You are on page 1of 69

Characterisation of Reject Red Clay Bricks in Khartoum as a Pozzolan Material

Dissertation for a partial fulfilment of BSc in Civil Engineering Civil Engineering Department Faculty of Engineering Academic Year 2010-2011 Student Names: Ashraf Osman Rahama Khalid Adam Habeeb Magdi Mohammed Khatir Mohammad Alhassan Mohammed Osman AlKhalifa

Suprvisor: Dr Yousif Hummaida Ahmed

1

ABSTRACT

Due to the poor quality control in traditional Sudanese bricks industry, the percentage of product rejected for sale mounts up to about 25%. This is about 0.7 billion bricks that represent a potential reserve for producing pozzolans in the Sudan.

In this study, random samples of reject red bricks were collected from various traditional brick kilns at Elgerief-Umdoum and Shambat in the Khartoum estate. The reject red bricks were crushed and sieved on #200 sieve (75µm). The amount passed through the sieve was further pulverised using ball mill until more than ⅔ passed through #325 sieve (45µm). The powdered materials for reject bricks collected from Elgerief-Umdoum specimens were denoted as red brick powder (RBP)-G-U, while those from Shambat specimens were denoted as RBP-SH.

The properties of the RBPs, namely fineness, strength activity indices of blended binders containing 20%RBP and 80% Portland cement (OPC) (by mass) and water requirement using methods described in with ASTM C311-02. In addition chemical compositions the RBPs were measured using methods described in BS EN 196-2 and BS EN 196-21. The values of these properties were compared with specifications in ASTM C 618-05, which classifies pozzolans in categories of class F, Class C and Class N.

This study found that according to ASTM C618-05 specifications, RBPs in Khartoum estate were of Class F, producing more than 75% of the strength of control OPC mortar. Also, unlike OPC, that there was a significant increase in the strength

between 28days and 91 days (more than 20%) for blended binders containing up to 30% gain. The gain for OPC mix was about 10% between 28 and 91days.

The study also investigated blended binders containing 30%RBP and 70%OPC (by mass) and found that the strength dropped with increasing the RBP dosage from 20% to 30%.

2

‫‪ACKNOWLEDGEMENT‬‬
‫بسم هللا الرحمن الرحٌم‬

‫(قل إعملوا فسٌرى هللا عملكم ورسوله والمؤمنون)‬
‫العظيم صدق هللا‬ ‫إلهي اليطيب الليل إال بشكزك واليطيب النهار إلً بطاعتك .. والتطيب اللحظاث إال بذكزك .. وال تطيب اآلخزة‬ ‫إال بعفىك .. وال تطيب الجنت إال بزؤيتك هللا جل جالله‬ ‫إلً من بلغ الزسالت وأدي األمانت .. ونصح األمت .. إلً نبي الزحمت ونىر العالمين..‬ ‫سيدنا محمد صلً هللا عليه وسلم‬ ‫واما بعد...البد لنا ونحن نخطو خطواتنا األخيرة في الحياة الجامعية من وقفة نعود إلى أعوام قضيناها في رحاب الجامعة مع‬

‫وقبل أن نمضي تقدم أسمى آيات الشكر واالمتنان والتقدير والمحبة إلى الذين حملوا أقدس رسالة في الحياة...‬ ‫والمعرفة...‬ ‫العلم‬ ‫طريق‬ ‫لنا‬ ‫مهدوا‬ ‫إلى جميع أساتذتنا األفاضل.......‬ ‫الذين‬ ‫إلى‬

‫أساتذتنا الكرام الذين قدموا لنا الكثير باذلين بذلك جهودا كبيرة في بناء جيل الغد لتبعث األمة من جديد...‬

‫"كن عالما. فإن لم تستطع فكن متعلما ، فإن لم تستطع فأحب العلماء ،فإن لم تستطع فال تبغضهم"‬

‫ع:‬ ‫وأخص بالتقدير والشكر مشرف هذا المشرو‬ ‫بشراك‬ ‫،‬ ‫له‬ ‫نقول‬ ‫الذي‬

‫وسلم:‬ ‫"‬

‫كذلك نشكر كل من ساعد على إتمام هذا البحث وقدم لنا العون ومد لنا يد المساعدة وزودنا بالمعلومات الالزمة إلتمام هذا‬ ‫و‬

‫الخير‬

‫عليه‬

‫الناس‬

‫معلم‬

‫اهلل‬

‫على‬

‫صلى‬

‫ليصلون‬

‫الدكتور : ٌوسف حمٌدة أحمد‬ ‫اهلل‬ ‫رسول‬ ‫قول‬ ‫،‬ ‫السماء‬ ‫في‬

‫والطير‬

‫البحر‬

‫في‬

‫الحوت‬

‫"إن‬

‫البحث ونخص بالذكر:‬

‫الدكتور: مجدي زمراوي رئٌس شعبة الطرق والتربة بكلٌة الهندسة جامعة الخرطوم‬ ‫االستاذ:عبد الرحٌم محمد ٌوسف(مشرف على التجارب المعملٌة لهذا المشروع)‬ ‫االستاذ:علً ٌحٌى آدم( مشرف معمل الخرسانة بجامعة امدرمان االسالمٌة)‬ ‫اسرة معمل الهندسة الصحٌة و البٌئٌة بكلٌة الهندسة جامعة الخرطوم‬ ‫ولكل الذين كانوا عونا لنا في بحثنا هذا ونورا يضيء الظلمة التي كانت تقف أحيانا في طريقنا.‬ ‫إلى من زرعوا التفاؤل في دربنا وقدموا لنا المساعدات والتسهيالت واألفكار والمعلومات، ربما دون يشعروا بدورهم بذلك فلهم‬

‫منا كل الشكر،:‬

‫.‬

‫3‬

Title……………………………………………………………………………………1 Abstract………………………………………………………………………………..2 Acknowledgement……………………………………………………………….…….3 Table of content…………………………………………………………….………….4 1. Background ............................................................................................................. 5 1.1. Portland Cement............................................................................................. 5 Introduction .............................................................................................. 5 Hydration of Portland Cement: ................................................................ 6

1.1.1. 1.1.2. 1.2.

Pozzolans ....................................................................................................... 7 Introduction .............................................................................................. 7 Chemical Process for the Pozzolanic Reaction ........................................ 9 Types of Pozzolans .................................................................................. 9

1.2.1. 1.2.2. 1.2.3. 1.3. 2. 3.

Utilization of Pozzolans in Sudanese Construction Industry ....................... 20

Aims and Objectives ............................................................................................. 22 Experimental Program .......................................................................................... 23 3.1. Materials ...................................................................................................... 23 Reject Red Bricks: ................................................................................. 23 Graded Standard Sand............................................................................ 28 Ordinary Portland Cement (OPC) .......................................................... 28

3.1.1. 3.1.2. 3.1.3. 3.2.

Test Methods & Results ............................................................................... 29 Fineness.................................................................................................. 29 Water Requirement ................................................................................ 29 Strength Activity Index (SAI) with Portland Cement............................ 33 Chemical Composition........................................................................... 37

3.2.1. 3.2.2. 3.2.3. 3.2.4. 4. 5. 6. 7. 8.

Discussion ............................................................................................................. 39 Conclusions .......................................................................................................... 49 Recommendations for further studies ................................................................... 51 References ............................................................................................................ 52 Appendices ........................................................................................................... 54 8.1. Appendix 1: ASTM C311 02 Standard Test Methods for Sampling and Testing Fly Ash or Natural Pozzolans for Use in Portland-Cement Concrete ......... 55 8.2. Appendix 2: ASTM C 618 05 Standard Specification for Coal Fly Ash and Raw or Calcined Natural-pozzolan for Use as a Mineral Admixture in Concrete. .. 65 8.3. Appendix 3: Standard Sand adapted from Table 4 of BS 882:1992............ 69 4

1. BACKGROUND
1.1. Portland Cement

1.1.1.

Introduction

Portland Cement (PC) is the binder used to make concrete, mortars and grouts for all types of building and construction. Unlike gypsum plaster, Portland cement is a hydraulic cement; i.e., it sets and hardens when mixed with water. The year 1824 is considered the beginning of modern Portland cements when the name „„Portland‟‟ was used in a patent by Joseph Aspdin. This term was used to reflect the similarity of concrete made with patented cement to Portland limestone. By the 1860s the Portland-cement industry was flourishing in the UK and Europe. The first patent in the US was filed by David Saylor in 1870.

The Portland cement is manufactured by mixing and crushing raw materials which are mainly limestone and siliceous clays. After the crushing process, which can be wet or dry, the mixture is fed through a rotating kiln with variable temperature zones, where the decomposition the raw material and fusing to produce what is known as Portland cement clinker.

