The Use of Cold Bitumen Stabilized Base Course

Mixes in Norway

Myre Jostein, Senior Engineer, Norwegian Public Rods Administration, Akershus,

Norway

Abstract

In Norway, penetrated crushed macadam, asphalt concrete, crushed gravel and

crushed rocks are the most common types of base course materials. However, in the
last 10-20 years the use of cold mixes has increased steadily.

The major difference between cold mixes and hot mixes is that the aggregates for cold
mixes are cold and damp during the production process. The type of binder in cold
mixes is also different as compared to hot mixes. Two types of binders are commonly
used in cold mixes: bitumen emulsion and foamed bitumen.

Cold bituminous base course mixes have a number of advantages compared to

unbound base course materials such as high structural coefficient, high stability during
the spring thaw season and reduced moisture susceptibility. The use of cold mixes in
the base course generally yields a long service life of the pavement.

In Norway, a complete mix design system for cold mixes has been developed. This
includes specimen preparation, curing and determination of indirect tensile strength.

Cold base course mixes may be produced by in-situ milling, in mix pavers or in
simple and highly mobile plants which are placed close to the construction site. Three
different types of plants are frequently used in Norway: free fall mixer, vertical mixers
and pugmills.

In Norway cold mixes are used for rehabilitation as well as for new pavement
structures. However, rehabilitation contributes to the greatest volumes. Recommended
AADT limits for the use of cold mixes in the base layer are presented in the paper.

Generally, cold mixes have a lower workability than hot mixes. Therefore, cold mixes
are more difficult to lay and to compact. Consequently, pavers with a heavy screed or
graders are recommended. Compaction trials have been carried out in order to find the
most efficient rollers for cold mixes. For cold base course mixes, vibratory rollers
with high amplitude yield the best compaction.

1
1 Introduction
This paper deals with the use of cold bitumen stabilized base courses in Norway. The
first cold mix project in Norway was carried out in 1983. Since then, the volume has
increased steadily to approximately 1,5 mill. m2 of milling stabilization and 50-
100,000 tons of plant mixed materials per year now.

2 Mix types

2.1 General
Three types of cold base course mixes are frequently used in Norway: Gravel
emulsion (Eg), foamed gravel (Sg) and bitumen stabilized gravel (Bg). These cold
mixes may be produced by plant mixing (Eg and Sg) or by in-situ milling (Bg), see
Chapter 4. Plant mixing (Eg and Sg) generally yields a more homogenous mix than in-
situ milling stabilization (Bg).

The binder content in all three mixes normally is in the range of 3.0-4.5 % (residual).
Table 1 shows recommended maximum traffic volumes for cold base course mixes, as
measured by annual average daily traffic (AADT).

Table 1: Recommended AADT Limits for Cold Mixes.

Mix type Production Recommended AADT limits for use in

method
Upper base course Lower base course
Gravel emulsion (Eg) or Plant mixing 3000 5000
foamed gravel (Sg) or mix pavers
Bitumen stabilized gravel In-situ milling 1500 3000
(Bg) stabilization

2.2 Gravel emulsion (Eg)

Gravel emulsion (Eg) consists of aggregates and bitumen emulsion. Eg is produced at
plants or in mix pavers. The maximum aggregate size is normally 16, 22 or 32 mm.
Penetration grades 145-210 or 300-430 are normally used for making emulsion. Rapid
(R), medium (M) or slow curing (S) emulsions should be selected on the basis of
mineralogy, grading and production method.

2.3 Foamed gravel (Sg)

Foamed gravel (Sg) is made from aggregates and foamed bitumen. The binder is
usually produced by penetration grades 145-210 or 300-430. Foamed gravel is usually
less susceptible than gravel emulsion to precipitation in the initial period subsequent
to laying. Foam bitumen is produced by adding small amounts of water, i.e. 2-5 % to
hot bitumen (above approximately 150 oC). This will cause an expansion of the
bitumen corresponding to approximately 15-20 times the original volume. The
amount of water added should be determined on the basis of the volume expansion
and the duration of the foam. Increased amounts of water yields a greater volume

2
expansion, but also a shorter duration of the foam, see Figure 1. Normally, addition of
3 % water yields an acceptable volume expansion and foam duration.
Before foaming an active adhesion agent has to be added to the bitumen. Normally,
this is not necessary for gravel emulsion because the emulsifier in most cases also
works as an adhesion agent.

