Successful Well Cement Standardisation Needs Research Underpinning

John Bensteda, Jasbir S. Lota b and Juliet Munn c

a
Materials Chemistry Centre, UCL, London WC1H 0AJ, UK;
b
Parmiters School, Garston, Watford WD25 0UU, UK;
c
(formerly of) Birkbeck College, University of London, London WC1E 7HX, UK

SUMMARY
Well cement standardisation is being developed globally and there are already
five international standards operative, which are applied globally but used locally.
Their development has benefited from background research, which gives greater
confidence in confirming existing limits or in assisting new developments. The
trend is to give more impetus to operational requirements, as has happened to
date with cements for deepwater well cementing and atmospheric foam cement
compositions. Emphasis is increasingly being placed upon having better test
methods for operational requirements and for technical report type standards
(like cement slurry rheology) to be produced pending further experimentation.

1. INTRODUCTION

Currently there are five ISO standards available globally under the generic title
‘Well Cements and Materials for Well Cementing’ and another standard is
nearing completion (Table 1).

Table 1: Petroleum and Natural Gas Industries – Well Cements and

Materials for Well Cementing:
Part ISO Number Title
1 10426-1 Specification
2 10426-2 Testing of Well Cements
3 10426-3 Testing of Deepwater Well Cement Formulations
4 10426-4 Preparation and Testing of Foamed Cement Slurries
at Atmospheric Pressure
5 10426-5 Determination of Shrinkage and Expansion of
Well Cement Formulations at Atmospheric Pressure
6 10426-6 Methods for Determining Static Gel Strength
of Well Cement Formulations (in preparation)

For these standards background research underpinning for verification of the test
procedures is highlighted in various areas.
2. ISO 10426-1, Part 1: Specification

In this well cement specification background research has been beneficial for
addressing limits for thickening (setting) time, compressive strength and free
fluid. Early hydration of HSR Class G cement was investigated at water/cement
ratio 0.44 alone, with 0.2% wt calcium lignosulphonate retarder, and then with
0.2% wt sugar-lignin proprietary product retarder,all at up to 5 hours at 25oC. The
ettringite formed was essentially amorphous at first and thereafter became more
crystalline. Retardation produced relatively more ettringite and less C-S-H, plus
an increased dispersion and hindered crystal growth. With retarder present, there
was more heat evolved in the presence of water. The sugar-lignin derivative gave
more ettringite than the cement without water, but less ettringite than with the
lignosulphonate retarder [1]. The prime cause of thickening was deposition of
some C-S-H, with ettringite at best playing a minor support role [1]. This situation
was comparable to that of final setting with a Portland cement investigated earlier
[2] with around 3-4% wt C-S-H being present and with this phase having the
main influence upon thickening.

Compressive strength was carried out at atmospheric pressure or at pressures

up to 20.7 MPa (3000 psi). It was only in more recent years that the 20.7 MPa
(3000 psi) pressure ceiling for well cement testing was generally found to be
inaccurate at higher temperatures (normally above 100oC). Pressure alone has
a very significant effect (although not quite as high as temperature does) upon
thickening time [3]. This is why well cement slurries for field use needed to be
based upon reliable pressure and temperature data, with the ISO compressive
strength specification limits being simply a basic yardstick for conformance and
not a realistic indicator of downhole compressive strength growth with time. For
more critical wells, computer simulations of pressure and temperature gradients
need to be evaluated before finalising the well cementing formulations.

The 30 minutes leeway in the ISO Schedule 5 thickening time test [4], which
requires thickening time to be in the range 90-120 minutes, is certainly useful for
minimising batch-to-batch inconsistency under downhole conditions of Class G
cement from the same source under the same manufacturing conditions. This
test is undertaken for achieving suitable additive responses from retarders, fluid
loss controllers, dispersants and defoamers and any other desirable additive for
securing a good seal in the downhole annulus in the field [5].

The old free fluid test (free water test) using a measuring cylinder was replaced
by the Erlenmeyer flask method, which uses a larger vessel and thus gives a
more reliable measurement, with a maximum free fluid of 5.9%. This test is
important for ensuring that free fluid is sufficiently low in order to prevent
segregation during thickening and hardening of the cement slurry. Such
segregation would be likely to give rise to unwanted events (such as gas
migration and influx of unwanted formation fluids) that would impede zonal
isolation and thus be detrimental for long term well durability.
3. ISO 10426-2, Part 2: Testing of well cements

This is a large document due to numerous tests having been included over the
years in the former API RP 10B from which ISO 10426-2 was developed, with the
result that they are not classified tidily. The ISO standard is being revised and is
likely to ‘tidy up’ the contents into subject areas for convenience of usage in the
laboratory. Such possible subject areas could include the following:
• General testing of well cements
• Additional thickening time testing
• Rheological and related testing
• Pozzolans in well cementing.

