Dipl.-Ing. L.

Langmaack
MBT International, Underground Construction Group, Zrich, Schweiz

Europe & Asia:

Application of new TBM Conditioning Additives

ABSTRACT:
The importance of shielded tunnelling is growing more and more. To do this in a successful
way, even Slurry- and Earthpressure Balance-TBMs need conditioning additives. Sometimes a
project only becomes a reality as a result of their use.
The growing product diversification concerning conditioning additives is a consequence of the
latest developments. This necessitates specific tests to evaluate the advantages or disadvantages
of their use. The use of conditioning additives has to be combined as effectively as possible with
the existing complexities TBM and Geology in order to result in optimum TBM advance.
New developments regarding tunnelfoams and polymers are presented here. In addition to the
more classic demands like modification of soil rheology and pressure stabilisation in the working
chamber, other points of interest which demonstrate the complexities are also playing an important role e.g. reducing the stickiness and adhesion in clayey soils or water ingress control in porous
soils. Another key factor is the proving of the performance of each conditioning additive by doing
practical laboratory tests.
The use of modern conditioning additives allows the extension of the classic operation areas
of Bentonite- and Earthpressure Balance TBMs. The global site applications presented by examples of Europe and Asia underline the successful use of conditioning additives, around the world.

1 Conditionig Additives - Overview 1.1 Is there a need for conditioning additives?

The latest literature [1] shows the following diagram (see figure 1.) concerning the limits of
EPB tunneling. In consequence soils with less than 10% clay, 10-60% silt and a gravel (fine
gravel) content of maximum 30% are up to now judged as suitable for EPB machines. But even
to reach these limits, additives have to be used to reduce stickiness in clayey soils or to plasticise
coarse, frictioned soil. Also for the silty sand in-between these limits conditioning additives show
important effects on reducing the torque and wear.

CLAY

SILT

MATERIAL RETAINED ON SIEVE [%]

FINE

SAND

MEDIUM

COARSE

FINE

MEDIUM

GRAVEL
FINE

COARSE

MEDIUM

COARSE

100
90

IF Sumax > 80 - 100 kPa

OR (SPT) Nmax > 8 - 10
THEN OPEN SHIELD
CAN BE USED

80
70
60

NO OR LOW-RATE
CONDITIONING

50
40
30

CONDITIONING DUE TO
COARSE, FRICTION-SOIL

CONDITIONING MAINLY
DUE TO ADHESIVNESS
OF SOIL

NOT SUITABLE FOR

EPB TUNNELLING

20
10
0
2

0.001

4 6 8

0.01

4 6 8

0.1

4 6 8

4 6 8

10

4 6 8

100

GRAIN SIZE d [mm]

Figure 1. Limits of EPB tunneling

1.1.1 The use of additives on EPB machines

Referring to figure 1 three main areas of conditioning by foam and / or Polymer can be identified:
coarse, frictioned soil
conditioning necessary to obtain plastic material for building-up a correct earth-pressure in the
working chamber
silty sands
conditioning shows positive effects regarding the reduction of abrasion and decreasing the
torque. This effects allow in consequence to increase the TBM speed.
clay
conditioning necessary due to adhesion and clogging problems
In case of very stiff and homogeneous clays open shields (air pressure mode) can be used. The application field of porous soils like gravelly sands are today mostly driven in Slurry mode, but some
site applications show that these soils are also possible to treat effectively with EPB machines.

1.1.2 The use of additives on Slurry machines

Additives for Slurry machines can be useful under the following conditions:
saline ground water (bentonite is loosing its capabilities)
sticky or swelling clays (clay lumps are produced by the cutterhead which are clogging together in the working chamber later, blocking the cutterhead and the mud pump)
segregation problems in the working chamber
These additives are mainly Polymers or special clay dispersants.
1.1.3 The use of additives on Hard Rock machines
Additives for Hard Rock machines are mainly foams for dust suppression or special anti abrasion
additives. The tunnelfoam has to be dry and stable to collect as much dust particles as possible.
The application of this foam may be on the conveyor belt sprayed on the excavated material or introduced in front of the cutterhead. This solution is maybe the most effective one because the dust
is binded where it is created, but not all hard rock TBMs are equipped for this kind of application.