25 16
14
Duration of the foam (s)

20
12

Expansion ratio
15 10
Duration of the foam (half life)
8
10 Expansion ratio
6
4
5
2
0 0
0 2 4 6
Water addded (%)

Figure 1: Duration of the foam (half life) and the expansion ratio.

2.4 Bitumen stabilized gravel (Bg)

Bitumen stabilized gravel (Bg) is produced at the construction site by milling and
stabilization of the existing gravel base course. This may also involve milling of
existing bituminous layers or imported unbound materials. The binder in Bg may be
foamed bitumen or bitumen emulsion. In Norway penetration grades 145-210 or 300-
430 are recommended.

2.5 Comparison of hot mixes and cold mixes

2.5.1 Curing
The aggregates for cold mixes are cold and damp during the production. However,
after laying the moisture content is reduced gradually as the mix is curing. Cold mixes
will therefore normally have a low structural coefficient in the initial period
subsequent to laying, and it may take up to 1-2 years before the materials have
developed full strength, see Figure 2.

3
 200

Momenthum (Nm)
150

100

50

0
0 100 200 300 400
Days after laying

Figure 2: Curing of a gravel emulsion mix /2/.

2.5.2 Workability
Cold mixes have a considerable lower workability than hot mixes, see Figure 3.
Therefore, cold mixes are more difficult to lay and to compact. Consequently, it is
necessary to use heavy pavers and screeds.

140

120
Maximum load (N)

100

80

60

40

20

0
Gravel Hot mix Foammix Emulsion
gravel

Figure 3: Workability (mix viscosity) for cold mixes and hot mixes /1/.

2.5.3 Selection of cold mixes versus hot mixes

In and around most densely populated areas in Norway there are many stationary hot
mix plants. Therefore, hot mixes generally are cheaper than cold mixes in such areas.
However, in remote areas the costs for cold mixes are often lower than hot mixes.
This is because:
• cold mix plants are simple and less space demanding than mobile hot mix
plants;
• local and marginal materials may be used for production of cold mixes;
• the need for hauling is reduced because the cold mix plant may in most
cases be established close to the construction site.

4
Thus, for small production volumes in remote areas the use of cold mixes provide a
cheaper solution. However, to get a true comparison of cold mixes and hot mixes it is
recommended to include both options in the tenders.

3 Mix design

3.1 General
In Norway a complete mix design system for cold mixes have been developed. The
main objectives of mix design are to optimize the mix composition, characterize mix
properties and to avoid “abnormal” behavior of the mix during production.

Mix design takes a minimum of nine days. In addition, it may also be necessary to
carry out relatively time consuming supplementary investigations, both in the field
and in the laboratory. The planning should therefore commence well before the job
itself has to be carried out.

For cold mixes, there are requirements pertaining to the structural coefficient (all
mixes) and the binder coverage (Eg). Durability, adhesion and fatigue properties
should also be assessed. The mix design process can be divided into the following
phases:
• sampling and evaluation of aggregates;
• selection of binder;
• selection of emulsion composition;
• test preparation;
• curing;
• testing;
• evaluation of results and selection of mix composition.

3.2 Sampling and evaluation of aggregates

In a stockpile, separation can easily take place, and the moisture content will often be
different on the surface, as compared to a position near the center of the stockpile.
Sampling of stone materials for mix design must be carried out in such a way that the
aggregate samples are representative. Therefore, it is absolutely necessary to use
mechanical equipment (tractor etc.).

The following properties of the aggregates should be evaluated:

• gradation;
• reactivity;
• dust coating;
• mechanical properties;
• mineralogy and humus;
• amount of crushed aggregates.

Gradation is probably the most important single factor for selection of mix type, i.e.
gravel emulsion (Eg) vs. foamed gravel (Sg). Recommended envelope curves are
specified in Figure 4 and 5. Materials which are stabilized by foamed bitumen (Sg and

5
also Bg) are normally produced using aggregate materials with a fines content
(<75µm) between 5-15%. One has however had positive experience also with fines
content (<75µm) between 2-20%. In addition, the amount less than 2 mm should
comprise a minimum of 30%.

In most cases mixes based on bitumen emulsion (Eg and also Bg) are produced using
aggregates with a fines content (<75µm) between 1-7%. The amount less than 2 mm
should also comprise a maximum of 20%.

The maximum aggregate size and amount of coarse aggregates are important with
regard to compaction and abrasion of equipment used for production and laying. For
plant mixing, it is easy to screen out oversize materials to comply with recommended
envelope curves. Problems with large stones are therefore first and foremost
connected to in-situ milling and stabilization of existing roads. Due to abrasion of the
milling equipment, the maximum aggregate size should not exceed 100 mm, and the
amount greater than 64 mm should not exceed 20-50%. Addition of virgin materials
must be evaluated if the gradation does not comply with the above mentioned
recommendations.

