Life cycle assessment of cubic boron nitride grinding wheels

Accepted Manuscript Life cycle assessment of cubic boron nitride grinding wheels Marius Winter, Suphunnika Ibbotson, Sami Kara, Christoph Herrmann PII:

S0959-6526(15)00656-3

DOI:

10.1016/j.jclepro.2015.05.088

Reference:

JCLP 5602

To appear in:

Journal of Cleaner Production

Received Date: 13 November 2014 Revised Date:

15 May 2015

Accepted Date: 16 May 2015

Please cite this article as: Winter M, Ibbotson S, Kara S, Herrmann C, Life cycle assessment of cubic boron nitride grinding wheels, Journal of Cleaner Production (2015), doi: 10.1016/j.jclepro.2015.05.088. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Life cycle assessment of cubic boron nitride grinding wheels

Marius Winter a, c*, Suphunnika Ibbotson a, b, Sami Kara a, b, Christoph Herrmann a, c

Joint German-Australian Research Group on Sustainable Manufacturing and Life Cycle Management

b

Sustainable Manufacturing and Life Cycle Engineering Research Group, School of Mechanical and

RI PT

a

Manufacturing Engineering, The University of New South Wales, Sydney, NSW 2052, Australia c

Sustainable Manufacturing and Life Cycle Engineering Research Group, Institute of Machine Tools and Production Technology (IWF), Technische Universität Braunschweig, Langer Kamp 19b, 38106

SC

Braunschweig, Germany

M AN U

* Corresponding author

Abstract:

AC C

EP

TE D

To reduce the environmental impacts caused by manufacturing processes science, and industry want to identify hotspots and to derive improvement measures. One of those contributing manufacturing processes is grinding using synthetically produced cubic boron nitride (cBN). CBN grains are broadly applied for super abrasive materials in the production of high-precision grinding wheels. The shape, size and volume concentration of the cBN grains have a major impact on the technological results (workpiece roughness, tool wear, temperature, etc.) of the grinding process and on the economic value of the grinding wheel. Despite the technological results and the economic value, the contribution of cBN grinding wheels to the overall environmental impact has not been fully investigated and understood. A key method to calculate the environmental impact is life cycle assessment. However, an essential requirement is the availability of the used material and energy data during the life cycle stages of grinding wheel material, production, application and disposal. This paper gives an overview regarding the needed materials and energy during the different life phases of a cBN grinding wheel. On this basis, the detailed environmental impact of the grinding process is presented in a case study.

Keywords:

1 of 26

grinding wheel; grinding process; life cycle assessment; cubic boron nitride; cradle-tograve

ACCEPTED MANUSCRIPT

1

Introduction

RI PT

Grinding is a finishing process with a geometrically undefined cutting edge. This process is used to achieve certain technological workpiece characteristics such as a fine surface finish, high geometrical accuracy and specific material properties. In industrial countries, this process accounts for 20 to 25 % of their total expenditures of all machining processes [1]. Approximately 28 % of the machine tool stock (about 740.000) installed in the European Union (EU27) are grinding machines (state 2009) [2]. The process can be used for the machining of hard-to-machine materials such as cemented carbide, carbon and alloy steels and austenitic nickel-chromium-based super alloys. For this purpose, different workpiece shapes can be ground utilising surface and rotational process kinematics.

AC C

EP

TE D

M AN U

SC

The grinding process can be described based on the interactions between the input process variables and the achieved output objectives. These complex interactions are divided into three main compositions and are depicted in Figure 1. The interactions include the relations of the process input variables (left side), the grinding process (centre) and the influences on the output objectives (right side). The left side of the figure presents the three groups of process input variables including the workpiece properties, the process parameters and different enabling factors. The centre of the figure illustrates the interactions of the groups, which are decomposed into three layers. The top layer presents the main purpose of the grinding process with the product attributes transformation. This layer is influenced by the required workpiece properties. The achievement of these properties is a function of the selected process parameters, such as cutting depth (ae), cutting speed (vc), workpiece speed (vw) and dressing feed (vfad). The enabling factors have, in connection with the process parameters, influence on the resource and energy conversion during the grinding process. Subsequently, the layers of a grinding process influence the three output objectives on the right side. They are distinctively classified as technological (surface roughness, geometrical accuracy, etc.), economic (cost and time) and environmental (carbon footprint, resource depletion, etc.) objectives.

Figure 1:

Composition of a grinding process (adapted from [3] and [4]).

The technological and economic objectives are the main goals of the grinding process. However, the environmental objective is of increasing importance due to changed legislation, regulations and customer requirements [5]. It is a challenge to achieve all objectives due to their antagonistic effects. To achieve a good surface roughness, for example, the cutting depth should be low. However, a low cutting depth increases the process time and inclines the cost and environmental impact due to a higher energy demand. 2 of 26

ACCEPTED MANUSCRIPT A number of studies have been conducted to investigate the environmental impacts of a grinding process. However, they often considered the assessment for the carbon dioxide equivalent (CO2eq.) of the raw materials and production of a grinding wheel in a simplified manner [3]. They also excluded the overall environmental impact, or only considered the environmental impact due to a combination of energy, cutting fluid or grinding wheel demand, while not changing the process parameters. An environmental impact assessment of the grinding wheel in details has not been possible due to the limitation of data availability as a result of the confidentiality issue [4].

SC

RI PT

To reduce the environmental impact, an understanding of the system is required to avoid problem shifting and to handle goal conflicts. It is possible to demonstrate the importance of system understanding and how to derive recommendations for industry and future research by using the grinding process as an example. Therefore, this research collected extensive data for the grinding wheel production as well as experimental data to examine the needed resources and energy flows during the operation of the grinding process. A life cycle assessment (LCA) is used to calculate the environmental impact of a grinding wheel in different life cycle phases including:

M AN U


impact assessment of one grinding wheel (cradle-to-gate)
Impact assessment of the grinding process for producing 12,000 workpieces (cradle-to-grave).

EP

TE D

This paper provides a theoretical background of a cBN grinding wheel and LCA in section 2, which also highlights previous research related to the LCA of cBN grinding wheels and research gaps. Subsequently, section 3 presents materials and methods. The background information and data of materials and production processes that are used to produce a cBN grinding wheel as well as material flows of the main ingredient, namely cBN abrasive grains, bond systems, pore builder and other additives are described in section 3.1. A methodology of LCA and case studies of both cradleto-gate and cradle-to-grave are also given in section 3.2. The information is based on input data obtained from extensive literature as well as experimental data, which is then used to generate results in Section 4. LCA results are discussed for both, a grinding wheel and its life cycle, considering varying grinding process parameters during the usage life cycle stage. Section 4 demonstrates results found in the hotspot analysis and the sensitivity analysis. The conclusion of this investigation is presented in section 5.

2.1

Theoretical Background

AC C

2

Application, structure and materials of grinding wheels

The composition and structure of the grinding wheel are the major influencing factors in achieving the prescribed technological characteristics. Four main grinding wheel components can be distinguished: the abrasive grains, the bond, the pores and (depending on the grinding wheel design) the wheel hub [8]. The abrasive grains are embedded and linked by the bond. Due to the irregular shape of the abrasive grains, pores are naturally created or can be induced by artificial pore builder. Pores are needed for cutting fluid transport into and chip clearance out of the process zone [9]. The abrasive grains and the bond are mixed and pressed to a green body to form either a full body abrasive wheel or an abrasive layer or segments. Subsequently, the body, the layer or the segments are cured and then coated on a wheel hub (see Figure 2).

3 of 26

ACCEPTED MANUSCRIPT

RI PT

Abrasive / bond / pore mix (layer) (full body)

Figure 2:

Structure of grinding wheels.

Abrasive Porosity grains

M AN U

Bond

SC

Wheel hub

TE D

The grinding wheel performance and characteristics are significantly influenced by the volume ratio of abrasive, bond and pore to the total wheel volume. For example, the increase of bond content, if the abrasive content is kept constant, leads to a reduction of the pores and to an increase of the grinding wheel hardness. The reason is the creation of stronger bond links between the abrasive grains [9]. The increase of the pore volume leads either to a reduction of cutting edges or bond strength. Apart from this technological impact, the volume ratio of abrasive and bond to pore has also an environmental and economic impact due to the amount of used materials.

AC C

EP

The abrasive grain materials are distinguished into two groups: the conventional abrasives and the super abrasives. The conventional abrasives commonly enfold grains made from fused or non-fused aluminium oxide, zirconia alumina and black or green silicon carbide. The super abrasive grains are made either from diamond or cubic boron nitride (cBN) [9]. Diamond is the hardest material followed by cBN, silicon carbide, aluminium oxide and zirconia alumina. Regarding the bond material, different types can be distinguished: vitreous, organic/synthetic resin, metallic sintered and electroplated bond systems. Depending on the wheel design and application scenario, the abrasive bond mixture can be coated on a wheel hub. As previously mentioned, different combinations of the mentioned components are mixed to produce a grinding wheel depending on the grinding process requirements. The resulting technological advantages and disadvantages are comprehensively described in literature, including Klocke and König [10], Marinescu et al. [11], Rowe [12] and Malkin and Guo [1]. 2.2

Life cycle assessment

LCA is widely used in evaluating an environmental impact of the entire life cycle stages of a product or a process. The life cycle stages include materials, the manufacturing process, usage and end-of-life (EOL). An LCA is conducted in four steps, according to ISO 14040 [13]. The first step is the ‘scope and 4 of 26

ACCEPTED MANUSCRIPT

2.3

LCA of machining processes and cBN grinding wheels

SC

RI PT

goal definition’ and defines the purpose, limitations and the functional unit of the analysis. ‘Inventory’ is the second step, which is carried out to collect a set of input data from various sources, including life cycle inventory (LCI) databases, to generate the LCI of input and output substances for the product life cycle. The challenges of this step are commonly caused by the limitations of input data and LCI databases. The next step is the ‘impact assessment’ involving the application of welldefined life cycle impact assessment (LCIA) methods to convert LCI into meaningful environmental impact results. Such results can be interpreted in different levels including: characterisation results of different environmental impact categories with specific units (e.g. climate change in kg CO2eq.), normalisation results in a dimensionless unit; weighted results in terms of the main damage categories (e.g. human health) and single score results summarising the damage category results. Midpoint LCIA methods generate the characterisation and the normalisation results, while endpoint LCIA methods can additionally produce the damage category and single score results. The last step, ‘interpretation’ is mainly the presentation and analysis of the produced results to illustrate the main findings, hotspots and uncertainties of the assessment.