Portland cement clinker consists mainly of four oxides: CaO (lime), denoted by cement chemists as C, SiO2 (silica), denoted as S, Al2O3 (alumina), denoted as A and Fe2O3 (iron oxide) denoted as F. Moreover, H2O is denoted as H and SO3. The levels of the four clinker minerals can be estimated using Bogue‟s equation. The method assumes that all the F is combined as C4AF. The remaining A (i.e. after deducting that combined in C4AF) is combined as C3A. The C combined in the calculated levels of C3A and C4AF and any free lime are deducted from the total C and the level of S determines the proportions of C3S and C2S. The procedure is expressed mathematically (in mass %) as follows (Moir 2003): 5 represents

C3S = 4.071(C – free lime) – 7.600S – 6.718A – 1.430F C2S = 2.867S – 0.7544C3S C3A = 2.65A – 1.692F C4AF = 3.043F In general when calculating the compound composition of cements (rather than clinkers) the normal convention is to assume that all the SO3 present is combined with Ca+2 (i.e. is present as calcium sulphate). The total CaO content is reduced by the free lime level and by 0.7×SO3 (Moir 2003) There are different types of Portland cements including Ordinary Portland Cement (OPC), rapid hardening PC (RHPC), low heat PC, sulphate resisting cements (SRPC), pozzolan cement, etc..

1.1.2.

Hydration of Portland Cement:

The hydration of the PC is described briefly as the sum of the hydration process of its four main crystallized compounds. Both C3S and C2S react with water to produce an amorphous calcium silicate hydrate known as C–S–H gel. This gel is the main „cementing compound‟ that binds the sand and aggregate particles together in concrete. The process is simplified in the following equations: C3S + 4.3H ⇒ C1.7SH 3 + 1.3CH C2S + 3.3H ⇒ C1.7SH3 + 0.3CH C3S is much more reactive than C2S and under „standard‟ temperature conditions. At 20°C approximately half of the C3S in PC cement will hydrate by 3 days and 80% by 28 days. In contrast to C3S, C2S hydration proceeds to a noticeable extent after 14 days. Both C3S and C2S has a typical Ca to Si ratio around 1.7. This is noticeably lower than the 3:1 ratio in C3S. The excess Ca is precipitated as calcium hydroxide (CH) crystals. C2S hydration also produces some CH formation albeit in lower quantity.

6

The hydration reactions of C3A with water, in absence of gypsum is very violent and causes flash setting. When mixed with water, in absence of gypsum, C3A and (to a lower extent C4AF) reacts rapidly in a highly exothermic reaction to form the phases C2AH8 and C4AH19. These hydration phases subsequently convert to C2AH6. However, the rate of reaction of the aluminate phases is significantly reduced by addition of finely ground gypsum (CaSO4·2H2O) or hemihydrate (CaSO4·0.5H2O) prior to mixing with water. This reduction in rate of reaction is due to the formation of a protective layer of ettringite (denoted in cement chemistry notation as AFt) on the surface of the C3A crystals. The reaction can be summarized as C3A + dissolved calcium (Ca2+) + dissolved sulfate (SO4-2) + water ⇒ ettringite which is written in cement chemistry notation as:

1.2.
1.2.1.

Pozzolans
Introduction

The term “pozzolan” was derived from volcanic ash firstly used by the Romans from deposits close to the village of Pozzuoli, near nowadays Naples, Italy. Pozzolans are primarily vitreous siliceous materials which react with calcium hydroxide to form calcium silicates and other cementitious products formed depending on the chemical composition of the pozzolan.

A pozzolan is a siliceous or aluminosiliceous material, which is not crystallized, i.e. highly vitreous or glassy. By its own, it has no cementitious properties, however when mixed with CH or a lime rich medium such as OPC, it exhibits cementitious properties. These cementitious properties are due to the reaction of silicates with lime to form calcium silicate hydrates with a low C/S ratio that gain gradual strength usually after 7 days. Pozzolans are commonly used as cement replacement material.

The rate of pozzolanic reaction is slower than OPC hydration reaction; this is normally reflected as lower short-term strength of concrete made with pozzolans, compared to the pure OPC concrete. This difference between the strengths of blended cement concrete and OPC concrete is reduced in long-term. However, highly reactive 7

pozzolans, such as silica fume (SF) and high reactivity metakaolin can produce "high early strength" concrete that may match or exceed the strength of OPC concrete.

The rate of gaining strength depends upon the chemical composition of the pozzolan and its degree of fineness as well the percentage of replacing OPC. The greater the percentage of vitrious alumina and silica, the better the pozzolanic reaction and the higher the strength development.

The replacement of Portland cement by pozzolans imparts many beneficial characteristics to the produced concrete. They increase the long-term strength and durability. Also they reduce the overall cost besides reducing the CO2 emission which will reduce the impact on global warming. Pozzolans have potential to make a significant contribution towards provision of low-cost building materials and consequently affordable houses for the poor masses.

It has been known for more than a century that several benefits can be derived by admixture of a suitable pozzolan with Portland cement, particularly for use in hydraulic structures. The optimum proportion of pozzolan is usually between 10 and 30 % by weight of the blended cement, however, ground granulated slag (GGBS) can be added to a much higher levels up to 85% (Calcrete 2008).

In general, the benefits of using pozzolans may be summarised as: 1. 2. 3. 4. 5. Reducing heat generation in massive structures concreting. Reducing cost of construction by saving in Portland cement constituent. Increasing concrete tensile strength. Reducing concrete permeability. Improving some properties of plastic concrete, such as improving workability and lowering the tendency to segregation and water gain. 6. 7. Increasing the resistance of concrete to sulphate attack. Reducing the risk of alkali aggregate reaction (AAR).

8

1.2.2.

Chemical Process for the Pozzolanic Reaction

Pozzolanic reaction can simply be classified as simple acid-base reaction between CH, also known as Portlandite, and silicic acid (H4SiO4, or Si(OH)4). as follows: Ca(OH)2 + H4SiO4 → Ca2+ + H2SiO42- + 2 H2O → CaH2SiO4 2 H2O or summarized as CH + SH → CSH

The ratio Ca/Si, or C/S of the CSH as well as the number of water molecules may differ depending on the stoichiometry and the chemical composition of the pozzolans. The density of CSH is lower than that of CH and pure silica, resulting in a swelling of the reaction products.

1.2.3.

Types of Pozzolans

Historically the two most abundant and widely used pozzolans are calcined clays (see section ‎ .2.3.1) and volcanic ash (1.2.3.2). Calcined clay in form of crushed fired 1 clay bricks, tiles or pottery has traditionally been used for improving properties of lime mortars and renders. The most commonly used pozzolans today are artificial such as fly ash (by product of burning coal in power-generating stations see section 1.2.3.3), ground granulated blast furnace slag (see section 1.2.3.4) and silica fume (see section 1.2.3.5). Other less common types are: high-reactive metakaolin and ashes of agricultural wastes including rice husk ash (RHA) & baggase ash which is a bye product of sugar industry.

Many pozzolans available for use in construction today were previously viewed as waste products. The use of pozzolans reduces the demand for Portland cement in producing concrete. This is more environmentally friendly and current practice may permit up to a 40% reduction of Portland cement. Also an optimisation of the mix design may permit utilization of pozzolans without significantly reducing the final compressive strength or other durability performance criteria.

9

1.2.3.1.

Calcined Clays and Shales

Clays or shales suitable for use as pozzolan are widespread and are readily available worldwide. They have been utilized as cement replacement materials on large scale construction programmes in the US, Brazil, Egypt and India. For example, in Egypt, a lime-calcined clay mortar was used in the core of the first Aswan dam built in 1902. Also OPC-calcined clay mixture was used in the construction of the Sennar dam in Sudan in 1919 to 1925 (Salih et al. 1999, Practical Action Technical Brief 2011). The blended calcined clay & lime mixture is known as surkhi in India, homra in Egypt and semen merah in Indonesia. Shales are harder than clays and have similar mineral contents resulting in similar pozzolanic properties.

However, this large-scale utilization has declined in the last three decades, due to the availability of pozzolan such as ground granulated blastfurnace slag (GGBFS) and pulverized fuel ash (PFA). In areas these where artificial pozzolans are not available, for example in Sudan, the use of calcined clay still has considerable potential.

Sandy-clays are often used as a pozzolan, frequently in the form of crushed fired clay bricks. It is worth mentioning that the coarser sand is not reactive. The pozzolanic activity is imparted by the finer clay mineral fraction, and sandy clays may not produce the best pozzolan.

