See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/305929635

A Comparative Study on the Effect of Cement Source on Engineering

Properties of Concrete

Conference Paper · October 2009

CITATIONS READS

0 2,576

2 authors, including:

Dr. Nurdeen Altwair

36 PUBLICATIONS   211 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

this is time, I am working on UHSC. View project

‫ دراﺳﺔ اﻟﺨﻮاص اﻟﻤﻴﻜﺎﻧﻴﻜﻴﺔ ﻟﺮﻛﺎم ﺧﺒﺚ اﻻﻓﺮان اﻟﻌﺎﻟﻴﺔ ﺑﻤﺼﻨﻊ اﻟﺤﺪﻳﺪ و اﻟﺼﻠﺐ ﺑﻤﺼﺮاﺗﺔ‬View project

All content following this page was uploaded by Dr. Nurdeen Altwair on 06 August 2016.

The user has requested enhancement of the downloaded file.

 5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

A Comparative Study on the Effect of Cement Source on Engineering Properties

of Concrete

Nurdeen M. Altwair
Civil Engineering Department, Faculty of Engineering,
Al-Merghab University, Al-Khums, Libya
: [email protected]
and
Shahid Kabir*
School of Civil Engineering, Engineering Campus,
University Sciences Malaysia,
14300 Nibong Tebal, P. Pinang, Malaysia

[email protected]

Abstract
Ordinary Portland cement is one of the most important materials used in manufacturing concrete in Libya. In
view of the increasing demands for this material on the local market, the Libyan government, as well as some
businessmen, has imported this material from outside countries, such as Romania (Romanian cement), Egypt (Aen
Sokhnah and Alqumia cements) and Tunisia (Alsonoper cement), to supplement the use of local Almerqeb and
Lebda cements. It is known that the source of raw materials for manufacturing any product may affect the properties
of that product. Consequently, the aim of this paper is to study the chemical and physical characteristics, as well as
the engineering properties, such as compressive strength and absorption, of concrete made from local and imported
cements in order to provide a comparison between the various cements. Furthermore, this research examines the
possibility of mixing local cements with some imported cements, such as Alqumia and Romanian cements. The
results indicate that the best types of cement, according to their source, are the Romanian cements, followed by local
cements.
Key words: ordinary Portland cement; chemical and physical properties; compressive strength; absorption

‫ـــــــــــــــــــــــــــــــــــــــ‬
*Corresponding author:
Civil Engineering Department, Faculty of Engineering, Al-Merghab University, Al-Khums, Libya
Mob. . 218927253938;
E-mail: [email protected]
Classification: Civil Engineering, Structural Engineering, Concrete Technology.

1. Introduction

1
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

Libya produces about 7,530 thousand tons of cement annually [1]; however, this amount does not meet the
growing demands for this material. This is due to the increase in building projects developed to deal with
growing infrastructure requirements. The vast amount of cement used in these projects has rendered the national
production of cement inadequate for satisfying Libya's self sufficiency. Over the last decade, many quantities of
cement have been imported from different countries, such as Romania (Romanian cement, RC), Egypt (Aen
Sokhnah and Alqumia cements, ASC and AC) and Tunisia (Alsonoper cement, ARC), to supplement the use of
local Almerqeb (AMC) and Lebda cements (LC). These imported cements have not been evaluated according to
scientific and economic standards dealing with production quality in order to assess their suitability for use in
local conditions, which has led to a negative effect on the users, especially in the construction of building
skeletons. This research aims to assess the technical quality of imported ordinary Portland cement and to
compare it with local cements. In addition, this research studies the engineering properties of both concrete and
mortar using the imported cements, especially when added to the local material, and to study the influence of
blending local and imported cements. The results of this study are used to provide some standard index in order
to improve users' trust in imported cements.

