*3.1. Friction Coefficient Evaluation*

The friction coefficient (μ) of the specimens shows specific behavior for different sliding lengths. These phenomena may be mainly caused by the wear mechanisms involved in each stage of the friction process. In this way, through analyzing the obtained values for the friction coefficient, three stages were detected, with different behavior during the sliding course, Figure 3.

**Figure 3.** Friction coefficient and pin temperature as a function of sliding length for 1000 m.

The first stage is observed up to 200 m of sliding length, where the "stick-slip" phenomena takes place. This effect is developed in the dynamic contact between both surfaces, resulting in unstable movements along the sliding track. In the first friction instant, a significant increase in the contact force and μ were caused by the detachment of initial asperities which come from the surface material of the tribological pair. These asperities may cause specific roughness values, leading to a smaller contact surface between the pin and the disc, resulting in higher contact pressures. The action of high contact pressures on rough surfaces tend to remove soft asperities, favoring the detachment of wear debris on the sliding track, and giving place to high amplitude in the friction coefficient values and the appearance of the effects of abrasive wear mechanisms. In this aspect, an important increase of amplitude values implies a growth in the instability during the initial length of the process, especially because of the lack of uniformity of the circular trace, Figure 4.

**Figure 4.** First stage sliding track with abrasive wear effects.

The increase in μ values is mainly caused by the movement of the pin over wear debris from initial aluminum asperities, involving a quick growth of the temperature in the contact area. A combination of friction and temperature conditions also favor adhesive phenomena from the aluminum particles to carbide surface.

The next stage starts with a decrease in μ values and a stabilization of the pin temperature. An important reduction in the oscillations amplitude of μ was also detected. This behavior is mainly due to a growth of the adhesion layer of aluminum alloy formed from the slip track over the pin in stratified sections. Under these conditions, adhesion of wear debris over the slip track and pin surface induces the modification towards softer topographies between contact elements, resulting in a sliding process where surfaces of the same material contact each other, Figure 5.

**Figure 5.** Second stage sliding track with abrasive wear effects.

In the last stage, the detachment is produced by aluminum adhered debris from the pin surface, as a result of reaching an excessive critical volume. Under these conditions, the adhered layer becomes unstable and may be removed as a consequence of the friction forces on the contact area. In this way, the carbide surface is subjected to wear from the adhered aluminum layer, promoting the lost of particles from the hemispheric surface. Because of this effect, we ar debris of aluminum and carbide (harder than the disc) are deposited again over the sliding track. The material debris on the wear trace results in an increase of the μ values and the temperature of the pin, Figure 6. During this stage, the described effect is repeated as a continuous cyclic and dynamic behavior of adhesion wear mechanisms, as is described in previous works [12].

**Figure 6.** Third stage sliding track, with an adhered debris layer over the sliding way, and deposited material in the outer edges of the track.

When the carbide pin's surface is analyzed, the existence of three different sections subjected to specific wear mechanisms are observed, Figure 7. In this aspect, similar wear effects are noticed from the morphological adhesion behavior over all of the tested specimens.

**Figure 7.** SEM (100×) micrography of the Pin for test 500 m sliding distance. Detailed areas for the adhesion and punctual Energy Dispersion Spectroscopy (EDS) of the different wear phenomena.

Section 1 is especially characterized by the existence of Al–Cu alloy worn particles. These particles have been mechanically adhered to the hemispherical surface, giving place to abrasive wear phenomena in the first instants of the tribological tests.

Section 2 is formed by the primary layer developed in the first instants of the tribological test. EDS analysis shows that the composition of this layer is close to pure Al. According to previous research [13], this adhesive mechanism is mainly associated with a thermomechanical effect.

Section 3 is composed of the secondary layer, specifically described by a stratification of wear debris which is adhered over the primary layer through mechanical effects due to thermomechanical phenomena. Cu composition percentages near to 2.65% were observed.

#### *3.2. Wear Effects Evaluation*

The material volume loss was selected as the control parameter in order to study the sliding wear effects on the aluminum specimens. In this way, we ight variations caused by the Pin-on-Disc test were measured on test probes with different sliding length configurations. The material volume loss was obtained by using the aluminum alloy density:

$$
\Delta V = \frac{\Delta w}{\rho\_A} = \frac{w\_F - w\_0}{\rho\_A} \tag{1}
$$

where *wF* is the weight of the tested samples, *w*<sup>0</sup> is the weight of the samples before the sliding test, and *ρ<sup>A</sup>* is the UNSA92024 aluminum alloy density.

In order to simplify analysis considerations and following the indications of the ASTM G99 Standard [6], the wear volume loss was considered negligible for the harder material (WC–Co). However, adhesive wear phenomena from the aluminum discs to carbide hemispherical pins have been evaluated

Figure 8 shows the analysis of the volume variation for different samples as a function of sliding length (*Ld*).

**Figure 8.** Volume variation vs. sliding length: (**a**) pin; (**b**) disc.

As was expected, a relevant increase in the values of volume lost for the aluminum discs was detected as a function of sliding length, fitting to a linear behavior and confirming previous research findings [7]. Regarding the pins, an estimation of the main wear mechanism was proposed. On the basis of the results indicated in the Figure 3, an important growth tendency was detected for the adhered material volume on the pins contact surface as a function of the sliding length. This fact may corroborate the raised hypothesis about the importance of secondary adhesion mechanisms in the analyzed tribological pair, detecting a progressive increase of the wear effects regarding the interaction time.

