Eqs. 1-2 still apply at high H/b values.
Hindered and Boycott settling have been demonstrated for years using a training device
known as a Zag Tube. Various sizes have been built, but the most effective is constructed
from three 2-ft long, clear acrylic tubes connected by 135° elbows (Hanson et al. 1990).
Light-reflecting particles (glitter) are suspended in a clarified xanthan-gum solution for the
demonstration. The left picture in Fig. 2 was captured from a video showing Boycott kinetics
under static conditions (no pumping). As highlighted in Fig. 1, there are three definite zones:
clarified layer, upward flow region, and a downward slumping bed.

Fig. 2 – Static Boycott settling in a Zag Tube (left); accelerated Boycott settling in a 2-inch flow
loop at 130 ft/min velocity (right).

The value of the Zag Tube illustration is evident when compared to a similar demonstration
suspension pumped in a 2-inch flow loop at 130 ft/min velocity (right-hand video capture in
Fig. 2). The clarified slit visually vanishes, but three distinct zones are visible. The main flow
stream covers the entire upper half of the tube; the center zone appears stationary, but
continually feeds particles to the slumping bed; the slumping bed moves opposite the
direction of flow at a velocity higher than when the suspension was static.
The significance of this so-called “accelerated” Boycott settling at low flow rates is that it can
exacerbate barite sag. A plausible explanation is that low flow rates break the gels, reduce

4
the viscosity of the shear-thinning fluid, and promote sag. Fluid rheology, especially at low
shear rates, plays a key role in minimizing this effect. The center zone gradually vanishes as
flow rate increases, but high flow rates would be required to eliminate bed formation and
slumping. This would be especially true in an eccentric annulus without the positive benefit of
pipe rotation.
Rheology Effects. Additional complications arise when typical drilling fluids are considered,
but the basic kinetics remain the same. Most weighted muds used in directional wells are
rheologically complex, exhibiting thixotropic and yield-power-law flow behaviour. Some
exhibit viscoelastic properties. Furthermore, they are specially treated chemically to achieve
multiple desired properties required for the drilling process, including mitigation of sag.
Numerous technical papers have addressed the role of rheology in both water-based and

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

non-aqueous drilling fluids, including Saasen et al. (1991), Dye et al. (1999a), and Tehrani et
al. (2004). The industry as a whole continues to look at rheological properties in hopes that
identification of key parameters and/or methods to achieve the right rheology will cure or at
least minimize sag potential. Seeberger et al. (1989), Saasen et al. (1991), Bern et al.
(1996), and Dye et al. (1999b) are among those who have documented the contribution of
low-shear-rate rheology. The 6 and 3-rpm dial readings (and their variants) taken on
standard field viscometers and yield-stress values determined by measurement or
extrapolation are commonly used as guidelines.
Most intriguing is the work of Tehrani, et al. (2004) and others with regard to the positive
influence of viscoelastic properties on sag reduction. Consistent with the kinetics discussed
earlier, data suggest that static and dynamic sag decrease as elastic properties increase.
Viscoelastic behavior can be achieved in most drilling fluids, but current equipment available
at the wellsite cannot make the necessary rheological measurements.

MEASUREMENT OF BARITE SAG

Measurement of barite sag has proven to be as challenging as eliminating the problem itself.
The fundamental concern is the density stratification that can occur in directional wells.
Failure to diligently make this measurement on critical wells or lack of response to warning
signs easily can exacerbate and lead to serious drilling problems. The accepted and most
practical method to quantify this stratification is to continually measure the mud weight while
circulating bottoms up after a trip, logging run or other operation requiring the mud to remain
static for an extended period. However, this is not without its own set of problems and
uncertainties.
Various wellsite and laboratory analytical tests have been developed in support of the field
measurement. Others have been effective in studies to validate sag mechanisms, identify
key parameters and their interrelationships, and measure a given drilling fluid’s response to
these parameters. Some are used in research to help evaluate and develop new mud
products and systems. While the distinction between laboratory and wellsite testing is
somewhat fuzzy, it is safe to say that most wellsite tests can and are often also used in the
laboratory. On the other hand, certain laboratory equipment cannot be used at the wellsite if
excessively large, expensive, complex, time-consuming, or otherwise unsuitable for the
wellsite environment.
A major goal of analytical tests is to target certain elements of barite sag and determine the
correlation with behavior in the field. Parametric studies with laboratory-scale flow loops in
conjunction with field data have been instrumental in establishing sound best practices. As
such, most barite sag analytical tests, based on number and use, focus on drilling fluid
characteristics. This may be a reason that barite sag often is mischaracterized as solely a
mud problem.
In the end, direct correlation between analytical tests and field results may prove to be too
daunting. One option is to use the same approach that the industry uses to predict downhole
pressure losses, where key rheological measurements taken in the laboratory and in the field
drive sophisticated hydraulics simulators under steady-state and transient conditions.

