MQL versus Dry Machining: A comparative analysis in a turning process using LCA Shadi Shams

(1)

DEGREE PROJECT IN CHEMICAL ENGINEERING AND TECHNOLOGY, FIRST LEVEL

STOCKHOLM, SWEDEN 2018

MQL versus Dry Machining:

A comparative analysis in a turning process using LCA

Shadi Shams

(2)

DEGREE PROJECT

Bachelor of Science in

Chemical Engineering and Technology

Title: MQL versus Dry Machining: A comparative analysis in a turning process using LCA

Swedish title: MQL eller torrskärning -‐

en

jämförande studie för en svarvningsprocess med hjälp av LCA

Keywords: Life cycle assessment, turning process,

environmental impact, process-‐based LCA, fast track LCA

Work place: KTH, Department of Production Engineering

Supervisor at

the work place: Associate Professor Amir Rashid

Supervisor at

KTH: Associate Professor Amir Rashid

Student: Shadi Shams

Date: 2018-‐06-‐12

Examiner: Sara Thyberg Naumann

(3)

Abstract

During the last decades the challenge of sustainability has become more urgent and environmental impacts of different processes in manufacturing industry have received more attention. Life cycle assessment (LCA) has become an important and useful tool to evaluate the environmental impact of products and processes.

In this study the environmental impact of two cooling techniques in a turning process has been evaluated using LCA. Turning is used for shaping metal parts by removing material. The compared cooling techniques in this study are dry cutting and Minimum Quantity Lubrication (MQL).

The inputs and output in each technique are considered in form of material flows and energy consumption as well as waste flows. The Ecoinvent database has been used in order to quantify, evaluate and compare the environmental impacts of the two cooling techniques. Environmental impact categories considered in this study are Carbon footprint (CO2 kg equivalent), Cumulative Energy Demand (CED), Total eco-cost in Euro and ReCiPe. ReCiPe is a method used to evaluate multiple environmental impact categories and it covers impact categories related to human health, ecotoxicity and material depletion.

Calculations and analysis of the results show that MQL has significantly lower environmental impact compared to dry cutting whereas energy consumption is the main contributor in the considered environmental impact categories.

(4)

Sammanfattning

Under de senaste åren har hållbar utveckling blivit mer relevant och miljöpåverkan av olika tillverkningsprocesser i industrin har således fått mer uppmärksamhet.

Livscykelanalys (LCA) har blivit ett viktigt och användbart verktyg för att analysera och utvärdera miljöpåverkan av produkter och processer.

I det här examensarbetet har miljöpåverkan av två olika kylmetoder vid svarvning utvärderats med hjälp av livscykelanalys (LCA). Svarvning används för att forma metalldelar. De jämförda kylmetoderna är torrskärning (dry cutting) utan kylvätska och minimalsmörjning (Minimum Quantity Lubrication - MQL) där en liten mängd smörjmedel används.

Tillfört material, energiförbrukning och avfall vid varje kylmetod har betraktats.

Ecoinvent-databasen har använts för att kvantifiera, utvärdera och jämföra miljöpåverkan av de två kylmetoderna. Miljöpåverkanskategorierna som behandlas i denna studie är koldioxidavtryck (CO2 kg ekvivalent), kumulativt energibehov (CED), totala miljökostnader i Euro och ReCiPe. ReCiPe är en metod som används för att utvärdera flera olika miljöpåverkanskategorier inkluderande människors hälsa, miljögifter och förbrukning av naturresurser.

Beräkningarna och analysresultaten visar att MQL har betydligt lägre miljöpåverkan

än torrskärning och att energiförbrukningen är den mest avgörande faktorn.

(5)

Acknowledgements

I would like to thank Professor Amir Rashid for the opportunity to have this topic as my degree project where I could apply my previous learning to a real case as well as broaden my knowledge by learning about machining processes. I would also like to thank my program director Sara Thyberg Naumann for all her support, care and dedication during the past three years. Last but not least I would like to thank Marta Garcia Tierno for helping me through this work by introducing me to the machining process and patiently answering my questions.

(7)

1. Introduction

During the last decades there has been more emphasis on sustainability [1]. More attention is therefore paid not only to advancing technologies in various industries but also to how sustainable theses developments are. In each industry large amounts of resources are used on a daily basis to produce certain products while enormous amount of waste is generated that in some cases can be problematic to dispose at the same time. It is interesting that during the past few years the regulations are not only driven by the governments but also consumers. Today consumers demand products that are more environmental friendly and create lower amounts of waste and emissions as well as resource demands [2]. Life cycle analysis (LCA) can in this context be an incredibly useful tool to estimate and evaluate the environmental impacts of a product or a process during its lifetime [3]. Therefore, LCA results can support decision-making and lead to lower damage to the environment.

1.1 Background

Different types of metal machining processes are used in various industries and especially in automotive industry for shaping metals [4]. In this context, turning is a frequently used machining process in industry, which is used to form metal parts through material removal [5]. During the turning process heat is generated as a result of friction between the cutting tool and the metal part that is being processed.

Therefore, it is important to decrease the heat since high temperature in the cutting zone can decrease tool life [6]. There are two cooling techniques of interest in this study: dry cutting and Minimum Quantity Lubrication (MQL). The former refers to the technique in which cooling or lubricating liquid is not used. The latter refers to using a small amount of liquid (i.e. few ml/h) for cooling and lubricating purposes [7].

Each alternative has its own technical and environmental advantages and disadvantages.

1.2 Purpose and objectives

This study quantifies the environmental impacts of dry cutting and MQL using LCA.