Reactivity is used to quantify the surface charge of the particles (aggregates). The
aggregates are most reactive immediately after crushing. The reactivity then decreases
and stabilizes at a level which is almost constant, see Figure 6. The reactivity is
important for the selection and application of emulsion. Newly crushed materials may
require a different emulsion composition than if the same stone materials were
crushed a long time ago. To ensure the most stable reactivity possible for stone
materials during production and laying, one should wait at least two to three weeks
after crushing before producing gravel emulsion (Eg), to ensure that the reactivity gets
down to a stable level, see Figure 6. It has not been proven that the reactivity is of a
corresponding importance when using foamed bitumen.

2,5

2
Reactivity (Mb-number)

1,5

0,5

0
0 50 100 150 200 250
Days after crushing

Figure 6: Reactivity of aggregates.

6
Dust coating on coarse aggregates may be a problem during production of gravel
emulsion. Dust coating on aggregates can be removed by washing. Experience shows
that problems with dust coating increase when unsorted materials are left in a
stockpile for a period of time. To split the aggregates immediately after crushing is the
simplest and most effective measure for reducing the problem with dust coating (such
as in 0-8 mm and 8-11 mm).

Humus may result in uneven curing of bitumen emulsions. If there is humus in the
aggregates, the binder content probably has to be increased. This should be taken into
account when comparing costs connected to the use of different aggregates.

3.3 Selection of Bitumen Emulsion or Foamed Bitumen

Bitumen emulsion usually contains 30-40 % water. Bitumen emulsion in combination
with varying fines contents may result in uneven curing and low stability of the mix,
particularly in the initial period after laying. In addition, the fines will retain water.
This may imply increased curing time for the mix. The main rule is therefore that
bitumen emulsion should not be used in combination with high fines contents.
Foamed bitumen contains much less water than bitumen emulsion and does not entail
the same problems as bitumen emulsions for materials rich in fines.

The climate and the moisture content of aggregates may also be important for the
selection of binder. Should the aggregate materials have a high moisture content at
the outset, and one does not have the opportunity to reduce this through aeration, the
use of emulsion may result in too high a moisture content in the mix, regarding
optimal compaction and potential emulsion flow problems. In such cases foamed
bitumen is preferable. This is also the case if the climate is wet with a high risk of
precipitation during laying. Precipitation may result in instability of the mix.

Selection of binder type is also an economic question. Usually, the cost of binder is
reduced substantially through the use of foamed bitumen, as compared to bitumen
emulsion. In cases where, on the basis of available production equipment, aggregates,
economy, etc., the type of binder is not immediately apparent, mix design should be
carried out with foamed bitumen as well as bitumen emulsion.

3.4 Selection of emulsion composition

Selection of emulsion composition is carried out after the aggregates have been
assessed and found suitable for gravel emulsion. Selection of emulsion composition
includes the following phases:
• selection of emulsifier;
• selection of emulsion type (amount of binder and water);
• selection of curing rate.

7
Emulsion composition and type of emulsifier are selected based on viscosity, storage
stability of the binder, binder coverage, adhesion and workability for the mix.

3.5 Determination of moisture content

The moisture content of foamed gravel and gravel emulsion is very important in order
to achieve:
• good binder coverage;
• good compaction;
• high structural coefficient numbers.

Trial mixing in the laboratory is recommended in order to determine the moisture

content. Mixing at different moisture contents should be followed by evaluation of:
• workability;
• binder coverage;
• curing.

In Norway, it is recommended to keep the total moisture content after mixing within
the following limits: OMC to OMC - (0.5 · residual binder content). The optimum
moisture content OMC, for non-stabilized materials is determined by Modified
AASHTO.

3.6 Determination of binder content

The binder content is determined on the basis of requirements related to structural
coefficient (all mix types) and binder coverage (Eg). There are also minimum
requirements for the binder content related to the fines content:

Pa = 0.14 · p75 + 2.6

where
Pa = minimum binder content in weight percent (residual) (pa > 3.0%)
p75 = percentage of aggregate material less than 75µm.
Figure 7 shows the required binder content as a function of the fines content (%
material < 75µm). As can be seen, the minimum binder content for cold base course
mixes is 3 % (residual).

8
 Binder content (%-residual) 5

4,5

3,5

2,5
0 5 10 15 20
Fines content (%<75 µm )

Figure 7: Minimum binder content (residual).

