Environmental Analysis of Sustainable and Traditional Cooling and Lubrication Strategies during Machining Processes

: Due to rising demands of replacing traditional cooling strategies with sustainable cooling strategies, the development of sustainable strategies such as minimum quantity lubrication (MQL) of nano-cutting ﬂuids (NCFs) is on the rise. MQL of NCFs has received a lot of attention due to its positive impact on machining process e ﬃ ciency. However, environmental and human health impacts of this strategy have not been fully investigated yet. This work aims to investigate the impacts of MQL of molybdenum disulﬁde (MoS 2 ), multi-walled carbon nanotubes (MWCNTs), titanium dioxide (TiO 2 ), and aluminum oxide (Al 2 O 3 ) NCFs by employing a cradle-to-gate type of life cycle assessment (LCA). Besides, this paper provides a comparison of the impacts and machining performance when utilizing MQL of NCFs with other cooling strategies such as traditional ﬂood cooling (TFC) of conventional cutting ﬂuids and MQL of vegetable oils. It was found that NCFs have higher impacts than conventional cutting ﬂuids and vegetable oils. The impacts of TiO 2 -NCF and MoS 2 -NCF were lower than the impacts of MWCNTs-NCF and Al 2 O 3 -NCF. MQL of NCFs presented higher impacts by 3.7% to 35.4% in comparison with the MQL of vegetable oils. TFC of conventional CFs displayed the lowest impact. However, TFC of conventional cutting ﬂuids is contributing to severe health problems for operators. MQL of vegetable oils displayed higher impacts than TCFs of conventional cutting ﬂuids. However, vegetable oils are considered to be environmentally friendly. According to the ﬁndings, the MQL of vegetable oils is the most sustainable strategy for machining processes with associated low / medium cutting temperatures. While MQL of TiO 2 and MoS 2 NCFs are the sustainable strategy for machining processes associated with high cutting temperatures.


Introduction
Usage of cutting fluids (CFs) results in increasing the machining process efficiency in terms of power reduction, rust control, and tool life improvement. CFs improve tool life and workpiece surface quality, during any metal cutting operation by decreasing the cutting temperatures [1]. A certain CF's lubrication property is significant, due to its influence on the friction between the cutting tool-workpiece interface. Appropriate lubrication decreases the friction and prompts smooth chip flow from the workpiece over the cutting tool [1]. Besides, proper lubrication decreases the effects of built-up-edge, which results in surface finish improvement [1]. In addition, CFs can remove formed chips from the nanoparticles are added to improve the heat transfer capabilities. It has been found that even a small amount of appropriate nanoparticles can result in substantial improvements in the fluid's ability to remove heat from the tool-workpiece interface [11]. The production of these fluids can be divided into three distinct parts: production of the base fluid, production of nanoparticles, and dispersion. Depending on the type of particles that are needed, a series of chemical and mechanical processes are required to produce usable nanoparticles. Common nanoparticles that have been explored for use in machining fluids include TiO 2 , Al 2 O 3 , MoS 2 , single-walled carbon nanotubes, and multi-walled carbon nanotubes [12]. Vegetable oils are shown to have very good tribological properties due to the structure of their molecules. The triglyceride structures allow strong lubricant films to be formed at the tool-workpiece interface during machining, significantly reducing the friction and cutting forces [8]. However, these same structures feature strong intermolecular interactions, causing a lower heat transfer ability than what is desired. Therefore, vegetable oils are excellent base fluids, since the introduction of nanoparticles can improve the thermal conductivity of the fluid. Previous work showed the effectiveness of adding 1 wt.% of MoS 2 nanoparticles to existing vegetable oils. Where heat transfer that resulted in tool wear and surface roughness improved by 15.5% and 22.5% for the same process [13]. Graphite is a well-known lubricant additive, as the structure of graphite allows planes of carbon atoms to slide over each other with ease. It was shown that this behavior remains intact when graphite nanoparticles are introduced to a cutting fluid [14].
Machining processes can be conducted under different types of environments, which can be categorized into commonly used traditional flood cooling (TFC), minimum quantity lubrication (MQL), cryogenic cooling, and dry machining. Minimum quantity lubrication (MQL) technique is also known as near dry lubrication [2,15], micro lubrication, and near-dry machining [15]. MQL is viewed as an environmentally friendly cooling technique and it is economically justifiable to use [5,16]. MQL techniques apply a fine mist of air-CF mixture into the cutting zone. The mixture contains minimal amounts of CFs, and it is applied through a nozzle that is around 1 mm in diameter. The mixture is applied at an approximate pressure of 600 kPa and flow rates between 50 mL/h to 2 L/h [2], unlike TFC that uses flow rates in the range of 50 L/h to 1000 L/h [16]. Advantages of MQL include its elimination of disposal and recycling problems associated with CFs, because of the tiny amounts of CFs used/consumed in cutting processes.
During the last decade, nano-cutting fluids with MQL (MQL-NCFs) have received increasing attention. This strategy has the potential to enhance heat transfer and tribological properties by using nano-cutting fluids using a small quantity of CF. Wusiman et al. [17] studied heat transfer properties of dispersed multi-walled carbon nanotubes (MWCNTs) with two types of surfactants in distilled water. NCF of 0.5 wt.% CNT with 0.25 wt.% of sodium dodecylbenzene sulfonate showed enhanced thermal performance of 2.8% in comparison to the base fluid of distilled water. Hegab [18] discussed NCFs' potentials to enhance process performance while machining Inconel 718 when employing the MQL technique. MWCNTs and Al 2 O 3 nano-additives were utilized with a base fluid of rapeseed oil. MWCNTs nano-additives displayed smaller tool wear than Al 2 O 3 nano-additives. Das et al. [19] investigated mixing aluminum oxide nanoparticles with radiator coolant, then applying the MQL cooling technique during machining 4340 AISI alloy steel. The used NCF displayed better machining performance in terms of flank tool wear, in comparison to soluble water and compressed air coolants. Rapeti et al. [20] analyzed mixing molybdenum disulfide (MoS 2 ) particles with three types of vegetable oils: coconut oil, sesame oil, and canola oil. These NCFs were utilized when turning AISI 1040 steel. Coconut oil with 0.5% MoS 2 nanoparticle inclusion presented the highest performance, which reduced cutting forces, tool wear, surface roughness, and cutting temperature by 37%, 44%, 39%, and 21%, respectively. Gupta et al. [21] discussed machining performances of Inconel 800 alloy when applying sunflower oil with three types of nano-additives: Al 2 O 3 , MoS 2 , and graphite. Results indicated that graphite had the highest machining performance in comparison to the other nano-additives. Sen et al. [22] studied the machining performance when using MQL during milling of Inconel 690. Different concentrations of Al 2 O 3 nanoparticles were investigated, which ranged between Sustainability 2020, 12, 8462 4 of 22 0.5% to 5%. Palm oil was the base fluid for the nanoparticles. In conclusion, Al 2 O 3 nanoparticles with a concentration of 2.5% displayed the best performance, in comparison to other concentrations, dry machining, and flood cooling of pure palm oil.
Researchers have claimed that the usage of the hybrid strategy of MQL-nano-vegetable base fluids is one of the most sustainable machining techniques [17][18][19][20][21][22][23]. Previous works have focused only on machining performance enhancement when using this strategy without studying the environmental and human health impacts. However, it is commonly known that the manufacturing processes of nanoparticles are energy-intensive processes [24]. Few studies have been conducted to evaluate the environmental and human health impacts of the nanoparticles, but environmental inventory data about the life cycle stages of nanoparticles is not available, especially the use and end-of-life stages [24]. Furthermore, information about exposure routes, exposure potentials, and the level of toxicity of the nanoparticles is limited [25]. Besides, there is a need to compare the impacts MQL of NCFs with other cooling strategies such as traditional flood cooling (TFC) of conventional CFs and MQL of vegetable oils. Due to the mentioned gaps and potentials of using nano-vegetable base fluids, there is a need to investigate the environmental and human impacts of this process thoroughly.
The goal of this paper is to investigate the environmental and human health impacts associated with NCFs. Multiple like cycle assessments (LCAs) were conducted for various types of NCFs including nanoparticles of molybdenum disulfide (MoS 2 ), multi-walled carbon nanotubes (MWCNTs), titanium dioxide (TiO 2 ), and aluminum oxide (Al 2 O 3 ). Two vegetable oils (soybean and rapeseed oils) were utilized as a base fluid for each type of NCF. LCAs covered material extraction and manufacturing stages of the base fluids. Besides, the conducted LCAs included synthesis processes of the nanoparticles and the preparation process (dispersion process) of the NCFs. Additionally, LCA comparisons between two conventional CFs, two vegetable oils, and eight NCFs are conducted and presented. The functional unit in this work is based on CFs in a volumetric content of 150 mL including 1 wt.% of the nano-additives. Besides, LCA comparisons between the three cooling strategies were conducted and presented. The compared cooling strategies are flood cooling of conventional cutting fluids, MQL of vegetable oils, and MQL of NCFs. The conducted LCAs of the different cooling strategies covered the required quantities of the CFs and the related power consumption for machining (turning process) an industrial application of a single AISI 4043 bolt. Besides, this paper offers an overall evaluation of the considered cooling strategies through human health impacts at machining process, environmental and human health impacts at end-of-life stage, and machining performance. The layout of this paper is broken down into the following. Section 2 presents the methodology used in this work. Section 3 presents the results and discussion of the conducted LCAs. Section 4 presents an overall evaluation of the considered cooling strategies. Finally, the conclusion section provides a summary of the findings and possible points for future work.

