Experimental investigation of ultrasonic vibration-assisted cryogenic minimum quantity lubrication for milling of Ti-6Al-4V and grinding of Zerodur

Abstract

Increasing demands on component properties are leading to the development of high-performance materials for which conventional production methods are reaching their limits from an economic and ecological point of view. In recent years, two technologies have been developed that show great potential compared to conventional machining processes, particularly in machining high-performance materials such as the titanium alloy Ti-6Al-4V. Ultrasonic-assisted machining leads to reduced cutting forces and increased tool life. Cryogenic minimum quantity lubrication prevents the occurrence of high machining temperatures and allows higher material removal rates without a negative impact on tool life. This paper shows the influence of ultrasonic-assisted milling and grinding processes in combination with cryogenic minimum quantity lubrication on the machinability of the high-strength materials Ti-6Al-4V and Zerodur. The investigation addressed cutting forces, tool wear, and surface roughness. The superposition of the technologies resulted in longer tool life and lower tool wear for both milling and grinding. However, the surface roughness was consistently higher due to the ultrasonic superposition. Nevertheless, machining with ultrasonic vibration-assisted cryogenic minimum quantity lubrication has great potential for difficult-to-machine materials, especially due to the reduction in tool wear.

Full text

Vol.:(0123456789)
1 3
Production Engineering
https://doi.org/10.1007/s11740-023-01214-6
PROCESSES
Experimental investigation ofultrasonic vibration‑assisted cryogenic
minimum quantity lubrication formilling ofTi‑6Al‑4V andgrinding
ofZerodur
JacquelineBlasl1 · KlausLichtinger2· FabianVieltorf3 · MichaelF.Zaeh3 · NicoHanenkamp1
Received: 9 March 2023 / Accepted: 20 June 2023
© The Author(s) 2023
Abstract
Increasing demands on component properties are leading to the development of high-performance materials for which con-
ventional production methods are reaching their limits from an economic and ecological point of view. In recent years, two
technologies have been developed that show great potential compared to conventional machining processes, particularly in
machining high-performance materials such as the titanium alloy Ti-6Al-4V. Ultrasonic-assisted machining leads to reduced
cutting forces and increased tool life. Cryogenic minimum quantity lubrication prevents the occurrence of high machining
temperatures and allows higher material removal rates without a negative impact on tool life. This paper shows the influence
of ultrasonic-assisted milling and grinding processes in combination with cryogenic minimum quantity lubrication on the
machinability of the high-strength materials Ti-6Al-4V and Zerodur. The investigation addressed cutting forces, tool wear,
and surface roughness. The superposition of the technologies resulted in longer tool life and lower tool wear for both mill-
ing and grinding. However, the surface roughness was consistently higher due to the ultrasonic superposition. Nevertheless,
machining with ultrasonic vibration-assisted cryogenic minimum quantity lubrication has great potential for difficult-to-
machine materials, especially due to the reduction in tool wear.
Keywords Vibration-assisted machining· Ultrasonic vibration· Cryogenic cooling· Minimum quantity lubrication·
Ti-6Al-4V· Zerodur
1 Introduction
Increasing demands on components in terms of strength
and quality are leading to a growth in use of high-strength
construction materials. These include the glass–ceramic
Zerodur and the titanium alloy Ti-6Al-4V, which are used
in the aerospace industry. During manufacturing, machining
is most commonly applied for the removal of excess mate-
rial. Consequently, the efficiency of machining technologies
plays a significant role in the economical processing of
high-performance materials. Conventional machining of
these materials results in high cutting forces, increased tool
wear, and a low material removal rate [1]. These disadvan-
tages can be counteracted by ultrasonic vibration-assisted
machining (UVAM) [2–4] and cryogenic minimum quantity
lubrication (CMQL) [5–8]. In vibration-assisted machin-
ing, a high-frequency oscillation – often in the ultrasonic
(US) range – with an amplitude of a few micrometers is
superimposed on the machining process. The advantages of
vibration-assisted machining are reduced cutting forces, an
increased tool life, and an improved component quality [1,
9–12]. CMQL has become the focus of many investigations
in recent years, which have demonstrated that adding oil to
carbon dioxide (CO2) cooling has an additional lubricating
effect. The consequences are a longer tool life and a higher
material removal rate [5–8].
This paper presents a superposition of both UVAM and
CMQL (UCMQL) in milling and grinding, based on recent
publications investigating UVAM and CMQL separately.
* Jacqueline Blasl
jacqueline.blasl@fau.de
1 Institute ofResource andEnergy Efficient Production
Systems, Friedrich-Alexander-Universität Erlangen-
Nürnberg (FAU), Fürth, Germany
2 Institute forPrecision Machining andHigh Frequency
Technology, Deggendorf Institute ofTechnology (DIT),
Deggendorf, Germany
3 Institute forMachine Tools andIndustrial Management,
Technical University ofMunich (TUM), Munich, Germany