1.2 Characterisation of Tunnel-Foams

To create a tunnelfoam, foaming solution and air has to be pumped simultaneously through a foam
generator. The foam generator may be filled with grids, membranes or a granular material. It shall
be equipped with flow meters and pressure meters to allow a complete monitoring of the injected
foam quantity.
This is very important to know exactly what happens in the working chamber. If the foam parameters are not set correctly, the following situation may appear:
too much air injected:
a. soil becomes too dry in the chamber, temperature and torque increases, insufficient speed.
b. creation of a big air bubble in the top of the TBM with non-homogeneous soil support and
risk of blow-out
too much liquid or Foam injected:
a. soil can be squeezed trough the extraction screw
b. muck handling becomes difficult
c. increasing costs per linear meter of tunnel
too less conditioning:
a. soil may plug the cutterhead and / or chamber, insufficient TBM speed
b. pressure loss in front of the TBM
c. water income
In order to use soil conditioning with tunnelfoam under defined conditions, the following parameters have to be defined (see figure 2):
cF (Concentration of foaming solution)
which determines the concentration of the foaming concentrate in water
FER (Foam Expansion Ratio)
which characterises the ratio of Air mixed with the foaming solution
FIR (Foam Injection Ratio)
which characterizes the ratio of foam mixed into soil

F.Concentrate

Foaming Concentrate
Water

Foaming Solution

cf = 100 x

m (foaming concentrate)
m (foaming solution)

Air

Foam

FER =

V (compressed air)
V (foaming solution)

Soil

Foam + Soil

figure 2. key parameters of tunnelfoam use

FIR = 100 x

V (foam)
V (soil)

The foaming concentrate which is needed to create the foaming solution, contains surface active
substances called surfactants. These surfactants are molecules with a combination of a hydrophobic chain and a hydrophilic head. Figure 3 illustrates how the surfactants are working:

Tunnel - Foam
hydrophobic
chain

hydrophilic
head

Air

Foam
Water

Air

Generator
Water

Figure 3 Presentation of Foam

Both parameters hydrophobic chain and hydrophilic head can be chemically varied: different
chain structures (length, steric structure) and different head characters (anionic, non-ionic, cationic,
amphoter) are possible. The different chemical characters come to different properties of a tunnelfoam: modification of
superficial / interfacial tension
force of dispersion
solubility
emulsification
foaming capacity
foam stability, etc.
Each soil type, from stiff clay to sandy gravel, requires more or less his own type of foam to work
properly. The type of foam which has be used for a specific site should be determined by laboratory tests with the original in situ type of soil.

1.3 Characterisation of Polymers

There is a wide range of application for polymers. Examples can be:
Structurising a soil
Reduction of stickiness
Reduction of adhesion to metal surfaces
Drying out a soil
Reduce soil segregation in the working chamber
In consequence there exists a wide range of products called polymers on the market, for example:
Polyacrylamides (not rated due to environmental problems)
Polyacrylates
CMCs
Biopoylmers
These product types are totally different from each other and not all are suitable for TBM use or
not up to date. Regarding the use of additives on site it is strongly recommended that they fulfill at
least the following criteria:
liquid state for easy and controllable dosage

 no blockage of the foam generator in case of a combined use with foam

develop its reactivity within minutes to be able to react as quick as necessary
not dangerous for the environment
A quite new and exciting range of additives are the Biopolymers, which show some additional effects like thixotrophy or pseudoplasticity and can be used in Slurry- and EPB-TBMs.

1.4 Clay Dispersants

Dispersing agents are mainly added to stiff clay in order to support the destructuring / dispersing
properties of the foam, but they might also be introduced without foam.
To fulfil the desired job, the dispersing agents have to adsorb on the soil particle surface. They
have to carry a high charge density to separate the soil particles and they should create a steric barrier.
These demands can be fulfilled both by surfactants and dispersants, but dispersants are more efficient due to a higher amount of loads per molecule (as illustrated in figure 4).
The choice for which geology it is better to use foam whether to inject dispersing agents can only
be made by specialists in the laboratory by studying the original soil and combine these knowledge
to the effects visible on the TBM itself.

Dispersants and Surfactants

SO
3

C C C C C
O O O O O
O- O- O- O- O polymeric

low molecular wight

numerous ionic groups

usually only one ionic group

high deflocculating power

limited deflocculating power

figure 4: Dispersants and Surfactants

2 Site Examples
After the given background information regarding which types of additives exist and for which application they may be needed, this chapter indicates some examples of their successful world-wide
use on site.