Materials and Methods
This study aims to investigate the environmental and human health impacts of three cooling strategies. These cooling strategies are flood cooling using conventional CFs, MQL using vegetable oils, and MQL using nano-cutting fluids. Figure 1 displays three cooling strategies. Life cycle assessment (LCA) is the tool that is used to assess the environmental and human health impacts of each cooling strategy. LCA covers all stages of any product which are: material extraction stage, manufacturing stage, usage stage, and end-of-life stage. This tool is commonly used to analyze the environmental impacts [26][27][28]. Proper usage of this tool provides users with the knowledge to select the most desirable sustainable machining process for their specific jobs. LCAs analyze a product's impacts on three areas: natural environment (ecosystem), human health, and natural resources. The guidance standard and principles of LCA are explained in ISO 14040 [29]. The execution of LCA includes four main steps: Sustainability 2020, 12, 8462 5 of 22 (a) The first step includes the goal and scope definition of the assessment, as interpreted in ISO 14041 [30]. It includes the purpose of the LCA, the definition of the considered product, and the needed resources to conduct the LCA. (b) The second step represents the inventory analysis according to ISO 14041. It shows the product's material flow. It also provides information during all four stages of the product's life in regard to input quantities, consumed amounts of different types of energy and water, and consumed resources during transportation during each of the product's four life stages. (c) The third step is the impact assessment according to ISO 14042 [31]. During this step, the environmental and health impacts of products are calculated. This step is considered to be the most important one, due to its crucial impact on the conducted LCA's results. (d) The fourth and final step is used to represent the results interpolation according to ISO 14043 [32].
Results of conducted LCA are analyzed and conclusions are provided to the end-user.
Sustainability 2020, 12, x FOR PEER REVIEW 6 of 22 In this paper, IMPACT 2002+ is employed to calculate the LCA impacts of the cooling strategies. This life cycle impact assessment methodology is a hybrid model, which combines both the midpoint model and the endpoint model [35]. Besides, it covers a wide range of the impact categories (midpoints and endpoints), and a large number of substances. Four damage categories and 14 midpoint impact categories are included in this model, as listed in Table 1. Each damage category contains several midpoint impact categories. The midpoint impact categories are utilized in LCAs of the primary components of the CFs. While damage impact categories are used in LCAs of the considered CFs and cooling strategies. The damage categories are human health, ecosystem quality, climate change, and resources. The human health damage category includes the human toxicity (carcinogens and non-carcinogens), respiratory, ozone layer depletion, and ionizing radiation

The System Boundary of the Cooling Strategies
There are various systematic approaches to conduct LCAs, and these approaches are referred to as system boundaries. System boundaries are also categorized as follows: Cradle-to-grave, cradle-to-gate, gate-to-gate, and gate-to-grave. The cradle-to-grave approach includes all four life cycle stages. The cradle-to-gate approach includes the first two stages of the four life cycle stages, which are the Sustainability 2020, 12, 8462 6 of 22 material extraction and manufacturing stages. The gate-to-gate approach covers only the manufacturing stage, while gate-to-grave covers all stages to expect the material extraction stage.
In this study, the cradle-to-gate system boundary is considered within the conducted LCAs for 12 CFs. The material extraction and manufacturing stages for the CFs were considered, as shown in Figure 1. Besides, the other conducted LCAs for 12 cooling strategies covered the use stage (machining process). Regarding traditional flood cooling (TFC) of conventional CFs, the material extraction stage includes crude oil extraction, and the manufacturing stage covers oil desalting, distillation, and purification processes. Considering MQL of vegetable oils, the material extraction stage includes oilseeds harvesting and cleaning processes, while the manufacturing stage includes pre-treatment processes of the oilseeds, extraction, and the refining processes. For MQL of NCFs, the material extraction stage and the manufacturing stage of vegetable oils are considered. Besides, the material extraction stage also includes extraction inputs of various nano-additives. Included nano-additives are molybdenum disulfide (MoS 2 ), aluminum oxide (Al 2 O 3 ), multi-walled carbon nanotubes (MWCNTs), and titanium dioxide (TiO 2 ). The manufacturing stage of MQL of NCF also covers nano-additive fabrication techniques and the two-step approach used in preparing the nano-fluids. Furthermore, executed LCAs includes the usage stage of the three cooling techniques during machining processes. The use stage includes the related cutting energy, pumping energy, compressor energy, and the quantity of coolant for turning a single AISI 4043 bolt.