Production Engineering
1 3
The research hypothesis was that an even greater increase
in tool life and material removal rate can be expected due to
the assumption that the lubricant can better reach the cutting
edge through the periodically interrupted cutting process.
The findings were compared to milling and grinding opera-
tions with a conventional flood cooling with emulsion (FC).
2 State oftheart
Both UVAM and CMQL have been the focus of investiga-
tions with respect to their application for different kinds of
machining processes. This chapter gives a brief overview of
the relevant literature concentrating on milling and grinding.
Vibration superposition has already been applied to
machining processes like turning, milling, drilling, and
grinding. It can be generated by vibrating the tool or the
workpiece. Most investigations thus far have focused on
turning because it is easier to apply a vibration superposition
on a stationary tool than on a rotating one (e.g., milling).
Rinck etal. [1], however, studied the influence of longitudi-
nal vibration-assisted milling on machining of Ti-6Al-4V.
Peripheral milling and slot milling were investigated. For
peripheral milling, the cutting forces decreased by 44.3%.
The surface roughness of the side wall improved, and the
bottom surface roughness in the slot-milled experiments
decreased with a vibration amplitude of 4µm but increased
with a higher amplitude. The surface roughness of the side
wall in the peripheral milling experiments also improved.
The tool wear of the minor cutting edges was increased,
whereas the tool wear of the primary cutting edges was
reduced. Ni etal. [10] used UVAM for milling of Ti-6Al-
4V. They concluded that the cutting force components could
be reduced by up to 46% and surface roughness by up to 48%
compared to conventional milling. Razafar etal. [13] inves-
tigated UVAM for AISI1020 steel by vibrating the work-
piece. A reduction in cutting forces and surface roughness
in comparison with conventional machining was achieved in
these investigations as well.
Reif etal. [14] investigated the influence of US-assisted
grinding of the glass–ceramic Zerodur. They showed that for
peripheral grinding the machining force and the workpiece
roughness can be reduced by a superposition of a 10.8µm
US amplitude. Sun etal. [15] studied the effects of the ori-
entation of the US superposition during grinding of the glass
material Ultra Low Expansion Glass. When the direction
of the vibration was in the feed direction, the machining
forces applied were 47% lower and the surface roughness
was 23% lower compared to conventional grinding. With a
vibration direction perpendicular to the machined surface,
the machining forces were reduced by 69%, but in this case
the surface roughness increased by about 23%.
Various publications have dealt with CMQL machining of
difficult-to-machine-materials. CMQL combines cryogenic
cooling of the contact zone, usually by liquid CO2 or liquid
nitrogen, with a small amount of oil added to ensure lubrica-
tion. Gross etal. [5], for example, investigated the perfor-
mance of different minimum quantity lubrication (MQL)
oils for CMQL milling of Ti-6Al-4V and compared them
to flood and cryogenic cooling. The tool wear as well as
the mechanical tool load (i.e., the bending moment) were
lower for CMQL milling regardless of the oil choice. Bagh-
erzadeh etal. [6] compared different lubricating/cooling
methods during slot milling of Ti-6Al-4V. They concluded
that CMQL enhances the machinability of Ti-6Al-4V and
lowers flank wear, cutting forces, and surface roughness
compared to separate MQL or cryogenic machining, espe-
cially for higher cutting speeds. Wu etal. [7] investigated
the application of CMQL in milling thin-walled Ti-6Al-4V
components. They found that for CMQL machining cutting
forces, surface roughness, and tool wear decreased compared
to MQL and flood cooling, particularly at higher spindle
speeds.
For grinding, investigations into CMQL are rather scarce
and have focused on machining of metals rather than ceram-
ics. Sanchez etal. [8] achieved a lower surface roughness
and a reduction in tool wear when grinding tool steel with
CO2-based CMQL. Arafat etal. [16] demonstrated higher
surface roughness on the workpiece compared to conven-
tional machining, but lower forces during the machining pro-
cess when grinding roller bearing steel with a combination
of supercritical CO2 and MQL.
The literature contains few initial studies for milling and
no initial studies for grinding in which UVAM was com-
bined with MQL or cryogenic cooling. Namlu etal. [17]
studied a combination of UVAM and MQL for machining
of the aluminum alloy EN AW-6061-T6. They found that
the simultaneous application of both technologies led to the
lowest surface roughness and tool marks on the workpiece
in their test series. Ni etal. [18] investigated the tool wear
mechanism and the corresponding machined surface charac-
terization in the milling of Ti-6Al-4V. They studied UVAM
in combination with MQL to improve the lubrication and
cooling performance. Compared to conventional machining
and UVAM, the combination of both technologies resulted
in the lowest surface roughness. Madakar etal. [19] used a
US-assisted MQL for grinding Ti-6Al-4V. The addition of
the US assistance led to lower grinding forces but higher sur-
face roughness compared to a conventional MQL. No studies
are known to the authors in which UVAM and CMQL were
combined for milling or grinding processes.

