A Comparative Study on the Wear Mechanisms of Uncoated and TiAlTaN-Coated Tools Used in Machining AMPCO^® Alloy

Abstract

A consistent evolution in materials developed for the industry and chip-start cutting processes has been acknowledged over the years. Cutting tool improvement through applying advanced coatings has proven very effective, enabling tool life (TL) extension while ensuring better surface quality. TiAlTaN coating enhances TL and surface quality in machining processes. However, only minimal research has been dedicated to comprehending the interaction between workpieces composed of Cu-Be and diamond tools. AMPCO^®, a Cu-Be alloy, plays a crucial role in moulding inserts, offering high wear resistance and contributing to extended mould longevity and improved productivity. The main objective of this work is to assess, identify, and quantify tool wear (TW) mechanisms evaluation while machining AMPCO^® with WC-Co uncoated tools and TiAlTaN-coated tools by physical vapour deposition (PVD). Evaluating tool performance while varying cutting length (L_cut) and feed rate (f) at three distinct levels and analysing the surface roughness (SR) produced in the machined surface were the primary purposes of this work. The results obtained with coated tools were distinct from those obtained with uncoated tools. While uncoated tools suffered from substrate abrasion and adhesion, the coated tools suffered mainly from delamination, followed by chipping. Furthermore, f and L_cut significantly influence the quality of the machined surface. TiAlTaN-coated tools performed significantly worse than uncoated tools, proving that the coating needs significant improvements to be considered as an alternative in milling Cu-Be alloys.

1. Introduction

Injection moulds are pivotal in manufacturing plastic injection moulding (PIM) and highly ductile metal components. Given the harsh conditions of elevated temperatures (T) and pressures during production [1], these moulds face significant wear and potential cracking challenges. The incorporation of Cu [2,3] and its alloys into the injection moulding process is driven by several key attributes encompassed in these materials: (1) high thermal conductivity (k), (2) corrosion and wear resistance, (3) dimensional stability, (4) electrical resistivity (ρ_R), and (5) substantial mechanical strength [4]. Beryllium (Be) [5] is recognised for its intrinsic hardness and granularly state substance. This innate attribute presents challenges in achieving polished and pristine surfaces due to the abrasive nature of the hard particles, which augments TW when machining injection moulding inserts of Cu-Be alloy by amplifying the costs and intricacy associated with manufacturing and repair/service procedures [6,7]. Incorporating Cu-Be alloys or AMPCO^® [8] variants into injection moulding tools can enhance the efficiency and quality of the moulding process [9,10], making them a valuable choice for high precision and durability, such as in the automotive and electronics sectors. Xu et al. [11] sought to investigate the wear mechanisms of the C17200 alloy [12] when in contact with GCr15 steel under dry conditions and a 3.5% NaCl solution. X-ray diffractometry (XRD), scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDX-SEMs), a three-dimensional (3D) profilometer, and X-ray photoelectron spectroscopy (XPS), were employed to assess wear damage. In dry wear conditions, the primary wear mechanisms observed for the C17200 alloy [12] were predominantly adhesive wear, oxidation wear, and delamination. Researchers and engineers continue to explore ways to optimise the use of these alloys to rapidly remove heat from plastic parts manufactured by the moulds, potentially reducing cycle times by 15 to 50% while maintaining good hardness and limiting the need for repair and refurbishment of the moulds, which improves the manufacturing process further. Ensuring the surface quality of these injection moulding materials is paramount for optimum performance. Conventional manufacturing (CM) prevails as the preferred method to fabricate injection moulds, guaranteeing and complying with the required dimensions and surface finish. Achieving the desired surface quality is crucial for injection moulds. A smooth surface finish is essential to produce high-quality plastic parts with minimal defects; likewise, geometrical and dimensional tolerances are vital for mass production. Not only Cu-Be alloys or AMPCO^® can be found in injection moulding, but also various other Cu alloys. Some of the most widely known and used Cu alloys are described in

Table 1. Summary of applications and characteristics of some Cu-based alloys.

Base Material	Subgrouped Material	Industry Applications	Characteristics
Cu	Bronze	Bearings and bushings [13].	Alloy of Cu and Sn, often including other elements like Al, P, or Si. Known for its mechanical strength and corrosion resistance [14,15,16].
	Brass	Plumbing and fittings [17].	Composed of Cu and Zn, brass is highly malleable and has a bright gold-like appearance [18,19].
	Cupronickel	Marine equipment [20], currencies.	Typically a Cu-Ni alloy with various compositions, it is used for its corrosion resistance, particularly in marine applications [21].
	Cu-Be	Aerospace components, springs and connectors, plastic moulds and injection moulding, valve seats.	Combines Cu with Be, known for its high mechanical strength, electrical conductivity, and fatigue resistance [12].
	Cu-Ni-Si	Electrical connectors.	Often used for electrical connectors and components, it is an alloy of Cu, Ni, and Si.
	Cu-Al	Electrical transmission.	An alloy of Cu and Al, used due to its electrical conductivity and corrosion resistance.
	Cu-Zn-Si	Electrical components.	An alloy containing Cu, Zn, and Si, used for applications that require high electrical conductivity.
	Cu-Mg	Marine propellers, high-strength components (gears, shafts).	Combines Cu and magnesium, offering high strength and resistance to corrosion.
	Cu-Ti	Heat exchangers, aerospace components.	Cu alloyed with Ti, used for its strength and corrosion resistance.
	Cu-Mn	Electrical applications.	Alloy of Cu and Mn, used for its mechanical and electrical properties.

, presenting some industrial applications and characteristics.