IMPACT 2002+
Outputs and emissions of life cycle stages contribute to negative impacts in three areas: (a) The ecosystem, (b) resources, and (c) human health [33]. For modeling life cycle impacts, there are different approaches such as the midpoint assessment model, endpoint assessment model, and a hybrid assessment model. The midpoint model depends on determining the number of impacts for each area. The endpoint model (damage approach) concludes with an impact for each impacted area. The hybrid model combines midpoint and endpoint models. These models are used to convert LCA inventory analyses into LCA impact indicators during the third step of conducting LCA. These impact indicators are calculated by considering the characterization factor for each emission. Menoufi [34] discussed the differences between these modeling methods for life cycle impact assessment (LCIA). The midpoint model is adopted in CML, TRACI, and EDIP 2003. The endpoint model is considered in EPS 2000, EI99, Eco Scarcity, and JEPIX. The hybrid model is employed in IMPACT 2002+, RECIPE, LIME, and LUCAS. Various available software packages include LCA inventory database and LCIA methods such as SimaPro, OpenLCA, and GaBi LCA software where the databases describe the production of several materials, chemicals, and energies. Besides, they contain databases for describing the post-use stage of materials.
In this paper, IMPACT 2002+ is employed to calculate the LCA impacts of the cooling strategies. This life cycle impact assessment methodology is a hybrid model, which combines both the midpoint model and the endpoint model [35]. Besides, it covers a wide range of the impact categories (midpoints and endpoints), and a large number of substances. Four damage categories and 14 midpoint impact categories are included in this model, as listed in Table 1. Each damage category contains several midpoint impact categories. The midpoint impact categories are utilized in LCAs of the primary components of the CFs. While damage impact categories are used in LCAs of the considered CFs and cooling strategies. The damage categories are human health, ecosystem quality, climate change, and resources. The human health damage category includes the human toxicity (carcinogens and non-carcinogens), respiratory, ozone layer depletion, and ionizing radiation categories. The ecosystem quality damage category includes aquatic ecotoxicity, terrestrial ecotoxicity, terrestrial nitrification, aquatic acidification, aquatic eutrophication, and land occupation. Both human health and ecosystem quality damage categories cover the photochemical oxidation midpoint category. Climate change damage only includes a global warming category. Resources damage considers non-renewable energy and mineral extraction midpoint categories. category. To calculate the impact of any category, a characterization factor of each chemical's emission for each specific impact category must be considered. Damage characterization factors of the midpoint characterization potential have been mentioned in [35], where authors considered emissions per person per year for western Europe as normalization factors for damage categories. SimaPro 7 software package has been utilized in this study to conduct IMPACT 2002+ methodology for three cooling strategies. Besides, Table 1 presents normalization factors for the damage categories used in SimaPro 7.

Functional Unit
A functional unit acts as a reference that is used to relate the system's inputs and outputs. This enables comparisons between different systems. The functional unit must be clearly defined, and it is considered to be a key element for conducting LCAs. Two functional units are presented in this paper. The first functional unit is used when conducting LCAs for CFs, while the second one is used when conducting LCAs for cooling strategies.
The first functional unit was selected based on the available information in literature considering quantities of preparing nano-cutting fluid. Padmini et al. [36] considered using 100 mL to prepare different types of nanofluid. Different vegetable oils and MoS 2 nanoparticles were included during the preparation process. Produced fluids were presented with different contents of nano-additives, ranging between 0 vol% to 1 vol%, with increments of 0.25%. Machining performances of each nano-additive fluid were examined. Wusiman et al. [17] employed 50 mL of distilled water with 0.5 wt.% of CNT. Hegab [37] investigated the usage of Al 2 O 3 and MWCNTs when machining difficult-to-cut materials. Two percent and 4% of nano-additives were added with rapeseed oil during the preparation of 100 mL of nanofluids. Zhu el al. [38] investigated and discussed the thermal conductivity characteristics of water with 0.1% Al 2 O 3 of nano-additives in 150 mL volumetric content. Das et al. [19] utilized 2.5 g of Al 2 O 3 with 500 mL of base fluid of radiator coolant and distilled water with a ratio of 1:4. Based on the reviewed literature, the functional unit in this work is based on CFs in a volumetric content of 150 mL. For the NCFs, 1 wt.% of the nano-additives is considered as the functional unit. The second functional unit used in this work is an AISI 4043 industrial machined bolt. The bolt has a length of 200 mm and a diameter of 42 mm. Besides, the used bolt is manufactured by local small to medium-sized enterprises (SME) in Western Australia [39]. The conducted LCAs in this paper cover the machining a single AISI 4043 bolt.

Results and Discussion
Results of LCAs for 12 CFs and 12 LCAs cooling techniques are presented, compared, and discussed in this section. Executed LCAs can be categorized into two parts. The first part covers the LCAs of the considered CFs, while the second part covers the whole LCAs of the considered cooling strategies. The first part included LCAs for two conventional CFs, two vegetable oils, and eight NCFs. The NCFs used were combinations of four nano-additives and the used vegetable oils. Each vegetable oil was used as the base fluid for each nano-additive. In addition, LCAs of CFs were based on a volumetric amount of 150 mL. The second part was an expansion of the first category, which included CFs' cooling techniques. Cooling techniques included in the second category were: flood cooling for two conventional CFs, MQL cooling technique for two vegetable oils, and MQL cooling technique for eight NCFs. Finally, CFs cooling strategies' LCAs were based on machining a single AISI 4043 bolt.

Life Cycle Assessment (LCA) of Conventional Cutting Fluids
LCAs were conducted for two types of conventional CFs: pure mineral oil and a benchmark CF that consists of mineral oils and petroleum feedstock surfactants [40]. The LCA of the pure mineral oil includes offshore extraction processes of crude oil, primary refining processes, transportation of the extracted oil to the coast via pipelines, transportation of the crude oil to the refinery, and the refining processes of the oil. Benchmark CF consists of 0.75 wt.% of petroleum oil, 0.3 wt.% anionic surfactant of sodium petroleum sulfonate, 0.4% of nonionic surfactant of diisoproponalamine, and 98.55 wt.% water. The benchmark CF is considered as a microemulsion of petroleum oil in water. The sodium petroleum sulfonate is obtained from the reaction of sulfur with linear alkyl, and diisoproponalamine is obtained from ammonia and propylene oxide. Figure 2a displays conducted LCAs' midpoint impacts of 150 mL of pure petroleum oil and benchmark CF, respectively. Petroleum oil presents a high impact on non-renewable energy (i.e., resources damage category) and lower impacts for the other midpoints. Results for benchmark cutting fluid displays high impacts for ionizing radiation, aquatic ecotoxicity, and non-renewable energy and lower impacts for the other midpoints. Non-renewable energy has the highest midpoint impact at 13 MJ primary in petroleum oil's LCA, the same midpoint impact in benchmark CF is lower at 0.19 MJ primary. Aquatic ecotoxicity has the highest midpoint impact at 0.191 kg TEG water in benchmark CF's LCA, the same midpoint impact in petroleum oil's LCA is higher at 0.490 kg TEG water.

