SPE-183693-MS

Coffee as a Cement Retarder

Rahul Jadhav, Tamal K. Sen, and Sumukh Deshprabhu, Halliburton

Copyright 2017, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Middle East Oil & Gas Show and Conference held in Manama, Kingdom of Bahrain, 6-9 March 2017.

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
Retarders are one of the most important additives in oilwell cementing for achieving the desired pumping
time of cementitious slurry. Retarders are mainly extracted from natural sources, or they are chemically
synthesized. However, synthesized retarders can have significant environmental effects. This paper proposes
using coffee powder as a biocompatible retarder in oilwell cementing.
This paper discusses a study wherein coffee powder was used as a cement retarder and its effect on
other cement slurry properties was evaluated. Cement slurry was prepared and tested as per API RP 10B2
(2005) procedures for thickening time, compressive strength, fluid loss, and rheology. The performance of
coffee powder was tested at various concentrations and temperatures up to 350F. Its retardation activity
was compared to a commonly used lignosulfonate-based retarder as well as a synthetic retarder.
Coffee powder exhibited a predictable and linear thickening time behavior with respect to concentration
and temperature variation. Testing at different temperatures revealed that coffee powder alone can be
used at temperatures up to 350F without need for a retarder intensifier. A concentration sensitivity
study was performed at 125F, and the coffee powder displayed expected results. To achieve similar
performance, lower amounts of coffee powder are necessary compared to the lignosulfonate-based retarder.
The performance of coffee powder was similar to the synthetic retarder. Furthermore, coffee powder does
not cause adverse effects on other cement slurry properties, such as compressive strength development,
fluid loss, and rheology.
This work presents coffee powder performance as a cement retarder. Coffee is readily available, and
it demonstrates a high performance and linear concentration response without hampering other cement
properties.

Introduction
Portland cement has four main components: tricalcium silicate (C3S) and tricalcium aluminate (C3A), which
provide early strength to the system, dicalcium silicate (C2S), which is responsible for the final strength,
and tetracalcium aluminaferrite (C4AF). In addition, gypsum is added during the cement manufacturing
process to control the set time of cement. Various additives are also added to cement slurry to obtain desired
properties.
Retarders are one such widely used additive in cement slurry. Their primary function is to keep the
cement slurry in liquid form and provide the necessary pumping time. Typically, standard cement slurry
2 SPE-183693-MS

without any retarder using Class H or Class G cement can be pumped up to 8,000 ft where the bottomhole
circulation temperature (BHCT) is less than 125F (Brothers et al. 1991). However, the hydration rate of
cement increases with increased temperature, thereby decreasing thickening time. Thickening time can be
defined as the time during which cement slurry remains in a fluid state and is capable of being pumped.
The thickening time of the cement slurry is determined using a high-pressure/high-temperature (HP/HT)
consistometer.
Retarders used in oilwell cementing contain compounds, such as starch derivatives, hydroxyethyl
cellulose, calcium lignosulfonate, sodium lignosulfonate, sodium tetra borate decahydrate (borax), weak
organic acids, polyacrylic acid, etc. A retarder is selected on the basis of downhole temperature, cement
type, cost, and its effect on other slurry properties. Usually, a retarder needs to increase thickening time
without having a significant effect on compressive strength development, fluid loss, and other cement
slurry properties. Lignosulfonates are the most commonly used retarders. However, these retarders are not
applicable at high temperatures and, to extend the temperature limit, intensifiers are added, such as organic
acids or water-soluble borates, in combination with the lignosulfonate-based retarder.
Retarders can be divided into two categoriesnaturally derived and synthetic. Performance of naturally
derived retarders varies depending on the source. Synthetic retarders are preferred because they perform
more consistently compared to naturally derived retarders. However, some synthetic polymeric retarders
can significantly affect the environment. Thus, the need exists for an alternative biocompatible retarder that
provides consistent results and causes less significant environmental effects.
This paper establishes the use of coffee as a biocompatible, high-performance retarder. The performance
of coffee powder as a retarder is compared to a lignosulfonate and synthetic polymeric retarder. Coffee
is manufactured from roasted coffee beans, which are the seeds of berries from the Coffea plant. All
coffea plants are classified in the large family Rubiaceae. Although more than 80 coffea species have
been identified worldwide, only two are economically important. Coffea arabica, also known as Arabica
coffee, comprises approximately 70% of the global coffee market, and Coffea canephora, or Robusta coffee,
accounts for the remainder (Farah 2012).