Cu-Be alloys amalgamate the attributes of Cu and Be to form a robust alloy exhibiting heightened wear resistance while maintaining k and ρ_R. This alloy integrates strength and resistance [9,22], showcasing elevated hardness and exceptional resistance to seizing and galling [23]. Due to these characteristics, Cu-Be alloy is an excellent choice for applications in bearings and bushings [23,24]. Its alloys find substantial demand within the aerospace sector, rendering them highly suitable for crafting tools and moulds in hazardous operational settings. A specific variant, the C17200 Cu-Be alloys, detailed in ASTM B 194-15, Standard Specification for Copper-Beryllium Alloy Plate, Sheet, Strip, and Rolled Bar [12], sees widespread use across diverse applications [8,25]. Regrettably, only minimal research has been dedicated to comprehending the interaction between workpieces composed of Cu-Be and diamond tools [26,27]. Nonetheless, the processing of these alloys, particularly during machining, can present considerable challenges, given the material adhesion propensity into tool surfaces, leading to speedy degradation of the tools. Zuo et al. [28] performed a single-factor experiment based on adhesive rate variation trend and crystal adhesive hypothesis to understand the adhesive effect on WC-Co tool damage, finding out that a high-rate adhesive environment effectively reduces the friction environment. Research efforts have been concentrated on mitigating these issues, exploring techniques such as EDM and using diamond tools. Dong et al. [29] proposed an efficient approach for machining microholes with high aspect ratios in C17200 beryllium copper alloy. This method combines micro-electrical discharge machining (EDM) using deionised water and in situ powder mixed electrical discharge machining (PMEDM) with boron carbide additives. The surface integrity of microholes with high aspect ratios can be enhanced through in situ PMEDM employing boron carbide mixed in deionised water. After in situ PMEDM, the arithmetic average of profile height deviation (R_a) decreased from 2.65 μm to 0.92 μm. However, alloys of this kind present distinct hurdles in processing due to their mechanical properties, particularly the notable high ductility, which can hinder machining operations. Consequently, numerous studies have delved into processing these alloys, specifically for applications in injection moulding, with a specific focus on EDM for Cu-based alloys. This method has proven to be highly effective in producing mould cavities [8]. Highly dependable engineered materials, particularly alloys, have found extensive application across various engineering domains [6]. Owing to the elevated hardness and fatigue strength at high T exhibited by Cu-Be alloys, conventional machining processes face issues such as TW, build-up edge (BUE), and grain lattice microtearing. AMPCO^®, the commercial designation for Cu-Be alloys, is extensively employed in plastic injection moulds owing to their elevated k and hardness, substantially reducing injection moulding cycles. Nonetheless, Be particles with high hardness present difficulties in achieving smooth surfaces when employing conventional machining techniques [30,31]. Cu and its alloys stand as pivotal materials across numerous industries, with their surface quality being a critical determinant of their performance; although showcasing distinctive properties, they introduce complexities in processing and manufacturing owing to their hardness and toxic nature. Ongoing research is essential to surmount these obstacles and enhance these materials’ effective adoption and long-term viability. In Boyer [32], Cu-Be alloys are addressed as the strongest known of the Cu alloy family. This high strength and good thermal conductivity make Cu-Be alloys (also known as Be-Bronze or Be-Cu) particularly suited for injection moulding applications [32]. Meng et al. [33] experimentally assessed two valves composed of distinct Cu-Be alloys, sharing remarkably similar compositions and subjected to identical heat treatment procedures. The valves were examined extensively using standard analytical methods, including metallography, XRD, chemical analysis, microhardness testing, and thermal conductivity measurements. Despite undergoing equivalent heat treatment processes, these alloys exhibited disparate thermal and mechanical characteristics due to their minor chemical composition differences. The alloy exhibiting a significantly longer lifespan, with an extension of 40%, showcased an enhanced thermal conductivity of k = 280.3 W (m·K)⁻¹, approximately 2× that of the other alloy. Lai et al. [34] carried out rotary bending fatigue (RBF) tests at different T (25 °C, 150 °C, 350 °C, and 450 °C). The tensile strength (σ_u) was determined, and associations between the applied bending stress magnitudes and the quantity of fatigue-induced fracture cycles were fitted to stress–life (S-N) curves, from which the respective equations were derived. The findings indicate that the fatigue resistance of C17200 alloy specimens under RBF conditions diminishes with escalating test T. In Ogawa et al.’s [35] investigation, the adaptability of a precipitation-hardened Cu-Be alloy to high-pressure gaseous hydrogen was assessed through a series of examinations, including slow strain rate testing (SSRT), fatigue-life analysis, fatigue crack growth (FCG) testing, and fracture toughness evaluations. The alloy showed no hydrogen-induced degradation and has large potential to be utilised for high-pressure hydrogen gas components.

The choice of milling AMPCO^® with TiAlTaN-coated [36,37] tools for the investigation stems from its paramount importance in the injection mould industry, making research in this domain precious for industry optimisation. In fact, AMPCO^® offers excellent surface finish and thermal conductivity, which are fundamental for mould cavities, allowing for better surface finishing of the injected parts and reduced injection cycle times. However, its machining aspect holds intrigue due to the dearth of information and scientific research about this alloy. Addressing this deficiency presents an essential and captivating challenge, offering potential contributions to this field’s engineering scientific community.

2. Materials and Methods

Given that the AMPCO^® alloy finds extensive use in injection mould production, it is imperative to develop new machining solutions to remain competitive. The primary objective focuses on evaluating the effect of L_cut of 26.8 m, 53.6 m, and 73.7 m and two levels of f, namely 750 mm/min and 1500 mm/min on the surface roughness (SR) of the AMPCOLOY^®83 machined workpiece while analysing the TL of cutters coated with TiAlTaN in comparison to the same uncoated tools. It is essential to scrutinise this coating and its influence on wear behaviour, given that it incorporates Ta, an element not commonly found in such coatings. Moreover, it is essential to qualitatively assess the primary wear mechanisms on the tool and quantify the corresponding flank wear (VB), analyse the influence of the chosen parameters, and understand which ones lead to improved results, both in terms of tool performance and the quality of the machined surface. This section will describe the used and tested materials and the methods employed throughout the experimental work. A detailed list of necessary equipment, materials, and procedures will be drawn as AMPCOLOY^®83 SR, TW, and TL are assessed for TiAlTaN-coated and WC-Co uncoated tools. Subsequently, all the results obtained will be documented, including identifying the main TW mechanisms, the performance of the tools in production quality, and their respective SR values.

2.1. Material to Be Machined

The stock material in this work was a Cu-Be alloy called AMPCOLOY^®83, supplied by AMPCO^® Metal Portugal, Ltd. (Porto, Portugal) in its forged state. It was supplied as a parallelepiped block, and the information regarding this material is summarised in

Table 2. Characteristics of the AMPCOLOY^®83 block.

Characteristics	Value	Units
Alloy	AMPCOLOY®83
Dimensions (L × W × H)	150 × 150 × 101.6	mm3
Mass		kg
State	Forged

. AMPCOLOY^®83 is a Cu-Be alloy with a significant element weight percentage (wt%) of Cu, as evident in

Table 3. Chemical composition of AMPCOLOY^®83.

Element	wt%
Cu	Bal.
Be	2.0
Cobalt + Nickel (Co + Ni)	0.5
Others	0.5 (max.)

Bal.—balance.

. This alloy is excellent for good ρ_R, k, wear resistance, and high hardness.

Table 4. Physical and mechanical properties of the provided AMPCOLOY^®83.

Mechanical Property	Value	Units
E	128	GPa
σu	1140	MPa
σy	1000	MPa
HV	376	HV
εu	5	%
ρ	8260	kg/m3

ε_u—strain at fracture.

provides this alloy’s specifications and physical and mechanical properties according to the supplier.

2.2. Cutting Tools

The cutting tools used in this work are made of tungsten carbide (WC, grade 6110) with an average grain size of 0.3 μm and a Co binder of wt% = 6%. The tools used for milling AMPCOLOY^®83 have dimensions shown in

Table 5. Dimensions of the milling cutters in mm (except for z).