Production Engineering
1 3
3 Objective, materials andmethods
The objective of the investigations leading to this paper
was an assessment of the combination of UVAM and
CMQL for machining processes. The superposition of both
techniques has the potential to further increase tool life
and decrease machining forces. It had to be investigated
whether a cryogenic cooling fluid has better contact with
the cutting zone in such a process due to the contact inter-
ruption between the workpiece and the cutting edge. This
was examined in a milling and a grinding process. The
measurement values during machining were the cutting
forces, the tool wear, and the surface roughness of the
workpiece. These were used to evaluate the process effi-
ciency. For comparison, a separate application of UVAM
and CMQL and a conventional process were studied.
In the following section, the experimental setup is
described. First, the equipment used for both the milling
and grinding experiments is outlined. Afterwards, param-
eters and methods specific to the milling or the grinding
process are described separately.
3.1 General experimental setup
The experiments were performed on a five-axis machin-
ing center, a DMG MORI ULTRASONIC 40 eVo linear
(Bielefeld, Germany), which was able to perform UVAM
in the longitudinal direction along the z-axis of the tool.
For the CMQL experiments, the mixing unit described in
Gross [20] was employed. The unit allowed a separate sup-
ply of liquid CO2 and a minimum quantity lubrication oil
into the machine and the transportation of the correspond-
ing mixture to the cutting zone after mixing.
The tool wear analysis was carried out on a Keyence
VHX-6000 (Osaka, Japan) digital microscope. The rough-
ness values Ra, Rt, and Rz were used to evaluate the work-
piece surface. They were measured in feed direction using a
Mahr MarSurf LD 260 Y machine (Goettingen, Germany)
for the milling experiments and using an optical Leica DCM
3D surface roughness system (Wetzlar, Germany) for the
grinding tests. A quartz crystal 3-component dynamometer
Kistler 9257 B (Winterthur, Switzerland) was applied to
continuously record the cutting force components in three
spatial directions during the machining process with a sam-
pling rate of 1600Hz for the milling tests and 1000Hz for
the grinding experiments. The force components were ana-
lyzed with the software DynoWare from Kistler (Winterthur,
Switzerland). The cutting force was calculated as a vector
addition of the three cutting force components.
During a previous investigation (see [21, 22]) using
CMQL without the addition of US-vibrations, a suitable
oil was chosen out of several experimental blends by
Jokisch (Berlin, Germany), which differed both in base
oil (mineral- or alcohol-based) and additives (with or with-
out). Operating parameters of the aforementioned mixing
unit, nozzle positioning, and nozzle orientation were also
selected according to previous testing results and kept con-
stant for all experiments.
The focus of this article are the experiments which uti-
lized the chosen oil in a US-assisted CMQL while keeping
the setup used in the previous experiments. The experimen-
tal setup for the milling process is displayed in Fig.1. It is
identical for the grinding process. Each milling and grinding
experiment was conducted three times with a new milling
tool or a dressed grinding tool.
3.2 Milling
For milling, the titanium alloy Ti-6Al-4V (annealed) was
employed as workpiece material. Experiments were con-
ducted using two different tools for which suitable machin-
ing parameters were chosen in previous tests. Both the tool
and the machining parameter specifications are listed in
Table1.
The nozzle was positioned in front of the tool in the
feed direction and kept at a constant angle (see Fig.2). The
experiments were conducted by using a milling process in
CMQL nozzle
workpiece
ultrasonic
actuator
Kistler
dynamometer
tool
10 mm
Fig. 1 Experimental setup

Production Engineering
1 3
climb milling. Entry into the workpiece was performed by
rolling into the cut, which leads to a reduction in vibration
and tool wear when milling difficult-to-machine materi-
als [23]. Afterwards, the workpiece material was removed
under constant contact conditions. A schematic of the mill-
ing process is shown in Fig.3.
Tool wear was analyzed in terms of the width of the
wear marks on the main cutting edge. Tool wear as well
as workpiece surface roughness analyses were conducted
after reaching predefined machining paths (see Table2).
The arithmetic mean and standard deviation were calcu-
lated out of the analysis of the three tools.
The cutting force was measured continuously throughout
the machining process. The data then was filtered with a low
pass Butterworth filter with an edge frequency of 50Hz and
a filter order of 2. For displaying the progression, the mean
cutting force was calculated out of the data from one line of
milling at the measurement points used for tool wear and
surface roughness (see Table2).
Regardless of the tool type, the maximum flank wear land
width VBmax was selected as the tool life criterion. If VBmax
exceeded a value of 200µm on at least two cutting edges of
the tool, a test series was completed. The test series was also
terminated if a cutting edge fractured or a milling distance
of 500m was achieved without significant wear phenomena.
Table 1 Tool and machining parameters (milling)
Tool and machining parameters Unit Value
Tool 1 Tool 2
Product development state Serial Prototype
Tool diameter dmm 10 8
Number of teeth z– 4 5
Material and coating – Solid carbide, TiAlN coating Solid carbide, TiAl-based coating
Feed per tooth fzmm 0.05 0.04
Cutting depth apmm 5
Cutting width aemm 0.5
Cutting speed vcm/min 240 125
Feed rate vfmm/min 1528 995
CO2 mass flow mCO2 kg/h 3.0–3.5
Oil volume flow Voil ml/min 0.8
CMQL oil – Mineral-based oil without additives Alcohol-based oil with additives
Reference tests – Flood cooling with emulsion (Oemeta NOVAMET 760 at 3.3%)
US amplitude A
(resonance frequency: approx. 21kHz)
µm 3 2–3
X
Z
Y
35°
25°
3
5
5
°
tool
workpiece
CMQL nozzle
Fig. 2 CMQL nozzle positioning (milling)
Y
Z
v
f
Y
tool workpiece
CMQL nozzletool path
n
X
Y
direction
of UVAM
Z
surfacefor
roughness
measurement
Fig. 3 Milling process schematic