LCA of Vegetable Oils
LCAs were conducted for two types of vegetable oils: rapeseed oil and soybean oil. LCAs for both vegetable oils include material extraction and manufacturing stages. The material extraction stage includes the following processes: Soil cultivation, sowing, weed control, fertilization, harvesting, and grain drying. In this study, the allocation factor for 1 kg of rapeseed oil is 74.3% of oil and 25.7% of the meal, while the allocation factor for 1 kg of soybean oil is 34.5% of oil and 65.5% of the meal. Both vegetable oils have different densities at 22 • C where the rapeseed oil has a density of 913.3 kg/m 3 , and soybean oil has a density of 915.7 kg/m 3 [41]. The 150 mL of each oil-weighted differently due to their different densities. Rapeseed seeds weighed 0.2211 kg and soybean seeds weighed 0.4725 kg. The manufacturing stage includes seeds (i.e., oil mill) processing and purification processes of the extracted oil. Figure 2b displays conducted LCAs midpoint impacts of 150 mL of rapeseed oil and soybean oil, respectively. Rapeseed oil presents high impacts for aquatic ecotoxicity and terrestrial ecotoxicity and lower impacts for the other midpoints. Soybean oil displays higher impacts on non-renewable energy, land occupation, and ionizing radiation. Lower impacts are displayed for the other midpoints. Terrestrial ecotoxicity has the highest midpoint impact at 741 kg TEG soil in rapeseed oil's LCA, the same midpoint impact in soybean oil is lower at −87.1 kg TEG soil. Ionizing radiation has the highest midpoint impact at 4.25 Bq C-14 eq in soybean oil's LCA, the same midpoint impact in rapeseed oil's LCA is higher at 6.81 Bq C-14 eq.

LCA of Nano-Cutting Fluids
LCAs of NCFs consider the impacts of base fluids (vegetable oils), nano-additive synthesis processes, and the fluid's preparation processes. Soybean oil and rapeseed oil are used as base fluids. LCAs for base fluids have already been conducted in the previous section. LCAs for nano-additives are presented in Section 3.3.1. Four nano-additives are presented: Molybdenum disulfide (MoS2), multi-walled carbon nanotubes (MWCNTs), and titanium dioxide (TiO2), and aluminum oxide (Al2O3). NCFs' preparation was the top-down (two-step method) preparation process approach.

LCA of Nano-Cutting Fluids
LCAs of NCFs consider the impacts of base fluids (vegetable oils), nano-additive synthesis processes, and the fluid's preparation processes. Soybean oil and rapeseed oil are used as base fluids.
LCAs for base fluids have already been conducted in the previous section. LCAs for nano-additives are presented in Section 3.3.1. Four nano-additives are presented: Molybdenum disulfide (MoS 2 ), multi-walled carbon nanotubes (MWCNTs), and titanium dioxide (TiO 2 ), and aluminum oxide (Al 2 O 3 ). NCFs' preparation was the top-down (two-step method) preparation process approach.

Nanoparticle Synthesis
Nanoparticle synthesis processes are available through physical, chemical, and biological methods. Detailed synthesis processes for nanoparticles were included in the presented results of the conducted SimaPro software LCAs. Information on the synthesis processes was gathered from the literature. Gathered information for synthesis processes including the processes inputs/outputs, and energy consumption.
Molybdenum Disulfide (MoS 2 ) Nanoparticle Synthesis MoS 2 nanoparticle's LCA used in this paper is mainly based on the wet synthesis technique used in previous work [24]. The aqueous solution of citric acid is utilized with ammonium molybdate to result in molybdenum with the citric acid solution. Then, ammonium sulfide is added to form MoS 2 . The functional unit used is 1 g of the MoS 2 powder. Inputs and outputs quantities required to obtain a 1 g of the MoS 2 powder are presented in Table 2. In addition, the electricity used during mixing, spinning, and drying processes are presented in Table 2. The chemicals and electricity supply inventory database in SimaPro software has been utilized to conduct the LCA of MoS 2 nanoparticles. Figure 3a displays the midpoint impact categories of MoS 2 nanoparticle's LCA. Aquatic ecotoxicity has the highest impact midpoint in comparison to other midpoints at 37.8 kg TEG water.

Multi-Walled Carbon Nanotubes (MWCNTs) Nanoparticles Synthesis
MWCNTs nanoparticle's LCA used in this paper is based on the catalytic chemical vapor deposition (cCVD) used in previous work [42]. This synthesis process is used to obtain 99% pure organized MWCNTs. The functional unit used is the quantity of the batch production, which is 300 mg. Inputs and outputs quantities required to obtain 300 mg of MWCNTs are presented in Table 3. Inputs consist of 0.2 g ferrocene dissolving in 10 mL of toluene to form a solution. Then, a syringe pump is used to inject the formed solution into a quartz tube at a rate of 10 mL/h and at 790 • C. A quartz tube is used during the injection process and Ar and H 2 are presented at 450 sccm and 50 sccm, respectively. In addition, Table 3 displays the total energy consumption which includes energies consumed in heating the furnace, mechanical movements of the syringe pump Ar and mass H 2 flow controllers. In addition, the LCA includes the production stage of the catalyst, feedstock preparation, heating of the furnace, the reaction period, furnace cooling, and purification. Figure 3a displays the midpoint impact categories of MWCNTs nanoparticle's LCA. Aquatic ecotoxicity has the highest impact midpoint in comparison to other midpoints at 395 kg TEG water.    TiO 2 nanoparticle's LCA used in this paper followed the Altairnano hydrochloride synthesis processes used in previous work [43]. Like sulphate process, ilmenite ore is the main feedstock in the Altairnano hydrochloride process. Ilmenite is introduced into the unit of digestion during the existence of concentrated hydrochloric acid. Then, a small amount of iron powder is added to the mixture and the iron chloride is separated from the output steam and filtration is conducted. After filtration, the trialkyl phosphine oxide is used as solvent extraction and the sodium hydroxide is used to remove the chloride iron for the product steam. Finally, TiO 2 nanoparticles are obtained by the reaction of titanium oxychloride with water. During the synthesis of TiO 2 nanoparticles, the dehydration process consumes the highest amounts of methane. The functional unit used 1 kg of the TiO 2 nanoparticles. Inputs and outputs quantities are required to obtain 1 kg of the TiO 2 nanoparticles are presented in Table 4. Figure 3b displays the midpoint impact categories of TiO 2 nanoparticle's LCA. Aquatic ecotoxicity has the highest impact midpoint in comparison to other midpoints at 4510.  [44]. The sol-gel synthesis technique is known as chemical solution deposition, it is a wet-chemical technique which widely used to produce Al 2 O 3 nanoparticles. The first step to produce Al 2 O 3 nanoparticles requires preparing a solution of distilled water with citric acid and aluminum nitrite. Then, the solution is placed for 2 h over the thermal plate with magnetic agitation at 60 • C. To maintain constant agitation, the solution is heated for additional 1.5 h at 80 • C. After that, an oven is utilized to evaporate the water content from the obtained gel. Finally, the produced gel is burnt for 2 h at 1000 • C. The functional unit used is 1 g of Al 2 O 3 nanoparticles. Inputs and outputs quantities required to obtained 1 g of the Al 2 O 3 nanoparticles are presented in Table 5. Figure 3b displays the midpoint impact categories of Al 2 O 3 nanoparticle's LCA. Aquatic ecotoxicity has the highest impact midpoint in comparison to other midpoints at 3480 kg TEG water.

