Waste Manage Res 2002: 20: 279–289
Printed in UK – all rights reserved
The production of synthetic aggregate from a
quarry waste using an innovative style rotary kiln
The large volumes of wastes generated by industrialised
society has led to efforts to find practical uses for these
wastes, whilst also offsetting the consumption of natural
resources. This paper describes the use of an innovative
rotary kiln to produce synthetic aggregates from a variety
of waste streams. The main waste used was a quarry fines
which was blended with either paper sludge, clay, or a
dredged harbour sediment. The different combinations
were extruded and fired in the kiln to produce a material
suitable for natural aggregate replacement. Two of the
synthetic aggregates produced were tested by incorpora-
tion in to concrete as coarse aggregate replacement. The
concrete 28-day compressive strengths achieved were
above 40 N mm–2 and compared favourably with control
concretes made with natural aggregates and a commer-
cially available lightweight aggregate (Lytag). Leaching
tests have also been carried out to assess the potential
environmental impact of utilisation. Although not
finalised, these tests have also given favourable results.
Introduction
The extraction of aggregates, coal and other mineral
resources produces large volumes of waste material. These
wastes arise due to a number of factors including; conta-
mination of the natural resource, the
existence of local fault zones and overburden, and the
processing of the mineral resource itself (Rockliff 1996,
Woolley 1994). The composition of the waste depends on
the mineralogy of the source rock and on any associated
sediments but will predominantly consist of silica (SiO2),
aluminium oxide (Al2O3), iron oxide (Fe2O3) and alkali
metal (K, Na, Ca, Mg) oxides (Woolley 1994). The pro-
duction of natural aggregates from sand and gravel
deposits produces a waste consisting largely of silt and clay
Downloaded from wmr.sagepub.com at UCSF LIBRARY & CKM on April 10, 2015
Copyright © ISWA 2002
Waste Management & Research
ISSN 0734–242X
P. J. Wainwright
D. J. F. Cresswell
School of Civil Engineering, University of Leeds, Woodhouse
Lane, Leeds, LS2 9JT, United Kingdom
H. A. van der Sloot
Soil and Waste Research, ECN-BCM, P.O.Box 1, Westerduin Weg
3, 1755 ZG Petten, Petten, Netherlands
Keywords – Synthetic aggregate, quarry waste, waste-paper
sludge, contaminated dredgings, concrete, rotary kiln,
leaching, wmr 392–4
Corresponding author: P. J. Wainwright, School of Civil
Engineering, University of Leeds, Woodhouse Lane, Leeds, LS2
9JT, United Kingdom
Received 27 March 2000, accepted in revised form 03 December
2001
slurry, whereas hard rock quarries generate a waste which
ranges in size from a very fine dust to a coarse sand. The
amount of waste produced will depend on the geology of
the natural resource and the methods used for extraction
and processing.
Similar waste streams to those referred to above, in terms
of mineralogical composition and grain size, are materials
dredged from waterways used for navigation and sediments
extracted to prevent the pollution of water resources. The
dredged material is generally of clay to sand in its size dis-
tribution and is composed predominantly of rock forming
minerals. The fine-grained dredged material may be conta-
minated with heavy metals and organic compounds, espe-
cially if located near to or downstream from industrial areas.
Waste Management & Research 279
P. J. Wainwright, D. J. F. Cresswell, H. A. van der Sloot

Volumes arising
Quarry and mining waste
The total volume of quarry and mining wastes produced in
Europe in 1995 according to the European Environment
Agency (EEA) (1998), was estimated at 700 million
tonnes (mt). In the UK the 1996 figure was estimated to
be 74 mt, with 30 mt of this arising from the quarrying of
hard rock, sand and gravel (DETR, 1998).
Dredged material
The volumes of dredged materials, although somewhat
smaller than the quarrying and mining wastes, are still sig-
nificant. For the five EU nations of Belgium, France,
Germany, Netherlands and the UK the volume of dredged
material when added together is in excess of 130 million
tonnes per annum (OECD 1997, DETR 1997, DETR
1998, BMV 1991, AKN 1990, Dearnaley et al. 1995).
Current disposal and management methods
Quarry and mining waste
The current management of quarry and mining waste is
relatively simple in that a large proportion of it is used on
site for the production of embankments to decrease the
visual intrusion associated with the extraction industries.
Other disposal methods include: the filling of voids left by
extraction processes and the creation of above ground tips
(spoil heaps) and tailings ponds (settling lagoons) that are
then landscaped (DoE 1995). The DoE (1995) also esti-
mates that only about 5% of quarry and mining waste was
used for purposes other than on site use, this figure may
well have increased in the last 5 years, but there are still
large volumes in stockpiles and lagoons.
Dredged material
The potential for the re-use of dredged material is very
much dependant on its particle size, as any contaminants
present tend to adhere to the clay sized fraction.
A number of physical processing technologies exist (see for
example Köthe 1995) to separate the coarser size fractions
(ie. above 200 microns). Such material has been re-used
(albeit in small quantities) for example, as a fine aggregate
in concrete, mortar and for other construction purposes.
Unfortunately, often less than 30% (van Gemert et al.
1988) of the dredged material is of this size range. The
finer (potentially more heavily contaminated) fraction
is dumped in sites where the risk of water resource conta-
mination is reduced, or is used for land reclamation
purposes.
The use of wastes and the production of synthetic
aggregates
The utilisation, in construction (or other industries), of
quarry and mining wastes, and river/harbour dredgings etc.
is important to allow for the most efficient use and preser-
vation of mineral resources. Products that can incorporate
wastes to produce high quality construction materials will
have the largest impact upon reducing resource depletion.
The production of lightweight (or synthetic) aggregates
from certain waste streams and virgin materials is a well-
established process (Owens 1993). For example, sintered

