Engineering Geology 196 (2015) 2436

Contents lists available at ScienceDirect

Engineering Geology
journal homepage: www.elsevier.com/locate/enggeo

Effect of high hyperbaric pressure on rock cutting process

M. Alvarez Grima a,, S.A. Miedema b, R.G. van de Ketterij a, N.B. Yenigl a, C. van Rhee b
a
b

MTI Holland B.V., Smitweg 6, 2963 AW Kinderdijk, The Netherlands

Delft University of Technology, Section of Dredging Engineering, Mekelweg 2, 2628 CD Delft, The Netherlands

a r t i c l e

i n f o

Article history:
Received 2 July 2014
Received in revised form 5 June 2015
Accepted 20 June 2015
Available online 27 June 2015
Keywords:
Rock cutting
Hyperbaric pressure
Fracture propagation
Time dependence

a b s t r a c t
When cutting rock hyperbaric, two cases may occur. The rock may encounter dilation or compaction due to shear.
Dilation results in pore under pressures, while compaction results in pore overpressures. Dilation will increase
the cutting forces considerably, while compaction may decrease the cutting forces. In both cases the cutting process is supposed to be cataclastic. To dimension cutting tools for deep sea mining, the worst case should be investigated, which is the dilatant case. To understand the cutting mechanism experiments are carried out in a
pressure tank, simulating the hyperbaric conditions. Hyperbaric cutting appears to be very different from atmospheric cutting due to the pore water pressures. The experiments have revealed that the cutting mechanism
changes from a chip type mechanism under atmospheric conditions to a cataclastic (crushed) type under hyperbaric conditions, resulting in higher cutting forces.
An analytical model is presented to estimate the cutting forces under high hyperbaric conditions. The results obtained with the analytical model agree rather well with the experimental data.
2015 Elsevier B.V. All rights reserved.

1. Introduction
When cutting rock hyperbaric, two cases may occur. The rock may
encounter dilation or compaction due to shear. Dilation results in pore
under pressure, while compaction results in pore over pressures. Dilation will increase the cutting forces considerably, while compaction
may decrease the cutting forces. In both cases the cutting process is supposed to be cataclastic. To dimension cutting tools for deep sea mining,
the worst case should be investigated, which is the dilatant case. To understand the cutting mechanism experiments are carried out in a pressure tank, simulating the hyperbaric conditions.
Hyperbaric cutting appears to be very different from atmospheric
cutting due to the pore pressures. The experiments have revealed that
the cutting mechanism changes from a chip type mechanism under atmospheric conditions, to a cataclastic (crushed) type under hyperbaric
conditions.
In front of the chisel the rock is crushed and shearing of the crushed
rock results in dilation, resulting in pore under pressures. These under
pressures increase the effective stress and thus also the frictional
shear stress. These under pressures depend on the magnitude of the dilation or the magnitude of the dilation and the permeability of the
crushed rock and are limited by the water vapor pressure. Because of
the very low permeability of the crushed rock, cavitation is expected
to be in effect of low to very low cutting velocities. The experiments

Corresponding author.
E-mail address: m.alvarezgrima@mtiholland.com (M. Alvarez Grima).

http://dx.doi.org/10.1016/j.enggeo.2015.06.016
0013-7952/ 2015 Elsevier B.V. All rights reserved.

were carried out at different velocities in order to quantify this effect.

From these experiments it was found that cavitation occurred already
at low velocities and that the forces can be predicted well with the modied sand cutting equations. Further it appeared that the cutting mechanism has changed in more than one way. Not only the mechanism
become cataclastic, but also the 3D chip pattern with a sideways
shape has become more a box cut, just following the shape of the chisel.
The experiments have proven that in the type of rock chosen, strong hyperbaric effects occur, which in terms of cutting forces, can be described
for the cavitating case, with the theory given in the paper.
The increase in cutting forces can be explained by analyzing the
combined effect of cutting speed and hyperbaric pressure during the
rock cutting process.
According to previous studies reported in the literature, it appears
that the understanding of the cutting mechanism at high hyperbaric
pressures is rather limited. Most of the studies conducted are mainly
concerned with drill bits; cutting a very thin layer of rock (b1 mm).
An understanding of the mechanism of rock cutting at large water
depth is a requisite for proper design of rock cutting tools cutting a
layer of centimeters.
Several rock cutting theories have been published in the literature
such as (Evans, 1961, 1962, 1965; Goktan, 1995, 1997; Nishimatsu,
1972). These theories concern rock cutting under dry and/or atmospheric conditions. They cannot be used to estimate the rock cutting
forces under hyperbaric conditions because the pore water pressure
and the effect of the cutting speed are not taken into account.
Kaitkay and Lei (2005) conducted lab experiments on the inuence
of hydrostatic pressure on rock cutting with drill bits on Carthage

