Natural pozzolans mixed with lime were used in concrete construc-tion long before the invenconstruc-tion of portland cement because of their contribution to the strength of concrete and
Trang 1ACI 232.1R-00 supersedes ACI 232.1R-94 and became effective December 6, 2000 Copyright 2001, American Concrete Institute.
All rights reserved including rights of reproduction and use in any form or by any means, including the making of copies by any photo process, or by electronic or mechanical device, printed, written, or oral, or recording for sound or visual reproduc- tion or for use in any knowledge or retrieval system or device, unless permission in writing is obtained from the copyright proprietors.
ACI Committee Reports, Guides, Standard Practices,
and Commentaries are intended for guidance in planning,
designing, executing, and inspecting construction This
document is intended for the use of individuals who are
competent to evaluate the significance and limitations of
its content and recommendations and who will accept
re-sponsibility for the application of the material it contains
The American Concrete Institute disclaims any and all
re-sponsibility for the stated principles The Institute shall
not be liable for any loss or damage arising therefrom
Reference to this document shall not be made in
con-tract documents If items found in this document are
de-sired by the Architect/Engineer to be a part of the contract
documents, they shall be restated in mandatory language
for incorporation by the Architect/Engineer
232.1R-1
Use of Raw or Processed Natural Pozzolans
in Concrete
ACI 232.1R-00
This report provides a review of the state-of-the-art use of raw or processed
natural pozzolans in concrete and an overview of the properties of natural
pozzolans and their proper use in the production of hydraulic-cement
con-crete Natural pozzolans mixed with lime were used in concrete
construc-tion long before the invenconstruc-tion of portland cement because of their
contribution to the strength of concrete and mortar Today, natural
poz-zolans are used with portland cement not only for strength, but also for
economy and beneficial modification of certain properties of fresh and
hardened portland-cement concrete.
This report contains information and recommendations concerning the selection and use of natural pozzolans generally conforming to the applica- ble requirements of ASTM C 618 and CSA A23.5 Topics covered include the effect of natural pozzolans on concrete properties, a discussion of qual- ity control and quality assurance, and guidance regarding handling and use of natural pozzolans in specific applications References are provided that offer more information on each topic.
Keywords: alkali-silica reaction; cement; concrete; concrete strength;
diatomaceous earth; lime; natural pozzolan; pozzolan; pozzolanic activity; sulfate attack (on concrete).
CONTENTS Chapter 1—General, p 232.1R-2
1.1—History1.2—Definition of a natural pozzolan1.3—Chemical and mineralogical composition1.4—Classification
1.5—Examples
Reported by ACI Committee 232
Gregory M Barger* Allen J Hulshizer Sandor PopovicsBayard M Call Tarif M Jaber Jan Prusinski Ramon L Carrasquillo Jim S Jensen Dan Ravina James E Cook Elizabeth S Jordan* D V Reddy Douglas W Deno Paul Klieger* Harry C Roof George R Dewey Steven H Kosmatka Della Roy Edwin R Dunstan, Jr Ronald L Larson John M Scanlon‡
William E Ellis, Jr V M Malhotra Ava Shypula*
Karen A Gruber* Bryant Mather* Robert SparacinoWilliam Halczak Richard C Mielenz* Michael D A Thomas
G Terry Harris, Sr Tarun R Naik Samuel S Tyson
R Douglas Hooton* Terry Patzias† Orville R Werner, II
Paul J Tikalsky*Chairman
Morris V Huffman*Secretary
* Subcommittee members for this report.
† Subcommittee chairman for this report.
‡ Deceased.
Note: Special thanks is extended to P K Mehta and Caijun Shi for their help with this document.
Trang 21.6—Chemical and physical properties
1.7—Uses
Chapter 2—Effects of natural pozzolan on
concrete properties, p 232.1R-8
2.1—Concrete mixture proportions
2.2—Properties of fresh concrete
2.3—Properties of hardened concrete
Chapter 3—Specifications, test methods, quality
control, and quality assurance, p 232.1R-16
3.1—Introduction
3.2—Chemical requirements
3.3—Physical requirements
3.4—General specification provisions
3.5—Methods of sampling and testing
3.6—Quality control and quality assurance
Chapter 4—Concrete production using natural
6.1—Grouts and mortars
6.2—Controlled low-strength materials
Lime and limestone are among the oldest materials used
by mankind for construction purposes Structures built of
limestone include the pyramids of Egypt Long before the
in-vention of portland cement in 1824, mortars and concretes
composed of mixtures and fillers and raw or heat-treated
lime were used for construction throughout the world
(Mali-nowski 1991)
Malinowski et al (1993) report that the oldest example of
hydraulic binder, dating from 5000-4000 B.C., was a mixture
of lime and natural pozzolan, a diatomaceous earth from the
Persian Gulf The next oldest reported use was in the
Mediter-ranean region The pozzolan was volcanic ash produced from
two volcanic eruptions: one, sometime between 1600 and
1500 B.C on the Aegean Island of Thera, now called Santorin,
Greece; the other in 79 A.D at Mt Vesuvius on the bay of
Na-ples, Italy Both are volcanic ashes or pumicites consisting of
almost 80% volcanic glass (pumice and obsidian)
According to the Roman engineer Marcus Vitruvius Pollio
(Vitruvius Pollio 1960), who lived in the first century B.C.,
the cements made by the Greeks and the Romans were of perior durability, because “neither waves could break, norwater dissolve” the concrete In describing the building tech-niques of masonry construction, he indicated that the Ro-mans developed superior practices of their own from thetechniques of the Etruscans and the Greeks The Greek ma-sons discovered pozzolan-lime mixtures sometime between700-600 B.C and later passed their use of concrete along tothe Romans in about 150 B.C During the 600 years of Ro-man domination, the Romans discovered and developed avariety of pozzolans throughout their empire (Kirby et al.1956)
su-During archaeological excavations in the 1970s at the cient city of Camiros on the Island of Rhodes, Greece, an an-cient water-storage tank having a capacity of 600 m3 (785 yd3)was found Built in about 600 B.C., it was used until 300 B.C.when a new hydraulic system with an underground watertank was constructed For almost three millennia this watertank has remained in very good condition, according to Ef-stathiadis (1978)
