ArticlePDF Available

Hole quality improvement in CFRP/Ti6Al4V stacks using optimised flow rates for LCO2 and MQL sustainable cooling/lubrication

Authors:

Abstract

Carbon fibre reinforced polymer/titanium stacks (CFRP/Ti6Al4V) are employed in aeronautics due to their excellent weight-to-strength ratio and corrosion properties. However, these same material properties present challenges for hole making which cannot be solved using conventional water-based metalworking fluids (MWFs), as they cause degradation of the composite. Moreover, environmental and health concerns require exploration of alternative cooling/lubrication solutions. In this study, a controlled mixture of liquid carbon dioxide (LCO2) and minimum quantity lubrication (MQL) was supplied through the drilling tool. The effect of varying LCO2 and MQL flow rates was evaluated on cutting forces, temperatures, and several hole quality outputs. The optimal flow rates were then determined through multi-objective optimisation. The results show that the cooling/lubrication flow rate greatly affects the measured outputs, and that supplying LCO2 + MQL with optimised flow rates helps achieve superior quality holes in CFRP, Ti6Al4V and CFRP/Ti6Al4V stacks.
Composite Structures 329 (2024) 117687
Available online 2 November 2023
0263-8223/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
Hole quality improvement in CFRP/Ti6Al4V stacks using optimised ow
rates for LCO
2
and MQL sustainable cooling/lubrication
I. Rodriguez
a
,
b
,
*
, P.J. Arrazola
a
, M. Cuesta
a
, F. Puˇ
savec
b
a
Mondragon Uniberstitatea, Faculty of Engineering, Loramendi 4, Arrasate-Mondrag´
on 20500, Spain
b
University of Ljubljana, Faculty of Mechanical Engineering, Ljubljana, Slovenia
ARTICLE INFO
Keywords:
CFRP/Ti stacks
Cutting temperature
LCO
2
assisted drilling
Surface integrity
Optimisation
ABSTRACT
Carbon bre reinforced polymer/titanium stacks (CFRP/Ti6Al4V) are employed in aeronautics due to their
excellent weight-to-strength ratio and corrosion properties. However, these same material properties present
challenges for hole making which cannot be solved using conventional water-based metalworking uids (MWFs),
as they cause degradation of the composite. Moreover, environmental and health concerns require exploration of
alternative cooling/lubrication solutions. In this study, a controlled mixture of liquid carbon dioxide (LCO
2
) and
minimum quantity lubrication (MQL) was supplied through the drilling tool. The effect of varying LCO
2
and MQL
ow rates was evaluated on cutting forces, temperatures, and several hole quality outputs. The optimal ow rates
were then determined through multi-objective optimisation. The results show that the cooling/lubrication ow
rate greatly affects the measured outputs, and that supplying LCO
2
+MQL with optimised ow rates helps
achieve superior quality holes in CFRP, Ti6Al4V and CFRP/Ti6Al4V stacks.
1. Introduction
The European aviation industry is committed to reach net zero CO
2
emissions from all ights departing from the EU by 2050 [1]. One
possible pathway to achieve this is to reduce fuel consumption by
reducing take-off weight. Consequently, multilayer composite/metal
stacks have been increasingly used in the manufacture of aircraft
structural components, due to their high weight-to-strength ratio,
corrosion resistance, and thermal stability. Carbon bre reinforced
polymers (CFRP) and alpha/beta titanium alloys, like Ti6Al4V, are two
representative materials used to produce these sandwich materials. In
modern aircrafts, as the Airbus A350 and Boeing 787 Dreamliner, these
materials are found in stacked conguration in some parts of the fuse-
lage, central wing box, and wing spar panels [2].
Rivets or screws are typically used to join composite and metal
components and form stacks in aerospace structural applications, mak-
ing mechanical drilling an essential machining operation in the aero-
space industry [3]. Drilling CFRP and Ti6Al4V plates in one operation
reduces positioning errors and cycle times. However, the disparate na-
ture of the materials lead to machinability and hole quality issues. The
abrasive carbon bres wear down the tool coating [4] increasing the risk
of adhesion wear when machining the Ti6Al4V phase [5]. This
combined abrasive and adhesive wear increases the tool wear rate
compared to drilling both materials separately [2,3].
Hole defects when drilling CFRP/Ti6Al4V stacks depend on the
drilling direction. Xu et al. have demonstrated experimentally [3] and
using nite element method (FEM) simulations [6] that the drilling from
CFRP to Ti6Al4V produces the best quality holes. Yet, even in this
drilling order, several hole quality issues persist (Fig. 1a). For instance,
peel-up delamination appears at the entrance of the CFRP phase as the
rake face of the drill pulls up the rst plies of the laminate like a cork-
screw [2,4]. Fibre pull-out also occurs in the machined surface, as both
the utes of the drill bit and the titanium chips being evacuated pull out
bre bundles as the hole is drilled [7]. Additionally, the heat generated
when drilling Ti6Al4V (about 500 C at the tool lip at V
c
=30 m/min in
dry [8]), can lead to carbonisation of the composite matrix at the
interface region as it exceeds its glass transition temperature (180 C)
[9,10]. Drilling induced defects in the Ti6Al4V phase include poor
surface roughness and excessive burr height. These problems can affect
component mechanical properties and aircraft part assembly, resulting
in the need to re-machining or rejecting parts.
Various solutions have been explored to overcome these machin-
ability and hole quality issues [11]. For instance, special tool geometries
(Fig. 1b) that minimise thrust force and thus the drilling induced
* Corresponding author.
E-mail address: irodriguez@mondragon.edu (I. Rodriguez).
Contents lists available at ScienceDirect
Composite Structures
journal homepage: www.elsevier.com/locate/compstruct
https://doi.org/10.1016/j.compstruct.2023.117687
Received 22 August 2023; Received in revised form 17 October 2023; Accepted 30 October 2023
Composite Structures 329 (2024) 117687
2
delamination in composite materials [12]. Brad point and double point
angle tools are the most popular, and are even commercially available
(SECO FeedMax drills for CFRP/Ti6Al4V stacks). Brad point tools reduce
the thrust force thanks to peripheral cutting edges, while double point
angle tools have a small point angle at the secondary stage which helps
to minimise delamination in the composite and reduce the length of
metal chips due to a change in chip ow direction through the double
cutting edges [4]. These tools can also be made from super-hard mate-
rials such as polycrystalline diamond (PCD), or diamond coated tung-
sten carbide to resist the abrasive wear from the hard reinforcement
bres.
Techniques which focus on improving chip evacuation, such as vi-
bration assisted drilling (VAD, in Fig. 1c) and orbital drilling (OD, in
Fig. 1d), have been also widely studied for improving hole quality in
Fig. 1. Overview of advances in mechanical drilling of CFRP/Ti6Al4V stacks: a) Applications and hole quality defects; b) e): Most common hole quality
improving techniques.
Table 1
Summary of the outputs measured by the most relevant studies concerning drilling of CFRP and Ti6Al4V materials employing techniques to supress machinability and
hole quality problems.
Study Technique Material Forces Temperature Hole quality
Hole entry/exit
(Delamination,
burr height etc.)
Hole surface (Fibre pull-out,
surface roughness, etc.)
Dimensional accuracy
(Diameter, roundness)
[40] Vibration assisted drilling CFRP
[41] Vibration assisted drilling CFRP
[39] Vibration assisted drilling Ti6Al4V
[7] Vibration assisted drilling CFRP/Ti6Al4V stack
[42] Vibration assisted drilling CFRPTi6Al4V stacks
[43] Vibration assisted drilling CFRP/C45E stack
[10] Vibration assisted drilling CFRP/Ti6Al4V stack
[13] Vibration assisted drilling CFRP/Ti6Al4V stack
[14] Vibration assisted drilling CFRP/Ti6Al4V stack
[15] Orbital drilling CFRP
[44] Orbital drilling CFRP/Ti6Al4V stack
[45] Orbital drilling CFRP
[46] MQL CFRP/Ti6Al4V stack
[47] MQL CFRP/Ti6Al4V stack
[48] MQL CFRP/Ti6Al4V stack
[19] LN
2
CFRP
[20] LN
2
Ti6Al4V
[49] LN
2
CFRP and Ti6Al4V
(separately)
[5] LN
2
and LCO
2
(separately)
Ti6Al4V
[18] LCO
2
Ti6Al4V
[50] LCO
2
CFRP/Ti6Al4V stack
[23] LCO
2
+lubricant media CFRP and Ti6Al4V
(separately)
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
3
CFRP/Ti6Al4V stacks. In VAD a periodic displacement is applied to the
tool to promote discontinuous chip formation, in turn the drilling tem-
perature and the mechanical abrasion and thermal damage produced by
titanium chips on the CFRP phase are reduced [10,13,14]. OD on the
other hand, consists of using a milling tool with a diameter smaller than
the hole following a helical trajectory. This improves chip evacuation
and lowers the cutting temperature. The axial load is also reduced, as the
larger force component is directed in the tangential direction, reducing
CFRP delamination [15].
Metalworking uids (MWFs) are also used to reduce temperature and
improve chip evacuation and lubrication when cutting a wide range of
materials. Nevertheless, conventional oil and water-based MWFs are
unsuitable for drilling CFRP/Ti6Al4V stacks, since the moisture ab-
sorption by the composite can cause swelling and degradation of matrix-
reinforcement bond [16]. Additionally, conventional MWFs can cause
various environmental and health risks, such as skin diseases and res-
piratory problems [17]. Alternatives like minimum quantity lubrication
(MQL), or the use of liqueed gases, namely liquid nitrogen (LN
2
) or
liquid carbon dioxide (LCO
2
) have been researched for drilling CFRP/
Ti6Al4V stacks (Fig. 1e). MQL is the most researched technique, as it is
easier to implement in machine-tools than cooling using liqueed gases.
