1 Introduction
Ceramic fabrication using additive manufacturing (AM) is a developing technology that allows the rapid processing of complex 3D structures designed via a CAD model across diverse fields including micro-mechanics, microfluidics, and micro-optics [
1‐
3]. The process uses a sequential top-down assembly of dimensionally pre-determined layers resulting in a controlled deposition of curable materials in physical space representing the original design, namely, the green body. In ceramic AM, the green body is then subjected to a controlled thermal treatment phase producing a fully densified ceramic component. Although the process allows for the formation of complex shapes that cannot be manufactured through conventional techniques, it does however present challenges in terms of the quality and reproducibility compared to established methods [
4,
5]. Nevertheless, the advantages ceramic AM offer, e.g., complex structural features, make it a useful tool to create prototype designs for concept validation.
Widely used in the medical, aerospace and precision manufacturing industries, austenitic stainless steel 316L exhibits excellent physiochemical properties which make it a material of choice for applications requiring high hardness, corrosion resistance and biocompatibility [
6,
7]. It is however, as a difficult-to-machine material, and aside from its high hardness, it also has poor thermal conductivity, high tensile strength and produces high abrasion in the cutting tools [
8]. This results in a high heat transfer rate from the workpiece back into the cutting tools. Abrasion in particular, both flank and crater wear, occur in the tool with the presence of hard particles within the 316L matrix contributing to the high cutting temperatures. This form of abrasive wear has a signature abrasive profile parallel to the flow chip removal direction [
9]. Therefore, a particular type of machining approach is required to address this issue.
Aluminium oxide (Al
2O
3), is a widely used oxide ceramic material, characterised by high hardness, chemical inertness, and excellent thermal resistance [
10]. Its exceptional physicochemical properties allow for successful applications in the precision machining of hard to cut materials such as 316L [
6]. Although ceramics cutting tools can be used in dry machining conditions, the use of cooling methods to enhance the heat transfer in the cutting zone has shown the improvement of the tool wear profile and workpiece surface finish [
11,
12].
Previous studies have shown that the use of internal cooling methods to remove heat from the tool edge offer promising results [
5,
13‐
15]. To apply an internal cooling method to a ceramic material, traditional manufacturing methods (such as molding or, casting) are not feasible, whereas AM offers a flexibility not afforded by conventional forming processes.
There has been much interest by researchers in developing new methods to reduce the heat in the cutting tool using internal cooling mechanisms [
16,
17]. Closed internal cooling techniques offer an encapsulated system approach to remove the heat during the cutting process [
17].
In tandem with the tool insert is the concept of modular design in cutting tools. This method affords a systematic holistic approach as the tool insert cooling unit and toolholder body are now defined and distinct allowing for separation of the parts and ease of replacement [
17].
What follows is a review of current developments in internal cooling techniques. Wu et al. [
17] built a 3D finite element model (FEM) of the internal cooling system, thereby assessing the thermal performance of the proposed design. The authors used purified liquid water coolant with a 10 mm/s velocity under applied heat conditions of 50 W/mm
2 over 1 mm
2 contact area. Results showed the local tool tip temperature without the applied coolant was 155.8 °C and with the coolant active, 67.9 °C was observed.
Yao et al. [
18] using purified water as a coolant, designed a closed internal cooling system driven by a mechanical pump in a carbide insert. Numerical modelling simulated the fluidic flow and corresponding temperatures at the tool tip. A thermal imager measured the average temperature distribution across the tool tip, indicating a drop from 433.5 °C to 258.5 °C. This represents a 30% drop in the maximum temperature at the tool tip when compared to dry cutting. However, both methods outlined used liquid water as the cooling agent. The limitations are the thermal conductivity and specific heat capacity of the water.
The geometrical specifications of the internal channel and its proximity to the cutting edge of the tool is an important factor in optimising the design of the internal cooling system. Li et al. [
19] designed a topological internal channel that considered the cutting tool dimensions in the design process. In their study, the authors showed that through optimisation of the internal cooling channel, the heat transfer effectiveness could be improved by a margin of 16.32 °C relative to a conventional design, using a flow rate of 1 mm/s through a liquid water coolant. The results emphasise the importance of the geometry and position of the internal channel relative to the heat source when designing the cooling system.