an-Examination of the materials used for this structure vealed that the concrete blocks and mortar used were madeout of a mixture of lime, Santorin earth, fine sand (<2 mm[<0.08 in.]) and siliceous aggregates with sizes ranging be-tween 2 and 20 mm (0.08 and 0.79 in.) The fresh concretewas placed into wooden sidewall molds The compressivestrength of a 20 mm (0.79 in.) cubic specimen was found to
re-be 12 MPa (1740 psi) Mortars like these were known tohave a composition of six parts by volume of Santorin earth,two parts by volume of lime, and one part by volume of finesand These mortars were used as the first hydraulic cements
in aqueducts, bridges, sewers, and structures of all kinds.Some of these structures are still standing along the coasts ofItaly, Greece, France, Spain, and in harbors of the Mediter-ranean Sea The Greeks and Romans built many such struc-tures over 2000 years ago Examples of such structures arethe Roman aqueducts as well as more recent structures such
as the Suez Canal in Egypt (built in 1860) (Luce 1969), theCorinthian Canal (built in 1880), the sea walls and marinestructures in the islands of the Aegean Sea, in Syros, Piraeus,Nauplion, and other cities, and the harbors of Alexandria inEgypt, Fiume, Pola Spalato, Zara on the Adriatic Sea, andConstanta (Romania) on the Black Sea All of these struc-tures provide evidence of the durability of pozzolan-limemortar under conditions of mild weathering exposure Ro-man monuments in many parts of Europe are in use today,standing as a tribute to the performance of lime-pozzolanmortars (Lea 1971)
1.2—Definition of a natural pozzolan
Pozzolan is defined in ACI 116R as:
“ a siliceous or siliceous and aluminous material,which in itself possesses little or no cementitious valuebut will, in finely divided form and in the presence ofmoisture, chemically react with calcium hydroxide at or-dinary temperatures to form compounds possessing ce-mentitious properties.”
Trang 3Natural Pozzolan is defined as:
“ either a raw or calcined natural material that has
poz-zolanic properties (for example, volcanic ash or pumicite,
opaline chert and shales, tuffs, and some diatomaceous
earths).”
ASTM C 618 and CSA A23.5 cover coal fly ash and
nat-ural pozzolan for use as a mineral admixture in concrete The
natural pozzolans in the raw or calcined state are designated as
Class N pozzolans and are described in the specifications as:
“Raw or calcined natural pozzolans that comply with the
applicable requirements for the class as given herein, such
as some diatomaceous earth; opaline chert and shales;
tuffs and volcanic ashes or pumicites, any of which may or
may not be processed by calcination; and various materials
requiring calcination to induce satisfactory properties,
such as some clays and shales.”
Similar materials of volcanic origin are found in Europe,
where they have been used as an ingredient of
hydraulic-ce-ment concrete for the past two centuries
Raw or processed natural pozzolans are used in the
pro-duction of hydraulic-cement concrete and mortars in two
ways: as an ingredient of a blended cement, or as a mineral
admixture This report deals with the second case Blended
cements are covered in ACI 225R Fly ash and silica fume
are artificial pozzolans and are covered in ACI 232.2R and
234R
1.3—Chemical and mineralogical composition
The properties of natural pozzolans vary considerably,
de-pending on their origin, because of the variable proportions
of the constituents and the variable mineralogical and
phys-ical characteristics of the active materials Most natural
poz-zolans contain substantial amounts of constituents other than
silica, such as alumina and iron oxide, which will react with
calcium hydroxide and alkalies (sodium and potassium) to
form complex compounds Pozzolanic activity cannot be
de-termined just by quantifying the presence of silica, alumina,
and iron The amount of amorphous material usually
deter-mines the reactivity of a natural pozzolan The constituents
of a natural pozzolan can exist in various forms, ranging
from amorphous reactive materials to crystalline products
that will react either slowly or not at all Because the amount
of amorphous materials cannot be determined by standard
techniques, it is important to evaluate each natural pozzolan
to confirm its degree of pozzolanic activity There is no clear
distinction between siliceous materials that are consideredpozzolans and those that are not Generally, amorphous sili-
ca reacts with calcium hydroxide and alkalies more rapidlythan does silica in the crystalline form (quartz) As is thecase with all chemical reactions, the larger the particles (thelower the surface area per unit volume) the less rapid therate of reaction Therefore, the chemical composition of apozzolan does not clearly determine its ability to combinewith calcium hydroxide and alkalies
Volcanic glasses and zeolitic tuffs, when mixed with lime,produce calcium silicate hydrates (CSH) as well as hydratedcalcium aluminates and calcium aluminosilicates These ma-terials were proven to be good pozzolans long ago Naturalclays and shales are not pozzolanic, or only weakly so, asclay minerals do not react readily with lime unless their crys-talline structure is partially or completely destroyed by cal-cination at temperatures below 1093 C (2000 F)
High-purity kaolin may be processed to form a ity pozzolan called high-reactivity metakaolin Italian re-searchers who have studied volcanic glasses and therelationship to pozzolanic activity believe that “reactiveglass originated from explosive volcanic eruptions” like theones from the volcanoes of Thera and Mount Vesuvius,which produced the natural pozzolans with unaltered alumi-nosilicate glass as their major component (Malquori 1960).Both are pumicites, one third of which is in the amorphousstate (glass), and are highly reactive with lime and alkalis atnormal temperatures
high-qual-1.4—Classification
Mehta (1987) classifies natural pozzolans in four ries based on the principal lime-reactive constituent present:unaltered volcanic glass, volcanic tuff, calcined clay orshale, and raw or calcined opaline silica This classification
catego-is not readily applicable to pozzolans of volcanic origin egories 1 and 2) because volcanic tuffs commonly includeboth altered and unaltered siliceous glass These are the sole