AC C

EP

TE D

M AN U

The LCA approach can support decision making in the product design stage for many industries by quantitatively calculating environmental impact results for different design alternatives. In the field of grinding and other machining processes, an LCA has been conducted by different authors for the cutting fluid and machine tool. For instance, Dettmer conducted an LCA to compare the environmental impact of non-water miscible cutting fluids on mineral oil base with cutting fluids on the basis of animal fat and used cooking oil. The fluid’s life cycle was investigated and comprehensively modelled. However, due to the focus being on cutting fluid, the investigation lacks the consideration of the fluid’s influences on the tool or process [6]. Clarens et al. compared the environmental impact of water-based cutting fluids with gas-based cutting fluids. The authors considered the fluid’s life cycle and investigated the interrelationships of the fluid’s application parameters (e.g. flow rate) and the replacement intervals [7]. Yet, the study lacks the consideration of the cutting fluid is interrelations with process and tool. Studies regarding the environmental impact of a machine tool mainly focus on the impact assessment due to the electrical energy demand. For example, Winter et al. calculated the environmental impact of the machine tool based on the electrical energy demand during the grinding process [14]. This approach did not consider the environmental impact due to the life phases of the machine tool. The overall environmental impact of a grinding process is composed of the demand of three enabling factors cutting fluid, machine tool and grinding wheel [3]. The demand rate depends on the machined workpiece properties, the selected process parameters and the interdependency between the three enabling factors. Li et al. performed one of the first environmental impact calculations of a grinding process by considering the cutting fluid and machine tool. Their calculation considered only the material phase by including only a few of the main materials used for the cBN grain synthesis. The investigation is based on a simple estimation and background data available in the Ecoinvent 2.0 database [15]. Other kinds of material and energy use in the material phase were omitted such as the entire cBN synthesis process and other life cycle phases (i.e. the grinding wheel production, use and disposal phases). These exclusions were made due to the limitation of LCI data inputs [15].

5 of 26

ACCEPTED MANUSCRIPT

RI PT

Linke and Overcash proposed a framework of inputs and outputs required in the grinding process when performing an LCA at a theoretical level due to the limited LCI data regarding the grinding tool [16]. Murray et al. performed an LCA of a resin bonded cBN grinding wheel by considering the cBN grain synthesis, grinding wheel production, use and end of life phase [17]. In their study, the LCA was based on the LCI values from the Ecoinvent 2.0 and the Simapro 7.1 databases. The materials, which were not included in the two databases, were calculated by the authors. Their environmental impact calculation of the cBN synthesis grains was based on the assumption that the cBN synthesis is similar to the synthesis of synthetic diamonds. This is correct; however, it is not clear if the different synthesis times for either cBN or diamond and the associated environmental impact were considered correctly. Murray et al. stated that it takes four days to synthesise a diamond of an average 2.5 carats and needs 20 kWh per carat [18]. The synthesis of cBN takes only 10 to 20 minutes under comparable temperature and pressure conditions to the diamond synthesis [18].

M AN U

SC

Based on the investigation of Murray et al. and Li, a simplified LCA was performed by Winter et al. for a vitrified bonded grinding wheel, involving the raw materials and the production process of the cBN grains. Nonetheless, they assessed the environmental impact merely based on the selective input data of the grinding wheel raw materials and production process only for the global warming potential in a unit of carbon dioxide equivalent. The impact of vitrified bonded materials, grinding wheel production process, use and end of life phase were not considered due to an absence of data [14].

TE D

Another study was conducted by Aurich et al. to overcome the drawback of absent data for the production process of vitrified bonded grinding wheels. Aurich et al. presented, for the first time, measurement data of the electrical energy demand [19]. They presented some valuable LCI data. However, the energy demand of the cBN grain synthesis and the needed raw materials were not included. The exclusions of the previous studies indicate that an environmental impact towards human health, ecosystem and resource use remains unclear. Therefore, it is essential to conduct a screening LCA that comprehensively calculates all environmental impact categories which are recommended in ISO 14040. Research aim

EP

2.4

AC C

As discussed in the previous section, it is important to determine and evaluate the consumed energy and resource flows of the process from an environmental perspective to understand and improve the grinding wheel and the grinding process. However, a number of input data in previous studies have been excluded for the material and electrical energy demand of the grinding wheel, the cutting fluid and the entire grinding system. The achievement was not possible due to:
limited knowledge of the energy and materials flows of the grinding wheel production process due to absent data;
limited consideration of all phases of a grinding wheel life cycle when performing an LCA;
limited consideration of different impact categories when presenting the grinding wheel environmental impact. Therefore, this study aims to overcome the stated limitations by analysing and modelling a screening LCA for the production process of cBN grinding wheels (cradle-to-gate) and the application and disposal of the grinding wheel (cradle-to-grave). The external cylindrical grinding process is used in a 6 of 26

ACCEPTED MANUSCRIPT case study. The outcomes of this research can be beneficial for the grinding wheel manufacturers to improve the environmental performance of the grinding wheel, as well as for the engineers and operators who are involved in operating the external cylindrical grinding process.

SC

RI PT

The relevant materials and processes of the grinding wheel production are included in the screening LCA through extensive literature review. Two different cBN synthesis catalysts and three different wheel hub materials, which have different disposal processes, are investigated. Based on a case study, the process conditions of the external cylindrical grinding strategies are varied to represent two selected grinding speeds and four different specific material removal rates. Results of the screening LCA are modelled using Umberto software in different environmental impact categories based on the CML2001 and the Eco-Indicator 99 H/A methods. Both results are generated to: 1) compare their results with the previous studies and 2) to evaluate the human health, ecosystem quality and resource use damage according to ISO 14040s of LCA.

Materials and Methods

3.1

Materials for the manufacturing of grinding wheels with cubic boron nitride abrasives

M AN U

3

According to Figure 2, a grinding wheel consists of a cBN abrasive layer and a wheel hub. The cBN layer is formed by three main components, namely cBN grains, bond (system) and porosity (pore builder and other additives). Figure 3 presents a simplified process and material flow chart for the production of a cBN grinding wheel with a vitreous, organic/synthetic resin and metallic sintered bond. Hexagonal boron nitride Mixing hBN/catalyst powder HPHT synthesis

Compound 1

Pore builder

EP

cBN grains

TE D

Catalyst

Other additives

Compound n Mixing Bond

AC C

Mixing

Legend Material Process step

Figure 3: 7 of 26

cBN Mixture Moulding cBN green body

Compound 1

Compound j

Curing

Preparation

cBN element (full body / layer / segment)

Wheel hub Finishing & testing cBN grinding wheel

Simplified material and process flow in producing vitreous, organic/synthetic resin and metallic sintered bonded cBN grinding wheels.

ACCEPTED MANUSCRIPT The materials and process steps shown in the figure are presented in greater detail in the following sections. 3.1.1

Production process of a cubic boron nitride grinding wheel

RI PT

Figure 3 shows the general process steps (mixing, moulding and curing) applied for the production of vitreous, organic/synthetic resin and metallic sintered bonded cBN elements. The difference lies within the used materials, therefore it requires process step specific parameters (time, temperature and pressure). The process steps are not valid for the electroplated bond, due to a different process route using the electro deposition procedure.

M AN U

SC

CBN abrasive grains, bond, pore builders and other additives are mixed at a defined ratio and then pressed into a green body by a moulding process. The curing or calcinations of a vitreous bonded cBN element can be performed by applying a cold pressing process or a hot pressing process. In the cold pressing process, the cBN element is cured within several hours at a temperature range between 700 and 950 °C with no additional pressure [20]. At hot pressing, the process takes only several minutes under the pressure of up to 150 MPa within a temperature range between 700 and 950 °C [21]. Organic/synthetic resin bonded cBN green body elements are commonly cured in a hot press mould for 30 to 120 minutes, at a temperature of at least 150 °C and a pressure between 35 and 105 MPa [9]. The metallic sintered bonded cBN elements can be produced similarly to the hot or cold pressed vitreous cBN elements. The cBN green body element is cured within 15 to 60 minutes, at temperatures between 600 °C and higher than 1,100 °C and a pressure between 14 and 140 MPa [9].