The appearance in the first instants of several negative values in the variation of volume of the pins should be noticed; this indicates the existence of a slight abrasive wear process prior to the adhesive phenomena. In this respect, SEM confirmed this behavior (Figure 9).

**Figure 9.** SEM (90×) micrography of the pin for test condition 900 m sliding distance. Detail of the abrasion produced by particles drag (600×).

This wear effect was observed in the first instant of sliding tests, as has been commented previously, in which the stick-slip phenomena takes place. The abrasive wear effect finishes when the surface tension of the material breaks, stabilizing the forces involved in the process and allowing the appearance of the first stage of adhesive mechanisms [14]. Furthermore, from the registered data in the tests, the Archard wear ratio [15] was determined, using the following expression:

$$K\_{\mathbb{S}} = \, a \cdot L\_d^{-1} \tag{2}$$

where *Ks* is the Archard coefficient, *Ld* is the sliding distance and:

$$a = \frac{\Delta V \cdot H}{10^3 \cdot N} \tag{3}$$

where Δ*V* is the volume variation, and *H* is the softer material hardness [16].

The coefficient exposed in Equation (2) allowed us to carry out the marginal analysis of Archard ratio as a function of the sliding length, taking special care of the fact that the proportionality coefficient is not a constant value, where direct [Δ*V*, *Ld*] and indirect variables [*H*, *N*] are involved. In fact, the initial hardness of the material may vary as the sliding distance increases, mainly because of a superficial softening effect on the material, favored by the temperature increase in the contact area.

These considerations can differentiate the theoretical model from the real conditions. For this reason, marginal studies can be carried out by the approximation of the experimental results to an empirical model. In this way, a potential model has been selected, following the research lines with a specific interest in material removal [17–20].

$$K\_s' = a' \times L\_d^b \tag{4}$$

This equation can be linearity expressed by logarithmic expression:

$$
\log \mathsf{K}'\_{\mathfrak{s}} = \log a' + b \times \log L\_d \leftrightarrow \mathcal{y} \; = \; m \times \mathbf{x} + n \tag{5}
$$

The potential model is shown in Figure 10 as a function of sliding distance for the different *Ks* coefficients obtained from empirical and theoretical models.

**Figure 10.** Archard wear coefficient vs. sliding distance.

Comparing the Archard theoretical model (*Ks*) and empirical model (*Ks'*), a relevant difference can be observed in the exponent which govern the sliding length. This disparity may be justified by the existence of specific wear mechanisms that are not taken into account in the theoretical model.

Material hardness (*H*), normal load (*N*), and wear volume (Δ*V*) are considered constant in this model. However, these components are direct or indirect variables of the process, making *a* coefficient not constant.

In this way, the *Ks* values tendency obtained are located in the Archard range for compatible and/or similar materials subjected to adhesive phenomena [21]. The first consideration may be justified by the friction behavior between the aluminum disc and adhered particles from the alloy to the carbide pins surface. The second one may be justified by the compatibility between Wolfram (W) with Aluminum (Al). This compatibility can be evaluated by the Rabinowicz relation [22], showing higher solubility values (>1%) and a high tendency towards adhesive mechanisms, Figure 11.

**Figure 11.** Partial reproduction of the Rabinowicz's table (adapted from [22]).

#### **4. Conclusions**

Wear mechanisms are the main responsible factors of cutting tool wear, being present on a wide temperature range. The main mechanism for the tribological pair Al–Cu and WC/Co is secondary adhesion, where the part material is removed and added to the cutting tool surface in the first step. After that, it brings with it the cutting tool's own particles, increasing the wear effects.

This work studied the tribological interference, simplifying to lab conditions (Pin-on-Disk) of a machining process. This allowed us to isolate the wear due to continuous friction between the contact pair, making it easy to characterize and to verify wear behavior.

The obtained results show an initial abrasion mechanism in the WC/Co pin, which is followed by the secondary adhesion of the aluminum alloy.

This adhesion takes place in two different stages. Firstly, thermomechanical effects (pre-fusion/adhesion) lead to the generation of a thin layer of pure aluminum, which comes from the aluminum matrix. After that, other layers with a similar composition to the Al–Cu alloy are adhered in a stratified way over the first one. This adhesion is due to mechanical effects. With the PoD tests it has been verified that, for pressures close to the ones achieved in finishing machining, the Built-Up Edge and Built-Up Layer effect can be studied for UNS A92024 alloy and WC–Co tribology pair. They are the main wear mechanisms for this pair against abrasion or erosion.

Furthermore, the classical model for the evaluation of Archard wear coefficient do not provide solid results, not taking into account variation of the process as it happens with the superficial hardness of the disc or the orientation of the pin.

**Author Contributions:** J.S. and M.B. conceived and designed the experiments; J.M.V.-M., I.D.S and J.S. performed the experiments; M.B. and J.M V.-M. analyzed the data; J.S. and J.M V.-M. wrote the paper.

**Funding:** This research was funded by the Spanish Government (MINECO/AEI/FEDER, UE), grant number [DPI2017-84935-R] and the Andalusian Government (PAIDI).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