2. Background
It is estimated that about 1.8 billion tonnes of OPC are manufactured annually worldwide [2]. Natural cements
first appeared in England in 1796 when James Parker patented his „„Roman cement.‟‟ However, the importance
of natural cements was eclipsed by the appearance of Aspdin‟s ordinary Portland cement in 1824 [3]. Cement is
a basic material for concrete used in the construction of a concrete skeleton, which represents about 30% of the
components of a concrete structure. Cement is the adhesive which holds together the construction material, such
as bricks and tiles. But more importantly, it is the largest ingredient in the modern miracle we refer to as
concrete. Because concrete is the most widely used of all construction materials in the world today, the
manufacture of cement is widespread. Each year, almost one ton of concrete is poured, per capita in developed
countries. Cements may be used alone (i.e., “neat,” as grouting materials), but the normal use is in mortar and
concrete, where the cement is mixed with an inert material known as aggregate. Mortar is cement mixed with
sand or crushed stone that must be less than approximately 5mm (3/16 inch) in size [4]. Concrete is a mixture of
cement, sand or other fine aggregate, and a coarse aggregate, which for most purposes is up to 20mm to 25mm
(3/4 to 1 inch) in size, but the coarse aggregate may also be as large as 150mm (6 inches) when concrete is
placed in large masses such as dams. Mortars are used for binding bricks, blocks, and stones in walls or as
surface renderings. Concrete is used for a large variety of construction purposes. Mixtures of soil and Portland
cement are used as a base for roads. Portland cement is also used in the manufacture of bricks, tiles, shingles,
pipes, beams, railroad ties, and various extruded products. These products are prefabricated in factories and
supplied ready for installation.
The raw materials are quarried or mined and transferred to the manufacturing facility to be crushed and milled
into a fine powder before entering a preheater and eventually a large rotary kiln where materials reach
temperatures greater than 1400°C. Therefore, there are four stages in the manufacture of Portland cement: (1)
crushing and grinding the raw materials, (2) blending the materials in the correct proportions, (3) burning the
prepared mix in a kiln, and (4) grinding the burned product, known as “clinker,” together with some 5% of
gypsum to control the curing time of the cement. There are three processes known as the wet, dry, and semidry
processes: the wet process is when the raw materials are ground wet and fed to the kiln as a slurry, the dry
process is when the materials are ground dry and fed as a dry powder, and the semidry process is when the
materials are ground dry and then moistened to form nodules that are fed to the kiln, no material other than
gypsum, water, and grinding aids may be added after burning [5]. Portland cement consists essentially of
compounds of lime (calcium oxide, CaO) mixed with silica (silicon dioxide, SiO2) and alumina (aluminum
oxide, Al2O3). The lime is obtained from a calcareous (lime-containing) raw material, and the other oxides are
derived from an argillaceous (clay-type) material. Additional raw materials such as silica sand, iron oxide
(Fe2O3), and bauxite (containing hydrated aluminum, Al[OH]3 may be used in smaller quantities to get the
desired composition ([6,7].
The commonest calcareous raw materials are limestone and chalk, but others, such as coral or shell deposits, are
also used. Clays, shales, slates, and estuarine muds are the common argillaceous raw materials. Marl, a compact
calcareous clay, and cement rock, both contain the calcareous and argillaceous components in proportions that
sometimes approximate cement compositions. Another raw material is blast-furnace slag, which consists mainly

2
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

of lime, silica, and alumina and is mixed with a calcareous material of high lime content. Kaolin, white clay that
contains little iron oxide, is used as the argillaceous component for white Portland cement. Industrial wastes,
such as fly ash and calcium carbonate from chemical manufacture, are other possible raw materials, but their
use is small compared with that of the natural materials. The magnesia (magnesium oxide, MgO) content of raw
materials must be low because the permissible limit in Portland cement is 4% to 5%.
Other impurities in raw materials that must be strictly limited are fluorine compounds, phosphates, metal oxides
and sulfides, and excessive alkalis. Another essential raw material is gypsum, some 5% of which is added to the
burned cement clinker during grinding to control the setting time of the cement. Portland cement can also be
made in a combined process with sulfuric acid using calcium sulfate or anhydrite in place of calcium carbonate.
The sulfur dioxide produced in the flue gases on burning is converted to sulfuric acid by normal processes.
Typical compositions of ordinary Portland cement are given in Table 1.
Table 1. Typical compositions of ordinary Portland cement
Basic Compound Oxide Formula Symbol Content (%)

Silica SiO2 S 17-25

Calcium Oxide (lime) CaO C 60-67
Aluminum Oxide Al2O3 A 3-8
Ferric Oxide Fe2O3 F 1.5-4.5
Magnesium Oxide MgO M 0.5-4.5
Potassium Oxide K2O K 0.1-1.5
Sodium Oxide Na2O N 0.1-1.0
Sulphur Trioxide SO3 S 1-3
Water H2O H -

There are four compounds that were recognized as the major compounds, silicates and aluminates, in Portland
cement clinker through the research [8]. The composition can be represented in a simple way; for example
dicalcium silicate can be represented as 2CaO.SiO2, as a combination of two separate oxides as Ca2SiO4, or as
the calcium salt of orthosilicic acid H4SiO4 [7]. Table 2 shows the main compounds of Portland cement [9].