(cat-or primary sources of pozzolanic activity in siliceous glass,opal, zeolites, or clay mineralsthe activity of the last twobeing enhanced by calcination In Table 1.1, the chemical
Table 1.1—Typical chemical and mineralogical analysis of some natural pozzolan (Mehta 1987)
Pozzolan
Ignition Loss, %
crystalline matter, % Major crystalline minerals SiO2 Al2O3 Fe2O3 CaO MgO Alkalies*
Non-Santorin earth 65.1 14.5 5.5 3.0 1.1 6.5 3.5 65 to 75 Quartz, plagioclase Rhenish trass 53.0 16.0 6.0 7.0 3.0 6.0 — 50 to 60 Quartz, feldspar, analcime Phonolite 55.7 20.2 2.0 4.2 1.1 10.8 3.6 — Orthoclase, albite, pyroxene, calcite
Roman tuff 44.7 18.9 10.1 10.3 4.4 6.7 4.4 — Herschelite, chabazite, phillipsitesNeapolitian glass 54.5 18.3 4.0 7.4 1.0 11.0 3.1 50-70 Quartz, feldspar
Trang 4and mineralogical composition is given for some of the
well-known pozzolans
A classification of natural pozzolans based on the identity of
the pozzolanic constituents was devised by Mielenz, Witte,
and Glantz (1950) Substances that are pozzolanic or whose
pozzolanic activity can be induced by calcination were
clas-sified as volcanic glass, opal, clays, zeolites, and hydrated
oxides of aluminum Activity type 3 (clays) was subdivided
into five subtypes: 3a kaolinite, 3b montmorillonite, 3c illite,
3d clay mixed with vermiculite, and 3e palygorskite
1.5—Examples
Following is a discussion of some natural pozzolans
pro-duced in various parts of the world
Santorin earth—Santorin earth is produced from a natural
deposit of volcanic ash of dacitic composition on the island
of Thera, in the Agean Sea, also known as Santorin, which
was formed about 1600-1500 B.C after a tremendous
explo-sive volcanic eruption (Marinatos 1972)
Pozzolana—Pozzolana is produced from a deposit of
pumice ash or tuff comprised of trachyte found near Naples
and Segni in Italy Trachyte is a volcanic rock comprised
pri-marily of feldspar crystals in a matrix of siliceous glass
Poz-zolana is a product of an explosive volcanic eruption in 79 A.D
at Mount Vesuvius, which engulfed Herculaneum, Pompeii,
and other towns along the bay of Naples The deposit near
Pozzuoli is the source of the term “pozzolan” given to all
ma-terials having similar properties Similar tuffs of lower silica
content have been used for centuries and are found in the
vi-cinity of Rome
Rhenish trass—Rhenish trass, a natural pozzolan of
volca-nic origin (Lovewell 1971), has been well known since ancient
Roman times The material is a trachytic tuff that differs from
place to place and is found in the Valley of the Rhine River in
Germany Similar tuffs have been used in Bavaria
Gaize—Gaize is a pozzolan found in France that is not of
volcanic origin but a porous sedimentary rock consistingmainly of opal The material is usually calcined at tempera-tures around 900 C (1620 F) before it is used as a pozzolan
or as a component of portland-pozzolan cement
Volcanic tuffs, pumicites, diatomaceous earth, and line shales—In the United States, volcanic tuffs and pumic-
opa-ites, diatomaceous earth, and opaline shales are foundprincipally in Oklahoma, Nevada, Arizona, and California.Natural pozzolans were investigated in this country byBates, Phillips, and Wig as early as 1908 (Bates, Phillips,and Wig 1912) and later by Price (1975), Meissner (1950),Mielenz, Witte, and Glantz (1950), Davis (1950), and others.They showed that concretes containing pozzolanic materialsexhibited certain desirable properties such as lower cost,lower temperature rise, and improved workability Accord-ing to Price (1975), an example of the first large-scale use ofportland-pozzolan cement, composed of equal parts of port-land cement and a rhyolitic pumicite, is the Los Angeles aq-ueduct in 1910-1912
The studies of natural pozzolans by the United States reau of Reclamation (USBR) in the 1930s and 1940s encour-aged their use for controlling heat of hydration and alkali-silica reaction of concrete in large dams Siliceous shales ofthe Monterey Formation in Southern California have beenproduced commercially and used extensively in the sur-rounding areas Price (1975) also states that sources of natu-ral pozzolan that do not require calcining to make themactive are located mainly west of the Mississippi River Gen-erally the pozzolanic deposit was in the vicinity of the partic-ular project and the amount required was sufficient to supportmining and processing costs The deposit was usually aban-doned at the completion of the project
Bu-Large deposits of diatomite were discovered decades ago
in the coastal ranges of central California and the peninsularranges of southern California The largest reserves of fresh-water diatomite are in the northeastern counties of Shasta,Siskiyou, Modoc, and Lassen (Burnett 1991) Diatomite con-sists of microscopic opaline silica frameworks Some diatoma-ceous shale deposits contain hydrocarbon impregnants thatprovide some of the fuel for their calcination (see Table 1.2)
In 1993, a study was undertaken that appraised as a source
of pozzolan a lacustrine deposit located about 48.3 km (30 mi)north of Reno, Nevada The material is an intermingling of di-atomaceous earth and dacite pumicite The raw material wascalcined and ground for marketing under the trade name Las-senite It was used (1970-1989) for the concrete construction
of structures, bridges, roadways, the trans-Canada highway,the Auburn dam, and the Los Melones dam and power plant
It has also been used in research projects by the Department
of Transportation of the State of California during the periodfrom January 1987 to August 1991
Pumicite is a finely divided volcanic ash composed ofangular and porous particles of siliceous glass and varyingproportions of crystal fragments differing from pumice only
in grain size Pumicites are mainly rhyolitic or dacites incomposition They occur as stratified or massive deposits,commonly as lake beds
Table 1.2—Mineral admixtures and structures that
used them (Elfert 1974)
Name
Date completed Type of pozzolan Arrowrock Dam 1915 Granite*
Lahontan Dam 1915 Siliceous silt*
Elephant Butte Dam 1916 Sandstone*
Davis Dam 1950 Calcined opaline shale
Glenn Anne Dam 1953 Calcined oil-impregnated diatomaceous shale
Cachuma Dam 1953 Calcined oil-impregnated diatomaceous shale
Tecolote Tunnel 1957 Calcined oil-impregnated
diatomaceous shale Monticello Dam 1957 Calcined diatomaceous clay
Twitchell Dam 1958 Calcined diatomaceous clay
Flaming George Dam 1963 Calcined montmorillonite shale
* By present standards, these materials have very little pozzolanic activity.
