University
stringclasses 19
values | Text
stringlengths 458
20.7k
|
|---|---|
Chalmers University of Technology | B
Genetic algorithm
It is an ample task to analyse complex GAs, and they may evidently be extensively
tailored to suit the optimisation problem at hand, eliminating any hopes to derive
general proofs. To this end, this appendix introduces a set of assumptions that
trades the applicability to the algorithms used in this project for a manageable
analysis. Thus, the following proofs are intended to serve as approximate indicators
for algorithm settings and performance.
B.1 Run-time estimation of simplified GA
Consider a population consisting of a single binary chromosome. The reproduction
is strictly limited to mutation and elitism is employed (i.e. the evolution of the
fitness value is non-decreasing). Let n denote the number of genes, out of which
m are 0s. Furthermore, assume that the fitness function is the one-max function 1.
Theoretically, each gene is independently mutated with some specified probability,
p . For the mutated chromosome to have an associated fitness value that exceeds
m
that of the parent, more 0s than 1s must be mutated. The probability of improve-
ment, P , is approximately the probability of at least one 0 mutating while no 1
i
does, which mathematically can be expressed as
P (p ,m) = (1−p )n−m(1−(1−p )m). (B.1)
i m m m
The estimated time for an improvement is thus
1
E(K (p )) = . (B.2)
m m
P (p ,m)
i m
As will be shown in B.2, the optimal mutation rate is 1/n. However, a bit more
generality may be introduced by defining the mutation rate to be a/n, with a (cid:28) n.
Assuming that there is no bias in the initialisation of the chromosome, the expected
number of 1s in the initial chromosome is n/2. Furthermore, according to (B.1), an
1The one-max function is the number of 1s in the chromosome. The aim of the (binary) GA is
then to maximise the sum of the genes in the chromosome.
III |
Chalmers University of Technology | Master of Science Thesis in the Master’s Programme Chemistry and Biosciences
ANNA ERIKSSON
PER JOHANSSON
Department of Civil and Environmental Engineering
Water Environment Technology
Chalmers University of Technology
Abstract
Contaminated soils are a problem all around the world. Only in Sweden it is estimated that
there is 80 000 contaminated sites. The most common remediation technique is excavating
and landfilling, thus just shifting the problem to a new location. Another problem with this
technique is that possibly valuable contaminants, most commonly metals, are lost. A more
sustainable soil treatment would be chemical soil washing with recovery of the contaminants,
i.e. washing the soil with liquid; in this case acidic process water.
In this study the aim was to leach copper from heavy contaminated soil and bark, from two
sites in Sweden: Björkhult and Köpmannebro. The washing media used was acidic process
water from the flue gas cleaning process in a municipal solid waste incineration plant. The
leaching process was optimized with the parameters L/S-ratio and leaching time, and further
on with evaluation of possible benefits with stepwise leaching. The optimal settings where
then used for batch experiments and includes two leaching steps followed by a washing step
where Milli-Q water is used instead of the process water leaching agent.
The leaching experiments were successful extracting more than 90% of the initial copper
concentration in the one-step leaching. The best parameters were proved to be L/S 10 with a
leaching time of 30 minutes. The two-step leaching, only involving the ash samples, gave a
higher extraction yield allowing for a cheaper disposal method of the ash.
The results show a good leaching of copper, but also that the cleaned soil still has
contaminants above the Swedish guidelines for non-hazardous soils. The key to solve this
probably lies in improving the washing step and by this enable a less expensive alternative for
landfilling the soil residue. The leaching itself will be hard to improve further since it already
gives an almost total leaching of copper and therefore could be used for recovery and this
should be seen as an environmental advantage.
Key words: Contaminated soil, acidic soil leaching, soil wash, copper
V |
Chalmers University of Technology | Jordtvätt: Optimering för sur lakning av koppar i förorenad jord
Examensarbete inom masterprogrammet Chemistry and Biosciences
ANNA ERIKSSON
PER JOHANSSON
Institutionen för Bygg- och miljöteknik
Vatten Miljö Teknik
Chalmers tekniska högskola
Sammanfattning
Förorenad mark är ett problem över hela världen. I Sverige uppskattas att det finns 80 000
förorenade områden. Den vanligaste metoden för omhändertagande är att gräva upp och
deponera den förorenade jorden. Denna lösning förflyttar dock bara problemet till en ny plats:
deponin. Ytterligare ett problem är förlusten av eventuellt värdefulla föroreningar, vanligtvis
metaller. En mer hållbar jordreningsmetod är kemisk jordtvätt där de värdefulla
föroreningarna återvinns. Jordtvätt innebär att man tvättar jorden med en vätska och i denna
studie har surt processvatten använts.