The main disadvantage of MQL is its low heat extraction capacity, but
this could be improved up to three times by combining it with sub-zero
liqueed gases, as proven by Jamil et al. [18]. LN
2
has proven to reduce
the thermal damage and increase hole accuracy when drilling CFRP [19]
as well as Ti6Al4V [20], due to its cold temperature (-197 C). On the
downside, since LN
2
needs to be kept at 197 C and at ambient pres-
sure, storing and/or delivering it through the tool are major technical
challenges. LCO
2
, on the other hand, can be stored at room temperature
and a pressure of 57 bar, which gives rise to the possibility of mixing it
with lubricating oils or delivering it through the machine spindle. When
LCO
2
is rapidly expanded to ambient pressure, its temperature instantly
drops to 79 C [21]. The relative ease of storage and management of
LCO
2
, and the ability to mix it with lubricating oils or deliver it through
the spindle has motivated the development of industrial-grade LCO
2
cooling/lubrication solutions [22].
Table 1 summarizes the main studies in which the techniques
described previously have been used to overcome tool life and hole
quality problems when drilling CFRP/Ti6Al4V stacks. Most studies
conclude that drilling from a CFRP to Ti6Al4V direction provides the
best hole quality results, provided that sufcient titanium chip evacu-
ation is ensured. VAD, OD and MQL lubrication are the most widely
researched techniques for CFRP/Ti6Al4V stacks. While LN
2
and LCO
2
cooling has been mostly studied for drilling CFRP and Ti6Al4V separate
plates, there is little published data on LCO
2
+MQL cooling/lubrication
for drilling stacked aeronautical materials (Table 1).
Regarding the outputs, few studies exist that monitor the cutting
temperature throughout the entire cutting process. Additionally, apart
from the studies by Xu et al. [10] and Hussein et al. [13], there is a lack
of research works that provide a comprehensive assessment of the
drilled hole quality, and link those results to cutting forces and
temperature.
Therefore, the present work aims to optimise LCO
2
+MQL delivery
to ensure the best hole quality when drilling CFRP, Ti6Al4V and CFRP/
Ti6Al4V stacks. For this purpose, a sensitivity analysis of independently
varying the LCO
2
and MQL ow rates (q
LCO2
and q
MQL
) was carried out
on cutting forces and temperatures. The analysis also considered several
hole quality defects, such as delamination (push-out for individual CFRP
plates and peel-up for CFRP/Ti6Al4V stacks), bre pull-out, and epoxy
matrix carbonisation for CFRP, as well as surface roughness and burr
height for Ti6Al4V. Dimensional accuracy (hole diameter and round-
ness) was monitored at different heights for each material.
A multi-objective optimisation (MOO) was conducted with the
monitored outputs for CFRP, Ti6Al4V and CFRP/Ti6Al4V stacks (the
measured hole quality parameters varied depending on the workpiece
material) to determine the optimal LCO
2
and MQL cooling and
lubrication ow rates (q
LCO2
and q
MQL
) for drilling each material. To
reduce the experimental campaign, an initial series of tests consisting of
16 different q
LCO2
/q
MQL
combinations was tested on separate CFRP and
Ti6Al4V plates, and the optimal cooling/lubrication ow rates were
employed to dene the input conditions for the CFRP/Ti6Al4V stack
drilling experiments.
The results discussed in this paper help to dene a process window
for drilling CFRP/Ti6Al4V stacks with LCO
2
based cooling/lubrication
to improve the resulting quality of the part and minimise the mechanical
and thermal loads on the tool. This study also adds to the knowledge of
heat generation when drilling composite/metal stacks under dry and
sub-zero cooling/lubrication.
2. Tools and materials
A Brad point tool (SECO SD205A-6.032-6R1-C2) with internal
cooling channels specially designed for through the tool LCO
2
delivery
was used for the tests. This tool geometry minimises defects in one of the
most crucial parts of composite/metal stacks, which is peel-up delami-
nation at the entry plane of the composite. By having a tip angle of 180
at the periphery, the forces get distributed in the axial direction, pushing
down the bres at the entrance, and preventing delamination. Addi-
tionally, SECO Brad point drill bits proved suitable to machine Ti6Al4V
with minimal microstructural damage [23]. The tools also have a
Table 2
Geometrical parameters of the Brad point drill.
Reference SD205A-6.0-32-6R1-C2
Target material CFRP/Ti6Al4V stacks (CFRP to Ti6Al4V)
Coating Diamond CVD
Point angle,
σ
[] 120/174
Diameter, Ø [mm] 6
Coolant channels 2 ×Ø 0.5 mm
Edge radius, r
c
[
μ
m] 11 ±2
Wedge angle, β [] 43.9 ±5.8
Tool geometry
Table 3
Mechanical properties of reinforcement bres and epoxy matrix of the aero-
nautical grade CFRP composite at room temperature.
Reinforcement bres Reference Sigratex CW205 TW2/2
Areal weight [g/m
2
] 205
Tensile strength [MPa] 2400
Tensile Modulus [GPa] 300
Fibre volume [%] 50
Epoxy matrix Reference Hexcel Hexow RTM 62
Tensile strength [MPa] 75
Tensile Modulus [GPa] 2.9
Strain [%] 3.4
Density [g/cm
3
] 1.14
Table 4
Chemical composition (% by weight) and mechanical properties of the Ti6Al4V
alloy, at room temperature.
Chemical
composition
Fe 0.15
V 4.14
Al 6.50
C 0.003
O 0.18
N 0.004
Y <0.001
Mechanical
properties
Tensile strength, Yield [MPa] Long.: 921; Trans: 967
Tensile strength, ultimate [MPa] Long.: 973; Trans.: 967
Elongation [%] Long.: 12.5; Trans.: 15
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
4
Chemical Vapour Deposition (CVD) diamond coating to resist the
abrasion caused by carbon bres, and a cutting edge radius of 11 ±2
μ
m. The cutting edge of the tools was characterized using an Alicona
Innite Focus SL optical 3D microscope, using the rotation unit with a ×
10 magnication, to obtain the 3D geometry of the tool. The main
geometric features of the tool are summarised in Table 2.
CFRP and Ti6Al4V plates of the same dimensions (40x250x5 mm
3
)
were used for the experiments. The CFRP plates were fabricated using
aeronautical grade carbon bre reinforcement bres and epoxy resin
and a Resin Transfer Moulding (RTM) press. 20 layers of reinforcement
were used per plate to achieve a 50% bre volume and the mould was
pressed at 5 Tn and 140 C for 180 min to cure the resin. After the epoxy
cured, a post-curing treatment was performed in an oven at 180 C for 2
h. The mechanical properties of the CFRP plates are summarized in
Table 3 and those of Ti6Al4V in Table 4.
Fig. 2. Scheme of the experimental procedure carried out: input parameters and output parameters monitored for the MOO in CFRP Ti6Al4V and CFRP/
Ti6Al4V stacks.
Table 5
Output parameters analysed in the study for different workpiece materials.
Output parameter CFRP Ti6Al4V CFRP/Ti6Al4V stacks
Thrust force, F
z
Torque, M
z
Temperature, T
Hole diameter, Ø
Roundness,
Arithmetic mean height, S
a
Maximum valley depth, S
v
Delamination, F
da
(Push-out) (Peel-up)
Matrix carbonisation
Average surface roughness, R
a
Maximum peak to valley height, R
t
Ten point burr height, S
10z
LCO
2
ow rate, q
LCO2
MQL ow rate, q
MQL
Table 6
Cutting forces and temperatures acquisition details.
Forces Equipment Kistler 9273
Acquisition rate [Hz] 2400
Filter Low pass at 150 Hz
Temperature Equipment Telops Fast M3k
Temperature ranges [C] 0 to 185; 114 to 377; 254 to 680
Image format [pixel] 256 ×320
Spatial resolution [
μ
m/pixel] 180
Acquisition rate [Hz] 1200
Exposure time [
μ
s] CFRP: 25
Ti6Al4V: 4
Reected temperature [C] 21.5
Emissivity,
ε
[%] 85
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
5
3. Methodology
3.1. Experimental plan
Fig. 2 summarizes the xed and variable inputs employed in the
experiments carried out on individual CFRP and Ti6Al4V plates and
CFRP/Ti6Al4V stacks, as well as the outputs monitored for each mate-
rial. The experimental campaign was divided into two stages: (i) cool-
ing/lubrication ow rate optimisation for drilling individual CFRP and
Ti6Al4V plates, and (ii) cooling/lubrication ow rate optimisation for
drilling CFRP/Ti6Al4V stacks.
The tool geometry described in section 2, cutting parameters (V
c
=
45 m/min; f
z
=0.1 mm/rev recommended by SECO Tools), and work-
piece dimensions were xed for all experiments. The MQL oil used was
Rhenus Lub SSM neat oil (viscosity 3.5 mm
2
/s at 20 C), which ensures
good mixing with LCO
2
, as demonstrated by[21,24]. For the cases in
which only MQL was delivered, the oil was supplied to the cutting zone
using room temperature gaseous CO
2
as the carrier medium at a ow
rate of 10 g/min. The variable input parameters and monitored output
parameters varied depending on the drilled workpiece material. These
are detailed in the following subsections.
To minimise the effect of tool wear on the evaluated outputs a new
tool was used after a certain number of holes for each material (25 for
CFRP, 16 for Ti6Al4V, and 12 for CFRP/Ti6Al4V stacks). The tools were
checked regularly between tests for signs of wear on the chisel or pe-
ripheral cutting edges.
3.1.1. Experiments in individual CFRP and Ti6Al4V plates
As Fig. 2 shows, 16 LCO
2
and MQL ow rates (q
LCO2
and q
MQL
) were
tested for individual CFRP and Ti6Al4V plates (black and blue dots).
Each experimental condition was repeated three times to determine the
uncertainty. The output parameters evaluated for individual plate dril-
ling experiments are schematically shown in the black (CFRP) and blue
(Ti6Al4V) columns in Fig. 2. Cutting forces and the temperature at the
exit plane were measured for both materials. Regarding the hole quality,
Fig. 3. Temperature calibration procedure: a) Calibration setup; b) Thermal eld measurements of the workpiece acquired with the infrared camera; c) Temperature
drop measured using the thermocouples and infrared camera with calibrated emissivity; d) Calibration of the reected temperature.