Trang 5A deposit in the Upper Fox Hills, 9.7 km (6 mi) north and
east of Linton, North Dakota (Fisher 1952, Manz 1962), was
examined at the University of North Dakota by N N
Koha-nowski of the Geology Department and was found to be
al-tered pozzolanic volcanic ash Crawford (1955) describes
similar deposits in Saskatchewan and refers to them as
pum-icite, which he described as a finely divided powder of a
white to gray or yellowish color composed of small, sharp,
angular grains of highly siliceous volcanic glass, usually
rhyolitic in composition
Stanton (1917) described the Cretaceous volcanic ash bed
on the Great Plains near Linton, North Dakota, as severalconspicuous white outcrops that suggest chalk or diatoma-ceous earth At one exposure, 1.6 km (1 mi) southeast of Lin-ton, the thickness of the white bed is 8 m (26 ft) and the rock
is very fine-grained and mostly massive, although it containssome thin-bedded layers A sample examined by G F.Loughlin consisted of 80% volcanic glass, 15% quartz andfeldspar, and 2 to 3% biotite
The Linton area ash bed is generally overlain by sand andunderlain by shale Contamination of the ash by this adjacentmaterial is detrimental If the ash is carefully mined, with noadmixture of sand or shale, the volcanic ash need only bedried at 100 C (212 F) and finely ground to comply withASTM C 618 Tests were performed in 1961 on compositesamples of volcanic ash, crushed and ground in a ball mill andcalcined at 538, 760, and 927 C (1000, 1400, and 1700 F), re-spectively, for 15 min and 1 h The results are shown in Ta-bles 1.3 and 1.4 Based on these tests conducted on thesamples submitted, the material, when calcined at 760 C(1400 F), complied with ASTM C 618
Rice husk ash—Rice husk ash (RHA) is produced from
rice husks, which are the shells produced during the ing operation of rice Rice husks are approximately 50% cel-lulose, 30% lignin, and 20% silica A scanning electronmicrograph illustrating the typical cellular structure of ricehusks where the silica is retained in noncrystalline formshown in Fig 1.2 To reduce the amount of waste materials,rice husks are incinerated by controlled combustion to re-move the lignin and cellulose, leaving behind an ash com-posed mostly of silica (retaining 20% of the mass of ricehusks) as seen in Fig 1.3
dehusk-Table 1.3—Cretaceous volcanic ash from North
Dakota (copy of report submitted to Minnesota
Electronics Company, St Paul, Minn.) *†
Testing parameters Samples ASTM C 618
µ m (No 325) sieve, % 7.85 — 10.26 34.0 max.
Strength activity index: with
lime at 7 days, MPa (psi),
50 x 100 mm cylinders
(2 x 4 in.)
50 mm cubes (2 in.)
4.2 (611) 4.6 (665)
4.7 (680)
7.1 (1030) 7.7 (1120)
—
—
Strength activity index:
with portland cement, at
Increase of drying
shrink-age of mortar bars at
28 days, difference, in %
over control
— — 0.025 0.03 max.
* By the Northwest Laboratories, Seattle, Wash., in 1960.
† These tests were performed on composite samples of volcanic ash from 20 test holes.
The portions from each test hole are taken from 0.3 m to 7 to 9 m (1 ft to 23 to 30 ft)
levels The material was crushed, ground in a ball mill, and calcined at 538 and 760 C
(1000 and 1400 F) for 15 min.
Table 1.4—Test results of North Dakota volcanic ash
Testing
tion 61-1 61-1 61-1 61-5 61-13 ASTMC 618Processed calci-
Specifica-nation ture
tempera-100 C (212 F)
760 C (1400 F)
927 C (1700 F)
100 C (212 F)
100 C (212 F) —
Density, Mg/m3 2.37 2.50 2.39 — — — Amount retained
on 45 µ m (No
325) sieve, %
Strength activity index with lime at
7 days, MPa (psi),
50 x 100 mm inders (2 x 4 in.)
cyl-6.6 (9.52)
9.5 (1375)
7.0 (1015)
7.5 (1090)
7.0 (1.10) —
Strength activity index with port- land cement at 28 days, % of con- trol
Water ment, % of con- trol
require-110 112 114 110 110 115 max.
Color of sample Light gray Light buff Dark buff Light gray Light gray —
Note: The materials tested were grounded with a muller Calcining was done at 760 C (1400 F) and 927 C (1700 F) for a period of 1 h.
Fig 1.1—Scanning electron micrograph of rice husk (Mehta
1992).