Despite their variable pozzolanic reactivity, the use of ground underfired or reject bricks and tiles as a pozzolan is likely to continue on a small scale due to the low cost of these waste materials. Plastic clays, used in tile manufacture or pottery, may produce better pozzolan, although the composition of good pozzolanic clays is variable. Table 1 shows chemical composition (on an oven-dry basis) of some raw clays in India that are suitable for producing pozzolans conforming to the Indian Standard for calcined clay pozzolan (IS 1344.1981).

Fineness similar or slightly greater than that of OPC is usually recommended for pozzolans although some have been ground considerably finer. The minimum

fineness recommended by the Indian standards for pozzolan (IS 1344. 1981. Calcined

10

clays) is 320 and 250m2/kg for grade 1 and 2 pozzolans respectively, measured by the Blaine air permeability test (Practical action 2011).

Once the pozzolan has been ground, it must be blended with lime and/or OPC to produce a pozzolanic cement. The grinding can be accomplished by human or The

animal-powered methods but full homogeneity is unlikely to be achieved.

strength and consistency of cements blended in this manner will vary considerably. Mechanical techniques, preferably intergrading in a ball mill or, as a second option, dry blending in a pan or concrete mixer, may give better results in terms of both strength and consistency.

Pozzolans can be used with either lime and/or OPC for non-structural purposes replacement of up to 50% can be used. A larger percentage of OPC may be required if only poor quality pozzolan as are available. The exact ratio of the ingredients will depend upon the quality of the respective raw materials and on the required characteristics of the concrete or mortar made from the cement.

With lime pozzolan cements mixtures of 1:1 to 1:4 (lime : pozzolan) by weight are used. A good calcined clay pozzolan may produce a cement, when mixed with a good quality lime to produce concretes and mortars with 7 and 28 days strengths exceeding 2 and 4 MPa respectively.

Pozzolan : lime mixture strength development is slow and long-term strengths should be considerably higher, as much as 15Mpa by 2 years. The addition of 5-10% of OPC to lime : pozzolans mixtures will improve strength and decrease setting times. Strength development can be accelerated if up to 4 per cent of fine gypsum is added to the lime-pozzolan mix (Practical Action Technical Brief).

11

Table 1: Indian Standard IS 1344 1981 specification for chemical composition for clays (on oven dry basis) that are suitable for producing pozzolan upon calcination. Constituent Silica + Alumina + Iron oxide Silica Calcium Oxide Magnesium Oxide Sulpher Trioxide Water Soluble Alkalis Water soluble materials Loss On Ignition (LOI) Content by weight (%) ≥ 70.0 ≥ 49.0 ≤10.0 ≤3.0 ≤3.0 ≤0.1 ≤1.0 ≤10.0

IS 1344.1981 states that raw materials for calcined clay pozzolan should be free from coarse sand or gravel larger than about 0.6mm in diameter.

In tropical climates, such as in Sudan, clay deposits are often subjected to a form of chemical weathering that leaches out the silica and alkalis. This leaves a soil rich in ferric (e.g. lateritic i.e. iron bearing) and aluminium hydroxides (e.g. bauxitic i.e. aluminium bearing). However, despite their low silica content, both soils may exhibit some pozzolanic reactivity after calcination. The reactivity of laterite is lower than bauxite. The latter exhibits reasonable results and can be used as pozzolan in lack of silica-bearing clays.

Several researchers reported that the utilization of crushed clay bricks in construction industry. The use of rejects from these industries have been researched as

construction materials in Argentina, Spain and Thailand (Araceli et al. 2009, Sanchez de Rojas et al. 2006 and Torkittikul & Chaipanich 2010). For example, Monteiro et al. 2003 (cited by Ummi Kalsum et al. 2008) investigated the effect of ground clay brick to replace cement mass and found it effective in suppressing the alkali-silica reaction (ASR) expansion in mortar.

Other use for crushed bricks and tiles in construction industry have been investigated by several researchers, e.g. Poon et al. 2006 cited by Ummi Kalsum et al. 2008

12

investigated the use of crushed clay bricks as unbound road sub-base. Also, the recycled clay bricks were also used as aggregates in asphalt concrete (Khalaf et al., 2005 cited by Ummi Kalsum et al. 2008). Gutovic et al. 2005 (cited by Ummi Kalsum et al. 2008) later reported the strength development of autoclaved OPC-clay brick blends, where different varieties of claybrick were used. The optimum

compressive strengths were achieved at 50 mass % clay brick additions. Poon et al. 2006 (cited by Ummi Kalsum et al. 2008) presented a study on the investigation of blending recycled concrete aggregate and crushed clay brick as aggregates on the production of paving blocks. The results indicated that the incorporation of crushed clay brick reduced the density, compressive strength and tensile strength of the paving blocks. Recently, Kenai et al. 2007 (cited by Ummi Kalsum et al. 2008) investigated the use of crushed brick as coarse and fine aggregate for a new concrete. Results showed that it is possible to manufacture good concrete containing crushed bricks provided that the percentage of recycled aggregates is limited to 25% and 50% for the coarse and fine aggregates, respectively.

In Sudan, Hamid (2002) researched the use of several local calcined clays and reject red bricks (grog). The calcined clays sources were from the Blue Nile banks, kaolin and black cotton soil. He studied the mixtures hydrated lime (CH) from three

different resources namely Rabak, El-fao and Atbara. He tested various levels of replacing OPC by grog namely 10%, 20% and 30%. He found that for replacing OPC by 20% grog in a mortar, the strength was 19.59, 25.79, 38.78 & 39.78 N/mm2 for 3, 7, 28 and 60 days respectively. For 1:2:4 concrete containing 20% proportion of grog, the strength was 13.30, 23.23, 36.70 & 43.90 N/mm2 for 3, 7, 28 and 60 days respectively.

The calcination process of calcined clay pozzolans is similar to the moulding and firing process of clay bricks, tiles or pottery. The optimum calcination temperatures for clay pozzolans are slightly below those for clay bricks or tiles. It is much better to design specific moulding and firing process for producing pozzolans. The Indian Standard (IS1344 1981) gives the following range of temperatures for different types of clays: Montmorillonite type 600° to 800°C, Kaolinite type700° to 800°C and Illite type 900 to 1000°C. In practice, most clay soils consist of a mixture of minerals, and a calcination temperature of 700-800°C is normally considered suitable. The 13

optimum period of calcination depends on the clay type but is normally about one hour or less.

Rotary kilns have been the most common means of calcining clays and have been extensively used in the US and Brazil. Natural gas or oil is normally used as a fuel and outputs vary from 12.5 to 100 tonnes per day. Several Indian institutions,

including the Central Road Research Institute and the National Building Organization, have researched designs on kilns specifically for clay pozzolan production with production rates between 5 and 20 tonnes per day.

Calcined clays are ground to produce fine powder. This grinding can be performed with human or, more commonly, animal-powered methods. Ball mills are more suited to large-scale applications. Some calcined clays, such as kaolin, are softer than others and require less grinding in order to achieve the desired fineness.

The majority of brick making industries in the Sudan are operating along the banks of the rivers descending from the Ethiopian Rift. They operate during the dry season between October and July of the year. Fired clay bricks are mainly produced by conventional methods using biomass fuels such as fuelwood and dung cake (zibalaa). The modern brick industries produce less than 2% of the total annual brick production where fuel oil is used.

In 2006, the total annual production of fired clay bricks in the Sudan was estimated to be about 2.8 billion bricks (Alam 2006). About 88.5% of this amount was produced in Khartoum and Central States (Gezira, Sinnar, White Nile and Blue Nile). The brick industries located on the Blue Nile banks produced about 82.4% and those on the River Nile bank produced about 9.5%.

Alam (2006) estimated the total annual consumption of wood by the brick making industries of the Northern Sates of Sudan to be about 549,000 stacked m3 equivalent to 183,000 tonnes of fuelwood. From this total, the brick making industries of Khartoum and Central States consumed 46.2% and 42.3% respectively.
3 3

Wood

consumption per 1000 bricks varied widely from 0.117 m to 1.6 m . wood consumption per 1000 bricks in Kassala and Kordofan States is approximately more 14

than 10 times those consumed in Khartoum and Central States. The total number of workers employed in this sector was about 35,000; of which 50% were employed in Khartoum and 38% in Central States.

Due to the poor quality control in traditional Sudanese bricks industry, the percentage of product rejected for sale mounts up to about 25%. This is about 0.7 billion bricks that represent a potential reserve for producing pozzolans in the Sudan. These rejects are usually broken to smaller sizes up to 50mm aggregate size and mixed with quick lime & OPC and used as a lightweight waterproof layer on building roofs. The mixture is known locally as Khafja.