Målet för denna studie var att laka ur koppar från starkt förorenad jord och bark från två olika
områden i Sverige; Köpmannebro och Björkhult. I denna studie användes surt processvatten,
från rökgasreningen vid den kommunala avfallsförbränningen vid Renova, som
lakningsvätska. Lakningsprocessen optimerades med avseende på två parametrar: L/S-kvot
och lakningstid. Optimeringen fortsatte genom att utvärdera eventuella fördelar med stegvis
lakning. De optimala parametrarna användes sedan för batchexperiment vilka inkluderade två
lakningsteg följt av ett tvättsteg där Milli-Q vatten användes istället för processvatten.
Lakningsexperimenten var framgångsrika i vilka mer än 90% av den initiala koncentrationen
extraherades när ett lakningssteg användes. De bästa parameterinställningarna från dessa
försök var L/S 10 med en lakningstid på 30 minuter. Tvåstegslakning utvärderades bara för
askproverna, för vilka de gav ännu högre lakningsutbyte jämfört med ett lakningssteg. Detta
medför eventuellt en billigare deponikostnad för askan.
Resultaten visar att processvattnet har mycket goda lakningsegenskaper för de aktuella jord-
och askproverna, men också att den tvättade jorden fortfarande har metallhalter som
överstiger de svenska riktlinjerna för brukbar jord. Lösningen på detta problem ligger med
stor sannolikhet i att förbättra tvättsteget för att billigare deponeringsalternativ ska bli
aktuella. Lakningsteget är dock i sin nuvarande form svår att förbättra med nära total
urlakning av koppar, vilket i sig bör ses som en miljömässig fördel då kopparen kan
tillvaratas.
Nyckelord: Förorenad jord, sur jordlakning, jordtvätt, koppar
VI |
Chalmers University of Technology | 1. Introduction
In 2008 the Swedish Environmental Protection Agency (SEPA) estimated that there are
80 000 potentially polluted sites in Sweden (SEPA, 2009a). This is one of the main obstacles
to achieve the environmental goal ”A non-toxic environment”, one of the 16 environmental
objectives set by the Swedish government to be accomplished by 2020 (SEPA, 2013). The
process is now in an inventory phase where all of the potentially polluted sites are divided
into classes according to the origin of the pollution, normally depending on what type of
industry that exist/existed, the degree of pollution and the toxic effect. This is a very time
consuming work but the ambition is that the inventory phase should be finished by 2013. The
process is obstructed by the fact that new polluted sites are identified and formed
continuously (SEPA, 2009a, Ohlsson et al., 2011).
Parallel to the inventory phase the intervention process is running, which is when the actual
remediation of the contaminated sites occurs. This is a very time consuming and costly
process. In 2008 250 million SEK was distributed to the different counties administration
boards for their remediation of contaminated sites (SEPA, 2009a). This corresponds to 970
ongoing investigations and 170 interventions during the same year (SEPA, 2009b).
The pollution situation at the different sites differs widely. The SEPA has calculated the
distribution between different pollutants based on the top 216 prioritized sites in 2008, as seen
in Figure 1.1.
13%
Arsenic
25%
Metals
8%
Dioxins
2%
Halogenated
hydrocarbons
Oils
11%
PAH
11% 30% Others
Figure 1.1. The estimated distribution of pollutants in contaminated sites in Sweden (SEPA, 2009a).
Metal and arsenic contamination contributes to about 55% of the total pollution. Metals are a
natural part of the ecosystem, but here their levels are elevated. Elevated metal amounts can
be directly toxic to organic life as well as indirect, pointing to that metals are non-
biodegradable and thus accumulates in biological tissue.