5
It is useful to broadly classify laboratory and wellsite sag tests as direct or indirect
measurements. Simply stated, direct measurements are expressed in density units (such as
lbm/gal), while indirect measurements seek to infer the presence of sag. A subdivision of this
classification scheme is based on the recent shear history of the test fluid or the shear
condition during the sag measurement: static, dynamic, or a combination of the two.
Field Density Stratification Measurement. The proposed API definition of barite or weight-
material sag is “recognized as a significant (>0.5 lbm/gal) drilling fluid density variation, lighter
followed by heavier than the nominal fluid density, measured when circulating bottoms up
where a weighted fluid has remained uncirculated for a period of time in a directional well.” In
this case, the condition is very aptly defined by the measurement. An important caveat in this
definition is that the proposed sag value is the difference between the maximum recorded
mud weight and the mud weight normally in circulation. Previously, some used the difference

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

between the maximum and minimum densities recorded during the bottoms-up circulation. It
follows that field measurement of barite sag involves continual (~15 min intervals) or
continuous density measurement of the annular mud column as it is circulated out of the
hole.
Fig. 3 is a trip-report graph of mud weight (corrected to 85°F) and flowline temperature
versus time after the mud was static for a 24-hr period. Maximum, nominal, and minimum
recorded mud weights were 13.59, 13.3, and 12.55 lbm/gal, respectively. The NAF (non-
aqueous fluids where oil or synthetic is the external phase) was circulated at 207 gal/min
inside 9⅝-inch casing inclined to a maximum of 60°. Rheological properties at 80°F and
150°F are given in Tab. 1. The well was drilled in 3,100 ft of water in the Gulf of Mexico.
Technically, this was not classified as a sag incident because the density difference was <0.5
lbm/gal . No problems were encountered on this well due to the density variation.

Fig. 3 – Density stratification after 24-hr static period while circulating a 13.3-lbm/gal NAF inside
9⅝-inch casing at 60°inclination in Gulf of Mexico.

Tab. 1: Rheological properties for 13.3 lbm/gal NAF in Fig. 3

Temperature PV YP LSYP 10-s Gel 10-m Gel
(°F) (cp) (lbf/100ft²) (lbf/100ft²) (lbf/100ft²) (lbf/100ft²)
80 63 24 11 15 30
150 30 17 10 15 26

Care must be taken when reporting and analyzing a reported sag incident – details are
important. In addition to the fore-mentioned temperature correction and pressurized balance,
preceding operations that could have promoted sag or provided a false reading need to be
logged and evaluated.

6
Additionally, there are several unknowns and uncertainties related to this field measurement
of barite sag during bottoms-up circulation worthy of open discussion. Does static settling
play a more important role than is currently thought? Are most, if not all, of the barite beds
circulated out? If the stratification can be circulated out, can the beds be hydraulically and
mechanically removed prior to tripping out? Is the circulation conducted such that the
process itself exacerbates the problem? Is there a practical method to measure barite-bed
accumulations in a well in-situ?
Laboratory Measurements. Devices used for measuring barite sag in laboratory settings
vary widely in design; however, differences are limited in scope. For example, most
laboratory tests are conducted either in test cells or tubes. Dynamic conditions are created
by pumping and/or pipe rotation, although induced vibrations to break gels has been
considered. Sample densities are determined by weighing after manual extraction or in-situ,