Table 1. Some commercially available lightweight aggregates (UK) {adapted from Owens 1993}

Commercial Raw material Manufacturing process Shape Typical dry Typical concrete strength
name loose bulk (N mm–2)
density (kg m–3)

Foamed slag Foamed Foaming bed Angular 750 <40

slag

Leca/Fibo Expanded Rotary kiln Rounded 425 <30

clay

Lytag Sintered Sinter strand Rounded 825 >40

pfa

Pellite Blastfurnace Pelletisation Irregular 900 >40

slag

Liapor Expanded Rotary kiln Rounded 650 >40

shale

280 Waste Management & Research

Downloaded from wmr.sagepub.com at UCSF LIBRARY & CKM on April 10, 2015
 The production of synthetic aggregate from a quarry waste using an innovative style rotary kiln

pulverised fuel ash (pfa) and expanded clay/shale are used
to produce aggregates sold in North Western Europe under
the trade names of ‘Lytag’ and ‘Fibo’ respectively.
Examples of other commercially available lightweight
(synthetic) aggregates are shown in Table 1. The use of
lightweight aggregates is highlighted by Price (1995) who
described the use of one such aggregate, Lytag, to produce
a high quality concrete used in the construction of oil field
base tanks for operation in the North Sea. Much research
has also been carried out into the production of synthetic
aggregates from other waste streams. For example Osada
(1997) studied the melting and crushing of municipal solid
waste incinerator (MSWI) bottom ash. Yasuda (1991) car-
ried out similar work on incinerated sewage sludge ash.
Wainwright and Boni (1981, 1983) and Wainwright and
Robery (1991) reported on the use of MSWI bottom ash
to produce a synthetic aggregate by grinding, pelletising
and firing in a rotary kiln. The aggregate produced was
incorporated into concrete. This concrete was tested over
a 4 1/2 year period with satisfactory results and compres-
sive strengths upto 50 N mm–2.
There has been little published work concerning the
utilisation of quarry and mining wastes or dredgings as syn-
thetic aggregates. This paper reports on the initial stages of
research into the production of aggregate from a granitic
quarry waste, using a number of different wastes and low-
grade materials as a binder.
Experimental work
Materials
The following materials were used in the study.
Table 2. Major element composition (as oxide) of Glensanda granite
washings
Oxide % different aggregates including Dutch sourced harbour
Silicon SiO2 66
Aluminium Al2O3 17
Iron Fe2O3 3 matrix. The mobility of Elements such as Cr, V, Mo may
Magnesium MgO 1 however be increased.
Calcium CaO 2
Sodium Na2O 3 Binders
Table 3. Particle size distribution of Glensanda granite washings
Sieve size (microns) Cumulative % passing
300 99.5
150 96.0
75 82.8
Glensanda granite washings filter cake
This material arises from the washing of crushed stone
coarse aggregate and sand from the Glensanda super quar-
ry off the West Coast of Scotland. The quarry produces
upto 10 million tonnes of aggregate per year and 1 million
tonnes (dry weight) of the washings cake. At present this
material is either stored in a lagoon on site or stockpiled at
an off site processing plant in the south east of England.
The major oxide composition of the washings is shown in
Table 2 and the grain size distribution in Table 3.
Dutch sourced river dredgings
This dredged material arises due to the removal of
contaminated sediment from a river harbour in the
Netherlands. The material is fine grained with more than
90% passing 100 microns, Table 4 shows the major ele-
mental analysis and the concentration of the contaminat-
ing elements. The sediment has been well characterised as
it poses a large disposal problem due to heavy metal cont-
amination. The dredgings are thermally treated to a point
where they undergo partial melting and are then cooled
rapidly. This will incorporate some of the contaminating
elements into a silicate matrix, making them less available
to pollute the immediate environment.
The extent to which this incorporation of contaminants
occurs can be measured by conducting leaching tests on
the aggregate. This has been undertaken on a number of
dredgings (van der Sloot et al. 2000). Results show that a
number of elements are incorporated into the silicate

Potassium K2O 5
• Paper sludge from a paper de-inking plant, this materi-
al is contaminated with heavy metals such as As, Cd,
Phosphorous P2O5 0.2
Cr, Cu, Mo, V and Zn.
Titanium TiO2 0.5
• Low-grade clay, a carbon rich material obtained from a
Manganese MnO 0.1
‘ball clay’ quarry.