Fig. 4. Advanced delamination factor (F
da
) proposed by Davim et al. [32].
Fig. 5. Representation of individual desirability function.
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
6
the diameter was evaluated both for CFRP and Ti6Al4V plates, however,
different parameters were measured to assess the quality of the
machined surface and the defects at the exit plane of the hole depending
on the workpiece material.
For instance, when drilling Ti6Al4V the burr height was measured at
the exit plane using the ten-point height (S
10z
) parameter, in order to
reduce the uncertainty of irregular burrs [25]. The surface roughness
parameters measured were those which dene fatigue strength (average
surface roughness, R
a
and maximum peak to valley height, R
t
) [26].
Unfortunately, the roughness parameters measured to evaluate the
quality of the machined titanium surface cannot be used for CFRP. The
coexistence of reinforcement bres and epoxy resin makes the machined
surface inhomogeneous, and evaluating the surface roughness through
proles becomes difcult, since the results vary depending on the po-
sition and orientation of the prole. Duboust et al. [27] showed in their
study that the arithmetic mean height of area roughness (S
a
), and the
maximum valley depth (S
v
) surface texture parameters accurately
represent the average roughness of the surface and the presence of voids
due to bre pull-out on the machined CFRP surface. Therefore, in the
present research, S
a
and S
v
were measured to evaluate the surface
quality. No burr is formed when drilling CFRP, however the composite
plies are usually delaminated, and hence, push-out delamination was
evaluated to assess the quality of the hole at the exit plane.
In addition to all these outputs, the LCO
2
and MQL consumption of
each experimental condition were also considered to perform a multi-
objective optimisation (MOO) using the response surface method
(RSM). This helped to obtain the q
LCO2
and q
MQL
combination that
ensured the optimal overall value of the output responses, while
ensuring minimal LCO
2
and MQL consumption. The optimized q
LCO2
and
q
MQL
values for drilling CFRP and Ti6Al4V were employed as input
values for the set of experiments on CFRP/Ti6Al4V stacks, to reduce the
number of experimental runs. The output parameters taken into
consideration for the CFRP and Ti6Al4V individual plate drilling ex-
periments are listed in Table 5.
3.1.2. Experiments in CFRP/Ti6Al4V stacks
For the stack drilling experiments, a drilling direction from CFRP to
Ti6Al4V was chosen as recommended by SECO Tools and studies from
literature [3,6]. In addition to the optimised q
LCO2
and q
MQL
values for
drilling separate CFRP and Ti6Al4V plates obtained in the rst experi-
mental stage (section 3.1.1), two more cooling/lubrication conditions
were tested for CFRP/Ti6Al4V stack drilling, namely dry cutting, and
pure LCO
2
cooling (green dots in Fig. 2). Since fewer cooling/lubrication
conditions were tested out in CFRP/Ti6Al4V stacks (four conditions)
than in individual plates (16 conditions), the MOO was performed by
comparing all output results, without using RSM.
Additionally, a more comprehensive temperature analysis was car-
ried out for CFRP/Ti6Al4V stack drilling, which recorded the cutting
temperature from two different positions as it is shown in the green
column in Fig. 2: (i) from the exit plane, recording the temperature
through a pre-drilled 1 mm pilot hole eccentric to the drill bit axis, and
(ii) from the side plane, following the studies by [10,13,14,28].
Fig. 6. Different setups employed in the experiments: a) Layout for tool exit and pilot hole temperature recording; b) Temperature elds when recording the exit
plane; c) Temperature elds when recording through pilot hole; d) Layout for side wall temperature recording; e) Temperature elds when recording from the
side plane.
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
7
Therefore, each experimental condition was repeated six times (three for
each camera position).
The hole quality analysis for drilled CFRP/Ti6Al4V specimens was
also more detailed than for separate plates, as the roundness was
measured in addition to the hole diameter. Also, as the CFRP was sup-
ported by the Ti6Al4V, the delamination only occurred at the entry
plane (peel-up delamination). The defect measured at the exit plane of
the CFRP was the matrix carbonisation generated by the heat of the ti-
tanium plate under the CFRP phase. Burr height (S
10z
) and machined
surface morphology parameters (S
a
, S
v
for CFRP and R
a
, R
t
for Ti6Al4V)
were measured same as in the individual plate experiments. The output
parameters monitored for CFRP/Ti6Al4V stack drilling are summarized
in Table 5.
3.2. Evaluation of output parameters
3.2.1. Cutting forces and temperature
The detailed parameters for the force and temperature acquisition
systems can be found in Table 6. Thrust force and torque values were
measured with a Kistler 9273 dynamometer. The cutting force signals
were ltered with a 150 Hz low-pass lter, since the rst natural damped
frequency of the workpiece clamping system is 1500 Hz and we applied
a lter at 10% of this frequency. The temperature during the drilling
process was recorded using a Telops Fast M3k high speed infrared
camera (IR Cam), at a frame rate of 1200 Hz. The thermal imaging
equipment was calibrated using the Telops blackbody free calibration
technique [29] to minimise the effects of the surrounding environment.
However, it is known that the accuracy of thermal imaging can be
affected by numerous factors such as the workpiece material, and even
the ambient temperature [30]. When the temperature desired to mea-
sure is close to the ambient one, the radiation emitted by surrounding
objects can affect the readings by the IRCam, therefore the reected
temperature must be properly characterised [31].
The emissivity calibration of the CFRP and Ti6Al4V materials was
carried out following the study of Segurajauregui et al. [28]. The
workpieces were painted with a heat resistant black spray to minimise
reection, and two K-Type thermocouples (TC) (ElPro NiCr-Ni-K) were
embedded at two different depths and a distance of 0.5 mm from the
surface (Fig. 3a). Then, the temperature readings of the TCs were
compared to those of the IR Cam at the position of the TC tip (Fig. 3b),
and the emissivity of the camera was varied until the temperature
readings matched those of the TCs. The camera lens was placed at a
distance of 180 mm from the workpiece, similar position to the drilling
experiments. The CFRP samples were heated up to 140 C to avoid
damaging the epoxy matrix, while the Ti6Al4V samples were heated up
to 260 C (maximum temperature of the furnace). The temperature
change due to conductivity across the 0.5 mm of material between the
surface of the workpiece and the position of the TC was taken into ac-
count. The temperature readings did not change for different depths of
the TC. As seen in Fig. 3c an emissivity of
ε
=0.85 matched best the
temperature of the thermocouple inserted into the CFRP and Ti6Al4V
specimens painted in black. Regarding the reected temperature, it was
calibrated for every experiment following the diffuser reector method
[31]. The drilled workpiece was employed as the heating source, and the
reection in the dynamometer walls was used to estimate the reected
temperature (Fig. 3d). A value of 21.5 ±1.5 C was measured in all
experiments where the cutting temperature was lower than 120 C.
3.2.2. Hole quality parameters
As described in subsection 3.1 defects at the entry and exit planes of
the hole (peel-up and push-out delamination, matrix carbonisation, burr
height and hole diameter and roundness) were characterised depending
on the workpiece material. After that, the workpieces were cut by water
jet to reveal the cross section and measure surface integrity parameters
(S
a
, S
v
for CFRP and R
a
, R
t
for Ti6Al4V). The hole diameter and round-
ness were measured using a Renishaw Cyclone Series 2 coordinate
measurement machine. Delamination and matrix carbonisation in CFRP
were observed using a Keyence VXH 6000 digital microscope with a ×
30 magnication. The delamination factor was evaluated following the
method proposed by Davim et al. [32], since it takes into account the
contribution of areal damage as well as thin cracks to the delamination.
A graphical representation of the delamination factor is shown in Fig. 4,
and it is calculated using the following equations, as described in [32]:
Fda =Fd+Adelam(F2
dFd)
AMAX ANOM
(1)
Where,
Fd=DMAX
DNOM
(2)
The burr height in Ti6Al4V single plates and stacks, as well as the
machined surface defects in CFRP and Ti6Al4V (both separately and in
stacks) were characterised using an Alicona Innite Focus SL 3D optical
microscope with a ×10 magnication.
3.3. Multi-objective optimisation
Due to the high volume of input conditions in separate CFRP and
Ti6Al4V plate drilling experiments, multi-objective optimisation (MOO)
using the response surface method (RSM) was carried out to determine
the q
LCO2
and q
MQL
values that give the best overall results for each
material. For the RSM, the regressions dening each of the response
surfaces (y
i
) were obtained, using a full quadratic model.
Desirability functions were calculated to optimise the response re-
gressions. By dening the objectives and constraints of the optimisation,
an individual desirability function (d
i
) can be dened for each of the
responses. These functions are dened as a piecewise function (Eq. (3),
and the intervals that make up the equation vary according to the
objective dened for the response (maximisation, minimisation, target
value). For every d
i
, the lower limit value (L
i
), upper limit value (U
i
) and
lower and upper thresholds of the target value (T
r1,i
and T
r2,i
) were
dened. The equation limitation levels for the different objectives are
shown in Fig. 5.
di=
0yi<Li
yiLi
Tr1,iLi
LiyiTr1,i
1Tr1,iyiTr2,i
Uiyi
UiTr2,i
Tr2,iyiUi
(3)
The output of the desirability function is a quantitative score of the in-
dividual optimisation result. The scores are arranged on values from 0 to
1 (the higher, the better). However, in order to reach MOO, the indi-
vidual desirability functions must be combined into one overall desir-
ability function, D(y). The overall desirability function (Eq. (4) is the
weighted geometric desirability for each of the individual responses. In
the current application, nine individual desirability functions will form
the overall desirability function for each individual CFRP and Ti6Al4V
plates (one per every output parameter listed in Table 5 for each ma-
terial). In order to eliminate non-linearities, the target approximation
relationship of all individual desirability evaluations was set as linear (p
i
=1). On the other hand, the weights (w
i
) used to calculate the overall
desirability function were calculated using the Entropy Weighting
Method (EMW) explained by Kumar et al. [33]. The EMW enables to
determine the objective weights of the responses by calculating the
probability, entropy value and degree of divergence of each individual
response.