1.2.3.2.

Volcanic Ash

Deposits of volcanic ash exist wherever there are active or recently active volcanos, e.g in the Mediterranean, the Pacific region, and central and eastern Africa. The form of volcanic ashes may range from loose fine material to coarse deposits containing quite large particles. Deposits may be loose, with an appearance and texture similar to a compacted coal or wood ash. Other deposits are cemented, sometimes with appearance and properties similar to stone, and in this form they are normally referred to as tuffs or trassy. The colour of deposits can vary from off-white to dark grey.

Table 2: Composition of volcanic ash suitable for use as a pozzolan Constituent Alumina + Iron oxide Silica Content by weight (%) 15-30 45-65

Calcium Oxide + Magnesium Oxide + Up to 15 Alkalis Loss On Ignition (LOI) Up to 12%

The pozzolanic reactivity of ash deposits can vary considerably. The quality of material may also vary within a single deposit or a single geologically consistent stratum, with variations in depth being common. Regular testing is therefore required if volcanic ash is to be used as a pozzolan and this has been a constraint to its 15

commercial exploitation. However, volcanic ash is, or has recently been, successfully used as a pozzolan in many countries including the US, Germany, Japan, Italy, Kenya and Indonesia, with pilot plants tested in Tanzania and Rwanda. For example, 200,000 tonnes of volcanic pozzolan were used in the construction of the Glen Canyon dam in the US, completed in 1964.

In Sudan, resources of natural pozzolans material in form of volcanic tuffs and pumice are reported in Bayouda deserrt, Jebbel Marra and other volcanic fields. Sulieman (2008) reported several quarry sites of Bayouda Desert natural pozzolans, hereinafter denoted as BDNP, located in the Northern State, some 63Km South of Merowe city. The area of the pozzolan raw material occurs in ebel Hebeish and ebel azrub and bounded by 26 28 . atitudes 18 29 01 . 18 29 26 and ongitudes 32º 32 28 50 .1 E. The amount of reserve are estimated to be 10 million

tonnes at the former and 33 million tonnes at the latter. These reserves are accessible from Khartoum by Shrian Al Shimal paved road to Merowe and 63Km by unpaved road or from Atbara to Merowe by paved road across Bayouda Desert and 46km towards the north by unpaved road. The area is permanently accessible by trucks during all season of the year. Other potential pockets of NP exists in Elgeili

(Sabaloga), Jabel , Jabbel Katool in Kurdofan and Jabbel Marra in Darfur (Sulieman 2008).

Once the deposits of natural pozzolans are excavated most volcanic ashes will require only minor processing before use as a pozzolan. Many ashes are only loosely

cemented and can easily be excavated by hand, although others may need mechanical or pneumatic equipment. Some lithic tuffs may require blasting with explosives. The ash may require drying, and in dry sunny climates this can simply be achieved by spreading the ash in a thin layer on a specially prepared drying floor, similar to those commonly used to dry crops. Alternatively, in wet climates, and for large quantities, inclined rotary driers are normally used (Practical Action 2011).

If the ash is cemented it will need to be crushed before entering the dryer. Some volcanic ashes may be in form of very fine and loose powders that may not require crushing or grinding. Other ashes may be of sufficient fineness but be cemented together. These will require milling or crushing. Coarse ashes and lithic tuffs will 16

need to be ground in a ball mill or similar pulverising apparatus (Practical Action 2011). 1.2.3.3. Pulverised Fly Ash (PFA)

Fly ash from a coal-fired power station is a pozzolan that results in low-permeability concrete, which is more durable capable of resisting ingress of deleterious chemicals. When coal burns in a power station furnace between 1250°C and 1600°C, the incombustible materials combine to form spherical glassy droplets of silica (S), alumina (A), iron oxide (F) and other minor constituents.

The first reference to the idea of utilizing coal fly ash in concrete was by McMillan and Powers in 1934 and in subsequently researched by others research. These

researches were culminated by the construction of the Lednock, Clatworthy and Lubreoch Dams during the 1950s with fly ash as a partial cementitious material. These structures are still in excellent condition, after some 50 years (Lewis et al. 2003).

In Sudan, recently in 2005, the power generation at Gerri Power Stations started to produce fly ash. Unfortunately due to the use of petroleum coke as fuel, the generated fly ash is not suitable for using in construction. This is attributed to its very high carbon content. Further research and optimisation of the burning process and the fuel type are required in this field.

1.2.3.4.

Ground Granulated Blastfurnace Slag (GGBS)

Blastfurnace slag is a by-product of manufacturing iron in a blastfurnace. It results from the fusion of a limestone flux with ash from coke. The slag is composed of the siliceous and aluminous residue remaining after the reduction and separation of the iron from the ore.

Two alternative processes, namely granulation or pelletization can be utilized to convert molten slag into a form suitable for use as a cementitious material. Either process requires that the molten slag to be rapidly cooled to form a glassy disordered 17

structure. If the slag is allowed to cool too slowly this allows a crystalline wellordered structure to form which is stable and non-reactive.

In the granulation process the stream of molten slag is forced over a weir into highpressure water jets of below about 50°C. The water slag ratio is normally between ten and twenty to one. This process produces glassy granules of cooled slag no larger than about 5 mm in temperature diameter. Following the granulation the granulate is dried and ground to cement fineness in a conventional cement clinker grinding mill.

In the pelletization process, the molten slag is poured onto a water-cooled steelrotating drum of approximately 1m diameter. The drum has fins projecting from it, which the slag is thrown through the air inside a building where water is sprayed onto it. This results in rapid cooling and produces pelletized particles from about 100 mm down to dust. However, large particles tend to be crystalline in nature and have little or no cementitious value. Particles larger than about 6 mm are therefore screened off and used as a lightweight aggregate in concrete and only the finer fraction (< 6 mm) used for the manufacture of GGBS.

Granulation is produces materials with a high glass content, but the capital costs of a granulator are about six times greater than that of a pelletizer. Both materials can be used as raw feed for GGBSs and most Standards do not differentiate between them.

The hydraulic potential of ground granulated blastfurnace slag (GGBS) was first discovered in 1862 in Germany by Emil Langen. Commercial production of limeactivated GGBS was started in 1865 and around 1880 GGBS was first used in combination with Portland cement (PC). It has been used extensively in many

European countries, such as Holland, France and Germany. In the UK the first British Standard for Portland Blastfurnace Cement (PBFC) was produced in 1923 (Lewis et al. 2003)

The chemical composition of slag depends on the source of the raw materials and the blastfurnace conditions. The major oxides exist within the slag glass (formed as the result of rapid cooling) in the form of a network of calcium, silicon, aluminium and

18

magnesium ions in disordered combination with oxygen. The oxide composition of GGBS is shown in Table 3 compared to the Portland cement.

Table 3: Comparison between oxide composition of European Portland Cements and GGBS (after Lewis et al. 2003)

1.2.3.5.

Silica Fume (SF)

Various names including silica fume, microsilica and volatilized silica are often used to describe by-products extracted from the exhaust gases of silicon, ferrosilicon and other metal alloy smelting furnaces. The terms microsilica and silica fume (SF) are used to describe those condensed silica fumes that are of high quality, for use in the cement and concrete industry. The latter term, SF is adopted in the European Standard – prEN 13263-1 (Lewis et al. 2003).

SF was first produced in Norway, in 1947, when environmental policy made the filtering of the exhaust gases from the furnaces compulsory. The major portion of these fumes was a very fine powder containing high percentage of silicon dioxide. Large-scale filtering of gases began in the 1970s and the first standard, NS 3050, for use in a factory-produced cement, was granted in 1976. As the pozzolanic reactivity for SiO2 was well known, extensive research was carried out at the Norwegian Institute of Technology. There is extensive research 3000 papers now available that detail work on silica fume and silica fume concrete (Lewis et al. 2003).

Due to the high S content of SF, the dosage of SF that would consume all the CH produced by PC hydration is not exceeding 10% (Lewis et al. 2003) 19

1.3.

Utilization Industry

of

Pozzolans

in

Sudanese

Construction

A workshop held in 1999 on pozzolan utilization in the construction industry in Sudan, jointly organized by the Sudanese Organization of Economic Alternatives, the Building & Road Research Institute of University of Khartoum and Sudan Engineering Society estimated that pozzolan alone, as an alternative for OPC have a capital business share of around 6,000,000USD per annum (Salih et al. 1999).