Retarder Mechanism
The retarder mechanism is based on five proposed theories: adsorption, precipitation, nucleation,
complexation, and dissolution-precipitation (Nelson and Guillot 2006).

Adsorption theory: According to the adsorption theory, the retarder becomes adsorbed on the
surface of the hydration products (C-S-H gel, generated by hydration of C3S and C2S), inhibiting
its contact with water and rendering it hydrophobic (Coveney and Humphries 1996).
Precipitation theory: According to this theory, the retarder reacts with calcium and/or hydroxyl
ions and forms a protective semipermeable sheet covering cement grains. The semipermeable
membrane is formed on the surface of C-S-H gel, which slows the migration of water to the
mineral surface. Ultimately, the protective semipermeable coating ruptures because of the high
pressure developed by osmosis-driven diffusion. Consequently, hydration continues at a regular
rate (Double et al. 1983).
Nucleation theory: The nucleation theory suggests that the retarder slows the growth rate of
hydration products by adsorbing on their nuclei. The retarder mainly inhibits the growth of the
initially formed highly reactive nano nuclei of cement hydrates by occupying their reactive faces
(Thomas and Birchall 1983).
Complexation theory: This states that calcium ions are chelated by the retarder, preventing the
formation of the nuclei. Here, the retarder reduces the concentration of available calcium ions
existing in pore solution through chelating the calcium ions, thus inhibiting the formation of cement
hydrates, such as, for example, ettringite (Ramachandran et al. 1981).
SPE-183693-MS 3

Dissolution-precipitation theory: The dissolution-precipitation process is mainly applicable

for organic phosphonic acid-based retarders. First, dissolution of calcium ions is enhanced by
extraction. Thereafter, a layer of water-insoluble calcium phosphonate precipitate is formed, which
creates a diffusion barrier to water and inhibits nucleation through binding to the surface of the C-
S-H gel (Bishop and Barron 2006).
For a particular retarder, it is difficult to conclude which specific mechanism is responsible. However, it
is likely that all of the mentioned mechanisms are involved to some extent during the retardation process
and the theories complement one another.

How Coffee Acts as a Retarder

The basic composition of coffee is water, carbohydrates, proteins, free amino acids, lipids, minerals, organic
acids, chlorogenic acids, trigonelline, and caffeine (Table 1) (Farah 2012). Of these, the compounds of
chlorogenic acids, caffeine, trigonelline, soluble fiber, and diterpenes from the lipid fraction are most likely
to be bioactive (Panzella et al. 2016), and they may also be important contributors to the beverage flavor
after roasting.

Table 1Chemical composition of green Coffea arabica and Coffea canephora seeds.

Component Concentration (g/100 g)

Coffea Arabica Coffea Canephora

Carbohydrates/fiber

Sucrose 6.0 to 9.0 0.9 to 4.0

Reducing sugars 0.1 0.4

Polysaccharides 34 to 44 48 to 55

Lignin 3 3

Pectin 2 2

Nitrogenous compounds

Protein/peptides 10.0 to 11.0 11.0 to 15.0

Free amino acids 0.5 0.8 to 1.0

Caffeine 0.9 to 1.3 1.5 to 2.5

Trigonelline 0.6 to 2.0 0.6 to 0.7

Lipids

Coffee oil (triglycerides with

15 to 17.0 7.0 to 10.0
unsaponifiables, sterols/tocopherols)

Diterpenes (free and esterified) 0.5 to 1.2 0.2 to 0.8

Minerals

Minerals 3.0 to 4.2 4.4 to 4.5

Acids and esters

Chlorogenic acid 4.1 to 7.9 6.1 to 11.3

Aliphatic acid 1 1

Quinic acid 0.4 0.4

Amongst the coffee constituents, sucrose, polysaccharides, lignin, and acids are well known for their
retardation activity in cement chemistry. The current existing retarders used for oilwell cementing are mostly
4 SPE-183693-MS

acids and lignin based. Thus, the retardation activity of coffee can be attributed to the presence of these
components.

