Sie können Operatoren mit Ihrer Suchanfrage kombinieren, um diese noch präziser einzugrenzen. Klicken Sie auf den Suchoperator, um eine Erklärung seiner Funktionsweise anzuzeigen.
Findet Dokumente, in denen beide Begriffe in beliebiger Reihenfolge innerhalb von maximal n Worten zueinander stehen. Empfehlung: Wählen Sie zwischen 15 und 30 als maximale Wortanzahl (z.B. NEAR(hybrid, antrieb, 20)).
Findet Dokumente, in denen der Begriff in Wortvarianten vorkommt, wobei diese VOR, HINTER oder VOR und HINTER dem Suchbegriff anschließen können (z.B., leichtbau*, *leichtbau, *leichtbau*).
Dieser Artikel untersucht die Fräsleistung austenitischer NiTi-Formgedächtnislegierungen, wobei der Schwerpunkt auf den Auswirkungen von Schnittgeschwindigkeit, Vorschubgeschwindigkeit und axialer Schnitttiefe liegt. Die Studie zeigt, dass die axiale Schnittkraftkomponente Fz dominiert, wobei die Schnittkraftkomponenten bis zu 2044,92 N und maximale Spantemperaturen von 230 bis 340 ° C erreichen. Die Forschung identifiziert die Bildung einer martensitischen Schicht mit tiefen martensitischen Nadeln nach dem Trockenfräsen und unterstreicht die Bedeutung nachfolgender technologischer Operationen, um diese induzierte Schicht zu entfernen. Die Studie untersucht auch den Werkzeugverschleiß, die Spänemorphologie und die Tiefe der bearbeitungsinduzierten Schicht und liefert wertvolle Erkenntnisse zur Optimierung des Fräsprozesses für NiTi-Legierungen. Die Ergebnisse bieten ein umfassendes Verständnis der Herausforderungen und Möglichkeiten, die Fräsleistung zu verbessern und die Qualität der Fertigprodukte sicherzustellen.
KI-Generiert
Diese Zusammenfassung des Fachinhalts wurde mit Hilfe von KI generiert.
Abstract
NiTi shape memory alloys hold significant potential for advanced applications in biomedical and high-performance engineering fields, making the development of reliable and efficient manufacturing technologies increasingly critical. The face milling process of NiTi alloys requires thorough investigation, both as a pre-processing operation—particularly in the context of assessing technological heritage—and as a standalone technique. This study investigated the face milling performance of austenitic Ni56.5Ti43.5 (wt.%) shape memory alloy as a function of cutting conditions in dry machining with uncoated cemented carbide cutting inserts. Tests were performed at different cutting speeds (20, 35, and 50 m/min), feed rates (0.05, 0.1, and 0.15 mm/tooth), and axial depths of cut (0.5, 0.75, and 1.0 mm). The article describes the cutting force components, temperature, surface roughness of the milled surface, tool wear, microstructure, and depth of the machining-induced layers, as well as the chip morphology. Cross-section analysis revealed a heterogeneous surface layer with a predominant martensite content and a deeper subsurface layer with irregular, deep martensite needles. The influence of cutting conditions on the depth of each of these layers is discussed. The failure modes of the cutting insert and the wear mechanism are determined. The findings deepen the understanding of the thermo-mechanical responses of NiTi during face milling, especially the extensive phase transformations that occur in the surface layer. The study emphasizes the importance of identifying and removing the mechanically affected layer in post-machining steps to preserve the alloy’s functional properties.
Andrey Manokhin and Steffen Ihlenfeldt contributed equally to this work.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
1 Introduction
The NiTi alloy, also known as nitinol, titanium nickelide, or nickel-titanium, is a binary alloy consisting of approximately 50 at.% nickel and 50 at.% titanium. It is one of the most well-known shape-memory alloys (SMAs) with unique physical properties such as shape memory and hyperelasticity, high strength, high corrosion resistance, high damping capacity, and excellent biocompatibility. Its potential applications include biomedical implants, precision instruments, robotic actuators, aerospace components, and micro-electro-mechanical systems.
However, the machining of NiTi is challenging for several reasons, including high strain hardening, unconventional stress–strain characteristics, severe adhesion, and poor chip breaking [1]. Inherent material properties such as low thermal conductivity, high specific heat, and low effective Young’s modulus also present challenges to machinability. These overall characteristics lead to high specific cutting energy and forces, severe tool wear, and excessive burr formation—ultimately compromising the surface integrity of the finished product.
Anzeige
Phase transformation significantly affects the machining of NiTi alloys [1]. These materials have a narrow transformation temperature window (~ 80 K) influenced by composition, heat treatment history, manufacturing method, and cold working. Even slight thermal changes can alter the material properties significantly, making machinability highly sensitive to machining parameters [2]. In the austenite phase, NiTi is hard and rigid due to its cube symmetry, while in the martensitic phase it becomes nonsymmetrical and somewhat ductile and softer. This behavior results in a narrow window of proper machining parameters, which must be determined experimentally for specific cutting conditions.
The functional performance of NiTi SMAs—shape memory, pseudoelasticity, actuation, and damping properties—is of great importance for their application. Machining parameters can affect the functional behavior of NiTi, as its phase transformation is sensitive to stress and temperature. The high temperatures and severe plastic deformation induced during machining can alter surface and subsurface properties, which in turn influence the material’s functional performance. A study [3] revealed that the phase transformation response of the austenitic NiTi alloy depends on the cutting speed under both dry and cryogenic turning conditions. After cutting, the first 500 µm of material from the machined surface exhibited different phase transformation behavior compared to the bulk material. NiTi SMAs are commonly used for thin-walled parts in the biomedical and aerospace industries. The phase transformation response of such components can be altered through machining.