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

the latter providing the benefits of automated, continuous measurement. Some devices have
no temperature control (measured densities must be corrected for thermal expansion), while
others operate under HTHP conditions.
The earliest tests run at temperature using static aging cells are still used today. It has been
suggested that the term “sag” was coined from these tests in an effort to distinguish from
“hard settling”. The density of samples extracted from near the top and bottom of the
stratified fluid column are compared to determine the “degree of sag”. This test is still used
often for tender evaluations.
Because of the insensitivity of static-aging cells to density stratifications, similar static tests
have been run in relatively long, narrow glass tubes (e.g., 4-ft x 20-mm) placed in ovens for
extended periods. Commonly called “slant” tubes, they can be placed at an angle in large
ovens to undergo classical Boycott settling. The most common procedure involves a time-
consuming extraction of the fluid to permit measurement of the density profile.
An improvement of the slant-tube test was presented by Jamison et al. (1990) in order to
automate analysis of static sag. This device relied on changes in the center of mass of the
slant tube containing the sample as the principal variable as a function of time. At that time,
the industry considered barite sag to be a static settling problem. However, a recent
development improved on this concept by providing two significant advancements (Murphy et
al. 2008). A rotatable, inner pipe was incorporated into an HTHP design in order to simulate
dynamic conditions. The improved device permits automated static and dynamic testing at
temperatures and pressures to 350°F and 10,000 psi, respectively.
Flow loops of different designs have been developed to simulate barite-sag kinetics under
static and dynamic conditions. An interesting distinction among them is the method used to
“measure” the barite sag. The flow loops by Hanson et al. (1990) and Saasen et al. (1995)
described previously in this paper relied on manual extraction from different ports along the
test section to validate the cross-sectional and longitudinal density profiles associated with
Boycott settling. Other designs relied on careful extraction of barrel equivalents in an effort to
simulate the characteristic signature of barite sag. An improved version used a segment test
section (segments could be physically isolated at the end of the test) for the same purpose.
A flow loop used in two joint industry projects helped formulate best practices that still are
widely accepted by the industry (Bern et al. 1996 and Bern et al. 1998). This flow loop
measures the density of the circulating fluid at different angles, flow rates, pipe rotation, and
eccentricity using an inline mass flow meter. The process is sufficiently accurate to
continuously measure transient density variations created by minor changes in operating
parameters.
Fig. 4 presents superimposed results from testing a NAF at four different inclinations and
seven combinations of annular velocity, and pipe rotation. Combinations 1 and 7 represent
the maximum parameter values (no measureable sag); combinations 2 and 3 represent low
annular velocities without rotation (circulating density decreases and beds grow with time);
and combinations 4-6 progressively increase velocity and rotation (beds are gradually re-
suspended). Combinations 4-6 provide valuable information regarding the characteristics of
the bed and suggest how they might respond to remedial operations.

7
 Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021
Fig. 4 – Superimposed NAF sag flow loop tests at 4 different inclinations and 7 annular-velocity
(AV) and rotary-speed (R) combinations.
A recently developed device (www.graceinstrument.com) is designed to evaluate dynamic
settling in an HTHP cell. Based on an inclined pressure cell with internal rotation capabilities,
a 220-mL fluid sample is subjected to temperatures up to 500°F, pressures to 30,000 psi,
and rotary speeds to 600 rpm. Sediment samples are extracted from a collection site inside
the cell without requiring depressurization. The unit also can be used to measure HTHP
rheology for drilling fluids and cements.
Wellsite Measurements. As if by universal decree, wellsite tests are expected to be
“performed quickly and with simple apparatus,” preferably at minimal cost. Undoubtedly, the
tests also need to be intrinsically safe. These constraints have made it difficult to develop
tests for barite sag usable by field engineers at the wellsite.
The need for a dynamic wellsite test was first addressed by Jefferson (1991). The test
sample is placed in a thermocup and stirred by a conventional field viscometer in order to
simulate dynamic conditions at temperatures up to 180°F. Samples taken by syringe from the
bottom of the heat cup at the beginning and end of the test determine the density increase
(relative sag) after 30 min. Several modifications and versions of this so-called “Jefferson
cell” followed. Other companies developed modified Jefferson cells. In one example,
changes included flow ports at the bottom of the heat cup, a peristaltic pump to circulate
fluid, and a densitometer to measure temperature and density of the circulating fluid (Dye et
al. 2002).
A more recent, low-cost improvement involves insertion of a thermoplastic “sag shoe” in the
bottom of the thermocup prior to running an otherwise conventional Jefferson test (Zamora
and Bell, 2004). The insert accelerates settling of weight material, provides a singular
location for sampling, and permits measurement of the fluid’s ability to pick up an existing
barite bed. Fig. 5 (left) shows the key tools required to run this Viscometer Sag-Shoe Test
(VSST), with the exception of the digital balance required to determine sample density. The
right-hand picture presents CFD results 31 sec into a test. The VSST is under consideration
by the API for inclusion in the proposed recommended practice.