Waste Management & Research 281

Downloaded from wmr.sagepub.com at UCSF LIBRARY & CKM on April 10, 2015
P. J. Wainwright, D. J. F. Cresswell, H. A. van der Sloot

Table 4. Composition by metal oxide of the Arnhem harbour dredgings, Netherlands

Metal mg kg–1 {% as oxide} Metal mg kg–1 {% as oxide}

Lithium 54.00 Iron 37,980 {5.4}

Boron 81.55 Cobalt 17.65
Sodium 4,43 {1.2} Nickel 70.00
Magnesium 10,108 {1.7} Copper 250.40
Aluminium 54,393 {10.0} Zinc 2,392
Silicon 208,421 {45.0} Strontium 217
Phosphorous 3,484 Molybdenum 22.25
Sulphur 3,814 Cadmium 20.20
Potassium 16,590 {2.5} Tin 2,955
Calcium 40,481 {5.7} Barium 3,334
Titanium 3,632 Lead 1,738
Vanadium 205.80 Fluorine 681.5
Chromium 305.00 Chlorine 3,387
Manganese 958.50

• Dutch sourced river dredgings. The fine-grained nature
of this sediment means that it is cohesive and as such
makes a good binder.
Aggregate production
The Trefoil Kiln
This project is specifically aimed at using the innovative
design of the Trefoil kiln, which is much more thermally
efficient than traditional rotary kilns. Traditional kilns are
composed of a thick steel drum lined with heavy manual-
ly laid fire-bricks, which have a limited life and which need
regular maintenance. They are extremely thermally
inefficient and very slow to respond to firing adjustments,
which can make it difficult to fire material either consis-
tently or quickly. By contrast, the Trefoil kiln has been
described by Parks and Sherwen (1986) as, "a significant
advance in rotary kiln technology". The kiln compresses a
ceramic fibre insulation between a thin steel alloy hot face
and a tensioned thin mild steel cold face to form a car-
tridge with a triple-lobed cross section (see Fig. 1) to
achieve the following:
• remarkable thermal efficiency (an inner kiln tempera-
ture of 1200°C and an outer temperature of 60°C);
• a much smaller and lighter structure which does not
need costly maintenance or relining and which is easier
to manufacture, transport, install, remove and replace;
• much faster heat-up and cool-down periods (50 times
quicker than conventional kilns);
• vastly improved responsiveness (operating temperatures
can be increased or decreased in minutes);

Table 5. Mix ratios used for the production of green pellets by

extrusion and ‘balling’

Mix Code Glensanda Washings Binder / Additive

Dry weight % Dry Weight %

Glensanda 01 100% -----------

Glensanda 02 95% 5% Clay
Glensanda 03 90% 10% Clay
Glensanda 04 87% 13% Clay
Glensanda 05 83% 17% Clay
Glensanda 06 90% 10% Dutch River Dredgings
Glensanda 07 95% 5% Paper Sludge
Glensanda 08 90% 10% Paper Sludge
Glensanda 10 85% 15% Paper Sludge
Fig. 1. Cross-section of the Trefoil rotary kiln