The environmental impact criteria taken into consideration in this case are CO2

emission using Carbon footprint, consumed energy using Cumulative Energy Demand (CED), Total eco-cost in Euro and ReCiPe impact categories.

The objective of this study is to compare and evaluate the two cooling techniques dry cutting and MQL with regards to their environmental impacts.

(8)

1.3 Scope and limitations

In this study the turning process is considered as a main system of interest where the flows entering and exiting this system are included in the LCA. The physical components of the machining equipment and the MQL instruments are not considered. Since the focus in this study is put on the environmental impacts, the technical details and parameters of the turning process are not considered. Only LCA- relevant data will be taken into account to evaluate the cooling techniques used in the turning process.

1.4 Outline of the thesis

This thesis consists of six chapters. Chapter 1 is an introductory chapter where the background of the topic as well as objectives and purpose of the study are formulated.

Also, the scope and the limits in the study are defined in chapter 1. In chapter 2 relevant theories of LCA and metal processing within the scope of the study are explained. Chapter 3 contains the method of the LCA study and includes steps taken during the study. The results of the LCA study are presented in chapter 4 where the environmental impacts of the two cooling techniques are compared. Chapter 5 includes a discussion where the results of the study are summarized and evaluated.

Finally, in chapter 6 conclusions are drawn based on the objectives of the study and suggestions to future work are made. Figure 1 shows the outline of the entire thesis.

Figure 1. Outline of the thesis Chapter 1

Introduction

Chapter 2

Relevant concepts and theories

Chapter 3 Methods

Chapter 4

Comparative analysis of dry cutting and MQL using LCA

Chapter 5 Discussion

Chapter 6 Conclusion

Background, objectives and scope

Review of relevant literature

The approach of the study based on relevant literature

Results of the study

Summary and evaluations of results

Concluding remarks and future work

(9)

2. Relevant concepts and theories

This chapter contains the results of literature review in form of theory related to LCA as well as the turning process. First, the steps of an LCA study, types of LCA and relevant LCA concepts are explained. Then the turning process and the two cooling techniques considered in the study are described briefly.

2.1 Life Cycle Assessment (LCA)

Life Cycle Assessment (LCA) according to ISO 14040 and 14044 is defined as “…the environmental aspects and potential impacts throughout a product’s life (i.e. cradle- to-grave) from raw material acquisition through production, use and disposal. The general categories of environmental impacts needing consideration include resource use, human health, and ecological consequences.” [8]. The general structure of an LCA is shown in figure 2 and consists of the following four steps:

1. Goal and Scope Definition 2. Life Cycle Inventory Analysis 3. Life Cycle Impact Assessment 4. Interpretation

Figure 2.LCA phases according to ISO 14040 [8]

Goal and Scope Definition: In this step the purpose and the scope of the study as well as the system boundaries are determined based on a real case. Also, the depth of the study is defined [9].

(10)

Life Cycle Inventory Analysis (LCI): This step in an LCA refers to gathering input and output data relevant to processes included in the system boundary [8]. The input data can be referred to as material and energy supplied to the system and the output data as waste and emissions to the environment [2]. Establishing an LCI is usually the most time consuming step in an LCA [10].

Life Cycle Impact Assessment (LCIA): In this step the results of the LCI are evaluated in order to quantify and determine the environmental impacts of the processes [8] [9]. The “translation” of LCI data into LCIA is normally done by using databases and/or LCA software programs since this step can be complicated and time consuming [2].

Interpretation: In this final step the results of the LCI and LCIA are discussed and a conclusion is drawn which can be used for further evaluation or decision-making purposes [8] [9].

2.2 Fast track approach

As it was mentioned earlier an LCA consists of four main steps and the most time consuming steps are LCI and LCIA steps. In order make an LCA more time-efficient the fast track LCA is a convenient approach. In a fast track LCA input and output of the considered process or product are multiplied by single indicator values from look- up tables or databases [11]. In such a case an indicator can be from a specific environmental category, for example Carbon footprint. The inputs and outputs are the material, energy and waste related to a process or a product. By simplifying the classic LCA steps, the fast track approach reduces the time needed for classification of environmental impact categories. Therefore a fast track LCA is more suitable for comparing two products or processes in a short amount of time while providing same results as a classic LCA [11].

Some examples of databases used in a fast track LCA are Ecoinvent and Idemat. The Ecoinvent database is gathered and compiled in Switzerland and contains over 4500 of LCIs that are associated with different processes and products from Switzerland as well as from Europe [9] [11]. The Idemat database is gathered and compiled by Delft University of Technology and contains an extension as well as a selection of LCIs in Ecoinvent databases. Peer reviewed articles are also used in Idemat [12]. As an alternative, computer programs can also be used for LCA calculations. Simapro is a well-established program in which the Ecoinvent database is incorporated. Another LCA program is GaBi, which can be used as database or software and is compiled at the University of Stuttgart. The data in GaBi is mostly related to processes and products in automotive industry [9].

(11)

2.3 Types of LCA

LCA can be generally divided into two major groups of process-based LCA and input-output-based LCA. This categorization is based on different types of data [13].

In the following sections these two types of LCA are described.

2.3.1 Input-output-based LCA

In an input-output (I-O) based LCA the data is collected using data that is based on statistics of production and consumption in specific industrial sectors [10]. The data includes economical transactions of goods and services involved in specific industries, for example automotive industry. These transactions are then used together with the environmental accounts in order to determine environmental impacts [13]. In I-O-based LCA all relevant industry sectors are included in datasets and therefore system boundaries are not necessary. The disadvantages of I-O based LCA is that it has less certainty comparing to process-based LCA since it takes longer to publish I-O datasets (1-5 years) [10] and it is therefore less frequently used [13] . The advantages of it on the other hand is that it is faster comparing to process-based LCA [10].