Production Engineering
1 3
3.3 Grinding
The tests were performed as slot grinding with a diamond
mounted point as grinding tool. Regarding the workpiece
material, the glass–ceramic material Zerodur was used. The
cutting width corresponded to the outer diameter of the
grinding tool of 9.1mm. The cutting depth was 0.2mm.
The surface roughness of the workpiece was measured at
the bottom surface of the ground slot. A schematic of the
grinding program can be seen in Fig.4.
Tool wear was determined using a Keyence VHX-6000
digital microscope. For this purpose, the contour of the
lateral surface was recorded at three positions distributed
equidistantly around the circumference before each test and
after 528 slots with a length of 0.12m each, which equals
63.36m machining path. The difference between those con-
tours equals the wear volume at that measuring position (see
Fig.5). This analysis ensures both length reduction and step
formation on the tool are taken into account. The mean value
of the three measuring positions resulted in the wear volume
of the grinding tool.
To ensure comparable tool conditions, the grinding tool
was dressed before each test and got a run-in of 1.32m
machining distance. For the evaluation of the cutting forces,
first the mean value was calculated out of the data from one
line of grinding. In order to compare the process forces for
the different grinding strategies, the mean cutting force
value over the course of the investigated machining path of
63.36m was then calculated.
The machining parameters for tool dressing, run-in and
the different test types are shown in Table3. For the rough-
ness tests, surface roughness was measured in the stable
Table 2 Measurement intervals
(milling) Process quality criterion Measurement interval (machining path)
Tool wear 1m, 5m, 10m, 25m, …
(each 25m until 200m, then each 50m until 500m)
Surface roughness
Rz, Ra, Rt
1m, 5m, 10m, 25m, …
(each 25m until 200m, then each 50m until 500m)
Cutting force Continuously
surfacefor
roughness
measurement
ultrasonic
vibration
v
f
n
X
Y
Z
workpiece
grinding tool
CMQL nozzle
Fig. 4 Grinding process schematic
(a)Evaluated tool wear volume
(b)Tool wear volume measuringpositions
contourline
aftertest
contour line
before test
tool wear
volume
0°
120° 240°
tool
measuring
positions
Fig. 5 Grinding tool wear analysis

Production Engineering
1 3
process window (after 10m). For the wear tests, a higher
feed-per-revolution-ratio and a longer machining path were
chosen to accelerate the wear behavior.
According to the previous studies, a CO2 mass flow of
10kg/h, an MQL volume flow of 0.8ml/min and a nozzle
position of 0° in feed direction led to a stable process, espe-
cially for longer grinding durations. With lower CO2 and
MQL flow rates, the grinding tool was thermally overloaded.
The wear study of the previous CMQL tests without US
and with different CMQL fluid types had a substantial influ-
ence on the process results (see [22]). The alcohol-based
media lead to less tool wear than the mineral-based media.
Therefore, all further tests were conducted with alcohol-
based CMQL fluid.
4 Results anddiscussion
In the following sections, the results are first presented and
thereafter discussed regarding possible cause-and-effect
relations due to the different machining processes.
4.1 Milling
4.1.1 Tool life andtool wear
The average tool life of the test series with conventional
flood cooling and CMQL with the associated standard devia-
tions are shown in Fig.6. The change in cooling lubrication
strategy already had a positive influence on the achievable
tool life with the given test setup for both tools. For tool 1,
compared to FC, a tool life extension from an average of
58.8–137.8m was achieved with CMQL. With UCMQL,
a further extension to 185.0m was realized. For this tool,
VBmax exceeded 200µm on at least two cutting edges at the
end of each experiment. In relation to FC, tool life exten-
sions of 134% with CMQL and 215% with UCMQL machin-
ing were therefore possible. Furthermore, the deviation
between the tool lives achieved in the three test runs was
reduced using US assistance. The standard deviation for the
UCMQL was 25% lower than for FC and 62% lower than for
the regular CMQL. For tool 2, compared to FC, a tool life
extension from an average of 193.2–437.6m was achieved
with CMQL. As can be seen from the large standard devia-
tion, no statement can be made as to whether the process
combination contributed to a further increase in tool life due
to the additional US assistance. However, with the UCMQL,
a machining distance of 500m was achieved in two test
runs, which was the highest achieved milling path in the
whole study. At this point, the VBmax only reached values of
58.8µm and 85.4µm, respectively.
For tool 1, the initial tool wear as well as the wear pro-
gression were lower in tests with UCMQL compared to FC.
Considerable differences in the maximum wear value or
in the deviation of the measured wear across the four cut-
ting edges of a tool, which were observed for FC, were not
detectable with the UCMQL. Figure7 displays examples of
various tool wear phenomena that were observed with the
two machining strategies. With FC, flank wear, adhesion
of workpiece material, and chipping occurred. These wear
phenomena were mainly present at the end of the engaged
Table 3 Machining parameters
for dressing, run-in, roughness
test and wear test (grinding)
Parameter Unit 1) Dressing 2) Run-in 3a) Rough-
ness testing
3b) Wear testing
Cutting width aemm 9.1 9.1 9.1 9.1
Cutting depth apmm 1 0.2 0.2 0.2
Feed / Revolution frev mm 0.2 0.05 0.05 0.10
Cutting speed vcm/min 132.5 265 265 265
US Amplitude A (resonance
frequency: approx. 22.5kHz)
µm 0 0 0; 3 0; 3
0
100
200
300
400
m
600
0
25
50
75
100
125
150
m
200
MWFMWF+USCMQLCMQL+US
CuttinglengthLc(tool 2)
CuttinglengthLc(tool 1)
tool 1
tool 2
100%
72%
201%
227%
100%
99%
315%
234%
FC FC+USUCMQL
Fig. 6 Tool life with different machining technologies