Ref.	D1	D2	D3	L1	L2	L3	z
288.060.00	6.00	6.00	5.50	57	13	21	4

according to the scheme presented in Figure 1. Six uncoated WC-Co tools and six TiAlTaN-coated tools were used for the machining tests. A thicker and longer shank was intended to minimise vibrations during machining as much as possible. The milling tools were supplied by InovaTools, S.A. (Marinha Grande, Portugal). The tool reference used can be followed by “T0”, representing the WC-Co uncoated tools, or “T1” for the TiAlTaN-coated ones. It is then accompanied by “L”, “F”, and “S”, representing L_cut, f, and s, respectively.

It is also crucial to understand the geometry of the cutting tool and its constituent faces. As shown in Figure 2, three distinct faces are identified: the clearance face (CF), the rake face (RF), and the top face (TOP), following the ISO 3002-1:1982 [38] standard. TiAlTaN coating is used to protect the milling cutters’ surface in machining and was fabricated by the PVD (physical vapor deposition) technique, utilising a high-power impulse magnetron sputtering (HiPIMS) power source, a CemeCom CC800, in TeandM company (Coimbra, Portugal).

2.3. Milling Process

Machining tests were conducted on a computer numerical control (CNC) machine DMG/Mori Seiki, model DMU 60 eVo, Deckel Maho. The technical specifications of the CNC are detailed in

Table 6. Technical specifications of the CNC machine, DMG/Mori Seiki, model DMU 60 eVo, Deckel Maho.

Property	Value	Units
Max. rpm	18,000	rpm
Pmax	54.4	kW
X-axis max. range	600	mm
Y-axis max. range	500	mm
Z-axis max. range	500	mm
No. of axes	5

P_max—nominal power.

. Regarding the experimental plan, there was a need to establish machining parameters beforehand, with only a few remaining constant across different tests. These tests were conducted to observe the tools’ behaviour and assess the TW mechanisms. The various setups are explained in

Table 7. Experimental plan with milling parameters used in the setups.

Tool Reference	Coated	s (rpm)	f (mm/min)	Lcut (m)	Cooling
T0L27F750S126	No	126	750	26.8	Yes
T0L54F750S126			750	53.6
T0L74F750S126			750	73.7
T0L27F1500S126			1500	26.8
T0L54F1500S126			1500	53.6
T0L74F1500S126			1500	73.7
T1L27F750S126	Yes		750	26.8
T1L54F750S126			750	53.6
T1L74F750S126			750	73.7
T1L27F1500S126			1500	26.8
T1L54F1500S126			1500	53.6
T1L74F1500S126			1500	73.7

. The radial depth of cut (a_e or RDOC) was defined considering 60% of the tool’s diameter (Ø) and the axial depth of cut (a_p, or ADOC) was defined to be 0.5 mm, as finishing milling operations were the goal to simulate.

2.4. SR Testing

The SR analysis relates to machined surface wear analysis and is pivotal during this project’s development. SR measurements, such as the maximum height of the profile (R_z, which is the average of the maximum heights between five consecutive peaks) and the total height of the profile (R_t, height between the deepest valley and the highest peak on the evaluation length) were taken from the milling with six uncoated tools and six coated tools and performed according to ISO 21920-2:2021 [40] standard. Each test used a cut-off (λ_c) of 0.8 mm as the base, with a length equivalent to seven cut-off segments, amounting to 5.6 mm [41]. Due to the probe’s inertia and considering that the probe’s acceleration and deceleration can introduce errors in roughness measurements, the first and last segments are not considered, eliminating potential measurement errors. Machining process stability was also a crucial factor to consider. Consequently, the SR of the machined workpiece was measured in the transverse and the longitudinal direction, concerning the machining direction, immediately after each machining trial.

2.5. Sample Preparation for SEM Analysis

SEM was used to analyse the predominant TW mechanisms in each cutting tool and the chemical composition of the uncoated and coated tool surfaces (t). For this purpose, an FAE Quanta 400FEG scanning electron microscope equipped with an EDAX Energy-Dispersive X-ray spectroscopy system was used. A cutting tool was selected to be cut in a cross-sectional sample that facilitated the analysis of the coating’s composition and t. Cutting [42], mounting in an epoxy medium [43], sanding (decreasing grit size of sandpaper: 220, 500, 800 and at last 1200, as for ISO 6344-2:2021(E) [44] and ISO 6344-3:2021(E) [45]), polishing [46], and cleaning [47] were the processes taken to have a SEM sample to analyse.

3. Results

This section will present the obtained results during the experimental procedure into three subsections: Section 3.1, Coating Characterisation; Section 3.2, Machined Surface; and Section 3.3, Tool Wear Assessment.

3.1. Coating Characterisation

This first subsection evaluates the produced coating’s morphology, t, and chemical composition. The coating was analysed using SEM (Figure 3a) and energy-dispersive X-ray spectroscopy (EDS, Figure 3b,c), provided by the Centre for Materials of the University of Porto (CEMUP, Porto), to confirm its chemical composition. A beam with an electric potential (U) of 15 kV, occasionally reduced to 10 kV, was employed to minimise interaction and reduce noise levels in the spectra. It is important to note that this approach may have some limitations concerning precision for quantitative analyses. However, it was deemed suitable for confirming the chemical composition of the coatings under examination, thus avoiding using more expensive techniques.

According to the chemical analysis conducted on the top layer of the coating, it is evident that this layer, zone 2, is rich in Ti, Al, Ta, and N, aligning with the expected composition of a TiAlTaN coating, and zone 1 corresponds to the tool substrate (WC-Co). The TiAlTaN coating t measurement was obtained by averaging several measurements, using SEM, taken in different zones of the coating. An average value of t = 3.369 μm can be assumed, with a relatively insignificant standard deviation (SD) of ±0.045 μm. This t is suitable for milling tools and is not critical in compromising adhesion; a point criticality generally begins at approximately t = 5 μm for PVD coatings. This t allows for some abrasion of the coating without exposing the substrate and losing the properties exhibited by the coating during machining. It is also worth mentioning that the t used in this Cu-Be alloy machining study aligns with the total t used in Huang et al.’s [48] work.

3.2. Machined Surface Assessment

3.2.1. SR Obtained Results

This second subsection addresses the machined surface section, the TW values measured, and the analysis of the corresponding SR under various cutting conditions and cutter typology will be presented. For the analysis of the machined surface, a profilometer was used to measure transverse and longitudinal roughness values after each test, enabling subsequent analysis.

Table 8. Measurements of transverse SR.