Experimental Procedure and Methods

Tests were performed using Dyckerhoff Class G cement/Joppa Class H cement, a fluid-loss additive, a free-
fluid control additive, and silica flour. Coffee powders from four different brands (Nescafe, India; BRU,
India; Douwe Egberts, UK; and Hintz, Germany) were used. Cement slurries were mixed and tested as
per API RP 10-B2 (2005). The dry additives were blended with cement and the liquid additives with tap
water. The tests were performed using an HP/HT consistometer (Model 290), atmospheric consistometer
(Model 165AT), ultrasonic cement analyzer (Model 304), static fluid-loss apparatus, and FANN viscometer
(Model 35).
Experimental data were gathered for coffee performance as a retarder at 125F with varied concentrations.
Likewise, coffee was tested between temperature ranges of 120 and 350F by varying its loading in the
cement slurry. Furthermore, a study was performed comparing coffee with a lignosulfonate-based retarder
and synthetic polymeric retarder. Additionally, the effect of coffee powder from four different brands on
thickening time and strength development was evaluated.

Results and Discussion

Predictable Thickening Time Behaviour
Table 2 shows the thickening time results obtained for various coffee concentrations in cement slurry.
Here, 15.8-lbm/gal neat cement slurry was prepared using fresh water, Dyckerhoff Class G cement, and
coffee powder. Neat cement slurry was used to eliminate the effect of other additives and study the coffee
concentration effect in isolation. The thickening time test was performed using an HP/HT consistometer
at 125F and 5,200 psi. The thickening time for 0% by weight of cement (BWOC) coffee was observed
to be approximately 2 hours, whereas it was 9:45 for 0.3% BWOC. The neutral concentration of coffee,
where it did not influence the thickening time of the slurry, was approximately 0.05% BWOC, where a 2:10
thickening time was achieved. Fig. 1 depicts that thickening time value increased predictably by increasing
the coffee powder concentration.

Table 2Coffee concentration vs. thickening time of cement slurry.

Coffee Concentration Thickening Time,

Sr. No. (% BWOC) 70 Bc (hr:min)

1 0 01:50

2 0.05 02:10

3 0.10 02:50

4 0. 20 05:20

5 0. 30 09:45
SPE-183693-MS 5

Figure 1Thickening time for various coffee concentrations.

Effect of Coffee Concentration on Thickening Time at Various Temperatures

Coffee was evaluated as a retarder in 16.5-lbm/gal cement slurry at various temperatures: 120, 150, 200,
250, 300, and 350F. Cement slurry was prepared using Premium Class H cement, 4.30 gal/sk of fresh
water, 35% BWOC coarse silica flour, and 0.5% BWOC fluid-loss additive. A thickening time test was
performed in an HP/HT consistometer at 10,200 psi. Fig. 2 shows the thickening time as a function of
coffee concentration in cement slurry at temperatures between 120 and 300F. The performance of coffee
as a retarder was further tested at 350F, which showed a thickening time of approximately 3 hours at a
concentration of 3.5% BWOC. It can be observed that the thickening time of cement slurries was consistent
and predictable. Furthermore, coffee provided a right-angle set to the cement slurry at all temperatures
tested. Figs. 3 and 4 show the thickening time at 150 and 300F, respectively. At 150F (Fig. 3), a 7:23
pump time was achieved with a concentration of 0.15% BWOC coffee. At 300F (Fig. 4), a 4:30 pump time
was achieved with 2% BWOC coffee.

Figure 2Thickening time vs. coffee concentration at various temperatures.

6 SPE-183693-MS

Figure 3Thickening time at 150F with 0.15% BWOC coffee.

Figure 4Thickening time at 300F with 2.0% BWOC coffee.

Coffee vs. Lignosulfonate and Synthetic Retarder

Three stable and pumpable slurry designs were conceived using different retarders: a lignosulfonate retarder,
a synthetic retarder, and coffee powder. The retarder concentration was adjusted to achieve a thickening
time between 5 to 6 hours. A thickening time test was performed using an HP/HT consistometer at 190F
and 5,200 psi. Compressive strength was measured using a UCA at 210F and 3,000 psi. Table 3 compares
performance of the three retarder types.
SPE-183693-MS 7

Table 3Performance comparison of coffee with lignosulfonate and synthetic retarder.