Fang and Obikawa [
14], employed a differential pressure flow using high pressure liquid water in a cemented carbide insert, with the internal channels fabricated through electro discharge machining. The non-uniform distribution of the fluid was analysed using computational fluid dynamic (CFD) tools, which indicated a corresponding increase in the cooling rate as the pressure increased. This was observed experimentally on an Inconel 718 workpiece, with reduced tool notch wear and flank wear found on the insert with the internal cooling active. The authors also found a connection between the angular positioning of the internal cooling channel and the flow rate behaviour, which in turn can enhance the heat dissipation.
Shu et al. [
20] used a tungsten carbide insert with an insert wall thickness of 1.8 mm and internal wall thickness of 0.7 mm towards the flank face, to spray the cutting zone with liquid water coolant when machining the aluminium alloy 6061. The authors employed a composite turning tool integrated with a pressurised internal and spray cooling mechanism. CFD analysis was performed to ascertain the effectiveness of the design with subsequent testing using a heat flow rate of 10 W/mm
2 into the insert, and an inlet liquid velocity of 1 m/s at 20 °C. Combining simulation and Taguchi methods, the optimal geometric configuration of the internal channel dimensions was found. By measuring the local temperature with a K-type thermocouple, it was observed that as the tool temperature increased, the spray and internal cooling mechanism were more effective at heat transfer. It should be noted that this method of combining internal and spray cooling is not a closed cooling system, and therefore produces (albeit at much reduced levels), an external coolant to the cutting zone area.
Chen et al. [
21] adopted a combined minimum quantity lubrication (MQL) and internal cooling approach machining the nickel based super alloy GH4169. This material exhibits high strength and low thermal conductivity, displaying similar properties to 316L. Using a modified SiAlON cutting tool, an FEM, based on Newtonian cooling interactions, was developed to investigate the temperature distribution field with subsequent validation achieved through machining tests. This design used an internal chamber with two microchannels of the same dimensions, which is fed by a regulated integrated pressurised water based semi-synthetic fluid, with the cooling inlet at the tool end. The first microchannel consisted of a cooling and lubrication dispenser, whereas the function of the second microchannel was primarily to assist chip removal. As expected, the authors found the highest temperature distribution was located at the tool tip, with reduced temperature profile as this region extended radially from the tool-chip zone. It was also found that the region of highest temperature was in the primary and secondary deformation zones, respectively. The results showed an 80 °C drop in tool temperature using the combined MQL and internal cooling system when compared with the dry cutting regime. However, as in Ref. [
20], this method used the coolant in combination with the MQL to effectively spray the cutting zone, as such, it was not, strictly speaking, a closed internal cooling system.
Singh and Sharma [
22] investigated the temperature variance between dry machining and internal cooling with laminar and turbulent flows models, relating to the cutting tool tip distance from the workpiece depth. The internal cooling system showed a temperature drop of ~ 29% for laminar flow behaviour, and ~ 53% for turbulent flow within the confines of the internal channel. The authors also noted that the temperature cooling effect was more pronounced at the tool tip and this difference radially reduced as the distance from the tool tip increased.
Shu et al. [
23] developed a closed internal cooling system in a carbide insert for dry machining using liquid water as the coolant. Finite element analysis found that the optimised thickness of the internal channel walls that could withstand the mechanical loads was 0.1 mm for the rake face and 0.7 mm at the flank face. Applying a heat flux of 20 W/mm
2, it was shown that the maximum temperature dropped from 381.62 °C to 273.9 °C using an inlet liquid velocity of 0.15 m/s. The authors noted that at higher coolant circulation velocities, the relative effective temperature
\(\left( {T_{{{\text{eff}}}} - T_{{{\text{max}}}} } \right)/T_{{{\text{ref}}}}\) was the most effective in heat absorption. The results showed the performance of the internal cooling system depends on the geometry of the internal channel along with the velocity and cooling properties of the liquid [
23]. These results again indicate internal cooling is an effective means to remove thermal energy in the cutting process, however, it is not stated the material workpiece parameters used in the study.