2.3.2 Process-based LCA

In a process-based LCA the focus is put on the processes involved in a system in which raw material and energy are used as input and products and emissions are produced as output [13]. In this type of LCA a system boundary is defined in order to limit the LCA to specific number of processes of interest [13]. In case of missing data of specific processes included in the system boundary, the processes may be grouped into a subsystem for which data is available [10]. One advantage of process-based LCA is that it provides more accurate results using more up-to-date data while the disadvantage of it is that it is more time consuming which results in higher costs [10].

2.4 Relevant LCA elements

2.4.1 System boundaries

A system boundary in the context of LCA refers to limiting the investigation by defining boundaries for it including the processes involved in the system for which the environmental evaluations are performed. Each system consists of various processes and sub-processes. It is important to define the boundaries appropriately from the beginning in order to arrange correct calculations and results [11].

(12)

2.4.2 Streamlining

In LCA streamlining is used when comparing two or more products or processes. It refers to reducing the number of processes in the system boundary in such a way that the focus is put on the differences of the processes involved [11]. Hence, less time is spent on those processes that are equal in all products or processes.

2.4.3 Functional unit

The functional unit is the unit used to describe the functionality of a system and is used for the LCA calculations [11]. Some examples for functional units are “per day”,

“per piece” or “per kg”. It is important to define a functional unit for the LCA in the beginning of an LCA study since it helps in determining the boundaries for the system [9].

2.5 Metal processing machining

Turning is one of the most extensively used machining processes for shaping and cutting metals in manufacturing industry [5]. In the turning process the cutting tool moves axial with a certain speed and cutting force in order to remove material from the turning work piece. Figure 3 shows a schematic picture of a turning process, which is relevant for this study. During the turning process material is removed from the work piece in form of chips [5]. During turning the friction between the cutting tool and the metal part increases the temperature. High temperature in the cutting zone can affect the tool life and degrade the cutting process. It is therefore important to decrease the heat and the friction in the cutting zone [6]. Metal working fluids (MWF) have been used for decades for this purpose and also to remove the chips from the cutting zone [4]. In MWF water or petroleum oil is used and therefore it is not the best choice from an environmental point of view [6]. During recent years due to stricter environmental regulations and also increasing cost of cutting fluids more environmental friendly options such as dry cutting and near dry cutting options have been introduced to the industry [4]. These techniques are described in the following sections.

Figure 3. Schematic representation of turning process [14]

(13)

2.5.1 Dry cutting

In a turning processes dry cutting refers to a process in which liquid is not used as lubricant or coolant [4]. It is therefore considered as a more environmental friendly cooling technique compared to MWF. In dry cutting the heat in the cutting zone is reduced by decreasing the friction between the cutting tool and the metal part. This can be done for example by using a coating layer with low friction coefficient on the tool material. Since dry cutting does not require any cooling liquid and generated waste in form of liquid is minimized, it is considered as a sustainable technology [4].

2.5.2 Minimum Quantity Lubrication (MQL)

Minimum quantity Lubrication (MQL) is a near dry machining technique. This technique refers to a metal cutting technique in which a small amount of liquid is mixed with compressed air, which is then sprayed on the cutting zone for cooling and lubricating purposes [7]. The MQL technique can be used in many metal machining process such as turning, milling and drilling as well as for cutting different materials such as steel, aluminum and cast irons [4]. In this study the considered liquid is a vegetable oil (rapeseed oil) and the technique for using the liquid is MQL.

In summary, it is anticipated that dry cutting may result in higher environmental impact through energy consumption. On the other hand, MQL requires additional fluids, which may impact the environment negatively. In the following section a method is outlined to compare both techniques based on a fast track LCA approach.

(14)

3. Method

In order to evaluate and compare the environmental impacts of the two techniques dry cutting and MQL in machining metal parts using turning process the four steps of LCA according to ISO 14040 have been followed. As it was mentioned in the previous section the second and the third step, i.e. LCI (Life Cycle Inventory) and LCIA (Life Cycle Impact Assessment), are the most time consuming steps in an LCA.

Since the purpose of this study is to compare two different cooling techniques (dry cutting and MQL), the fast track LCA approach has been followed. An overview of how the main four steps of LCA have been pursued is described below.

Step 1: Goal and Scope Definition

• Process-based LCA is used in this study where the focus is put on turning process as a process unit and its inputs and outputs are considered. The inputs are supplied material and energy to the system and outputs are generated waste from the system.

• The functional unit is defined.

• A system boundary is sketched and relevant inputs and outputs are identified.

Step 2: Life Cycle Inventory Analysis (LCI)

• The system boundary for each of the two techniques is filled with data for all inputs and outputs. The amount of required materials and the consumed energy were considered as inputs and the waste and the product as outputs.

• Reasonable assumptions are made for the scope of the study.

• Streamlining is used in order to make the comparison of the two techniques more effective, so that only the differences in the LCIs are highlighted in the LCA study.

• For quantifying the amount of material, waste and energy consumption required data was collected. The sources of data are from previous lab experiments run in the Production Engineering department at KTH. When adequate data was not available reasonable assumptions are made to estimate values for the further LCA calculations. In some cases the values are taken from related scientific articles containing similar experiment results, in order to make the calculations more realistic.

(15)

Step 3: Life Cycle Impact Assessment (LCIA)

• The Ecoinvent database is used in order to determine and quantify the environmental impacts. The version 2.2 of the database is used in this study from year 2012, which is the latest publicly available version, and has been downloaded from ecocostsvalue website that belongs to Delft University of Technology [12].