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
8
D(y) = (n
i=1((di)pi)wi)1
wi(4)
3.4. Experimental setup
Fig. 6 shows an overview of the setups created to carry out the ex-
periments on a Doosan NX 6500 II CNC machining centre. Two different
setups that allow simultaneous recording of cutting forces were
employed: (Fig. 6a) positioning the camera at the exit plane (used for
individual plates and CFRP/Ti6Al4V stacks, and (Fig. 6b) positioning
the camera at the side plane (used for stacks only). The xture described
in Fig. 6a enabled the clamping of individual CFRP and Ti6Al4V plates
and CFRP/Ti6Al4V stacks on the Kistler 9273 dynamometer. Examples
of CFRP/Ti6Al4V experiments are shown in Fig. 6a, however the same
xture was employed for the experiments with CFRP and Ti6Al4V. This
xture allowed drilling in three different positions in the transversal
direction, while having free movement in the longitudinal direction. The
support plate had a 10 mm hole which provided homogeneous support
to the workpiece, reducing the uncertainties for push-out delamination
in CFRP and exit burr in Ti6Al4V, while allowing to record the tem-
perature at the exit plane.
Positioning the camera at a distance of 180 mm from the exit plane
(Fig. 6a), allowed to measure the temperature of the workpiece as the
tool was exiting in individual CFRP and Ti6Al4V plate drilling experi-
ments (Fig. 6b), as well as monitoring the temperature of the entire
cutting process through an eccentric 1 mm pilot hole when drilling
CFRP/Ti6Al4V stacks (Fig. 6c). A similar procedure was employed by
Ueda et al. [34], showing that is possible to record temperature of the
cutting edge throughout the whole cutting process with very little effect
on cutting forces. Measuring the temperature from the side plane
(Fig. 6d), allowed to observe the temperature evolution on the work-
piece during the cutting process (Fig. 6e). The camera was placed at 180
mm from the side of the workpiece, and the distance between the hole
perimeter and the side wall of the workpiece (e) was 0.5 mm, similar to
other research studies [10,13,14,28].
4. Experimental results and discussion
4.1. Individual CFRP and Ti6Al4V plates
4.1.1. Cutting forces and temperatures
The effect of varying q
LCO2
and q
MQL
on the cutting forces when
drilling CFRP and Ti6Al4V is shown in Fig. 7. As it can be seen an in-
crease in q
LCO2
led to a rise in the thrust force (F
z
) component when
drilling CFRP or Ti6Al4V (Fig. 7a and b). This might be due to a rise in
the ow stress of the material when being machined at colder temper-
atures than in dry drilling [19]. Spraying MQL on the cutting zone
helped to slightly reduce the rise in F
z
, however it is smaller than the
increase caused by LCO
2
.
Applying MQL also helped to reduce the torque (M
z
) for both CFRP
and Ti6Al4V, probably due to a reduction of the friction coefcient
(Fig. 7c and d). However, through the tool LCO
2
cooling brought
different M
z
results depending on the machined material. When drilling
CFRP, an increase in q
LCO2
created a rise in M
z
, while for Ti6Al4V case,
helped to reduce it. Applying LCO
2
cooling usually increases the friction
Fig. 7. Effect of varying q
LCO2
and q
MQL
on cutting forces when drilling individual plates: a) Thrust force in CFRP; b) Thrust force in Ti6Al4V; c) Torque in CFRP; d)
Torque in Ti6Al4V.
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
9
coefcient at the tool-chip interface, thus increasing the torque [35].
However, the improvement in chip evacuation generated by supplying
LCO
2
through the tool at 5.7 MPa could have helped to reduce M
z
when
drilling Ti6Al4V [36]. The effect of torque reduction due to chip evac-
uation was not observed for CFRP drilling since the powder-like chips of
the composite are not problematic to evacuate.
The torque generated by the tool was also affected by the hole
diameter. As Fig. 8 shows, the cooling effect of LCO
2
helped to achieve
holes closer to the nominal value (6 mm). This generated a reduction in
M
z
when drilling Ti6Al4V, since the thermal shrinkage decreases and
bigger holes than in dry drilling are achieved. But on the other hand, an
increase in M
z
is observed when drilling CFRP, as smaller holes are
machined with LCO
2
compared to dry drilling. Similar results were
achieved by Xia et al.[19], who achieved holes with dimensions closer to
the nominal diameter of the tool, and thus an increase in M
z
when
drilling CFRP with sub-zero cooling. Merzouki et al. [20] also reported
that the sub-zero temperature of the coolant helped to reduce the
thermal shrinkage of the hole, and hence the torque when drilling
Ti6Al4V.
The cutting temperature was recorded at the exit plane of the
workpiece. The key frame selected for the analysis was the previous one
to the tool exit. The highest temperatures appeared during the last 50
frames before the tool exit, and the variation between them was smaller
than 20 C, as the frame rate was 1200 Hz. The response surfaces of the
maximum temperatures recorded at the tool-workpiece interface when
drilling CFRP and Ti6Al4V individual plates are shown in Fig. 9. The
temperature elds of the key frame at constant q
LCO2
=0 g/min and
varying MQL, and constant q
MQL
=0 mL/h and varying LCO
2
are also
shown for each material.
The maximum temperatures measured in dry drilling are consistent
with the results of other studies, about 120 C for CFRP [37] and 500 C
for Ti6Al4V [8]. As the cutting temperatures reached when drilling
CFRP (Fig. 9a) were considerably lower than the ones for Ti6Al4V
Fig. 8. Effect of varying q
LCO2
and q
MQL
on hole diameter when drilling indi-
vidual plates: a) CFRP; b) Ti6Al4V.
Fig. 9. Effect of varying different q
LCO2
and q
MQL
on maximum temperature at
the tool-workpiece interface when drilling individual plates: a) CFRP;
b) Ti6Al4V.
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
10
Fig. 10. Effect of varying q
LCO2
and q
MQL
on hole quality when drilling CFRP individual plates: a) Arithmetical mean height; b) Maximum valley depth; c) Push-out
delamination.
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
11
Fig. 11. Effect of varying q
LCO2
and q
MQL
on hole quality when drilling Ti6Al4V individual plates: a) Average surface roughness; b) Maximum peak to valley height;
c) Ten point average of burr height.
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
12
(Fig. 9b), temperatures below zero were reached in CFRP when using
q
LCO2
above 150 g/min. Also, a considerable temperature reduction was
achieved when supplying MQL without LCO
2
cooling. When drilling
Ti6Al4V, the effect of the MQL on heat reduction was less pronounced.
However, with LCO
2
cooling, the cutting temperature was reduced by
half when LCO
2
ow rates of q
LCO2
=150 g/min were used. In Fig. 9b the
temperature elds of LCO
2
cooling show that the tool tip is considerably
colder than the cutting edges, since the coolant came through the tool.
4.1.2. Hole quality in CFRP
The drilling induced push-out delamination and the surface
morphology are shown in Fig. 10. As can be seen, surface roughness and
voids related to bre pull-out (Fig. 10a and b) and delamination
(Fig. 10c) decreased when MQL was applied, and increased when the
LCO
2
ow rate rose. Applying LCO
2
through the tool had a detrimental
effect on the quality of CFRP holes. The increase in thrust force gener-
ated by the LCO
2
, might have generated greater push-out delamination,
as also shown by other studies [4,19]. The rise in bre pull-out created
by LCO
2
might be related to the increase in transverse elastic toughness
of the epoxy matrix at low temperatures [38], in combination with the
brittle fracture that governs chip formation mechanisms in composites
[6]. In Fig. 10, the delaminated areas at the exit plane of the composite
are marked in red in. When looking at the 3D surface morphology, a
greater number of peaks (in yellow and red) and valleys (in dark blue
and purple) indicate a greater number of pulled-out bres.
4.1.3. Hole quality in Ti6Al4V
The response surfaces of R
a
, R
t
and burr height, as well as the surfaces
measured with the Alicona Innite Focus SL, are given in Fig. 11. LCO
2
through the tool cooling produced machined surfaces of poorer quality
than when only MQL was supplied or dry machining, although the LCO
2
coolant helped to improve the hole diameter and reduce the torque. This
might have occurred due to the worse tribological properties of LCO
2
[35]. On the other hand, the temperature reduction achieved by
increasing q
LCO2
contributed to the reduction in burr height (Fig. 11c).
This could be related to the damage mechanism of Ti6Al4V, which is
dominated by elastoplastic deformation [6]. At lower temperatures the
plasticity of the titanium decreases, thus reducing the burr height.
Similar results were observed by Gao et al. [39] using VAD. When q
MQL
increased, an improvement in both surface roughness and burr height
was observed. This could be due to the reduction in torque and cutting
temperature achieved by combining cooling and lubrication in the
cutting zone with LCO
2
+MQL.
4.1.4. Multi objective optimisation
In order to achieve optimal q
LCO2
and q
MQL
values for drilling sepa-
rate CFRP and Ti6Al4V plates, the output parameters listed in Table 5 for
each material (cutting forces, temperatures, different hole quality
Table 7
Individual objectives, constraints, and weights for every output parameter employed in the multi-objective optimisation of the LCO
2
and MQL ow rates for drilling
separate CFRP and Ti6Al4V plates.