Table 2. Main Compounds of Portland Cement

Name of compound Oxide composition Abbreviate

Tricalcium Silicate 3CaO.SiO2 C3S

Dicalcium Silicate 2CaO.SiO2 C2S

Tricalcium Aluminate 3CaO.Al2O3 C3A

Tetracalcium Auminoferrite 4CaO.Al C2O3.Fe2O3 C4AF

The most important hydraulic constituents are the calcium silicates, C2S and C3S. Upon mixing with water, the
reaction between cement and water leads to the formation of a number of hydration products including an
amorphous or poorly crystalline calcium silicate hydrate C–S–H (3CaO.2SiO2.3H2O), calcium hydroxide CH
(Ca[OH]2) [10,11]. and the hydration reaction can be crudely represented by the following reactions
(Neville,1982):

C3S + H2O → C-S-H + Ca[OH]2 (1)

C2S + H2O → C-S-H + Ca[OH]2 (2)

3
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

where C3S is preferentially formed upon elevating the temperature and increasing amount of added burned lime,
CaO. C3S is responsible for short term strength development (days to months) while C2S displays the better
long term strength development performances [12].
During the initial stage of hydration, the parent compounds dissolve, and the dissolution of their chemical bonds
generates a significant amount of heat. Then, for reasons that are not fully understood, hydration comes to a
stop. This quiescent or dormant period is extremely important in the placement of concrete. Without a dormant
period there would be no cement trucks; pouring would have to be done immediately upon mixing.
Following the dormant period, which can last several hours, the cement begins to harden, as CH and C–S–H are
produced. This is the cementitious material that binds cement and concrete together. As hydration proceeds,
water and cement are continuously consumed. Fortunately, the C–S–H and CH products occupy almost the
same volume as the original cement and water; volume is approximately conserved, and shrinkage is
manageable [13]. Although the formulas above treat C–S–H as a specific stoichiometry, with the formula
C3S2H3, it does not at all form an ordered structure of uniform composition. C–S–H is actually an amorphous
gel with a highly variable stoichiometry [14]. The ratio of C to S, for example, can range from 1:1 to 2:1,
depending on the mix design and curing conditions. The reaction of pure C3A with water is very violent and
leads to immediate stiffening of the paste, known as flash setting. To prevent this from happening, gypsum
(CaSO4.2H2O), is added to cement clinker. Gypsum and C3A react to form insoluble calcium sulphoaluminate
(3CaO.Al2O3.3CaSO4. 3H2O), but eventually a tricalcium aluminate hydrate is formed. Thus the final form of
the reaction can be written as [9]:

C3A+H2O→C4AH13+C2AH8→C3AH6 (3)

The presence of C3A in cement is undesirable, since it contributes little or nothing to the strength of cement
except at early ages, and when hardened the cement paste is attacked by sulphates, and expansion due to the
formation of calcium sulphoaluminate from C3A may result in a disruption of the hardened paste. However,
C3A acts as a flux, thus reducing the temperature for burning the clinker and facilitating the combination of lime
and silica. For these reasons C3A is useful in the manufacture of cement. Also gypsum does not only react with
C3A; with C4AF it forms calcium sulphoferrite as well as calcium sulphoaluminate, and its presence may
accelerate the hydration of the silicates [15]
The progress of hydration of the cement can be determined by different means, such as through measuring the
amount of Ca[OH]2 in the paste, the amount of heat evolved by hydration, the specific gravity of the paste, the
amount of chemically combined water, the amount of unhydrated cement present (using X-ray quantitative
analysis) and indirectly, from the strength of the hydrated paste [16]. The heat of hydration studies can be used
for characterizing the setting and hardening behaviour of cements, and predicting temperature rise [17].
The fineness of cement is a measure of the size of the particles in the cement and is expressed in terms of
specific surface. Thus, the rate of hydration depends on the fineness of the cement particles; cements with a
higher fineness will tend to hydrate faster and develop a high early strength [18]. Finer cement leads to a
stronger reaction with alkali-reactive aggregates [9,18]. . An increase in fineness increases the amount of
gypsum required for proper retardation, since more C3A is available for early hydration in finer cement. The
water content of a paste of standard consistency is greater for the finer cement. In the past, the fraction of
cement retained on a 90μm test sieve was determined, and the maximum residue was limited to 10% by weight
for ordinary Portland cement and 5% for rapid hardening Portland cement (No. 170ASTM). A more recent
method of determining the specific surface of cement is the air permeability method, using an apparatus
developed by Lea and Nurse [15]. , which is based on the relation between the flow of a fluid through a
granular bed and the surface area of the particles comprising the bed [9]. This is the method of measurement
prescribed by BS 4550.
By changing the chemical composition of the cement through varying the percentage of the four basic
compounds, it is possible to obtain several types of cement, each with some unique characteristics.
Consequently, cement can be manufactured for the required performance of a specific application.
Ordinary Portland cement is admirably suitable for use in general concrete construction when there is no
exposure to sulphates in the soil or in ground water. Therefore, it is considered the most common type of
cement in use. A further requirement of BS 12:1978 for the chemical composition of ordinary Portland cement

4
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

is that the magnesia content should not exceed 4.0%. Formerly, the ratio of Al2O3/Fe2O3 was specified to not be
less than 0.66. In addition, the insoluble residue was not to exceed 1.5% and the loss on ignition was limited to
3% in temperate climates and 4% in the tropics. The maximum gypsum content was also a specified maximum
of 2.5% when the C3A content is not more than 5% and 3.0% when the amount of C3A exceeds 5%. Over the
years, there have been some changes in the characteristics of ordinary Portland cement. In particular, modern
cements have a higher C3S content and a greater degree of fineness [15].