Few studies have investigated the cutting temperature during orthogonal cutting of austenitic NiTi. The numerical modeling of the orthogonal cutting process of austenitic NiTi carried out in DEFORM-2D [2] showed that as the cutting speed was increased from 12.5 to 100 m/min, the maximum cutting temperature increased from 358 to 1140 °C. Cutting temperature measured by the tool thermocouple method during orthogonal turning of austenitic NiTi (56% Ni, Ti-balance, wt.%) was obtained in [4]. The temperature in the cutting zone increased from 500 to 950 °C as the cutting speed increased from 5 to 100 m/min. This is due to high internal friction and deformation resistance, which lead to high heat generation when machining NiTi alloys.
Several studies have analyzed the cutting processes of NiTi SMAs. The study in [3] investigated the effects of cutting speed in cryogenic and dry orthogonal cutting on the surface integrity of NiTi (55.82% Ni, Ti-balance, wt.%, in the austenitic phase at room temperature (RT)). They found that higher cutting speeds reduced subsurface hardness and increased latent heat for phase transformation. The depth of the machining-induced layer decreased with increased cutting speed. The chips exhibit similar behavior with regard to transformation response and microhardness.
Anzeige
The study [2] presents a modified Helmholtz free energy-based phase transformation model for chip formation in orthogonal machining of austenitic NiTi. Their results show a strong link between cutting speed and phase transformation, predicting a deeper machining-induced layer at higher cutting speeds. This result contradicts their previous findings [3]. They also show that the phase transformation heavily influences the chip morphology. The exact chemical composition of the alloy was not specified.
The work [5] examined the effects of various cutting speeds and conditions (dry, CO2, high-pressure, and flood cooling), on the surface integrity of the NiTi alloy (55.82% Ni, Ti-balance, wt.%, in austenitic phase at RT) after turning. Their findings reveal that the depth of the affected layer decreased with cutting speed in all cutting conditions because of the annihilation effect from high cutting temperatures. Similar effects were reported for NiTi SMA (50.8% Ni, Ti-balance, at.%, in austenitic phase at RT) in [6], where the mechanical effect was the dominant influence on surface integrity at low cutting speeds, while the thermal effect dominated at high cutting speeds. Work hardening was more significant at low cutting speeds.
The study [7] investigated the turning performance of NiTi alloy (49.9% Ni, Ti-balance, at.%, in martensitic phase at RT) under precooled cryogenic, dry, minimum quantity lubrication (MQL), and preheated machining conditions. Findings reveal that precooled cryogenic machining, which maintained the martensitic phase, profoundly reduced cutting force components, notch wear, and surface roughness. Machining in the austenite state via preheating offered no benefit over dry or MQL machining. All processes were generally less effective than cryogenic machining, particularly at higher cutting speeds. Cryogenic machining also reduced tool wear, cutting force components, and tool-chip contact length.
The phase characteristics of NiTi SMA (55.81% Ni, Ti-balance, wt.%, in austenitic phase at RT) related to turning parameters were studied in [8]. The surface microstructure remained austenitic across cutting speeds of 5.28…96 m/min at a feed rate of 0.15 mm/r and depth of cut of 0.2 mm. In contrast, [9] illustrated that turning NiTi alloy (49.9% Ni, Ti-balance, at.%, in martensitic phase at RT) under different cutting and cooling conditions affects the surface integrity and phase transformation responses.
The study [10] examined how cutting speed affects the shape recovery of work material in turning NiTi alloy (56% Ni, Ti-balance, wt.%, in austenitic phase at RT), which can reduce dimensional accuracy. At low cutting speeds, the temperature remained below the phase transformation threshold, causing severe shape recovery throughout the whole cutting process due to the phase transformation. When the cutting speed increased to 100 m/min, the temperature of the work material near the cutting point exceeded the threshold temperature of phase transformation; thus, the work material did not generate an obvious shape recovery because it could not undergo any form of phase transformation during the stable part of the cutting process and after the feed stopped. The authors consider increasing the cutting speed as a possible approach to improving dimensional accuracy by inhibiting shape recovery of the work material in the cutting process of NiTi alloys.
The chip shape and microstructure during turning of NiTi SMA (50.8% Ni, Ti-balance, at.%, in austenitic phase at RT) were analyzed in [11] to understand material flow behavior. The martensitic phase transformation was found to be essential for the material flow and the chip formation. At low cutting speeds, strain hardening dominated the material flow in the chip. Increased cutting speed enhanced the thermal softening effect and martensitic transformation, which promoted overall material softening.
It is known that studies exist on the micromilling of NiTi alloys. However, the cutting conditions and thus the stresses and temperatures that are typical for micromilling are in a completely different range of values compared to conventional milling. In addition, the microcutting process is characterized by different chip formation conditions due to the size and ploughing effects. In this regard, known works primarily focus on the control of burr formation [12], ensuring minimum roughness and high-quality surface microtextures [13], specific cutting forces [14], and the investigation of tool wear [15], and only a few studies examine the phase transition in the subsurface layer of the workpiece [16].
In contrast, conventional milling is used for larger NiTi components, such as medical bone plates, yet studies on milling of NiTi remain very limited. The influence of the end milling parameters on the surface roughness and strain hardening of the NiTi alloy (50.8% Ni, Ti-balance, at.%) was studied in [17]. The research shows that the medium range of cutting speed selection minimizes work hardening and surface roughness, though the phase composition of the workpiece was not reported.
Altas et al. [18] studied surface roughness during face milling of the NiTi SMA (55.8% Ni, Ti-balance, wt.%, in austenitic phase at RT) using cutting tools with different nose radii under dry cutting conditions. They found that higher cutting speeds and feed rates increase flank wear on the cutting tool and increase surface roughness. Minimum surface roughness Ra (0.346 μm) was achieved at 20 m/min cutting speed and 0.03 mm/tooth feed rate using cutting tools with 0.8 mm nose radius.