• The environmental impact categories in this study are Carbon footprint (CO2

kg equivalent), Cumulative Energy Demand (CED) in MJ, Total eco-cost in Euro and ReCiPe impact categories. The environmental impact categories are described below:

o Carbon footprint (CO2 kg equivalent)

Carbon footprint is a measure for amount of greenhouse gases (GHG) that are produced during various activities [12]. The unit for measuring the Carbon footprint is carbon dioxide where the amount of GHG emissions is converted into CO2 kg equivalents [15].

o Cumulative Energy Demand (CED)

Cumulative energy demand also known as “primary energy consumption”, refers to a sum of all direct and indirect energy demands associated with extraction, production, use and disposal of a an economic good [9]. In the database used in this study CED is measured in MJ.

o Total eco-cost in Euro

“Eco-cost is a measure to express the amount of environmental burden of a product on the basis of prevention of that burden” [12]. It is the cost of reducing an environmental impact, for example, the cost of reducing environmental burden of 10 kg CO2 emission [12].

o ReCiPe

ReCiPe is a method for evaluating multiple environmental impact categories as a single indicator [16] where the impact categories belong to three main impact groups: human health, eco-toxicity and materials depletion. The unit used in ReCiPe to express the environmental burden of product or a process is points (Pt), where higher points indicate higher environmental burden. The environmental impact categories in ReCiPe are listed below and are described in appendix C in more detail.

- Climate change – effect on humans

(16)

- Photochemical oxidant formation - Particulate matter formation - Ionizing radiation

- Climate change – effect on ecosystems - Terrestrial acidification

- Fresh water eutrophication - Terrestrial ecotoxicity - Fresh water toxicity - Marine ecotoxicity

- Agricultural land occupation - Urban land occupation - Natural land transformation - Metal depletion

- Fossil depletion

• The amount of inputs and outputs (in kg for the case of material and waste and in MJ for the case of energy consumption) are multiplied with each corresponding indicator value in order to quantify the environmental impacts.

In case indicator values are not available they are substituted with values from another indicator with similar characteristics.

Step 4: Interpretation

• First, the results from previous steps 1 through 3 are used to compare the overall environmental impacts of both techniques in all impact categories.

• The results are also used to identify which inputs and outputs are the major contributors in terms of environmental impact.

• Lastly, suggestions to minimize the environmental impacts are provided based on the results of the study.

(17)

4. Comparative analysis of dry cutting and MQL using LCA

4.1 Goal and Scope Definition

The goal of this LCA is to compare the environmental impacts in a turning process with dry cutting and MQL as cooling techniques. Figure 4 shows the turning machine that uses the two cooling techniques.

Figure 4. The turning machine and its tools

Figure 5 shows the associated system boundary to the turning process.

(18)

The functional unit is defined as “environmental impact per produced metal part” to highlight the environmental impacts associated with machining one metal part made of cast iron.

4.2 Life Cycle Inventory Analysis (LCI)

In this section the system boundaries for each cooling technique are detailed.

4.2.1 System boundaries

Figure 6 shows the system boundary for the turning process using dry cutting. The center of the system is the turning process (in blue). The inputs to the turning process are electricity, the metal part and the cutting inserts. The outputs are chips, which are assumed to be recycled, and ceramic cutting inserts that are disposed. The finished metal part is another output of the turning process.

Figure 6. System boundary for turning process using dry cutting

Figure 7 shows the system boundary for the turning process using MQL. In comparison to the dry cutting process MQL has additional inputs of oil and output of oil, which is disposed.

!"#$%&

'()*#*+&

$),-.%%&

!(/*+&#*%.)0%&&

123₄5₆7& 8#%$,%93&

:.-;-3#*+&

!(/*+&#*%.)0%&

1234567&

<.093&

$9)0&

=3.-0)#-#0;&

>#*#%".?&

@.093&$9)0&

(19)

Figure 7. System boundary for turning process using MQL

The equipment used for the MQL technique is a booster shown in figure 8. However, the MQL booster equipment itself is not included in the scope of the study.

Figure 8. MQL booster

4.2.2 Assumptions

The following assumptions are made:

• The amount of produced chips (removed material) and as a result the mass and volume of the product (the finished metal part) after the turning process is the same during both cooling/lubricating techniques.

!"#$%$&' (#)*+,,' -./'

0%,(),12' 0%,(),12' 3+*4*2%$&' -+512'

(1#5' 6"7$&'%$,+#5,'

892_:;_<='

>2+*5#%*%54'

6?%(,'

;%2' 6"7$&'%$,+#5,'

892_:;_<='

;%2'

>2+*5#%*%54'

@%$%,?+A' B+512'(1#5'

(20)

• The chips produced after machining using vegetable oil in MQL are clean and therefore no cleaning process is considered in the LCA.

• The vegetable oil after use is disposed.

• The ceramic cutting inserts are disposed after replacement.

• Metal cutting process takes approximately 8 seconds in both techniques.

4.2.3 Streamlining dry cutting and MQL

Based on assumptions streamlining is performed leaving only the processes that differ when comparing both techniques. Since the cutting parameters are the same in both scenarios, the amount of removed material is equal and the final product has the same volume after the process in both cases. Therefore as a result of streamlining, the metal part, the produced chips and the product, which all consist of cast iron, are not included in LCA calculations. Figure 9 shows the relevant inputs and outputs for this LCA marked with red circles.

Figure 9. System boundaries after streamlining in dry cutting and MQL (relevant processes marked in red circles).