3. Experimental program
3.1 Materials
Two types of water were used in this study, water used to determine the chemical and physical properties of
cement, and water used for the preparation of concrete containing cement. The water used to determine the
chemical and physical properties of cement conforms to the specifications in BS 3148: 1980, such as dissolved
solids in the water not exceeding 2000 parts per million (ppm), Alkali carbonate or bicarbonate not exceeding
1000ppm, Chloride not exceeding 500ppm and Sulphates not exceeding 100ppm. The water used in the
preparation of the concrete was normal drinking water, containing dissolved solids not exceeding 2000ppm.
Standard sand was employed, consisting of pure siliceous material of uniform size with all particles nearly
spherical; the sand was free from organic impurities [16], and at least 90% of it was retained on a 600μm sieve
(No. 25 BS) [15]. This sand was used to determine the chemical and physical properties of the cement.
Mechanical properties of the aggregates are presented in Table 3. In order to be within the specification limits,
the gradation of the aggregates and the mix proportions were set at 0.25 fine + 0.3 size 3/4" + 0.45 size 1",
where the maximum aggregate size is 25mm for size 1" and the maximum aggregate size is 19 mm for size
3/4". The grading curve of the combined aggregates is shown in Figure 1.

Table 3. Mechanical properties of the aggregates

Results
Test Specification Limits
3/4" 1"
Specific gravity 2.46 2.65 2.7-2.6 (ASTM.C33)
Crushing value 24% 23% < 25% (BS.812)
Impact 14.6% 16.2% ≤ 30% (ASTM.C33)
Los Angeles abrasion value 13.0% 12.0% <40% (BS.812)
Loss of aggregate 25.6% 27.9% < 40% (BS.812)
Absorption 1.9 % 1.4% ≤ 3.44% (ASTM.C33)

3.2 Specimen preparation

The samples for this research were randomly obtained at different periods, so as to ensure that no sample was
from the same shipment and to guarantee no resemblance in production time of the samples of cement. All
samples of Romanian cement (RC) were taken from shipments at Alkhoms port; two samples were taken from
every cargo and the time between each shipment was 30 days. All samples of Al-qumia (AC), Aen Sokhnah
(ASC) and Al-Sonoper (ARC) cements were taken from the local markets in the cities of Tripoli, Alkhoms and
Musratah, at different periods, ranging from one month to three months. Samples of Lebda (LC) and Al-merqeb
(AMC) cements were taken directly from Lebda and Almerqeb factories after mobilization. Each sample of
blended cement was produced from a mix of local and imported cement at a ratio of 1:1 by weight. The samples
weighed about 50kg each, and were divided into two parts; the first part weighed about 5.00kg and was
employed for the chemical and physical experiments, and the second part weighed about 45.00kg and was used
for the compressive strength and concrete absorption experiments.

5
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

Fig. 1. Sieve analysis of combined aggregates: 0.25 fine aggregate, 0.3 size 3/4” and 0.45 size 1”.

3.3 Testing procedures

The manufacture of cement requires quality control, and a number of tests are performed in the cement works
laboratory to ensure that the cement is of the desired quality and that it conforms to the requirements of the
relevant national standards. It is desirable for the purchaser or for independent laboratories to prepare quality
tests or to examine the properties of a cement to be used for some special purpose. Figure 2 shows a scheme
work of tests for cement paste and concrete, which has been carried out in this study.

Fig 2. Scheme of testing program.

6
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

3.3.1 Specification of chemical tests

3.3.1.1 Determination of the main compounds in cement

The calculation of the main compounds of ordinary Portland cement is based on the work of Bogue [15]. .
Bogue‟s equations for the percentages of main compounds in cement are shown in equations 1 to 4. The
specification limits of the main compounds of OPC according to BS EN 197-1, are C3S (40% to 65%), C2S
(10% to 30%), C3A (6% to 12%), C4AF (6% to 13%).

C3S=4.07(CaO)-7.60(CaO.SiO2)-6.72(Al2O3)2.85(SO3) (4)
C2S=2.87(CaO)-0754(CaO.SiO2) (5)
C3A=2.65(Al2O3)-1.69(Fe2O3) (6)
C4AF=3.04(Fe2O3) (7)

3.3.1.2 Lime saturation factor and free calcium oxide

The specifications for ordinary Portland cement are given in BS 12/1996, and in Libyan standards 97/340. The
lime saturation factor is not to be greater than 1.02 and not less than 0.66. This factor can be defined as below:

(8)

where terms in brackets denote the percentage by weight of the given compound present in the cement. Free
calcium oxide is limited to 2% in both specifications.