3.7 Specimen preparation

3.7.1 General
Mix design is carried out by preparing cylinder shaped test specimens to the same size
as the Marshall specimens. In Norway static compaction or a gyratory compactor
may be used for compaction of specimens. The molds must be perforated so that
moisture can come out. This is of particular importance when preparing specimens of
gravel emulsion.

3.7.2 Static compaction

The following procedure is used for static compaction:
• the load is increased gradually from 0 to 8 tons within two minutes;
• the maximum load of 8 tons is maintained for two minutes;
• the specimen is unloaded and demolded.

3.7.3 Gyratory compactor

The following compaction preliminary procedure with the gyratory compactor is
proposed in Norway:
• first a test series with a minimum of three test specimens, is compacted
using a total of 200 rotations;
• subsequently, a new test series is compacted to 96% of the density of the
previous test series.

The latter test series is used to determine the structural coefficient based on static
indirect tensile tests. The apparatus settings during compaction are as follows:
• 1o angle;
• 600 kPa static load (on specimen surface);
• 30 rotations per minute.

9
3.8 Curing and testing
Curing is carried out as follows:
• The specimens are stored (dry) in an oven for 7 days at 40° C. In the event
of lack of time, one may with the approval of the Client alternatively cure
the specimens for 3 days at 60° C;
• The specimens are then exposed to 8 freeze/thaw cycles: 4 cycles per day
for 2 days.

Before testing the specimens are tempered in a water bath at 25º C for 30-40 minutes.
The indirect tensile test is then used to determine the stiffness of the specimens. The
indirect tensile strength is calculated according to the following equation:

St = (636.62 · Pmax)/(t · D)

where
St = indirect tensile strength (kPa)
Pmax = maximum load at break (N)
t = specimen height (mm)
D = specimen diameter (mm)

4 Production and laying

4.1 Introduction
Cold mixes may be produced at plants or by in-situ milling. For new roads, cold
mixes are normally produced at stationary or mobile plants. Whereas, for
reconstruction of existing roads, the materials may also be produced by in-situ milling
stabilization.

4.2 Stabilization by in-situ Milling

The objective of milling is to stabilize existing unbound base course materials through
the addition of bitumen emulsion or foamed bitumen. The mix produced by in-situ
milling is designated bitumen stabilized gravel (Bg), see Chapter 2.4.

The milling widths for stabilizing machines vary from approximately 2.2 to 4.4
meters, depending on type of machine. Normally, the maximum milling depth is 20-
40 cm depending on the milling machine and the stiffness and thickness of the
bituminous layers. However, typical milling depths in Norway are 10-15 cm.

Milling machines for stabilization can be equipped with one or two drums. Dual drum
machines normally provide a more homogeneous mix after one milling pass. The
homogeneity is also related to the rate of progression; slow progression generally
yields a more homogeneous mix than high rates of progression.

10
4.3 Plant Mixing

4.3.1 General
The following production equipment is frequently used in Norway:
• free fall mixer;
• vertical mixer;
• pugmill.

In Norway, it is recommended to use two or more fractions when producing cold

mixes by plant mixing. Fractioning will contribute to better homogeneity. The
following fractions are typical: 0-4 mm, 0-6 mm, 0-8 mm, 4-11 mm, 6-11 mm, 6-16
mm, 8-11 mm, etc.

4.3.2 Free fall mixer

The free fall mixer consists of a mixing chamber with interior sieves and spraying
bars, see Figure 8. The aggregates are fed into the top of the free fall mixer via a
conveyor belt and split into two fractions, usually 0-6 mm and +6 mm. Spraying of
binder takes place through three spraying bars, placed under the sieves. The amount
of bitumen is adjusted depending upon the amount of aggregate materials which pass
through the different sieves. Separate spraying of binder to the fine and coarse
fractions ensures that also the coarse fractions are coated by binder. This may
otherwise be a problem when binder is added to unsorted materials. The binder may
be foamed bitumen or bitumen emulsion.

4.3.3 Vertical mixer

The vertical mixer consists of a mixing chamber with interior mixing arms and
mixing blades, see Figure 9. The equipment can be classified as a combined gravity
mixer and a mechanical pugmill. The binder (bitumen emulsion or foamed bitumen)
is added through spraying bars mounted at up to three different levels in the mixer.

4.3.4 Pugmill
A pugmill is a highly mobile plant which most often is equipped with 2-3 dosing
pockets and conveyor belts for separate addition of coarse and fine fractions to the
mixer (split material flow). Pugmills may also have spraying bars for separate
addition of binder to coarse and fine fractions and separate binder lines for the
addition of two binder types.