Tools Reference	Coated	Ra (μm)	Rz (μm)	Rt (μm)
T0L27F750S126	No	0.370 ± 0.047	1.992 ± 0.199	3.542 ± 0.600
T0L54F750S126		0.428 ± 0.016	1.992 ± 0.178	3.576 ± 0.537
T0L74F750S126		0.470 ± 0.022	2.228 ± 0.109	3.654 ± 0.552
T0L27F1500S126		0.747 ± 0.031	3.410 ± 0.487	6.194 ± 0.893
T0L54F1500S126		0.780 ± 0.092	3.468 ± 0.249	6.194 ± 0.372
T0L74F1500S126		0.812 ± 0.040	4.152 ± 0.479	7.146 ± 0.656
T1L27F750S126	Yes	0.453 ± 0.035	2.264 ± 0.308	3.108 ± 0.692
T1L54F750S126		0.484 ± 0.019	2.334 ± 0.324	3.992 ± 1.059
T1L74F750S126		0.584 ± 0.055	3.120 ± 0.164	4.192 ± 0.557
T1L27F1500S126		0.816 ± 0.063	4.136 ± 0.267	5.054 ± 0.772
T1L54F1500S126		0.940 ± 0.036	4.834 ± 0.438	6.872 ± 1.057
T1L74F1500S126		1.092 ± 0.084	5.256 ± 0.362	7.078 ± 0.585

and

Table 9. Measurements of longitudinal SR.

Tools Reference	Coated	Ra (μm)	Rz (μm)	Rt (μm)
T0L27F750S126	No	0.248 ± 0.035	1.502 ± 0.205	1.660 ± 0.196
T0L54F750S126		0.266 ± 0.055	1.614 ± 0.262	2.088 ± 0.256
T0L74F750S126		0.360 ± 0.017	1.926 ± 0.662	2.090 ± 0.931
T0L27F1500S126		0.363 ± 0.016	2.066 ± 1.099	2.440 ± 1.268
T0L54F1500S126		0.384 ± 0.050	2.192 ± 0.278	2.490 ± 0.240
T0L74F1500S126		0.432 ± 0.060	2.654 ± 0.452	2.962 ± 0.553
T1L27F750S126	Yes	0.175 ± 0.050	1.404 ± 0.582	1.574 ± 1.127
T1L54F750S126		0.334 ± 0.012	1.790 ± 0.305	1.938 ± 0.295
T1L74F750S126		0.496 ± 0.039	2.540 ± 0.709	3.230 ± 0.728
T1L27F1500S126		0.799 ± 0.079	4.488 ± 0.477	4.710 ± 0.674
T1L54F1500S126		1.054 ± 0.027	4.532 ± 0.802	4.742 ± 1.753
T1L74F1500S126		1.347 ± 0.097	4.784 ± 1.522	6.060 ± 1.555

’s tool references for TiAlTaN-coated and WC-Co uncoated tools correspond to three tools used for each setup. R_a, R_z, and R_t presented are the mean values and SD of three SR measurements in each tool’s longitudinal and transverse directions. The difference between R_a against R_z and R_t is the presence of more pronounced roughness peaks, while the rest of the surface is relatively regular with much lower peaks.

3.2.2. R_a Analysis on L_cut Influence

Discrepancies between the R_a values of the respective TiAlTaN-coated and WC-Co uncoated tools are displayed in the graphs from Figure 4, which resumes the data for R_a as a function of L_cut for the transverse direction, and Figure 5, which demonstrates the longitudinal direction results to better visualise the final and aggregated results. A tendency of L_cut to increase leads to an SR increase for WC-Co and TiAlTaN-coated tools in the transverse and longitudinal directions. A four-series organisation is provided: the yellow and orange series are for WC-Co tools and TiAlTaN-coated tools at f = 750 mm/min, and the red and pink series are for WC-Co tools and TiAlTaN-coated tools at f = 1500 mm/min, respectively. Regarding R_a in the transverse direction, WC-Co uncoated tools consistently result in lower values than the TiAlTaN-coated ones (Figure 4). An increase in R_a was observed as L_cut increased. The rise is attributed to the increased TW caused by the longer L_cut. The greater the wear, the higher the SR, compromising the quality of the machined surface. The maximum R_a value is obtained for the TiAlTaN-coated tool that managed L_cut = 73.7 m at f = 750 mm/min. It is noticeable that L_cut has a more significant influence on TiAlTaN-coated tools. It is observable that R_a values are also higher in the transverse direction (Figure 4), as shown in Table 8 and Table 9. This phenomenon is easily understood because the profilometer is perpendicular to the machining direction, thus encountering significant irregularities (or burrs) on the surface while travelling between tool paths. Also, a better performance from the TiAlTaN-coated tools for L_cut = 27.8 m can preview a much better behaviour for lower L_cut or an outlier. Anyhow, a tendency for TiAlTaN-coated tools to have a greater TW rate than WC-Co uncoated tools.

Similar to what was observed for f = 750 mm/min, albeit with a more negligible difference, the SR values increase with the increase in the L_cut. Lower SR values continue to be observed for WC-Co uncoated cutters compared to coated ones, indicating that for this tool substrate, the coating has not been advantageous thus far. Regarding the obtained longitudinal R_a values for f = 1500 m/min condition for TiAlTaN-coated tools; a more severe TW can be assumed upon the measured values, relatively comparable to the longitudinal condition of f = 750 mm/min. Additionally, the experimental tendency of having lower R_a values for shorter L_cut values is verified for WC-Co and TiAlTaN-coated tools. The coating degradation was more pronounced for L_cut = 73.7 m, resulting in poorer machined surface quality. From Figure 4, the best performance was achieved for WC-Co tools at f = 750 mm/min for all considered L_cut. The overall performance for both types of tools is better for shorter L_cut. SR increases with the increase in f and L_cut; however, L_cut influences more TW and SR than f.

From Figure 5, the deterioration of the coating becomes notably accentuated when L_cut = 73.7 m, resulting in deterioration of the machined surface’s quality. Inadequate adhesive strength, as indicated by L_C2 < 22 N, triggers the delamination of the coating due to a cohesive flaw [49]. This detachment renders tool coatings susceptible to the effects of three-body abrasion, ultimately diminishing both tool performance and the quality of the machined surface. This elucidates the significantly elevated R_a values observed for f = 1500 mm/min.

3.2.3. Ra Analysis on f Influence

Figure 6 and Figure 7 illustrate a comparative graph of R_a obtained with WC-Co tools as a function of f for the transverse and longitudinal directions.

It can be observed that in both the transverse and longitudinal directions, as L_cut increased, the quality of the machined surface decreased. Also, the transverse values remained higher than the longitudinal ones for the WC-Co tools. The SR values for the f = 750 mm/min machining condition are significantly lower than those for the f = 1500 mm/min machining condition for transversal and longitudinal directions. A curious tendency of higher R_a values is seen in the longitudinal direction for f = 1500 mm, compared to the transverse direction. For higher f, the tool coating deteriorates more, causing irregularities on its surface due to more severe TW mechanisms. Consequently, the SR values will be higher, with an uneven end mill geometry due to the already-mentioned three-body abrasion phenomenon. Thus, it can be concluded that the higher the f, the greater the TW exhibited by the milling–cutting tools, with the WC-Co uncoated ones being a better option, as they showed lower R_a values than the TiAlTaN-coated ones. Figure 7 presents the R_a values about f and tool typology for the longitudinal direction.