3.1.2

AC C

EP

TE D

A full body cBN grinding wheel or a wheel hub coated with a cBN layer or segments can be produced depending on the grinding wheel design, size and working speed requirements. The full body cBN grinding wheels are mostly common for small grinding wheels and low grinding wheel speed, as used in an internal grinding process. With increasing grinding wheel size and grinding wheel speed, a cBN layer/ring or cBN segments are coated on a wheel hub. The wheel hub materials are commonly composed of low alloyed steel, aluminium, bronze, ceramic or synthetic resin (fibre-reinforced; with metallic or non-metallic fillers) [10]. The selection of the wheel hub material depends on the requirements of the grinding process regarding the maximal burst resistance value and grinding wheel speed. The burst resistance is tested, as a final process step, to make sure that the grinding wheel meets the defined specifications. Synthesis of cBN abrasive grains

The synthesis of cBN grains, discovered by R.H. Wentorf in 1957 [22], is based on the transformation of boron nitride from the hexagonal to the cubic crystal structure with or without using a catalyst or solvent. The hexagonal boron nitride (hBN), also called α-BN, g-BH or white graphite, represents a binary compound of boron and nitrogen (the neighbouring elements of carbon). Furthermore, the compound has a layered structure with stacked sheets of six-membered rings similar to graphite. This similarity led to the assumption that hBN can be transformed into a zinc-blend cubic structure, using a high pressure and high temperature (HPHT) synthesis, similar to the diamond production [18]. Besides the application as a primary material for the cBN grain application, hBN powder can be used as a solid lubricant, a mould-release agent for glass, an insulating and heat-dissipating material and a material for cosmetics [23]. The hBN powder can be produced in an industrial scale by three main reactions: the reaction of boric oxide with ammonia, the reaction of boric oxide with organic 8 of 26

ACCEPTED MANUSCRIPT nitrogen compounds (i.e. urea, melamine) and the nitridation of calcium hexaboride in the presence of boric oxide [24].

M AN U

SC

RI PT

The synthesis of cBN can be performed with and without employing a catalyst or solvent. It can be argued if the term catalyst or solvent is correct, because these substances are not active as a catalyst or a solvent, they rather have the function as a flux precursor. This flux precursor creates a eutectic with hBN, in which the boron nitride is partially dissolved [25]. However, in literature, the term catalyst is common and therefore used in the following sections as well. The application of a catalyst has influence on the equilibrium pressure and temperature when the hBN is transformed to cBN. Therefore, the synthesis with catalyst leads to a reduction of the pressure and temperature level by half, compared to the synthesis without catalyst [26]. Guo et al. reported that about 50 kinds of catalysts can be used to support the hBN transformation to cBN (state 2010) [27]. These catalysts are based on alkali metals and alkali earth metals. Furthermore, the nitrides and boron nitrides of both metal types can also be applied. The most positive effects on the transformation process have been found when using lithium, calcium, barium, magnesium and their boron nitrides (e.g. Li3BN2, Mg3B2N4, Ca3B2N4, LiCaBN2 and LiBaBN2). In an industrial scale, catalysts are widely used in the transformation process to increase the productivity and to reduce the production costs, due to energy demand and die tool wear.

TE D

As mentioned before, HTHP are necessary to break and change the nature of the chemical bonds for the synthesis of cBN from hBN, with or without applying a catalyst. This can be conducted using HPHT machines. Different HPHT apparatus design types can be distinguished, such as toroid type (also called Bridgman type) [28], belt type [29], tetrahedral type [30] or cubic anvil type [31]. The level of pressure and temperature needed for the cBN synthesis depends on several factors such as particle size, oxygen content and purity of the used starting material and the application of a catalyst [25]. This relationship is exemplarily presented in Figure 4.

Direct

EP

6

AC C

Pressure (GPa)

7

cBN (stable)

Eutectic Ca3B2N4 - BN Eutectic Mg3B2N4 - BN

Li3N

Ca / Ca3B2N4

5

Bundy-Wentorf Equilibrium line

4

1000

Figure 4:

Eutectic Li3BN2 - BN

hBN (stable) 1500 Temperature (K)

2000

The formation regions of cBN with and without application of catalysts [25].

The figure shows the hBN to cBN conversion regions in a (P, T) phase diagram with and without the presence of a catalyst (Ca3B2N4, Mg3B2N4, Li3BN2, Li3N, Ca and Ca3B2N4); additionally to the mentioned 9 of 26

ACCEPTED MANUSCRIPT catalysts, up to 50 different kinds can be used [27]. According to Vel et al., stable cBN can be obtained in conversion processes between temperatures of 1,200 to 2,000 °C and pressures of 2.5 to 7.5 GPa [25]. Besides the HPHT process, cBN can also be produced with a low-pressure synthesis using chemical or physical vapour deposition [32]. 3.1.3

Bond systems, pore builders and other additives

M AN U

SC

RI PT

Four different bond systems can be distinguished (vitreous bond, organic/synthetic resin bond, metallic sintered bond and electroplated bond). Alkali-earth borosilicate glasses are commonly used as a vitreous bond system. This glass is composed of silica, boric oxide, aluminium oxide, alkali metals (e.g. sodium oxide or potassium oxide) and alkali-earth metals (e.g. calcium oxide or magnesium oxide). The organic/synthetic resin bond systems can be produced from plastic, polyimide resin, phenolic resin or phenol-aralkyl formulations. The metallic sintered bond systems are made from bronze in the copper-tin alloy with an alloy composition range between 60:40 and 85:15 and additional fillers [11]. The electroplated bond system enfolds a process where a single layer of cBN abrasive grains is directly bonded to the wheel hub by an electro deposition of nickel or a nickel alloy [9].

AC C

EP

TE D

As mentioned before, the pores are either naturally formed or induced by pore builder additives. By means of pore builder, a defined porosity and permeability of the grinding wheel can be achieved. Both porosity and permeability are important regarding the metal removal process performance. To obtain a certain level of porosity, three process methods can be conducted: the burn-out method, the closed cell or bubble method and the wash- or melt-out method. The burn-out method creates pores by the addition of an organic pore builder, which is thermally decomposed during the curing process of the grinding wheel and leaves free areas within the tool matrix. Such materials are ground walnut shells, beads of naphthalene or plastic, paraffin wax and other organic granules. Pores are created in the cell or bubble method by using hollow materials embedded in the abrasive bond mix. Such materials are foamed glass particles, bubble mullite and bubble alumina, and combinations of those materials [33]. In the wash- or melt-out method, paraffin wax beads [34] or salts (e.g. oxalic acid) [35] are used as pore builder. The wax beads or salt crystals are mixed with the abrasive grains and bond, which are then formed into a grinding wheel green body. Then, the wax is removed by liquefaction at a higher temperature (<350 °C). If salt is used instead of wax, then the green body is heated above the salt’s decomposition temperature to achieve a change from the solid to the gaseous phase (<400 °C). Beside the advantage to create a defined pore volume, there are also some drawbacks using pore builder, for example closed porosity, high spring-back, health and safety hazard, reburning, high moisture sensitivity and incomplete thermal decomposition [33]. Other additives can be used beside the pore builder to facilitate or to restrain specific grinding wheel characteristics during the manufacturing process or application. To enhance the cohesion of the green pressed grinding wheel, a temporary binder is added (e.g. dextrin, water, paraffin wax, animal protein glue, starch and ethylene glycol). Great portions of these binders are thermally decomposed during the curing process. Secondary abrasive grains as well as secondary active and inactive fillers are used to keep a specific content of primary cBN abrasive grains, while keeping a defined amount of bond. Secondary abrasives enfold grains made of aluminium oxide, silicon carbide, flint, garnet grains etc. Different halogen salts or metal sulphides can be applied as secondary filler. To enhance 10 of 26

ACCEPTED MANUSCRIPT the grinding wheel performance during the material removal process, graphite, hBN or molybdenum disulfide can be added to achieve a self-lubricating effect [36]. 3.2

Method

3.2.1

Case study - Objective and scope definition

RI PT

Objective of this study is the determination of the environmental impact for one grinding wheel within a cradle-to-gate and a cradle-to-grave perspective respectively. The assessment is based on the demanded energy and resource flows.

Material extraction

Production

Figure 5:

Grinding process

Production

Usage

End-of-Life

End-of-Life

EP

Cradle-to-gate perspective of one grinding wheel

Capital goods

Auxiliary goods

Material extraction

TE D

Cradle-to-grave perspective of one grinding wheel

M AN U

SC

The two system boundaries of the screening LCA for the grinding wheel and the whole life cycle of the grinding wheel are depicted in Figure 5. They are cradle-to-gate and cradle-to-grave. The cradleto-gate assessment highlights the environmental impact upon material extraction and production of one grinding wheel. The cradle-to-grave perspective covers all life cycle stages for the horizontal flow of the grinding wheel enfolding a cradle-to-grave evaluation with material extraction, production, usage and EOL stage. During this usage phase, the grinding wheel is applied in the grinding system consisting of an external cylindrical grinding machine, a cutting fluid filter and an exhaust air filter. In this system, the required primary energy and resource flows, such as demand of electrical energy and cutting fluid, are considered.

Horizontal and vertical interactions of the grinding wheel with the grinding system.