!"#$%$&' (#)*+,,' -./'

0%,(),12'

0%,(),12' 3+*4*2%$&' -+512'

(1#5'

6"7$&'%$,+#5,' 892_:;_<='

>2+*5#%*%54'

6?%(,'

;%2' 6"7$&'%$,+#5,'

892_:;_<='

;%2'

>2+*5#%*%54'

@%$%,?+A' B+512'(1#5'

Dry cutting

MQL

!"#$%&

'()*#*+&

$),-.%%&

!(/*+&#*%.)0%&&

123₄5₆7& 8#%$,%93&

:.-;-3#*+&

!(/*+&#*%.)0%&

123₄5₆7&

<.093&

$9)0&

=3.-0)#-#0;&

>#*#%".?&

@.093&$9)0&

(21)

4.2.4 Material flows

In MQL a small amount of oil is mixed with compressed air. The mixture is then sprayed onto the cutting zone. During the turning process ceramic cutting inserts are used (see figure 10). The material of the cutting inserts in mainly aluminum oxide [17]. After machining a certain number of metal parts the cutting inserts are replaced with new ones since they are not usable any longer. An experimental study at KTH shows that cutting inserts need to be replaced more often in dry cutting cooling technique compared to MQL [17]. The difference of the wear-out of the cutting inserts for the two techniques is considered in the scope of the study.

Figure 10. Ceramic cutting insert made of aluminum oxide (Al2O3)

In the case of dry cutting, since there is not any cooling liquid, the only material used as resources is considered to be the cutting inserts. The metal part and the chips produced are not considered according to assumptions. Table 1 contains a list of relevant materials for the study (see appendix A1 and A2 for detailed calculations).

Material Kg/metal part

MQL

Vegetable oil ^{7.235 ∗ 10}^!! (calculated)

Cutting insert Al2O3 3.21 ∗ 10^!! kg (calculated)

Dry cutting

Cutting insert Al2O3 8.13 ∗ 10^!! kg (calculated) Table 1. Material resources used in MQL respective dry cutting technique

4.2.5 Energy consumption

The cutting energy in MQL respective dry cutting is considered as the energy consumption during the process. Since values from the real case are difficult to

(22)

reference where the energy consumption in MQL and dry cutting was measured by Power Sight Manager (PSM). A comparative ratio has been calculated in order to estimate the energy consumption for the case of cast iron material (see appendix A3 for detailed calculations). Table 2 shows these values for the two techniques.

Cooling/lubricating technique

Energy

consumption/metal part [MJ]

MQL 0.0129 [18]

Dry cutting 0.0289 [18]

Table 2. Energy consumption using in MQL respective dry cutting technique

4.2.6 Waste flows

The waste that is generated during MQL includes the oil and the replaced cutting inserts. The produced chips after the turning process are clean and therefore do not need to be cleaned. The residual oil after the turning process is disposed. The ceramic cutting inserts can be used for machining a certain number of metal parts, however the number of maximum machined metal part is different in MQL and dry cutting. The used cutting inserts are disposed after replacement. In the case of dry cutting the only generated waste is considered to be the cutting inserts. Results of previous lab experiments [17] have been used in order to estimate the amount of generated waste.

Table 3 contains the amount of waste generated in both techniques (see appendix A1 and A2 for detailed calculations).

Waste Kg/metal part

MQL

Vegetable oil 7.235 ∗ 10^!! (calculated)

Cutting insert Al2O3 3.21 ∗ 10^!! (calculated)

Dry cutting

Cutting insert Al2O3 8.13 ∗ 10^!! (calculated) Table 3. Waste flows in MQL respective dry cutting technique

In order to quantify environmental impacts, the amount of material flows, energy consumption and waste flows per functional unit is multiplied with the indicator value for each process in the database. A detailed dataset is attached in appendix B.

(23)

4.3 Life Cycle Impact Assessment (LCIA)

The environmental impact categories in this LCA are Carbon footprint (CO2 kg equivalent), Cumulative Energy Demand (CED) in MJ, the ReCiPe environmental impact categories (Pt) as well as Total eco-cost in Euro. The results of environmental impacts of the turning process using dry cutting and MQL are presented in figures 11 through 14. ReCiPe environmental impact categories are explained in detail in appendix C.

Figure 11 shows the Carbon footprint (CO2 kg equivalent) in the turning process using dry cutting respective MQL. MQL has approximately 55% lower Carbon footprint compared to dry cutting. The major part in both dry cutting and MQL consists of electrical energy consumption with 72%. The usage of ceramic cutting inserts as material has the second largest impact with 28% in dry cutting and with 25% in MQL. The disposal of ceramic cutting inserts in dry cutting as well as MQL does not have any major impacts on CO2 emission. The oil has a minor impact of 3%

in MQL as material. However, the disposal of the oil can be neglected.

Figure 11. Carbon footprint CO2 kg

Figure 12 shows the Cumulative Energy Demand (CED) in dry cutting and MQL.

MQL has 55% lower CED than dry cutting. For both dry cutting and MQL electrical energy consumption has a significant effect on CED with 95% respective 94% of the overall impact. On the other hand the material and waste (i.e. the oil and ceramic cutting inserts) have minor effects on CED impact category. In both dry cutting and

28%

25%

72%

3%

0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012

Dry cutting MQL

Carbon footprint CO₂ kg

Waste: Lubrication oil

Material: Lubrication oil

Waste: Cutting insert Al2O3

Energy: Cutting energy consumption

Material: Cutting insert Al2O3

!""#$#

(24)

Figure 12. Cumulative Energy Demand (CED) in MJ

Figure 13 shows the Total eco-cost in Euro. Similar to previous impact categories, MQL has 56% lower eco-cost than dry cutting. The most significant contributor in both methods is electrical energy consumption with 54% for dry cutting and 55% for MQL. The second and third significant effects result from ceramic cutting inserts. In dry cutting the ceramic cutting inserts have an impact of 31% as material and 15% as waste. Similarly, in MQL the effects of ceramic cutting inserts are 28% as material and 13% as waste. The oil has minor impacts of 4% as material while the impact of the oil disposal can be neglected.