CFRP Ti6Al4V
Output parameter Objective Constraints [min max] wi [%] Objective Constraints [min max] wi [%]
Thrust force, F
z
Minimise 91.6 138.7 7.9 Minimise 533.6 670.6 7.8
Torque, M
z
Minimise 0.102 0.154 6.7 Minimise 0.952 1.343 13.2
Temperature, T 25 C 21.3 104.3 16.4 Minimise 187.6 487.9 21.3
Hole diameter, Ø Minimise 6.007 6.08 20.1 6 mm 5.99 6.057 4.8
Roundness,
Arithmetic mean height, S
a
Minimise 2.45 4.84 9
Maximum valley depth, S
v
Minimise 16.9 35.52 7.5
Delamination, F
da
Minimise 2.05 3.01 5.9
Matrix carbonisation
Average surface roughness, R
a
Minimise 0.35 0.682 6
Maximum peak to valley height, R
t
Minimise 5 7.24 6.6
Ten point burr height, S
10z
Minimise 129 248.6 13.8
LCO
2
ow rate, q
LCO2
Minimise 0 200 12.3 Minimise 0 200 12.3
MQL ow rate, q
MQL
Minimise 0 75 14.2 Minimise 0 75 14.2
Fig. 12. Overall desirability surfaces and optimal q
LCO2
and q
MQL
values for
drilling: a) CFRP and b) Ti6Al4V.
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
13
parameters depending on the material, and LCO
2
and MQL consump-
tion), needed to meet the optimisation objectives. As seen in Table 7, the
objective for most output parameters was minimisation, however, as the
damage in CFRP plates increased with sub-zero temperatures, a target
temperature of 25 C was set. This way, when drilling CFRP with opti-
mised cooling/lubrication parameters, the cutting temperature will be
close to the ambient one. The constraints used for the MOO were the
maximum and minimum measured values for each of the outputs. The wi
values calculated by EMW for every output are listed in Table 7.
The overall desirability surfaces for CFRP and Ti6Al4V are shown in
Fig. 12. The highest desirability value corresponds to the combination of
q
LCO2
and q
MQL
that better satises all the objectives and constraints set
for the different output responses in each of the materials. As it can be
observed, the best results for drilling CFRP were obtained when q
LCO2
is
very close to 0 g/min and q
MQL
is about 50 mL/h (Fig. 12a). This
conguration of cooling/lubrication parameters ensures good hole
quality, while preventing the workpiece from freezing. On the other
hand, when drilling Ti6Al4V, as the cutting temperature is higher, the
optimal q
LCO2
value was around 100 g/min, while a MQL ow rate of 50
mL/h was found to be optimal (Fig. 12b).
4.2. CFRP/Ti6Al4V stacks
An experimental plan consisting of the optimized q
LCO2
and q
MQL
values for drilling individual CFRP and Ti6Al4V, and two extended
conditions (dry drilling and pure LCO
2
assisted drilling) was carried out
on CFRP/Ti6Al4V stacks. The cooling/lubrication parameters employed
for each test are detailed in Table 8.
4.2.1. Cutting forces and temperatures
Fig. 13 shows the evolution of the cutting forces and temperature of
the workpiece for different cooling/lubrication techniques during the
entire drilling process of the CFRP/Ti6Al4V stack. The maximum tem-
perature prole was acquired by recording the temperature through the
pilot hole (Fig. 6c). Similar to the individual plate drilling experiments,
an increase in F
z
was observed when applying LCO
2
(Fig. 13c and d), in
comparison to DRY and MQL cases (Fig. 13a and b). Supplying LCO
2
also
increased M
z
in the CFRP phase, while reducing it when the drill was
machining through the Ti6Al4V due to better chip evacuation and
reduced thermal shrinkage [20,36]. The cutting temperature was
considerably lower for the LCO
2
and LCO
2
+MQL cases (Fig. 13c and d),
showing that the cutting temperature when CFRP/Ti6Al4V stack drilling
can be reduced by half when delivering LCO
2
through the tool.
The temperature elds recorded from the side of the stack shown in
Fig. 14, revealed that the titanium chips heated up the CFRP phase while
being evacuated. This temperature value was similar or higher than the
glass transition temperature of the epoxy matrix (180 C) in DRY and
MQL conditions (Fig. 14a and b), which increases the risk of matrix
Fig. 13. Cutting force and temperature evolution when drilling CFRP/Ti6Al4V stacks under different cooling/lubrication conditions: a) DRY; b) MQL; c) LCO
2
; d)
LCO
2
+MQL.
Table 8
Selected cooling/lubrication parameters to test in CFRP/Ti6Al4V drilling
experiments.
Experiment name q
LCO2
[g/min] q
MQL
[mL/h]
DRY 0 0
MQL 0 50
LCO
2
100 0
LCO
2
+MQL 100 50
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
14
carbonisation as demonstrated by [38]. For LCO
2
and LCO
2
+MQL
conditions (Fig. 14c and d), the temperature of the Ti6Al4V chips was
considerably lower. As a result, the temperature of the CFRP and risk of
matrix carbonisation decreased, but without reaching freezing temper-
atures. This could be due to the heat generated as the Ti6Al4V plate is
being machined.
4.2.2. Hole quality in CFRP/Ti6Al4V stacks
The hole diameter and roundness obtained at eight depths, with the
different cooling/lubrication techniques, are shown in Fig. 15. The error
in roundness and diameter is the greatest at the entry of the CFRP phase,
which is due to the delamination of the rst plies of the laminate
creating an irregular surface in the rst microns of depth. The reduction
in cutting temperature due to LCO
2
cooling contributed to decrease the
thermal shrinkage in titanium and improve the hole roundness and ac-
curacy in the stack.
In Fig. 16 the different hole quality parameters analysed, and their
location in the CFRP/Ti6Al4V stack can be seen. Peel-up delamination
was greater when using LCO
2
cooling (Fig. 16a), probably due to an
Fig. 15. Dimensional accuracy of holes drilled in CFRP/Ti6Al4V stacks under
different cooling/lubrication conditions.
Fig. 14. Temperature elds when drilling CFRP/Ti6Al4V stacks under different cooling/lubrication conditions: a) DRY; b) MQL; c) LCO
2
; d) LCO
2
+MQL.
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
15
Fig. 17. Visual comparison of hole quality of CFRP and Ti6Al4V phases when drilling stacks under different cooling/lubrication conditions: a) DRY; b) MQL; c) LCO
2
;
d) LCO
2
+MQL.
Fig. 16. Evaluation of hole quality of CFRP and Ti6Al4V phases when drilling stacks under different cooling/lubrication conditions.
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
16
increase in the thrust force. However, using through the tool LCO
2
,
greatly improved the surface roughness (S
a
) and reduced the voids (S
v
),
which are related to bre pull-out (Fig. 16b and c). This could be due to
an improvement in titanium chip evacuation by the LCO
2
, which
reduced thermal and mechanical damage produced by the chips on the
machined CFRP surface. CFRP matrix carbonisation at the interface
(Fig. 16d) was also greatly reduced when using LCO
2
and LCO
2
+MQL.
This was expected as lower temperatures were recorded at the CFRP
phase for these cooling/lubrication cases (Fig. 14c and d). The hole
quality in the Ti6Al4V phase of the stack followed the same trend as for
individual titanium plates (Fig. 11). Surface roughness was improved
when using MQL lubrication (Fig. 16e and f), and the burr height was
reduced due to the cold temperatures of LCO
2
(Fig. 16g).
In Fig. 17 the visual comparison of the hole quality is shown. The
damage on CFRP phase by titanium chips, as well as the matrix car-
bonisation is clear for DRY and MQL cases (Fig. 17a and b), since the
cutting temperature was higher, and the titanium chip evacuation was
worse. For LCO
2
and LCO
2
+MQL cases (Fig. 17c and d), less scratches
and carbonisation are observed on the CFRP. No push-out delamination
was observed at the exit of the CFRP, due to the support provided by the
Ti6Al4V.
5. Conclusions
The present work analyses the effect of LCO
2
and MQL ow rates,
and denes the optimal values for drilling CFRP, Ti6Al4V and CFRP/
Ti6Al4V stacks, by performing a multi-objective optimisation of cutting
forces, temperature, and various hole quality parameters. The conclu-
sions are as follows:
The results show that LCO
2
cooling is detrimental for drilling CFRP
plates, since the cutting temperature was not critical. Pure MQL
lubrication produced the best hole quality, as it lubricated the cut-
ting zone without freezing it. On the other hand, for Ti6Al4V and
CFRP/Ti6Al4V stack drilling using LCO
2
+MQL at q
LCO2
=100 g/
min and q
MQL
=50 mL/h achieved the best results. When drilling
stacks, the cooling and lubrication by LCO
2
+MQL helped to reduce
the thermal damage in the CFRP phase and the burr height in the
Ti6Al4V.
Supplying LCO
2
+MQL through the tool created a decrease in the
torque due to better chip evacuation and reduced hole shrinkage
when drilling Ti6Al4V individually and in the stack. On the down-
side, LCO
2
cooling increased the ow stress of the workpiece mate-
rial, and thus the thrust force. However, this effect was reduced when
adding MQL to the LCO
2
.
Machining stacks in the direction from CFRP to Ti6Al4V resulted in
suppression of delamination defects in the CFRP phase. LCO
2
+MQL
supplied through the tool signicantly reduced the degradation of
the machined CFRP surface by improving evacuation of the titanium
chips and reducing the cutting temperature. Additionally, the burr
height of the titanium was also reduced due to the decrease in plastic
deformation of the workpiece promoted by LCO
2
cooling.
Future efforts should focus on varying the ow rate of LCO
2
+MQL
during the process to provide the optimum mix of coolant at every
stage of the machining operation. However, limitations in the time
required to stabilize the LCO
2
+MQL ow rate should be considered.
Funding sources
This research was nancially supported by the ARRS national sci-
ence agency within research program 20266 (Advanced manufacturing
technologies for high quality and sustainable production), and the
projects CRYOMACH (INNO-20182049) from Smart Advanced
Manufacturing call, OPTICED (KK-2021/00003) from Elkartek call, and
EIT Manufacturing: Transitioning to a waste-free production interna-
tional cryogenic +MQL machining activity (2021, 2023).