Slurry density, 14.5 lbm/gal; fresh water, 5.35 gal/sk; Dyckerhoff Class G cement; fluid loss additive, 1.5 gal/sk; free
fluid control additive, 0.03 gal/sk; defoamer, 0.02 gal/sk

Retarder Lignosulfonate Synthetic Coffee No Retarder

Retarder amount (% BWOC) 0.19 0.09 0.09

Thickening time at 190F

Thickening time 70 Bc (hr:min) 05:28 05:29 05:43 03:00

UCA compressive strength at 210F

50-psi Strength (hr:min) 06:09 05:46 06:24 03:39

500-psi Strength (hr:min) 07:03 07:07 07:31 04:48

24-hr Compressive strength (psi) 1,569 1,386 1,447 1,614

48-hr Compressive strength (psi) 1,764 1,624 1,643 1,826

API fluid loss at 190F

API fluid loss (mL/30 min) 64 70 52 54

According to Table 3, 0.09% BWOC coffee powder was necessary to achieve a thickening time of 5:43.
For the lignosulfonate retarder to achieve a similar thickening time, 0.19% BWOC was necessary, more
than double that of the coffee powder. However, the synthetic retarder time was close to that of the coffee
powder to achieve a similar thickening time.
The coffee powder design achieved a 500-psi ultrasonic compressive strength (UCS) in 7.5 hours and
developed a compressive strength of 1,447 psi in 24 hours. However, for the lignosulfonate and synthetic
retarders, the time to achieve 500-psi compressive strength was approximately 7 hours. After 24 hours, the
lignosulfonate and synthetic retarders developed a compressive strength of 1,569 and 1,386 psi, respectively.
This shows that coffee powder does not adversely affect the strength development of cement slurry.
API fluid loss at 190F for cement slurry containing coffee powder was 52 mL, whereas cement slurries
containing lignosulfonate and synthetic retarder provided 64 and 70 mL fluid loss, respectively. This shows
that the effect of coffee powder on fluid loss is similar to the other retarders.
Rheological measurements were performed immediately after mixing the slurry at 80F. Thereafter, the
slurry was conditioned in an atmospheric consistometer at 190F for 30 minutes. Afterward, rheological
measurements were performed at 190F. Figs. 5 and 6 show that no significant difference can be observed
in the rheogram of coffee powder, lignosulfonate retarder, and synthetic retarder at 80F. At 190F, no
significant difference can be observed in the rheogram of the coffee powder and lignosulfonate retarder,
while the viscosity of the synthetic retarder increased.
8 SPE-183693-MS

Figure 5Effect of shear rate on shear stress at 80F.

Figure 6Effect of shear rate on shear stress at 190F.

An overall comparison shows that approximately 50% less coffee powder is necessary compared to the
lignosulfonate retarder to achieve a similar thickening time. However, the performance of coffee powder and
synthetic retarder was comparable. Additionally, coffee powder had no detrimental effects were observed
on other slurry properties, such as compressive strength development, fluid loss, and rheology.

Effects of Various Coffees on Thickening Time and Strength Development

Cement slurry was prepared using four different coffee powders as retarders. A thickening time test was
performed in an HP/HT consistometer at 190F and 5,200 psi. Compressive strength was measured using
a UCA at 210F and 3,000 psi. Table 4 compares the performance of each coffee powder retarder.
SPE-183693-MS 9

Table 4Performance comparison of four different coffee brands.