1.1. Aim and objectives
The main aim with this thesis work is to optimize the soil washing process for contaminated
soil and ash from bark as well as evaluate the possibility of recovering copper. The specific
goals are to:
1 |
Chalmers University of Technology | Investigate the parameters L/S ratio, leaching time and the possible advantages with
two-step leaching by several leaching experiments.
Measure the success of the leaching by the amount of leached copper, the amount left
in the solid residue but also how stable the soil residue is to further leaching as this is
equally important.
The samples that is used for experiments consist of clay soil and bark from the polluted site
Köpmannebro south of the Swedish city Mellerud and Björkhult close to the Swedish city
Kisa. Both of these sites are heavily contaminated with metals, foremost copper, from the
former wood processing industry. The bark samples are be incinerated before leaching due to
its high organic content, which makes it illegal at landfill, but also since previous studies
indicate that the copper becomes more accessible with incinerated samples (Tateda, 2011,
Karlfeldt Fedje et al., 2013).
As leachate, acidic process water from the flue gas cleaning process of the municipal waste
incineration at Renova in Gothenburg, Sweden is be used. After optimization the cleaned soil
is evaluated in terms of quality compared to the Swedish guidelines for landfill and
contaminated soils. Depending on the degree of contamination, soils are divided into two
different categories KM, “känslig markanvändning”, and MKM, “mindre känslig
markanvändning”. The KM is less contaminated soil, which do not apply any boundaries for
what kind of activities or buildings that can reside in the area. The MKM corresponds to more
contaminated soil, which restricts the area to be used for industry, offices and other activities
where people for example only spend their working hours. The KM and MKM limitations for
Cu are 80 and 200 mg Cu per kg soil, as comparison the soil in Köpmannebro has measured
values as high as 51600 mg Cu/kg soil (Kemakta, 2012).
The aspiration of the project is to find a remediation method for contaminated soil, where
large amount of the copper can be recovered and reused. Due to the very acidic process
water’s effect on the soil samples, the intention for these samples are not to be used as soil
again but rather as construction material and thereby avoiding the landfill alternative.
1.2. Limitations
This project will focus on the leaching of copper although other metal contaminants will be
measured to some extent. The focus will also be on the specific site at Köpmannebro even but
the optimal settings from this site will be evaluated for the Björkhult site as well. The
optimization will be set on using the acidic process water from the flue gas cleaning of
municipal waste from Renova and Milli-Q water as leachates.
1.3. Main research questions
What settings of L/S ratio and leaching time give the best leaching of Cu from the
contaminated soil of Köpmannebro?
Is it an advantage to perform the leaching in one more step?
Are the pollution levels of the cleaned soil below the Swedish limits for toxic waste?
If not, are the soil and ash matrixes stable enough to prevent heavy leaching of
contamination to the surroundings?
2 |
Chalmers University of Technology | 2. Theory and background
2.1. Remediation methods
Different pollutions need different methods of remediation. This report focuses on the
remediation of metal-contaminated soils. This process is complicated, mainly because of three
facts:
A. The contamination is seldom homogenous; the metals are unevenly distributed in the
soil.
B. Metals are non-degradable and cannot be destroyed.
C. The large variation of the forms the metals exist in as ions, salts etc., as well as the
variation in soil matrixes. This yields multiple interactions as bonding, partitioning,
chemical reactivity, mobility etc., between the soil and the metal contamination that
derives from the soil characteristics as particle size, cat-ion exchange capacity, pH,
mineralogy, organic content, and the form of the metal (Dermont et al., 2008).
The by far most common soil remediation technique in Sweden, as well as internationally, is
soil excavating and landfilling. This is due to tradition, availability and economic reasons
(Ohlsson et al., 2011, Dermont et al., 2008). The problem with this method is that it does not
primarily solve the underlying issue, rather relocate the problem because it does not remove
the contamination from the soil; just shift the contaminated soil to a different location even
though the potential leaching is controlled within the landfill.
Dermont et al., 2008, gives a review of the existing techniques for remediation of metal
contaminated sites, and divides them into two main groups: stabilization/isolation of metals
and extracting metals. Each of these main groups can be further divided into off site and on
site techniques, thus excavation and landfilling, where the contaminated soil is dug up and
relocated to a landfill and the contaminated site is refilled with clean soil, are examples of ex
situ stabilization/isolation techniques.