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

marble. They found that the increase of conning pressure transformed

the cutting process from a brittle to a ductilebrittle failure mode. Longer chips were formed and the cutting forces increased with high hydrostatic pressure.
Van Kesteren (1995) examined the effect of pore water pressure in
rock by distinguishing two limiting conditions: drained and undrained.
In the drained condition, pore water ow due to pore water pressure gradient is possible without affecting the porous system itself. In the undrained condition pore water is not allowed to ow through the pores
and the pore water pressure will affect the stress state in the rock fabric.
The drained condition, undrained condition, and the transition zone are
determined by the Peclet number. It relates the cutting speed, cutting
depth, and the diffusivity coefcient, which in turns depends on the permeability, compressibility of the rock skeleton, compressibility of water,
compressibility of rock grains, and porosity. Additionally van Kesteren
(1995) explained that the pore pressures are of inuence on crack initiation and propagation. He argues that at the strain rates occurring during
dredging (order of 103 s1), the pore water is able to ow towards the
crack tips (drained condition), the transition towards undrained condition occurs at strain rate larger than 105 s1. This implies that crack initiation is not impeded by the resistance of the water pressure. At high
hyperbaric conditions, however, a different situation might occur regarding the speed of the process, strain rate, and ambient pressure. This in
fact, constitutes the main objective of this study. The magnitude of this
hyperbaric effect will probably depend on the ratio of the hydrostatic
pressure and the unconned compressive strength of the rock.
Detournay and Atkinson (2000) investigated the inuence of pore
pressure on the drilling response in low-permeability shear dilatant
rocks. Different pore pressure regimes were identied, which were controlled by a dimensionless number. They found that in the high speed
regime, the rock in the shear zone is undrained and pressure drops induced by shear-induced dilatancy, which leads to cavitation.
Huang et al. (1999) investigated the effect of cutting depth numerically by using DEM. They found that the ductile failure mode in rock cutting is characterized by a steady ow of a crushed material ahead of the
cutter. The brittle failure is characterized by the coalescence of micro
cracks and possibly formation of chips. They concluded that the transition between ductile and brittle failure mode in rock cutting depends
on the depth of the cut.

25

Al-Shayea et al. (2000) investigated the effect of conning pressure

and temperature on mixed-mode (III) fracture toughness of a limestone. Tests were conducted under an effective conning pressure of
28 MPa, and a temperature of up to 116 C. They found a substantial increase in fracture toughness under conning pressure. The pure mode-I
fracture toughness KIC increased by a factor of about 3.7 under a conned pressure of 28 MPa compared to that under atmospheric pressure.
The pure mode-II fracture toughness KIIC increased by a factor of 2.4 for a
conning pressure of 28 MPa. The effect of temperature was only 25%
more for KIC at 116 C.
Sang et al. (2003) investigated the strain-rate dependency of the dynamic tensile strength of rock. The fracture processes were analyzed at
various strain-rates. They found that higher strain rates generated a
large number of micro cracks, which interfered with the formation of
the fracture plane. The observed increase in dynamic strength at high
strain rate was caused by crack arrests due to the generation of a large
number of micro cracks.
Funatsu et al. (2004) studied the combined effect of increasing temperature and conning pressure on the fracture toughness of clay bearing rocks. They found that the fracture toughness of sandstone
increased by approximately 470% at 9 MPa connement over its value
at atmospheric pressure.
Schmidt and Huddle (1997) investigated the effect of conning pressure on fracture toughness of Indiana limestone. They found that KIC increased from 0.93 MN m3/2 at atmospheric pressure to 4.2 MN m3/2
at a conning pressure of 62 MPa.
Zijsling (1987) found that due to low permeability, cavitation will
occur in the crushed zone, even with very small layer thicknesses. The
result, combined with narrowing to a box cut, implies that the full
width of the cut has to be covered with chisels/pick points. So different
rows of chisels have to be staggered in contrary with atmospheric cutting where there is overlap due to the 3D effect.
From the studies presented in the literature, it can be concluded that
high hyperbaric pressure affects the rock behavior and particularly the
fracture toughness, crack initiation and propagation. This of course
will depend on the rock material properties such as porosity, permeability, and elasticity (plasticity). It is expected that cutting rock at large
water depth will have a strong effect on the magnitude of the cutting
forces and energy required.

2. Time dependency of fracture initiation and propagation

This section discusses the impact of time and speed on the cutting process. The cutting process is divided into three different subprocesses (i)
forming of a crushed zone, (ii) fracturing by shear failure, and (iii) fracturing by tensile failure. Each of these sub-processes is inuenced by the speed
of the cutting process. A dimensional analysis has been carried out to estimate the impact of the different factors on the cutting process and establishing in this way the basis for the selection of the parameters that will be investigated in the laboratory (see Table 1). The dimensional analysis was
carried out by using the Buckingham theorem.
Fig. 1 schematizes the phenomena involved in the chip forming process during rock cutting for shallow (Verhoef, 1997), and for deep water conditions. It is assumed that under high hyperbaric pressure (about 20 MPa) the failure mechanism will be predominantly shear, in contrast to a typical
shallow cutting process where the failure mechanism will be predominantly tensile. The extension of the crushed zone in front and below the cutting
tooth is expected to be larger for high hyperbaric pressure.
(i) Forming of a crushed zone: In this part of the process the cutting tool penetrates the rock and crushes it. The rock compressive strength will be
exceeded. Grains will be pulled out of the joints, pulverized, and pushed into the pores of the material further away from the cutting tool. The
water in the pores will be pushed further away into the material, resulting in high pore pressure. Water ow in the pores is governed by
Darcy's law. If the cutting speed is increased, the uid velocity in the pores will have to increase too. This can only occur if pressure difference
increases. The pore pressure near the tool tip has to decrease considering a pore pressure away from the tool tip approximately equal to the
hydrostatic pressure. This has an inevitable inuence on the required cutting force, which will increase linearly with pore pressure difference
near the tip.
Table 1 lists the parameters inuencing the cutting process. When the hydrostatic pressure (Phyd), the volume of crushed zone (Vcr), and viscosity () are chosen as running variables the following relation emerges by using the derived dimensionless numbers:
"
#
3
3
3
P hyd
t Dgrain Egrain Ematrix P hyd
V 2=3
Encr
p k L3
V

cr
11 12 13 Ltool
;
y

;
;
;
;
;
;
;
;
;
;
;
:
P hyd
V cr
P hyd V 2cr V cr P hyd
V 1=3

V cr P hyd P hyd
P hyd P hyd P hyd V cr
2
cr

26

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

Table 1
Parameters inuencing the cutting process.
Symbol

Parameter

Units

t
Dgrain
Phyd
Egrain
Ematrix

11
22
33
Vcr
Encr
Ltool
k
p
L

V
Enf
Lf
KIC
pp
Vx
wf
vl

Time
Diameter of the grains
Hydrostatic pressure
Youngs' modulus of the grains
Youngs' modulus of the material
Density water
Stress in the material
Stress in the material
Stress in the material
Volume crushed zone
Energy crushed zone
Length dimension of the tool
permeability
Pressure difference
Distance over which the pressure difference acts
Viscosity
Speed of the water through the material
Energy fracture zone
Fracture length
Fracture toughness
Fluid pressure in fracture
Fluid speed in fracture
Fracture width
Leak-off coefcient

s
m
N/m2
N/m2
N/m2
N s2 m4
N/m2
N/m2
N/m2
m3
Nm
M
m2
N/m2
m
N s/m2
m/s
Nm
m
Nm3/2
Nm2
ms1
m
Nm1/2

t

It becomes clear from the dimensionless number t hyd

that time has an equivalent inuence on the crushing process as the hydrostatic
Encr
also increases when either
pressure. Also the dimensionless number Encr Phyd