CRediT authorship contribution statement
I. Rodriguez: Conceptualization, Methodology, Validation, Investi-
gation, Writing original draft. P.J. Arrazola: Conceptualization, Su-
pervision, Project administration, Funding acquisition, Writing review
& editing. M. Cuesta: Conceptualization, Supervision, Project admin-
istration, Writing review & editing. F. Puˇ
savec: Conceptualization,
Supervision, Project administration, Funding acquisition, Writing re-
view & editing.
Declaration of Competing Interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Data availability
The data that has been used is condential.
Acknowledgements
The authors would like to express their great appreciation to SECO
Tools for providing the tools necessary to carry out this research work.
References
[1] ACARE, Flightpath 2050 Europes vision for aviation: Report of the high level
group on aviation research,2011. doi: 10.2777/50266.
[2] Xu J, Mkaddem A, El Mansori M. Recent advances in drilling hybrid FRP/Ti
composite: A state-of-the-art review. Compos Struct 2016;135:31638. https://doi.
org/10.1016/j.compstruct.2015.09.028.
[3] Xu J, Li C, Chen M, El Mansori M, Davim JP. On the analysis of temperatures,
surface morphologies and tool wear in drilling CFRP/Ti6Al4V stacks under
different cutting sequence strategies. Compos Struct 2020;234:111708. https://doi.
org/10.1016/j.compstruct.2019.111708.
[4] Geier N, Davim JP, Szalay T. Advanced cutting tools and technologies for drilling
carbon bre reinforced polymer (CFRP) composites: A review. Compos A Appl Sci
Manuf 2019;125:105552. https://doi.org/10.1016/j.compositesa.2019.105552.
[5] Shah P, Khanna N, Singla AK, Bansal A. Tool wear, hole quality, power
consumption and chip morphology analysis for drilling Ti-6Al-4V using LN2 and
LCO2. Tribol Int 2021;163:107190. https://doi.org/10.1016/j.
triboint.2021.107190.
[6] Xu J, Lin T, Li L, Ji M, Davim JP, Geier N, Chen M. Numerical study of interface
damage formation mechanisms in machining CFRP/Ti6Al4V stacks under different
cutting sequence strategies. Compos Struct 2022;285:115236. https://doi.org/
10.1016/j.compstruct.2022.115236.
[7] Pecat O, Brinksmeier E. Low damage drilling of CFRP/titanium compound
materials for fastening. Procedia CIRP 2014;13:17. https://doi.org/10.1016/j.
procir.2014.04.001.
[8] Lazoglu I, et al. Thermal analysis in Ti-6Al-4V drilling. CIRP Ann - Manuf Technol
2017;66(1):1058. https://doi.org/10.1016/j.cirp.2017.04.020.
[9] Cao S, et al. Investigation of CFRP damages induced by the interface high
temperature and mixed tool wear mechanism in drilling of thin-walled CFRP / Ti
stacks. Compos Struct 2022;323:2023.
[10] Xu J, Li C, Chen M, Ren F. A comparison between vibration assisted and
conventional drilling of CFRP/Ti6Al4V stacks. Mater Manuf Process 2019;34(10):
118293. https://doi.org/10.1080/10426914.2019.1615085.
[11] Jiao F, Li Y, Niu Y, Zhang Z, Bie W. A review on the drilling of CFRP/Ti stacks:
Machining characteristics, damage mechanisms and suppression strategies at stack
interface. Compos Struct 2022;305:2023. https://doi.org/10.1016/j.
compstruct.2022.116489.
[12] Shu L, et al. Study on dedicated drill bit design for carbon ber reinforced polymer
drilling with improved cutting mechanism. Compos A Appl Sci Manuf 2020;142:
2021. https://doi.org/10.1016/j.compositesa.2020.106259.
[13] Hussein R, Sadek A, Elbestawi MA, Attia MH. Low-frequency vibration-assisted
drilling of hybrid CFRP/Ti6Al4V stacked material. Int J Adv Manuf Technol 2018;
98(912):280117. https://doi.org/10.1007/s00170-018-2410-2.
[14] Li C, Xu J, Chen M, An Q, El Mansori M, Ren F. Tool wear processes in low
frequency vibration assisted drilling of CFRP/Ti6Al4V stacks with forced air-
cooling. Wear 2019;426:161623. https://doi.org/10.1016/j.wear.2019.01.005.
[15] Sadek A, Meshreki M, Attia MH. Characterization and optimization of orbital
drilling of woven carbon ber reinforced epoxy laminates. CIRP Ann - Manuf
Technol 2012;61(1):1236. https://doi.org/10.1016/j.cirp.2012.03.089.
[16] Turner J, Scaife RJ, El-Dessouky HM. Effect of machining coolant on integrity of
CFRP composites. Adv Manuf Polym Compos Sci 2015;1(1):5460. https://doi.
org/10.1179/2055035914Y.0000000008.
I. Rodriguez et al.
Composite Structures 329 (2024) 117687
17
[17] Pereira O, Rodriguez A, Fern´
andez-Abia AI, Barreiro J, L´
opez de Lacalle LN.
Cryogenic and minimum quantity lubrication for an eco-efciency turning of AISI
304. J Clean Prod 2016;139:4409. https://doi.org/10.1016/j.
jclepro.2016.08.030.
[18] Jamil M, et al. Inuence of CO2-snow and subzero MQL on thermal aspects in the
machining of Ti-6Al-4V. Appl Therm Eng 2020;177:115480. https://doi.org/
10.1016/j.applthermaleng.2020.115480.
[19] Xia T, Kaynak Y, Arvin C, Jawahir IS. Cryogenic cooling-induced process
performance and surface integrity in drilling CFRP composite material. Int J Adv
Manuf Technol 2016;82(14):60516. https://doi.org/10.1007/s00170-015-7284-
y.
[20] Merzouki J, Poulachon G, Rossi F, Ayed Y, Abrivard G. Effect of cryogenic
assistance on hole shrinkage during Ti6Al4V drilling. Int J Adv Manuf Technol
2020;108(910):267586. https://doi.org/10.1007/s00170-020-05381-z.
[21] Grguraˇ
s D, Sterle L, Krajnik P, Puˇ
savec F. A novel cryogenic machining concept
based on a lubricated liquid carbon dioxide. Int J Mach Tool Manu Oct. 2019;145.
https://doi.org/10.1016/j.ijmachtools.2019.103456.
[22] Transitioning to a waste-free production international cryogenic+MQL
machining activity - EIT Manufacturing.https://www.eitmanufacturing.eu/ne
ws-events/activities/transitioning-to-a-waste-free-production-international-cry
ogenicmql-machining-activity/ (accessed Oct. 11, 2023).
[23] Rodriguez I, Arrazola PJ, Cuesta M, Sterle L, Puˇ
savec F. Improving surface integrity
when drilling CFRPs and Ti-6Al-4V using sustainable lubricated liquid carbon
dioxide. Chinese J Aeronaut 2023;36(7):12946. https://doi.org/10.1016/j.
cja.2022.09.004.
[24] Grguraˇ
s D, Sterle L, Malnerˇ
siˇ
c A, Kastelic L, Courbon C, Puˇ
savec F. Media ow
analysis of single-channel pre-mixed liquid CO2 and MQL in sustainable
machining. Stroj Vestnik/Journal Mech Eng 2021;67(12):310. https://doi.org/
10.5545/sv-jme.2020.7076.
[25] Kacalak W, Lipi´
nski D, R´
o˙
za´
nski R, Kr´
olczyk GM. Assessment of the classication
ability of parameters characterizing surface topography formed in manufacturing
and operation processes. Measurement 2020;170:2021. https://doi.org/10.1016/j.
measurement.2020.108715.
[26] Arola D, Williams CL. Estimating the fatigue stress concentration factor of
machined surfaces. Int J Fatigue 2002;24:92330. https://doi.org/10.1016/S0142-
1123(02)00012-9.
[27] Duboust N, Watson M, Marshall M, ODonnel GE, Kerrigan K. Towards intelligent
CFRP composite machining: Surface analysis methods and statistical data analysis
of machined bre laminate surfaces. Proc Inst Mech Eng Part B J Eng Manuf 2021;
235(10):160217. https://doi.org/10.1177/0954405420960920.
[28] Segurajauregui U, Arrazola PJ. Heat-ow determination through inverse
identication in drilling of aluminium workpieces with MQL. Prod Eng Res Dev
2015;9(4):51726. https://doi.org/10.1007/s11740-015-0631-x.
[29] Telops Inc., Telops Infrared Camera Blackbody-Free Calibration Technique.htt
p://info.telops.com/Website-content-requests_05—LP—Infrared-Camera-Blackbod
y-Free-Calibration-Technique.html (accessed Mar. 29, 2022).
[30] Fu R, et al. Drill-exit temperature characteristics in drilling of UD and MD CFRP
composites based on infrared thermography. Int J Mach Tool Manu 2018;135:
2437. https://doi.org/10.1016/j.ijmachtools.2018.08.002.
[31] Sela A, Ortiz-de-zarate G, Soler D, Germain G, Gallegos L, Arrazola PJ.
International Journal of Heat and Mass Transfer Adiabatic self-heating
determination for Ti6Al4V at different temperatures. Int J Heat Mass Transf 2023;
204:123747. https://doi.org/10.1016/j.ijheatmasstransfer.2022.123747.
[32] Davim JP, Rubio JC, Abrao AM. A novel approach based on digital image analysis
to evaluate the delamination factor after drilling composite laminates. Compos Sci
Technol 2007;67:193945. https://doi.org/10.1016/j.compscitech.2006.10.009.
[33] Kumar R, et al. Revealing the benets of entropy weights method for multi-
objective optimization in machining operations: A critical review. J Mater Res
Technol 2021;10(4):147192. https://doi.org/10.1016/j.jmrt.2020.12.114.
[34] Ueda T, Nozaki R, Hosokawa A. Temperature measurement of cutting edge in
drilling -Effect of Oil Mist-. CIRP Ann - Manuf Technol 2007;56(1):936. https://
doi.org/10.1016/j.cirp.2007.05.024.