Isik [
24] machined the nickel-based superalloy Waspaloy, using a prototype tool holder with a coated carbide insert. This workpiece material is particularly difficult to machine due to its high shear strength, high chemical affinity, and low thermal conductivity [
24]. Employing a 2 mm wide internal channel, with purified liquid water at 18 °C as the cooling agent, the fluid was mechanically pumped in a closed circulatory system with a flow rate of 0.5‒2.0 m
3/h. Using a pyrometer to measure the average tool temperature over varying cutting speeds, experimental tests indicated dry cutting, with a speed of 95 m/min, produced a temperature of 641 °C. With the liquid coolant applied, there was a reduction in temperature to 587 °C, representing a decrease of 9%. Flank wear was the dominant form of tool wear observed due to the high heat generated, along with the low thermal conductivity of Waspaloy. Overall, the authors indicated a 12% increase in tool longevity using the internal cooling system was achievable. The surface finish was also improved with the internal cooling applied. Using the maximum cutting speed of 95 m/min with a fluidic velocity of 1.6 m/s at a depth of cut of 0.5 mm, produced a surface roughness of
\(R_{{\text{a}}}\) 0.699 µm [
24]. This represents a 13% increase in surface quality compared to the dry machining tests. Again, the highest flow rate and cutting speed produced the best effect in terms of surface quality and heat transfer.
Öztürk et al. [
25] used an internal cooling technique to machine 1040 steel and compared the data for tool tip temperature differential and dry machining relating to the average value of the surface roughness. The authors employed an in/out liquid water coolant flow that dispersed onto the base of the insert. Combined with a coolant reservoir, aluminium blocks were used in conjunction with integrated Peltier modules operating as a fan. A CFD model was established and subsequently validated by machining tests. The resultant data showed a 107 °C local temperature drop in the tool tip when compared to the dry test. The measured surface roughness on the 1040 steel obtained a range of values from 0.18 µm to 2.05 µm. Although the results showed a 107 °C drop when compared to dry machining experimentally, it was a relatively complex design.
To ensure good structural conformity in the insert, it is necessary to design the internal channel dimensions within the boundaries of the mechanical reliability parameters. Li et al. [
26] used a topological design method to deduce the fluidic behaviour and optimise the structural design of the internal coolant features. The results showed a 180.4 °C drop in temperature using the internal cooling system with the topological design, in comparison to the dry cutting regime.
Ingraci Neto et al. [
27] performed experimental machining tests using a prototype cutting tool with a two-phase pump cooling system on 1045 steel during uninterrupted turning. This approach implemented a closed loop design whereby the circulating liquid water vaporised upon contact with a silver interface that acted to cool the cutting tool during machining. It forms with the inlet channel, a tilted annular section of 30 mm
2 that prevents vapor entrapment and has 106 mm
2 of heat transfer surface area. The 54 mm
2 base is in contact with the silver interface. Condensing of the vaporised liquid occurred through forced convection at 25 °C which was stored in an accumulator and then pumped back (via a peristaltic pump) into the internal channel system. The water is pumped at a feed rate of 1.78 min
−1. Three thermocouples were fixed in contact with the tool to measure the temperature change. The results showed that the temperature of the cutting tool with internal cooling was 79 °C lower than the maximum temperature reached in dry cutting.
Uhlmann and Meier [
28] developed a numerical model of the heat transfer mechanism using standardised industrial inserts. In their study, the heat flow is directed orthogonal to the rake face which is then conducted through the tool into a copper heat sink. The accumulated heat is then transferred through forced convection into a dynamic fluid consisting of water/water and glycol flowing parallel to the rake face. The numerical modelling results showed a maximum tool temperature reduction of 21% with the water/water glycol agent relative to water.