3.3. TW Assessment

After the stipulated experimental plan concluded, the tools were subjected to SEM analysis to assess TW and identify the main wear mechanisms, as addressed in Pedroso et al. [50]. Various tool faces were examined, including CF, RF, and TOP. TW was quantified both quantitatively and qualitatively. ISO 8688-2:1989 [51] standard, Annex C, was considered to analyse all TW phenomena, and through this analysis, to understand which milling parameter has the most significant influence on the type of wear generated. The TW mechanism section will analyse the exact mechanisms and the obtained VB values in the cutting tools. Since 36 tools were used, SEM image analysis will be performed by sampling.

3.3.1. TW SEM Image Analysis

In this subsection, a detailed characterisation of all TW mechanisms of each cutting tool is performed for the respective test conditions. It should be noted that throughout the analysis of the images, some “stains” related to the presence of the base material adhered to the tool will appear; moreover, in some instances, these areas represent only contamination or dirt and should be ignored. When contaminated areas appear in the images, they are not identified as any wear mechanism. Images were taken using SEM for at least two teeth of each cutting tool, ensuring that the TW mechanisms were consistent throughout the tool. For TW analysis, the cutting tools were grouped by testing conditions, disregarding whether they were WC-Co uncoated or TiAlTaN-coated tools. For instance, tools T0L27F750S126 and T1L27F750S126 (see reference in Table 8) were subjected to f = 750 mm/min and L_cut = 26.8 m, and this will be the first pair under analysis. All other pairs will be analysed for the same test parameters.

T0L27F750S126/T1L27F750S126 Tools

T0L27F750S126 and T1L27F750S126 tools were tested for L_cut = 27 m and f = 750 mm/min. In the case of T0L27F750S126 (WC-Co uncoated), various types of TW are visible, abrasion being predominant; moreover, there is also evidence of material adhesion in different parts of the cutting tool (Figure 8a,c), and occasionally, substrate detachment, chipping, and microchipping (Figure 8b). This chipping may have been caused primarily by material adhesion to the surface.

An EDS analysis was conducted in the adhesion area (Figure 8c) to verify whether the observed stains were indicative of this wear mechanism (adhesion) or were related to the tool’s image or surface defects. As Figure 9a shows, two distinct zones are visible: Z1 corresponds to the machined material, and Z2 corresponds to the base material of the tool. This conclusion can be extracted since in Figure 9b, the highest peak corresponds to Cu, indicating AMPCOLOY^®83; in Figure 9c, the highest peak corresponds to W, the substrate material of the cutting tool.

Regarding tool T1L27F750S126, it can be stated that the main wear mechanism is delamination, which is easily visible on all faces of the mill. Most likely due to this factor, there are various areas of substrate detachment (marked as chipping), either in more significant portions (Figure 10c) or in small amounts of material (Figure 10d), tooth (4). Abrasion is also evident on the faces of the tool, particularly visible on the CF.

After the initial discussion regarding the tools under the test conditions of L_cut = 26.8 m and f = 750 mm/min, one uncoated and one coated tools, it is possible to provide a final summary based on the previous analysis and Figure 11. It is evident from Figure 11 that the tool that suffered more severe wear was the coated one. Although in the case of the uncoated tool, there is a significant amount of abrasion and adhered material, the coated tool exhibits more severe wear, including the detachment of a large portion of material at the blade’s vertex, both from the coating and the substrate, driven by the cracking observed in Figure 11d. Additionally, delamination of the coating is likely to influence the degradation of the cutters. In fact, with the detachment of the coating, harsh particles are released, which, upon contact, facilitate easier abrasion of the tool, leading to the chipping mechanism of substrate wear.

T0L54F750S126/T1L54F750S126 Tools

Tools T0L54F750S126 and T1L54F750S126 were tested for L_cut = 54 m and f = 750 mm/min, with the first being uncoated and the second coated.

In the case of the T0L54F750S126 tool, the same wear mechanisms were observed once again: abrasion on the exit face and adhesion of machined material. Three different teeth were analysed, as shown in Figure 12. For the CF, images were taken of blades 1 and 2. On the RF, teeth 2 and 3 were analysed (Figure 12c,d), revealing a slight substrate detachment (chipping) on tooth number 2 and a more pronounced detachment on tooth number 3. Areas with abrasion and adhered material were also identified. In comparison with the T0L54F750S126 tool, the T1L54F750S126 tool exhibits more severe wear, predominantly with coating delamination. Numerous areas with adhesion of machined material are visible, occurring more frequently in this coated tool than in the uncoated one (Figure 13). In the Figure 13 case, taking a broader analysis of the second pair of tools under investigation, it can be inferred that the uncoated tool experienced more wear than the previous parameters (L_cut = 54 m and f = 750 mm/min). However, it proved slightly more effective than the coated tool under the same conditions.

Although abrasive wear is present, the predominant TW mechanism for these cutting conditions was delamination and chipping for both tools. The portion of material detached was approximately in the same region and with the same geometry for both. It is important to note that the dark spots visible in the images (particularly in Figure 14) are to be disregarded as they are not an integral part of a wear mechanism but rather contamination or dirt.

T0L74F750S126/T1L74F750S126 Tools

Tools T0L74F750S126 and T1L74F750S126 were tested for L_cut = 74 m and f = 750 mm/min, with the former being uncoated and the latter coated. Tool T0L74F750S126 was one of the ones that showed the most VB among all the coated ones due to the significant portion of surface detachment, as can be seen in Figure 15a. However, abrasive wear and chipping were also observed to be persistent mechanisms. Similar to what happened with the previous tools, the coated tool under these machining conditions is also more worn. In Figure 16c, a significant material detachment can be observed, possibly originating from the cracking observed in the same image. Along with delamination, chipping was also the most common mechanism in this tool.

Up to this point, the least worn milling cutters are depicted in Figure 17. The coated one followed the earlier pattern, showing higher degradation than the uncoated one. In the uncoated tool, slight zones of abrasion and adhered material can be observed, whereas in the coated tool, delamination and mild abrasion prevail once again.

T0L27F1500S126/T1L27F1500S126 Tools

The milling cutters T0L27F1500S126 and T1L27F1500S126 were tested for a L_cut = 27 m and a f = 1500 mm/min, with the former being uncoated and the latter coated. Regarding the tool T0L27F1500S126, as shown in Figure 18, the main wear mechanisms are abrasion of the tool’s exit face and chipping. This tool has a significant amount of material detachment at the vertices of each blade. Adhesion of machined material is also observed in this milling cutter.

An EDS analysis was performed on the surface to dispel any doubt regarding Figure 18b to identify the material in the darker region and confirm whether it was material adhesion, contamination marks, or surface defects. As Figure 19b clearly shows, there is a high wt% of Cu in the darker region, confirming that it is material from the AMPCOLOY^®83 wrought-stock, and therefore, adhesion. Figure 20 is a SEM image of the T0L27F1500S126 tool and Figure 21a–c are EDS spectra of the three different zones of Figure 20.