V cr shows that the energy required for the crushing process
L3
shows
that an increase of one of
the hydrostatic pressure, or the volume of the crushed zone increases. The dimensionless number Ltool Vtool
cr
the important dimensions of the cutting tool such as the width, or the contact area, has a signicant inuence on the volume of the crushed
zone and consequently on the energy involved in the crushing process.
(ii) Shear failure: Perpendicular to the crushed zone, approximately perpendicular to the upper surface of the tooth, a shear failure, which is induced by shear stresses caused by the tool and the changes in the crushed zone will occur (see Fig. 1). This fracturing in sliding mode, refereed
as shear fracture (van Kesteren, 1995) occurs when the mode II stress intensity factor near the tip of the micro fracture in the material reaches
a critical value.

Shallow Water

Tensile Failure

Produced chip

Shear failure

Crushed zone

Deep Water

Produced chip

Tensile Failure
Shear failure

Crushed zone

Fig. 1. Phenomenological description of the chip forming process.

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

27

Table 2
Experimental program.
Test no.

Chisel width
(mm)

Wear angle
(o)

Hyperbaric pressure
(MPa)

Cutting velocity
(m/s)

Cutting depth
(mm)

Cutting angle
(o)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

21
21
21
21
21
21
21
21
21
21
21
21
21
21
21

10
10
10
10
10
10
10
10
10
10
10
10
10
10
10

0
18
1.5
0
1.5
3
6
18
18
18
0
3
6
18
6

0.2
0.2
0.2
2.0
2.0
2.0
2.0
2.0
0.6
0.01
0.01
0.2
0.2
1.2
1.2

20
20
20
20
20
20
20
20
20
20
20
20
20
20
20

68
68
68
68
68
68
68
68
68
68
68
68
68
68
68

(iii) Tensile failure: Tensile stress increase as the horizontal distance from the tool tip increases. At a certain moment, the ratio between the mode I
and mode II stress intensity factor near the tip of the propagating shear fracture will reach a critical level. At this point the shear fracture will
bifurcate into a tensile fracture. As soon as the fracture starts propagating, the pressure prole in the fracture inuences the fracture propagation to a large extent. The pressure prole in the fracture is determined by three factors: mass balance in the fracture, viscous uid ow, and
elastic (or plastic) deformation of the fracture (Weijers, 1995). In view of the high pressures, the pressure dependency of the uid density
needs to be considered.
Fluid leak-off from the surrounding rock into the fracture (or from the fracture into the surrounding rock) typically follows the equation:
Kl
vl constant
p
tt i x

where Kl is the leak-off coefcient, which is linearly dependent on the permeability of the material. The term (t ti) indicates the time elapsed since
the fracture tip passed a certain part of the fracture wall. If inertia, compressibility and gravity are neglected the NavierStokes equation for uid ow
in x-direction simplies to a linear relation between uid velocity and pressure gradient:
vx

w2 px


12
x

where w is the local fracture width and is the uid viscosity.

Finally the pressure will depend on the entrance width, fracture length, Young's modulus of the material, and over pressure distribution in the
fracture.
Once a dimensional analysis using the hydrostatic pressure (Phyd), fracture length (Lf) and uid viscosity () as running variables (see Table 1) is
performed, then the following dimensionless groups are obtained.
"
p#
P hyd
L2f
pp vx
w f L f

vl
P hyd
t
K IC

:
p
p
;
y

;
;
;
;
;

P hyd P hyd
L f L f
2
P hyd
L f
P hyd
P hyd
L3f
Enf

Table 3
Rock properties at atmospheric conditions.
Test no.

UCS MPa

E (GPa)

()

BTS
(MPa)

k liquid
(m/s)

n
(%)

s (Mg/m3)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

7.92
7.92
7.92
8.75
8.75
8.75
8.75
9.29
10.62
10.64
8.86
8.86
8.86
10.54
10.54

5.95
5.95
5.95
7.53
7.53
7.53
7.53
5.89
8.32
9.01
8.20
8.20
8.20
9.98
9.98

0.31
0.31
0.31
0.25
0.25
0.25
0.25
0.27
0.23
0.27
0.31
0.31
0.31
0.33
0.33

0.88
0.88
0.88
1.09
1.09
1.09
1.09
1.15
1.05
1.13
0.86
0.86
0.86
x
x

3.1E06
3.1E06
3.1E06
8.5E07
8.5E07
8.5E07
8.5E07
1.4E07
2.8E07
2.2E08
1.5E07
1.5E07
1.5E07
3.4E09
x

37.86
37.86
37.86
34.64
34.64
34.64
34.64
33.17
31.66
33.92
35.12
35.12
35.12
35.89
x

2.78
2.78
2.78
2.76
2.76
2.76
2.76
2.76
2.78
2.79
2.77
2.77
2.77
2.80
x

28

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

Fig. 2. a) Experimental test set-up, hyperbaric tank, Ifremer, Brest. b) Cutting rig and hyperbaric tank Ifremer (Brest, France). c) Position of the sensors in the cutting rig.