Altas et al. [19] investigated the surface integrity of NiTi SMA (55.8% Ni, Ti-balance, wt.%, in austenitic phase at RT) after face milling with cryogenically treated/untreated cemented carbide cutting tools under dry and MQL cutting conditions. Optimal results were achieved at a cutting speed of 50 m/min and feed rate of 0.03 mm/tooth using boron-added cutting fluid (EG + %5BX) with deep cryogenic heat-treated (− 196 °C) CVD-coated S40T grade cutting tool. The obtained results indicate a phase transformation in the machined material under different cutting conditions. Improved performance was attributed to better heat absorption by cryo-treated tools from the cutting zone, which reduced microstructural changes. The reduced thermal and mechanical loads resulted in a smaller deformation zone under the machined surface. Tools protected by a hard coating were recommended to increase their wear resistance. However, in this work, the effect of cutting conditions on the cutting forces, temperatures, and microstructure of the machined near-surface layer of NiTi was not investigated. This represents a gap in this area and indicates the importance of further research.
The effect of feed rate on surface integrity when machining NiTi alloy (50.8% Ni, Ti-balance, at.%) with a carbide single-tooth milling cutter coated with AlTiN/TiN with a general-purpose synthetic coolant was studied in [20]. The thickness of the subsurface white layer was the largest at the lowest feed rate. The authors attribute this to the longer duration of mechanical loading at low feed, which in turn leads to a deeper phase transformation. The authors suggest that the white layer is in an austenitic state since the detwinned martensite caused by considerable plastic strain during cutting will transform back to austenite above the Af (finish austenite formation temperature). It was also assumed that the presence of the martensite phase in the subsurface was formed by minor plastic strain in the subsurface. However, the initial microstructure was not reported, and cutting forces and temperatures—key factors for surface integrity—were not investigated.
Kaya et al. [21] investigated the effect of cutting speed and cutting tool materials on the surface integrity and functional properties in the milling of NiTi shape memory alloy (50% Ni, 50% Ti, at.%, in austenitic phase at room temperature). The authors found that an increase in cutting speed resulted in less affected transformation enthalpy and less affected transformation temperature hysteresis. The authors address alterations to the surface layer of the workpiece material; however, they do not examine the mechanical forces and temperatures within the cutting zone.
Numerous studies have examined the effects of cutting conditions on the surface roughness, tool wear, phase transformations, and subsurface integrity of NiTi alloys, primarily in turning and micromilling operations, as evidenced by the literature. However, due to the distinct cutting mechanics, thermal conditions, and geometric scales involved in face milling, these findings cannot be directly applied. Research specifically addressing face milling of NiTi alloys is limited, particularly with regard to surface roughness, tool wear, and subsurface microstructure. Furthermore, the cutting force components, temperature in the cutting zone, and chip morphology during face milling have not been systematically studied. This study aims to evaluate how cutting speed, feed rate, and axial depth of cut influence the cutting force components, temperature, roughness of the milled surface, tool wear, microstructure, and the depth of the layer induced by the machining process, and the chip morphology of austenitic Ni56.5Ti43.5 (wt.%) SMA during face milling.
Section 2 provides a comprehensive description of the materials employed and the methodological framework adopted in the study. Section 3 reports the results obtained along with their interpretation and discussion. Finally, Sect. 4 synthesizes the principal findings.
2 Materials and methods
Section 2 is organized into three subsections, which detail the workpiece material, machine tool, and tooling, experimental setup, and cutting parameters, as well as the measurement setup and data analysis.
2.1 Workpiece material, machine tool, and tooling
The material used in this study is a commercially available NiTi alloy with a composition of Ni56.5Ti43.5 (wt%). Its chemical composition, mechanical, and thermal properties are given in Tables 1 and 2. The samples were supplied as 30 × 30 × 30-mm cubes.
Table 1
The chemical composition of Ni56.5Ti43.5 (wt%) SMA
Ni
Ti
O
C
Fe
N
Si
H
56.05
43.65
0.019
0.007
0.006
0.003
0.001
0.001
Table 2
Mechanical and thermal properties of Ni56.5Ti43.5 (wt%) SMA [10]
Property
Hardness
HV
283
Thermal conductivity
W·(m·̊C)−1
20 (at 20 ̊C)
Expansion
ˣ10−6
11.3 (at 20 ̊C)
Young’s modulus
GPa
83 (at 20 ̊C)
Poisson’s ratio
0.3
Specific heat capacity
J/kg·K
230…314 (at 20 ̊C)
Yield strength
MPa
280
Tensile strength
MPa
630
Phase transformation behavior is defined by the following characteristic temperatures: martensite start (Ms = − 23.6 °C), martensite finish (Mf = 41.2 °C), austenite start (As = − 17.9 °C), and austenite finish (Af = − 2.5 °C) [10]. In addition, the martensite desist temperature (Md), approximately 150 °C, marks the upper limit for stress-induced martensitic transformation. Below Md, the austenite phase can transform into martensite through plastic deformation—a phenomenon known as strain-induced martensitic transformation. The material exhibits superelasticity in a temperature range of − 2.5 to 150 °C. XRD (X-ray diffraction) and metallographic analysis (Figs. 1 and 2) before cutting confirmed that the samples were in the austenite state at RT and capable of superelastic deformation. Porosity was also observed.
As face milling performance on austenitic NiTi shape memory alloy has not been investigated before, it is advisable to start with a baseline setup using uncoated carbide inserts under dry cutting conditions with conservative parameters. This provides a reference point against which to evaluate fundamental tool–workpiece interactions, independent of the influence of coatings or cutting fluids. NiTi has low thermal conductivity and a high tendency to work-harden; it is also prone to adhesive wear and built-up edge formation. Coatings or lubrication may initially mask the underlying cutting behavior. Uncoated inserts generally have a smaller edge radius, which reduces adhesive wear in initial trials. Thus, the methodology used minimizes variables and isolates the effects of cutting parameters and tool geometry before introducing lubrication or coatings.