. Preparation Process of Nano-Cutting Fluids
NCFs preparation process is the process where nano-sized particles are mixed with a base fluid. Proper execution of this process results in fluids with desirable properties such as homogeneity, physical and chemical stability, durability, and dispersibility. There are two commonly used approaches for the preparation process of NCFs: The bottom-up and the top-down approaches [45]. The bottom-up approach is known as the one-step method, while the top-down approach is known as the two-step method. NCFs presented in this paper are prepared via the two-step method, similar to the procedure followed in [18]. During the one-step method, preparation of nanoparticles and their suspension into the base fluid are executed at the same time. The two-step method is more commonly used and is cheaper, which produces NFCs in two steps. The first step involves producing nanoparticles via synthesis techniques. Then, during the second step nanoparticles are dispersed into the base fluid. There are common apparatuses that are used during the dispersion of solid nanoparticles into the base fluid, such as magnetic stirrers, ultrasonic bath, homogenizers, high-shear mixers, and bead mills.
According to the first functional unit, the preparation volume for the nano-fluids is 150 mL. This volume consists of three components: vegetable oil (base fluid), nano-additives (1 wt.%), and surfactant of sodium dodecyl sulfate (0.3 g). The purpose of using a surfactant is to increase the NCF's stability. The two steps method followed in [18] utilized an ultrasonic device (AQUASONIC-50HT) for 3 h at 60 • C. Then, a magnetic stirrer (Hot Plate Stirrer-3073-21) is used for 30 min to assure for complete dispersion of nanoparticles within the base fluid. The ultrasonic device consumed 1.5 kWh and the magnetic stirrer consumed 0.09 kWh.

LCA Comparison of Cutting Fluids
LCAs are broken down into LCAs for two conventional CFs (i.e., benchmark and petroleum oil), two vegetable oils (i.e., soybean oil and rapeseed oil), and eight NCFs (i.e., soybean-MoS 2 -NCF, soybean-MWCNTs-NCF, soybean-TiO 2 -NCF, soybean-Al 2 O 3 -NCF, rapeseed-MoS 2 -NCF, rapeseed-MWCNTs-NCF, rapeseed-TiO 2 -NCF, rapeseed-Al 2 O 3 -NCF). Figure 4 presents the damage categories in the normalized unit (µPt) for two conventional CFs, soybean vegetable oil, and four soybean vegetable oil-based NCFs. Figure 5 presents damage categories in µPt for two conventional CFs, rapeseed vegetable oil, and four rapeseed vegetable oil-based NCFs. Four damage categories are included in Figures 4 and 5, which are human health, ecosystem quality, climate change, and resources. Presented damage categories are normalized for each CF. Figures 4 and 5 indicate that benchmark CF has the lowest impacts on all damage categories, followed by petroleum oil, in comparison to the other CFs. Petroleum oil has a higher impact on the resource's category, because of its non-renewable energy sources. Soybean oil has a high impact on the ecosystem quality category, because of the high consumptions of water during soil's cultivation and fertilization. However, it has a medium level impact on the resources category, due to its impact on the non-renewable energy midpoint category as shown in Figure 2. Rapeseed oil has a higher impact on human health, ecosystem quality, and resource categories, in comparison to soybean oil, while it has a higher allocation factor. However, it has a slightly smaller impact on the climate change category, in comparison to soybean oil. In regards to NCFs, Soybean-TiO 2 has the lowest impacts on all damage categories, in comparison to the other seven NCFs, except for rapeseed-TiO 2 -NCF and rapeseed-MoS 2 -NCF, which have lower impact factors in the climate change category. Soybean-TiO 2 has a slightly larger impact than soybean oil CF. Moreover, on average between all damage categories, rapeseed-NCFs have higher impacts on the damage categories than soybean-NCFs, due to the higher impact their base rapeseed oil has, in comparison to the soybean oil.  Figure 6 displays a single score for each analyzed CF. Out of the 12 investigated CFs, conventional benchmark CF presents the lowest impact on the four damage categories, and rapeseed-Al2O3-NCF presents the highest impact. Additionally, vegetable oils have higher impact scores than conventional CFs due to due to their cultivation and the fertilization processes, where high amounts of water are used in the soil. However, in terms of the resource's category, conventional petroleum CF has a higher impact than vegetable oils, due to its extraction method from non-renewable energy sources. In comparison with the vegetable oils, rapeseed oil has higher impact scores than soybean oil, mainly because of its higher score within the ecosystem quality category. Moreover, NCFs have impacts higher than vegetable oils that range between 41% to 470%. TiO2 and MoS2 NCFs impacts were lower than the MWCNTs-NCF and Al2O3-NCF. NCFs have the highest impacts out of all other CFs, due to their energy-intensive synthesis and preparation processes.
Nanoparticles synthesis processes are energy-intensive processes. Energy consumed by nanoparticles synthesis processes depends on three parameters: material of nanoparticles, synthesis process used, and quality of the output particles. Used synthesis processes presented a significant impact on NCFs damage impact scores. Mainly because NCFs are produced in a laboratory (small scale) setting, while conventional CFs are produced in industrial (large scale) settings. Laboratory settings mainly focus on researching and developing NCFs. Industrial settings focus on optimizing CFs production techniques which reduce the energy required per unit volume.   Figure 6 displays a single score for each analyzed CF. Out of the 12 investigated CFs, conventional benchmark CF presents the lowest impact on the four damage categories, and rapeseed-Al2O3-NCF presents the highest impact. Additionally, vegetable oils have higher impact scores than conventional CFs due to due to their cultivation and the fertilization processes, where high amounts of water are used in the soil. However, in terms of the resource's category, conventional petroleum CF has a higher impact than vegetable oils, due to its extraction method from non-renewable energy sources. In comparison with the vegetable oils, rapeseed oil has higher impact scores than soybean oil, mainly because of its higher score within the ecosystem quality category. Moreover, NCFs have impacts higher than vegetable oils that range between 41% to 470%. TiO2 and MoS2 NCFs impacts were lower than the MWCNTs-NCF and Al2O3-NCF. NCFs have the highest impacts out of all other CFs, due to their energy-intensive synthesis and preparation processes.
Nanoparticles synthesis processes are energy-intensive processes. Energy consumed by nanoparticles synthesis processes depends on three parameters: material of nanoparticles, synthesis process used, and quality of the output particles. Used synthesis processes presented a significant impact on NCFs damage impact scores. Mainly because NCFs are produced in a laboratory (small scale) setting, while conventional CFs are produced in industrial (large scale) settings. Laboratory settings mainly focus on researching and developing NCFs. Industrial settings focus on optimizing CFs production techniques which reduce the energy required per unit volume.  Figure 6 displays a single score for each analyzed CF. Out of the 12 investigated CFs, conventional benchmark CF presents the lowest impact on the four damage categories, and rapeseed-Al 2 O 3 -NCF presents the highest impact. Additionally, vegetable oils have higher impact scores than conventional CFs due to due to their cultivation and the fertilization processes, where high amounts of water are used in the soil. However, in terms of the resource's category, conventional petroleum CF has a higher impact than vegetable oils, due to its extraction method from non-renewable energy sources.
In comparison with the vegetable oils, rapeseed oil has higher impact scores than soybean oil, mainly because of its higher score within the ecosystem quality category. Moreover, NCFs have impacts higher than vegetable oils that range between 41% to 470%. TiO 2 and MoS 2 NCFs impacts were lower than the MWCNTs-NCF and Al 2 O 3 -NCF. NCFs have the highest impacts out of all other CFs, due to their energy-intensive synthesis and preparation processes.
Nanoparticles synthesis processes are energy-intensive processes. Energy consumed by nanoparticles synthesis processes depends on three parameters: material of nanoparticles, synthesis process used, and quality of the output particles. Used synthesis processes presented a significant impact on NCFs damage impact scores. Mainly because NCFs are produced in a laboratory (small scale) setting, while conventional CFs are produced in industrial (large scale) settings. Laboratory settings mainly focus on researching and developing NCFs. Industrial settings focus on optimizing CFs production techniques which reduce the energy required per unit volume. Sustainability 2020, 12, x FOR PEER REVIEW 15 of 22 Figure 6. The single score of all damage categories for the considered cutting fluids.