The dimensionless number (t

P hyd
t
)

in Eq. (4) is the most interesting parameter for the purpose of this study. It suggests that time impact is

equivalent to the impact of the hydrostatic pressure, and inversely related to the uid viscosity. In other words, the inuence of time on the fracturing
process is equivalent to the inuence of pressure on the fracturing process.
x
When fracture length and hydrostatic pressure are chosen as running variables, another interesting dimensionless number arises: vx t
v
L f . This
dimensionless number shows that time is equally important as uid speed in the material, but more importantly, inversely proportional to fracture
length. As fracture length is proportional to the thickness of the layer cut during the cutting process, time is also inversely proportional to the layer
thickness.
Table 4
Results of executed tests.
Test
no.

Chisel
width
(mm)

Cutting
depth
(mm)

Hyperb.
pressure
(MPa)

Cutting
velocity
(m/s)

Actual cutting
velocity (m/s)

Average cutting
force
(kN)

Max cutting
force
(kN)

Min cutting
force
(kN)

Cuttin
cross
(mm2)
2
)

Specic
energy
(MJ/m3)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

21
21
21
21
21
21
21
21
21
21
21
21
21
21
21

20
20
20
20
20
20
20
20
20
20
20
20
20
20
20

0
18
1.5
0
1.5
3
6
18
18
18
0
3
6
18
6

0.20
0.20
0.20
2.00
2.00
2.00
2.00
2.00
0.60
0.01
0.01
0.20
0.20
1.20
1.20

0.188
0.178
0.200
1.826
1.717
1.740
1.702
1.577
0.618
0.010
0.017
0.202
0.207
1.238
1.188

7.22
9.25
10.42
8.09
11.17
12.23
13.19
20.70
22.72
4.94
4.72
11.36
11.29
12.74
10.90

9.8
13.1
14.4
12.5
12.6
14.9
18.3
28.5
12.3
7.5
6.5
16.2
15.3
17.4
12.6

5.1
5.2
5.6
3.8
8.1
10.6
9.8
15.8
13.1
2.4
2.5
7.5
7.4
8.2
7.9

811
729
795
820
577
543
655
562
581
831
1070
675
833
541
596

8.90
12.69
13.11
9.87
19.36
22.52
20.14
36.83
39.13
5.94
4.41
16.83
13.55
23.54
18.28

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

29

Table 5
Particle size distribution of cutting debris.
Test no.

Chisel width (mm)

Cutting depth (mm)

Hyperbaric pressure (bar)

Cutting velocity (m/s)

Gravel
N2 mm (%)

Sand
b2 mm
N63 m (%)

Fines
b63 m
N20 m
(%)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

21
21
21
21
21
21
21
21
21
21
21
21
21
21
21

20
20
20
20
20
20
20
20
20
20
20
20
20
20
20

0
180
15
0
15
30
60
180
180
180
0
30
60
180
60

0.20
0.20
0.20
2.00
2.00
2.00
2.00
2.00
0.60
0.01
0.01
0.20
0.20
1.20
1.20

83.55
76.65
76.68
80.30
54.55
54.76
51.12
57.17
68.98
88.84
85.05
78.77
66.32
66.86
73.07

16.43
23.35
23.32
19.69
45.45
45.24
48.86
42.78
30.89
11.14
14.90
21.23
33.67
33.14
26.93

0.02
0.01
0.00
0.01
0.00
0.00
0.01
0.05
0.13
0.01
0.05
0.00
0.01
0.00
0.00

3. Experimental investigation
This section presents the laboratory cutting experiments performed
on Savonnieres limestone. The laboratory work focuses on investigating
the effect of the hyperbaric pressure on the magnitude of the cutting
forces and consequently the required power. The parameters that
were varied in the rst experimental series are the cutting speed and
the ambient pressure. The cutting depth, tooth geometry and cutting
angle were kept constant (see Table 2).
Table 3 lists the rock properties determined at atmospheric conditions on the Savonnieres limestone samples used in the lab experiments. The unconned compressive strength UCS, Young's modulus E,
Poisson's ratio , Brazilian tensile strength, BTS, permeability k, porosity
n, and solid density s are measured rock properties from the tests performed by Delft University of Technology, Faculty of Civil and Geo Engineering. The tests were done according to the ASTM (American Society
for Testing and Materials).

3.1. Rock cutting experiments

The rock cutting tests were executed in a hyperbaric tank having a
height of 2.2 m and a diameter of 1 m. The tests were done by Deltares
at Ifremer, Brest in France (Fig. 2a) (van Kesteren, 2009). The pressure
rating is 1000 bar. The experiments were conducted at scale 1:1. This allows investigating the effect of the hyperbaric pressure on the cutting
process regardless of the scaling effects. Therefore, the formation of a

3.2. Results of experiments

Table 4 lists the results of the cutting experiments average horizontal cutting forces, minimum and maximum cutting forces, cut cross
section as determined with a laser device, and the specic energy. The
specic cutting energy SE (MJ/m3) is dened as the amount of energy
required per volume of excavated rock. Table 5 lists the particle sizes
of the cutting debris in terms of gravel, sand and nes fraction as
percentage.

25

20
Cutting forces, Fh (kN)

realistic crushed zone as well as a proper crack initiation and propagation

could be achieved without masking the effect of the ambient pressure.
The tooth used has a width of 21 mm, a clearance angle of 10 and a
cutting angle of 68. The cutting forces were measured with strain
gauges with maximum range of 50 kN in tension. The sensor was tailor
made by Deltares to accommodate the requirements of the Ifremer
tank. There is one horizontal sensor and two vertical sensors, which
measures the torque as well. The horizontal sensor is located behind
the chisel and the vertical sensors are located at the two sides of the cutting rig. This enables the measurement of the position of the cutting
force on the chisel (Fig. 2b, Fig. 2c). For each test the cutting velocity, displacement, cutting forces, and ambient pressure were recorded. After
completion of each test, the cutting debris were collected and analyzed
in order to relate the effect of the cutting speed and pressure with the
size of the cutting debris and tooth production.
The particle size distribution of coarse samples (63 m31 mm) was
determined by sieve analysis according to NEN 5753. Particles larger
than 31 mm were weighed individually whereas the distribution of
ne particles, less than 63 m were determined on a sub-sample of
3 gram dry soil and measured with Malvern Mastersizer 2000. A laser
device was used to scan the linear cutting grooves and to measure the
groove prole cross section. This allows getting more insight into the
combined effect of the cutting speed and pressure on the side breakout angle.