Upon analysing Figure 22, various wear mechanisms can be observed in tool T1L27F1500S126. Once again, identical to what was observed previously, delamination is the predominant mechanism for this coated tool, along with chipping, adhesion, and abrasion. Once again, to clarify the meaning of the colour differences in the obtained SEM image (Figure 22b), a surface chemical composition analysis was performed. As can be observed in Figure 21a, there are three distinct zones: a lighter one (Z1), an intermediate one (Z2), and a darker one (Z3). Zone 1 corresponds to the tool substrate (WC-Co), zone 2 to the machined material (AMPCO^®), and finally, zone 3 to the coating of the cutter (TiAlTaN).

As Figure 23 shows, there is a significant detachment of the base material for the uncoated tool. On the other hand, there was an initial abrasion and subsequent delamination of the substrate for the coated tool, with no material detachment from the tool’s surface, only coating deterioration. Contrary to the milling cutters tested with f = 750 mm/min, the tool in worse condition is the uncoated one, in this case. A significant chipping zone can be observed, where a large portion of the tool substrate was detached without prior cracking. This phenomenon could be due to a defect in both the tool and the machined material, where machining a groove or cavity might have caused increased vibration at the tool’s edge, resulting in detachment. This factor could also be associated with the coating, which influences the initial stage of the test and, after detachment or wear, allows the substrate to undergo chipping, suggesting that the duration of the coating effect may be too short-lived, ceasing to have an effect after a relatively short L_cut. In the coated tool, delamination is again noticeable across the edge of the tool’s exit face.

T0L54F1500S126/T1L54F1500S126 Tools

Tools T0L54F1500S126 and T1L54F1500S126 were tested for L_cut = 54 m and f = 1500 mm/min, with the former being uncoated and the latter coated. Close to what was observed in the previous uncoated cutting tool for the same L_cut (T0L54F750S126), in tool T0L54F1500S126, there is a significant detachment of material from the edge of one of the teeth (Figure 24a). Consequently, the predominant wear mechanism is chipping.

Regarding tool T1L54F1500S126, as shown in Figure 25, the delamination of the substrate is visible. Additionally, in Figure 25c, it is evident that cracking occurred in various areas, which could have been a significant cause of chipping. Adhesion and abrasion are also present. In terms of wear, coated tools display a more aggressive level of cracking. In Figure 26d, it is possible to see that cracking was indeed the cause of the substrate detachment from the milling cutter. The uncoated milling cutter shows minor areas of abrasive wear and a small chipping zone, although much smaller when compared to the coated milling cutter.

T0L74F1500S126/T1L74F1500S126 Tools

Milling cutters T0L74F1500S126 and T1L74F1500S126 were tested for L_cut = 74 m and f = 1500 mm/min. The uncoated tool was primarily subjected to abrasion and adhesion on the exit face, as evident in Figure 27. Additionally, adhesive wear and cracking are observed in smaller amounts, with the latter likely being the cause of material detachment.

The T1L74F1500S126 tool exhibited one of the worst wear behaviours, showcasing nearly all types of wear, each in significant amounts. Figure 28 shows various TW mechanisms on the CF and RF (abrasive wear, adhesive wear, chipping, delamination, and cracking). Finally, in the last pair of tools (Figure 29), referring to the uncoated tool, it seems that they experienced less apparent wear under what would theoretically be the worst machining conditions. The present wear mechanisms are abrasion, adhesion, and minor zones of cracking and chipping that could be attributed to the longer L_cut. At some point, the tool may have significantly increased the contact area, stabilising the geometry due to low contact pressure. Although the difference is not substantial in the coated tool, it appears more worn, with significant delamination of the coating along the edge and abrasion of the substrate.

3.3.2. TW VB Analysis

This section will present and analyse VB measured on the CF for TiAlTaN-coated and WC-Co uncoated tools, following the guidelines set by the ISO 8688-2:1989 [51] standard. The measurements were accomplished using the ImageJ 1.54 (a program developed at the National Institutes of Health and the Laboratory for Optical and Computational Instrumentation (LOCI, University of Wisconsin)). The presentation of VB values will be compartmented by the different f values used in milling. The measured VB values are presented in

Table 10. VB average values obtained for TiAlTaN-coated and WC-Co uncoated tools at a f = 750 mm/min.

Tool Reference	Coated	VB (μm)
T0L27F750S126	No	10.674 ± 0.984
T0L54F750S126		14.244 ± 0.085
T0L74F750S126		122.54 ± 0.974
T1L27F750S126	Yes	68.612 ± 1.820
T1L54F750S126		98.876 ± 1.295
T1L74F750S126		132.81 ± 1.685

, corresponding to the mean of five consecutive measurements, with the respective associated standard deviation.

The VB values for f = 1500 mm/min are presented in

Table 11. VB average values obtained for TiAlTaN-coated and WC-Co uncoated tools at a f = 1500 mm/min.

Tool Reference	Coated	VB (μm)
T0L27F1500S126	No	59.792 ± 1.830
T0L54F1500S126		103.784 ± 3.474
T0L74F1500S126		141.958 ± 2.779
T1L27F1500S126	Yes	82.802 ± 2.297
T1L54F1500S126		121.252 ± 0.965
T1L74F1500S126		145.814 ± 0.914

, calculated by averaging readings and obtaining the corresponding mean and SD.

Figure 30 graph shows the evolution of the VB for the milling cutters tested at f = 750 mm/min as a function of L_cut for each test condition. As is depicted, the TiAlTaN-coated tool exhibited a significantly higher VB compared to the WC-Co uncoated tool, with a notable discrepancy in the values for the first two noncoated tools (corresponding to T0L27F750S126 and T0L54F750S126, first and third bars, respectively). Hence, it can be concluded that VB values are meagre for L_cut of 26.8 m and 53.6 m since the tools deal with less demanding machining parameters, both f and L_cut. Regarding the coated tools, as L_cut increases, the expected trend in VB between L_cut values is observed, with an approximately equal increase. For conditions of f = 750 mm/min and L_cut = 73.7 m, for both TiAlTaN-coated and WC-Co uncoated tools, VB is at its maximum, slightly higher in the TiAlTaN-coated one. For the setups of f = 1500 mm/min, an increase in L_cut leads to an increase in VB, consistently showing higher values for TiAlTaN-coated tools and superior performance from WC-Co uncoated tools. Thus, the trend of increasing VB with increased L_cut is consistent.

Thus, it can be concluded that from the L_cut = 26.8 m to L_cut = 53.6 m conditions, there was no significant increase in wear for WC-Co uncoated tools. Nonetheless, there was a sharp increase for the L_cut = 73.7 m condition, which occurs thanks to the point of total failure reached as the cutter travels a longer L_cut. The tool becomes increasingly worn until it breaks. Among the three analysed conditions, the WC-Co uncoated tool with a L_cut = 73.7 m has a longer machining time, resulting in a large portion of substrate being detached, justifying the high VB value. In the case of TiAlTaN-coated tools, the values of VB were consistently higher than the ones recorded from WC-Co uncoated tools, glimpsing that the coating is contributing none. Chipping and delamination are the wear mechanisms that show the most coating detachment from the substrate. Therefore, it would be expected that tools exhibiting a higher presence of these wear mechanisms would also show higher VB values, as confirmed by Figure 30. As seen in Section 3.3.1, tools with the references T1L54F750S126, T1L54F1500S126, T0L74F750S126, T1L74F750S126, T0L74F1500S126, and T1L74F1500S126 had the highest area of detached coating and damaged substrate, aligning with the previously established theoretical framework.