LCA of Cooling Strategies
Twelve LCAs cooling techniques are presented, compared, and discussed in this section. LCAs are broken down into LCAs for TFC and MQL cooling techniques. TFC was considered for two conventional CFs (i.e., benchmark and petroleum oil). MQL cooling technique was considered for 10 CFs which are; two vegetable oils (i.e., soybean oil and rapeseed oil), and eight NCFs (i.e., soybean-MoS2-NCF, soybean-MWCNTs -NCF, soybean-TiO2-NCF, soybean-Al2O3-NCF, rapeseed-MoS2-NCF, rapeseed-MWCNTs-NCF, rapeseed-TiO2-NCF, rapeseed-Al2O3-NCF). The LCAs' functional unit that is adapted to compare cooling technique is the machining of a single M42 200 AISI 4043 bolt. The industrial bolt was presented and discussed, in the small to medium-sized enterprise (SME) in Western Australia [39]. Also, SME provided inputs and outputs data, which are utilized during the execution of this works LCAs. SME provided inventory information for machining a M42 200 AISI 4043 bolt. The information included: cutting energy, pumping energy, compressor energy, the quantity of coolant, and tool weight reduction. SME cutting energy and tool weight reduction information were excluded during the development of the presented LCAs, because their provided amounts are influenced by CFs [39] used. In regard to the pumping energy, TFC consumed 5.124 × 10 5 kWh per bolt, while the MQL cooling technique consumed 0.051 kWh during CFs pumping and air compressing. Although MQL uses approximately 7.6% of the amount of cutting fluid in TFC, it consumes considerably higher energy than TFC, due to the energy consumed by MQL's air compressor [39]. Figures 7 and 8 display the impacts of CF strategies on four damage categories, which are: human health, ecosystem quality, climate change, and resources. Damage categories are provided in µPt. Based on the displayed impacts, the TFC technique of benchmark CF has the lowest impact on the damage categories, followed by the TFC technique of petroleum oil. TFC of petroleum oil has a noticeably high impact on the resource category, because of the used extraction method of this type of CF. MQL technique of rapeseed oil displays a higher impact on the human health category by 14% and ecosystem quality category by 58%, than the MQL technique of soybean oil. However, in terms of climate change and resource categories both vegetable oil MQL techniques present similar impacts. Besides, there is a slight increase in the damage categories when comparing MQL-NFCs to MQL of vegetable oils. Slight increases range within: 2%-37%, 0%-10%, 3%-37%, and 3%-37% on human health, ecosystem quality, climate change, and resources categories, respectively.

LCA of Cooling Strategies
Twelve LCAs cooling techniques are presented, compared, and discussed in this section. LCAs are broken down into LCAs for TFC and MQL cooling techniques. TFC was considered for two conventional CFs (i.e., benchmark and petroleum oil). MQL cooling technique was considered for 10 CFs which are; two vegetable oils (i.e., soybean oil and rapeseed oil), and eight NCFs (i.e., soybean-MoS 2 -NCF, soybean-MWCNTs -NCF, soybean-TiO 2 -NCF, soybean-Al 2 O 3 -NCF, rapeseed-MoS 2 -NCF, rapeseed-MWCNTs-NCF, rapeseed-TiO 2 -NCF, rapeseed-Al 2 O 3 -NCF). The LCAs' functional unit that is adapted to compare cooling technique is the machining of a single M42 × 200 AISI 4043 bolt. The industrial bolt was presented and discussed, in the small to medium-sized enterprise (SME) in Western Australia [39]. Also, SME provided inputs and outputs data, which are utilized during the execution of this works LCAs. SME provided inventory information for machining a M42 × 200 AISI 4043 bolt. The information included: cutting energy, pumping energy, compressor energy, the quantity of coolant, and tool weight reduction. SME cutting energy and tool weight reduction information were excluded during the development of the presented LCAs, because their provided amounts are influenced by CFs [39] used. In regard to the pumping energy, TFC consumed 5.124 × 10 5 kWh per bolt, while the MQL cooling technique consumed 0.051 kWh during CFs pumping and air compressing. Although MQL uses approximately 7.6% of the amount of cutting fluid in TFC, it consumes considerably higher energy than TFC, due to the energy consumed by MQL's air compressor [39]. Figures 7 and 8 display the impacts of CF strategies on four damage categories, which are: human health, ecosystem quality, climate change, and resources. Damage categories are provided in µPt. Based on the displayed impacts, the TFC technique of benchmark CF has the lowest impact on the damage categories, followed by the TFC technique of petroleum oil. TFC of petroleum oil has a noticeably high impact on the resource category, because of the used extraction method of this type of CF. MQL technique of rapeseed oil displays a higher impact on the human health category by 14% and ecosystem quality category by 58%, than the MQL technique of soybean oil. However, in terms of climate change and resource categories both vegetable oil MQL techniques present similar impacts. Besides, there is a slight increase in the damage categories when comparing MQL-NFCs to MQL of vegetable oils. Slight increases range within: 2%-37%, 0%-10%, 3%-37%, and 3%-37% on human health, ecosystem quality, climate change, and resources categories, respectively.   Figure 9 displays a single score for each analyzed cooling strategy. Out of the 12 investigated cooling strategies, TFC of benchmark CF presents the lowest impact on the four damage categories, and MQL of rapeseed-Al2O3-NCF presents the highest impact. Also, MQL of vegetable oils has higher impact scores than TFC of conventional CFs, even though they have similar damage impact scores on the resource category. In regard to vegetable oils, MQL of rapeseed oil has a higher impact than MQL of soybean oil by 11.9%. Additionally, damage impact scores increase by 14.13% by the introduction of nanoparticles. MQL NCFs of rapeseed oils have higher impact scores than the MQL NCFs of soybean oils, because of their base fluids impact scores. Due to the small used quantity of CFs and high used energy in MQL technique, the ecosystem quality scores are relatively small in comparison with scores of the other damage categories. Besides, it can be seen that the small quantity of cutting fluid associated with MQL reduce the increasing percentage of the total score of MQL of NFCs by the range of 3.7% to 35.4% in comparison with MQL of vegetable oils.   Figure 9 displays a single score for each analyzed cooling strategy. Out of the 12 investigated cooling strategies, TFC of benchmark CF presents the lowest impact on the four damage categories, and MQL of rapeseed-Al2O3-NCF presents the highest impact. Also, MQL of vegetable oils has higher impact scores than TFC of conventional CFs, even though they have similar damage impact scores on the resource category. In regard to vegetable oils, MQL of rapeseed oil has a higher impact than MQL of soybean oil by 11.9%. Additionally, damage impact scores increase by 14.13% by the introduction of nanoparticles. MQL NCFs of rapeseed oils have higher impact scores than the MQL NCFs of soybean oils, because of their base fluids impact scores. Due to the small used quantity of CFs and high used energy in MQL technique, the ecosystem quality scores are relatively small in comparison with scores of the other damage categories. Besides, it can be seen that the small quantity of cutting fluid associated with MQL reduce the increasing percentage of the total score of MQL of NFCs by the range of 3.7% to 35.4% in comparison with MQL of vegetable oils.  Figure 9 displays a single score for each analyzed cooling strategy. Out of the 12 investigated cooling strategies, TFC of benchmark CF presents the lowest impact on the four damage categories, and MQL of rapeseed-Al 2 O 3 -NCF presents the highest impact. Also, MQL of vegetable oils has higher impact scores than TFC of conventional CFs, even though they have similar damage impact scores on the resource category. In regard to vegetable oils, MQL of rapeseed oil has a higher impact than MQL of soybean oil by 11.9%. Additionally, damage impact scores increase by 14.13% by the introduction of nanoparticles. MQL NCFs of rapeseed oils have higher impact scores than the MQL NCFs of soybean oils, because of their base fluids impact scores. Due to the small used quantity of CFs and high used energy in MQL technique, the ecosystem quality scores are relatively small in comparison with scores of the other damage categories. Besides, it can be seen that the small quantity of cutting fluid associated with MQL reduce the increasing percentage of the total score of MQL of NFCs by the range of 3.7% to 35.4% in comparison with MQL of vegetable oils. Sustainability 2020, 12, x FOR PEER REVIEW 17 of 22 Figure 9. The single score of all damage categories for the considered cooling strategies.