Production Engineering
1 3
cutting edge, but they also occurred partially at the cut-
ting corner. In general, UCMQL initially only led to flank
wear, especially at the end of the engagement area. At the
end of the tool life, titanium adhesions and built-up edges
were more prevalent. In addition, small chipping sometimes
occurred on the cutting edge.
Similar observations were obtained for tool 2. Over
the course of the machining process, flank wear and tita-
nium adhesions formed, which were more pronounced for
UCMQL than for FC. In contrast, chipping occurred in FC
but not with a UCMQL.
4.1.2 Cutting forces
The cutting force progression for tool 1 is shown in Fig.8 for
both series of tests with FC and UCMQL. After 5m milling
path, when the tool wear was still low and therefore compa-
rable in all tests, the cutting force was between 63 and 73N
for FC. In comparison, forces between 58 and 63N were
documented for UCMQL. In relation to FC, a reduction of
the cutting force of up to 18% was possible.
When comparing the cutting forces for tool 2, no differ-
ences could be identified between the experiments of the
different machining variations. The mean cutting forces were
almost identical at the beginning (FC: 79.7 N; UCMQL:
80N), and increased with incipient tool wear.
4.1.3 Surface roughness
Since the roughness values did not change over the mill-
ing path, the values were averaged over the complete tool
life. Due to the US assistance, higher surface roughness val-
ues were obtained in all experiments (FC + US, UCMQL)
compared to the respective tests without US assistance (FC,
CMQL). Compared to FC, the averaged Rz value for tool
1 for the UCMQL increased by 59% from 2.13 to 3.39µm
(a)FC(cuttinglengthL
c
=72m)
1mm
flank wear
1mm
(b)UCMQL (cutting length L
c
=190 m)
chipping
titanium
adhesion
Fig. 7 Tool wear phenomena using FC and UCMQL (tool 1)
0
50
100
150
200
250
300
N
400
0100 200 300m500
Cuttingforce
Milling path
tool 1: FC (1)
tool 1: FC (2)
tool 1: FC (3)
tool 1: UCMQL(1)
tool 1: UCMQL(2)
tool 1: UCMQL(3)
tool 2: FC (1)
tool 2: FC (2)
tool 2: FC (3)
tool 2: UCMQL(1)
tool 2: UCMQL(2)
tool 2: UCMQL(3)
Fig. 8 Cutting forces using FC and UCMQL

Production Engineering
1 3
and the averaged Ra value by 62% from 0.37 to 0.60µm.
For tool 2, the average Rz value increased by 33% from 2.4
to 3.2µm and the Ra value by 16% from 0.44 to 0.51µm
(see Fig.9). However, the measured values for the UCMQL
were significantly less scattered across the test series than
for FC for both tools.
4.1.4 Discussion
The experiments carried out confirm observations pub-
lished in literature about the positive impact of CMQL on
the machinability of Ti-6Al-4V [5–7] and the differences in
wear behavior depending on tool and oil combination [24].
While a pure CO2-cooling can be insufficient because of the
lack of lubrication, CMQL improves the results, especially
with oils whose properties ensure good transportability into
the cutting zone [5]. In the case of UCMQL, the high-fre-
quency vibration aids the oils transport into the cutting zone
and helps to improve the results further.
Overall, with tool 1, UCMQL increased tool life by 215%
compared to FC. In addition, the cutting forces with compa-
rable tool wear were around 10% lower with the technology
combination. The measured surface roughness was slightly
higher with USassistance than without it. In the case of tool
2, an average tool life extension of 127% was achieved with
CMQL compared to FC. Due to the large scatter of the tool
life of the UCMQL tests and because two tests were stopped
at 500m machining path, no statement can be made about
the tool life extension. However, the highest tool life was
achieved with the technology combination using this tool.
In addition, a reduction in wear phenomena was noted with
UCMQL machining.
In general, a higher surface roughness was always meas-
ured with US assistance than without, regardless of the
machining strategy; this is in accordance with the findings of
Rinck etal. [1]. With the US assistance along the tool axis, a
high-frequency contact interruption of the secondary cutting
edges and therefore a high-frequency hammering of the tool
onto the workpiece surface occurred in the investigated side
milling process. This effect causes a higher surface rough-
ness and increased tool wear, especially at the end of the
engagement area and at the cutting corner, as shown with
the previous series of tests.
For the investigated experiments with constant contact
conditions between the milling tool and the workpiece, US
assistance enables a more targeted supply of the cooling
lubricant to the actual cutting zone due to the high-frequency
vibration. However, because of the high cutting speeds and
the low US amplitudes, no complete cutting edge detach-
ment of the main cutting edges took place. The low ampli-
tudes are the reason for the less pronounced cutting force
reduction compared to the literature. In the case of a longitu-
dinal excitation of the tool, the critical cutting speed, below
which a cutting interruption takes place, can be calculated
according to
with fUS being the frequency of the tool, Al the amplitude
of the vibration and β the helix angle of the tool [1]. With
a frequency fUS of approximately 22kHz, an amplitude of
3µm, and a helix angle of approximately 40°, the critical
cutting speed is calculated to be 20.88m/min. This value is
far below the cutting speeds of 125m/min and 240m/min
tested in the process. Consequently, a complete cut interrup-
tion was not possible in any of the tests. However, the sec-
ondary cutting edges periodically detached from the work-
piece surface, which allowed the CMQL to be introduced
into this area with a potential transfer of the medium to the
main cutting edges via the flutes. These improved lubrica-
tion conditions are the reason for the reduced tool wear and
therefore longer tool life with UCMQL.
4.2 Grinding
4.2.1 Tool life andtool wear
Figure10 displays the tool wear volume after 63.36m
machining path for different cooling strategies with a US
superposition at an amplitude of 3µm. With FC, US-
assisted machining results in a wear increase of 29%.
With CMQL the wear was reduced by 24% through the US
assistance and reached an even lower level than with FC.
The standard deviations with CMQL are higher compared
(1)
vcrit =2𝜋
⋅
fUS
⋅
Al
⋅
tan (𝛽)
0
0,5
1
1,5
2
2,5
3
3,5
µm
4,5
FC FC+USCMQL UCMQL
Surfaceroughness
tool 1tool 2
154%
Ra
100%
Ra
100%
Rz
100%
Rz
150%
Rz
142%
Rz
63%
Rz
70%
Rz
116%
Ra
159%
Rz 133%
Rz
162%
Ra
68%
Ra
55%
Ra
122%
Ra
100%
Ra
Fig. 9 Surface roughness for different machining technologies and
different cooling and lubrication strategies