4.3.5 Mix pavers

Mix pavers are used for production and laying of cold materials in-situ. A mix paver
may also be placed on ramps and used as an ordinary cold mixing plant. Mix pavers
are equipped with binder tanks and storage bins for aggregates and binder tanks.
Foamed bitumen or bitumen emulsion may be used.

11
4.3.6 Laying
Cold mixes have a considerable lower workability than hot mixes, see Chapter 2.
Therefore, it is necessary to use heavy pavers and screeds.

4.3.7 Rolling
Extensive compaction trials have been carried out in Norway in order to find the most
efficient rollers for cold mixes, see example of results in Figure 10. For cold base
course mixes vibratory rollers with high amplitudes yield the best compaction.
2,4

2,3
Static roller
2,2
Density (g/cm**3)

Dynamic roller;
2,1 high amplitude
Dynamic roller;
2
low amplitude
1,9

1,8

1,7
0 2 4 6 8
No of roller passes

Figure 10: Compaction trials for a gravel emulsion mix (Eg) /1/.

5 Use of cold mixes

5.1 Terms for use of cold mixes

Cold mixes may be used for construction of new roads or for rehabilitation of existing
roads. Cold mixes have a relatively low initial strength. The fines, however, are bound
and accordingly the mix is less moisture susceptible than corresponding materials
without binder. Additionally, the structural coefficient increases quickly as the
moisture content is reduced and the mix is gradually being cured, see Figure 2. To
ensure this, the stabilized material has to be well drained. In case of excessive
moisture contents the structural coefficient may not increase as anticipated.

Due to the cold climate in Norway, it is advantageous to produce and lay cold mixes
early in the summer, so that there is enough warmth and traffic on the stabilized layer.
This will promote a favorable development of the strength in the mix before frost sets
in and before the next spring thaw. If laying late in the season, or in parts of the
country where long warm periods are rare, it is beneficial to wait one year or more
before a permanent wearing course is laid. For roads with little or moderate traffic,
temporary seals or surface treatments may be laid on top of the cold bitumen
stabilized base course.

12
5.2 Construction of new roads
For new roads, it is common practice to produce the cold mix in plants or mix pavers.
Normally, the same cold mix is used in the upper and lower base course. However, as
there are more stringent requirements higher up in the road structure, it may be
relevant to use different types of materials in the lower and upper base course for
technical or economic reasons. This is particularly true for roads with heavy traffic.

Bitumen stabilized materials may also be used as temporary surfacings during the
construction stage. Temporary surfacings of cold mixes can easily be cleaned by
sweeping and/or hosing before the next layer is laid. Whereas for unbound materials,
excavation of materials which are contaminated during the construction stage may be
necessary.

5.3 Rehabilitation
Both plant mixed materials and in-situ stabilized materials are used for rehabilitation
purposed. Production of cold mixes in plants or mix pavers is recommended when:
• there is a nearby quarry with suitable materials for stabilization;
• the distance to the closest hot mix plant is long;
• the existing road structure does not contain sufficient amounts of suitable
materials for milling (in-situ);
• an increase in total thickness of the pavement is acceptable (which often imply
widening of the road).

On the other hand, stabilization by in-situ milling is particularly suited in cases where:
• widening of the existing road should be avoided because of environmental
considerations (farm land) or because of extra costs in connection
with land acquisition, other connecting roads along the section or
adjustment of technical installations such as guardrails;
• there is a shortage of suitable materials for plant production;
• the existing road has a moisture susceptible gravel base course.

Some years ago, low quality natural gravel was frequently used as a base course on
low volume roads in Norway. This has resulted in a short pavement service lives
which can often be related to the gravel base course being moisture susceptible. In-
situ milling stabilization is particularly well suited in such cases, and may result in
long pavement service lives and low maintenance costs. In addition, the technique
allows good use of limited resources, in that existing materials are used.

However, due to abrasion on the equipment, in-situ milling stabilization is not suitable
if there is a high content of large stones in the existing base course. Virgin materials
should be added if the gradation does not comply with the recommendations described
in Chapter 3.2.

13
6 References

/1/ Publication no. 92; The cold asphalt development project (AUT), Public Roads
Administration, NRRL, Oslo, 1998 (in Norwegian with an English summary).

/2/ Manual no. 198; Cold base course mixes, Public Roads Administration, NRRL,
Oslo, 2000.

14