282 Waste Management & Research

Downloaded from wmr.sagepub.com at UCSF LIBRARY & CKM on April 10, 2015
 The production of synthetic aggregate from a quarry waste using an innovative style rotary kiln
Table 6. Concrete mix design ratios by weight
Glensanda 03 Glensanda 06
Coarse aggregate 2.51 2.00
Fine aggregate 2.23 2.23
Ordinary Portland cement 1 1
Water : cement 0.53 0.53
Slump (mm) 140 150
• higher load factors; and
• greater bed-load mixing allowing reduced thermal
gradient in the bed and improved firing consistency.
Although a traditional rotary kiln could be used to pro-
duce the synthetic aggregates used here, the benefits of the
Trefoil Kiln are clear and as such are much more likely to
make an impact on the treatment of mineral by-products
to produce a consistent and economically viable product.
Extrusion and ‘balling’
The first stage in the aggregate production process was to
blend the quarry waste with a binder to enable the mixture
to be extruded and balled (pelletised). In this study the
binders used were clay, river dredgings and paper sludge.
The proportions of the different materials used are shown
in Table 5. Once blended the mixtures were extruded and
pelletised by means of a rotary drum pelletiser that also
incorporated a burner/drier. The combined action of the
drier and the rolling motion of the drum produced a skin
on the pellets that aided green pellet strength and created
a thin denser outer skin on the fired pellet. The
production of aggregate from the contaminated Dutch
sourced river dredgings was also investigated. However,
the large clay content meant that this material was
difficult to pelletise using the machinery that was avail-
able. It was therefore used only as a binder. Hand-rolling
was used to produce a small sample of green pellets from
the dredgings, with no binder or additives, to establish the
firing characteristics of the dredgings and the potential
physical characteristics of the aggregate produced.
Aggregate firing
Firing of the synthetic aggregate consisted of two stages;
the first is known as the ‘burnout’ stage where the carbon
content of the ash/binder mix is allowed to combust
(burnout) at a temperature below that where the inert
Downloaded from wmr.sagepub.com at UCSF LIBRARY & CKM on April 10, 2015
Lytag Control
2.1 3.64
2.23 2.23
1 1
0.53 0.53
10 20
material in the pellet begins to sinter. From previous expe-
rience it was known that this stage can take between 5 and
40 minutes and occurs between 700°C and 800°C,
depending on the amount and type of carbon (organic
material) present and the internal structure of the pellet.
The carbon content of a lightweight aggregate produced
from pfa needs to be below 4% to conform to BS 3797 (BSI
1990) and it was considered appropriate to adopt this fig-
ure in this study. To ensure that this 4% level was reached
before the sintering stage begins, samples of the aggregate
were taken at various times during burnout and analysed
for the carbon content. It was assumed that the remaining
carbon would be burnt during further firing, providing a
volatile material for the production of gases during sinter-
ing, this may form voids helping to give a lightweight struc-
ture to the pellet. The second stage is the sintering of the
green pellets at temperatures between 1200°C and
1230°C, to fuse the particles of the granitic filter cake into
a single mass. All aggregates produced were tested for
relative density and water absorption in accordance with
BS 812 (BSI 1975).
Casting
Two of the aggregates (Glen 3 and Glen 6) were produced
in larger quantities and were cast in concrete as a 100%
replacement of the coarse natural aggregate. The concrete
mix designs (Table 6) were altered on a saturated surface
dried loose bulk density basis to keep equal volume ratios
of coarse aggregate, fine aggregate and Portland cement
(PC). The synthetic aggregates were soaked in water and
drained prior to casting to ensure they were in a saturated
surface dried condition before mixing. Once mixed, the
concrete was cast in 100 mm cubes and kept at 20°C and
100% relative humidity for compressive strength testing.
Concrete was also cast using natural coarse and 8–12 mm
Lytag. These were used to compare the performance of the
synthetic aggregate concrete.
Waste Management & Research 283
P. J. Wainwright, D. J. F. Cresswell, H. A. van der Sloot

Leaching leaching trials on some of the aggregates have been con-

A number of questions have to be asked to assess the full
potential environmental impact of using waste materials to
produce synthetic aggregate. These are:
• What are the leaching characteristics of the waste and
the synthetic aggregate?
• Which leached elements are of concern?
• How do the raw (waste) materials translate into the
environmental quality of the aggregate?
• What is the relationship between the physical proper-
ties and leaching behaviour of the aggregate?
• How consistent is the aggregate quality?
• How does aggregate behave in a cement matrix?
• How to assess the environmental quality quickly?
• What is the effect of different utilisation scenarios
(stockpile, loose fill, aggregate in concrete, end of life)
on contaminant release?
To answer these questions a number of different leach-
ing test methods have to be adopted. These may include
pH dependence leaching test (CEN TC 292), column
leaching test (NNI, 1995), batch leaching test after differ-
ent contact times, CEN TC 154 procedure, concise leach-
ing tests (Kosson et al. 1997) for quality control purposes
and a tank leaching test for monolithic materials (NEN
7345), such as concrete. The application of these leaching
tests are described by van der Sloot et al. (1997). So far
only initial waste characterisation tests and pH-stat tests
on the aggregate have been fully conducted. Column
ducted according to the Dutch NEN 7343 procedure
(NNI 1995).
During the pH-stat procedure extractions were taken
from a liquid:solid ratio (L S–1) 10 (l kg–1) over 48 hours at
set pH values.
In the column leaching test a number of leachate sam-
ples were collected at a range of L S–1 from 0.1 – 10 l kg–1.
The leachate was passed over the unbroken aggregate in
an upward flow direction, through a column 280 mm high
and 100 mm in diameter.
Results
Aggregate properties
The densities and water absorption figures for all the
Glensanda synthetic aggregates produced in this study are
shown in Table 7 where they are compared with natural
aggregate and with Lytag. It is not possible to show results
of synthetic aggregates made using a quarry waste in a
traditional kiln, as these are not commercially available.
The proportion and type of binder used with the granite
washings does appear to have an effect on the properties of
the resulting aggregate. The relative densities range from
1.63 to 2.23 with the majority being above the figure of 1.8
for Lytag and all lying below the figure of 2.6 for natural
aggregate. The results for water absorption and loose bulk
density show similar trends with most values lying between
those for Lytag and natural aggregate. To be classified as a

Table 7. Physical properties of synthetic aggregates made using the Trefoil kiln

Aggregate 24 Hour water absorption Relative density Loose bulk density

kg m–3
(% dry mass) (saturated surface dry) (saturated surface dry)