15

10

0
0

10

15

Hyperbaric pressure (MPa)

Fig. 3. Cutting forces versus hyperbaric pressure.

20

3.2.1. Effect of hyperbaric pressure and cutting speed on cutting forces

Fig. 3 and Table 4 show that in general the cutting forces increase as
the hyperbaric pressure increases. Under high hyperbaric pressure the
cutting process changes into an apparent ductile (cataclastic) mode
and failure of the rock will be predominately shear. In brittle cutting
process, the chip formation is dominated by tension cracks as commonly encountered in shallow rock cutting. However, when ductile behavior
is prevailing, the crack formation is mostly along shear planes and more
force will be required to create a chip. Besides, high hyperbaric pressure

30

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

(a)

(b)

(c)

(d)

Fig. 4. Overview of complete cut a) at atmospheric pressure with low cutting velocity (0.2 m/s) b) at high hyperbaric pressure (180 bar) with low cutting velocity (0.2 m/s), c) at atmospheric pressure with high cutting velocity (2 m/s) and d) at high hyperbaric pressure (180 bar) with high cutting velocity (2 m/s).

will lead to high friction forces between the material cut and the top
surface of the cutting tool (i.e., apparent ow mechanism). The crushed
zone below and in front of the tooth will increase.
The scatter in the magnitude of the cutting forces at 18 MPa pressure
in Fig. 3 is attributed to the effect of the cutting speed in combination
with the drained and/or undrained behavior of the material in relation
to its permeability and porosity. It was observed that at lower cutting
speeds the effect of the pressure is counteracted. The experimental results indicate that for low cutting speeds (v b 1 m/s) despite the ductile
behavior of the rock, much less changes in the magnitude of cutting
forces are observed. The results showed that the increase in cutting
forces varies from 4.7 kN at atmospheric conditions up to 22.7 kN at
18 MPa hyperbaric pressures for a cutting speed equal to 2 m/s. The increase in cutting forces is approximately ve times higher than the magnitude of the cutting forces at atmospheric conditions. Fig. 4 shows an
overview of complete cut, where the differences in side-break out
angle between atmospheric and hyperbaric pressures can be seen for
different cutting speeds.
As an example two selected tests are presented showing the cutting
forces versus time for both atmospheric conditions and hyperbaric conditions (Figs. 5 and 6). They correspond with test 1 and test 8 as listed in
Table 4. Clearly the gures show the combined effect of pressure and
cutting speed on the magnitude of the cutting forces.

3.2.2. Effect of hyperbaric pressure and speed on cutting debris size

Fig. 7 shows the average cross sectional area, Acr of a groove. The
cross sectional area is normalized with the applied cutting area of the
tooth, which equals to the product of the tooth width and the cutting
depth. As can be seen in Fig. 7a (speed = 0.2 m/s, pressure equals to atmospheric) a large side-break out angle occurred indicating a short
shear path reaching a bifurcation point from shear to tension earlier
than what is shown in Fig. 7d (speed = 2 m/s, hyperbaric pressure =
18 MPa).
Fig. 8 shows the ratio between the average cutting cross sectional
area and the cutting area of the tooth versus the cutting velocity. Fig. 9
shows the ratio between the average cutting cross sectional area versus
pressure times speed to illustrate the combined effect on tooth production. As can be seen in Figs. 8 and 9, at low cutting speed the area is
about 2.5 times the applied cutting area of the tooth, which is caused
by the small crushed zone and large chips that are formed (Figs. 10a
and 10b). This still holds for the tests conducted at atmospheric condition and high cutting speed (Fig. 10c), where the area is about 2 times
the applied cutting area of the tooth. At high cutting velocity, however,
when the hyperbaric pressure increases (i.e., p = 18 MPa) the production ratio drops to 1.3. In this situation rather than chips, lumps with the
appearance of clay are formed (Fig. 10d). This is a clear evidence of a
more ductile process with increasing cutting speed and pressure.

Fig. 5. Cutting forces versus time (Test 1, atmospheric conditions). Black solid line horizontal cutting force, red solid line vertical cutting force, blue solid line cutting speed.

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

31

Fig. 6. Cutting forces versus time (Test 8, 18 MPa). Black solid line horizontal cutting force, red solid line vertical cutting force, blue solid line cutting speed.

In general the grain size of the debris becomes ner as the hyperbaric pressure and cutting speed increases. The gravel fraction decreases
and the sand, and ne fractions increase with increasing cutting speed
(Table 5).

failure an equivalent shear strength c is determined, which is based on

the tensile strength of the rock.
Fig. 11 illustrates the forces on the layer of rock cut. The forces acting
on this layer are:

4. Analytical models

A normal force acting on the shear surface N1 resulting from the grain
stresses.
A shear force S1 as a result of internal ction N1 tan().
A shear force C as a result of the shear strength (cohesion) c. This force
can be calculated by multiplying the cohesive shear strength c with
the area of the shear plane.
A force normal to the tooth N2 resulting from the grain stresses.
A shear force S2 as a result of the soil/steel friction N2 tan() or external friction.