The five-axis CNC vertical machining center (DMG DMU 80 eVo) was used in the face milling tests. The setup is displayed in Fig. 3. All tests were conducted under dry conditions. Cutting was performed using a Sandvik Coromant face milling cutter CoroMill 300 (R300-052C5-08H) equipped with eight uncoated cemented carbide round inserts (grade H13A). The milling inserts had an 8 mm inscribed circle diameter, a 2.78 mm thickness, a 0.1 mm face land width, and a 15° face land angle. A new cutting edge was used for each test to ensure consistent conditions.
Symmetrical dry face milling was performed. All specimen surfaces were prepared using the same tool and cutting parameters (cutting speed V = 20 m/min, feed rate fz = 0.15 mm/tooth, depth of cut ap = 1 mm). The cutting length for all experiments was 30 mm. The runout of the tool’s cutting edges was monitored at each insert change and did not exceed 5 μm. The machining parameters used in the cutting tests are shown in Table 3. Each test was performed three times, and the results were averaged.
Table 3
Experimental details
Levels
V (m/min)
fz (mm/tooth)
ap (mm)
1
20
0.05
0.5
2
35
0.10
0.75
3
50
0.15
1.0
2.3 Measurement setup and data analysis
Cutting forces and chip temperature were measured during each cutting pass. A Kistler 9257B multi-component dynamometer was used to measure the cutting force components Fx, Fy, and Fz in the machine coordinate system. The analog output signal was amplified using a Kistler type 5070 charge amplifier and then digitized using a Goldammer USB-Basic G0S-1034–4 digital analog quantifier (DAQ). Subsequent filtering and processing were performed using the National Instruments DIAdem software suite.
Chip temperature was measured using an Optris PI 450i thermal imaging camera at an ambient temperature of 20 ± 1 °C. The maximum temperature was measured within a 175 × 75 mm area covering the cutting zone between the cutting tool and the workpiece. It should be noted that the infrared camera captures the chip temperature after material separation, rather than directly measuring the tool–chip interface or primary shear zone. As a result, the reported chip temperatures likely underestimate the peak temperatures in the cutting zone, where phase transformations and thermal softening are initiated. Nevertheless, chip surface temperature trends provide valuable comparative data between cutting conditions and can be correlated with finite element predictions and other indirect methods.
The surface roughness was measured along the feed direction using a Mitutoyo Surftest SJ-410 profilometer (Fig. 4). X-ray diffraction analysis of the as-received samples was performed using a Bruker D8 Discover X-ray diffractometer. The microstructure of the as-received specimen and the machining-induced layers of specimens after milling were determined by metallographic analysis using a KEYENCE VHX digital microscope with the differential interference contrast (DIC) method. The areas of the milled surfaces for the roughness and microstructure studies are shown in Fig. 4.
Fig. 4
Roughness and microstructure investigation areas (mm)
Wear on cutting inserts and the shape of the chips were assessed using a KEYENCE VR-6000 optical profilometer. Scanning electron microscopy (SEM) of cutting inserts and chips was performed on a JSM-6490LV microscope with an INCA Energy 350XT energy dispersive spectrometer and backscattered electron diffraction pattern registration system (HKL Channel 5 EBSD, JEOL Ltd.).
There are some limitations regarding the research methodology. Other factors such as cutting insert geometry, tool material, coating, coolant, and chemical and phase composition of the workpiece material are not included in this research. This study did not involve directly evaluating the functional properties of NiTi shape memory alloys after machining, such as the shape memory effect or superelasticity, as well as direct phase analysis of the chips (e.g., by X-ray diffraction or electron backscatter diffraction).
3 Results and discussion
This section presents the results of the experimental investigation, organized into four subsections: (1) cutting forces, chip temperature, and surface roughness; (2) microstructure of the machined surface layer, including the depth of the affected zone; (3) tool wear; (4) chip shape and morphology.
3.1 Cutting forces, temperatures, and surface roughness
The experimental values of the cutting force components (Fx, Fy, and Fz), equivalent cutting force F, chip temperature T, and average roughness of the milled surfaces Ra are shown in Table 4.
Table 4
Experimental results (average values)
V (m/min)
fz (mm/tooth)
ap (mm)
Fx (N)
Fy (N)
Fz (N)
F (N)
T (°C)
Ra (µm)
1
50
0.05
0.5
315.24
414.43
614.99
755.56
250
0.617
2
20
0.05
1.0
668.03
881.96
1108.4
1414.55
300
0.62
3
50
0.05
1.0
710.45
739.75
923.83
1248.27
320
0.58
4
50
0.15
1.0
1200.26
1492.92
1805.42
2413.57
340
1.347
5
20
0.05
0.5
273.13
496.83
529.05
707.20
230
0.463
6
20
0.15
1.0
1293.03
1727.91
2044.92
2705.26
257
1.433
7
50
0.15
0.5
649.11
915.53
1234.13
1559.30
245
0.97
8
20
0.15
0.5
505.68
969.54
1020.75
1363.41
245
0.953
9
35
0.1
0.75
773.01
986.94
1203.86
1594.66
255
0.773
As shown in Table 4, the components of the cutting force vary significantly when the cutting conditions change during face milling. In all tests, the dominant component of the cutting force was Fz, while Fx was the smallest. This distribution of the cutting force components is attributed to the peculiarities of chip formation and chip flow mechanics when machining high-strength, ductile NiTi, the positive geometry of the round inserts, the relationship between feed rate and depth of cut, and the kinematics of the face milling process.
The cutting force components reached high values (Fz = 2044.92 N, test 6), indicating high loads on both the cutting tool and the machined surface. This ultimately leads to the breakage of the insert and deterioration of the surface integrity parameters.