A concern with this test is the relatively high 100-rpm viscometer speed selected to provide
the dynamic environment. This speed was chosen strictly for practical reasons. Standard
field viscometers for critical wells normally operate at six speeds and the next lower speed (6
rpm) may not be sufficient to prevent gels from forming. At 100 rpm, shear rate between the

8
rotating sleeve and bob is 170 s–1; however, average shear rate for the bulk of the fluid
contained between the sleeve and the thermocup is about 40 s–1.

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

Fig. 5 – Equipment for VSST (left) and CFD analysis of test (right).

A more critical concern is correlation with field results. After all, the test essentially measures
the relative degree of weight-material settling under dynamic conditions. Direct correlation
should not be expected in the same manner that yield point should not correlate with
pressure loss without consideration of numerous well parameters. In the same manner as
Jefferson (1991), a set of constants have been developed from field data and hundreds of
flow loop tests to calculate a Sag Index. The new constants are given in Tab. 2.

The intent is to calculate the product of VSST results and the four constants in the table
depending on field conditions. Note that inclination constant Ka parallels the profile versus
angle described for Boycott settling in the discussion of kinetics. Also, Kv at annular
velocities below 50 ft/min and Kr at rotary speeds less than 75 rpm reflect the impact that
these low ranges have on promoting sag. Although the constants are based on considerable
data, those wanting to use this technique are encouraged to adjust the constants or add new
ones as necessary to improve correlation in local areas.

Tab. 2: Sag Index Constants for VSST method.

Angle Ann Vel Rotary Length

Ka Kv Kr Kz
(°) (ft/min) (rpm) (ft)
<5 0 <2 1.0 <5 1.0 <1000 0.5
5-10 0.03 2-50 1.2 5-75 1.1 1000-2000 0.7
10-30 0.3 50-100 0.9 75-100 0.8 2000-5000 1.0
30-40 0.7 100-150 0.6 100-150 0.5 >5000 1.2
40-70 1.0 150-250 0.3 >150 0.3
70-80 0.8 >250 0.1
80-90 0.6

Suppose a VSST result of 1.2 lbm/gal is measured for a well drilling a 3,100-ft interval at 57°
inclination. Annular velocity opposite drill pipe is 140 ft/min and rotary speed is 120 rpm. The
Sag Index would be 1.2 * 1.0 * 0.6 * 0.5 * 1.0 = 0.36 lbm/gal. This result would be compared
to the difference between the maximum and nominal mud weights recorded while circulating
bottoms up.

9
Use of methods such as the VSST is increasing, but the most common wellsite tests involve
drilling mud rheological properties, particularly the yield stress or other low-shear-rate
parameter and the gel strengths. These are labeled as indirect test methods because they
infer a relationship between the measurement and sag potential. The 6-rpm dial reading R6 is
sometimes substituted for the yield stress; however, the LSYP (low-shear yield point) is
usually a better choice if a quality value for yield stress is unavailable. LSYP is calculated
from field viscometer measurements by 2*R3 – R6.
Dye et al. (1999b) and others have demonstrated the value in taking rheological
measurements at shear rates considerably below those available on standard field
viscometers. New-generation, commercial viscometers capable of operating at shear rates
as low as 0.017 s–1 have been developed for wellsite use. A recent paper by Dye et al. (2006)
describes the use of a dynamic sag window (DSW) reproduced in Fig. 6 from that paper.

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

Mud viscosities superimposed over the graph above the lower limit correlate with minimum
sag potential. Viscosity levels above the upper limit indicate a low potential, but are
considered excessive for controlling dynamic barite sag. The authors also suggest a
minimum annular velocity of 100 ft/min to maximize value from the DSW. The technique is
being considered for inclusion in the proposed API recommended practice.

Fig. 6 – Dynamic sag window for minimizing sag potential (from Dye et al. 2006).