This method requires an external pump to drive the circulation of the fluid, a heat exchanger, and a chiller. It therefore is relatively complex compared to more simple designs.
Shu et al. [
29] performed numerical and experimental studies on a closed looped internal cooling tool. Using a tungsten carbide insert, modelling showed that a cutting-edge thickness of 1 mm and a wall thickness of 0.7 mm from the flank face, was able to withstand the thermomechanical loading. Thermocouples were used to measure the temperature in the insert during the experiments. Simulation results showed that the maximum temperature reduction of 82.68 °C was achieved with the internal cooling system. Experiments were not conducted in an online machining process, but the authors instead used a modified experimental set-up via heat induction by a solder iron. The results indicate that the effectiveness of the liquid water-cooling system increases as the inlet velocity, heat flux and tool-chip contact area are increased [
29].
State of the art review and analysis in internal cooling mechanisms in cutting inserts have revealed the following.
(i)
Internal cooling fluids have a positive effect on the heat transfer rate in cutting tools, and this relates to a reduction in tool wear and improved workpiece integrity overall.
(ii)
The location and geometrical dimensions of the internal channel has a significant impact on the effectiveness of the cooling system. Furthermore, it can be said that as follows.
①
There is a direct correlation between the distance of the cooling channel from the main source of heat on the cutting edge, and the effectiveness in heat transfer within the channel.
②
The location and dimensions of the internal channel can reduce the structural integrity of the cutting insert, therefore considered design of the geometry is required to enable the retention of the structural strength, whilst providing for optimum cooling is paramount in the design process.
③
Research suggests that the higher the temperature within the cutting zone, the faster the temperature transfer is achieved when using internal cooling systems.
(iii)
The state of the art in internal coolant systems in cutting inserts is primarily centred on liquid water as the coolant fluid of choice. This limits the effectiveness of the fluid in terms of its thermal conductivity.
(iv)
To accurately model the thermomechanical effects on the cutting insert it is necessary to employ a combined analysis, this is the basis of conjugate heat transfer. This requires modelling of the solid-liquid phase boundaries.
(v)
To date, all the internal cooling methods used a mechanical pump to drive the liquid around the defined inner chamber. There exists no other method that has been successfully implemented in a model or prototype that can achieve cyclic circulation, and removal of the heat currents, without an integrated external power source.
(vi)
Based on the current research, it can be extrapolated that the use of liquid metals as an internal cooling substance in cutting inserts is novel. Furthermore, the use of a magnetohydrodynamic drive integrated into the internal cooling system is original in its proposal.
This work describes the design and fabrication process involved in creating an aluminium oxide cutting tool with an internal cooling channel formed through a ceramic additive manufacturing method. This study aims to investigate the design, analysis and performance of an aluminium oxide cutting insert using a developed thermomechanical numerical model combined with controlled experiments on a custom-made turning machine. Liquid gallium as an internal coolant, combined with permanent magnets to generate a homogeneous magnetic field, forms the basis of a heat transfer mechanism through a magnetohydrodynamic drive. Thus, allowing enhanced heat transfer within the boundaries of the defined geometrical structure of the internal channel, without the need for external coolants or mechanical power input.
Experimental results showed at \(V_{{\text{c}}}\) = 250 m/min, the corner wear VBc rate observed was 75 µm with the coolant off, and 48 µm with the coolant on. When increased to \(V_{{\text{c}}}\) = 900 m/min, the corner wear VBc rate showed 357 µm with the coolant off, and 246 µm with the coolant on. To provide further validation of the new internal cooling system, experimental tests were compared against the results of the liquid gallium coolant versus external liquid water coolant. At \(V_{{\text{c}}}\) = 250 m/min, the difference between the tool wear rate reduction with the internal coolant relative to the external coolant was 29%. Increasing this to \(V_{{\text{c}}}\) = 900 m/min, the difference observed between the internal liquid gallium coolant relative to the external coolant was 16%.