Depending on the cutting conditions, the maximum values of the measured temperatures ranged from 230 to 340 °C, and the surface roughness Ra was found to be between 0.58 and 1.347 µm.
Multiple linear regression models were developed to describe the dependence of cutting force components and equivalent cutting force F, as well as temperatures and surface roughness, on the cutting conditions, based on experimental data and analysis of variance:
According to the data (Appendix), feed is the predominant factor influencing the cutting force components Fy, Fz, and the equivalent cutting force F, while the depth of cut has a secondary effect. In contrast, the Fx component is primarily influenced by the depth of cut. Within the range of the tested cutting conditions, cutting speed had no significant influence on the force components or the equivalent cutting force F. Temperature, however, was found to depend significantly on both the depth of cut and the cutting speed, while surface roughness showed a significant dependence on both the feed rate and the depth of cut.
Table 5 evaluates the reliability and accuracy of the developed regression models. The R2 of all fitting regression equations was greater than 0.75, indicating that the regression models are statistically acceptable.
Table 5
Coefficient of determination (R2) for each response variable
Variation in source
Coefficient of determination R2
Fx
0.9663
Fy
0.9782
Fz
0.9524
F
0.9704
T
0.7596
Ra
0.9578
With the regression models validated, the effect of the cutting conditions on the equivalent cutting force F, the roughness of the machined surface, and temperature can be interpreted as shown in Figs. 5, 6, and 7. The equivalent cutting force reaches a maximum of 2700 N at a feed rate of 0.15 mm/tooth and a depth of cut of 1 mm.
Fig. 5
Effect of the feed rate fz and the depth of cut ap on equivalent cutting force F (V = 20 m/min)
The obtained results on surface roughness are in agreement with the data of [18], which examined the dependence of roughness on feed per tooth during dry face milling of NiTi using uncoated carbide inserts of the same grade (H13A) but with different geometry (rectangular, with a cutting edge angle of 90°). The surface roughness Ra obtained in [18] varied from 0.346 to 0.736 µm depending on the cutting conditions. However, unlike [18], cutting speed had no significant effect on surface roughness.
Measured temperatures exceeded 240 °C, indicating that face milling of NiTi generates extremely demanding thermal and mechanical conditions [4]. They can critically influence tool life and machined surface integrity. However, a direct comparison of the obtained data on cutting forces and temperatures with other studies was not possible, since no relevant data for the face milling of NiTi alloy were found in the reviewed studies.
3.2 Microstructure of the surface layer and depth of the induced layer
Thermal and mechanical effects during machining are the main reasons for microstructure changes in the machined surface layer due to phase transformations and plastic deformation [3]. The depth of the affected subsurface layers was determined by measuring the depth of the twinned martensite zone in cross-sections.
Metallographic analysis showed a non-homogeneous microstructure with deep, irregularly oriented martensitic needles in the subsurface layer after machining (Fig. 8).
Fig. 8
Microstructure of the machined surface layer. (test 4: Vc = 50 m/min, fz = 0.15 mm/tooth, ap = 1.0 mm)
Unlike former studies, which observed either no changes [8], the formation of twinned martensite [5], or reported visual changes in the microstructure [9], this work documents distinct martensite needle formation in the surface layer of NiTi after dry milling. Such martensitic transformation can degrade the shape memory effect and functional performance of NiTi components. Therefore, machining conditions should be selected to minimize possible martensitic transformation. When face milling is used as the initial processing step, subsequent technological operations must ensure complete removal of the induced surface layer.
Three distinct microstructural layers were identified in the cross-sectional view of NiTi after milling (Fig. 9). Differential interference contrast imaging revealed a distinct, high-contrast (gold-colored) surface layer 3 with a dense, twinned morphology. This is consistent with a deformation-induced martensitic structure. This transitions to a more homogeneous, low-contrast structure 2 (blue-colored), which was consistent with austenite, with individual deformation-induced martensitic structures present at a greater depth. Zone 1 was unaffected and consistent with austenite. The depths of the affected layers were determined by averaging multiple measurements taken with a digital optical microscope.
Fig. 9
Microstructure of the machined subsurface layer and the machining-induced affected layers (μm): a test 5: V = 20 m/min, fz = 0.05 mm/tooth, ap = 0.5 mm; b test 6: V = 20 m/min, fz = 0.15 mm/tooth, ap = 1.0 mm; c test 1: V = 50 m/min, fz = 0.05 mm/tooth, ap = 0.5 mm; d test 4: V = 50 m/min, fz = 0.15 mm/tooth, ap = 1.0 mm
The results indicate that the cutting speed has the most significant effect on the depth of the machining-induced layers (Fig. 9). As the cutting speed increases from 20 to 50 m/min, the depth h increases significantly by 53.8 to 57.1%, while the total depth t decreases significantly by 32.7 to 42.5%. This aligns with previous results obtained for other machining methods [5], where higher speeds produce shallower deformation zones due to increased temperatures and reduced contact time [23].
In contrast, increasing the feed rate from 0.05 to 0.15 mm/tooth and the cutting depth from 0.5 to 1.0 mm, the depth of both layers increases gradually by 7.7 to 10% for h and by 2.5 and 20% for t (Fig. 9). These results partially align with prior results in [23] regarding the increase in the depth of the induced layer with increasing feed. Notably, the results contradict the conclusions of [19], which state that smaller feed rates produce a thicker white layer because it imposes higher mechanical and thermal loading, which results in phase transformation in the deeper subsurface. However, that conclusion lacked metallographic evidence. They did not analyze the initial microstructure, and relied on assumptions about phase composition.
The effect of the cutting depth on the depth of the induced layer when cutting NiTi alloy has not yet been investigated separately. The results suggest that greater cutting depths require more mechanical work due to increased plastic shear deformation, leading to higher cutting temperatures. Thus, the depth of the machining-affected layer also increases.