3.3.1.3 Alumina modulus (AM)

In both specification BS 12/1996 and Libyan Standards 97/340, the value of the alumina modulus should not be
less than 0.66. The AM factor is defined as.

(9)

3.3.1.4 Sulphate content and Magnesia content

BS EN 196 specifies the value of the Sulphate content to be between 3% and 3.5%. Libyan Standards 97/340
gives the value of the Sulphate content for two cases: when Tricalcium Aluminate is less than 8%, the sulphate
content should not be more than 2.5%, and when Tricalcium Aluminate is more than 8%, the sulphate content
should not be more than 3%. BS 12/1996 and Libyan Standards 97/340 give the Magnesia content to be not
more than 4%.

3.3.2 Specifications of Physical and mechanical testing

3.3.2.1 Consistency of Standard Paste, setting time soundness and specific surface

Neat cement paste of a standard consistency has to be used for the determination of the initial and final setting
times, as well as for the Le Chatelier soundness test. The consistency is measured through the Vicat apparatus,
using a 10mm diameter plunger fitted into the needle holder. The water content of the standard paste is
expressed as a percentage by weight of the dry cement. According to Libyan standards 97/340, the range of
values should be between 26% and 33%.
The setting times of cement are also measured using the Vicat apparatus with different penetrating attachments.
For the determination of the initial set, a round needle with a diameter of 1.13 ± 0.05mm is used. This needle,
acting under a prescribed weight, is used to penetrate a paste of standard consistency placed in a special mould.

7
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

When the paste stiffens sufficiently for the needle to penetrate only to a point 5 ± 1mm from the bottom, initial
set is said to have taken place. Initial set is expressed as the time elapsed since the mixing water was added to
the cement. A minimum time of initial set of 45 minutes is prescribed by BS 12/1996 and Libyan Standards
97/340. Final set is determined by a similar needle fitted with a metal attachment hollowed out so as to leave a
circular cutting edge 5mm in diameter and set 0.5 mm behind the tip of the needle. The final setting time is
considered from the moment when mixing water is added to the cement, and is required by the relevant British
Standards and Libyan Standards 97/340 to be not more than 10 hours.
Unsoundness of cement is not apparent until after a period of months or years; thus, it is essential to test the
soundness of cement in an accelerated manner. The Le Chatelier apparatus is used to determine the soundness
of cement. According to Libyan Standards 97/340 the soundness of ordinary Portland cement should be less
than 10mm. In this study the Blaine test was also used; it measures the volume of air passing at a prescribed
average pressure, with the rate of flow diminishing steadily. The time (t) for the flow to take place is measured;
for a given apparatus and a standard porosity of 0.500, the specific surface is given by Equation 10.

(10)

Where K2 is a constant.
BS 12/1996 and Libyan Standards 97/340 specify the minimum specific surface as 2250 cm²/gm for ordinary
Portland cement.

3.3.2.2 Strength of Cement

In this test, the mortar proportions were 1:3 (555gm sand + 185gm cement). Mixing water was used at a
water/cement ratio of 0.4 for all mixes by weight (74gm water). A standard procedure, prescribed by BS 4550:
Part 3: Section 3.4 : 1978, was followed in mixing the mortar, and 70.7mm cubes were made using a vibrating
table with a frequency of 200 Hz applied for two minutes. The cubes were demoulded after 24 hours and further
cured in water, until tested, in wet-surface conditions [16]. BS12:1978 and Libyan Standards 97/340
requirements of minimum strengths after 3 days (average values for three cubes) are 23MPa and 21MPa, and
after 28 days, 41MPa and 39MPa, respectively.
3.3.3 Concrete tests

3.3.3.1 Absorption test

Absorption cannot be used as a measure of the quality of concrete, but most good concrete has an absorption
level well below 10% (Roβler & Odler, 1985). The water absorption of a concrete cube is determined by
measuring the weight loss of an oven-dried cube (A), when immersed in water for 90 days (surface water
removed) (B). Using the ratio of the increase in weight to the weight of the dry cube, the absorption of
concrete can be obtained through equation 11 as follows:

(11)

3.3.3.2 Compressive strength for concrete

In this study, a concrete mix was designed with a compressive strength equal to 30 MPa after 28 days;
calculations were done using the volumetric design equation. The concrete mix proportions are as shown in
Table 4.