Overall Evaluation of the Cooling Strategies
This subsection provides an overall qualitative evaluation of the considered cooling strategies. This evaluation covers the conducted cradle-to-gate LCA of cooling strategies, human health impacts during the machining process, environmental and human health impacts at end of life stage, as well as the machining performance as shown in Figure 10. The first three evaluation factors are based on the "lower-the-better criterion," while the fourth factor is based on the higher-the-better criterion. Besides, the evaluation has been conducted over three levels which are high, medium, and low.
Cradle-to-gate LCA: TFC of conventional CFs displayed the lowest impact on the four damage categories. Accordingly, this strategy has a low level of impact on cradle-to-gate LCA. The MQL of vegetable oils strategy achieved a medium level in this evaluation factor which presented higher impact scores compared to TFC of conventional CFs, however, they displayed close damage impact scores on the resource category. In addition, MQL of NCFs showed higher impact scores than MQL of their base fluids, and that is mainly due to the used synthesis processes. Thus, this strategy displays a high score in cradle-to-gate LCA as shown in Figure 10.
Human health impacts during the machining process: The usage of conventional CFs displays high human health impacts due to the high rates of occupational infections related to the development of cancer, respiratory distress, and skin disorders. Besides, the big quantity of the used conventional CFs in the TFC cooling strategy leads to sever impact for the human health factor for the machining performance factor as shown in Figure 10. Vegetable oils are organic products extracted from plant seeds and consist of triglycerides, which are long chains of fatty acids combined with glycerol, and therefore, that makes the exposure of these fluids safe to the operators in the machining process. Utilizing the MQL technique for applying vegetable oils leads to low impacts of this cooling strategy. For the nano-CFs, nanosized particles are more toxic than larger particles, because nano-particles are smaller in size in comparison to human body cells, and this feature allows the body cells to capture these particles and negatively impact the human health. Although the vaporization of NCFs during MQL leads to small amounts of particles to be airborne, this cooling strategy showed a high level in human health factor at the machining process as shown in Figure 10 due to lack of information about exposure assessment of these particles.

Environmental and human health impacts at end-of-life stage:
Conventional cutting fluids are made of combinations of oils, water, and other additives. Adding antimicrobial (biocides) and antifoam additives to control these problems leads to critical environmental problems due to the disposal process of these additives. Besides, the disposal of used CFs which included other additives of EP and corrosion inhibiting is only allowed in special incineration sites due to the generated toxins [2]. Consequently, the disposal of used conventional cutting fluids has been linked to pollution in water systems, air quality, and soil. Thus, TFC of conventional CFs displayed high impacts on environmental and human health impacts at end-of-life stage. Vegetable oils satisfy the aspects of CFs sustainability and are considered to be renewable, biodegradable, and can easily be degraded in soils due to their limited toxicity. Consequently, the MQL of vegetable oils achieved a low level of