Of all the cutting conditions, cutting speed has the most pronounced impact on the total depth of the induced layer. The effects of feed rate and depth of cut are less substantial and have not yet been investigated in sufficient depth. Unlike other authors who studied the effect of cutting speed only on the total depth of the machining-induced layer, the presented study distinguished two zones.
As suggested in [5], the upper surface layer of the workpiece after machining is influenced by both mechanical and thermal effects, while the layer below it is affected primarily by mechanical stress (stress-induced martensite). That is why the formation of stress-induced martensite cannot be completely avoided, even with intensive cooling.
Thus, the depth h (Fig. 10a) rises significantly with increasing cutting speed due to temperatures reaching the phase transformation threshold, promoting the depth and density of martensitic needles. With increasing feed rate and depth of cut, a gradual growth of the depth h is observed due to elevated stresses from deforming a larger volume of material. These stresses contribute to a slight temperature increase in the chip formation zone.
Fig. 10
Average depth of machining-induced layers under different cutting conditions
In contrast, depth t decreases with increasing cutting speed. This results from a shorter duration of mechanical loads, which decreases the formation of martensitic needles in the deeper surface. At the same time, higher feed rates and depth of cut slightly increase depth t, as greater mechanical loads occur (Fig. 10b).
The formation of a non-homogeneous layer with deep martensitic needles during milling can have a detrimental impact on the shape memory effect and compromise the functional properties of NiTi parts. Therefore, machining parameters should be selected in such a way as to minimize the depth of the non-homogeneous induced martensite layer. Subsequent technological operations must involve the removal of the induced surface layer to its full depth, which can reach hundreds of micrometers.
3.3 Cutting insert wear
Insert wear on the rake and flank faces was investigated under variable cutting conditions. In all tests, wear on the insert’s rake face was predominant. An example of the insert after milling a length of 30 mm is shown in Fig. 11.
Fig. 11
Worn cutting insert (30 mm milling length):. a test 6: V = 20 m/min, fz = 0.15 mm/tooth, ap = 1.0 mm; b test 4: V = 50 m/min, fz = 0.15 mm/tooth, ap = 1.0 mm; c test 8: V = 20 m/min, fz = 0.15 mm/tooth, ap = 0.5 mm
At all cutting conditions, pinkish-purple spots of varying sizes appeared on the rake face of all inserts. These spots, removable with ethanol, are located outside the chip contact zone, above and in the opposite direction to the cutting speed. Their size and density increase with increasing cutting forces and temperature (Fig. 11). Their formation can be attributed to a thin film of cobalt salt, resulting from the reaction between cobalt in the carbide and the atmosphere during heating, then deposited on cooler areas of the rake face. Notably, this phenomenon was not observed when the same cutting inserts were utilized for milling titanium alloy grade 2 in a wide range of cutting conditions. This suggests that dry milling NiTi generates exceptionally high cutting temperatures, underscoring the importance of effective cooling strategies.
The profiles of the new insert and worn areas are shown in Fig. 12. Measurements of the wear areas revealed that the wear on the rake face reached a maximum value of 0.756 mm (test 6, Fig. 11a) and the wear on the flank face had a maximum value of 0.097 mm (test 4, Fig. 12c).
Fig. 12
Profile of the worn areas of the cutting insert (cross-section): a new cutting insert; b test 6: V = 20 m/min, fz = 0.15 mm/tooth, ap = 1.0 mm; c test 4: V = 50 m/min, fz = 0.15 mm/tooth, ap = 1.0 mm
Figure 13 shows the rake face of another worn cutting insert from a set of inserts mounted on the milling cutter. The failure modes were chipping of the cutting edge, flaking, notch wear, thermal cracks, and built-up edge (BUE).
The worn face of the carbide shows deep, smooth grooves and sliding strips, formed by removing the adhesion layer. Thermal cracks were observed at localized areas along the cutting edge. Cutting-edge chipping prevailed at maximum cutting force values, while at more moderate loadings, notch wear and BUE were dominant. It was found that both adhesion and abrasion contributed to the wear.
During cutting with uncoated carbide inserts, cobalt layers are first removed from the insert surface due to their lower hardness and higher chemical activity than tungsten. Energy dispersive X-ray analysis (EDX) (Fig. 14) confirmed that there is a severe decrease in cobalt concentration on the rake face within the wear zone. This removal causes micro-roughness from protruding carbide grains (Fig. 14b, spectrum 4). EDX also indicates that NiTi alloy components penetrate into the surface layers of the carbide and predominantly accumulate in the grooves formed after cobalt removal (Fig. 14b, spectrum 2). The oxygen content in the worn surface layers (1.43–3.56%, Fig. 14b, spectra 1–4) exceeds that of the base carbide (1.36%, Fig. 14a), suggesting oxygen diffusion and the formation of an oxide layer.
Fig. 14
EDX spectrum of cutting inserts: a new cutting insert; b wear area
The results obtained are generally consistent with the findings of [18] for dry face milling of NiTi using uncoated rectangular cutting inserts made of H13A carbide. The difference between the wear modes of round inserts is the presence of flaking, notch wear, and thermal microcracks. In both cases, both adhesion and abrasion wear mechanisms occur.
3.4 Chip shape and morphology
The shape of the chip provides insights into the machining process. Figure 15 shows the general appearance of the chips produced during face milling under different cutting conditions. In all cases, discontinuous, short, serrated, cone-like chips were formed. Because mechanical and thermal loads affect chip formation differently and are the main causes of phase transformations in NiTi alloys, significant differences in chip shape and color are visible.