Trang 6Mehta (1992) has shown that RHA, produced by
con-trolled incineration under oxidizing conditions at relatively
low combustion temperatures and short holding time, is
highly pozzolanic with high surface area (50 to 100 m2/g by
nitrogen adsorption), and consists mainly of amorphous
sili-ca By varying the temperature, RHA can be produced with
a range of colors, from nearly white to black The chemical
analysis of fully burnt RHA shows that the amorphous silica
content ranges between 90 and 96% It is a highly active
poz-zolan, suitable for making high-quality cement and concrete
products The average particle size of ground RHA varies
from 10 to 75 mm (No 1500 – 200 sieve)
To obtain lower-permeability concrete, RHA can be added
in amounts of 5 to 15% by mass of cement The benefits of
using RHA, as shown by Mehta and Folliard (1995) and
Zhang and Malhotra (1996), are higher compressivestrength, decreased permeability, resistance to sulfate attack,resistance to acid attack, reduction of surface cracking instructures, excellent resistance to chloride penetration, andexcellent performance under freezing-and-thawing cycling
Metakaolin—Metakaolin (Al2O3:2SiO2) is a natural zolan produced by heating kaolin-containing clays over atemperature range of about 600 to 900 C (1100 to 1650 F)above which it recrystallizes, rendering it mullite(Al6Si2O13) or spinel (MgAl2O4) and amorphous silica (Mu-
poz-rat, Ambroise, and Pera 1985) The reactivity of metakaolin
is dependent upon the amount of kaolinite contained in theoriginal clay material The use of metakaolin as a poz-zolanic mineral admixture has been known for many years,but has grown rapidly since approximately 1985 The aver-age particle size of metakaolin varies and can be controlledduring the processing to change the properties of the fresh con-crete In general, the average particle size of high-reactivitymetakaolin ranges from 0.5 to 20 µm
The pozzolanic properties of metakaolin are well mented Kostuch, Walters, and Jones in 1993 indicate thatcalcium hydroxide released during cement hydration is con-sumed if the formulation contains a sufficient quantity ofhigh-reactivity metakaolin (Fig 1.3) The consumption ofcalcium hydroxide causes the formation of calcium silicatehydrate (CSH) and stratlingite (C2ASH8) DeSilva andGlasser (1991) report that metakaolin can react with sodium,potassium, and calcium hydroxides, as well as gypsum andportland cement Gruber and Sarkar (1996) confirm the re-duction of calcium hydroxide by the use of high-reactivitymetakaolin, having an average particle size of about 2 µm.From 1962-1972, approximately 250,000 metric tons(227,300 tons) of calcined kaolinitic clay was used in theconstruction of four hydroelectric dams in Brazil (Saad , An-drade, and Paulon 1982) In the United Kingdom, large-scale
docu-Fig 1.3—Effect of replacing part of portland cement in
crete by metakaolin on calcium hydroxide content of
con-crete as it cures (Kostuch, Walters, and Jones 1993).
Fig 1.4—Effect of high-reactivity metakaolin at 0.4 w/cm ratio on compressive strength of concrete (Hooton, Gruber, and Boddy 1997).
Fig 1.2—Scanning electron micrograph of rice husk ash
(Mehta 1992).
Trang 7trials have been conducted using high-reactivity metakaolin
concretes subjected to aggressive environments (Ashbridge,
Jones, and Osborne 1996) Their research shows excellent
strength development, reduced permeability, and chemical
resistance In addition, strength, pozzolanic activity, and
ce-ment hydration characteristics have been studied in
super-plasticized metakaolin concrete (Wild, Khatib, and Jones
1996)
In the United States, metakaolin has been evaluated as a
pozzolan in various research studies as well as in the field
In one air-entrained high-performance concrete mixture, the
metakaolin-containing concrete showed increased strength
and reduced chloride penetration compared to the portland
cement control design, while maintaining good workability
and an air-void system that produced good resistance to
cy-cles of freezing and thawing and to deicer scaling
(Cal-darone, Gruber, and Burg 1994) Benefits of using
high-reactivity metakaolin in ternary systems with ground
granu-lated blast-furnace slag and fly ash have also been reported
(Caldarone and Gruber 1995) Fig 1.4 and 1.5 shows the
ef-fect of a high-reactivity metakaolin on compressive strength
of concrete (Hooton, Gruber, and Boddy 1997) Mixtures
with 8 to 12% metakaolin replacement at 0.4 to 0.3
water-ce-mentitious materials ratio (w/cm) greatly improved the
com-pressive strength at all ages Hooton, Gruber, and Boddy
(1997) showed that high-reactivity metakaolin enhanced
re-sistance to chloride ingress
1.6—Chemical and physical properties
When a mixture of portland cement and a pozzolan reacts,
the pozzolanic reaction progresses like an acid-base reaction
of lime and alkalies with oxides (SiO2 + A12 O3 + Fe2O3) of
the pozzolan Two things happen First, there is a gradual
de-crease in the amount of free calcium hydroxide with time,
and second, during this reaction there is an increase in
for-mation of CSH and calcium aluminosilicates that are similar
to the products of hydration of portland cement (Fig 1.6).According to Lea (1971), the partial replacement of portlandcement by pozzolan of high SiO2/R2O3 (R2O3 = Al2O3 +
Fe2O3) ratio has been found to increase the resistance of crete to sulfate and seawater attack (R2O3 is approximatelythe summation of the Al2O3 and Fe2O3 contents) This is, inpart, attributable to the removal of free hydroxide formed inthe hydration of portland cements
con-The result is that the hardened cement paste contains lesscalcium hydroxide, more CSH, and other products of low po-rosity Research on the hydration of blended cements madewith natural pozzolans of volcanic origin (Santorin earth,pozzolana) indicated that pore refinement resulting frompozzolanic reaction is important for enhancing chemical du-rability and mechanical strength (Mehta 1987)
The shape, fineness, particle-size distribution, density, andcomposition of natural pozzolan particles influence the prop-erties of freshly mixed unhardened concrete and the strengthdevelopment of hardened concrete Most natural pozzolanstend to increase the water requirement in the normal consis-tency test as a result of their microporous character and highsurface area Natural pozzolans can improve the perfor-mance of both fresh and hardened concrete when used as aningredient of portland-pozzolan cement or as an admixture toportland-cement concrete
1.7—Uses
Pozzolans of natural origin have been used in mass crete on large projects in the United States, and where theyare locally available they are used in concrete constructionand manufacture of concrete products Such uses of poz-zolans of natural origin are more widespread in Europe than
in the United States Natural pozzolans are now used in crete in a variety of ways, depending upon their reactivity.The natural pozzolans may be used as partial replacementsfor portland cement or in addition to portland cement Somenatural pozzolans have been used in much the same way as
con-Fig 1.5—Effect of high-reactivity metakaolin at 0.3 w/cm
ratio on compressive strength of concrete (Hooton, Gruber,
and Boddy 1997).