Glensanda 01 1.58 2.12 1,080

Glensanda 02 1.90 2.09 1,067
Glensanda 03 2.91 1.8 913
Glensanda 04 2.42 2.00 1,108
Glensanda 05 1.37 1.84 1,022
Glensanda 06 6.37 1.63 724
Glensanda 07 1.95 2.23 1,208
Glensanda 08 2.90 2.21 1,180
Glensanda 09 6.19 2.10 1,129
Lytag 12.3 1.8 875
Natural coarse 0.8 2.6 1,520
Hand-rolled Dutch dredgings 27 1.12 --

284 Waste Management & Research

Downloaded from wmr.sagepub.com at UCSF LIBRARY & CKM on April 10, 2015
 The production of synthetic aggregate from a quarry waste using an innovative style rotary kiln
lightweight aggregate the loose bulk density of the materi-
al needs to be below 1000 kg m–3 and the results show that
the majority of the aggregates produced in this study just
fail to meet this requirement.
The only difference between aggregate types Glensanda
1 to 5 is the percentage clay, which ranges from 0%
(Glensanda 1) to 17% (Glensanda 5). The results show no
clear trend between clay content and any of the three
aggregate properties measured. Where water absorption is
concerned the results would indicate optimum clay con-
tent for a maximum value of water absorption. Ideally
however one is looking for a minimum value of water
absorption, which in this case occurs at the maximum clay
content but is only just below the figure of 1.58% for zero
clay content.
By comparing Glensanda aggregates 3, 6 and 8, all of
which have a 10% binder content, the influence of binder
type can be seen. The difference between clay and paper
sludge appears to be limited, the aggregates have the same
water absorption figures of 2.9% and slightly different rel-
ative density values (1.8 cf 2.21 respectively). Using the
Dutch sourced river dredgings as the binder (Glensanda 6)
produces a significant increase in water absorption (6.37%
cf 2.9%) but a reduction in relative density (1.63 cf 1.8
and 2.21). This suggests that the pore structure in
Glensanda 6 is more open than in the other two, particu-
larly that at or near the surface of the aggregate particles.
Ideally an aggregate with low water absorption and low rel-
ative density would be desirable which is the case with
both clay and paper sludge as the binder. Increasing the
content of paper sludge from 5% through to 15%
(Glensanda 7,8,9) results in an increase in water absorp-
tion and a corresponding reduction in relative density.
When comparing all the synthetic aggregates produced
in this study with Lytag it can be seen that although the
majority of them have higher relative and loose bulk den-
sities the differences are only small, yet at the same time
they all have significantly lower water absorption figures.
The highest value of water absorption was recorded with
Glensanda 6 (90% granite fines, 10% river dredgings) but
even this figure of 6.37% is almost 50% lower than the fig-
ure of 12.3% for Lytag. At the same time the bulk density
and relative density of this aggregate (Glensanda 6) are
slightly lower than those of Lytag (724 cf 875 and 1.63 cf
1.80 respectively). The lower values of water absorption
are most likely due to the smooth closed surface texture of
the synthetic aggregates compared to Lytag which is more
Downloaded from wmr.sagepub.com at UCSF LIBRARY & CKM on April 10, 2015
porous and open textured. These results are encouraging
when bearing in mind the good track record that Lytag has
when used as an aggregate in concrete. The reason for the
decrease in density of the Glensanda 6 aggregate is due to
the ‘bloating’ (Riley 1951) nature of the dredgings.
The Dutch river dredgings that were pelletised by
‘hand-rolling’ were also fired and the relative density and
water absorption results are also shown in Table 7. The
reason for this reduced density is the bloating characteris-
tics of the Dutch sourced harbour sediments and is
suspected of having little to do with being ‘hand rolled’.
The limited data show very high water absorption for this
aggregate. If this water absorption can be decreased and
the relative density maintained, then the Dutch sourced
harbour dredgings are potentially an ideal candidate for
lightweight aggregate production.
It must be remembered that all the tests so far report-
ed on the aggregates were performed on only small samples
(no more than 4 kg for each aggregate type – 12 kg for
samples 3 and 6). It would be preferable to use larger sam-
ples for testing, and the results presented here may not
therefore be truly representative of the aggregate proper-
ties. However, the Glensanda material is very homoge-
neous and the firing conditions are well established and
repeatable.
The two synthetic aggregates chosen for larger scale
production for subsequent use in concrete were Glensanda
3 and 6 made with 10% clay and 10% river dredgings
respectively (see Table 5).
Concrete workability
The workabilities of all four mixes made to nominally the
same water/cement ratios of 0.53 are shown in Table 6. All
mixes, with the exception of the control mix, had similar
densities and it is reasonable therefore to make compar-
isons between the Lytag mix and those two mixes made
with the synthetic aggregates manufactured in this study.
It can be seen that using either of the synthetic aggregates
as a coarse aggregate replacement leads to a significant
increase in workability when compared with Lytag (140
and 150 mm cf. 10 mm). This is due not only to the
smoother more closed surface texture of these aggregates
referred to above but also to their particle shape which is
much more rounded than Lytag.
Compressive strength
It is only possible at this stage to present strength results
Waste Management & Research 285
P. J. Wainwright, D. J. F. Cresswell, H. A. van der Sloot