Calcined clays were used in 1920s to construct Sennar dam. Khafja , a local term for a mixture of wasted bricks aggregate plus lime and/or cement is normally used for roofing in traditional Sudanese building industry. Chinese PFA was imported for constructing Merewe dam, despite the presence of Bayouda Dessert Natural Pozzolans (BDNP) vicinity of 50Km. Apart from these cases, there are neither sound engineering practice nor a drive and will to utilize pozzolans in domestic construction industry. Also due to the existence of large sugar industry in Sudan, there is a potential of using baggase ashes as pozzolanic material in Sudan (Salih et al. 1999)

The lack of using pozzolans as cement replacement material may be attributed to scarce information on engineering properties of pozzolans blended cements. This trend is common almost in all African countries, apart from limited experiments in Egypt, Rwanda, Tanzania and Kenya (Salih et al. 1999). Also, it may be due to the industry‟s ignorance of benefits imparted by blended pozzolans cement. However, Mamoun & Zamzami (2010) recently published a paper characterizing samples of BDNP using Indian standard. Their study was carried out from a geological

perspective and involved chemical analysis, x-ray diffraction and compressive strength of lime and BDNP mixes. The need for cement in the Sudanese construction industry has increased significantly in the last few years. This increase is due to the initiation of a vast public work construction programme. Thus large quantities of cement have been imported to cover the specific needs of the construction sector. Despite the recent openings of 20

several new cement factories, the cost of Sudanese cement per tonne is still far higher than its average international cost. International average cost per tonne is 70USD while producing Sudanese cement costs about 100USD and about 130USD to the end consumer in Khartoum (Assalam Cement Factory communication 2010). Management, Personal

Moreover, one of the challenges facing the Sudanese cement and concrete industry is production of durable concrete especially in coastal areas and economical concrete nationwide. The utilization of Bayouda desert volcanic ash and other similar

resources may be the key for addressing these challenges. Bearing in mind the vicinity of Bayouda dessert to the Port-Sudan and the accessibility to Khartoum, there is a potential market for using this material in coastal area for resisting chloride and sulphate attacks.

21

2. AIMS AND OBJECTIVES

As mentioned in Section 1.2.3.1, it is estimated that about 0.7 billion bricks are rejected annually due to the poor quality control in traditional Sudanese bricks industry. These reject bricks represent a potential reserve for producing pozzolans in the Sudan.

This aim of this study is investigating the potential of using ground reject fired clay bricks in Khartoum as pozzolan for replacing cement. The objectives of this research are: 1. Collection of samples of reject bricks from different traditional red brick industries located at Elgerief-Umdoam and Shambat in Khartoum, Sudan. 2. Crushing and pulverising samples collected in Paragraph 1 to produce fine Red Brick Powder (RBP). 3. Measuring the properties of RBP, as pozzolanic material, according to the scope of Standard Test Methods for Sampling and Testing Fly Ash or Natural Pozzolans for Use in Portland-Cement Concrete ASTM C311-02 (See Appendix 1). Namely, Chemical composition, water requirement, fineness and strength activity index with OPC will be assessed. 4. Interpreting the test results determined in Paragraph 3 using Specifications in ASTM C 618-05 (see Appendix 2) to characterise RBP as pozzolanic material. The ASTM C 618-05 provides specification for Coal Fly Ash and Raw or Calcined Natural-pozzolan for Use as a Mineral Admixture in Concrete.

22

3. EXPERIMENTAL PROGRAM
3.1.
3.1.1.

Materials
Reject Red Bricks:

Random samples of reject red bircks (weighing about 50Kg) were collected from various Brick kilns located on the east bank of the Blue Nile river, namely El-gerief east & Umdoum. These samples, hereinafter, bear the suffix G-U (short for El-gerief east & Umdoum). Also other 50Kg samples were collected randomly from various bricks kilns located on the east bank of the river Nile, namely in Shambat in Khartoum North. These samples bear the suffix SH (short for Shambat). Photographs 1 to 8 document for trips to collect reject brick samples.

The reject red bricks were firstly crushed using a crusher at Building and Road Research Institute (BRRI), University of Khartoum. Then they were sieved on # 200 sieve (75µm). The amount passed through the sieve was further pulverised using ball mill crusher at the Corporation of Geological Researches Laboratory, Ministry of Mining located at El-neel Avenue Khartoum. The powdered materials for the Elgerief-Umdoum specimens are denoted as RBP-G-U, while those from Shambat specimens are denoted as RBP-SH.

Clause 5.2 of ASTM C 311-05 (see Appendix 1) recommends that the hydrated lime used in the tests shall be reagent-grade calcium hydroxide, 95 % minimum calculated as Ca(OH)2 and have a minimum fineness of 2500 m2/kg as determined in accordance with Test Method C 204. However, in this study, quick lime was used, purchased form Al-gamieer on the west bank of the River Nile at Umdurman. The quick lime was dissolved in water at 23oC to produce hydrated lime. This reaction is exothermic and was carried out using protective eye and hand gear.

23

Photograph 1: A platform for moulding green bricks (front) and a yard for sun drying of the green bricks. (back).

Photograph 2: Stacking of dried bricks for firing.

24

Photograph 3: Start of firing the bricks after covering the stacked bricks with broken bricks cemented with cow dung (Zibala).

Photograph 4: Cooling down of a fired brick kiln.

25

Photograph 5: A finished fired brick kiln at El-gerief East, Khartoum.

Photograph 6: Firing holes for feeding fuelwood at the bottom of a red brick kiln.

26

Photograph 7: Unloading the fired bricks from a kiln -1

Photograph 8: Unloading the fired brick from the kiln-2

27

3.1.2. The AST

Graded Standard Sand C311 requires that “The sand used for making test specimens for the

activity index with Portland cement shall be natural silica sand conforming to the requirements for graded standard sand in Specification ASTM C 778”. However, in this study, the standard sand was prepared according to BS 882:1992, the grading was conforming to the specifications of BS 882:1992 as described in Appendix 3 3.1.3. Ordinary Portland Cement (OPC)

Ordinary Portland Cement (OPC) of class 42.5N produced by Atbara Cement (http://www.atbaracement.sd). The chemical composition of the cement is listed in Table 4.

Table 4: Chemical Composition of OPC Oxide Al2O3 CaO CL content Fe2O3 Insoluble Residue (IR) Loss on Ignition (LOI) MgO SiO2 SO3 Alkaline (Na2O & K2O) Total % by Weight 7.17 62.5 0.004 3.28 2.12 0.88 1.10 20.16 2.75 00 99.964

28

3.2.
3.2.1.

Test Methods & Results
Fineness

RBP G-U and RBP-SH specimens were wet-sieved in accordance with ASTM 593-69 through sieve #325 (45µm) as required. The results are shown in Table 5

Table 5: Fineness Results for RBP-G-U and RBP-SH Specimen Ref RBP-G-U RBP-SH Weight (gm) 50 50 Weight retained on sieve #325 (45µm) 15.67 16.00 34.33 34.00 68.66 68.00 %retained %passed

3.2.2.

Water Requirement

The ASTM C 1437, the Standard Test Method for Flow of Hydraulic-Cement Mortar, determines how much a mortar sample flows when it is unconfined and consolidated. The mortar flow is similar to a slump test is a relative measure of workability. Changes in flow indicate variability in the materials and/or the batching process that may not be observed from slump testing alone. ortar flow is most sensitive to water

content and air content. It is also more sensitive than the slump test for stiff concrete mixtures. ortar flow is a process control test procedure and should not be The test is sensitive to several parameters

considered as an acceptance criteria.

including water content fine aggregate gradation cementitious chemistry mixing time air content and concrete temperature all interact to affect mortar flow.

The apparatus for the flow test is shown in Photograph 9. The apparatus consists of flow table, flow mould, clipper, tamper, trowel, and ruler. The apparatus is Model ZI 1007 made by ZEAL International:1, Netaji Subhash Marg, Daryaganj, New Delhi, INDIA. The flow table consists of a 30 cm diameter polish steel plate with 3

engraved annular circles 7, 11 and 19cm diameter. The table top is arranged for a free fall of 12.5mm by a cam action. A brass conical mould with the following

dimensions: 65mm internal diameter. at base and 40mm internal diameter at top,

29

height of 90mm. The test was carried out at the Civil Engineering Laboratory of Umdurman Islamic University.

Photograph 9: Flow Table apparatus for testing mortar flow

Photograph 10: Transformation of the conical mould of mortar to a pancake shape after manual vibration.

30

The mortar specimen was placed inside 90mm tall conical brass mould in two 45mm. layers, each layer was tamped 20 times. The mortar on the top of the mould was stricken off and flushed with the top of the mould. The mould was removed and the mortar was vibrated manually as the flow table rises and drops 12.5mm., 15 times in 15 seconds. The mortar changed from a conical shape with a 65mmn. base to a “pancake.”. The diameter of the pancake was measured (see Photograph 10). The ortar flow was reported as a percentage of the original diameter based on the change in diameter from 4 in. to the final diameter of the mortar “pancake.”.