[35] Puˇ
savec F, Sterle L, Kalin M, Mallipeddi D, Krajnik P. Tribology of solid-lubricated
liquid carbon dioxide assisted machining. CIRP Ann 2020;69:6972. https://doi.
org/10.1016/j.cirp.2020.04.033.
[36] Sterle L, Krajnik P, Puˇ
savec F. The effects of liquid-CO2 cooling, MQL and cutting
parameters on drilling performance. CIRP Ann 2021;70(1):7982. https://doi.org/
10.1016/j.cirp.2021.04.007.
[37] Xu J, Lin T, Davim JP. On the Machining Temperature and Hole Quality of CFRP
Laminates when Using Diamond-Coated Special Drills. J Compos Sci 2022;6(2):
114. https://doi.org/10.3390/jcs6020045.
[38] Kubher S, Gururaja S, Zitoune R. Coupled thermo-mechanical modeling of drilling
of multi-directional polymer matrix composite laminates. Compos. Part A Appl.
Sci. Manuf. 2022;156:106802. https://doi.org/10.1016/j.
compositesa.2022.106802.
[39] Gao G, Xia Z, Yuan Z, Xiang D, Zhao B. Inuence of longitudinal-torsional
ultrasonic- assisted vibration on micro-hole drilling Ti-6Al-4V. Chinese J Aeronaut
2021;34(9):24760. https://doi.org/10.1016/j.cja.2020.06.012.
[40] Sadek A, Attia MH, Meshreki M, Shi B. Characterization and optimization of
vibration-assisted drilling of bre reinforced epoxy laminates. CIRP Ann - Manuf
Technol 2013;62(1):914. https://doi.org/10.1016/j.cirp.2013.03.097.
[41] Sun Z, Geng D, Meng F, Zhou L, Jiang X. High performance drilling of T800 CFRP
composites by combining ultrasonic vibration and optimized drill structure.
Ultrasonics 2023;134:107097. https://doi.org/10.1016/j.ultras.2023.107097.
[42] Shao Z, et al. The interface temperature and its inuence on surface integrity in
ultrasonic-assisted drilling of CFRP/Ti stacks. Compos Struct 2021;266:113803.
https://doi.org/10.1016/j.compstruct.2021.113803.
[43] Bleicher F, Wiesinger G, Kumpf C, Finkeldei D, Baumann C, Lechner C. Vibration
assisted drilling of CFRP/metal stacks at low frequencies and high amplitudes. Prod
Eng Res Devel 2018;12(2):28996. https://doi.org/10.1007/s11740-018-0818-z.
[44] Pan Z, Wang L, Fang Q, Sun Z, Qu W. Study on tool deection compensation
method based on cutting force observer for orbital drilling of CFRP/Ti stacks.
J Manuf Process 2022;75:45060. https://doi.org/10.1016/j.jmapro.2021.12.058.
[45] Kong L, Gao D, Lu Y, Zhang P. Experimental investigation on machining
performance during orbital drilling of CFRP. Int J Adv Manuf Technol 2022;121
(34):161121. https://doi.org/10.1007/s00170-022-09437-0.
[46] Senthilkumar M, Prabukarthi A, Krishnaraj V. Machining of CFRP/Ti6Al4V stacks
under minimal quantity lubricating condition. J Mech Sci Technol 2018;32(8):
378796. https://doi.org/10.1007/s12206-018-0731-6.
[47] Xu J, Ji M, Chen M, El Mansori M. Experimental investigation on drilling
machinability and hole quality of CFRP/Ti6Al4V stacks under different cooling
conditions. Int J Adv Manuf Technol 2020;109(56):152739. https://doi.org/
10.1007/s00170-020-05742-8.
[48] Fern´
andez-P´
erez J, Cantero JL, Díaz-´
Alvarez J, Migu´
elez MH. Hybrid composite-
metal stack drilling with different minimum quantity lubrication levels. Materials
2019;12(3):448. https://doi.org/10.3390/ma12030448.
[49] Iqbal A, et al. Between-the-holes cryogenic cooling of the tool in hole-making of Ti-
6Al-4V and CFRP. Materials (Basel) 2021;14(4):119. https://doi.org/10.3390/
ma14040795.
[50] Rodríguez A, Calleja A, de Lacalle LL, Pereira O, Rubio-Mateos A, Rodríguez G.
Drilling of CFRP-Ti6Al4V stacks using CO2-cryogenic cooling. J. Manuf. Process.
2021;64:5866. https://doi.org/10.1016/j.jmapro.2021.01.018.
I. Rodriguez et al.
... Rimpault et al. [13] show that the axial force and the torque have both a significant effect on titanium burr height. Only few authors studied the influence of temperature on burr geometry [14], but not on burr formation. This paper presents the experimental analysis of exit burr geometry during the drilling of the titanium alloy Ti6Al4V. ...
Conference Paper
To improve the assembly process of aeronautical structures by the mean of One-Way Assembly strategies (no deburring of metallic parts before the installation of final fasteners), it is mandatory to monitor the burr size of drilled holes. Indeed, burrs can have a significant effect on the fatigue life of structures. It was shown that cracks are initiated from exit burrs. This paper presents an experimental analysis of the burr geometry through cutting forces and thermal imaging measurements. The effect of the tool wear on burr geometry is also analyzed to identify phenomena occurring during burr formation.
... The results showed that the tool wear was significantly reduced under nanofluid MQL. Rodriguez et al. [14] used MQL mixed with liquid carbon dioxide to drill CFRP/Ti6Al4Vlaminated materials. The results showed that the height of the burr on the cutting surface of titanium alloy was significantly reduced. ...
Article
Full-text available
A study of a hole-making experiment for AFRP was carried out using two cutting methods, drilling, and milling, respectively, to analyze the surface morphology, cutting force, hole wall morphology, and tool wear, as well as to conduct the test under MQL conditions, and to compare the situation with that in dry cutting. It was found that the drilling of holes had more burrs on the inlet surface, surface roughness and poor roundness, and fewer burrs on the outlet surface and better quality, and the opposite was true for milling, with fewer burrs on the inlet surface, good surface quality and roundness, and more burrs on the outlet surface and poor roundness. After the use of MQL, the cutting surface edge bulge expansion, free resin block, and resin smearing phenomenon significantly reduced or even disappeared, but the surface burr inhibition effect is limited. At the same time, the observation of the hole wall found that the MQL conditions of the shear effect are apparent, the fiber fracture surface is neat, and we can see the orthogonal weaving intersection of the upper and lower layers of fibers. By establishing the response surface model of cutting force, the drilling cutting force achieves the minimum value in the interval, and the milling cutting force achieves the maximum value in the interval. The study shows that the cutting force of AFRP hole-making shows a laminar solid nature, and its laminated structure has a strong reflection, especially the axial force signal; the cutting surface morphology has a great relationship with the cutting mode and the system stiffness/strength; the increase of the feed speed makes the single cutting fiber volume increase, and the extrusion of the cutting surface is enhanced, which inhibits the surface burr to a certain degree. After the weighting analysis, it was found that the interaction between cutting speed and feed rate was reduced, and the individual effect of each influencing factor was enhanced, which was conducive to exerting the cutting efficiency of the tool itself. In addition, the study also found that the contact point between the tool and workpiece is also one of the influencing factors of the cutting surface morphology and makes the surface burr problem more complicated.
Article
In order to explore the milling mechanism of aramid fiber reinforced polymer under minimal quantity lubrication and improve the milling quality of aramid fiber reinforced polymer, the milling experiments of aramid fiber reinforced polymer were carried out under dry cutting and minimal quantity lubrication, and the cutting force, fiber section, and surface quality were compared and analyzed. The experiment results show that in the dry cutting state, the surface burr is more, the surface quality is poor, the fiber section is rough, and the resin matrix has a melting trend, accompanied by the phenomenon of chip wrapping around the tool, tooltip collapse, and coating peeling. In minimal quantity lubrication, the surface burr is reduced, the surface quality is good, the fiber section is neat, and the resin interface is clean and tidy. The resin matrix has no melting trend, the tooltip is only slightly worn, and the phenomenon of chip entanglement is effectively avoided. Further research shows that the minimal quantity lubrication can reduce the temperature and friction, enhance the stiffness of the fiber, and change the fracture model of the fiber to a certain extent, that is, from bending to tensile fracture (or accompanied by tensile phenomenon) to shear fracture, which also leads to the overall high cutting force. The total milling force synthesized by the average value increases from 2.45 N to 20.93 N, with a maximum increase of 88.29%. However, due to the cooling and friction reduction of the lubricating fluid and the chip removal and diversion, the overall fluctuation of the cutting force under the minimal quantity lubrication condition is reduced. The overall cutting force is more stable than that under dry cutting conditions. The results show that the milling quality is improved under minimal quantity lubrication, the cutting force is more significant but more stable, and the milling process is more stable. This study has important reference significance for milling composites (especially resin matrix composites) under the condition of minimal quantity lubrication.
Article
Full-text available
Minimum quantity lubrication (MQL) is an effective method for addressing technical challenges associated with high cutting loads during machining. Although much research has explored the ability of MQL to reduce cutting forces, intuitive analyses of its developmental trends, leading researchers, and research hotspots remain limited. To address this, a bibliometric analysis was conducted on studies related to MQL and its role in cutting force reduction. Data from the Web of Science Core Collection database, covering the period from January 1, 2008, to December 31, 2023, were analyzed. Bibliometric tools such as VOSviewer, Scimago Graphica, and Bibliometrix were used to visually assess publication trends, source journals, countries, institutions, authors, references, and keywords. The results show that India leads in publication output with 231 papers, while Li CH from China is the most cited author, with 3200 citations. A detailed clustering analysis of keywords and their representative research areas was performed, summarizing the current findings on the effects of processing parameters on cutting forces. The study also analyzed the optimization of machining parameters to minimize cutting forces, identifying relatively optimal conditions. Finally, challenges and potential solutions were discussed, providing theoretical and technical guidance for the industrial application of MQL.