Leached (L/S=10, 24 hrs, mg kg–1)

Fig. 2. Compressive strength of concrete made with synthetic and
natural aggregates Fig. 3a. Total organic carbon data obtained from the paper de-inking
sludge
up to an age of 28 days and these are shown for all four
mixes in Fig. 2. As can be seen, there is little difference in
performance between any of the mixes; at 28 days
strengths range from 40 to 48 N mm–2 with the highest Leached (L/S=10, 24 hrs, mg kg–1)
and lowest results being achieved by the two mixes made
with the synthetic aggregates. The fact that all mixes were
made to nominally the same free water/cement ratio, and
that they all have similar strengths suggests that at this
stage the strength of the concrete is being controlled by
the strength of the cement paste, and not by the strength
of the aggregate. Clearly longer term strength data are
required to determine whether or not there are any dele-
terious reactions taking place between the aggregate and
the cement paste which might lead to some degradation in
strength. The results to date though are encouraging and Fig. 3b. Total inorganic carbon data obtained from paper de-inking
sludge
suggest that the aggregates could be used to produce
medium strength structural grade concrete. It must also
be remembered that had the concretes been made to
Leached (L/S=10, 24 hrs, mg kg–1)

nominally the same workability the water/cement ratios of

the two mixes made with the synthetic aggregates
produced in the Trefoil kiln could have been reduced. This
may have led to an increase in strength, providing that the
strength of the paste did not exceed that of the aggregate.

Leaching
A brief review of the data obtained from the waste mater-
ial by characterisation leaching tests are given, followed by
a comparison of the produced aggregate and the starting
materials The pH dependence test is used mainly for this
purposes, as it allows important changes in materials
leaching behaviour to be identified, which are not obvious Fig. 3c. Molybdenum leaching data obtained from pH-stat test of the
paper de-inking sludge
from dynamic leaching tests, such as a column or a tank
leach test. The methods allows to identify potentially crit- – – Dashed lines show Dutch limits for building materials, class 1 and 2

286 Waste Management & Research

Downloaded from wmr.sagepub.com at UCSF LIBRARY & CKM on April 10, 2015
 The production of synthetic aggregate from a quarry waste using an innovative style rotary kiln

ical elements as the entire pH range relevant to environ-

mental assessment for use as unbound aggregate (pH

Leached (L/S=10, 24 hrs, mg kg–1)

range 5 – 10) and as bound aggregate (pH 8 – 13) are
covered. Even for bound aggregate, the evaluation of
environmental properties is not limited to intact products
testing as the recycling stages and "end of life" scenarios
are also relevant (van der Sloot 2000).

Paper sludge
The paper de-inking sludge is characterised by a very high
Total Organic Carbon (TOC) content (approximately
10,000 mg kg–1) and a relatively high Total Inorganic
Carbon (TIC) content (approximately 1,000 mg kg–1 at
neutral pH) as compared to other starting materials, these
are shown in Figs. 3a and 3b respectively. The elements
Se, Mo, Ba and Cu only exceed critical limits as given by Fig. 5a. Strontium leaching obtained by pH-stat from Glensanda
the Dutch Building Materials Decree (1995) at pH’s washings (labelled washings) and Glensanda 8 aggregate (labelled
Glensanda)
greater than 10, Fig. 3c shows this for Mo. The material
has a strong buffering capacity, as 2 mmol g–1 of acid was
needed to increase the pH by one unit.
Leached (L/S=10, 24 hrs, mg kg–1)

Dutch river dredgings

For the elements Ni, Co and Zn a change in the pH of
one unit in either direction of its own pH of 7.3 leads to
significant changes in leachability, Fig. 4 illustrates this
trend for Nickel. The increase towards higher pH would
be critical in cement-based application of the untreated
material. However, after sintering to aggregate the dis-
Leached (L/S=10, 24 hrs, mg kg–1)

Fig. 5b. Chromium leaching obtained by pH-stat from Glensanda

washings (labelled washings) and Glensanda 8 aggregate (labelled
Glensanda)

solved organic carbon (DOC) causing this mobilisation

at high pH in the untreated material is destroyed and
metals are largely incorporated in the silicate matrix, thus
reducing leachability substantially.