AC C

The transport of workpieces and the disposal of grinding swarf and cutting fluid are excluded, as well as the production of capital goods (e.g. buildings, machinery and equipment) and auxiliary resources (e.g. hydraulic fluid, compressed air and filter paper for the cutting fluid filtration). The functional unit for the cradle-to-gate phase are the material extraction and production phase of a vitrified bonded grinding wheel with a straight profile, an external diameter of 400 mm, width of 15 mm and an abrasive layer thickness of 5 mm. Furthermore, a ceramic, a steel and an aluminium wheel hub are considered. Subsequently, the cradle-to-grave is an extension of the cradle-to-gate results for the three different types of hubs of the grinding wheels. The definition of the functional unit of the cradle to grave covers material, production, usage and end-of-life of grinding wheels to realise an external cylindrical grinding process of 12,000 workpieces. Two grinding strategies are considered, differing in regard to the selected cutting speed (vc) of 50 m/s and 100 m/s. These

11 of 26

ACCEPTED MANUSCRIPT grinding strategies are also tested under four different specific material removal rates (Q’w) which include 2.5, 5.0, 7.5 and 10.0 mm³/(mm·s). 3.2.2

Case Study - System description and life cycle inventory

RI PT

The LCI of the case study is mainly based on input data that was collected from various sources including experiments, literature, company data and the Ecoinvent 2.2 database [37]. The following subsections are provided to outline the LCI of the grinding wheel, grinding system and process as well as the cutting fluid. The grinding wheel subsection is presented to elaborate the LCI data used in both cradle-to-gate and cradle-to-grave analyses as portrayed in Figure 5. The next subsection summarises the LCI of grinding system and process including the cutting fluid for the cradle-to-grave analysis. Grinding wheel

M AN U

SC

For the grinding experiments, a vitrified cBN grinding wheel with the specification B 107 VSS T 140 was used. The tool had a diameter of 400 mm and a width of 15 mm. The thickness of the grinding wheel abrasive layer was 5 mm; however, it was assumed that only 4 mm of the abrasive layer can be used. The maximal utilization potential of 2,000 possible dressing times for one grinding wheel is calculated based on an assumption that the maximal acceptable radial grinding wheel wear (Δrs) is 10 µm before redressing is necessary.

AC C

EP

TE D

The used abrasive layer is composed of 31.25 Vol.-% cBN grains, 20.25 Vol.-% vitrified bond, 20.00 Vol.-% secondary grains (made of aluminium oxide (Al2O3)) and 28.50 Vol.-% pores. Figure 6 presents a simplified material and process flow for the production of a vitrified cBN grinding wheel. The input values and ratios of the raw materials and production processes were mainly collected from literature such as material data sheets, publications and patents. On this basis, the needed time, temperature and pressure of each process step was derived. The electricity demand during the process steps was derived from literature and associated company data by comparing their electricity demand and operating conditions such as operating time. Raw materials for the cBN grain production include hBN and a catalyst. The production of the hBN was modelled based on the reaction of boric acid with organic nitrogen compounds urea, according to Lipp et al. [24] and Ecoinvent Centre [39]. Two catalysts were considered in this research, namely lithium calcium boron nitride (LiCaBN2) and lithium barium boron nitride (LiBaBN2). These catalysts were selected to show the different impacts regarding cBN conversion rate and environment [40]. The production process of these catalysts was modelled in accordance with Eagleson [41]. The conversion ratio of input materials to cBN grains was 35.6 % for LiCaBN2 and 40.1 % for LiBaBN2 [42]. It was assumed that the different catalysts have no influence on the grinding wheel or the abrasive performance. For the HTHP synthesis, a pressure of 5 GPa, a temperature of 1,450 °C and a duration of 15 minutes were considered. The electrical energy demand during the synthesis was assumed based on Ali with 0.52 kW per gram cBN [18]. A vitrified bond composed of silicon dioxide (SiO2), aluminium oxide (Al2O3), boron trioxide (B2O3), calcium oxide (CaO), magnesium oxide (MgO) and sodium oxide (Na2O) was selected and modelled based on Li [43]. The other additives include secondary grains and temporary binder. The secondary grains are made of Al2O3 and have the same grain size as the cBN grains. Two kinds of temporary binder were considered with dextrin and urea formaldehyde resin [44]. The dextrin was modelled based on the chemical organic process, as there is no direct LCI database for dextrin while the urea 12 of 26

ACCEPTED MANUSCRIPT formaldehyde resin is based on a process case obtained from the Ecoinvent 2.2 database. Both binders were burned out during firing and were assumed to have no influence on the finished abrasive layer.

Secondary Al2O 3 grains (34.4 wt-%) Dextrin (1.0 wt-%) Urea formaldehyde resin (4.7 wt-%)

Mixing 0.5

25

1

hBN/catalyst powder HPHT synthesis 15

1,450 50,000 cBN grains (46.7 wt-%)

Pore builder (0.0 wt-%)

Calcium oxide (1.5 wt-%) Aluminium oxide (5.2 wt-%) Silicon dioxide (66.0 wt-%)

SC

LiBaBN2 or LiCaBN2 (23.8 wt-%)

M AN U

hBN (79.2 wt-%)

RI PT

In the further calculations, three different wheel hub materials were considered by using low alloyed steel, wrought aluminium alloy and ceramic. The composition of the ceramic hub complies with the composition of conventional full body grinding wheels. This composition enfolds 15.0 wt.-% of the vitrified bond, 79.4 wt.-% Al2O3 grains, 0.9 wt.-% dextrin and 4.7 wt.-% urea formaldehyde resin. It is assumed that similar mixing, pre-forming and burning conditions are applied for the production of the cBN abrasive layer. It is also assumed that the wheel hub material can be changed without any influence on the grinding process.

Other additives (40.1 wt-%)

Sodium oxide (5.0 wt-%) Magnesium oxide (0.1 wt-%) Boron trioxide (22.3wt-%)

Mixing 0.5

25

1

Bond (13.2 wt-%)

Mixing 1

25

1

TE D

cBN Mixture Moulding

0.67

Steel, ceramic or aluminium hub

Curing

Preparation

EP

950

1

cBN layer

Legend

AC C

Material

Process step

Figure 6:

20

cBN green body

960

Time [min]

25

Temp. [°C]

Pressure [bar]

Epoxy resin

Wheel hub

Assembly / Mounting 300

190

1

cBN grinding wheel

Simplified material and process flow for the production of a vitrified cBN grinding wheel (based on literature values).

The EOL options for the different grinding wheel materials were assumed in accordance with the common municipal waste management in Europe. The cBN waste and the ceramic wheel hub are assumed to be 100 % landfill whereas 100 % recycling was assumed for the aluminium and steel hub. The steel or aluminium hub scrap is transported to the recycler and is recycled into a secondary scrap. This scrap is then sent back as input for the ingot production. The distribution and disposal transportation was assumed as 100 km for the ease of assessment. Instead of the direct recycling of the steel or the aluminium hub, another alternative option could be the reuse of the metallic wheel 13 of 26

ACCEPTED MANUSCRIPT hubs. In this case, the wheel hubs are sent back to the grinding wheel manufacturer, the old abrasive layer is stripped off and a new abrasive layer is coated on the wheel hub. Yet, this option is not considered in this research to ensure that the EOL phase of the three hubs are compared based on a unified basis. Grinding system and process as well as cutting fluid

M AN U

SC

RI PT

The external cylindrical grinding process was performed on a Studer S40 CNC universal cylindrical grinding machine. A cutting fluid filter (gravity belt filter) and an exhaust air filter (mechanical filtration) were connected to the grinding machine. As mentioned earlier, two grinding wheel cutting speeds were used (50 m/s and 100 m/s). Due to a fixed velocity ratio of qs = 100, the workpiece speed was 0.5 m/s and 1 m/s respectively. For dressing, a diamond dressing roll with a radius of Rsp = 1.2 mm was used with an infeed of aed = 2 µm and dressing overlap of Ud = 5. The assumed number of possible dressing times was 2,000 dressing strokes. With each cutting speed four experimental series were performed, which differed in regard to the used specific material removal rates (Q’w = 2.5, 5.0, 7.5 and 10 mm³/(mm·s)). In the grinding process solid shafts of hardened carbon alloy steel (62 HRC) with the designation 1.3505 were used as workpieces. At each workpiece, the volume of material removed by cutting was Vw = 1,000 mm3.

TE D

Measurands of the grinding process were the electrical power (Pc), the surface roughness (Rz) and the radial grinding wheel wear (Δrs). The electrical power due to cutting and the base load of the grinding machine, cutting fluid filter and exhaust air filter were metered by a three-phase power analyser (Load Controls® PPC-3). The surface roughness was measured at four different points on the workpiece by a surface measurement device (Hommel-Etamic® T1000 basic). The radial grinding wheel wear was measured with the flat blank method.

EP

A mineral oil based emulsion with a concentration of 8 % was used as cutting fluid. The emulsion had a kinematic viscosity of ν40 °C = 1.2 mm2/s. A total volume of 1,200 litres of cutting fluid was stored in the cutting fluid filter. It was assumed that lost cutting fluid is replaced continuously. The rates of cutting fluids loss, due to drag out via the exhaust air, grinding swarf, evaporation and workpieces were experimentally determined. The environmental impact of the fluid was calculated on the basis of Ecoinvent 2.2 database [37] and Winter et al. [38].