Table 4. Concrete mix proportions by kg/m³

C W S Gravel 1" Gravel ¾"
346 211 446 803 535
Note: C= cement; W= water; and S= sand.
8
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

This study focuses on the influence of cement using local materials on the performance of cement in concrete,
especially for concrete mixes. All mixes were prepared with constant proportions of cement, water, coarse and
fine aggregate; only the cement type was different. Fifteen cubes (15×15×15 cm) were cast for each sample of
cement, all cubes were cast in layers, and each layer of concrete was compacted by vibration. The cubes were
cured in water at 19 to 20°C. These cubes were tested after 3, 7, 28, 56 and 90 days; each value of compressive
strength was taken from the average of three cubes.

4. Results and discussion

4.1Test results and discussion of cement

According to the Libyan standard 97/340 and BS 12/1996 the values for loss of ignition are similar in both
specifications. Figure 3 shows the loss of ignition for all the cement types. Asonoper cement (ARC) exhibits
values that are not within the specification limits, but there were other types of cement that had values below the
specification limits. For the two types of Egyptian cements, it can be noticed that the loss of ignition for
Alqumia cement (AC) is higher than that of Aen Sokhnah (ASC). If we compare the local cements from
Almerqeb (AMC) and Lebda (LC) factories, it can be noticed that LC shows a value slightly higher than the
AMC. These results confirmed the effect of the cement source on the loss of ignition. In comparing the blended
cements with the other types of non-blended cements, we notice the high values for loss of ignition for the
blended cements, which are similar to the value for Romanian cement (RC).
On the other hand, the Magnesia content for all types of cements was within specification limits, according to
BS 12/1996, as shown in Figure 4, with the exception of LC, which exhibited a value higher than the other
types. If we compare the Egyptian cements AC and ASC, we notice some similarity between them.

Fig. 3. Loss of ignition for studied cements.

9
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

Fig. 4. Magnesia content for studied cements.

The sulphate content for all the cement types is presented in Figure 5. The blended cement (LC+RC) shows a
higher value than the Libyan specification 97/340 limit, but this value is still within the maximum specification
limits of BS EN 196. Comparison between the Egyptian cements AC and ASC show a similar value for SO3
ratio. A difference in sulphate content for the local cements AMC and LC is noticed. These results show the
effect of cement source on sulphate content.
Figure 6 presents the Alumina Modules (AM), for which the results were within the Libyan standards and
BS12:1978 specifications. The Egyptian cement ASC exhibited a higher AM value than AC, which means that
ASC contains a high ratio of Al2O3. Comparing the different AM values for local cement shows that LC
exhibits a higher value than AMC. All these results show the effect of cement source on AM. Comparing
blended cements with other types of cement, we remark lower values of AM. These results show that blending
the cement significantly affects the AM. Figure 7 presents the C3S ratio, where the values ranged from 49% to
63%, which is within the specification limits. The highest value was obtained by AC; this is due to the high
ratio of CaO and the absence of homogeneity in the ingredients of C3S according to the specifications. LC has
a low C3S ratio due to its low CaO ratio. The ratio of C3S for AMC blended with the Egyptian cement
decreased. On the other hand, the ratio of C3S for LC blended with the same Egyptian cement increased
slightly. This indicates the effect of cement blending on the change of C3S ratio.

Fig. 5. Sulphate content of studied cements.

11
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

Fig. 6 Alumina Modulus of studied cements.

The percentages of C2S for the cement types considered in the present study are given in Figure 8. There were
some differences in the C2S ratio, which for AC shows a slightly lower value for C2S ratio according to the
specification limits, due to the high ratio of CaO. AR and ASC illustrated some similarity in C2S ratio. But LC
exhibited the highest value. Different results for LC are shown in figure 8, where the C2S ratio decreased for LC
blended cement. On the other hand, AC produced the same values as AMC in the blended cements; therefore,
there was no effect on the C2S ratio for this combination.
Figure 9 shows the percentage of the C3A. The values of C3A conform to the BS EN196-1 specifications; the
highest value of C3A observed was for the LC+AC blended cement, and the Tunisian ARC respectively. Figure
9 shows the percentage of the C3A. The values of C3A conform to the BS EN196-1 specifications; the highest
value of C3A observed was for the LC+AC blended cement, and the Tunisian ARC respectively.

Fig. 7. Percentage of tricalcium silicate of studied cement.

11
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

Fig. 8. Percentage of dicalcium silicate of studied cement.