Article
Full-text available
Drilling of high-strength T800 carbon fiber reinforced plastic (CFRP) are widely employed in current aviation industry. Drilling-induced damages frequently occur and affect not only the load carrying capacity of components but also the reliability. As one of effective methods to reduce the drilling-induced damages, advanced tool structures have been widely used. Nevertheless, it is still difficult to realize high machining accuracy and efficiency by this method. This paper compared three different drill bits to evaluate the drilling performance of T800 CFRP composites and the results showed that the dagger drill was a good choice to drill T800 CFRP considering the lowest thrust force and damages. On this basis, ultrasonic vibration was successfully imposed on dagger drill to further improve the drilling performance. The experimental results showed that ultrasonic vibration reduced the thrust force and surface roughness with a maximum decrease of 14.1 % and 62.2 % respectively. Moreover, the maximum hole diameter errors were decreased from 30 μm in CD to 6 μm in UAD. Besides, the mechanisms of force reduction and hole quality improvement by ultrasonic vibration were also revealed. The results suggest that the combination of ultrasonic vibration and dagger drill is a promising strategy for high performance drilling CFRP.
Article
Full-text available
Nowadays, numerical models are one of the most widely used techniques to predict material performance subjected to different manufacturing processes. However, to obtain accurate predictions, these models require reliable input data from thermomechanical tests. Nevertheless, during the test performance the material is self-heated due to a phenomenon known as adiabatic self-heating. Despite the proven relevance of a proper characterization, adiabatic self-heating is not properly taken into account during thermome-chanical tests. In addition, in the literature, two different definitions were found under the umbrella of adiabatic-self heating. On the one hand, it could be defined as the ratio between the heat spent to heat the sample to the plastic work, value commonly taken as 0.9. On the other hand, many authors define the adiabatic heating as the ratio between the heat experimentally measured to the total plastic work. This second approach, although seems easier, is neglecting heat losses. These two different approaches could lead to misunderstandings once this parameter is implemented in the models. This paper aims to clarify this issue. Moreover, the techniques found in literature aiming to measure this parameter are usually based on 2D approaches at low temperatures. In this paper, a 3D methodology to measure adiabatic self-heating is presented which considers all possible heat losses (conduction, convection, radiation and mass flux) through infrared measurements and Digital Image Correlation (DIC) technique. The adiabatic self-heating was measured for a widely used alloy (Ti6Al4V) obtaining promising results.
Article
Full-text available
In the quest for decreasing fuel consumption and resulting gas emissions in the aeronautic sector, lightweight materials such as Carbon Fiber Reinforced Polymers (CFRPs) and Ti-6Al-4V alloys are being used. These materials, with excellent weight-to-strength ratios, are widely used for structural applications in aircraft manufacturing. To date, several studies have been published showing that the use of metalworking fluids (MWFs), special tool geometries, or advanced machining techniques is required to ensure a surface quality that meets aerospace component standards. Conventional MWFs pose a number of environmental and worker health hazards and also degrade the mechanical properties of CFRPs due to water absorption in the composite. Therefore, a transition to more environmentally friendly cooling/lubrication techniques that prevent moisture problems in the composite is needed. This research shows that environmentally friendly cooling/lubrication techniques based on liquid carbon dioxide (LCO2) are suitable for drilling CFRP and Ti-6Al-4V aeronautical materials because they improve numerous surface integrity parameters compared to conventional dry machining. The results show that the best combination of tool geometry and cooling conditions for machining both materials is drilling with Brad point drills and lubricated LCO2. Drilling under these conditions resulted in a 90% improvement in fiber pull-out volume compared to dry machined CFRP holes. In addition, a 33% reduction in burr height and a 15% improvement in surface roughness were observed compared to dry drilling of titanium. Overall, lubricated LCO2 is a viable option to improve the quality of drilled CFRP and titanium aerospace components while maintaining clean work areas.
Article
Full-text available
As the most promising carbon fiber reinforced plastic (CFRP) hole making method, orbital drilling is widely concerned. This paper aims to understand the influence of the cutting parameters, tool diameters and ratio between milling and drilling (Rm&dRm&d{R}_{m\&d}) on machinability in CFRP orbital drilling. The effects of cutting parameters on thrust force and cutting temperature were studied by orthogonal experiments, and experiments were performed to investigate the variations of tool diameters, ratio between drilling and milling on thrust force, cutting temperature, tool wear and machining quality. The experimental results show that the spindle speed and axial feed rate have apparent effects on thrust force, and the tangential feed rate appreciably impacts on the cutting temperature. The selection of tool diameter and the Rm&dRm&d{R}_{m\&d} has specific influence on tool wear, machining quality and cutting temperature. The result is helpful for selecting cutting parameters and tool diameters for high quality hole machining in CFRP orbital drilling.
Article
Full-text available
Carbon fiber reinforced polymers (CFRPs) are attractive engineering materials in the modern aerospace industry, but possess extremely poor machinability because of their inherent anisotropy and heterogeneity. Although substantial research work has been conducted to understand the drilling behavior of CFRPs, some critical aspects related to the machining temperature development and its correlations with the process parameters still need to be addressed. The present paper aims to characterize the temperature variation and evolution during the CFRP drilling using diamond-coated candlestick and step tools. Progression of the composite drilling temperatures was recorded using an infrared thermography camera, and the hole quality was assessed in terms of surface morphologies and hole diameters. The results indicate that the maximum drilling temperature tends to be reached when the drill edges are fully engaged into the composite workpiece. Then it drops sharply as the tool tends to exit the last fiber plies. Lower cutting speeds and lower feed rates are found to favor the reduction of the maximum composite drilling temperature, thus reducing the risk of the matrix glass transition. The candlestick drill promotes lower magnitudes of drilling temperatures, while the step drill yields better surface morphologies and more consistent hole diameters due to the reaming effects of its secondary step edges.
Article
Carbon fiber reinforced polymer (CFRP)/titanium alloy (Ti) stacks have promising application in aerospace field due to their excellent mechanical properties. During integrated drilling of CFRP/Ti stacks, the machining process at interface between CFRP and titanium alloy is very complicated. The interaction among CFRP, titanium alloy and cutting tool leads to poor processing quality, serious tool wear and other problems. In this paper, in order to deeply analyze the influence mechanism of interface area on the machining process, the research achievements and progress of drilling CFRP/Ti stacks are reviewed. Special attention is paid to the machining characteristics and CFRP damage mechanism at stack interface during drilling and helical milling of CFRP/Ti stacks. The cutting force characteristics, temperature characteristics and tool wear characteristics of interface machining are analyzed, and the damage mechanism and evaluation method are summarized. In order to solve the existing problems in drilling CFRP/Ti stacks, the suppression strategies of interface damage in recent years are discussed from four aspects: machining parameters, machining environment, tool structure and advanced machining technology. The research trend of laminated interface machining is prospected, which provides an important reference value for the realization of precision, high efficiency and low damage integrated drilling of CFRP/Ti stacks.
Article
The influence of in-situ cutting temperature on machining induced damage and machining forces during drilling of multi-directional carbon fiber reinforced polymer (MD-CFRP) composites has been examined using a coupled thermo-mechanical finite element (FE) framework. The MD-CFRP laminate has been modeled ply-by-ply as an equivalent homogenous anisotropic material (EHM) using temperature dependent laminate elastic and fracture properties. Stress-based criteria has been adopted for element deletion simulating the drilling process. The intralaminar damage model for simulating various composite machining induced damage modes was implemented in the FE framework via a user material subroutine. A new composite damage criterion is proposed that accounts for the out-of-plane drilling damage behavior along tool feed direction. Additionally, a fracture mechanics approach has been used to simulate interlaminar delamination onset using surface based cohesive elements at the drill exit plies. The current numerical model predictions show a good agreement with drilling experiments for thrust force, delamination damage, and in-situ cutting temperature.
Article
Orbital drilling is an alternative hole-making machining process that combines drilling and milling, usually performed with specific devices and industrial robots to make holes in aircraft composite structures. In order to eliminate the adverse effects on the diameter precision of the tool deflection in the orbital drilling process of CFRP/Ti stacks, a novel method that fuses cutting force model prediction and feed-axis disturbance observation is proposed to estimate the radial cutting force. The tool deflection caused by the radial cutting force is calculated based on the equivalent diameter cantilever beam model. Then the tool path is corrected to compensate for the tool deflection and prevent borehole diameter deviation by adjusting the tool eccentricity in real-time. The orbital drilling experiments demonstrate that the geometrical error of the hole was reduced by 50%, and the manufacturing precision requirements of IT9 level can be satisfied after one single shot even the cutting force was increased to twice the initial value caused by tool wear, the proposed method effectively improved the hole-making precision of CFRP/Ti stacks.
Article
Composite/titanium stacks are extremely difficult to machine due to the generation of severe interface damage, being the critical defect to suppress. The present paper aims to use the finite element method (FEM) to investigate the interface damage formation mechanisms following the machining of CFRP/Ti6Al4V stacks with a particular focus on the distribution of stresses and temperatures. Its key objective lies in revealing the effects of different cutting sequence strategies on the interface damage formation to guide the design of the stack machining processes as well as the selection of cutting sequences. A micro-mechanical orthogonal cutting model and a 3D drilling model based on the macro-mechanical model of the CFRP/Ti6Al4V stacks were developed to explore the fundamental cutting edge/material interactions and the damage formation mechanisms under both the CFRP → Ti and Ti → CFRP strategies. The investigations confirm the dominant impact of the cutting sequence strategy on the stress and temperature distribution during the machining of CFRP/Ti stacks, which is mainly responsible for the diverse interface damage. Besides, the interface damage of a material layer becomes more serious when the material is machined as a second phase due to the adhesion and abrasion of previous material chips. The coupling effects of the matrix damage and the interface bending are the influential factors leading to severe interface damage under the Ti → CFRP strategy, thereby unfavorable for the interface quality.