4.1. Analytical model for atmospheric conditions

The model for rock cutting under atmospheric conditions presented
here is based on the ow type of cutting mechanism. Although in general rock will encounter a more brittle failure mechanism and the ow
type considered represents the shear failure mechanism, the ow type
mechanism forms the basis for all cutting processes. In the case of brittle

(a)

(b)

(c)

(d)

Fig. 7. Composition of laser scan cut geometry a) at atmospheric pressure with low cutting velocity (0.2 m/s) b) at high hyperbaric pressure (180 bar) with low cutting velocity (0.2 m/s),
c) at atmospheric pressure with high cutting velocity (2 m/s) and d) at high hyperbaric pressure (180 bar) with high cutting velocity (2 m/s).

32

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

should be used. On the blade a force component in the direction of cutting velocity Fh and a force perpendicular to this direction Fv can be distinguished.
F h K 2
sin

F v K 2
cos :

The following equations for the horizontal Fh and vertical Fv cutting

forces are found.
Fh

c
hi
w
cos
sin
sin
sin

Fv

c
hi
w
cos
cos
sin
sin

Fig. 8. Ratio of the average cut cross sectional area to the cutting area of chisel versus cutting velocity.

The normal force N1 and the shear force S1 can be combined to a

resulting grain force K1.
The forces acting on a straight tooth when cutting rock, can be distinguished as:
A force normal to the blade N2 resulting from the grain stresses.
A shear force S2 as a result of the soil/steel friction N2 tan() or external friction.
Combining the forces N2 and S2 will lead to a resulting force K2,
which is the unknown force on the blade. By taking the horizontal and
vertical equilibrium of forces, an expression for the force K2 on the
blade can be derived.
The force C due to the cohesive shear strength c is equal to:

c
hi
w
C
sin

where is the cutting angle, is the external friction angle, and is the
internal friction angle.
The cohesion c is assumed to be about 50% of the UCS value, when
the internal friction angle is small or not taken into account.
To determine the shear angle where the horizontal force Fh is at a
minimum, the denominator of Eq. (8) has to be at a maximum. This occurs when the rst derivative of Fh with respect to equals to zero, and
the second derivative is negative.
sin
sin
sin 2
0

:
2
2

Fig. 9. Ratio of the average cut cross sectional area to the cutting area of tooth as a function
of the product cutting speed and hyperbaric pressure.

11

This gives for the cutting forces:

Fh

2
c
hi
w
cos
sin
H F
c
hi
w
1 cos

12

Fv

2
c
hi
w
cos
cos
V F
c
hi
w:
1 cos

13

where c is the shear strength (cohesion), hi is the cutting depth, w is the

tooth width, and is the shear angle.
The factor in Eq. (5) is the velocity strengthening factor, which
causes an increase of the cohesive shear strength. In clay (Miedema,
1992, 2010) this factor has a value of about 2 under normal cutting conditions. In rock the strengthening effect is not reported, so a value of 1

10

Fig. 12 shows the values of the horizontal cutting force coefcient

HF as a function of the blade angle and the internal friction angle
of the rock. (See Fig. 11.)
4.1.1. Validation of experiments under atmospheric conditions
As shown in Table 3, the rock used in the experiments had a UCS
value between 7.92 and 10.64 MPa, resulting in a shear strength c between 4 and 5.3 MPa, when the angle of internal friction is not taken
into account. The tensile strength of the rock was between 0.86 and
1.15 MPa (BTS see Table 3). According to Miedema (2014) the failure
mechanism of such a rock will be shear failure. The internal friction
angle of the rock is unknown, but an estimated value between 15 and
30 gives a reasonable range. According to Fig. 12 the value of HF is between 2 and 3.5, resulting in horizontal cutting forces between 3.4 kN
and 6 kN with a shear strength of 4 MPa, and between 4.5 kN and
7.9 kN with a shear strength of 5.3 MPa. This gives a total range from
3.4 kN up to 7.9 kN. On the one hand the atmospheric experiments
showed a strong 3D failure pattern (Fig. 7) with a cross section much
larger than the box cut of the tooth resulting in underestimation of
the cutting forces, on the other hand the theory calculates the maximum
cutting force at the start of the shear failure resulting in an overestimation of the cutting forces. Assuming that these two effects more or less
compensate one to each other, the range of the theoretical atmospheric
cutting forces is 3.4 kN to 7.9 kN, matching the measured atmospheric
cutting forces ranging from 4.72 kN to 8.09 kN (see Table 4).

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

(a)

(b)

33

(c)

(d)

Fig. 10. Production of the cut a) at atmospheric pressure with low cutting velocity (0.2 m/s) b) at high hyperbaric pressure (180 bar) with low cutting velocity (0.2 m/s), c) at atmospheric
pressure with high cutting velocity (2 m/s) and d) at high hyperbaric pressure (180 bar) with high cutting velocity (2 m/s).

4.2. Analytical model for hyperbaric conditions

The differences between rock cutting under atmospheric conditions
and under hyperbaric conditions is concerned with the extra pore pressure forces W1 and W2 on the shear plane and on the blade as explained
below. Fig. 13 illustrates the forces on the layer of rock cut. The forces
acting on the layer are:
A normal force acting on the shear surface N1 resulting from the grain
stresses.
A shear force S1 as a result of internal ction angle N1 tan().
A force W1 as a result of water under pressure in the shear zone.
A shear force C as a result of the cohesive shear strength c. This force
can be calculated by multiplying the cohesive shear strength c with
the area of the shear plane.
A force normal to the tooth N2 resulting from the grain stresses.
A shear force S2 as a result of the external friction angle N2 tan().
A shear force A as a result of pure adhesion between the rock
and the tooth a. This force can be calculated by multiplying
the adhesive shear strength a of the rock with the contact
area between the rock and the tooth. In most rocks this force
will be absent.
A force W2 as a result of water under pressure on the tooth.