With respect to the Egyptian cement, the results show a higher value of C3A ratio for ASC. As for local
cement, the higher value was obtained for LC, but there is a convergence of values for the Egyptian cements,
which can also be seen for the local cements. This indicates the effect of cement sources on C3A ratio.
Higher values in the ratio of C3A for the LC+AC blend and for ARC were due to a low Fe2O3 ratio,
according to the Bogue equation.
The lime saturation factor (LSF) for all the cement types was within the specification limits; the values
ranged between 0.90 and 0.97. The high value of LSF for AC is due to some extent to the high CaO ratio.
For the local cements, the similarity between LC and AMC is due to their source being from the same area.
Figure 10 shows the relationship between LSF and the studied cements according to their source.
Figure 11 shows the specific surface of the cements. It is noticed that AC and ARC exhibit a higher value of
specific surface area. Comparing the local cements, it can be noted that LC obtained a higher value for
specific surface.

Fig. 9. Percentage of tricalcium aluminate of studied cement.

12
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

Fig. 10. Lime saturation factor of studied cements.

This shows the effect of cement sources on the specific surface of cement. On the other hand, all the blended
cements show a similarity in specific surface. This indicates that blending local and imported cements affects the
specific surface of cement and improves the product. Generally, the specific surfaces for all types are within the
specification limits.
The soundness of cement types considered in this study is presented in Figure 12, and was generally within
specification limits. ASC and ARC exhibit higher values than the other types of cement because of the high ratio
of free lime [F. CaO]. When comparing the blended cements with other types of cement, it can be noticed that
the soundness values approach each other; this shows that blending of local and imported cement leads to
soundness changes, at the least. The initial set for all types of ordinary Portland cement in this research is
illustrated in Figure 13. All results were within the specification limits. Compared with the considered types of
cement, low initial set values were exhibited by AC, RC and ARC due to high C 3S ratio. AMC and LC had a
high initial set compared with the other types of studied cements, due to low fineness. It is also indicated that
blending cements affects the initial set. Accordingly, variation in initial setting time could be controlled by
blending local and imported cements in appropriate proportions to obtain the intended setting time

Fig. 11. Specific surface (Blaine) of studied cements.

13
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

Fig. 12. Soundness (Le Chatelier) of studied cements.

Figure 14 shows that the final set of all types of studied cement are similar. Comparing local cements AMC
and LC, we notice a great similarity in the final set. Also, we find the same similarity for the Egyptian
cements AC and ASC; this shows the effect of cement sources on the final set of cement. As for the blended
cement there is a similarity in the final set; this again shows the effect of blending local and imported
cements on the final set. Generally, the final set of all types of cement is within the specification limits.
The compressive strength development of the studied cement mortar is presented in Figure 15, which shows
that all the results were within the specification limits. Comparing the results for all types of ordinary Portland
cements, it can be noted that the compressive strength development of RC exhibits higher values from 7 days
up to 28 days, but after 28 days, the compressive strength of LC starts increasing and holds the highest value
compared with the other types of cement considered. The compressive strength development of LC and AMC
cements are nearly equal before the age of 12 days, after which the compressive strength development of LC
increases. The compressive strength development of AC gained the lowest values before 7 days, but the
strength gradually increases up to 28 days. The strength development of ARC and the blended cements are
similar, but they are low if compared with the other types of cement. Hence, the compressive strength
development of RC and LC exhibited better results than AC and ASC. The lowest values of compressive
strength were obtained by ARC and the blended cements. It can be concluded that blending local and imported
cement does not produce better results in terms of strength development.

Fig. 13. Initial set of studied cements

14
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

Fig. 14. Final set of studied cements.

Fig. 15. Compressive strength development of studied cements.

2.4 Test results and discussion of concrete

4.2.1 Compressive strength development of concrete

Figure 16 shows the relationship between the strength of concrete made from local material and the age of
concrete in days. It can be noted that the strength of concrete made from RC exhibits a higher value at 3 days up
to the age of 28 days; after that the strength reduces. For the strength of concrete made from local cement, AMC
and LC, the results were very close to the strength development of the other types of cement at an early age, but
after 7 days, the development of strength increased; at 28 days, the strength development for local cement
exceeds the development of concrete strength for RC and the strength development continues at a slow rate
until 90 days. Concrete made from LC exhibited the highest values in strength at 28 days compared with the
strength of the other types of cement. The development strength of AC was lower at 3 days, and gradually
increased up to 7 days, exceeding the strength development of ARC, but lower than the strength development of
LC, AMC and RC. From these results it can be seen that the effect of local material is better when using local
and Romanian cements. However, using local materials to produce concrete made from ARC and AC provides
satisfactory results.

15
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

4.2.2 Water absorption of concrete

Figure 17 shows the water absorption of concrete made with different types of cement. It can be noted that
water absorption of the concrete made from local cement is low compared with other types of cement. AC and
RC show similar results in absorption, but concrete made from ARC exhibited higher values. This indicates that
the permeability of the concrete made from local cement is low. This confirms the effect of cement sources on
the permeability of concrete.

Fig. 16. Compressive strength development of concrete cubes with (0.55) w/c
ratios made with different types of ordinary portland cement according to source.