Fig 1.6—Changes in calcium hydroxide content of ing portland-pozzolan cement (Lea 1971).
hydrat-0.00 0.50 1.00 1.50 2.00 2.50
Portland-Pozzolan Cement Containing 40% Pozzolan Portland Cement
Trang 8fly ash Other natural pozzolans of high reactivity, such as
metakaolin, have been found to perform similarly to silica
fume, and are used in a similar manner
According to Mielenz, Witte, and Glantz (1950), in 1933
the USBR undertook an intensive study on using natural
pozzolans for the purpose of controlling the heat of
hydra-tion of concrete and other concrete benefits for mass
con-crete applications such as large dams Several investigations
revealed the effect of calcination of more than 200
prospec-tive natural pozzolans on their properties and performance in
concrete The following properties were reported:
1 Mineralogical and chemical composition;
2 Pozzolanic activity, water requirement, and strength; and
3 Expansion due to alkali-silica reactivity
Mielenz, Witte, and Glantz (1950) conclude that
calcina-tion of clay minerals was essential to develop satisfactory
pozzolanic activity, and the response to heat treatment varied
with the type of clay minerals present Many natural
poz-zolans were usable in the raw state If moist, they usually
re-quired drying and grinding before use The best natural
pozzolans owed their activity to volcanic glass with 70 to
73% SiO2 content, with 40 to 100% being in the form of
rhy-olitic glass Mielenz (1983) gives the history and
back-ground on mineral admixtures along with the use of natural
pozzolans (raw and calcined) Elfert (1974) describes the
ex-periences of the USBR in the use of large quantities of fly
ash and natural pozzolans in the western United States Table
1.2 lists the types of mineral admixtures used in concrete
dams, built during the time period 1915-1964
Today, blended cements consisting of portland cement
and pozzolan, as covered by ASTM C 595 and C 1157, are
used in concrete construction for economic reasons to help
reduce the energy consumption and to achieve specific
technical benefits
In the 1920s and 1930s, natural pozzolans were used as a
mineral admixture in concrete for the construction of dams
and other structures then being constructed by the Los
Ange-les County Flood Control District The California Division
of Highways used a specially made portland-pozzolan
ce-ment in several structures (bridges) because of its proven
re-sistance to sulfate attack from seawater and its lower heat of
hydration (Davis 1950)
Meissner (1950) reports that a portland-pozzolan cement
containing 25% interground calcined Monterey shale was
produced during the 1930s and 1940s The California
Divi-sion of Highways used this cement in the 1930s in several
structures, including the Golden Gate Bridge and the San
Francisco-Oakland Bay Bridge Another portland-pozzolan
cement, containing 25% interground calcined pozzolan, was
used in 1935 for the construction of the Bonneville Dam
spillway on the lower Columbia River In 1940 to 1942 the
USBR built the Friant Dam on the San Joaquin River in
Cal-ifornia with a portland cement-pozzolan combination The
pozzolan was a naturally fine rhyolite pumicite, which was
batched separately at the concrete mixer at the rate of 20%
by mass of cement This pozzolan was obtained from a
de-posit along the San Joaquin River near Friant
During the 1960s and early 1970s, natural pozzolan wasused at the rate of 42 kg/m3 (70 lb/yd3) in nearly all of theconcrete in the California State Water Project, including lin-ing of the California Aqueduct (Tuthill 1967, Tuthill and Ad-ams 1972) This was the most extensive use of a naturalpozzolan in a project in U.S history Requirements on thispozzolan exceeded those of ASTM C 618
A kaolin clay from Brazil has been used since 1965 as aningredient in concrete in the construction of large dams at acost of approximately 1/3 that of portland cement (Saad , An-drade, and Paulon 1982) This natural pozzolan is produced
by calcining kaolin clay and grinding it to a fineness of 700
to 900 m2/kg (380 to 490 yd2/lb) Because of this high ness and activity it can be used for cement replacement up to50% by volume, with 90-day compressive strength similar toconcrete made with portland cement At Jupia Dam, the use
fine-of this natural pozzolan, at 20 to 30% fine-of the volume fine-of ment, resulted in lower temperature rise, improved cohesion,and reduction of expansion due to alkali-silica reaction (An-driolo 1975) When first used for general concrete construc-tion the pozzolan replaced 30% of the cement by volume,and when used for structural concrete construction the rate ofreplacement was 20% The use of this high-reactivity poz-zolan in mass concrete construction provided substantialgains in cost and improved the concrete properties (Saad,Andrade, and Paulon 1982)
ce-CHAPTER 2—EFFECTS OF NATURAL POZZOLAN
ON CONCRETE PROPERTIES 2.1—Concrete mixture proportions
The most effective method for evaluating the performance
of a concrete containing a natural pozzolan and establishingproper mixture proportions for a specific application is theuse of trial batches and a testing program Because some nat-ural pozzolans perform better than others and project re-quirements differ, optimum proportions for a givencombination of pozzolan and portland cement cannot be pre-dicted When used as a replacement for a portion of portlandcement, natural pozzolan replaces an equal volume or equalmass of the cement Because the density of natural pozzolans
is typically less than the density of portland cement, mass placement results in a greater volume of total cementitiousmaterials than when volume replacement is used at a givenpercentage The mass of natural pozzolan employed may begreater than that of the replaced cement if the concrete is pro-portioned for optimum properties and maximum economy.Proportioning techniques for concrete including a finelydivided mineral admixture are similar to those used in pro-portioning concrete that does not include such an admixture.Proportioning techniques for concrete mixtures are given inACI 211.1 Specific procedures for proportioning mixturescontaining pozzolans were developed by Lovewell and Hy-land (1974) Finely divided mineral admixtures, whether nat-ural pozzolan or other finely divided material, should usually
re-be regarded as part of the cement paste matrix in determiningthe optimum percentages of fine and coarse aggregate.The effect of the natural pozzolan on the mixing water re-quirement should also be determined Some finely divided
Trang 9mineral admixtures cause a major increase in water
require-ment; others have little or no effect on water requirement,