Production Engineering
1 3
to FC, which may be due to more process instabilities of
the CMQL system. Nevertheless, over all test runs, the
two lowest wear values have been achieved with UCMQL.
Figure11 displays examples of the observed signs of
tool wear which include attritious wear, grain fractures
and bond wear. These phenomena were present to varying
degrees regardless of machining strategy.
4.2.2 Grinding forces
The mean grinding forces for the different machining
technologies in the 63.63m machining path are shown in
Fig.12. Grinding with FC led to an average grinding force of
87N, whereas US superposition reduced this to 65N. With
CMQL, US machining reduced the mean grinding forces
from 200 to 138N. Nevertheless, machining with cryogenic
cooling led to higher force levels. Without USassistance,
the values increased by 129%, while with US assistance the
values increased by 58% compared to FC. The lowest force
levels were achieved with FC and US superposition.
4.2.3 Surface roughness
The averaged surface roughness valuesRz and Ra of the
workpiece after 10m machining path are displayed in
Fig.13. Machining with FC led to a higher roughness than
grinding with CMQL. Compared to FC, the averaged Rz
value decreased by 36% with CMQL. The average Ra value
showed a similar behavior and is 37% lower with CMQL
than with FC. With US superposition this effect is lower, and
Rz and Ra decreased by 27% compared to FC.
Generally, with US superposition the surface roughness
increases. With FC the average Ra value increased with US
machining by 6%, and the Rz value by 3%. Machining with
UCMQL leads to 15% higher surface roughness (Rz and
Ra) compared to CMQL. The lowest roughness values of all
cooling strategies can be reached with CMQL.
4.2.4 Discussion
The wear tests with US superposition showed different
results depending on the cooling strategy. With US-assisted
FC, the tool wear increased compared to FC, whereas with
UCMQL it decreased compared to CMQL. Machining with
UCMQL reached the lowest wear values of all test series.
The tool wear with CMQL was much more sensitive than
with FC when applying US superposition. For both cool-
ing techniques, grinding forces could be reduced with US
superposition, which is in accordance with Reif etal. [14].
With FC, the US assistance reduced the cutting force by
26%, whereas with CMQL it decreased by 31%.
Generally, grinding with CMQL leads to over two times
higher grinding forces but lower surface roughness values
than with FC. The reason for this could be that with CMQL
attritious and bond wear was higher compared to FC because
of the different lubricating conditions. If the cutting edges
become rounder, the surface roughness decreases. The
reduced sharpness of the rounded grains also causes higher
grinding forces. The grinding dust produced is removed
more slowly by CMQL than by FC. This also affects the
engagement conditions.
US superposition led to higher roughness values in FC
and CMQL machining. The increase in surface roughness
is caused by a kind of hammering effect, which was dis-
cussed for milling in Sect.4.1.4. In this case, the grinding
tool is struck on the surface of the workpiece in ultra-
sonic mode; this is reflected in the surface structure and
increases the roughness values.
The current results are not sufficient to describe the
phenomena. Further research is needed, especially regard-
ing the behavior of workpiece and tool materials under the
thermal conditions when using CMQL.
5 Conclusion andoutlook
This paper presents a superposition of CMQL and US
assistance as a new technique for milling Ti-6Al-4V and
grinding Zerodur. The series of tests conducted indicated
great potential for UCMQL machining of these difficult-to-
machine materials. The superposition of the technologies
had a positive influence on the different target variables in
both machining processes.
While US superposition led to a faster tool failure in
the investigated milling process with FC, tool life could be
increased by up to 215% compared to conventional FC with
UCMQL. In an industrial environment, the focus is often
0,0
0,2
0,4
0,6
0,8
mm³
1,2
MWFMWF + US CMQL UCMQL
Tool wear volume
0
100%
129%
180%
76%
FC FC+US
Fig. 10 Tool wear volume after 63.36m machining path with FC and
CMQL with and without US superposition