Another area that is covered in the National Annex within the British version (BS
EN ISO 10426-2) is free lime [6], where the global situation is not overtly
straightforward for international standardisation purposes. It is well known that
high free lime containing well cements can produce unacceptably high
rheological characteristics like plastic viscosity and yield value, high downhole
fluid loss and interference with cement slurry retardation [7]. Interestingly, in
Brazil their national standard for MSR (medium sulphate resistant) and HSR
(high sulphate resistant) Class G well cement has a maximum limit of 2.0% [8].

In the British version [6], the free lime requirements were set out thus after
numerous experiments had been carried out on a wide range of cements from
different countries:

Free lime (CaO) in cements is measured either by the ethylene glycol method or
by X-ray diffraction.

Simple mandatory limits for free lime are not to date ascribed on a worldwide
basis, because of the many interactive effects from other cement constituents
which influence overall cementitious reactivity. High levels of free lime can affect:
a) additive (especially retarder) responses
b) rheological characteristics of well cement slurries (like plastic viscosity and
yield point)
c) fluid loss results (high).

Optimisation of free lime levels in well cements is achieved on a plant-by-plant

basis during manufacture. In most circumstances a useful “rule of thumb” guide
for free lime limits applies as follows:
a) If the magnesium oxide (MgO) content of the cement is 1.5% mass or less,
free lime is limited to 1.0% mass or below.
b) If the MgO content lies above 1.5% mass, the free lime is limited to 0.5%
mass or below.
There are some individual exceptions to this “rule of thumb”, for example where
higher contents of free lime give adequate retardation with appropriate retarder
concentrations.

This information given for free lime limits is, in standardisation terms, purely
informative (non-mandatory) and not normative (compulsory), because it is part
of a useful, but individual, National Annex to an International Standard.

The apparent discrepancies in desirable free lime limits can be explained thus.
In the United States, for example, well cements like Class G and H normally
contain much higher magnesia (MgO) levels than in Western Europe, because
of the common use of high magnesian limestones as a raw material for cement
production. As a result the US cements contain high (relatively speaking) MgO in
comparison with the Western European well cements. Although up to around
1.5% mass of the MgO content is normally contained in solid solution in the
clinker phases, most of the rest is present as free magnesia (periclase, MgO).
Free magnesia reacts slower with water than free lime, but the general effect is
to augment the undesirable properties given to cement slurries by free lime.
Thus only low levels of free lime can be tolerated, hence the desirability of a free
lime limit of 0.5% mass (or at most 0.6% mass) for high MgO-containing
cements. In Western Europe, where cement MgO levels in well cements are
normally below 1.5% mass, there is generally little free magnesia and free lime
values up to ca. 1.0% mass can usually be tolerated [9].

Most well cement producers of Class G and H cements in particular normally

check the amounts of free lime present, so that the well cements perform
satisfactorily in the field. Such international standardisation would need to be
addressed globally before mandatory limits could be set for free lime contents of
well cements. None of the rheological tests set any limits for plastic viscosity,
yield point or gel strength development. The Brazilian national standard for MSR
and HSR Class G cement has set limits for rheological properties at 27 oC and
52oC as given in Table 2, which are determined by rotational viscometry [10]:

Table 2: Rheological Properties for Class G Well Cement

Rheological Properties 27oC 52oC
Initial gel strength (10 seconds) at 3 rpm, Pa (max.) 12 12
Final gel strength (10 minutes) at 3 rpm, Pa (max.) 16.8 16.8
Consistency after 1 minute at 3 rpm, Pa (max.) 9.6 9.6
Consistency after 5 minutes at 3 rpm, Pa (max.) 9.6 9.6
Plastic viscosity, Pa.s (max.) 0.055 0.055
Yield point, Pa 14.4-33.5 14.4-38.3

Maximum increment of final gel strength with temperature rise from 27oC to 52 oC
is 4.8 Pa at 27oC rising to 10.0 Pa at 52 oC
Recommended maximum increment of yield point with temperature rise from
27oC to 52 oC is 9.6 Pa
The limits for these rheological tests are appropriate for Class G cements made
in Brazil for producing suitably pumpable well cement slurries downhole.
However, whilst the cements conforming to these limits also work satisfactorily
for European Class G cements, the aforementioned limits are not appropriate for
Class G or H cements made from high magnesian limestones, as commonly
experienced in the United States, for instance. This is because of the generally
higher gel strengths and yield points normally found for the latter Class G or H
cements. These increased rheological characteristics appear to arise through the
hydration of periclase augmenting that from free lime, but the cement slurries can
still be pumped satisfactorily downhole into the annular spaces of the wells
during the well cementing process. If full international standardisation were
required for rheological properties then the various limits would need to be
changed, so as to accommodate the Class G and H cements derived from high
magnesian limestones.