Figure 13. Total eco-cost in Euro.

5% 5%

95%

94%

1%

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

Dry cutting MQL

CED in MJ

!""#$#

31%

28%

54%

55%

15%

4% 13%

0 0.00005 0.0001 0.00015 0.0002 0.00025 0.0003 0.00035

Dry cutting MQL

Total eco-cost in Euro

!"#$%$

(25)

Figure 14 shows the ReCiPe environmental impact categories. Similar to Carbon footprint and CED, MQL has 53% lower score in the ReCiPe categories compared to dry cutting. The impacts are similarly distributed for both methods: climate change with effects on human (28% in dry cutting and 27% in MQL), fossil depletion (26%

in dry cutting and 24% in MQL), climate change with effects on ecosystems (13%) particulate matter formation (10%) and human toxicity (8%). The exact definitions of each of the ReCiPe impact categories are explained in appendix C.

Figure 14. ReCiPe environmental impact categories

4.4 Interpretation

4.4.1 Significant contributors

The LCIA results show that overall MQL has roughly 53-56% lower environmental impact compared to dry cutting and in all environmental categories (Carbon footprint, Cumulative Energy Demand CED, Total eco-cost in Euro and ReCiPe impact categories).

28%

27%

5%

4%

8%

10%

2%

13%

3%

2%

3%

7%

1%

26%

24%

1%

0 0.00001 0.00002 0.00003 0.00004 0.00005 0.00006 0.00007 0.00008 0.00009 0.0001

Dry cutting MQL

Recipe environmental Impact categories (Pt)

Fossil depletion Metal depletion

Natural land transformation Urban land occupation Agricultural land occupation Marine ecotoxicity

Fresh water ecotoxicity Terestrial ecotoxicity Fresh water eutrophication Terrestrial acidification Climate change, ecosystems Ionising radiation

Particulate matter formation Photochemichal oxidant formation Human toxicity

Ozone depletion Climate change, human

!"#$%$

(26)

The electrical energy consumption has considerably high effect on the results in all impact categories and in both dry cutting and MQL, as it accounts for the majority of the total inputs supplied to the system and outputs generated from it.

As can be seen in figures 11 through 14, ceramic cutting inserts have the second highest contribution after energy consumption in all impact categories and in both cooling techniques.

The additional oil supply that is used in MQL as a lubricant and cooling liquid does not have any major effect on the total environmental impact. One explanation for this fact may be the enormously large impact of electrical energy consumption, which by far exceeds the environmental impact of the oil in MQL.

4.4.2 Suggestions to minimize environmental impacts

Since the electrical energy consumption is the main contributor in all environmental impact categories, it may be worthwhile to consider technical energy optimization to reduce consumption and therefore impact on the environment. The results of this study indicate that MQL allows for lower electrical energy consumption and may therefore be a better choice, if the environmental burden of machining processes is supposed to be reduced.

As a second point, recycling of the ceramic cutting inserts may be considered.

Although the environmental impact of the ceramic cutting inserts is lower than the electrical energy consumption, the recycling of these inserts would decrease their environmental impact even further.

(27)

5. Discussion

As the results of the study show, energy consumption is the main contributor in all environmental impact categories. Therefore the electrical energy consumption has a significant effect on the LCA results. In this study the energy consumption has been estimated by adapting values from a titanium alloy case, which has been found in literature, to the given cast iron case. An accurate measurement of electrical energy consumption in MQL and dry cutting may therefore lead to more accurate results for environmental impacts associated to the energy consumption.

The results of the study also show that the energy consumption in the case of MQL cooling technique is approximately 50% lower than dry cutting. This difference in electrical energy consumption can be the result of reduced cutting force due to lubrication by the oil in the case of MQL technique. Another factor that may influence the energy consumption can be the cycle time, which may be shorter in the case of MQL technique. Since the cycle time in this study is assumed to be equal in both cooling techniques, the potential reduction of cycle time using the MQL cooling technique needs to be further investigated.

According to one of the assumptions in the study, the chips do not require any cleaning process before recycling. Even though the amount of oil is very small in the case of MQL cooling technique, in practice a new cleaning process may still be necessary to remove the oil to ensure proper recycling. Such a process may increase the environmental impact of the MQL technique.

Regarding the oil in MQL, the indicator that has been used from the database to calculate the environmental effects for the disposal scenario was a mix of fat and oil.

Since the MQL oil used in this study is biodegradable without any added chemicals a corresponding indicator value should be created in the database. This would lead to a result of even lower environmental impact associated to the disposal of oil.

Another factor that affects the accuracy of the study is the version of the database and the indicator values used in this LCA study. Since some indicator values were missing in the Ecoinvent database for two processes, they were replaced with similar processes. Lack of indicator values for disposal of vegetable oil in MQL and disposal of ceramic cutting inserts in both MQL and dry cutting may limit and affect the validity of the study. For this study it is expected that the overall trend of the results would remain unchanged, however access to an updated version of the database may result in more accurate estimations. Using LCA software programs like Gabi or Simapro can also be advantageous for more accurate calculations since access to the latest version of the Ecoinvent database is incorporated in the software license.