The normal force N1 and the shear force S1 on the shear plane can be
combined to a resulting grain force K1
K1

q
N21 S21 :

14

The forces acting on a straight tooth when cutting rock, can be distinguished as:
A force normal to the tooth N2 resulting from the grain stresses.
A shear force S2 as a result of the external friction angle N2 tan().
A shear force A as a result of pure adhesion between the rock and the
tooth. This force can be calculated by multiplying the adhesive shear
strength a of the rock with the contact area between the rock and
the tooth. In most rocks this force will be absent.
A force W2 as a result of water under pressure on the tooth.
Fig. 14 shows the abovementioned forces. When the forces N2 and S2
are combined to a resulting force K2, and the adhesive force and the
water under pressures are known, then the resulting force K2 is the unknown force on the tooth. By taking the horizontal and vertical equilibrium of forces an expression for the force K2 on the tooth can be derived.
K2

q
N22 S22 :

15

The force K2 on the tooth is:

K2

W 2
sin W 1
sin C
cosA
cos

:
sin

16
The forces on the tooth can be derived from Eq. (16). On
the tooth a force component in the direction of the cutting
velocity F h and a force perpendicular to this direction F v can be
distinguished.
F h W 2
sin K 2
sin

17

F v W 2
cos K 2
cos :

18

The pore pressure forces can be determined in the case of fullcavitation or in the case of no cavitation according to:

Fig. 11. The forces on the layer cut in rock (atmospheric).

W1

w
g
z 10
hi
w
sin

or W 1

P 1m
hi
w
sin

19

34

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

Fig. 12. The ductile (shear failure) horizontal force coefcient.

W2

w
g
z 10
hb
w
sin

or W 2

P 2m
hb
w
sin

20

where P1m is the pore water pressure on the shear zone and P2m is the
pore water pressure on the tooth, hb is the tooth blade height, hi is the
cutting depth, z is the water depth, w width of the tooth, w water density, g gravitational acceleration, and cutting angle.
The forces C and A are determined by the cohesive shear strength c
and the adhesive shear strength a according to:
C

c
hi
w
sin

21

a
hb
w
:
sin

22

Fig. 13. The forces on the layer cut in rock (hyperbaric).

4.2.1. Validation of experiments under hyperbaric conditions

First of all it is assumed that the adhesive shear strength of the rock is
zero. According to Miedema (2014) the shear angle is about 2025
for the case considered here. Taking an internal friction angle of 20
and a shear angle of 22, together with a shear strength of 5.3 MPa,
the cutting forces can be determined as shown in Fig. 15.
The measured horizontal cutting forces seem to increase rapidly as
the water depth increases starting at zero water depth (atmospheric
conditions), but having a less steep increase for water depths larger
than 60 m. This may be explained by two effects: The rst effect is the
fact that at zero water depth the failure mechanism is brittle shear failure, meaning that the average cutting force will be smaller than the theoretically calculated cutting force. This failure mechanism transits to
ductile shear failure at larger water depths. This transition takes place
at a water depth of about 50100 m. The second effect is the 3D side
way cross section of the rock cut, being bigger than the box cut of the

Fig. 14. The forces on the blade in rock (hyperbaric).

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

35

Fig. 15. The theoretical and corrected cutting forces and the measured horizontal cutting forces.

chisel. Figs. 4 and 7 show that this second effect decreases with increasing water depth (hyperbaric pressure). The second effect does increase
the cutting forces since the cross section cut is larger than the chisel
width times the layer thickness. The following empirical equation
(Eq. (23)) takes both effects into account. The rst term between
brackets gives the rst effect; the transition from brittle shear failure
to ductile shear failure. The second term between brackets gives the second effect; the decreasing 3D cross section at large water depth. The coefcients 1, 2, and 3 in Eq. (23) depend on the rock properties. The
ones used in this study are 1 = 3.33, 2 = 200, and 3 = 400.

Fh; c Fh
1

 

1
2

1
:
z 10
z 10 3

23

It is important to mention that Eq. (23) is an empirical equation based

upon a limited set of experiments. For the other types of rocks the coefcients 1, 2, and 3 might be different than the ones used in this
study.
4.2.2. Inuence of the cutting velocity
From the experiments it is clear that there is an inuence of the cutting velocity on the cutting process. In general a higher cutting velocity
gives a higher cutting force, but there is a lot of scatter. The theoretical
cutting forces in the previous section are determined assuming full cavitation, but the question is Was there full cavitation during all experiments?. Miedema (1987, 2013) derived an equation to determine the
transition cutting velocity Vc between the non-cavitating regime and
the fully cavitating regime based on pore water ow according to
Darcy equation.
Vc

d1 z 10km
:
c1 h1

24

The proportionality coefcients c1 and d1 have values close to 0.6 and

4, respectively (Miedema, 1987). As can be seen in Table 3, the permeabilities of the rocks show a lot of scatter. The permeabilities measured
(km) differ by a factor of 1000. So it is not possible to determine the transition velocities exactly, but it is still possible to see if Eq. (24) may explain

the velocity effect. The answer to this question is: at the lowest permeability the transition velocity is about 0.2 m/s (at z = 1800 m) with a dilatation (pore volume increase) of 0.01, and 0.02 m/s with a dilatation of
0.1. At the highest permeability the transition velocity is 187 m/s (at z =
1800 m) with a dilatation of 0.01 and 18.7 m/s with a dilatation of 0.1. The
cutting velocities used in the experiments are within this range, so the
only conclusion that can be drawn is that some tests had full cavitation
where the cutting forces do not depend on the cutting velocity, while
with other tests this was not the case and the cutting forces depended
on the cutting velocity. More detailed information has to be available to
give a better prediction about the drained or undrained behavior of the
rock during the cutting process. For now, the conclusion is that the prediction based on the equations for full cavitation gives a good upper
limit for the cutting forces and thus the required cutting energy and
power.
5. Discussion and conclusions
The purpose of this study was to investigate the effect of high hyperbaric pressures on rock cutting performance. The following conclusions
are drawn from the analysis of the results:
1. In general the cutting forces increase as the hyperbaric pressure increases. This can be explained by taken into account that under
high hyperbaric pressures the brittle behavior of the material and
the brittle cutting process changes into an apparent ductile mode.
This effect, however, is more noticeable for the combined effect of
high cutting speed and high pressure.
2. The experiments showed that the cutting forces at large hydrostatic
pressure (18 MPa) can be about four to six times higher than the cutting forces at shallow ambient pressure or atmospheric conditions.
3. It was observed that when cutting rock at high hyperbaric pressures,
the side-break out angle is much narrow (i.e., box cut) than the sidebreak out angle as commonly found when cutting rock at atmospheric conditions or dry conditions. This results in a decrease in tooth
production.
4. The experiments reveal that contrary to dry and or atmospheric rock
cutting the effect of speed in combination with the hyperbaric