4. ISO 10426-3, Part 3: Testing of deepwater well cementing formulations

This standard was the first of a number of new well cementing standards that are
being developed through ISO on a performance-related basis. Deepwater is
usually regarded as water of depths exceeding 400-500 m with ultra-deepwater
being water more than 1500 m deep [5].

There are numerous complexities involved in deepwater well cementing involving

a number of variables impacting the thermal history of a cement formulation
during placement and curing, plus the interdependence of many of those
variables. As a result, the user is directed to employ numerical heat-transfer
simulation or actual field measurement to determine the test temperature and the
temperature/pressure schedule required. In this way, the testing of the cement
formulation can reflect as closely as possible the actual temperature profile found
during field cementing operations. Numerical modelling may need to be used for
determining the relative magnitude of the input variables so that ‘most likely’ and
‘least likely’ scenarios of temperature history can be reliably assessed. These
procedures serve not only for the testing of well cements under deepwater well
conditions, but may also be used in those circumstances where low seafloor
temperatures are found at shallow water depths [11].

Well cements that can be used in deepwater well cementing [11] can include
those of ISO Classes A, C, G or H, high alumina cement, appropriate foamed
cements, various types of suitable ductile cement compositions [12] etc. In each
deepwater well cementing operation, the cement chosen needs to be fit for
purpose [11]. Such cementations are normally critical and extensive pre-planning
and pre-testing are essential for enabling the successful cementing of these
critical wells.
5. ISO 10426-4, Part 4: Preparation and testing of foamed cement slurries
at atmospheric pressure

This is another performance-related standard developed by ISO, which is closely

related to ISO 10426-3 in that foamed cementing compositions are often utilised
in deepwater well cementing operations. These two standards can often
complement each other during well cementing activities. Foamed cement slurries
have been described elsewhere in the overall context of ductile well cementing
formulations that can give rise to good zonal isolation downhole with critical well
cementing jobs [5,12].

Cements or cement blends used for foamed cement slurry preparation at

atmospheric pressure should be fit for purpose. Such cements could include well
cements of ISO Classes, high alumina cement, or other speciality cements. The
cements and blending materials should conform to appropriate standards. Where
ISO standards do not exist, conformance with other appropriate standards should
be made [13]. The test procedures employed to characterise the foamed cement
slurries produced are often similar to those used for deepwater cement slurries
[11,13].

6. ISO 10426-5, Part 5: Determination of shrinkage and expansion of well

cement formulations at atmospheric pressure

ISO 10426-5 [14] has been partially developed from the API Technical Report
10TR2 [15] as a base-line standard for determining shrinkage and expansion of
well cement formulations at atmospheric pressure. Two methods have been
described for use at atmospheric pressure [14]:
• The annular ring test under conditions of free water access
• The membrane test under impermeable condition.

a) The annular expansion mould is a device for measuring only the linear bulk
shrinkage or expansion properties of a cement formulation. The magnitude of
the expansion depends on the amount of expansion agent, cement powder,
slurry design and curing condition (pressure, temperature, time, fluid access).
It should be noted that expansion is strongly affected by boundary conditions.
The chemical process of mineral growth is strongly controlled by the state of
stress and mineral growth will tend to occur when the stress value is the
lowest, i.e. in pore space or empty spaces. Therefore, the degree of cement
shrinkage and expansion is dependent on a number of conditions, not all of
which can be uniquely defined. The test does not represent fully the annulus
of a well.

b) However, the purpose of the membrane test method is to measure the bulk
expansion or shrinkage when the cement is unable to imbibe water or gas.
This is commonly the case when the cement is placed within impermeable
boundaries within a well. An impermeable flexible membrane is used to
 measure shrinkage or expansion under appropriate impermeable conditions.
The membrane is sealed by tying a knot at the top in the experiments. The
material of the membrane shall not react in a high pH environment and shall
be capable of containing at least 150 ml +/- 30 ml of slurry.

Shrinkage can lead to unwanted wellbore phenomena such as a microannulus

that leads to a bad bond, interzonal communication, resulting in costly remedial
operations and lack of a hydraulic seal when utilising cement inflatable packers.
Shrinkage and expansion in cement can result from the formation of hydration
products having a density different from the compounded density of the reaction
components. This can result in changes in pore volume, pore pressure, sample
dimensions and internal stress [14].