MITIGATION OF BARITE SAG

Proper management in the form of best practices is the underlying key to mitigating barite
sag. This process involves (a) planning that considers sag potential within the constraints
created by other well-design requirements, (b) achievement of the right drilling fluid
properties for downhole well conditions, and (c) application of best drilling practices to
prevent or minimize any well problems. Ultimate success depends on how well the details of
this process are understood and applied by those charged with its implementation.
Best practices provided by Bern et al. (1996, 1998), and others have been effective in
reducing the frequency and severity of sag incidents. However, barite-sag incidents and their
consequences persist. Notably, most of the guidelines have remained relatively unchanged
since first introduced nearly two decades ago. Perhaps a re-categorization in light of the
kinetics and measurements discussed in this paper can help revitalize the management

10
process. This PDMR scheme logically categorizes key barite-sag best practices according to
Prevention, Detection, Mitigation, and Remediation.
Prevention. Industry consensus is that barite sag cannot be prevented. Based on kinetics,
however, sag actually can be prevented by eliminating the source of the problem – the solid-
particle weight material. This can be achieved in water-based muds by using heavy brines to
provide the required density. Sassen et al. (2002). for example, reported successful use of
15.9-lbm/gal, solids-free, drilling and completion fluids formulated with cesium formate.
A different approach must be taken with NAFs. Many of the benefits of solids-free weight
material can be achieved by using high-density particles with very low mass. Known as
“micronized weight materials” (MWM), fine and ultra-fine barite particles have been very
successful in drilling critical wells prone to barite sag, especially in the North Sea. Also, a

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

then-record ultra-ERD well was drilled in 2007 with a NAF weighted with micronized, coated
barite. Fig. 7, reproduced from Taugbøl et al. (2005), compares particle-size distribution of a
micronized barite slurry (d50 <2 μm) with API barite (d50 ~25 μm). Manganese tetroxide,
illmenite, and hematite are other fine-grind weight materials that have been used
successfully. Perhaps the next generation of weight materials will be based on
nanotechnology.

Fig. 7 – Particle-size distribution comparison of micronized barite slurry (labeled MBS) and API
barite (from Taugbøl et al. 2005).
Mitigation. Mitigation starts at the well-planning stage, where all options should be evaluated
meticulously. Well profile, downhole environment, mud type, hole sizes, and casing/drill-
string design are among the most important. Serious compromises usually are in order for
critical directional wells with narrow drilling windows, especially in ultra-ERD and ultra-
deepwater applications. During drilling, maintenance of mud rheology and annular velocities
at optimum levels to mitigate sag can be challenging when annular pressure losses must be
tightly controlled.
Field data suggest that NAFs are more prone to sag issues than their water-based
counterparts. Hypotheses include, but are not limited to, (a) viscosity loss under downhole
conditions, (b) beds more conducive to slumping, (c) settling caused by improper wetting of
the weight material, (d) settling caused by gas-stripping, (e) lower viscous drag on settling
particles due to better lubricity, and (f) less efficient viscosifiers and rheology modifiers in the
non-aqueous environment (especially to obtain elastic properties).
During drilling, low-shear-rate viscosity (as defined by the LSYP, true yield stress, or dynamic
sag window) and gel strengths are the most targeted rheological parameters because they
have been shown to correlate well with minimizing sag under dynamic and static conditions,
respectively. Ideally, gels should form quickly (to capture particles immediately after
circulation stops) and break easily (to minimize excess annular pressures when circulation
starts).

11
Small concentrations of clay materials in combination with viscosifying polymers have proven
effective for achieving desired rheological properties in both water-based muds and NAFs.
When viscosity levels have to be restricted due to pressure loss concerns, drill string rotation
(>75 rpm) can become very critical for mitigating sag.
Certain drilling operations are notorious for promoting barite sag and should be avoided if
possible. For example, adding unweighted premix to an NAF indiscriminately can over-thin
the mud and promote settling. More importantly, long periods of circulating at low annular
velocities and low (or no) pipe rotation can intensify sag problems, even if rheological
properties are ideal. This can occur during rig-repair and weather time delays, when
conditioning mud, and even when circulating bottoms up after a trip or logging run. For the
latter, it is important to circulate at rates close to those used during drilling; rotary speeds can
be lower, but preferably above the recommended 75-rpm threshold.