4. Discussion

This section will discuss the results, divided into three main subsections: Section 4.1, Comparison of the Machined Surface Condition; Section 4.2, Comparison with the Presented Wear; and Section 4.3, Comparison with Other Used Coatings. Thus, it is a conclusion on which tools exhibited less and more TW, the factors influencing each of these, and consequently, those that produced a better surface quality of the machined workpiece.

4.1. Comparison of the Machined Surface Condition

4.1.1. Analysing f‘s Influence

In this subsection, it will be discussed the results of R_a concerning the significant influence of f and L_cut on the final quality of the machined surface [52]. As is evident from Figure 6 and Figure 7, all values are presented in a single graph, grouped according to f, throughout the different L_cut values used and tools typologies. It is also noticeable in the respective legend that they are sorted based on the presence of coating, or lack thereof, and within this, they are sorted in ascending order of L_cut. It can immediately be reckoned that SR increased as f increased for each of the subgroups of TiAlTaN-coated tools (the final three bars in each set) and WC-Co uncoated tools (the initial three bars in each set). This trend is observed in all conditions and directions (transverse and longitudinal). In detail, starting in Figure 6 and Figure 7, by comparing WC-Co tools with TiAlTaN-coated ones, for f = 750 mm/min, the machining performance is very similar: TiAlTaN-coated tools somewhat have the upper hand when machining in shorter L_cut values once the coating protects the tool substrate and enhances the AMPCO^® cutting; WC-Co uncoated tools experienced much less TW, with abrasion being the predominant wear mechanism. The TW damage was far more severe to TiAlTaN-coated tools, having undergone more significant geometry changes due to substrate detachment and delamination. The TiAlTaN-coated tools naturally exhibited slightly higher SR values at higher f. The inadequate adhesion of the coating, likely due to the deposition parameters used and lack of an adhesion interlayer, compromised its performance. The detached particles act as third bodies after delamination, significantly increasing abrasion, as also observed in Silva et al. [53] and Sousa et al.’s [54] work. Moreover, the steps left by the detachments help in favouring adhesion phenomenon, with the deposit of AMPCO^® in the borders of the detachments. Furthermore, the detachment of the coating “weakens” the substrate in that area through micromodifications, which contributes to the faster deterioration of the substrate. For both tool typologies and the longitudinal and transverse roughness, f = 1500 mm/min resulted in significantly higher roughness values than f = 750 mm/min. Therefore, it can be concluded that the f parameter plays a significant role in the machined surface quality [52,55], as it is evident that higher f corresponds to higher SR values. This phenomenon may have been influenced by the significant abrasive wear on the WC-Co uncoated tools and delamination and chipping on the TiAlTaN-coated ones. These TW mechanisms alter the geometry of the cutter, thereby influencing the quality of the machined surface. However, despite this situation being the most commonly found one in the literature, Martinho et al. [55] reported that the behaviour of coated tools when machining duplex stainless steels was not very sensitive to cutting parameters, within the range used in that work.

Overall, it can be said that the quality of the machined surface was satisfactory, with reasonable results in terms of the R_a; although, it can be observed that doubling the f leads to an almost twofold increase in R_a values. Additionally, the recorded values for TiAlTaN-coated tools were consistently higher than those for WC-Co uncoated tools, meaning that the coating was not the most suitable.

4.1.2. Analysing L_cut’s Influence

Regarding L_cut, the trend observed in the f parameter remains consistent, although the intervals are not as wide. It is evident from Figure 4 and Figure 5 that as L_cut increases, both in the transverse and longitudinal directions, R_a also increases [56]. However, this difference is not as significant as observed previously with f, concluding that f is the most influential factor affecting the machined surface’s quality in this experimental procedure. For the same L_cut, in the case of the longitudinal direction and f = 1500 mm/min, it is evident that there is a much more significant discrepancy between TiAlTaN-coated and WC-Co uncoated. According to a study conducted by Wang et al. [57], the SR of a machined alloy was analysed, and it was also concluded that R_a increased with an increase in f. It is also observed that, in general, the values in the transverse direction are much higher than in the longitudinal direction, which in turn can be easily explained by the fact that during transverse measurement, the roughness gauge may have been placed between concavities or protrusions, being perpendicular to these irregularities and naturally measuring higher roughness in this direction (striae).

4.2. Comparison with the Presented Wear

4.2.1. TW Mechanisms

The main wear mechanism observed in most cutting tools was abrasion.

Table 12. Summary on the quantification of the TW visualised.

Tool Reference	Coated	Abrasion	Adhesion	Chipping	Microchipping	Delamination	Cracking
T0L27F750S126	No	5	4	3	4	1	1
T0L54F750S126		5	4	3	3	1	3
T0L74F750S126		5	3	3	1	1	1
T0L27F1500S126		4	3	3	1	1	1
T0L54F1500S126		5	2	4	3	1	1
T0L74F1500S126		5	2	3	1	1	3
T1L27F750S126	Yes	3	3	5	1	5	4
T1L54F750S126		3	3	4	1	5	1
T1L74F750S126		4	3	5	1	5	4
T1L27F1500S126		4	3	3	1	5	1
T1L54F1500S126		3	3	5	1	5	5
T1L74F1500S126		4	3	3	1	5	1

was created to provide an easier visualisation of the predominant TW mechanisms in each test condition. These mechanisms are labelled from 1 to 5 to quantify the visual qualification on the TW, with 1—none; 2—rare; 3—low; 4—moderate; 5—severe.

In the case of WC-Co uncoated tools, which experienced less TW, the main observable mechanism was substrate abrasion, which was already predominant among cutting tools [58]. In the case of coated tools, the precise main mechanism was delamination, followed by chipping. It is important to note that despite not being one of the predominant wear mechanisms in coated tools, adhesion was higher in these tools than in WC-Co uncoated ones. This observation may indicate that the introduction of Ta into the composition of the TiAlN coating leads to a greater metallurgical affinity between the coating and the material to be machined, in the specific case of the AMPCO alloy. Machining using TiAlN-based coatings carried out by PVD is dissected by Sousa et al. [37]. It is logical that the higher the wear of the tools, the worse the quality of the machined surface. For instance, the coated tools subjected to f = 1500 mm/min exhibited the most significant TW. Nearly all the tools exhibit all the wear mechanisms present throughout this procedure, although in small quantities. WC-Co uncoated tools do not show this wear mechanism regarding delamination, as it involves coating detachment. Cracking is the least observed wear mechanism and is not present in T0L27F750S126, T0L74F750S126, T0L27F1500S126, T0L54F1500S126, T1L54F750S126, T1L27F1500S126, or T1L74F1500S126 (see Table 8 for reference reminder), that is, in more than half of the conditions used. However, it is usually strongly represented when it is present, as in the case of tools T1L54F1500S126 and T1L74F750S126 (see Table 8 for reference reminder). Chipping and adhesion are present in all tools since adhesion is likely responsible for coating detachment, as the presence of adhered material tends to generate severe wear, compromising machining process performance and reducing surface quality [57]. Abrasion is the predominant wear mechanism. These wear mechanisms are common in machining operations like this.