Production Engineering
1 3
less on tool life and more on the productivity of the manu-
facturing processes. Tool life extension with the selected
parameter combination opens up the option of increasing
the cutting parameters instead; this increase might reduce
the previously mentioned tool life extension, but it improves
the material removal rate of the process.
Regarding the grinding of Zerodur, investigations into
using alternative cooling strategies have been rare. While
(a)Emulsion(b)Emulsion+US
(c)CMQL(d) UCMQL
100µm
100µm
100µm100µm
length
reduction
andstep
formation
attritious
wear
bond
wear and
grain
vacancies
splintered
grains
Fig. 11 Tool wear phenomena after 63.36m machining path with FC and CMQL with and without US superposition
0
50
100
150
N
250
MWFMWF + US CMQL UCMQL
Grinding force
100%
74%
229%
158%
FC FC+US
Fig. 12 Grinding force after 63.36 m machining path with FC and
CMQL with and without US superposition
0
5
10
15
20
25
µm
35
MWFMWF +USCMQLUCMQL
Surfaceroughness
Rz
Ra
Rz
Rz
Rz
Ra Ra Ra
100%
100%
64%
63%
103%
106%73%
73%
FC FC+US
Fig. 13 Surface roughness after 10 m machining path using FC and
CMQL with and without US superposition

Production Engineering
1 3
currently the grinding process of Zerodur with CMQL has
not yet achieved comparable or better results in terms of
force and wear than with FC, the conducted experiments
were able to demonstrate the potential of CMQL for grind-
ing brittle-hard materials. With UCMQL, the tool wear
could be reduced to a level comparable to FC or even lower.
The positive effects of the combination of CMQL and US
assistance are starting points for further developments of the
proposed machining method.
The investigations presented in this paper considered
only simple line geometries. In future studies, the machined
geometries can be expanded to more complex and closer
to industrial workpieces. Based on the investigations con-
ducted, the following conclusions can be drawn:
• UCMQL increases the tool life of milling and grind-
ing tools compared to FC, US-assisted FC, and regular
CMQL.
• Vibration-assisted milling and grinding lead to a higher
surface roughness.
UCMQL is a new approach for combining advanced
machining techniques. The first investigations indicated
great potential for machining high-performance materials,
but further studies are necessary for industrial uses.
Acknowledgements The results presented were developed within the
project KryoSonic (Investigation of the combination of ultrasonic-
assisted machining and cryogenic minimum quantity lubrication;
project number: AZ-1333-18), which was supported by the Bavarian
Research Foundation (BFS). We thank the BFS and the industry part-
ners for their support and the good cooperation during the project. The
authors would also like to thank Alpcan Güray and Manfred Dirnberger
for assisting during the experiments and the firefighting school Regens-
burg for the provision of a ventilator unit.
Funding Open Access funding enabled and organized by Projekt
DEAL. Bayerische Forschungsstiftung, AZ-1333-18.
Declarations
Conflict of interest The authors declare that they have no conflict of
interest.
Open Access This article is licensed under a Creative Commons Attri-
bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
included in the article's Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
the article's Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
References
1. Rinck PM, Gueray A, Kleinwort R, Zaeh MF (2020) Experimental
investigations on longitudinal-torsional vibration-assisted milling
of Ti-6Al-4V. Int J Adv Manuf Technol 108(11–12):3607–3618.
https:// doi. org/ 10. 1007/ s00170- 020- 05392-w
2. Chen W, Huo D, Shi Y, Hale JM (2018) State-of-the-art review on
vibration-assisted milling: principle, system design, and applica-
tion. Int J Adv Manuf Technol 97(5–8):2033–2049. https:// doi.
org/ 10. 1007/ s00170- 018- 2073-z
3. Hafiz MS, Kawaz MH, Mohamad WN, Kasim MS, Izamshah R,
Saedon JB, Mohamed SB (2017) A review on feasibility study of
ultrasonic assisted machining on aircraft component manufactur-
ing. IOP Conf Ser Mater Sci Eng 270:12034
4. Tong J, Wei G, Zhao L, Wang X, Ma J (2019) Surface microstruc-
ture of titanium alloy thin-walled parts at ultrasonic vibration-
assisted milling. Int J Adv Manuf Technol 101(1–4):1007–1021.
https:// doi. org/ 10. 1007/ s00170- 018- 3005-7
5. Gross D, Blauhoefer M, Hanenkamp N (2020) Milling of Ti6Al4V
with carbon dioxide as carrier medium for minimum quantity
lubrication with different oils. Procedia Manuf 43:439–446.
https:// doi. org/ 10. 1016/j. promfg. 2020. 02. 190
6. Bagherzadeh A, Kuram E, Budak E (2021) Experimental evalu-
ation of eco-friendly hybrid cooling methods in slot milling of
titanium alloy. J Clean Prod 289:125817. https:// doi. org/ 10. 1016/j.
jclep ro. 2021. 125817
7. Wu G, Li G, Pan W, Raja I, Wang X, Ding S (2022) Experimental
investigation of eco-friendly cryogenic minimum quantity lubrica-
tion (CMQL) strategy in machining of Ti–6Al–4V thin-wall part.
J Clean Prod 357:131993. https:// doi. org/ 10. 1016/j. jclep ro. 2022.
131993
8. Sanchez JA, Pombo I, Alberdi R, Izquierdo B, Ortega N, Plaza
S, Martinez-Toledano J (2010) Machining evaluation of a hybrid
MQL-CO2 grinding technology. J Clean Prod 18(18):1840–1849.
https:// doi. org/ 10. 1016/j. jclep ro. 2010. 07. 002
9. Maurotto A, Wickramarachchi CT (2016) Experimental inves-
tigations on effects of fre-quency in ultrasonically-assisted end-
milling of AISI 316L: a feasibility study. Ultrasonics 65:113–120.
https:// doi. org/ 10. 1016/j. ultras. 2015. 10. 012
10. Ni C, Zhu L, Liu C, Yang Z (2018) Analytical modeling of tool-
workpiece contact rate and experimental study in ultrasonic vibra-
tion-assisted milling of Ti–6Al–4V. Int J Mech Sci 142–143:97–
111. https:// doi. org/ 10. 1016/j. ijmec sci. 2018. 04. 037
11. Verma GC, Pandey PM (2019) Machining forces in ultrasonic-
vibration assisted end milling. Ultrasonics 94:350–363. https://
doi. org/ 10. 1016/j. ultras. 2018. 07. 004
12. Zhang Y, Zhao B, Wang Y, Chen F (2017) Effect of machining
parameters on the stability of separated and unseparated ultrasonic
vibration of feed direction assisted milling. J Mech Sci Technol
31(2):851–858. https:// doi. org/ 10. 1007/ s12206- 017- 0137-x
13. Razfar MR, Sarvi P, Zarchi MM (2011) Experimental investiga-
tion of the surface roughness in ultrasonic-assisted milling. Proc
Inst Mech Eng, Part B: J Eng Manuf 225(9):1615–1620. https://
doi. org/ 10. 1177/ 09544 05411 399331
14. Reif A, Rinck P, Sitzberger S, Rascher R, Zäh MF (2020) Zerspa-
nung von Hochleis-tungswerkstoffen mit ultrasonisch modulierter
Schnittgeschwindigkeit (translated title: “Machining of high-per-
formance materials with ultrasonic-modulated cutting speed”). Z
wirtsch Fabr 115(3):162–165. https:// doi. org/ 10. 3139/ 104. 112255
15. Sun G, Shi F, Ma Z (2020) Effects of axial ultrasonic vibration on
grinding quality in peripheral grinding and end grinding of ULE.
Int J Adv Manuf Technol 109(7–8):2285–2298. https:// doi. org/
10. 1007/ s00170- 020- 05761-5
16. Arafat R, Madanchi N, Thiede S, Herrmann C, Skerlos SJ (2021)
Supercritical carbon dioxide and minimum quantity lubrication