ISO 10426-5 is likely to be the first of a number of future standards on the

importance of shrinkage/expansion and is basically a yardstick standard for
testing at ambient temperature and atmospheric pressure. Standards still to be
worked on are likely to involve various high temperature-high pressure (HTHP)
scenarios, where the effects of increases in temperature and pressure downhole
can be better appreciated in the field than at present. Much laboratory work
needs to be undertaken in this area, which is important for the successful
cementing of ductile cement compositions [12].

7. ISO 10426-6, Part 6: Methods for determining static gel strength of

cement formulations (in preparation)

A Technical Report type of standard is currently being prepared, which discusses

various different methods for determining static gel strength of well cementing
formulations in order to disseminate current available information. There is
insufficient information available at present for more definitive standardisation at
this particular stage in time. This should form a future basis for adopting either a
suitable specification or a ‘recommended practice’ type of standard in the future,
when more data becomes available. ISO 10426-6 is expected to be published by
early 2008.

Note should be taken of the Brazilian Standard NBR 9831 [8] mentioned already
in this paper in Table 2, which utilises normative limits for rheological properties
of MSR or HSR Class G cement determined by rotational viscometry using
procedures described in the Standard NBR 9830 [10]. This could act as a basis
for starting to develop normative requirements in this particular area of
viscometry on a global basis.
8. Conclusions

This paper has shown how research underpinning for international standards for
oilwell cements has been very useful for producing good quality standards that
are relevant to the industries (oil and construction) that need to make use of
these documents. All future international standards and revisions of the existing
ones, when the needs arise, require the input of modern scientific research.

References
1. J. Bensted, R. Shaunak. Proceedings of the 11th International Conference on
Cement Microscopy, April 10-13, 1989, New Orleans, Louisiana, pp. 198-224.
International Cement Microscopy Association, Duncanville, Texas, 1989.
2. J. Bensted. Chemical aspects of normal setting of Portland cement, in
‘Characterisation and Performance Prediction of Cement and Concrete’, pp.
69-86. Engineering Foundation Conference, New England College, Henniker,
New Hampshire, 25-30 July 1982. Engineering Foundation, Washington DC,
1983.
3. S.R. Farris, G. Jackson, J.S. Lota, J. Bensted. Influence of pressure alone on
thickening of Class G oilwell cement (in Polish and English). Cement-Wapno-
Beton No. 3, 121-126, 2005.
4. International Organisation for Standardisation. Petroleum and natural gas
industries – Cements and materials for well cementing – Part 1: Specification,
ISO 10426-1:2005.
5. J. Bensted. Developments with oilwell cements, in ‘Structure and
Performance of Cements’, (Eds. J. Bensted and P. Barnes), pp.237-252.
Spon Press, London and New York, 2002.
6. British Standards Institution. Petroleum and natural gas industries – Cements
and materials for well cementing – Part 2: Testing of well cements, BS EN
ISO 10426-2:2003.
7. W.J. Caveny, W. Weigand. Use cement microscopy to quickly check product
quality. World Oil 201, No. 5, 79-84, 1985.
8. Associação Brasileira de Normas Técnicas. NBR 9831 – Cimento Portland
destinado à cimentação de poços petrolíferos – Especificação, 1993.
9. J. Bensted. Difficulties in specifying free lime limits for oilwell cements. 16th
Meeting of the API Task Group on Eastern Hemisphere Cementing,
Wiesbaden, Germany, 23 March 1995.
10. Associação Brasileira de Normas Técnicas. NBR 9830 – Cimento Portland
destinado à cimentação de poços petrolíferos – Determinação das
propriedades reológicas – Método de ensaio, 1993.
11. International Organisation for Standardisation. Petroleum and natural gas
industries – Cements and materials for well cementing – Part 3: Testing of
deepwater well cement formulations, ISO 10426-3:2003.
12. J. Bensted: Oilwell cements. Part 3. Ductile oilwell cement compositions for
better long term durability (in Polish and English). Cement-Wapno-Beton No.
1, 13-32, 2005.
13. International Organisation for Standardisation. Petroleum and natural gas
industries – Cements and materials for well cementing – Part 4: Preparation
and testing of foamed cement slurries at atmospheric pressure, ISO10426-
4:2004.
14. International Organisation for Standardisation. Petroleum and natural gas
industries – Cements and materials for well cementing – Part 5:
Determination of shrinkage and expansion of well cement formulations at
atmospheric pressure, ISO 10426-5:2005.
15. American Petroleum Institute. Shrinkage and Expansion in Oilwell Cements,
API Technical Report 10TR 2, First Edition, July 1997.