Overall Evaluation of the Cooling Strategies
This subsection provides an overall qualitative evaluation of the considered cooling strategies. This evaluation covers the conducted cradle-to-gate LCA of cooling strategies, human health impacts during the machining process, environmental and human health impacts at end of life stage, as well as the machining performance as shown in Figure 10. The first three evaluation factors are based on the "lower-the-better criterion," while the fourth factor is based on the higher-the-better criterion. Besides, the evaluation has been conducted over three levels which are high, medium, and low.
properties which are needed at low cutting speeds machining processes. Oil-based CFs (i.e., pure petroleum oil) act as lubricants which decrease the coefficient of friction decreasing the generated forces. While the aqueous-based conventional CFs (i.e., benchmark CF) show excellent cooling properties and are preferably used in operations in which high heat is generated and relatively high cutting speeds are required. Conventional CFs offer a wide range of lubrication and cooling properties that are needed for different types of machining processes. Accordingly, the conventional CFs have a high-performance level as shown in Figure 10. Vegetable oils show desirable lubrication properties for the machining processes. Their lubrication properties come from their triglyceride structure and their long and polar fatty acid chains which are beneficial to decrease the friction between the tool-workpiece. Furthermore, vegetable oils have a higher flash point which makes vegetable oils less flammable. However, vegetable oils have their drawbacks: they are less reliable under high loads, and they have poor rust control and low thermal stability. Consequently, MQL of vegetable oils displays a medium level in machining performance. NCFs have a significant increase in the friction reduction between the contact surfaces and also reduce cutting forces and tool wear. Besides, they have many special effects such as bearing effect, mending effect, protective film effect, and polishing effect. The introduction of nanoparticles can improve the thermal conductivity of the vegetable oils, which enhance the cooling properties and thermal stability of the base oil, and allow to use a small quantity of CFs by MQL technique. Therefore, MQL of NCFs achieved a high level in the machining performance factor as shown in Figure 10.
According to the overall evaluation of the considered cooling strategies, the MQL of vegetable oils is the most sustainable cooling strategy for machining processes with associated low/medium cutting temperature. While MQL of TiO2 and MoS2 NCFs are the sustainable cooling strategy for machining processes associated with high cutting temperature due to their enhanced lubrication and cooling properties and small impacts of their dispersed nanoparticles. Accordingly, MQL of TiO2 and MoS2 NCFs cooling strategies with high personal protection procedures to the operators are considered as sustainable strategies for machining difficult-to-cut material which have low thermal conductivity and high-speed machining processes.  Cradle-to-gate LCA: TFC of conventional CFs displayed the lowest impact on the four damage categories. Accordingly, this strategy has a low level of impact on cradle-to-gate LCA. The MQL of vegetable oils strategy achieved a medium level in this evaluation factor which presented higher impact scores compared to TFC of conventional CFs, however, they displayed close damage impact scores on the resource category. In addition, MQL of NCFs showed higher impact scores than MQL of their base fluids, and that is mainly due to the used synthesis processes. Thus, this strategy displays a high score in cradle-to-gate LCA as shown in Figure 10.
Human health impacts during the machining process: The usage of conventional CFs displays high human health impacts due to the high rates of occupational infections related to the development of cancer, respiratory distress, and skin disorders. Besides, the big quantity of the used conventional CFs in the TFC cooling strategy leads to sever impact for the human health factor for the machining performance factor as shown in Figure 10. Vegetable oils are organic products extracted from plant seeds and consist of triglycerides, which are long chains of fatty acids combined with glycerol, and therefore, that makes the exposure of these fluids safe to the operators in the machining process. Utilizing the MQL technique for applying vegetable oils leads to low impacts of this cooling strategy. For the nano-CFs, nanosized particles are more toxic than larger particles, because nano-particles are smaller in size in comparison to human body cells, and this feature allows the body cells to capture these particles and negatively impact the human health. Although the vaporization of NCFs during MQL leads to small amounts of particles to be airborne, this cooling strategy showed a high level in human health factor at the machining process as shown in Figure 10 due to lack of information about exposure assessment of these particles.
Environmental and human health impacts at end-of-life stage: Conventional cutting fluids are made of combinations of oils, water, and other additives. Adding antimicrobial (biocides) and antifoam additives to control these problems leads to critical environmental problems due to the disposal process of these additives. Besides, the disposal of used CFs which included other additives of EP and corrosion inhibiting is only allowed in special incineration sites due to the generated toxins [2]. Consequently, the disposal of used conventional cutting fluids has been linked to pollution in water systems, air quality, and soil. Thus, TFC of conventional CFs displayed high impacts on environmental and human health impacts at end-of-life stage. Vegetable oils satisfy the aspects of CFs sustainability and are considered to be renewable, biodegradable, and can easily be degraded in soils due to their limited toxicity. Consequently, the MQL of vegetable oils achieved a low level of impact in this evaluation factor. For the NCFs, they need special types of separation and filtration processes before disposal and recycling. These processes are utilized to separate the NCF to its components of vegetable oil and nanoparticles and lead to eco-friendly disposal of the base-fluid of vegetable oils. Besides, these processes drive to reusing the separated nanoparticles instead of disposing of the whole nano-fluids. Therefore, the MQL of NCFs achieved a medium level of impact at the end of life stage.
Machining performance: The oil-based type of conventional CFs shows excellent lubrication properties which are needed at low cutting speeds machining processes. Oil-based CFs (i.e., pure petroleum oil) act as lubricants which decrease the coefficient of friction decreasing the generated forces. While the aqueous-based conventional CFs (i.e., benchmark CF) show excellent cooling properties and are preferably used in operations in which high heat is generated and relatively high cutting speeds are required. Conventional CFs offer a wide range of lubrication and cooling properties that are needed for different types of machining processes. Accordingly, the conventional CFs have a high-performance level as shown in Figure 10. Vegetable oils show desirable lubrication properties for the machining processes. Their lubrication properties come from their triglyceride structure and their long and polar fatty acid chains which are beneficial to decrease the friction between the tool-workpiece. Furthermore, vegetable oils have a higher flash point which makes vegetable oils less flammable. However, vegetable oils have their drawbacks: they are less reliable under high loads, and they have poor rust control and low thermal stability. Consequently, MQL of vegetable oils displays a medium level in machining performance. NCFs have a significant increase in the friction reduction between the contact surfaces and also reduce cutting forces and tool wear. Besides, they have many special effects such as bearing effect, mending effect, protective film effect, and polishing effect. The introduction of nanoparticles can improve the thermal conductivity of the vegetable oils, which enhance the cooling properties and thermal stability of the base oil, and allow to use a small quantity of CFs by MQL technique. Therefore, MQL of NCFs achieved a high level in the machining performance factor as shown in Figure 10.
According to the overall evaluation of the considered cooling strategies, the MQL of vegetable oils is the most sustainable cooling strategy for machining processes with associated low/medium cutting temperature. While MQL of TiO 2 and MoS 2 NCFs are the sustainable cooling strategy for machining processes associated with high cutting temperature due to their enhanced lubrication and cooling properties and small impacts of their dispersed nanoparticles. Accordingly, MQL of TiO 2 and MoS 2 NCFs cooling strategies with high personal protection procedures to the operators are considered as sustainable strategies for machining difficult-to-cut material which have low thermal conductivity and high-speed machining processes.

Conclusions and Future Work
Available works in the literature have investigated the usage of MQL-NCF and presented its benefits in comparison to TFC techniques, in terms of machining process performance. However, environmental and human health impacts of NCFs and MQL have not been discussed or compared to TFC techniques. This paper presented and discussed 24 cradle-to-gate LCAs, which mainly focused on CFs and cooling strategies' impact on the environment and human personal health. Besides, it offers an overall evaluation of the considered cooling strategies through human health impacts at the machining process, environmental and human health impacts at end-of-life stage, and machining process performance. In terms of CFs, conventional benchmark CF presented the lowest impact scores on all damage categories, pure petroleum oil displayed a medium impact on the resources and very small impact on the other damage categories. Soybean and rapeseed oils showed high impacts on the ecosystem quality category. NCFs displayed higher impacts scores than their base vegetable oils that ranged between 41% to 470%, based on the used nano-additives, because of their impactful synthesis processes. TiO 2 and MoS 2 NCFs presented slightly higher impacts than the impacts of their base vegetable oils. However, TiO 2 and MoS 2 NCFs impacts were lower than the MWCNTs-NCF and Al 2 O 3 -NCF. Besides, NCFs of rapeseed oil displayed higher impacts than NFCs of soybean oil, because rapeseed oil had a higher impact score than soybean oil. In terms of cooling strategies, TFC of benchmark CF displayed the lowest impact on the four damage categories. The highest impact scores were displayed by MQL of rapeseed-Al 2 O 3 -NCF. In addition, MQL of rapeseed oil had a higher impact than MQL of soybean oil by 11.9%. MQL of NCFs had higher impact scores than MQL of their base fluids, mainly due to the used synthesis processes. MQL after adding 1 wt.% of nanoparticles to vegetable oil increased the total impact on average by 14.13%. TFC of conventional CFs displayed the lowest damage categories scores in comparison to MQL of vegetable oils and NCFs. However, TFC of conventional CFs is known to contribute to severe health problems for operators. Besides, its recycling is considered to be a challenging task. Conventional CFs offer a wide range of lubrication and cooling properties that are needed for different types of machining processes. MQL of NCFs displayed the highest damage categories scores compare to other cooling strategies, because of its energy-intensive synthesis process. Besides, operator exposure to nanosized particles can lead to respiratory distress and cancer. NCFs need special types of separation and filtering processes before disposal and recycling. MQL of NCFs achieved a high level of machining performance. MQL of vegetable oils displayed moderate impacts on the damage categories due to their cultivation and the fertilization processes. Besides, vegetable oils show a medium level of machining performance due to their low thermal stability. However, vegetable oils are considered to be environmentally friendly because they are renewable and biodegradable.
In terms of future work, the cradle-to-grave impacts of NCFs need more investigation along with developing the monitoring methodologies of the nanoparticles. Besides, research works are required to optimize the parameters of the synthesis process to reduce the energy consumption and impacts of NCFs. In addition, more attention is needed to develop new treatment techniques to eliminate the disposal problems of the NCFs to overcome their drawbacks.