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

Detection. By definition, detection requires monitoring the annular density variations while
circulating bottoms up after trips and regular measurement of rheological properties to
ensure key parameters are within specified guidelines. Major sag incidents have been the
consequence of failure to follow this plan.
A pressurized mud balance should be used if entrained gas is a concern. As such, gas units
also should be recorded. More importantly, the mud density must be corrected for
temperature. “Trip reports” useful for documenting field data versus pump strokes, time, or
well depth are available from Hanson et al. (1990) and others.
When a barite-sag incident has been recognized, all pertinent data should be collected and
evaluated to determine the best course of action. For example, no changes in rheology
usually are necessary if the incident was the result of drilling operations. Instead, it may be
necessary to retrain rig personnel or to develop options if the drilling operations were
constrained by well conditions. For situations where the density stratification can be
attributed to gas-stripping caused by a large gas influx into an NAF, the mud may have to be
treated with wetting agents and emulsifiers and subsequently weighted up to restore a
uniform mud weight.
Careful scrutiny of all well data is especially critical if another drilling problem has been
aggravated or caused by barite sag. An example is when lost circulation occurs after heavy
mud is circulated above a weak casing shoe. Clarification in this case is critical to prevent
this from reoccurring on subsequent intervals or wells.
Remediation. In this context, remediation refers to taking corrective action immediately
before a possible sag incident. If barite beds are suspected due to inappropriate drilling
operations or inadequate mud properties, it may be advantageous to attempt to re-suspend
the beds prior to tripping out of the hole. This would minimize the ill effects of bed slumping,
especially if the trip is to be followed by logging or casing operations.
Test procedures are available for the VSST method (Zamora and Bell, 2004) to characterize
bed consistency and estimate the effectiveness of using high flow rates and rotational
speeds. This information can assist with the decision on whether time spent remediating
barite beds is cost effective for the project.
Another common remediation procedure involves staging in the hole if the mud has remained
static in the hole for a long period. The process serves to minimize the full impact of a severe
sag incident, an important step if the tolerance between mud weight and fracture gradient is
small.
CONCLUSIONS
1. Barite sag is such a complex issue that it is worthwhile to revisit fundamental kinetics
to generate options for measurement and mitigation.
2. Qualitative analysis of barite-sag mechanisms suggests that settling under static
conditions may be more problematic than currently thought.

12
 3. The classical kinetic model for Boycott settling may not apply directly to drilling well
situations because of the large aspect ratio and deserves further study. Otherwise,
undefined limiting factors may be part of the sag phenomenon.
4. Practical procedures for monitoring barite-sag tendencies at the wellsite have been
developed and soon could be standardized in an API recommended practice.
5. Proper barite-sag management involves
a. well planning that considers all aspects,
b. achievement of the right drilling fluid properties, and
c. application of best drilling practices.

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

6. Corrective action taken immediately before a possible sag incident can reduce the
impact of barite sag and its consequences.
7. A scheme based on prevention, detection, mitigation, and remediation can assist with
organizing barite-sag best practices and training tools.

ACKNOWLEDGEMENTS
The author thanks the management of M-I SWACO for supporting this work and for
permission to publish this paper.