(28)

6. Conclusions

This study has evaluated the environmental impacts of a turning process using dry cutting and MQL techniques for cooling and lubricating purposes. As main method process-based LCA has been used. The evaluation has been done by considering the material supplied to the system, energy consumption by the cutting process and the waste generated after the process. The environmental impacts of the two cooling and lubricating techniques have been compared. The criteria of evaluating the environmental impact were Carbon footprint (CO2 kg equivalent), Cumulative Energy Demand (CED) in MJ, Total eco-cost in Euro and ReCiPe impact categories (Pt). As main conclusion of this LCA study, in all investigated environmental categories MQL outperforms dry cutting, which is based on the following findings:

• The CO2 emissions of MQL are 55% lower than dry cutting technique. The CO2 emissions are measured with Carbon footprint in kg CO2 equivalent.

• Cumulative Energy Demand (CED) of MQL is 55% lower than dry cutting technique.

• Total eco-cost (Euro) of MQL is 56% lower than dry cutting technique.

• Environmental impact categories of MQL in ReCiPe categories are 53% lower than dry cutting technique.

As future work, the energy consumption needs to be measured accurately in both cooling and lubricating techniques to obtain data for the life cycle inventory of the process. Secondly, the potential reduction of cycle time in the MQL cooling technique may also be further investigated. Thirdly, a newer version of the Ecoinvent database may lead to more up-to-date results. The results from LCI in this work together with accurate energy consumption measurements can be used to repeat the experiment with an LCA software program, for example Simapro, in order to compare and strengthen the findings.

(29)

References

[1] C. Pieragostini, M. C. Mussati, and P. Aguirre, “On process optimization considering LCA methodology,” J. Environ. Manage., vol. 96, no. 1, pp. 43–

54, 2012.

[2] S. Roy, Applying Life Cycle Assessment (LCA) in Process Industry—The Chemours Experience BT -‐ Reference Module in Earth Systems and Environmental Sciences. Wilmington: Chemours Company, 2017.

[3] Z. Zhang, H. Li, X. Hao, Y. Chen, C. Li, and C. Wu, “The life cycle assessment of metal materials used for automobile body materials and castings,” TMS (The Miner. Met. Mater. Soc., vol. 89–90, pp. 157–164, 2013.

[4] I. H. Mulyadi, “Improving the Performance of Minimum Quantity Lubrication in High Speed Milling and Environmental Performance Analysis,” The University of Manchester, 2013.

[5] T. Childs, K. Maekawa, T. Obikawa, and Y. Yamane, Metal Machining -‐

Theory and Applications. London: Arnold, a member of the Hodder Headline Group, 2000.

[6] S. Pervaiz, A. Rashid, I. Deiab, and C. Mihai Nicolescu, “An experimental investigation on effect of minimum quantity cooling lubrication (MQCL) in machining titanium alloy (Ti6Al4V),” Int. J. Adv. Manuf. Technol., vol. 87, no.

5, pp. 1371–1386, 2016.

[7] S. A. Lawal, I. A. Choudhury, and Y. Nukman, “A critical assessment of lubrication techniques in machining processes: A case for minimum quantity lubrication using vegetable oil-‐based lubricant,” J. Clean. Prod., vol. 41, pp. 210–221, 2013.

[8] ISO (International Standard Organization), “ISO 14040. Environmental Management – Life Cycle Assessment – Principles and Framework,” CEN European Committee for Standardization, Brussels, 2006.

[9] W. Klöpffer and B. Grahl, Life Cycle Assessment (LCA) A Guide to Best Practice. Weinheim, Germany: Wiley-‐VCH Verlag GmbH & Co, 2014.

[10] Q. Jiang, T. Li, Z. Liu, H. Zhang, and K. Ren, “Life Cycle Assessment of an engine with input-‐output based hybrid analysis method,” J. Clean. Prod., vol. 78, pp. 131–138, 2014.

[11] J. G. Vogtländer, A practical guide to LCA for students, designers and business managers, 1st ed. Delft: VSSD, 2010.

[12] Delft University of Technology, “The Model of the Eco-‐costs / Value Ratio

(EVR),” 2017. [Online]. Available:

(30)

[13] M. J. Yan, J. Humphreys, and N. M. Holden, “Evaluation of process and input-‐output-‐based life-‐cycle assessment of Irish milk production,” J. Agric.

Sci., vol. 151, no. 5, pp. 701–713, 2013.

[14] B. Buldum, U. Eşme, M. Kulekci, Tarsun-‐Mersin, A. ŞIK, Ankara, and Y.

Kazançoğlu, “Use of Grey-‐Taguchi Method for the Optimization of Oblique Turning Process of AZ91D Magnesium Alloy,” Mater. Test., vol. 54, pp. 779–

785, Dec. 2012.

[15] Climate Change Connection, “CO2 equivalents,” 2016. [Online]. Available:

https://climatechangeconnection.org/emissions/co2-‐equivalents/.

[Accessed: 01-‐May-‐2018].

[16] M. A.J. Huijbregts, Z. J. N. Steinmann, P. M. F. Elshout, G. Stam, F. Verones, M. D. M. Vieira, M. Zijp, and R. van Zelm, “ReCiPe 2016: A harmonized life cycle impact assessment method at midpoint and enpoint level -‐ Report 1 : characterization,” Bilthoven, 2016.

[17] M. Garcia Tierno, “Towards sustainability using Minimum Quantity Lubrication technique and Nano-‐cutting fluids in metal-‐machining processes,” Master Thesis, KTH Royal Institute of Technology, 2017.

[18] I. Deiab, S. W. Raza, and S. Pervaiz, “Analysis of lubrication strategies for sustainable machining during turning of titanium ti-‐6al-‐4v alloy,” Procedia CIRP, vol. 17, pp. 766–771, 2014.

[19] M. Goedkoop, R. Heijungs, M. Huijbregts, A. De Schryver, J. Struijs, and R.