Fig. 17. Absorption of concrete cubes with (0.55) w/c ratios made with different types of
ordinary Portland cement according to source.

5.Conclusions
Based on the data in this study, the following conclusions can be drawn.

1. Romanian, Lebda, Almerqeb and Aen Sokhnah cements, respectively show an improvement in their chemical
and physical properties; the results from Alsonoper and Alqumia were acceptable to some extent.
2. The raw material source has an effect on engineering properties of cement paste, and also has an effect on the
strength properties of concrete mixed with local material.
3. The blending of local and imported cement at a ratio of 50% by weight has an effect on the chemical and
physical properties; it improves the quality of concrete and gives better results compared to the imported
cements alone.
4. All properties of ordinary Portland cement were within the specifications, but there was a significant disparity
in properties within the limits allowed.
5. The difference in average compressive strength of concrete made from the studied cements was due to
differences in the sources of the cement; this was observed despite the use of the same source for local

16
5th National Civil Engineering Conference AWAM, kala lumbur '09, 27-29 October.

materials and the same proportions in the manufacture of concrete cubes, with the only difference being the
type of cement according to the source.
6. The volume of pores in concrete was estimated by measuring the water absorption; the results were
acceptable for all the mixes. The volume of pores varied with the progress of cement hydration and the results
show that the rate of hydration for local cement is higher than that of other types.

Acknowledgements
The authors would like to thank the Arab Company of Cement in Libya, the Factory of Al-Merqeb Cement, the
Body for Public works in Alkomes City in Libya, the Executive Body for Railway.

References

[1] A. Gali, The libyan standard specifications in the cement industry, National Conference of Building
Materials and Structural Engineering, Vol 1, Alkomes, Liby, 2004.
[2] M. Liska and A. Al-Tabbaa, Performance of magnesia cements in pressed masonry units with natural
aggregates: Production parameters optimisation, Construction and Building Materials, 22(2008),
pp1789–1797.
[3] M. J. Varas, A. M. Buergo, and R. Fort, Natural cement as the precursor of Portland cement:
Methodology for its identification, Cement and Concrete Research 35(2005), pp2055 – 2065.
[4] M. Sidney and F. J. Young, Concrete, Prentice-Hall Inc., New Jersey, 1981.
[5] A. Monshia and M. K. Asgaranib, Producing Portland cement from iron and steel slags and
limestone, Cement and Concrete Research, 29(1999), pp 1373–1377.
[6] A. Emanuelsona, S. Hansena, and E. A. Vigghb, Comparative study of ordinary and mineralised
Portland cement clinker from two different production units Part I: Composition and hydration of the
clinkers, Cement and Concrete Research, 33(2003), pp1613-1621.
[7] K. Svinning, A. Hǿskuldsson, and H. Justnes, Prediction of compressive strength up to 28 days from
microstructure of Portland cement, Cement & Concrete Composites 30(2008), pp138–151.
[8] H. LeChatelier, Experimental Researches on the Constitution of Hydraulic Mortars, English
translation by J. L. Mark, McGraw, Publishing Co., New York, 1905.
[9] A. M. Neville, Properties of Concrete, Fourth and Final Edition, Longman, 1995.
[10] A.V. Girao, I. G. Richardson, C. B. Porteneuve and R. M. Brydson, Composition, morphology and
nanostructure of C–S–H in white Portland cement pastes hydrated at 55 °C, Cement and Concrete
Research 37(2007), pp1571–1582.
[11] K. O. Kjellsen and H. Justnes, Revisiting the microstructure of hydrated tricalcium silicate––a
comparison to Portland cement, 26 (2004), pp 947-956.
[12] R. Ylmén, U. Jaglid, B. Steenari, and I. Panas, Early hydration and setting of Portland cement
monitored by IR, SEM and Vicat techniques, Cement and Concrete Research, 39(2009), pp433–439.
[13] Y. Chen and L. Odler, On the origin of Portland cement setting,Cement and Concrete Research,
22(1992), PP. 1130-1140.
[14] I. Soroka, Portland Cement Paste and Concrete, 1st edition, 1979.
[15] AM. Neville, Properties of concrete, Third Edition, Pitman Books Limited, London, 1982.
[16] P. Sandor, Concrete Materials, 2nd ed., 1992.
[17] A. Subasi, A. Serdar and Y. H. Binici, Prediction of early heat of hydration of plain and blended
cements using neuro-fuzzy modelling techniques, Expert Systems with Applications 36(2009),
pp4940–4950.
[18] S. Aytaç, A. H. Aytaç and K. Ramyar, Effects of fineness of cement on polynaphthalene sulfonate
based superplasticizer–cement interaction, Construction and Building Materials 23(2009), pp2402–
2408.

17

View publication stats