Fig. 15
Effect of cutting conditions on the chip morphology
The macroscopic chip images in Fig. 15 reveal brighter coloration at higher cutting speeds, indicating temperature-induced microstructure alteration (Fig. 15a–c). The only exception occurred at a high feed rate of fz = 0.15 mm/tooth and a depth of cut of ap = 1.0 mm (Fig. 15d). This alteration suggests a phase transformation associated with the formation of a white layer, which occurs when both stress aging and recrystallization are present. Under these conditions, the mechanical load caused by the cutting forces is sufficient to trigger phase transformation, even at low cutting speeds.
Rising cutting temperatures contribute to material softening, reflected in smoother inner chip surfaces. At a cutting speed of 50 m/min, the folds on the inner chip surface are shallower. As before, the exception is machining with a high feed rate and depth of cut (Fig. 15d). These results align with the findings from turning with constant feed and depth of cut in [11]. At the same time, the cutting speed affects chip formation by influencing the rate of plastic deformation.
The analysis of the effect of feed rate and depth of cut on the chip shape shows slightly different trends. As either parameter increases, the chip coloration becomes brighter—even at a constant cutting speed. This indicates that phase transformations can result from high mechanical loads alone. The inner surface of the chip exhibits deeper material folds with increasing feed and/or depth of cut, which can be explained by the larger undeformed chip thickness (from increased feed) and width (from increased depth of cut).
Under the most intensive cutting conditions—maximum cutting speed, feed rate, and depth of cut within the range of the experiments, the free edge of the chip takes on a nearly silver color similar to the base material (Fig. 15d). This suggests that phase transformations were induced by the combined effect of thermal and mechanical loads.
SEM micrographs of NiTi chips (Fig. 16) reveal a distinctive serrated morphology featuring regions of high plastic deformation, which are visible as adiabatic shear bands. The lamellae on the chip’s free surface are only a few micrometers wide. Adiabatic shear bands are a common failure mechanism in titanium and nickel-based superalloys under high strain rates during machining. These bands are usually very narrow, typically ranging from 5 to 50 µm, and consist of highly sheared material [22]. The outer chip face remains smooth.
Fig. 16
SEM images of chip morphology (test 6: V = 20 m/min, fz = 0.15 mm/tooth, ap = 1.0 mm)
This study investigated the effect of cutting conditions during dry face milling of Ni56.5Ti43.5 (wt.%) alloy on cutting forces, temperatures, surface roughness, microstructure, depth of the induced layer, tool wear, and chip morphology.
The axial cutting force component Fz was dominant in all tests, while the feed force Fx was the lowest. Cutting force components reached up to Fz = 2044.92 N, and maximum chip temperatures ranged from 230 to 340 °C. These critical thermomechanical loads on both the cutting tool and the machined surface contribute to the failure of cutting inserts and the degradation of surface integrity. Depending on the cutting conditions, the surface roughness Ra varied from 0.58 to 1.347 µm.
Multiple linear regression models were obtained for the dependence of the cutting forces, the chip temperature, and surface roughness on the cutting conditions. The feed rate predominantly influences the cutting force components Fy, Fz, and the equivalent cutting force F, while the depth of cut has a moderate effect. In the case of the component Fx, the predominant factor is the depth of cut. The cutting speed showed no significant effect on the cutting forces. A statistically significant correlation was observed between chip temperature and both the depth of cut and cutting speed, as well as between surface roughness and both the feed rate and the depth of cut.
For the first time, the formation of a martensitic layer consisting of deep martensitic needles was observed after dry face milling of NiTi. An inhomogeneous surface layer rich in martensite (up to 110 µm) and a deeper, inhomogeneous subsurface layer containing isolated, irregularly deep martensite needles (up to 205 µm) were identified. The surface layer forms from the simultaneous action of mechanical and thermal effects, while the subsurface layer results mainly from mechanical effects. Therefore, although the depth of the surface-induced upper layer can be influenced by applying various methods of intensive cooling to the cutting zone, this is unlikely to have an impact on the depth of the heterogeneous, stress-induced martensite lower layer. Subsequent technological operations must ensure the complete removal of the induced layer.
The cutting speed is the predominant influence on the depth of machining-induced layers, while feed rate and depth of cut have a lower impact. As the cutting speed increased from 20 to 50 m/min, the general depth of the affected layer decreased by up to 42.5%, due to reduced contact time between the cutting edge and workpiece, which contributed to reduced plastic deformation. The depth of the upper surface layer increased significantly by up to 57.1% with increasing cutting speed, attributed to higher temperatures that triggered phase transformations and deeper martensitic needle formation. Increasing feed rate from 0.05 to 0.15 mm/tooth and depth of cut from 0.5 to 1.0 mm led to moderate increases in both layers’ depths by 10 to 20%.
Rake face wear was dominant under all cutting conditions, with pinkish-purple color spots of varying sizes attributed to a thin film of cobalt salt from the reaction of the carbide binder cobalt with atmospheric oxygen. This was deposited on the cooler areas of the insert’s rake face during heating. Typical wear modes on the rake face included chipping of the cutting edge, flaking, notch wear, thermal cracks, and BUE. Both abrasion and adhesion contributed to wear. The cobalt layers are first removed from the insert surface, forming micro-roughnesses from protruding carbide grains. NiTi alloy components accumulate in the grooves. Oxygen diffusion and oxide layer formation occurred on the rake face.
NiTi chip morphology was strongly dependent on the face milling conditions. Higher cutting speeds produced brighter chips, indicating a rise in temperature-related material microstructure alteration. The gradual increase in cutting temperature with cutting speed contributes to material softening and results in a smoother inner chip surface. Increasing feed rate and/or the depth of cut also produces brighter chips. This shows that even at constant cutting speeds, high mechanical stresses can induce phase transformations. SEM micrographs revealed a serrated NiTi chip morphology with adiabatic shear bands—narrow, highly deformed regions—with lamellar width on the chip’s free surface of only a few micrometers.