First, the flow for a standard mix containing 500gm OPC, 242mL water (Water/Cement (W/C) ratio=0.48) and 1:3 cement: standard sand was calculated. Then water was added (on a trial basis starting from 255gm) to a 1:3 binder : standard sand mix. The binder is made of 400gm OPC and 100gm pozzolan (i.e. 20% RBP replacing OPC). The flow was measured targeting a flow level equals to that of the control OPC mix flow ± 5.

The data and the results for the mortar flow test are shown in Table 6.

31

Table 6: Data and Results of Mortar Flow Test. Mix Constituents (gm) Mix Ref. Sand (gm) OPC-Control RBP-G-U 20% RBP-G-U 20% RBP-G-U 20% RBP-Sh 20% 1375.00 Cement (gm) 500.00 RBP (gm) 0.00 100.00 100.00 100.00 100.00 Water (gm) 242.00 255.00 252.00 250.00 250.00 0.484 0.510 0.504 0.500 0.500 Water /Binder ratio (W/B) Original (cm) 6.50 6.50 6.50 6.50 6.50 Spread (cm) 13.00 14.50 13.50 13.00 13.00 (%) 100.0 123.1 107.7 100.0 100.0 Diameter Flow

1375.00 400.00 1375.00 400.00 1375.00 400.00 1375.00 400.00

32

The water requirement, which was later used for the Strength Activity Index with Portland Cement in Section 3.2.3 was calculated as a percentage of control as follows:

where: Y = water required for the test mixture to be ±5 of control.

As deducted from data in Table 6, the Y value for both RBP-G-U and RBP-SH was found to be 250gm resulting in water requirement = 103%

3.2.3.

Strength Activity Index (SAI) with Portland Cement

The test for strength activity index as described in section 4.1.1 of ASTM C311-05 was carried out to determine whether RBP produced an acceptable level of strength development when used with hydraulic cement in concrete.

The mix references, curing regime and proportions and shown in Table 7. Nine 70× 70× 70 mm mortar cubes of each mix were cast. The water binder ratios W/B were 0.484 for OPC control mix and 0.50 for OPC + RBP mixes as calculated from the Water Requirement Section 3.2.2. Clause 26.2 of the ASTM C311-05 requires that SAIs to be evaluated on 7 or 28 days on 20%RBP:80%OPC mortars cured in saturated lime. However, in this study extra SAIs for 20%RBP:80%OPC were

evaluated after 91days, as well as for water cured specimens. In addition extra mixes, containing 30%RBP:70%OPC binder were also cast and cured in water to investigate the effect of higher RBP dosage on SAIs.

It is worth noting that for charactering RBP only 6 cubes were required. Since ASTM C 618 specifies that “meeting the 7 day or 28 day Strength Activity Index will indicate specification compliance” only one age might be required. It is optional for the producer or the user after preparing six-cube batches to choose only three cubes of control and test mixtures need to be demoulded for either 7 or 28 day testing.

33

The apparatus for measuring SAIs was 3 sets of triplicate 70mmx70mmx70mm cubes joint together, mechanical concrete mixer, sensitive platform balance, blade, scoop, ruler, mechanical vibrator, graduated cylinder, cube crushing machine, water curing tank and curing cabinet. The curing cabinet model was Control 65-L0013/D CURACEM Cement Curing Cabinet supplied by CONTROLS S.R.L Via Aosta, 6 – 20063 Cernusco, Italy. The cabinet provides a curing environment between 95% and 100% relative humidity (RH). Its temperature was set to 26±1oC deviating from the ASTM C311 standard requirement of 23±2oC.

First the mix ingredients were weighed as shown in Table 7, then the dry ingredients were poured in the mixture and mixed for 1 minute. Then the water was added and mixed for 4 minutes. Then 3 sets of clean and oiled triplicate moulds were filled with one 35mm thick layer of mortar and vibrated by the mechanical vibrator for 1 minute. Another layer of 35mm thick was filled and vibrated for another 1 minute. The cast cubes were labelled and designated for triplicate testing on 7, 28 and 91 days.

The cast samples were sealed with cling film and placed immediately after vibration into the curing cabinet for 24 hours. After 24 hours, the specimens were demoulded and stored in the designated curing regime until testing dates at 7, 28 and 91 days. The strength activity indices (SAIs) with OPC were calculated using the following equation:

where: A B = = average compressive strength of test RBP mixture cubes (Mpa) and average compressive strength of OPC control mix cubes (Mpa)

The compressive strength of the cast mortars and their relevant strength activity indices (SAIs) are shown in Table 8.

34

Table 7: Mix References, proportions and curing conditions. Mix Ingredient (gm) Mix Ref. OPC-Control 2 Sand (gm) 1800.00 Cement (gm) 600.00 480.00 480.00 420.00 420.00 RBP (gm) 0.00 120.00 120.00 180.00 180.00 Water (gm) 288.00 300.00 300.00 300.00 300.00 Water /Binder ratio1 (W/B) 0.484 0.500 0.500 0.500 0.500 Curing solution (23oC±1) Water Water Water Water Water Number of 70x70x70 mm cubes 9 9 9 9 9

RBP-G-U 20%-W 1800.00 RBP-SH 20%-W 1800.00

RBP-G-U 30%-W 1800.00 RBP-SH 30%-W RBP-G-U 20%-L3 RBP-SH 20%-L4 1800.00

1800.00 1800.00

480.00 480.00

120.00 120.00

300.00 300.00

0.500 0.500

Saturated lime Saturated lime

9 9

1 2

For determining W/B ratios please see Section 3.2.2 for water requirement Control OPC Mix cured in water 3 El-gerif –Umdoum Red Brick Powder replacing 20% (by weight) of cement and cured in saturated lime for measuring ASTM 618 strength activity index with OPC 4 Shambat Red Brick Powder replacing 20% (by weight) of cement and cured in saturated lime for measuring ASTM 618 strength activity index with OPC.

35

Table 8: Strength Activity Indices for various RBP:OPC mixes Mix Ref. Average 7 Days Strength (N/mm2) 33.36 26.74 24.44 20.81 19.24 Strength Activity Average 28 Days Strength Activity Average 91 Days Index (7days) Strength (N/mm2) Index (28 days) Strength (N/mm2) 1.00 0.80 0.73 0.62 0.58 41.90 29.99 28.02 28.85 26.06 1.00 0.72 0.67 0.69 0.62 45.83 40.15 44.59 35.87 37.97 Strength Activity Index (91days) 1.00 0.88 0.97 0.78 0.83

OPC-Control RBP-G-U 20% (W) RBP-SH 20% (W) RBP-G-U 30% (W) RBP-SH 30% (W)

RBP-G-U 20% (L) RBP-SH 20% (L)

26.53 27.02

0.80 0.81

35.25 33.05

0.84 0.79

43.10 39.27

0.94 0.86

36

3.2.4.

Chemical Composition C311 states that “The chemical component determinations and

Clause 4.1.2 of AST

the limits placed on each do not predict the performance of a fly ash or natural pozzolan with hydraulic cement in concrete, but collectively help describe composition and uniformity of the material”. However, in this study the chemical compositions of RBP-G-U and RBP-SH were, determined using methods described in BS EN 196-2 and BS EN 196-21. The chemical compositions of RBP-G-U and RBPSH are shown in Tables 8 and 9 respectively.

Table 9: Chemical Composition of Red Brick Powder (RBP) from Elgerief East & Umdoum (RBP-G-U) Oxide Al2O3 CaO CL content Fe2O3 MgO SiO2 SO3 Na2O K2O Equivalent total alkali content to Na2O Insoluble Residue (IR) Loss on Ignition (LOI) Total % by Weight 9.40 3.30 Not determined (ND) 6.50 0.34 78.73 0.27 0.225 0.292 0.817 ND ND 99.057

37

Table 10: Chemical Composition of Red Brick Powder from Shambat (RBP-SH) Oxide Al2O3 CaO Fe2O3 MgO SiO2 SO3 Na2O K2O Equivalent Total alkali content to Na2O CL content Insoluble Residue (IR) Loss on Ignition (LOI) Total % by Weight 9.28 3.44 4.20 0.157 80.68 0.216 0.615 0.625 0.773 Not determined (ND) ND ND 99.213

38

4. DISCUSSION

The results in Table 8 were plotted in Figures 1 to 12. Figure 1 shows that Red bricks powders (RBP) produced from reject bricks collected from both El-gerief & Umdoum (G-U) and Shambat (SH) have strength activity indices (SAIs) with OPC of about 80% higher than 75%. This satisfies the ASTM C 618 criterion of SAI for 7 days. Figure 2 also shows that both RBPs satisfied the criterion for 28 days. However, RBP-G-U produced a higher SAI of 84% than RBP-SH which is 79%. Despite not required by ASTM C618, the SAIs for 91 days still showing that RBP-G-U performed better than RBP-SH.