Glensanda washings and synthetic aggregate

Fig. 4. Nickel leaching data (Ni) from pH-stat test. Duplicate (Duplo) The leaching data set obtained allows a comparison to be
data confirms the test results made between the produced aggregate (coded Glensanda

Waste Management & Research 287

Downloaded from wmr.sagepub.com at UCSF LIBRARY & CKM on April 10, 2015
P. J. Wainwright, D. J. F. Cresswell, H. A. van der Sloot
Leached (L/S=10, 24 hrs, mg kg–1)
Fig. 6a. Chromium leaching obtained by pH-stat test of the Glensanda
6 aggregate
Leached (mg kg–1)
L/S (I kg–1)
Fig. 6b. Chromium leaching obtained by column leaching of the
Glensanda 6 aggregate
7) and the initial waste materials (Glensanda granite
washings and paper de-inking sludge). Several elements –
Li, Ca, K, Mn, Sr and Ba are incorporated in the silicate
matrix in the sintering process (see Fig. 5a for Sr leaching
data). Most other metals such as Cr and V show little dif-
ference or are somewhat increased (see Fig. 5b for Cr)
compared to the raw Glensanda washings.
The column leaching test data undertaken on intact
aggregate are similar, at a liquid to solid ratio of 10
(L S–1=10) to the results of the pH-stat data (L S–1=10)
288 Waste Management & Research
Downloaded from wmr.sagepub.com at UCSF LIBRARY & CKM on April 10, 2015
(Figs. 6a and 6b show this). If the trend of 6b is continued
to L S–1=10 the leached concentration is similar to the fig-
ures obtained from the pH-stat test data shown in Fig. 6a.
This suggests a fairly rapid exchange between interior
particle surfaces and the bulk solution and implies that
for environmental impact assessment size reduction is
adequate to complete a test in 24 hours. A longer leaching
time is needed if intact pellets are leached (van der Sloot
et al.. 2000). Further work is in progress on assessing the
behaviour of aggregate in a bound form.
Conclusions
The work presented in this paper is limited in scale but is
broad in scope, dealing with the processing of waste mate-
rials, the production of synthetic aggregates, and study of
the physical and environmental performance of these
aggregates. The following conclusions cover the main
findings from the work and suggest areas for further
research:
• It was possible to process by extrusion and balling bina-
ry mixes made mainly from quarry waste, with smaller
volumes of clay, paper sludge, and harbour dredgings, to
act as a binder;
• The processed pellets were fired in a rotary kiln to
produce synthetic aggregates with slightly higher
relative densities but significantly lower water absorp-
tion compared to a commercially available synthetic
aggregate (Lytag);
• The 28-day compressive strengths of concretes made
from these synthetic aggregates were in excess of 40 N
mm–2 and compared favourably to purpose-made con-
trol concretes using natural aggregates and Lytag; and
• The initial data obtained on the leaching of contami-
nants from the synthetic aggregate have to date given
favourable results.
It is clear from these results that concretes with reason-
able strengths can be made with synthetic aggregates made
from the different waste streams investigated but more
work is needed to obtain the optimum mix proportions (in
terms of binder contents) and firing conditions to obtain
the optimum performance from the aggregates. Although
the strength results obtained to date are satisfactory, much
longer-term data are needed to determine whether or not
there are any potentially deleterious reactions taking place
between the aggregates and the cement paste. The work
 The production of synthetic aggregate from a quarry waste using an innovative style rotary kiln
programme currently underway is designed to address
these questions as well as providing more detailed infor-
mation on the leaching characteristics of the synthetic
References
AKN (Arbeitskreis Nassbaggerei der Wasser- und Schifffahrtverwaltung des
Bundes). (1990) Jahresbaggermengen 1990 in den Bundesstrassen des
Kuestengebiets (alte Bundeslaender) (Working Group on dredging of
the Federal Waterway and Shipping Administration: Annual volumes
of dredged material 1990 in Federal Waterways in Coastal Waters – old
Federal States).
Building Materials Decree. (1995). Staatsblad van het Koninkrijk der
Nederlanden, 567.
BMV Bundesministerium fuer Verkehr (1991) Jahresbaggermengen in den
Bundeswasserstrassen (alte Bundeslaender) – Erlass BW 15/52.06.00
vom 09.07.1991; Bonn. (Federal Ministry of Transport, 1991: Annual
volumes of dredged material from Federal Waterways – old Federal
States).
BSI (British Standards Institute) (1975) Methods for the Testing of
Aggregates for use in Concrete: Part 2. The Determination of Relative
Density and Water Absorption. BS 812.
BSI (British Standards Institute) (1990) Lightweight Aggregate for use in
Masonry Units and Structural Concrete. BS 3797.
Dearnaley, M. P., Burt N. T., Carpenter, K., Nottage, A. S. & Stevenson, J. R.