AC C

The experimental results of the external cylindrical grinding process are presented in Figure 7. The figure shows the measured surface roughness (Rz) in the upper diagrams, the cutting power (Pc) for the two cutting speeds in the centre diagrams and the radial grinding wheel wear (Δrs) in the lower diagrams. Four specific material removal rates were tested for each cutting speed by grinding six workpieces in a row. The grinding wheel was dressed after the sixth workpiece, which equates to an overall volume of Vw = 6,000 mm3 of workpiece material removed. The figures include the aforementioned boundary of acceptable parts, represented by the dashed line. The boundary of the surface roughness is Rz = 1.7 µm and for the radial grinding wheel wear Δrs = 10 µm. The experimental results indicate that increasing cutting speed leads to lower surface roughness, lower radial grinding wheel wear and increasing cutting power, which corresponds with results from Winter et al. [3] and Klocke and König [10]. The lower surface roughness and wear is caused by an 14 of 26

ACCEPTED MANUSCRIPT increasing cutting speed and a decreasing cutting depth given that the ratio between workpiece speed and grinding wheel speed is constant. The lower cutting depth leads to a reduction of the forces per grain and hence the overall grinding forces as well. Lower forces cause a lower grinding wheel wear and a lower surface roughness [45]. If the cutting speed increases, the number of grain passes over the workpiece surface and the friction between workpiece and grain will be inclined, accordingly the cutting power increases [45]. vc = 50 m/s and 100 m/s ap = 10 mm vw = 0.5 m/s and 1.0 m/s Ud = 5 aed = 2 µm

Q‘w = 2.5 mm³/(mm·s)

Cutting fluid: Tool: Specification: Workpiece: Dresser:

Mineral oil based emulsion (8 %) cBN, vitrified bonded B 107 VSS T 140 1.3505 (DIN 100Cr6), 62 HRC Diamond form roll

RI PT

2,00 2.00

Q‘w = 5.0 mm³/(mm·s)

Q‘w = 7.5 mm³/(mm·s)

Boundary for acceptable parts

1,50 1.50

Q‘w = 10.0 mm³/(mm·s)

SC

z

1,00 1.00

Values above the boundary are not considered

0,50 0.50

M AN U

Surface roughness Rz [µm]

Cutting speed: Grinding width: Workpiece speed: Dressing overlap: Dressing infeed:

0,00 0.00 1

2

3

4

5

6

1

2

3

4

5

6

1

2

3

4

5

6

Workpiece [-]

Q‘w = 2.5 mm³/(mm·s)

Q‘w = 5.0 mm³/(mm·s)

1000 c

750 500 250 0 1

2

3

4

5

TE D

Cutting power P c [W]

1250

6

1

2

3

4

5

Q‘w = 7.5 mm³/(mm·s)

6

1

2

3

4

EP

Q‘w = 5.0 mm³/(mm·s)

Boundary for acceptable parts

15,00 15.00 10,00 10.00

AC C

Radial tool wear ∆rs [µ m]

Q‘w = 2.5 mm³/(mm·s)

20,00 20.00

2

3

5

4

5

6

vc = 100 m/s

Q‘w = 10.0 mm³/(mm·s)

6

Workpiece [-]

25,00 25.00

1

vc = 50 m/s

1

2

3

vc = 50 m/s

Q‘w = 7.5 mm³/(mm·s)

4

5

6

vc = 100 m/s

Q‘w = 10.0 mm³/(mm·s)

Values above the boundary are not considered

5,00 5.00 0,00 0.00

1

Figure 7:

15 of 26

2

3

4

5

6

1

2

3

4

5

6

1

2

3

4

5

6

Workpiece [-]

Experimental results of the external cylindrical grinding process.

1

2

vc = 50 m/s

3

4

5

6

vc = 100 m/s

ACCEPTED MANUSCRIPT

4

Results and discussion

Results are assessed for: 1) one grinding wheel for cradle-to-gate (material and production) and for 2) a grinding process used to produce 12,000 workpieces for cradle-to-grave (all life cycle stages). 4.1

Cradle-to-gate impact assessment of one grinding wheel

RI PT

CML2001 and Eco-Indicator 99 H/A (EI99) were selected to evaluate the grinding wheel impact. These methods were selected as they were developed on the basis of a European database. CML2001 was chosen to generate the carbon footprint results in a unit of kilogram of carbon dioxide equivalent (kg CO2eq.). EI99 assesses the impact categories that are associated with three damage categories, namely human health, ecosystem and resources, which can then be summarised into a single score in a unit of points.

EP

TE D

M AN U

SC

Figure 8 depicts a Sankey diagram for the LiBaBN2 and LiCaBN2 production. It highlights the variations of the distributions of materials in kg and electricity in kWh for the production of LiBaBN2 (Figure 8 (a) and (b)) and LiCaBN2 (Figure 8 (c) and (d)). The model was developed using Umberto 5 software, where the square symbol represents the processes (e.g. T4 Electricity) involved in both productions. The green circle that has a vertical line on the left hand side represents inputs (e.g. P1 Raw materials) used by a process. The red circle with a vertical line on the right hand side (e.g. P2 Emissions), represents the output of a process. For example, T4 is an electricity process which uses raw materials from P1 and emits an output as electricity to P6 as well as emissions to P2. The yellow double circle (e.g. P6 Electricity in Figure 8 (a)) is a connection. The connection is created in the model to connect the output produced by the previous process (e.g. T5 Barium) which then became an input of the next process (e.g. T2 Ba3N2). This connection can be duplicated and used as an input for other processes, for example, connection P6 electricity is used as input for T1, T5, T2 and T3 processes in Figure 8 (a). The double square symbol represents a subnet. It enfolds the presentation of inputs, processes and outputs that are used within the subnet. For example, T5 Barium subnet uses P6 Electricity and P1 raw materials as inputs for a number of processes such as barite, hydrochloric acid, barium chloride processes which are used to produce outputs as P5 Barium and P2 Emissions. The LCI data of the model was mainly based on the Ecoinvent 2.2 database as well as input data collected from associated literature.

AC C

Generally, the cBN grain produced by using a LiBaBN2 catalyst (P7 in Figure 8) demands 11.22 % less electrical energy via a synthesis process, compared to the application of a LiCaBN2 catalyst. The impacts of these differences of both catalysts are reflected within the results of the hBN production (P8 in Figure 8). The impact of LiBaBN2 is made up by 50 % from electricity, 20% from nitrogen, 15% from barium and 12% from lithium, whereas for a LiCaBN2 catalyst 55 % stem from nitrogen, 34 % from lithium, 8 % from electricity and 3 % from calcium. On this occasion, nitrogen requires high electricity for processing, cooling water and waste heat.

16 of 26

ACCEPTED MANUSCRIPT a) LiBaBN2 material flow (kg) LiBaBN2 hBN Ba3N 2 Li3N Lithium, at plant [GLO] Nitrogen, liquid, at plant [RER] Barium (Ba (Ba2) 2)

b) LiBaBN2 electricity flow (kWh) P1: Raw material

P1: Raw material

P2: Emissions

T7:Nitrogen

Electricity, production mix DE

P2: Emissions

T7:Nitrogen

P6: Electricity

P6: Electricity 3.9 kWh

P1: Raw material

P3

T5:Barium P5:Barium

T6: Lithium

T5:Barium P5:Barium

T6: Lithium

P1: Raw material

P6: Electricity

P6: Electricity

P3

P6: Electricity T1: Li3N

P6: Electricity T1: Li3N

T4:Electricity

T4:Electricity

P18:Waste

P18:Waste 4.1 kWh

T2:Ba3N2 P4

P4

P2: Emissions

P2: Emissions

P8:hBN

P8:hBN

T3:LiBaBN2

T3:LiBaBN2

P7:LiBaBN2

P7:LiBaBN2

c) LiCaBN2 material flow (kg)

d) LiCaBN2 electricity flow (kWh)

P1: Raw material

Electricity, production mix DE T5:Calcium P5: Calcium

T6:Lithium

T4:Electricity

P8:hBN T3:LiCaBN2 P7:LiCaBN2

P8:hBN T3:LiCaBN2 P7:LiCaBN2

Sankey diagrams of the material and energy flows for the LiBaBN2 and the LiCaBN2 productions.

TE D

Figure 8:

P4

P2: Emissions

P2: Emissions

T2:Ca3N2 P18:Waste

P18:Waste P4

P5: Calcium P6: electricity

P6: Electricity T1: Li3 N

T2:Ca3N2

P6: Electricity T1: Li3N T4:Electricity

T5:Calcium

P3

P1: Raw material

P3

P1: Raw material

P2: Emissions

T7:Nitrogen

T6:Lithium

P6: Electricity

M AN U

Li3N Lithium, at plant [GLO] Nitrogen, liquid, at plant [RER] Calcium ion (Ca Ca22++)

P1: Raw material

P2: Emissions

T7:Nitrogen

SC

LiCaBN2 hBN Ca3N 2

RI PT

T2:Ba3N2

EP

Based on the models presented in Figure 8, a cradle-to-gate impact assessment for one grinding wheel is presented in Figure 9. The upper part of the figure presents the boundary conditions of the grinding wheel. The centre part shows the aggregated CML2001 and EI99 results. The lower part presents individual EI99 results for different impact categories. The figure focuses on the impact caused by the extracted materials, the production of the cBN abrasive layer and the wheel hub. Two different types of catalysts, namely LiBaBN2 and LiCaBN2, are compared.

AC C

Generally, both carbon footprint and EI99 results have the same trend. For the cBN layer, the production phase makes up the largest contribution due to the energy demand during the cBN synthesis. The grinding wheel made from the LiBaBN2 catalyst reduces the overall carbon footprint when compared to the LiCaBN2 catalyst by 10.34 % in the production phase and 3.31 % due to the extracted materials. Approximately 90 % was caused by the electricity used during the mixing and synthesis processes of the catalyst and hBN to produce the cBN grains (see the top left corner of Figure 3). The variations found between the two catalysts were caused mainly due to the fact that the LiBaBN2 catalyst has a higher yield during the cBN synthesis than the LiCaBN2 catalyst. The remaining share was the impact from electricity used to produce the cBN element from the cBN grains, pore builder, bond and other additives. The contribution during the material extraction phase for the cBN layer was originated from 85 to 90 % of the hBN production, where 5 % (LiCaBN2) and 13 % (LiBaBN2) came from the catalyst. The impact of hBN production is caused by approximately 60 % from electricity, 30 % from urea formaldehyde resin and 10 % from boric acid and calcium 17 of 26

ACCEPTED MANUSCRIPT nitride. Although the vitrified bond has an impact of less than 1 %, it is worth mentioning that boron trioxide contributes the most (60 % to 70 %), followed by sodium oxide (15 % to 25 %), aluminium oxide (8 % to 15 %) and the remaining is from quick lime and silica sand.