7 days Strength (N/mm2) and Strength Activity Indices with OPC (expressed as Ratio)
40.00 35.00 30.00

Ratio=1
Ratio=0.80 Ratio=0.81

25.00
20.00 15.00 10.00

5.00 0.00 OPC-Control RBP-G-U 20% (L) RBP-SH 20% (L)

Figure 1: Influence of Red brick powder source on seven days Strength Activity Indices (SAIs) with OPC.

39

28 Days Strength (N/mm2) and Strength Actvity Indices with OPC (expressed as ratio)
45.00 40.00 35.00 30.00

Ratio=1
Ratio=0.84

Ratio=0.79

25.00
20.00 15.00 10.00 5.00

0.00
OPC-Control RBP-G-U 20% (L) RBP-SH 20% (L)

Figure 2: Influence of red brick powder source on 28 days strength activity indices (SAIs) with OPC.

91 Days Strength (N/mm2) and Strength Actvity Indices with OPC (expressed as ratio)
47.00

46.00 45.00
44.00 43.00 42.00 41.00 40.00

Ratio=1

Ratio=0.94

Ratio=0.89

39.00 38.00
OPC-Control RBP-G-U 20% (L) RBP-SH 20% (L)

Figure 3: Influence of Red brick powder source on 91 days strength activity indices (SAIs) with OPC. NB the ASTM C618 does not require for 91 days.

Figures 4, 5 and 6 summarise the influence of increasing the RBP dosage from 20% to 30% when cured in water. It is clearly that increasing of RBP dosage lowered the strength.

40

Effect of RBP Dosage on 7 days Strength
40.00

35.00
30.00 25.00

Ratio =1 Ratio = 0.8 Ratio =0.73

Ratio=0.62

20.00
15.00

Ratio=0.58

10.00
5.00 0.00 OPC-Control RBP-G-U 20% RBP-SH 20% RBP-G-U 30% RBP-SH 30% (W) (W) (W) (W)
Figure 4: Influence of dosage level of red brick powder (RBP) on 7 days strength of water cured mortars.

Effect of RBP Dosage on 28 days Strength
45.00 Ratio =1 Ratio = 0.72

40.00
35.00 30.00 Ratio =0.67

Ratio=0.68

Ratio=0.62

25.00
20.00 15.00 10.00 5.00

0.00 OPC-Control RBP-G-U 20% (W) RBP-SH 20% (W) RBP-G-U 30% (W) RBP-SH 30% (W)

Figure 5: Influence of dosage level of red brick powder (RBP) on 28 days strength of water cured mortars.

41

Effect of RBP Dosage on 91 days Strength
50.00 45.00 40.00 35.00 30.00 Ratio =1

Ratio = 0.88

Ratio =0.91
Ratio=0.78 Ratio=0.83

25.00 20.00 15.00 10.00 5.00 0.00
OPC-Control RBP-G-U 20% (W) RBP-SH 20% (W) RBP-G-U 30% (W) RBP-SH 30% (W)

Figure 6: Influence of dosage level of red brick powder (RBP) on 91 days strength of water cured mortars.

Effect of Curing Solution on 7 days Strength & Strength Activity Indices of 20RBP:80OPC Mixes
40.00

35.00
30.00 25.00

Ratio=1
Ratio=0.80 Ratio=0.80 Ratio=0.81

Ratio=0.73

20.00
15.00 10.00 5.00 0.00

OPC-Control

RBP-G-U 20% (W)

RBP-G-U 20% (L)

RBP-SH 20% (W)

RBP-SH 20% (L)

Figure 7: Influence of curing regime on 7 days strength of 20%RBP:80%OPC mortars.

42

Effect of Curing Solution on 28 days Strength & Strength Activity Indices of 20RBP:80OPC Mixes
45.00 40.00 35.00 30.00 25.00 20.00 15.00 10.00 5.00 0.00 OPC-Control RBP-G-U 20% (W) RBP-G-U 20% (L) RBP-SH 20% (W) RBP-SH 20% (L) Ratio=1 Ratio=0.84 Ratio=0.79 Ratio=0.67

Ratio=0.72

Figure 8: Influence of curing regime on 28 days strength of 20%RBP:80%OPC mortars.

Effect of Curing Solution on 91 days Strength & Strength Activity Indices of 20RBP:80OPC Mixes
47.00 46.00 45.00 44.00 43.00 42.00 41.00 40.00 39.00 38.00 37.00

Ratio=1
Ratio=0.94 Ratio=0.91 Ratio=0.88 Ratio=0.89

OPC-Control

RBP-G-U 20% RBP-G-U 20% (W) (L)

RBP-SH 20% RBP-SH 20% (L) (W)

Figure 9: Influence of curing regime on 91 days strength of 20%RBP:80%OPC mortars.

Figures 7, 8 and 9 show that curing in saturated lime, produced higher strength than curing in water. This is attributed by the fact that curing in water will leach out Ca+2, while curing in saturated lime may preserve or increase the Ca+2 concentration within the mortar. As mentioned in Section 1.2.2 that the presence of Ca+2 is essential for the 43

pozzolanic reaction.

Figure 10 shows that mix containing 20%RBP-G-U (cured in lime) gained 32.86% extra strength from 7 days to 28 days, however this gain dropped to 22.27% from 28days to 91 days. For mix containing 20%RBP-SH these gains were 22.32% and 18.88%. This shows that the rate of pozzolanic reaction for both RBPs was exceeding 22% and more significant between 7 and 28days.

Figure 11 shows that mix containing 20%RBP-G-U (cured in water) gained 12.15% extra strength from 7 days to 28 days, however this gain increased significantly to 33.88% from 28days to 91 days. For mix containing 20%RBP-SH these gains were 14.65% and 59.14% respectively. This shows that the rate of pozzolan reaction for both RBPs was more significant and exceeding 33% between 28 to 91days. Hamid (2002) found that the strength of 20%RBP from blue Nile (Suba):80%OPC increased by 50.36% from 7 days to 28 days.

Figure 12 shows that mix containing 30%RBP-G-U (cured in water) gained 38.64% extra strength between 7 and 28 days. However this gain reduced to 24.33% from 28days to 91 days. For mix containing 30%RBP-SH these gains were 35.54%

between 7 and 28 days and increased to 45.70% between 28days and 91days. This shows that the rate of pozzolan reaction for both RBPs was more than 24% in the duration between 28 to 91days.

44

50.00 45.00 40.00 35.00

Strength (N/mm2)

30.00 25.00 20.00 15.00 10.00 OPC-control RBP-G-U 20% (L)

RBP-SH 20% (L)

5.00
0.00 1

Age (Days) 10

Log Scale 100

Figure 10: Influence of age on strength gaining of 20%RBP:80%OPC mortars cured in saturated lime.
50.00 45.00 40.00

Strength (N/mm2)

35.00 30.00 25.00 OPC-control 20.00 RBP-G-U 20% (W) 15.00 RBP-SH 20% (W) 10.00 5.00 0.00 1

Age (Days) 10

Log Scale 100

Figure 11: Influence of age on strength gaining of 20%RBP:80%OPC mortars cured in water.

45

50.00 45.00 40.00 35.00

Strength (N/mm2)

30.00 25.00 20.00 OPC-control 15.00 10.00 5.00 0.00 1 10 100 RBP-G-U 30% (W) RBP-SH 30% (W)

Figure 12: Influence of age on strength gaining of 30%RBP:70%OPC mortars cured in water.

46

Table 11 summaries ASTM C 618 05 Standard Specification for Coal Fly Ash and Raw or Calcined Natural-pozzolan for Use as a Mineral Admixture in Concrete. The assessed physical properties of both RBP, namely fineness, water requirement & strength activity indices (SAIs) and chemical compositions are compared to Table 11 values and summarised in Table 12. It is clear that both RBPs satisfy the requirement for class F. However, for full classification of the pozzolans, it is strongly

recommended to investigate other untested properties listed in Table 11.