Overall, this study improves the understanding of the NiTi SMA face milling process by identifying and quantifying the main challenges and ways to enhance milling performance, as well as ensuring the quality of finished products. In considering face milling as a pre-processing stage, it is important to ensure subsequent machining to remove the mechanically induced surface layer, as it can affect the functional properties of the final product. Further investigation of the machinability of NiTi alloys of other phase compositions (martensitic and mixed) will help optimize machining strategies and ensure the functional integrity of final components. Another important direction is the development of empirical models to predict the depth of the affected layer as a function of cutting parameters. A post-machining evaluation of the functional performance (e.g., shape memory effect or superelasticity), as well as direct phase analysis of the chips, are essential, given the sensitivity of NiTi to microstructural alterations.
Declarations
Conflict of interest
The authors declare no competing interests to disclose.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, 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://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Kaya E, Kaya İ (2019) A review on machining of NiTi shape memory alloys: the process and post process perspective. Int J Adv Manuf Technol 100:2045–2087. https://doi.org/10.1007/s00170-018-2818-8CrossRef
2.
Kaynak Y, Manchiraju S, Jawahir IS, Biermann D (2020) Chip formation and phase transformation in orthogonal machining of NiTi shape memory alloy: microstructure-based modelling and experimental validation. CIRP Ann 69(1):85–88. https://doi.org/10.1016/j.cirp.2020.04.025CrossRef
3.
Kaynak Y, Karaca H, Jawahir I (2015) Cutting speed dependent microstructure and transformation behavior of NiTi alloy in dry and cryogenic machining. J Mater Eng Perform 24(1):452–460. https://doi.org/10.1007/s11665-014-1247-6CrossRef
4.
Shizuka H, Sakai K, Yang H, Sonoda K, Nagare T, Kurebayashi Y, Hayakawa K (2020) Difficult cutting property of NiTi alloy and its mechanism. J Manuf Mater Process 4(4):124. https://doi.org/10.3390/jmmp4040124CrossRef
5.
Kitay O, Kaynak Y (2021) The effect of flood, high-pressure cooling, and CO2-assisted cryogenic machining on microhardness, microstructure, and X-ray diffraction patterns of NiTi shape memory alloy. J of Materi Eng and Perform 30:5799–5810. https://doi.org/10.1007/s11665-021-05854-6CrossRef
6.
Zhao YZ, Guo K, Sivalingam V et al (2021) Surface integrity evolution of machined NiTi shape memory alloys after turning process. Adv Manuf 9:446–456. https://doi.org/10.1007/s40436-020-00330-1CrossRef
7.
Kaynak Y, Karaca HE, Noebe RD (2015) The effect of active phase of the work material on machining performance of a NiTi shape memory alloy. Metall Mater Trans A 46:2625–2636. https://doi.org/10.1007/s11661-015-2828-1CrossRef
Kaynak Y, Karaca HE, Jawahir IS (2014) Surface integrity characteristics of NiTi shape memory alloys resulting from dry and cryogenic machining. Procedia CIRP 13:393–398. https://doi.org/10.1016/j.procir.2014.04.067CrossRef
10.
Yang H, Sakai K, Shizuka H, Kurebayashi Y, Hayakawa K, Nagare T (2021) Effect of cutting speed on shape recovery of work material in cutting process of super-elastic NiTi alloy. Int J Autom Technol 15(1):24–33. https://doi.org/10.20965/ijat.2021.p0024CrossRef
Balytska N, Tomashevskyi O, Shadura V (2025) Evaluation of cutting forces and temperatures in micro-milling of NiTi alloys using finite element modeling. Journal of Engineering Sciences (Ukraine) 12(1):A20–A27. https://doi.org/10.21272/jes.2025.12(1).a3CrossRef
Wang G, Liu Z, Huang W et al (2019) Influence of cutting parameters on surface roughness and strain hardening during milling NiTi shape memory alloy. Int J Adv Manuf Technol 102:2211–2221. https://doi.org/10.1007/s00170-019-03342-9CrossRef
18.
Altas E, Gokkaya H, Karatas MA, Ozkan D (2020) Analysis of surface roughness and flank wear using the Taguchi method in milling of NiTi shape memory alloy with uncoated tools. Coatings 10(12):1259. https://doi.org/10.3390/coatings10121259CrossRef
19.
Altas E, Altin Karatas M, Gokkaya H (2021) Surface integrity of NiTi shape memory alloy in milling with cryogenic heat treated cutting tools under different cutting conditions. J Mater Eng Perform 30(12):9426–9439. https://doi.org/10.1007/s11665-021-06095-3CrossRef
Kaya E, Kaya İ (2023) The effect of cutting tool and cutting speed on the surface integrity and functional properties in milling of NiTi shape memory alloys [NiTi şekil hafızalı alaşımların frezelenmesinde kesici takım ve kesme hızının yüzey bütünlüğü ve fonksiyonel özellikler üzerine etkisi]. Journal of the Faculty of Engineering and Architecture of Gazi University 38(4):2375–2384. https://doi.org/10.17341/gazimmfd.1095871
22.
Özel T, Ulutan D (2014) Effects of machining parameters and tool geometry on serrated chip formation, specific forces and energies in orthogonal cutting of nickel-based super alloy Inconel 100. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture 228(7):673–686. https://doi.org/10.1177/0954405413510291CrossRef
23.
Li B, Zhang S, Zhang J (2019) Plastic deformation and grain refinement in surface layer induced by thermo-mechanical loads for hard milling process. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture 233(10):2033–2046. https://doi.org/10.1177/0954405418815366CrossRef
Die im Laufe eines Jahres in der „adhäsion“ veröffentlichten Marktübersichten helfen Anwendern verschiedenster Branchen, sich einen gezielten Überblick über Lieferantenangebote zu verschaffen.