4.2.2. VB Assessment

In Figure 30, the VB values are represented as a function of the test conditions. The chart is organised with WC-Co uncoated tools first, followed by TiAlTaN-coated tools, in ascending order of L_cut. It is evident, for the same f, that VB increases proportionally with the L_cut in the case of TiAlTaN-coated tools, demonstrating the significant influence of the L_cut on VB. In contrast, for WC-Co uncoated tools, there is a slight increase from L_cut = 27.8 m to L_cut = 53.6 m, with both TW levels being minimal and predominantly abrasive. Conversely, a more pronounced increase is observed from L_cut = 53.6 m to L_cut = 73.7 m. Consequently, as is commonly concluded, the greater the TW, the lower the quality of the machined surface [59]. Regarding f, there is a general trend where an increase leads to a surplus in VB. Concerning TW, the most affected tool references were those subjected to higher f and longer L_cut, specifically T0L74F1500S126 and T1L74F1500S126 (see Table 8 for reference reminder). On the other hand, reference tools T0L27F750S126 and T0L54F750S126 (see Table 8 for reference reminder) exhibited very low TW, approximately 6 to 7× lower, in the range of 11–14 µm, when compared to the TiAlTaN-coated ones. Nevertheless, it is also known that flank abrasion leads to coating failure, and subsequently, increased wear on that coating [37,60].

4.3. Comparison with Other Used Coatings in Literature

As previously mentioned, AMPCOLOY^®83 is a material with a ductile behaviour, primarily induced by Cu, which has a strong tendency to induce material adhesion.

Based on the analysis of all the results obtained in this experimental procedure, it can be concluded that the coating did not positively impact the quality of the machined surface. The results for the TiAlTaN-coated tools were significantly worse than those presented by the WC-Co uncoated tools, both in R_a and VB. Sousa et al. [8] also studied the machining of a CuBe alloy but with diamond-like carbon (DLC)-coated tools, and came to the same conclusion that the coating did not have a positive effect on the quality of the machined surface. The coating under study is TiAlTaN, meaning that the presence of TiN theoretically should improve the resistance to coating delamination [37]. However, this was not the case, since the TiAlTaN-coated tools showed a much greater delamination tendency than the uncoated ones, which could be attributed to the presence of Ta, as coatings with Al typically yields more favourable results [61]. However, this is not a certainty that was observed throughout the procedure, and it may be a subject for future work.

Furthermore, one of the factors on which the quality of the machined surface depends is the level of residual compressive stresses to which it was subjected. If the HiPIMS deposition process is performed correctly, these stresses benefit low SR, promoting optimal coating cohesion and a tougher cutting edge [61]. What was observed throughout this procedure was precisely the opposite: poor coating adhesion, extensive delamination, cutting-edge abrasion, and cracking, which in many cases led to chipping. Improved mechanical properties are achieved when this technique is executed effectively, contributing to better coating performance [62,63]. Thus, it can be concluded that the PVD HiPIMS coating did not contribute to the excellent performance of the machined surface when performed under the deposition conditions used in this work.

5. Conclusions

The impact of f and L_cut on the roughness of machined surfaces and TW was evaluated for both WC-Co uncoated and TiAlTaN-coated tools. Some conclusions were drawn during this research, aligning with what is being reported in the literature regarding the influence of these parameters.

• In WC-Co uncoated tools, there was an evident influence of f and L_cut parameters on the R_a values. Regarding R_a values, the lowest values were observed for the test conditions of f = 750 mm/min and L_cut = 26.8 m (tool ref. T0L27F750S126), while the highest R_a values were for f = 1500 mm/min and L_cut = 73.7 m (tool ref. T0L74F1500S126), both in the longitudinal and transverse directions.
• This implies that the best R_a, R_t, and R_z values were achieved at lower f and L_cut values. Thus, it can be concluded that the machined surfaces with poorer quality were obtained for higher values of f and L_cut. Regarding VB, the most and least worn tools were observed for the same machining conditions that resulted in the highest and lowest R_a values, respectively. The main wear mechanisms observed were abrasion and adhesion of machined material.
• For TiAlTaN-coated tools: The result obtained for R_a w consistently higher compared to that observed in uncoated tools.
• This is likely due to coating defects, i.e., its rapid degradation during service. The R_a and R_a trends were comparable to those observed in uncoated tools, although with significantly higher values. The roughness and TW were lower for the lower parameters. Thus, the tool with the best roughness and VB values was tool ref. T1L27F750S126, while the least successful tool was ref. T1L27F1500S126. The main wear mechanisms observed were delamination, chipping, and abrasion,
• The maximum SR values are higher for higher f and longer L_cut. Additionally, in the case of coated tools, this value is more significant than in uncoated ones. The VB values for f = 750 mm/min and L_cut of 26.8 m and 53.6 m for the uncoated tools (tools ref. T0L27F750S126 and T0L54F750S126) were entirely satisfactory. Whereas in the other conditions, values ranging between approximately 60 μm and 150 μm were observed, these tools did not exceed 15 μm of VB, which is highly acceptable for VB.

The coating did not play a beneficial role in the machined surface quality, as significantly high R_a values were obtained, and cracks were present in several tools, representing a wear mechanism that the coating was intended to prevent. The failure of the coating is attributed to poor adhesion of the coating to the substrate and to the deposition conditions used, which were not adequately optimised. Material adhesion was found to accelerate the delamination of the coating. Furthermore, delamination was caused by the machined material’s adhesion to the substrate and the delamination boundary zone. This behaviour was observed in the coated tools, which exhibited delamination, resulting in a wear cycle on the tool. This wear cycle likely contributed to the tool failure.

The analysis of all the tools showed that the uncoated tools exhibited better cutting performance than the coated samples. The uncoated tools demonstrated relatively satisfactory results, with less pronounced wear values in VB and R_a.

Following the initial phase of the “Drivolution project”, limitations in work can be attributed to constraints related to minimal academic scientific information about all Cu-Be alloys due to their commercial status, constituting a restraint concerning both enriching the literature review on the alloy and a direct comparison and discussion of results with similar procedures, both at the alloy and machining process levels. Nonetheless, this study will be a great contribution to the endeavours of the injection mould industry, in the milling of AMPCO^® using TiAlTaN-coated tools, underscoring the significance of research for industrial optimization. The optimisation of the PVD deposition parameters need to be carried out because the works involving TiAlTaN coatings are not clear enough to draw a safe way to deposit very adherent coatings of this type [37,64].