Other stabilization/isolation techniques except from excavation are;
Stabilization/solidification: Stabilization and solidification neither remove the
contaminants, rather covers them. Solidification is to physically encapsulate the
contaminated soil e.g. bitumen, fly ash or cement are injected to the soil (Mulligan et
al., 2001). This can be done either on site or after the soil have been moved; the latter
more common. In stabilization different chemicals are used to stabilize the
contaminants, thus reduce their mobility. Often the chemical is a liquid monomer that
polymerize (Mulligan et al., 2001). The main advantages of these methods are their
relatively low costs though problems can occur if the soil for example has a lot of clay
or oily patches, which obstruct the mixing procedure.
Vitrification: Vitrification is similar to stabilization/solidification in the way that the
contaminants are not removed. Instead of using encapsulation or stabilizing media it
uses thermal energy. Electrodes are inserted in the soil and a glass or graphite frit is
placed on the ground. This frit initiates the vitrification process where the minerals in
the soil are melted due to the high induced current. The soil is then allowed to cool off
at which point an encapsulating glassy material is formed by the inorganic
compounds. Successful vitrification solutions exist for arsenic, chromium and lead
contamination, but problems concerning clay rich soils that lower the efficiency still
exist. Other problems are the hazards with toxic gases that could be released during
the process, the uncertainty in the vitrified end products leaching qualities that still has
3 |
Chalmers University of Technology | to be monitored and the high cost since the method is highly energy demanding.
However this could be a suitable method for large masses of contaminated soils in
shallow depths (Mulligan et al., 2001).
Chemical red/ox: This is a chemical treatment used to detoxify the contaminated soil.
It is especially applicable for reducing highly toxic Cr(VI) to less toxic Cr(III) or
oxidizing As(III) to less toxic As(V) or to adjust pH in acidic or basic soils (Mulligan
et al., 2001). This method is commonly used prior stabilization/solidification to lower
the toxicity. The major disadvantage of chemical treatment is that it is in need of
chemicals that could be both hazardous and expensive (Dermont et al., 2008).
Phytostabilization: Phytostabilization is a technique based on certain plants ability to
accumulate heavy metals. Implantation of such plants can thereby remediate
contaminated sites although the method is limited to root deep contamination and the
remediation has to be monitored during a long period of time. When the soil is
remediated the plants has to be taken care of as toxic waste. Advantages is that except
the remediation of the contamination on site the plants also prevent erosion, hence
preventing that the contamination is spread to ground water (Dermont et al., 2008,
Mulligan et al., 2001).
Monitored natural attenuation: This is the non-treatment option which might be
relevant where any action might lead to enhanced spread of contamination or the costs
exceeds the benefits. However, this demands continuous observations assuring no high
toxic compounds leaks to the surrounding environment.
The main advantage with the stabilization/isolation techniques is that they work for a wide
variety of soils and metals compared to extracting techniques. The drawbacks are many; most
important is that it is not a sustainable solution because the contaminants are not removed
from the soil. There is also a lack of research of the long-term stability of the stabilized
material, which means that the contaminated site or the landfill has to be monitored for a long
time period. Other problems are that the excavated area needs to be refilled with clean soil as
well as that the cement based solidification significantly increase the volume if it is sent to
landfill.
Therefore the extracting techniques have a promising future. Not only because the cleaned
soil sometimes can be used as soil once again but also because there is a possibility of
recovering the metals. Examples of extracting techniques are;
Physical separation: Physical separation is a good method when the contaminant is
dominant in one of the particle fractions. Equipment to perform the physical
separation varies from hydro cyclones, fluidized beds or flotation, all these well-
known methods from the ore industry. Another method is magnetic separation that
uses the magnetic qualities of many metals (Mulligan et al., 2001).
Chemical soil washing: When using soil wash the contaminated soil is excavated and
washed with various agents in either reactors or as heap leaching. Ideally the cleaned
soil is clean enough to be returned afterwards. Several different leaching agents have
been used, such as inorganic acids, organic acids, chelating agents or combinations of
earlier mentioned. Earlier test soils have showed that the method is most efficient with
sandy soils i.e. less than 10-20% clay and organic content (Mulligan et al., 2001).