Grinding wheel materials

Carcinogenics 10 Fossil fuels 1 0.1 0.01 Mineral 0.001 extraction 0.0001 0.00001

Ionising radiation

Ozone layer depletion Respiratory effects

Steel hub

10.0

7.5

5.0

RI PT

10 1 0.1 0.01 0.001 0.0001 0.00001

Fossil fuels

Mineral extraction

Land occupation

Ecotoxicity

TE D

Figure 9:

Eco-Indicator'99 (Points) Grinding wheel production

Carcinogenics

Acidification & eutrophication Aluminium hub

dh

Comparison cBN catalyst (Cradle-to-gate)

Climate change

Land occupation Ecotoxicity

Steel hub Ceramic hub

dhm bs

M AN U

Comparison hub material (Cradle-to-gate)

Aluminium hub

Steel hub Ceramic hub

ds

2.5

150

100

50

0

Carbon footprint (kg CO2eq)

Aluminium hub

0.0

Steel hub Ceramic hub

ds = 400 mm bs = 15 mm dh = 390 mm bh = bs mm dhm = 127 mm

SC

Aluminium hub

Steel hub Ceramic hub

Tool diameter: Tool width: Hub diameter: Hub width: Hub mounting:

Catalyst: LiCaBN LiCaBN2 2

Aluminium hub

cBN abrasive composition: cBN: 31.25 Vol.-% Bond: 20.25 Vol.-% Pores: 20.00 Vol.-% Secondary grains: 28.50 Vol.-%

Catalyst: LiBaBN LiBaBN2 2

Catalyst: Catalyst: LiCaBN LiCaBN2 2

B 107 VSS T 140 vitrified LiBaBN2 , LiCaBN2 Ceramic, steel, aluminium

Catalyst: Catalyst: LiBaBN LiBaBN2 2

Tool specification: Bond type: cBN catalyst: Hub material:

Ceramic hub

Climate change Ionising radiation Ozone layer depletion Respiratory effects

Acidification & eutrophication LiBaBN2 catalyst LiBaBN2

LiCaBN2 catalyst LiCaBN2

Cradle-to-gate impact assessment of one grinding wheel.

AC C

EP

The impact of the wheel hub presented in the bar charts of Figure 9 differs significantly depending on the types of hub materials. The application of a ceramic wheel hub leads to a low carbon footprint and EI99 value, followed by the steel and the aluminium hub. Figure 9 shows two distinctive patterns of their distribution. First, the aluminium hub has the material production as a major impact. This is due to a high energy and resource demand during the aluminium extraction process, such as the aluminium smelting process that produces the aluminium ingot, when compared to a low energy demand when machining the aluminium ingot into a hub billet. Second, the ceramic and the steel grinding wheel hub are dominated by the production phase. On this occasion, more energy and resources are used in machining the steel hub billet and curing the ceramic grinding wheel when compared to a lower energy demand for the material extraction (processing, smelting, casting and rolling) during the steel production and the ceramic production process (processing, pre-forming and burning). The lower left chart of Figure 9 shows the total EI99 results of the cradle-to-gate results for the three different hubs in different impact categories. These results are the aggregated results of the top right bar chart of the figure. The chart shows that all hub materials have similar contributions across all 18 of 26

ACCEPTED MANUSCRIPT

RI PT

impact categories. The ceramic hub has the lowest values in all impact categories. The aluminium hub has the highest values in most impact categories, namely carcinogens, climate change, respiratory effects, acidification and eutrophication, ozone layer depletion, and fossil fuels impact categories. This is due to the fact that these impact categories have high association with higher amounts of emissions such as particulate matter. GHG, SOx and NOx are produced by the higher demand of electricity during the extraction and processing processes of the aluminium and the steel hub when compared to the production of the ceramic hub. An exception is found for the ecotoxicity, land occupation and mineral extraction impact categories, where the steel hub has slightly higher values than the aluminium hub. This reflects the fact that higher mining activities are required to obtain the amount of steel used in producing one hub (17 kg) when compared to the aluminium hub (5 kg).

Sensitivity analysis and discussion

M AN U

4.2

SC

The lower right bottom chart of Figure 9 compares the results of the impact categories when using the two different cBN synthesis catalysts. It can be seen that the LiCaBN2 catalyst value is significantly higher compared to the LiBaBN2 for the climate change category. This is predominantly due to the large electricity demand during the production of nitrogen.

EP

TE D

The previous section highlighted the hotspots of the results, which were caused by the following factors: 1) the energy demand during the cBN synthesis, 2) the level of cBN concentration in the abrasive layer and 3) the energy demand for curing the ceramic wheel hub. In practice, these parameters may change due to the variations of production and applications. Hence, three sensitivity analyses were conducted to test such uncertainties of the three factors, which have significant influence on the LCA results. The carbon footprint and the EI99 results of the sensitivity analyses are presented in Figure 10 for one grinding wheel. This figure was calculated on the basis of the results presented in Figure 9. The figure shows the value for the presented status quo (left side) and the results of the sensitivity analyses (right side). Results for both status quo and the sensitivity analyses were based on the cradle-to-grave calculation of one grinding wheel made from a LiBaBN2 catalyst and a ceramic wheel hub. The material extraction (ME) and the production (PP) are highlighted with an individual bar, which explicitly shows the contributions of the impact caused by the cBN layer and the wheel hub.

AC C

According to the previous section, the production stage of cBN was the main hotspot when compared to other life cycle stages. This was caused by the assumed electricity demand of 520 kW for processing 1 kg of cBN from the catalyst and hBN, based on investigations of Ali [18]. An alternative number can be obtained from a study by Klimczyk et al. [46]. In this study, cBN based composites were produced from hBN via a HPHT synthesis. The synthesis of cBN based composites is similar to the synthesis of cBN grains in terms of the process structure. Therefore, it was assumed that the numbers provided by Klimczyk et al. can be used, and the electricity demand was calculated with 2,190 kW for processing 1 kg of cBN [45]. The results in Figure 10 show that the higher electrical energy demand during the synthesis process has a major impact on the overall carbon footprint and EI99 value. The carbon footprint and EI99 impact of the status quo were dominated by the grinding wheel hub material whereas the first sensitivity analysis shows that the impact is dominated by the cBN layer. The overall carbon footprint is increased by 80 % and the EI99 value by 60 % when applying the 19 of 26

ACCEPTED MANUSCRIPT electrical energy demand obtained from Klimczyk et al. This indicates that the LCA results may change in practice when the electricity demand during the production of cBN grains alters, owing to the differences of machines and operating systems.

Status quo

Grinding wheel: Specification: Bond type: cBN catalyst Hub material:

Sensitivity analysis

7.5 5.0

B 107 VSS T 140 vitrified LiBaBN2 Ceramic, steel, aluminium

RI PT

2.5 0.0 200

cBN abrasive composition: cBN: 31.25 Vol.-% Bond: 20.25 Vol.-% Pores: 20.00 Vol.-% Sec. grains: 28.50 Vol.-%

150

50 0

PP

ME

PP

ME

PP

ME

PP

M AN U

cBN cBN cBN Hub curing Hub curing energy content content energy energy Klimczyk -50% +50% 50% +50% Sensitivity analysis with ceramic wheel hub and LiBaBN2 catalyst

ME = Material extraction CBN layer

Figure 10:

ME

PP

Impact of:

ME

PP

ME

PP

ME

PP

Wheel Wheel with with Wheel aluminium with steel ceramic hub hub hub Status quo

Wheel hub

Grinding wheel design: Tool diameter: ds = 400 mm Tool width: bs = 15 mm Hub diameter dh = 390 mm Hub width: bh = bs mm Hub mounting: dhm = 127 mm

SC

100

ME

Eco-Indicator'99 (Points) Carbon footprint (kg CO2eq.)

Carbon footprint (kg CO2 eq.) Eco-Indicator‘99 (Points)

10.0

ds

dhm

dh

PP = Production process

bs

Results of the sensitivity analysis for one grinding wheel in comparison to the status quo.