Van Zelm, “ReCiPe 2008: A life cycle impact assessment method which comprises harmonised category indicators at the midpoint and the endpoint level,” 2009.

[20] World Health Organization (WHO), “Ozone and other photochemical

oxidants,” 2000. [Online]. Available:

http://www.euro.who.int/__data/assets/pdf_file/0018/123084/AQG2nd Ed_7_2ozone.PDF?ua=1. [Accessed: 02-‐May-‐2018].

(31)

Appendices

Appendix A. Calculations

A1. Calculation of amount of oil used for one metal part

Knowing the flow of the oil which is 2-5 ^!"_! [17], the average flow of oil 3.5 ^!"_! is then considered for further calculations:

3.5 ^!"_!= _!"##^!.! ^!"_! = 9.722 ∗ 10^!! ^!"_!

Machining one metal part takes ca. 8 s, therefore the volume of oil for machining one metal part can be estimated to be:

9.722 ∗ 10^{!! !"}_! . 8 s = 0.00778 ml

The amount (mass) of oil used for one metal part can be calculated knowing the density and the volume:

! = !

!

Where ! [_!"^!] is the density, ! [!] is the mass and ! [!"] is the volume.

!!"#$%$$& !"# ≈ 0.93 _!"^! [6]

! = !. ! = 0.93 _!"^! ∗ 0.00778 ml = 0.00723 g ≈ 7.235 ∗ 10^!! !"

(32)

A2. Calculation of ratio of cutting insert needed for machining one metal part

According to the experimental study carried out in the lab in department of Production Engineering at KTH the number of machined metal parts before breakage of cutting inserts are different in dry cutting and MQL [17]. When dry cutting was implemented the first time 69 and the second time 54 metal parts were machined before the cutting insert would break [17]. For the calculations the average number that is 61.5 is considered. The ratio of cutting inserts required for machining one metal part can then be estimated to:

1

61.5≈ 0.01626

Knowing the weight of the cutting insert which is 5 g, then the amount of the ceramic cutting insert required for machining one metal part in dry cutting can be calculated:

0.01626 ∗ 5 ∗ 10^!! ≈ 8.13 ∗ 10^!! kg

The number of successfully machined metal parts before breakage of cutting inserts in MQL the first time was 150 and the second time was 162 [17]. For the calculations the average number, 156 is considered. The ratio of cutting inserts required for machining one metal part is therefore:

1

156≈ 0.00641026

Similarly, the amount of the ceramic cutting insert required for machining one metal part in MQL can be calculated:

0.00641026 ∗ 5 ∗ 10^!! ≈ 3.21 ∗ 10^!! kg

(33)

A3. Calculation of energy consumption in MQL and dry cutting techniques

Since the data from the real case were not available during this study, the energy consumption is taken from an article in which the energy consumption of turning process using dry cutting and MQL are estimated [18]. In this experiment the Energy consumption using dry cutting and MQL techniques are estimated to be 0.01875 kWh respective 0.00833 kWh. Assuming that the duration of machining metal parts is the same, the energy consumption for machining one metal part using dry cutting can be calculated:

1 kWh = 3.6 MJ

0.01875 kWh = 0.01875 * 3.6 = 0.0675 MJ

Similarly, the energy consumption for machining one metal part using MQL can be calculated:

0.00833 kWh = 0.00833 * 3.6 = 0.029988 MJ

Since the material in the experiment is a titanium alloy and the material of the interest in this study is cast iron, a ratio is needed to estimate the energy consumption in this study. This ratio has been calculated by comparing the specific cutting energies and the cutting powers in the two cases (processing titanium alloy versus cast iron). First the cutting power is calculated for processing each material using the cutting parameters:

!_! = !_!×!_!×!_!×!_! 60037.2

Where !_! [!"] is the cutting power, !_! _!"#^! is the cutting speed, !_![!!] is the cutting depth, !_![^!!_!"#] is the feed per revolution and !_! [_!!^!_!] is the specific cutting force.

!_! (!"#$ !"#$) =!"# × !.! × !.! × !"##

!""#$.! = 1.56 kW

!_! (!"#$%"&' !""#$) = !"# × !.! × !.! × !"##

!""#$.! = 0.895 kW The specific cutting energy for both materials is then calculated:

!_! = !_!

!!"

(34)

Where !_! [_!!^!!!_!] is the specific cutting energy, !_! [!"] is the cutting power, and

!!" [^!!_! ^!] is the material removal rate.

!! (!"#$ !"#$) =1.56 × 10^! [!]

1300 [!!! ]^!

= 1.2 [! − !

!!^!]

!! !"#$%"&' !""#$ = 0.895 × 10^! ! 320 !!! ^!

= 2.8 ! − !

!!^!

Finally the ratio of the two specific cutting energies can be estimated as:

!"#$%&%$ !"##$%& !"!#$% !" !"#$ !"#$

!"#$%&%$ !"##$%& !"!#$% !" !"!#$"%& !""#$= 1.2

2.8= 0.4286

Using the ratio, the energy consumption of processing cast iron using dry cutting cooling technique can be estimated to be:

0.0675 MJ × 0.4286 = 0.0289 MJ

Similarly, the energy consumption of processing cast iron using MQL cooling technique can be estimated to be:

0.029988 MJ × 0.4286 = 0.0129 MJ

(35)

Appendix B. Data set for LCI and LCIA

re B1. Dataset for Carbon footprint (CO2 kg equivalent), Cumulative Energy Demand (CED) in MJ and Total eco-cost in Euro.

(36)

Figure B2. Dataset for ReCiPe environmental impact categories

MQL versus Dry Machining: A comparative analysis in a turning process using LCA Shadi Shams