Production Engineering
1 3
in pendular surface grinding – a feasibility study. J Clean Prod
296:126560. https:// doi. org/ 10. 1016/j. jclep ro. 2021. 126560
17. Namlu RH, Yılmaz OD, Lotfisadigh B, Kılıç SE (2022) An
experimental study on surface quality of Al6061-T6 in ultrasonic
vibration-assisted milling with minimum quantity lubrication.
Procedia CIRP 108(3):311–316. https:// doi. org/ 10. 1016/j. procir.
2022. 04. 071
18. Ni C, Zhu L, Yang Z (2019) Comparative investigation of tool
wear mechanism and corresponding machined surface characteri-
zation in feed-direction ultrasonic vibration assisted milling of
Ti–6Al–4V from dynamic view. Wear 436–437:203006. https://
doi. org/ 10. 1016/j. wear. 2019. 203006
19. Madarkar R, Agarwal S, Attar P, Ghosh S, Rao PV (2018)
Application of ultrasonic vibration assisted MQL in grinding of
Ti–6Al–4V. Mater Manuf Process 33(13):1445–1452. https:// doi.
org/ 10. 1080/ 10426 914. 2017. 14154 51
20. Gross D (2021) Untersuchungen zur kohlenstoffdioxidbasierten
kryogenen Minimalmengenschmierung (translated title: “Inves-
tigations into carbon dioxide-based cryogenic minimum quan-
tity lubrication”). Dissertation, Friedrich-Alexander-Universität
Erlangen-Nürnberg. https:// doi. org/ 10. 25593/ 978-3- 96147- 413-4
21. Blasl J, Vieltorf F, Hanenkamp N, Zäh FM (2021) Grundstein für
neuartige Bearbeitungsstrategie gelegt (translated title: “Corner-
stone laid for novel machining strategy”). VDI-Z 163(01–02):29–
31. https:// doi. org/ 10. 37544/ 0042- 1766- 2021- 01- 02
22. Lichtinger K, Reif A, Sitzberger S, Rascher R (2021) Zerodur
mit kryogener MMS und Ultraschall bearbeiten (translated title:
“Machining of Zerodur with cryogenic MQL and ultrasonic
vibrations”). VDI-Z 163(03):39–41. https:// doi. org/ 10. 37544/
0042- 1766- 2021- 03
23. Perez H, Diez E, Perez J, Vizan A (2013) Analysis of machin-
ing strategies for peripheral milling. Procedia Eng 63:573–581.
https:// doi. org/ 10. 1016/j. proeng. 2013. 08. 193
24. Meier T, Seydaack M, Gross D, Hanenkamp N (2022) Investiga-
tion of the influence of the additivation of a biological metalwork-
ing oil in cryogenic machining. Procedia CIRP 105:694–699
Publisher's Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.