AC C

EP

TE D

In the second sensitivity analysis, the cBN concentration changed from 125 % by ±50 %. Referring to Figure 3, the bond and pore content was assumed fixed, while the content of secondary grains (the other additives in Figure 3) changed in correspondence with the cBN grain concentration. For example, if the cBN concentration increases the secondary grain content decreases and vice versa. The selected cBN concentration has a major influence on the stock removal rate, the achieved surface roughness, the tool life, the profile retention and the procurement price. For example, a high cBN concentration leads to high profile retention and tool life. However, it also leads to a high procurement price. Both analyses were performed due to the high impact of the production phase of the cBN abrasive, as presented in Figure 9. The carbon footprint and EI99 impact decreases with declining cBN concentration. The opposite behaviour occurs when the cBN content increases. The change of the cBN content by ±50 % has a comparably low impact on the overall impact of the grinding wheel. The cBN content change results in a change of the carbon footprint by ±11 % and the EI99 value varies by ±13 %. The results show that the variation of the cBN content (75 % and 175 %) may considerably alter the impact of a grinding wheel by approximately 10 % due to the consequent amount of raw materials used in making a cBN layer. The third sensitivity analysis enfolds the impact of the curing energy used during the production of the ceramic wheel hub, which can be changed significantly in practice. In the sensitivity analysis, the actual curing energy demand of 15.52 kWh/kg was reduced and increased by 50 %. This analysis was performed due to the high impact of the production phase of the ceramic hub as presented in Figure 9. The results in Figure 10 indicate that an increase or decrease by ±50 % of the curing energy 20 of 26

ACCEPTED MANUSCRIPT demand during the ceramic wheel hub production process has a comparably high influence, though not as high as the impact found in the first sensitivity analysis. The carbon footprint and the EI99 results change by approximately ±22 % and ±29 % respectively.

Cradle-to-grave impact assessment of the grinding process

SC

4.3

RI PT

Three sensitivity analyses were performed to address the aforementioned uncertainties. Overall, the results indicate that the impact of the grinding wheel made from LiCaBN2 catalyst remains higher than the wheel using LiBaBN2 (see Figure 9). The results of the first two analyses indicated that the uncertainties may significantly change when compared to the majority of the status quo results but the ranking order of the three hub materials remains the same. The grinding wheel using the aluminium hub has always a higher impact than the steel and ceramic hub. Only the ranking order between the steel hub and the ceramic hub has changed when performing the third sensitivity analysis.

Table 1:

TE D

M AN U

As stated above, the functional unit of the analysis is the production of 12,000 workpieces. The number of required grinding wheels to grind this amount of workpieces depends on the selected cutting speed and specific material removal rate. Based on the experimental data presented in section 4.2.2, the necessary redressing times can be calculated and therefore the number of needed grinding wheels. The result of this calculation is presented in Table 1. The table shows the number of required grinding wheels depending on the selected cutting speed and specific material removal rate. As presented in Figure 7, the tool wear and the surface roughness increase with rising specific material removal rate, therefore the grinding wheel has to be redressed more often to fulfil the required technological demands (surface roughness, Rz ≤ 1.7 µm and radial grinding wheel wear Δrs ≤ 10 µm). By increasing the cutting speed, the demand of redressing decreased due to the aforementioned relationships. Number of required grinding wheels.

AC C

EP

Cutting speed (vc) [m/s] 50 100

Specific material removal rate (Q’w) [mm³/(mms)] 2.5 5.0 7.5 10.0 6 8 15 15 2 5 6 10

Based on the numbers presented in Table 1 and the results of Figure 9, the carbon footprint according to the CML2001 and EI99 assessment approach for a cradle-to-grave perspective can be calculated. The results are presented in Figure 11. The upper part of the figure presents the boundary conditions of the grinding wheel. The centre part shows the aggregated CML2001 results, and the lower part plots the aggregated EI99 results. Both aggregated results are presented for the three grinding wheels when applying different grinding speeds and specific material removal rates. The calculation is based on the application of LiBaBN2 during cBN synthesis. The demand of electrical energy and cutting fluid is considered in the usage phase. Moreover, it was also assumed that certain materials are also consumed during the usage phase such as abrasive materials, process energy and cutting fluid, which are independent from the materials used for the wheel hub.

21 of 26

ACCEPTED MANUSCRIPT

500 0

5

vc [m/s]

7.5 10 2.5 50

2.5

5

7.5 10 2.5

5

50

7.5 10

2.5

5

100

7.5 10 2.5

5

50

7.5 10 100

TE D

100 75 50

0

Q‘w [mm³/(mms)] 2.5

5

EP

25

7.5 10 2.5

5

7.5 10

50 100 Material extraction phase

AC C

Figure 11:

7.5 10

100

125

Eco-Indicator'99 (Points)

5

Aluminium wheel hub

M AN U

1000

Q‘w [mm³/(mms)] 2.5

Mineral oil based emulsion (8 %) cBN, vitrified bonded B 107 VSS T 140 1.3505 (DIN 100Cr6), 62 HRC Diamond form roll

Steel wheel hub

1500

vc [m/s]

Cutting fluid: Tool: Specification: Workpiece: Dresser:

Ceramic wheel hub

2000

Carbon Footprint (kg CO2eq.)

vc = 50 m/s and 100 m/s ap = 10 mm vw = 0.5 m/s and 1.0 m/s Ud = 8 aed = 2 µm

SC

Cutting speed: Grinding width: Workpiece speed: Dressing overlap: Dressing infeed:

RI PT

It is evident that both impact categories are dependent on the number of needed grinding wheels during the grinding process. The demand is therefore connected with the chosen grinding process parameters. The impact category results of the grinding process rises when the specific material removal rate is increased and decreases when there is an increase of the cutting speed. When comparing the results amongst the three hub materials, the aluminium wheel hub has the highest impact, followed by the steel wheel hub and the ceramic wheel hub. The presented carbon footprint is majorly dominated by the material extraction and the production phases of the grinding wheel. The EOL and the usage phase have a comparably low impact. Almost the same applies for the EI99 results; however, the impact of usage and EOL phase has a marginal higher share on the overall impact compared to carbon footprint. The usage phase has a comparably low influence on the two impact categories.

2.5

5

7.5 10 2.5

50 Production phase

5

7.5 10

100 Usage phase

2.5

5

7.5 10 2.5

5

50 End-of-Life phase

7.5 10 100

Cradle-to-grave impact assessment of the grinding process (Catalyst: LiBaBN2).

When comparing the results of the EOL phase presented in Figure 11, the aluminium hub has the highest impact in the EOL phase followed by the steel hub and the ceramic hub. The differences of their impact are influenced by the differences in their disposal processes. The ceramic hub is disposed within the common municipal waste which uses less energy when compared to the recycling process of the aluminium and steel hub that involved sorting and reprocessing the hub as secondary scrap materials. The scrap material can then be used afterwards as raw material for the production process of either aluminium or steel.

22 of 26

ACCEPTED MANUSCRIPT The abrasive layer made of either LiBaBN2 or LiCaBN2 catalyst, has a comparably low carbon footprint and EI99 values in the EOL phase. Both catalysts have the same value of impact as they are disposed in the same quantity with the same disposal strategy, the common municipal waste.

5

Conclusion

RI PT

Transparency is needed regarding the demand of energies and materials to calculate the environmental impact of a grinding process due to the usage of grinding wheels, cutting fluids and machine tools. The presented study investigated the impact of a grinding wheel with a vitrified bond and cBN grains produced by using two different synthesis catalysts and three different wheel hub materials. This combination was chosen due to its high application degree.

TE D

M AN U

SC

The needed raw materials, the production steps and influencing parameters to produce a cBN grinding wheel were presented based on comprehensive literature and patent study. The investigation included the synthesis of cBN from hBN using different catalysts, the production of a wheel hub and the manufacturing of the grinding wheel. On this basis, a model was developed to calculate the environmental impact regarding different impact categories in the context of an LCA approach. Data and results from previous studies in literature were compared and additional experiments were conducted to validate the calculated results of the sensitivity analyses. The calculation considered the application of different cBN synthesis catalysts and grinding wheel hub materials. Sensitivity analyses were conducted to test the impact of different cBN synthesis energies, cBN concentrations and ceramic grinding wheel hub curing energies. The results indicate a major impact, especially of the energy demand during the cBN synthesis and the curing of the ceramic hub. Another high impact factor is extracted materials and production processes to produce an aluminium or steel based grinding wheel hub.

EP

The derived data from the grinding wheel LCA model were applied in a case study to investigate the influence of different process parameters on an external cylindrical grinding process. The results indicate that high cutting speeds and low material removal rates lead to low overall environmental impact, due to a reduced tool wear.

AC C

To increase the accuracy, further investigations and measurements are needed particularly for the cBN grain production due to its comparably high impact. The study tried to cover all relevant steps during the grinding wheel production based on the process descriptions in literature. Nonetheless, some process steps may not be considered due to insufficient description in literature. This limitation can be overcome with more detailed process descriptions. A broad variety of different bond/grain/hub-combinations in industry can be further covered to achieve a use case-oriented calculation of the grinding process. In addition, further investigations regarding the demand of energies and materials during the production of wheel hubs and abrasive materials can be conducted to achieve more accurate results. This is to refine the results obtained from the current study that obtained all LCI results for the needed materials and energies over all life phases of a grinding wheel in an unprecedented detail. Lastly, an analysis of the economic impact can also be performed to complement the current findings for the environmental perspective of a grinding process.

23 of 26

ACCEPTED MANUSCRIPT

Acknowledgements The authors would like to acknowledge the Go8-DAAD scheme for supporting the Joint GermanAustralian Research Group on Sustainable Manufacturing and Life Cycle Management. The authors also kindly acknowledge the contribution of Mrs. Gerlind Öhlschläger.