Slurry density, 14.5 lbm/gal; fresh water, 5.35 gal/sk; Dyckerhoff Class G cement; fluid loss additive,
1.5 gal/sk; free fluid control additive, 0.03 gal/sk; defoamer, 0.02 gal/sk; coffee powder, 0.09% BWOC

Douwe Hintz Nescafe

Coffee brand BRU (India) No Retarder
Egberts (UK) (Germany) (India)

HP/HT thickening time

70 Bc (hr:min) 05:19 05:51 06:58 05:43 03:00

UCA compressive strength

50-psi Time (hr:min) 06:52 06:07 06:09 06:24 03:39

500-psi Time (hr:min) 08:08 07:24 07:29 07:31 04:48

24-hr Compressive strength (psi) 1,483 1,462 1,498 1,447 1,614

48-hr Compressive strength (psi) 1,719 1,692 1,718 1,643 1,826

Fig. 7 shows thickening time, time to reach 500-psi UCS, and 24-hour compressive strength for the
samples tested. The thickening times of all coffee powders were between 5 and 7 hours. The 500-psi
UCS was achieved in approximately 8 hours, and the samples achieved a 24-hour compressive strength of
approximately 1,500 psi. It is evident that variations in coffee powder do not significantly impact thickening
time and that coffee powder does not affect strength development of the cement slurry.

Figure 7Performance comparison of four different coffee brands.

Conclusions
The following conclusions are a result of this work:

Coffee acts as a cement retarder over a broad temperature range of 120 to 350F.

Coffee can be used successfully up to 350F without needing additional intensifiers, such as organic
acid or soluble borate salt.
Coffee provides a predictable thickening time without affecting other properties, such as
compressive strength development and fluid loss.
10 SPE-183693-MS

Coffee performed better than the lignosulfonate-based retarder, with a similar thickening time
achieved using less than half the amount of the lignosulfonate-based retarder.
The performance of coffee powder was comparable to the synthetic polymeric retarder when tested
for properties, such as thickening time, compressive strength development, and fluid loss.
The results of this study should encourage use of coffee as a biocompatible retarder.

Acknowledgement
The authors thank the management of Halliburton Technology Center, Pune for the opportunity to publish
this paper. They also thank and appreciate the strong support of the Halliburton Technical Paper Review
Board, Houston, Texas.
Nescafe is a registered trademark of Nestl, Vevey, Switzerland.
FANN is a registered trademark of Fann Instrument Company, Houston, Texas.

References
API RP 10B-2, Petroleum and natural gas industries - Cements and materials for well cementing - Part 2: Testing of well
cements, first edition. 2005. Washington, D.C.: API.
Bishop, M. and Barron, A.R. 2006. Cement Hydration Inhibition with Sucrose, Tartaric Acid, and Lignosulfonate:
Analytical and Spectroscopic Study. Ind Eng Chem Res 45 (21): 70427049.
Brothers, L.E., Chatterji, J, Childs, J.D. et al 1991. Synthetic Retarder for High-Strength Cement. Presented
at the SPE/IADC Drilling Conference, Amsterdam, The Netherlands, 1114 March. SPE-21976-MS. http://
dx.doi.org/10.2118/21976-MS.
Coveney, P.V. and Humphries, W. J. 1996. Molecular modelling of the mechanism of action of phosphonate retarders on
hydrating cements. Chem Soc, Faraday T 92: 831841.
Double, D.D. Hewlett, P.C., Sing, K.S.W. et al 1983. New Developments in Understanding the Chemistry of Cement
Hydration [and Discussion]. Phil Trans R Soc Lond A 310 (1511): 5366.
Farah, A. 2012. Coffee Constituents, in Coffee: Emerging Health Effects and Disease Prevention, ed. Y.-F. Chu. Oxford,
UK: Wiley-Blackwell. doi: 10.1002/9781119949893.ch2.
Nelson, E.B. and Guillot, D. 2006. Well Cementing. Sugar Land, Texas: Schlumberger Dowell.
Panzella, L., Cerruti, P., Ambrogi, V. et al 2016. A Superior All-Natural Antioxidant Biomaterial from Spent Coffee
Grounds for Polymer Stabilization, Cell Protection, and Food Lipid Preservation. ACS Sustainable Chem Eng 4 (3):
11691179.
Ramachandran, V.S., Feldman, R., and Beaudoin, J. 1981. Concrete Science: Treatise on Current Research. Philadelphia,
Pennsylvania: Heyden and Son. Ltd.
Thomas, N.L. and Birchall, J.D. 1983. The retarding action of sugars on cement hydration. Cem Concr Res 13 (6):
830842.