Table 11: Summary of ASTM C 618 05 Standard Specification for Coal Fly Ash and Raw or Calcined Natural-pozzolan for Use as a Mineral Admixture in Concrete. (see Appendix 2)

47

Table 12: Partial characterisation of red brick powders (RBPs) from El-gerief & Umdoum (G-U) and Shambat (SH) based on tested properties in this study. Chemical/Physical Property SiO2+Al2O3+Fe2O3 SO3 Moisture Content Loss on Ignition Available Alkalies Fineness #325 (percentage retained) Water Requirement Strength Activity Index with cement (SAI) -7 days (cured in lime) Strength Activity Index with cement (SAI) -28 days (cured in lime) ≥75% 84% 79% ≤105% ≥75% 103% 80% 103% 81% ≤1.5% ≤34% ASTM C618 F Class ≥70% ≤5% RBP-G-U 94.53% 0.27% ND ND 0.877% 31.34% RBP-SH 94.16% 0.216% ND ND 0.773% 32.00%

48

5. CONCLUSIONS

1. Reject brick were collected from Khartoum Estate brick industries, namely Elgerief East & Umdoum (G-U) and Shambat (SH) areas. Then they were crushed and pulverised to fine powders denoted as RBP-G-U and RBP-SH respectively.

2. The physical properties of both RBPs were assessed, namely fineness, water requirement & strength activity indices (SAIs) with Portland cement in accordance with ASTM C311.

3. The fineness of RBP-G-U when assessed as percentage passed through #325 (45µm) was found to be 68.66%. For RBP-SH the fineness was 68%.

4. For both RBPs, the water requirement assessed in accordance with clauses 26.1.1 & 30 Of ASTM C311 and ASTM C 1437 was found to be 103%.

5. Both RBPs have strength activity indices (SAIs) with OPC of about 80% assessed in accordance with clauses 26, 27, 28 and 29 of ASTM 311. This satisfies the ASTM C 618 criterion of SAI for 7 days, i.e. ≥75%. Also both RBPs satisfied the criterion for 28 days. However, RBP-G-U produced a higher SAI of 0.84 than RBP-SH which is 0.79. Despite not required by ASTM C618, the SAIs for 91 days still showed that RBP-G-U performed better than RBP-SH.

6. The mix containing 20%RBP-G-U (cured in lime) gained 32.86% extra strength from 7 days to 28 days, however this gain dropped to 22.27% from 28days to 91 days. For mix containing 20%RBP-SH these corresponding values were 22.32% and 18.88%. This shows that the rate of pozzolanic reaction for both RBPs was more significant in duration between 7 and 28days.

49

7. The mix containing 20%RBP-G-U (cured in water) gained 12.15% extra strength from 7 days to 28 days, however this gain increased significantly to 33.88% from 28days to 91 days. For mix containing 20%RBP-SH these values were 14.65% and 59.14% respectively. This shows that the rate of pozzolan reaction for both RBPs was more significant exceeding 33% in the duration between 28 to 91days.

8. The mix containing 30%RBP-G-U (cured in water) gained 38.64% extra strength from 7 days to 28 days, however this gain reduced to 24.33% from 28days to 91 days. For mix containing 30%RBP-SH these figures were

35.54% and increased to 45.70% respectively. This shows that the rate of pozzolan reaction for both RBPs was more than 24% in the duration between 28 to 91days.

9. The chemical composition of both RBPs was assessed in accordance with BS EN 196-21: 1989. The SiO2+Al2O3+Fe2O3≥70% namely 94.53% for RBP-GU and 94.16% for RBP-SH. The available alkalies were found to be 0.877% and 0.773% for RBP-G-U and RBP-SH for respectively. The SO3 were found to be 0.27% and 0.216%. for RBP-G-U and RBP-SH for respectively.

50

6. RECOMMENDATIONS FOR FURTHER STUDIES

1. Full characterisation of red brick powders (RBPs) including all of tests listed in ASTM C 311-02 and specified in ASTM C 618-05, namely the following: moisture content, loss on ignition, density, autoclave expansion, drying shrinkage, air entraining admixture demand, control of alkali silica reaction (ASR), sulphate resistance in moderate and high exposures.

2. Investigation of samples of reject bricks from other production areas in Khartoum estates, for example Suba, Algamier, Elgieli, Alkabbashi, etc.

3. Investigation of producing lime:RBPs mixtures in accordance with Indian Standard IS 1344.1981.

4. Investigation of concretes containing RBPS as cement replacement and also investigation of using coarse red bricks as aggregates in light concrete production.

5. Investigation of the economy of mass production of RBPs by preparing a feasibility study for a typical production unit.

51

7. REFERENCES
Araceli E. Lavat *, Monica A. Trezza, Mónica Poggi. Characterization of ceramic roof tile wastes as pozzolanic admixture. Waste Management 2009 : 29, pp. 1666– 1674.

Alam, Syed Ashraful. Use of biomass fuels in the brick-m akingindustries of Sudan: Implications for deforestation and greenhouse gas emission. MSc Thesis submitted for an M. Sc. Degree in Forest Ecology /Tropical Silviculture, Department of Forest Ecology, Viikki Tropical Resources Institute (VITRI), University of Helsinki, Finland 2006.

Assalam Cement Factory, Atbara, Sudan, Mr Ahmed Abdaljalil, Deputy General Manager, November 2010.

Calcrete. Computer based training (CBT) Software for learning about concrete. The Concrete Centre, Camberley, United Kingdom 2008.

Hamid, Mohammed Hussein (in Arabic).

Used of burnt

clays as pozzolanic materials. Journal of Building and Road Research. 2002 : 4, pp. 65-74.

Lewis, R, Sear, L, Wainwright, P and Ryle, R. Cementitious additions. In John Newman and Ban Seng Choo eds, Advanced Concrete Technology: Constituent Materials. Oxford: Butterworth-Heinemann, 2003, pp. 99-159.

Mamoun E.E. and Zamzami M.A. Characterization of natural pozzolanic deposits of Northern Bayouda Northern Sudan. Journal of Building and Road Research. 2010 : 9, pp. 20-27.

Moir, G (2003). Cements. In John Newman and Ban Seng Choo eds, Advanced Concrete Technology: Constituent Materials. 2003, pp. 18-62. Oxoford : Butterworth-Heinemann,

52

Practical Action Technical Brief.

Pozzolans: Calcined Clays and Shales, and

Volcanic Ash. The Schumacher Centre for Technology & Development, Bourton on Dunsmore, RUGBY,CV23 9QZ, United Kingdom. Tel.:+44(0)1926634400, website:

Fax:+44(0)1926634401,email

practicalaction@practicalaction.org.uk,

http://www.practicalaction.org, 2011.

Salih, et al.: Indicators about the future uses of pozzolans in Sudan, a paper presented in workshop, held in Khartoum on pozzolans uses in the construction industry in Sudan. 1999.

Sanchez de Rojas, M. I., Marin, F., Rivera, J. and Frias, M.

Morphology and

Properties in Blended Cements with Ceramic Wastes as a Pozzolanic Material. J. Am. Ceram. Soc., 2006 : 89 [12], pp. 3701–3705.

Suleiman, S. H.. Geological Report on Pozzolan Quarry Sites in the Bayuda Volcanic Field. Unpublished report Prepared for Nevkarsand for Investment Industry Transport & Construction Co. Ltd. 2008.

Specification for Calcined Clay Pozzolan (2nd revision), Indian Standard IS 1344: 1981, Bureau of Indian Standards, New Delhi, India, http://www.bis.org.in/

Torkittikul, Pincha and Chaipanich, Arnon. Utilization of ceramic waste as fine aggregate within Portland cement and fly ash concretes. Cement & Concrete Composites 2010: 32, pp. 440–449.

Ummi Kalsum, H.M.N, Mashitah, M.D. and Badorul, A.B. Recycling of clay based demolition wastes for the production of concrete block. International Conference on Environment 2008 (ICENV 2008).
‫مؤشرات حول مستقبل استخدامات البوزوالنا في السودان 1119 ورقة قدمت في ورشة عمل في استخدامات البوزوالنا‬ 9111 ‫في صناعة البناء بالسودان منظمة البدائل االقتصادية السودانية قاعة الشارقة جامعة الخرطوم 91 سبتمبر‬

53

8. APPENDICES

54

8.1.

Appendix 1: ASTM C311 02 Standard Test Methods for

Sampling and Testing Fly Ash or Natural Pozzolans for Use in Portland-Cement Concrete

55

56

57

58

59

60

61

62

63

64

8.2.

Appendix 2: ASTM C 618 05 Standard Specification for Coal
Fly Ash and Raw or Calcined Natural-pozzolan for Use as a Mineral Admixture in Concrete.

65

27

66

28

67

68

8.3.

Appendix 3: Standard Sand adapted from Table 4 of BS 882:1992

69