REFERENCES
Acrivos, A. and Herbolzheimer, E. (1979) “Enhanced Sedimentation in Settling Tanks with
Inclined Walls,” J. Fluid Mechanics, Vol 92, part 3, 435-457.
Bern, P.A., van Oort, E., Neusstadt, B., Ebeltoft, H., Zurdo, C., Zamora, M., and Slater, K.
(1998) “Barite Sag: Measurement, Modelling and Management,” IADC/SPE 47784,
IADC/SPE Asia Pacific Drilling Tech, Jakarta, Indonesia, 7-9 Sept., and SPE Drilling &
Completion, March 2000, 25-30.
Bern, P.A., Zamora, M., Slater, K.S., Hearn, P.J. (1996) “The Influence of Drilling Variables
on Barite Sag,” SPE 36670, SPE ATCE, Denver, 6-9 Oct.
Boycott, A.E. (1920) “Sedimentation of Blood Corpuscles,” Nature, Vol 104, 532.
Dye, W. and Mullen, G. (2002) “New Technology to Manage Barite Sag,” AADE-02-DFWM-
HO-12, AADE Technical Conference, Houston, 2-3 April.
Dye, W., Brandt, M. Wesling, T., Hemphill, T., and Greene, R. (1999a) “Rheological
Techniques to Minimize the Potential for Barite Sag,” AADE Annual Technical Forum,
Houston, 30-31 March.
Dye, W., Hemphill, T., Gusler, W., and Mullen, G. (1999b) “Correlation of Ultra-low Shear
Rate Viscosity and Dynamic Barite Sag in Invert-Emulsion Drilling Fluids,” SPE 56636, SPE
ATCE, Houston, TX, 3-6 Oct, and SPE Drilling & Completion, March 2001, 27-34.
Dye, W., Mullen, G. and Gusler, W. (2006), “Field-Proven Technology to Manage Dynamic
Barite Sag,” IADC/SPE 98167, IADC/SPE Drilling Conference, Miami, 21-23 Feb.
Hanson, P.M., Trigg, T.K., Rachal, G., and Zamora, M. (1990) “Investigation of Barite “Sag”
in Weighted Drilling Fluids in Highly Deviated Wells,” SPE 20423, SPE ATCE, New Orleans,
23-26 Sept.
Jamison, D.E. and Clements, W.R. (1990) “A New Test Method to Characterize Settling/Sag
Tendencies of Drilling Fluids in Extended-Reach Drilling,” ASME Drilling Technology
Symposium, PD 27, New Orleans, 14-18 Jan..
Jefferson, D.T. (1991) “New Procedure Helps Monitor Sag in the Field”, ASME paper 91-
PET-3, Energy-Sources Technology Conference and Exhibition, New Orleans, 20-24 Jan.

13
Murphy, R., Jamison, D., Hemphill, T., Bell, S. and Albrecht, C. (2006) “Measuring and
Predicting Dynamic Sag,” SPE 103088, SPE ATCE, San Antonio, 24-27 Sept., and SPE
Drilling & Completion, June 2008, 142-149.
Nakamura, H. and Kuroda, K. (1937) “La cause de l’accélération de la vitesse de
sédimentation des suspensions dans les recipients inclinés,” J. of Med. College in Keijo, Vol
8, 256-296.
Paslay, P.R., Sathuvalli, U.B., and Payne, M.L. (2007) “A Phenomenological Approach to
Analysis of Barite Sag in Drilling Muds,” SPE 110404, SPE ATCE, Anaheim, 11-14 Nov.
Ponder, E. (1925), “On Sedimentation and Rouleaux Formation,” Quart J. Expt. Physiol., Vol
15, 232-252.

Downloaded from http://onepetro.org/OMCONF/proceedings-pdf/OMC09/All-OMC09/1780879/omc-2009-105.pdf by Saint Francis University user on 02 March 2021

Saasen, A., Liu, D., Marken, C.D., Sterri, N., Halsey, G.W., and Isambourg, P. (1995)
“Prediction of Barite Sag Potential of Drilling Fluids from Rheological Measurements,” SPE
29410, SPE/IADC Drilling Conference, Amsterdam, 28 Feb – 2 March.
Sassen, A., Jordal, O.H., Burkhead, D., Berg, P.C., Pedersen, E.S., Turner, J., and Harris,
M.J. (2002) “Drilling HT/HP Wells Using a Cesium Formate Based Drilling Fluid,” SPE 74541,
IADC/SPE Drilling Conference, Dallas, 26-28 Feb.
Scott, P.D., Zamora, M., and Aldea, C. (2004) “Barite-Sag Management: Challenges,
Strategies, Opportunities,” SPE 87136, IADC/SPE Drilling Conference, Dallas, 2-4 March.
Taugbøl, K., Fimreite, G., Prebensen, O.I. (2005) “Development and Field Testing of a
Unique High-Temperature and High-Pressure (HTHP) Oil-Based Drilling Fluid with Minimum
Rheology and Maximum Sag Stability,” SPE 96285, SPE Offshore Europe, Aberdeen, 6-9
Sept.
Tehrani, A., Zamora, M., and Power, D. (2004) “Role of Rheology in Barite Sag in SBM and
OBM,” AADE-04-DF-HO-22, AADE Drilling Fluids Conference, Houston, 6-7 April.
Zamora, M. and Bell, R. (2004) “Improved Wellsite Test for Monitoring Barite Sag,” AADE-
04-HO-DF-19, AADE Fluid Technology Conference, Houston, 6-7 April.

14