[4] [5]

[6]

[7]

[8]

SC

AC C

[9]

M AN U

[3]

TE D

[2]

Malkin, S., Guo, C., 2008. Grinding technology: Theory and Application of Machining with Abrasives. Industrial Press Inc., New York, USA. Schischke, K., Hohwieler, E., Feitscher, R., König, J., Kreuschner, S., Wilpert, P., Nissen, N.F., 2012. Energy-Using Product Group Analysis - Lot 5, Machine tools and related machinery, Task 2 Report – Economic and Market Analysis. Winter, M., Li, W., Kara, S., Herrmann, C., 2013. Determining optimal process parameters to increase the eco-efficiency of grinding processes. Journal of Cleaner Production. 66:644645. Li, W., Winter, M., Herrmann, C., Kara, S., 2012. Eco-efficiency of manufacturing processes: A grinding case. CIRP Annals - Manufacturing Technology. 61/1:59-62. Duflou, J.R., Sutherland, J.W., Dornfeld, D., Herrmann, C., Jeswiet, J., Kara, S., Hauschild, M., Kellens, K., 2012. Towards energy and resource efficient manufacturing - A processes and systems approach. CIRP Annals - Manufacturing Technologies. 61/2:587-609. Dettmer, T., 2006. Nichtwassermischbare Kühlschmierstoffe auf Basis nachwachsender Rohstoffe (Non-water miscible cutting fluids on the basis of renewable resources). Dr.-Ing. Dissertation, Technische Universität Braunschweig, Vulkan Verlag, Essen, Germany (in German). Clarens, A.F., Zimmerman, J.B., Keoleian, G.A., Hayes, K.F., Skerlos, S.J., 2008. Comparison of Life Cycle Emissions and Energy Consumption for Environmentally Adapted Metalworking Fluid Systems. Environmental Science & Technology. 42/22:8534-8540. Webster, J., Tricard, M., 2004. Innovations in Abrasive Products for Precision Grinding. CIRP Annals - Manufacturing Technology. 53/2:597-617. Davis, J.R., 1995. ASM Specialty Handbook Tool Materials, ASM International, Materials Park, Ohio, USA. Klocke, F., König, W., 2005. Fertigungsverfahren II - Schleifen, Honen, Läppen (Manufacturing processes 2 - Grinding, honing, lapping). Springer-Verlag Berlin Heidelberg, Germany (in German). Marinescu, I.D., Hitchiner, M., Uhlmann, E., Rowe, W.B., Inasaki, I., 2007. Handbook of machining with grinding wheels. CRC Press, Taylor & Francis Group, Florida, USA. Rowe, W.B., 2009. Principles of modern grinding technology. William Andrew, Elsevier, UK. International Standards Organization, 2006. Environmental management – life cycle assessment principles and framework. ISO 14040:2006(E). Winter, M., Li, W., Kara, S., Herrmann, C., 2014. Stepwise approach to reduce the costs and environmental impacts of grinding processes. The International Journal of Advanced Manufacturing Technology. 71/5-8:919-931.

EP

[1]

RI PT

References

[10]

[11] [12] [13] [14]

24 of 26

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[15] Li, W., 2012. Energy and eco-efficiency of manufacturing processes. PhD Thesis. The University of New South Wales. Sydney, Australia. [16] Linke, B., Overcash, M., 2012. Life cycle analysis of grinding. Leveraging Technology for a Sustainable World, Proceedings of the 19th CIRP Conference on Life Cycle Engineering, University of California at Berkeley, Berkeley, USA, 293-298. [17] Murray, V.R., Zhao, F., Sutherland, J.W., 2012. Life cycle analysis of grinding: A case study of non-cylindrical computer numerical control grinding via a unit-process life cycle inventory approach. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture. 226/10:1604-1611. [18] Ali, S.H., 2011. Ecological comparison of synthetic versus mined diamonds. Working paper, Institute for Environmental Diplomacy and Security, University of Vermont. [19] Aurich, J.C., Linke, B., Hauschild, M., Carella, M., Kirsch, B., 2013. Sustainability of abrasive processes. CIRP Annals - Manufacturing Technology. 62/2:653-672. [20] Kubota, O., Furukawa, H., Kiskimoto, M., Ukai, N., 2006. Vitrified grinding wheel and method of manufacturing the same. European Patent 1634678 A4. [21] Keat, P.P., 1979. Ceramic bonded grinding tools with graphite in the bond. United States Patent 4,157,897. [22] Wentorf, R.H., 1957. Cubic Form of Boron Nitride. The Journal of Chemical Physics. 26/4:956. [23] Gohara, T., Koshida, T., Hiwasa, S., 2012. Hexagonal boron nitride powder and method for producing same. United States Patent US 2012/0196128A1. [24] Lipp, A., Schwertz, A., Hunold, K., 1988. Hexagonal Boron Nitride. Fabrication, Properties and Applications. Journal of the European Ceramic Society. 5:3-9. [25] Vel, L., Demazeau, G., Etourneau, J., 1991. Cubic boron nitride: synthesis, physiochemical properties and applications. Materials Science and Engineering. 10/2:149-164. [26] Solozhenko, V.L., Turkevich, V.L., 1997. High pressure phase equilibria in the Li3N-BN system: in situ studies. Materials Letters. 32/2–3:179–184. [27] Guo, W., Jia, X., Guo, W.L., Xu, H.W., Shang, J., Ma, H.A., 2010. Effects of additive LiF on the synthesis of cBN in the system of Li3N–hBN at HPHT. Diamond and Related Materials. 19/10:1296-1299. [28] Khvostantsev, L.G., Slesarev, V.N., Brazhkin, V.V., 2004. Toroid type high-pressure device: history and prospects. High Pressure Research: An International Journal. 24/3:371-383. [29] Hall, H.T., 1960. Ultra-High-Pressure, High-Temperature Apparatus: the “Belt”. Review of Scientific Instruments. 31/2:125-131. [30] Hall, H.T., 1958. Some High-Pressure, High-Temperature Apparatus Design Considerations: Equipment for Use at 100.000 Atmospheres and 3.000°C. Review of Scientific Instruments. 29/4:267-275. [31] Liebermann, R.C., 2011. Multi-anvil, high pressure apparatus: a half-century of development and progress. High Pressure Research: An International Journal. 31/4:493532. [32] Mishima, O., Era, K., 2000. Science and Technology of Boron Nitride. In: Electric Refractory Materials, edited by Y. Kumashiro. Marcel Dekker Inc., New York, USA. [33] Wu, M., 1998. Method for making high permeability grinding wheels. United States Patent 5,738,696. 25 of 26

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[34] Stabenow, R., Wenzel, C., 2011. Method of manufacturing a tool made from bound abrasive agents. European Patent 2251143 B1. [35] Christiani, S., Wenzel, C., 2013. Method of manufacturing a tool made from bound abrasive agents. European Patent 2540445 B1. [36] Mulone, R., 2013. Cutting element comprising an integrated lubricant. European Patent 2643123 A1. [37] Ecoinvent Centre, 2010. Database ecoinvent Data v2.2. Centre for Life Cycle Inventories. [38] Winter, M., Öhlschläger, G., Dettmer, T., Ibbotson, S., Kara, S., Herrmann, C., 2012. Using jatropha oil based metalworking fluids in machining processes: A functional and ecological life cycle evaluation. Leveraging Technology for a Sustainable World, Proceedings of the 19th CIRP Conference on Life Cycle Engineering, University of California at Berkeley, Berkeley, USA, 311-316. [39] Babl, A., Geng. H.J., 1969. Preparation of hexagonal boron nitride. United States Patent 3,473,894. [40] Iizuka, E., 1985. Process for producing boron nitride of cubic system. United States Patent 4,551,316. [41] Eagleson, M., 1994. Concise Encyclopedia: Chemistry. Walter de Gruyter, Berlin, Germany. [42] Nakano, S., Fukunage, O., 1993. New scope of high pressure-high temperature synthesis of cubic boron nitride. Diamond and Related Materials. 2, 1409-1413. [43] Li, R., 1995. Vitrified abrasive bodies. United States Patent 5,472,461 A. [44] 3M Innovative Properties Co., 2013. Bonded abrasive article. European Patent 2567784 A1. [45] Kassen, G., 1969. Beschreibung der elementaren Kinematik des Schleifvorgangs (Description of the elementary kinematics of the grinding process). Dr.-Ing. Dissertation, RWTH Aachen, Aachen, Germany (in German). [46] Klimczyk, P., Figiel, P., Petrusha, I., Olszyna, A., 2011. Cubic boron nitride based composites for cutting applications. Journal of Achievements in Materials and Manufacturing Engineering. 44/2, 198-204.

26 of 26

ACCEPTED MANUSCRIPT Life cycle assessment of cubic boron nitride grinding wheels

Marius Winter a, c, Suphunnika Ibbotson a, b, Sami Kara a, b, Christoph Herrmann a, c

Joint German-Australian Research Group on Sustainable Manufacturing and Life Cycle Management

b

Sustainable Manufacturing and Life Cycle Engineering Research Group, School of Mechanical and

RI PT

a

Manufacturing Engineering, The University of New South Wales, Sydney, NSW 2052, Australia c

Sustainable Manufacturing & Life Cycle Engineering Research Group, Institute of Machine Tools and Production Technology (IWF), Technische Universität Braunschweig, Langer Kamp 19b, 38106 Braunschweig, Germany

SC

*Corresponding author

M AN U

Highlights

Comprehensively assessing an environmental impact for a cradle-to-gate of a grinding wheel and cradle-to-grave of an external grinding process using life cycle assessment method to identify their environmental hotspots and the influencing factors.

•

Uncovering the inventory data for the production processes of a cubic boron nitride grinding wheel based on an extensive patent and literature study

•

Investigate the environmental performances and trade-offs when comparing with different abrasive materials and hub materials as well as with different grinding process parameters of a case study including cutting speed and depth of cut. This will shade some lights to support manufacturers and users in improving the environmental sustainability.

AC C

EP

TE D

•