(1995) Beneficial Use of Muddy Dredged Material. In: WODA (World
Organisation of Dredging Associations) 14th World Dredging Congress –
Dredging Benefits. 1, 333–344, Amsterdam, The Netherlands.
DETR (Department of the Environment, Transport, and the Regions) (1997)
Digest of Environmental Statistics number 19. London, HMSO.
DETR (1998) Digest of Environmental Statistics number 20. London, HMSO.
DoE (Department of the Environment and the Welsh Office) (1995)
Making Waste Work – A strategy for sustainable waste management in
England and Wales. London:HMSO.
EEA (European Environment Agency) (1998) Europe’s Environment: The
Second Assessment. Office for official publications of the European
Communities, Luxembourg. Elsevier Science Ltd.
Kosson, D. S. & Van der Sloot, H. A. Integration of Testing Protocols for
Evaluation of Contaminant Release from Monolithic and Granular
Wastes. In: Goumans, G.J. Senden & van der Sloot H.A.(ed.) Waste
Materials In Construction – Putting Theory into Practice. Studies in
Environmental Science 71. Elsevier Science Publishers, Amsterdam,
1997, 201–216.
Köthe, H. (1995) Criteria for beneficial use or the disposal of contaminated
dredged material in the Federal Republic of Germany. Central dredging
Association. In: WODA (World Organisation of Dredging
Associations) 14th World Dredging Congress – Dredging Benefits. 1,
459–474, Amsterdam, The Netherlands.
NNI (1995) Leaching characteristics of solid earthy and stony building and
waste materials; Leaching tests; Determination of the leaching of inor-
ganic components from granular materials with the column test. Dutch
standard NEN 7343.
OECD (1997) Environmental Data Compendium. Paris, OECD.
Osada, M. (1997) Direct Melting Process for MSW Recycling. In: Waste
Downloaded from wmr.sagepub.com at UCSF LIBRARY & CKM on April 10, 2015
aggregates. Future planned work will also look at other
properties of concretes made with these aggregates such
as: permeability, elasticity, shrinkage and creep.
Materials in Construction: Putting Theory into Practice. Goumans, J. J. J.
M.et al. (eds). Elsevier. Amsterdam. pp73–78.
Owens, P. L. (1993) Lightweight Aggregates for Structural Concrete. In:
Clarke, J. L. (eds.) Structural Lightweight Aggregate Concrete. London,
UK. Blackie Academic and Professional. pp 1–18.
Parks, L. F. & Sherwen, R. G. (1986) The Use of New Materials in an
Improved Design of Rotary Kiln. Materials & Design 7, 252–55.
Price, W. (1995) Lightweight Concrete for the North Sea: the BP Harding
(Forth) Field Gravity Base Tank. Quarterly Journal of the Federation de
la Precontrainte 2, 3–7.
Riley, C. M. (1951) Relation of Chemical Properties to the Bloating of Clays.
Journal of the American Ceramic Society. 34, 121–128.
Rockliff, D. (1996) Low-Grade quarry products, reclaimed aggregates and
inert wastes. Their use in unbound mixtures for road pavements. Waste
Management and Research 16, 83–85.
van der Sloot, H. A., Heasman, L. & Quevauviller, Ph. (eds.) (1997)
Harmonization of Leaching/Extraction tests. Studies in Environmental
Science. 70. Amsterdam. Elsevier Science. Amsterdam, p. 292.
van der Sloot, H. A., Hoede, D., Cresswell, D. J. F. & Barton, J. R. (2000)
Leaching behaviour of synthetic aggregates. In: Woolley, et al. (eds.)
Waste Materials in Construction: Science and Engineering of Recycling
for Environmental Protection. Waste Management Series, Vol. 1.
Pergamon. p.695.
van der Sloot, H. A. (2000) Comparison of the characteristic leaching
behaviour of cements using standard (en 196-1) cement-mortar and an
assessment of their long-term environmental behaviour in construction
products during service life and recycling. Cement & Concrete Research
30, 1–18.
van Gemert, W. J. Th., Quakernaat, J. & van Veen, H. J. (1988) Methods for
the Treatment of Contaminated Dredged sediments. In: Salomous, W.
& Foerster, U. (eds.) Environmental Management of solid waste.
Springer-Verlag. Berlin. p 44.
Wainwright, P. J. & Boni, S. P. K. (1981) Artificial Aggregate from Domestic
Refuse. Concrete. 5, 25–29.
Wainwright, P. J. & Boni, S. P. K. (1983) Some Properties of Concrete
Containing Sintered Domestic Refuse as a Coarse Aggregate. Magazine
of Concrete Research. 35, 75-85.
Wainwright, P. J. & Robery, P. (1991). Production and Properties of Sintered
Incinerator Residues as Aggregates for Concrete. In: Goumans, J. J. J.
M. et al. (eds.) Waste Materials in Construction – Proceedings of the
International Conference on the Environmental Implications of
Construction with Waste. Elsevier. Amsterdam. pp 425–432.
Woolley, G.R. (1994) State-of-the-art report. Use of waste materials in
Construction – Technical Development In: Goumans, J. J. J. M. et al.
(eds.) Environmental Aspects of Construction with Waste Materials.
Amsterdam. Elsevier. pp.963–977.
Yasuda, Y. (1991) Sewage Sludge Utilisation Technology in Tokyo. Water
Science Technology 23, 1743–1752.
Waste Management & Research 289