Soil flushing: Soil flushing is quite self-explanatory, a solution is flushed through the
soil via infiltrations systems, surface trenches or horizontal/vertical drains and leachate
collected at the bottom (Dermont et al., 2008). The technique is based on the
possibility to solubilize the contaminants and is preferably applied on soil with high
water permeability (Mulligan et al., 2001). The solution could vary depending on the
4 |
Chalmers University of Technology | type of contaminant, but most commonly used is water with or without additives.
Water being more environmentally friendly alternative since additives such as
chelating agents and surfactants could have a negative effect on the environment.
(Dermont et al., 2008). Soil flushing is quite similar to soil wash and is preferable if all
of the contaminated water can be collected at the site. Is this not the case soil washing
is the better choice
Biological extraction: Is similar to soil washing but with biological agents as bacteria
or algae used instead of earlier mentioned chemical agents (Dermont et al., 2008).
Biological extraction has not yet been used in any big scale remediation but successful
lab trials have been performed.
Electro kinetics: Electro kinetics involves passing a low electric current through the
soil; the current makes the positive ions move to the cathode and negative ions move
to the anode (Mulligan et al., 2001). This method is most efficient with saturated soils
since water enhance the conductivity of the soil.
There are some problems with the existing extraction techniques that stem from the earlier
mentioned problem with a large variety of soils as well as with the economical sustainability
(Dermont et al., 2008).
2.2. Sites used in this study
2.2.1. Köpmannebro
In Långö, south of the city of Mellerud in Dalsland Sweden, there was a wood processing
industry for telephone poles in the beginning of the 20th century. The processing was made
according to the Boucherie method, which involves injecting blue vitriol into the timber and
let it soak until saturation (de Vougy, 1856). Then the timber were decorticated and limbed
and the bark was left at the site, leading to accumulation of contaminated bark at the site. Blue
vitriol is a rest product from mining with sulfuric ores, and consists of one Copper(II)sulfate
molecule that is crystalline bonded to five water molecules [CuSO 5H O].
4 2
This industry resulted in the highly contaminated site of 8000 m2, were still no vegetation
exist (Kemakta, 2012). The core study performed by Kemakta, commissioned by Dalsland’s
office of environment, concludes that the copper content is elevated in all of the soil layers,
with 70% of the samples showing levels corresponding to toxic waste. The study suggests
several different treatment alternatives as landfilling or solidification. None of the suggested
treatments will recover the Cu from the site, which is the main goal with this project.
Therefore the site is fitting for this study, to determine if there is a method to actually recover
the large amounts of copper.
5 |
Chalmers University of Technology | Figure 2.1. The contaminated area at Köpmannebro (Kemakta 2009).
As seen in the Figure 2.1, the bark is not degraded to a high degree. The bark layer reached
from the surface to as deep as 1-1.5 m under which the clay layer could be found.
2.2.2. Björkhult
The other site investigated is Björkhult, situated on the south shore of the lake Verveln close
to the city Kisa in Östergötland, Sweden. From 1916 to 1944 there was a wood processing
industry for telephone poles, similar to the one earlier described at Köpmannebro. The site is
approximately 7000 m2, but differs from Köpmannebro in the case of vegetation. At Björkhult
the natural fauna seems to have recovered well, as seen in Figure 2.2, and there are a lot of
trees, grass and bushes which could be an effect of the different soil characteristics observed
at the two sites. This might be due to that the site has been covered with soil from an external
site since there is a well-defined soil layer above the bark.
6 |
Chalmers University of Technology | Figure 2.3. The dominating soil classes in Sweden according to FAO (Markinfo, 2006)
2.3. Criteria for contaminated material
2.3.1. The KM/MKM-criteria
The Swedish government has set up 16 environmental objectives to ensure a sustainable
environment in Sweden. Among these objectives is “A non-toxic environment”. The agency
responsible for these is the Swedish Environmental Protection Agency (SEPA). The problem
is seen in a long time perspective, 100 to 1000 of years ahead. When come to risk analysis
and planned use for a site it is hard to see more than 100 years ahead, but the SEPA tries to
make the demands higher to ensure risks in the future. (SEPA, 2009c)
The land use is divided into two main groups, sensitive land use and less sensitive land use.