2.1 Aviles Collector (Spain):

Porous sand under 3,0 bar seawater pressure
The Aviles Site works with a Lovat EBP machine, diameter 3,40 m. After facing stiff clayey silt,
the soil changed over a length of approximately 1.000m to pure gravely beach sand with a seawater pressure of nearly 3,0 bar. The grain size distribution is shown in figure 5:

MATERIAL RETRAINED ON SIEVE [%]

CLAY

SAND

SILT

GRAVEL

100
90
80
70
60
50
40
30
20
10
0
0.001

0.010

0.100

10

100

GRAIN SIZE d (mm)

figure 5: Grain size distribution curve of Aviles Sand

There was no possibility on the machine to introduce additional fine material into the working
chamber. An extra installation had been too costly and too time intensive. The alternative solution
was to use additional Polymers to make the soil as plastic as possible (to be able to install a
counter pressure against the seawater) and to reduce the water content of the outcoming soil.

2.1.1 Laboratory tests

Lab results present some segregation control tests (figure 6) to identify the searched plastic comportment and penetrometer tests (figure 7) to study the effects of an excess of water. All tests are
realised with the original Aviles Sand and are carried out with Wi=7%, d(org)=1,5.

Segregation of Aviles Soil

1

10

time [min]
100

1000

10000

0
segregation [ml]

20
40
60
80
100

Water
Foam
Foam + SLP 2

120

figure 6: segregation tests

Figure 6 shows the drastic problem of a coarse soil mixed with water. An almost instantaneous
segregation is obtained, impossible to work with on a TBM. With a very stable tunnelfoam the result can be improved, but still 20% segregation after 15 minutes is obtained too much for a secure work of the TBM. The segregation as well as the homogeneity of the soil paste can only be
controlled by the addition of a structuring biopolymer to the tunnelfoam.
The penetration depth of the cone penetrometer indicates the plasticity of the soil: the higher the
penetration depth the more liquid is the soil. The aim in this case is to maintain the penetration
depth even when adding water. These tests results are shown in figure 7:

Penetration depth [1/10mm]

400
350
300
250
200
150

water
water
water
water

100
50

+3%
+1%
+1%
+1%

SLF
SLF
SLF
SLF

45
45
45 + 3% P1
45 + 3% P2

0
0

10

Addition [%]
figure 7: cone penetrometer tests with Aviles Soil + Foam + Polymer
Only the use of foam will not be successful, the soil gets fluid far too quickly even when the concentration is decreased to 1%. The use of SLF P1 stabilises the soil up to a certain water content,
but looses its efficiency when the water content is too high. Whereas SLF P2 shows no change in
the penetration depth even when the water content increases.

2.1.2 Conclusion:
The decision was to inject stable foam together with a structuring biopolymer (SLF P2) in order to
keep the water away and to structurise the small amount of fine soil particles. The result was a stable, homogeneous and pasty soil. For security reasons a second polymer (SLF P1) was injected
into the screw conveyor in order to dry out again the excavated material.
The drive through this type of soil was very successful with the presented solution and reached
high advance rates.

MATERIAL RETRAINED ON SIEVE [%]

2.2 Hongkong:
conditioning of weathered granite
The most difficult soil for an EPB machine in this project was weathered granite which showed the
following grain size distribution curve (see figure 8) which is composed out of two different soil
graduations:
CLAY

SAND

SILT

100
90
80
70
60
50
40
30
20
10
0

GRAVEL

type 1
type 2

0.001

0.010

0.100

figure 8: grain size curve of weathered Hongkong granite

10

100

GRAIN SIZE d (mm)

In order to identify the right conditioning additives and injection parameters to create a cake development, permeameter tests were carried out with the results presented in figure 9. For these tests
the cylinder has been filled first with gravel, then with the soil which should be tested. Water was
finally added up to a certain height over the soil. The tests were carried out with an overpressure of
0,4 bars and with the soil type I which contains less fine particles.

total outflow [ml]

1400
1200
1000
800
600
400
200
0
0

soil
soil + SLF 30 (FER 10)

10

t [s]

15

20

soil + SLF 30 (FER 6)

soil + SLF 30 + P2

figure 9: Permeameter tests with p=0,4 bar

The untreated (only water added) soil porosity was calculated to k=5*10-3 [m/s]. The outflow
increases due to the outwashing effect of the fine particles. In the second step the soil was mixed
with foam (cf=3%, FER=6, FIR=40) which caused less outflow compared to the untreated soil.
But still insufficient to create a filtercake. In a second step the expansion ratio was increased from
6 to 10 to obtain a dryer foam. The first indications of a filter cake development are shown but it
was not stable enough. In the fourth trail the foam was reinforced by adding 2,5% of the Polymer
MEYCO Fix SLF P2. The result was a stable cake even at p=0,6 bar which is possible due to
the structuring effect of the Polymer.