and still others typically reduce the water requirement of
concrete in which they are used (Mather 1958) Natural
poz-zolans affect the water requirement of the concrete and
there-fore the cement content A natural pozzolan should be
considered as part of the cementitious material (U.S Bureau
of Reclamation 1975) The amount of natural pozzolan used
varies significantly based upon the activity of the pozzolan
Some natural pozzolans are used in a range of 15 to 35%
based upon the mass of the total cementitious material in the
concrete More reactive natural pozzolans can be used in
lower concentrations of 5 to 15% by mass of total
cementi-tious material; however, such low concentrations may
in-crease expansion resulting from the altered silica reaction in
the presence of some alkali-reactive aggregates (Stanton
1950) The optimal amount of natural pozzolan depends on
where the concrete is used and the specifications for the
work
2.2—Properties of fresh concrete
Most natural pozzolans produce a cohesive mixture that
maintains a plastic consistency, improving the workability
Typically, natural pozzolans absorb water from the mixture
and hold this water in the system allowing for improved
finishing
Where the available concrete aggregates are deficient in
finer particle sizes, particularly material passing the 75 µm
(No 200) sieve, the use of a finely divided mineral
admix-ture can reduce bleeding and segregation, and increase the
strength of concrete by supplying those fines missing from
the aggregate (ACI 211.1) When an appropriate quantity of
mineral admixture is used to correct such grading
deficien-cies, no increase in total water content of the concrete is
re-quired to achieve a given consistency or slump Drying
shrinkage and absorption of the hardened concrete are not
greatly affected A favorable particle shape, which is not flat
or elongated, and a satisfactory fineness of the mineral
ad-mixture, however, are necessary qualities if a low water
con-tent is to be achieved without use of a water-reducing
admixture For example, coarse pozzolan of poor particle
shape, such as finely divided pumicites, may require an
increase in water content of the concrete for a given
slump This may contribute to increased bleeding and
segre-gation of the fresh concrete
The use of finely divided mineral admixtures having
poz-zolanic properties can provide a major economic benefit in
that the use of these materials permits a reduction in the
amount of portland cement in the mixture For example,
Waugh (1963) reported that the U.S Army Corps of
Engi-neers experienced a major economic benefit through the use
of natural pozzolan; although, aside from a reduction in
wa-ter requirement, other technical benefits had not been
spec-tacular When the ratio of surface area of solids to volume
of water is low, the rate of bleeding is relatively high
More-over, most of the bleeding does not appear at the surface The
aggregate particles settle for a short period until they
estab-lish point-to-point contacts that prevent further settlement
The watery paste continues to bleed within the pockets fined by aggregate particles, leaving water-filled spaces atthe undersides of the particles Therefore, with such mix-tures, bleeding tends to reduce homogeneity of the concrete
de-In extreme cases, the lack of homogeneity is manifested byopen fissures large enough to be easily visible to the nakedeye in a cross section of the concrete under the aggregate par-ticles This lack of bond between paste and aggregate reduc-
es the potential strength of concrete and increasespermeability and absorption
These undesirable effects can be reduced by increasing theratio of surface area of solids to volume of water in the paste.This generally increases the stiffness of the paste and, at agiven slump, effects a wider separation of the aggregate par-ticles in the concrete Increasing the amount of a suitablepozzolan usually increases the ratio of surface area of solids
to volume of water
Natural pozzolans generally increase the cohesiveness ofthe mixture by producing a more plastic paste that allows theconcrete to consolidate readily and flow freely under vibration.The increased cohesiveness also helps to reduce segregation.Natural pozzolans should have physical characteristicsthat allow the portland cement-pozzolan paste to contain amaximum proportion of solid matter and a minimum propor-tion of water This requires that the mineral particles nothave too high a surface area The preferred shape would be asmooth, round particle instead of an irregular, rough-tex-tured particle that would have a higher water demand Thehigh water demand of bentonite, which has a surface areaconsiderably higher than cement, limits the use of that natu-ral pozzolan to smaller percentages than those used in con-ventional concrete mixture proportions
As is the case with other pozzolans, for example, fly ash(ACI 232.2R), the use of natural pozzolan may extend thetime of setting of the concrete if the portland cement content
is reduced The setting-time characteristics of concrete areinfluenced by ambient and concrete temperature; cementtype, source, content, and fineness; water content of thepaste; water soluble alkalies; use and dosages of other ad-mixtures; the amount of pozzolan; and the fineness andchemical composition of the pozzolan When these factorsare given proper consideration in the concrete mixture pro-portioning, an acceptable time of setting can usually be ob-tained The actual effect of a given natural pozzolan on time
of setting may be determined by testing, when a precise termination is needed, or by observation, when a less precisedetermination is acceptable Pressures on formwork may beincreased when concrete containing a natural pozzolan isused if increased workability, slower slump loss, or extendedsetting-time characteristics are encountered
de-2.3—Properties of hardened concrete
Concrete containing a pozzolan typically provides lowerpermeability, reduced heat of hydration, reduced alkali-ag-gregate-reaction expansion, higher strengths at later ages,and increased resistance to attack from sulfates from seawa-ter or other sources than concrete that does not contain poz-
zolan (Mather 1958) Mather (1982) reported that the sulfate
Trang 10resistance of mortar is highest when a silica fume or a highly
siliceous natural pozzolan is used
2.3.1 Strength—The effect of a natural pozzolan on the
compressive strength of concrete varies markedly with the
properties of the particular pozzolan and with the
character-istics of the concrete mixture in which it is used The
com-pressive strength development is a function of the chemical