36

M. Alvarez Grima et al. / Engineering Geology 196 (2015) 2436

pressure is signicant on the magnitude of the cutting forces and required energy. This is due to the crushed rock created in front of the
tooth resulting in dilation and pore under pressures.
5. The analytical models presented in this paper are an extension of the
models developed by Miedema (1987). The calculations show that
the analytical models can reproduce the measured values rather
well. It is important to mention, however, that the calculations
done with the hyperbaric cutting model assume full cavitation. The
results show a good upper limit for the cutting forces and thus the required cutting energy and power.
Acknowledgments
This research has been sponsored by the Royal IHC (IHC). We would
like to thank Mr. W.G.M. van Kesteren and Mr. J. Pennekamp from
Deltares for the execution of the rst series lab experiments of this
study. We also would like to thank Mr. Y Le Guen from the laboratory
of Ifremer, Brest, France.
References
Al-Shayea, N.A., Khan, K., Abduljauwad, S.N., 2000. Effect of conning pressure and temperature on mixed mode (III) fracture toughness of a limestone rock. Int. J. Rock
Mech. Min. Sci. 37, 629643.
Detournay, E., Atkinson, C., 2000. Inuence of pore pressure on the drilling response in
low-permeability shear-dilatant rocks. Int. J. Rock Mech. Min. Sci. 37, 10911101.
Evans, I., 1961. A theory of the basic mechanics of coal ploughing. Proceedings of the International symposium on mining research, University of Missouri, Oxford, Vol. 2,
pp. 761768.
Evans, I., 1962. Min. Res. 2, 761.
Evans, I., 1965. The force required to cut coal with blunt wedges. Int. J. Rock Mech. Min.
Sci. 2, 112.
Funatsu, T., Seto, M., Shimada, H., Matsui, K., Kuruppu, M., 2004. Combined effects of increasing temperature and conning pressure on the fracture toughness of clay bearing rocks. Int. J. Rock Mech. Min. Sci. 41, 927938.

Goktan, R.M., 1995. Prediction of drag bit cutting force in hard rocks. Proceedings of the
Third International Symposium on Mine Mechanization and Automation, Golden,
Colorado, Vol. 1, pp. 1038.
Goktan, R.M., 1997. A suggested improvement on Evans cutting theory for conical bits.
Proceedings of the Fourth International Symposium on Mine Mechanization and Automation, Brisbane, Queensland, Vol. 1 (p. A4:5761).
Huang, H., Detournay, E., Bellier, B., 1999. Discrete element modeling of rock. In: Amadei,
Kranz, Scoott, Smeallie (Eds.), Rock Mechanics for Industry. Balkema, Rotterdam.
ISBN: 90 5809 052 3.
Kaitkay, P., Lei, S., 2005. Experimental study of rock cutting under external hydrostatic
pressure. J. Mater. Process. Technol. 159, 206213.
Miedema, S.A., 1987. The Calculation of the Cutting Forces When Cutting Water Saturated
Sand (PhD thesis), Delft University of Technology, Delft (September).
Miedema, S.A., 1992. New Developments of Cutting Theories with Respect to Dredging,
the Cutting of Clay. WODCON XIII. World Dredging Association (WODA), Bombay,
India.
Miedema, S.A., 2010. New Developments of Cutting Theories with Respect to Offshore Applications. ISOPE. ISOPE, Beijing, China, p. 8.
Miedema, S.A., 2014. The Delft Sand, Clay & Rock Cutting Model. Delft University of Technology, IOS Press.
Nishimatsu, Y., 1972. The mechanics of rock cutting. Int. J. Rock Mech. Min. Sci. 9,
261271.
Sang, Ho Cho, Ogata, Yuyi, Kaneko, Katsuhiko, 2003. Strain rate dependency of the dynamics tensile strength of rock. Int. J. Rock Mech. Min. Sci. 40, 763777.
Schmidt, R.A., Huddle, C.W., 1997. Effect of conning pressure on fracture toughness of
Indiana limestone. Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 14, 289293.
van Kesteren, W.G.M., 1995. Numerical simulations of crack bifurcation in the chip
forming cutting process in rock. In: Bakker, G., Karihaloo, B.L. (Eds.), Fracture of Brittle
Disordered Materials: Concrete, Rock and Ceramics. ISBN: 0 419 19050 3.
Van Kesteren, W.G.M., 2009. Hyperbaric rock cutting experiments. Technical Report,
Deltares.
Verhoef, P.N.W., 1997. Wear of Rock Cutting Tools: Implications for the Site Investigation
of Rock Dredging Projects (PhD thesis). Delft University of Technology 905410434 1.
Weijers, L., 1995. The Near-wellbore Geometry of Hydraulic Fractures Initiated from Horizontal and Deviated Wells (PhD-thesis). Delft University of Technology.
Zijsling, D., 1987. Single cutter testing a key for PDC bit development (SPE 16529). Offshore Europe 87. Society of Petroleum Engineers, Aberdeen, Scotland.