2.3 Bangkok Metro (Thailand):

clogging problems of Bangkok Clay
The EPB machine suffered due to the stiff and adhesive Bangkok Clay. The contractor used a
Japanese Polymer and was not happy about the obtained effects itself and compared to the costs.
The solution was to introduce a material, which increases the clay dispersion in order to obtain
untreated clay lumps embedded in a more homogeneous soil paste, which does not stick. Laboratory ball rolling tests (clay balls + sand in water) were carried out and the effects are shown in figure 10.
On the first view the results are astonishing. Especially for clay designed dispersing agents or
foams like SLF 30 or 45 do not have the desired effect on Bangkok Clay. This illustrates once
more the complexity of the interaction between soil and conditioning agents. For the desired site
the use of foam SLF 20 showed the best results in the laboratory. This is well proven on site since
February 2000.

Mass loss [%] after 3h

60
50
40
30
20
10
0
c lay dispersant A

c lay dispersant B

c lay dispersant C

Foam SLF 20

Foam SLF 30

Foam SLF 45

figure 10: Ball Rolling test with Bangkok Clay

2.4 Wesertunnel (Germany):

adhesion and clogging of Lauenburger Clay
The most difficult soil condition this Slurry TBM has to work with is the Lauenburger clay. It
shows high adhesion and clogging properties on site. Additional problems occur due to the saline
groundwater which reduces the bentonite effects.
The aim of the laboratory tests have been to prevent the cutted clay lumps from clogging (in terms
of re-agglomeration) in the working chamber with secondary problems like segregation. The used
test method was again the rolling test: A cylinder was filled with the original bentonite suspension
made with the original water and freshly cut clay chips were added. The results are shown in figure 11.
Clay chips

Clay chips + Bentonite

Clay chips + Bentonite

+ 0,1% SLP P2

Clay chips + Bentonite

+ 0,1% SLP P1

Clay chips + Bentonite

+ 0,2% SLP P2

Clay chips + Bentonite

+ 0,2% SLP P1

figure 11: rolling test of Lauenburger Clay in bentonite suspension

If the clay chips are rolling in the untreated bentonite suspension, they are quickly agglomerating
and forming a big lump. This causes clogging of the cutterhead (followed by low advance speed),
segregation effects in the working chamber and finally blocking of the bentonite circuit. Is the
bentonite treated with Polymer P2, at 0,1% the clay agglomeration is drastically reduced at 0,2% it
increases slightly again. Is the bentonite treated with Polymer P1 at 0,1% the agglomeration is
nearly prevented but at 0,2% the dispersing effect increases visibly. This has negative influences
on the bentonite separation. This very positive laboratory results have to be transferred to site application.

3 Conclusion
The laboratory results combined with the 3 examples of world wide site uses prove the positive
effects of conditioning additives in shielded tunneling. Their use is one of the key factors for a successful and overall economic TBM drive.
In addition the market shows the tendency towards using EPB machines even in clay soils (London Heathrow T5 as another example) and also in porous soils (Aviles, DTSS Singapore, ...). This
results in new limits for EPB machines summarised in figure 12.
CLA Y

SILT

M ATER IAL R ETA IN ED ON SIEVE [% ]

F INE

SAND

M EDIUM

COARSE

FINE

M EDIUM

G RA VEL
FINE

COARSE

MED IUM

COA RSE

100
90

IF Sumax > 80 - 100 kPa

OR (SPT) Nmax > 8 - 10
THEN OPEN SHIELD
CAN BE USED

80
70

CONDITIONING DUE TO
COARSE, FRICTION-SOIL

60

LOW RATE
CONDITIONING

50
40
30

CONDITIONING MAINLY
DUE TO ADHESIVNESS
OF SOIL

NOT SUITABLE FOR

EPB TUNNELLING

20
10
0
2

0.001

6 8

0.01

6 8

0.1

6 8

6 8

10

6 8

100

GRAIN SIZE d [mm]

figure 12: EPB limits

Due to the new generations of conditioning additives, their easy use on site and thanks to their
high influence on the soil more and more EPB machines will be used instead of slurry machines.

4 Acknowledgements
The author acknowledge the following people for their excellent work regarding the presented
projects in this paper:
Peter Ellenberger, MBT International, Zrich
Dr. Mustapha Sari, Rhodia SA., Paris

5 References
[1] Advantages of soil Conditioning in shield tunneling: Experiences of LRTS Izmir
Jancsecz, Krause, Langmaack
ITA 1999 Oslo, S. 865 ff., Balkema ISBN 90 5809 063 9