interaction between the natural pozzolan and the portland
ce-ment during hydration For example, materials that are
rela-tively low in chemical activity generally increase the
strength of lean mixtures and decrease the strength of rich
mixtures On the other hand, cements and pozzolans
contrib-ute to strength not only because of their chemical
composi-tion but also because of their physical character in terms of
particle packing (Philleo 1986) When some pozzolanic
ma-terials of low chemical activity are used to replace cement on
an equal volume basis, early strengths may be reduced
These early strengths can be increased by substituting thepozzolanic material for the cement on an equal mass basis or
a volumetric amount greater than one-to-one for the cementreplaced, provided that the increase in the amount of poz-
zolanic materials does not significantly increase the w/cm so
that the required strength of the concrete is not achieved
A natural pozzolan of high chemical activity, such as takaolin, can sometimes increase early-age strengths, evenwhen used as a replacement for cement, either by an equalmass or by volume in an amount greater than one-to-one forthe cement replaced Caldarone, Gruber, and Burg (1994)compare the compressive strength of a concrete without poz-zolan with concrete containing a highly reactive metakaolin
me-at an addition level of 5 to 10% by mass of cement Figures2.1 and 2.2 show that at all testing ages, the concrete contain-ing this natural pozzolan provided higher compressive
strength than the control (w/cm = 0.38, 0.36, 0.38, and 0.36
compared with 0.41 for the control)
Zhang and Malhotra (1996) report on the physical andchemical properties of RHA, and a total of 10 air-entrainedconcrete mixtures were made to evaluate the effects of theuse of RHA as a cement replacement Their test results indi-cate that RHA is highly pozzolanic and can be used to pro-duce high-performance concrete The test results are shown
in Fig 2.3 through 2.5 Figure 2.3 shows the compressivestrength development of concrete with different percentages
of RHA Figure 2.4 shows the increase of compressive
strength of concrete containing RHA with decreasing w/cm
from 0.50 to 0.31 Figure 2.5 shows compressive strengths ofconcrete with RHA and silica fume compared with that ofcontrol concrete at various ages up to 730 days
It has been shown in Europe and the United States that theintergrinding of pozzolans with portland cement clinker inthe production of blended cements improves their contribu-tion toward strength Results from an investigation of theeffect of curing time on the compressive strength of ASTM
C 109 mortar cubes, made with portland-pozzolan cementscontaining 10, 20, and 30% Santorin earth, are shown inFig 2.6 and 2.7 by Mehta (1981) It is clear from these re-
Fig 2.1—Comparison of compressive strength of
high-reac-tivity metakaolin and silica fume concrete at 5% cement
replacement (Calarone, Gruber, and Burg 1994).
Fig 2.2—Comparison of the compressive strength of
high-reactivity metakaolin and silica fume concrete at 10%
cement replacement (Calarone, Gruber, and Burg 1994).
Fig 2.3—Development of compressive strength of concrete with different percentages of RHA as cement replacement (w/cm = 0.40) (Zhang and Malhotra 1996).
RHA=8%
RHA=10% RHA=15%
Trang 11sults that the contribution of the pozzolan to compressive
strength development occurs sometime after seven days of
hydration
At 28 days, the compressive strength of a concrete with
10% Santorin earth was higher than that of the reference
port-land cement concrete At 90 days, the concrete that used 10
and 20% pozzolan showed compressive strengths higher than
that of the reference portland cement concrete, and at 1 year,
the concrete that used 30% pozzolan was similar to that of the
reference portland-cement concrete, as shown in Fig 2.7 As
shown in Fig 2.8, Massazza and Costa (1979) reported
sim-ilar results on the effect of substituting varying proportions
of portland cement with an Italian natural pozzolan Figure
2.9 compares the compressive strength development of fly
ash concrete and concrete containing a calcined
diatoma-ceous shale natural pozzolan to the compressive strength ofthe control concrete (Elfert 1974)
2.3.2 Sulfate resistance—Use of natural pozzolans with
portland cement in concrete generally increases resistance
to aggressive attack by seawater, sulfate-bearing soil tions, and natural acid waters The relative improvement isgreater for concrete with a low cement content The use of
solu-a pozzolsolu-an with sulfsolu-ate-resistsolu-ant portlsolu-and cements msolu-ay notincrease sulfate resistance and, if chemically active alumi-num compounds are present in the pozzolan, a reduction insulfate resistance of the concrete may result
ASTM C 1012 is a suitable performance test method veloped to evaluate the performance of mortars made withportland cements, blended cements, and blends of portlandcements with fly ash, natural pozzolans, or slags in produc-
de-Fig 2.4—Development of compressive strength of concrete
with different w/cm (RHA content = 10%) (Zhang and
Mal-hotra 1996).
Fig 2.5—Development of compressive strength of concrete
with RHA and silica fume (w/cm = 40) (Zhang and
Mal-hotra 1996).
Fig 2.6—Effect of curing time on compressive strength of mortar cubes up to 28 days made with portland-pozzolan cements containing Santorin earth (Mehta 1981).
Fig 2.7—Effect of curing time on compressive strength of mortar cubes up to 12 months made with portland-pozzolan cements containing Santorin earth (Mehta 1981).
0 5 10 15 20 25 30 35
Age, days
Portland Cement 10% Pozzolan 20% Pozzolan 30% Pozzolan
Trang 12ing a sulfate-resisting cement mortar (Patzias 1987) A series
of ASTM C 1012 tests with 20 cements and blends of Type
I with Class F fly ash, Santorin earth, and silica fume showed
that blended cements containing highly siliceous natural or
artificial pozzolans, slags, or silica fume had better sulfate
resistance than portland cements having the same C3A
con-tent as calculated by the Bogue method (Fig 2.10) (Patzias1987)
An extensive research program at the USBR assessedvarious natural pozzolans for sulfate resistance (Elfert1974) Figure 2.11 shows the results of accelerated tests in2.1% sodium sulfate solution to predict the service life of var-
Fig 2.8—Effect of substituting Italian natural pozzolan for
portland cement on compressive strength of ISO mortar
(Massazza and Costa 1979).
Fig 2.9—Effect of pozzolan on compressive strength of crete (Elfert 1974).
con-Fig 2.10—ASTM C 1012 sulfate resistance results comparing blended cements and land cements having same C 3 A content as calculated by Bogue method (Patzias 1987).