The sensitive land use is for an area where the quality of the soil does not limit the
possibilities of land use. All groups of humans are out of harm no matter how much time they
spend there and most of the ecosystems, water and ground water systems included, are
protected (SEPA, 2009c). The less sensitive land use is for areas where the quality of the soil
does limit the possibilities. The risk analyses of these soils recommend that grown-ups should
not spend more than normal working hours there whereas children and elderly people should
not spend time there regularly. This less sensitive land use is for example offices, industries or
8 |
Chalmers University of Technology | roads. The contamination limits are set so that water and ground water systems are protected
in a distance of 200 m. The actual limits can be seen in Table 2.1.
The mobility of contaminants is strongly dependent on the surrounding soil, pH and the
chemical form of the contaminant. The general guidelines for sensitive and less sensitive land
use are set to not underestimate leaching of contaminants. In some cases a site-specific risk
analysis can be appropriate. These site specific limits should be set from leaching tests as well
as from comparing the existing content in soil and ground water. The site-specific limits are
not in any case intended to increase the allowed limits, rather the opposite, to decrease limits
if increased risks are suspected.
Table 2.1. Limits for sensitive land use; MK and less sensitive land use; MKM
KM [mg/kg MKM [mg/kg
Substance TS] TS]
As 20 40
Pb 200 400
B 7 20
Ba 160 260
Cd 4 20
Co 10 15
Cr 90 150
Cu 75 160
Sb 30 50
Se 1 5
Zn 300 450
Be 20 40
Hg 5/10 10/20
Mo 10 25
Ni 75 150
V 100 200
2.4. Incineration
Several studies have been done regarding chemical soil washing but none about leaching
metals from bark. A problem with performing “soil wash” on bark is the requirements of high
L/S-ratios owing to the barks high absorption ability and the high amount of organic matter
that can form strong bonds with metals (Thomas et al., 2013). A way to overcome this is to
incinerate the bark to ash, which not only accumulate the metal contamination to a smaller
mass but also burns the organic compounds and thereby releasing strongly adsorbed metals
from the complexes.
Another advantage with incineration is that contaminated bark, due to its high organic
content, is illegal for landfilling (SFS 2001:512). This is due to volatilization of the organic
compounds at 473-773 K depending on the compound properties. Most industrial combustion
of biomass is usually done at 1073 K or higher to assure a complete burnout of CO (van Loo,
2008).
9 |
Chalmers University of Technology | 2.4.1. Ash
The in biomass, such as bark, the ash-forming part is salts bound to the carbon backbone of
the organic compounds. However, since the bark to some extent is mixed with the underlying
soil, the ash-forming fraction will also come from mineral particles from the surrounding soil.
The ash can be divided into two fractions; the heavier part is called bottom ash which is the
part left on the grates consisting of sintered ash particles and impurities such as stone or sand
and the fly-ash which is coarser particles precipitated during the second combustion or in the
multi-cyclones and particles that precipitates later in the flue gas cleaning, often in the
electrostatic filter (van Loo, 2008). The amount of metals and salts in the bottom ash varies
from metal to metal. Volatile metals like Hg, Cd, Pb and Zn are for example often
accumulated in the fly ash (Hong, 2000, Nurmesniemi, 2007, van Loo, 2008).
The ratio between the fractions depends on several factors; type of incineration, excess air
ratio, fuel type, continuous or batch combustion to mention a few. A general rule is that the
fly ash-ratio increase with fluidized-bed combustion compared to fixed bed combustion. In
this study a larger fraction of bottom ash would be preferred since it is this fraction that will
be studied in the leaching optimization and thereby all the copper that goes with the fly ash is
lost. The ratio of copper content between fly ash and bottom ash differs amongst studies from
10-90% of the copper in the bottom ash (Sander 1997, van Loo, 2008).
2.4.2. Industrial combustion
The bark in this study was incinerated in batches with smaller furnaces due to the small
amount of sample and to generate a pure bark ash. However, a large-scale solution would
probably involve an industrial scale continuous furnace because of the large amount of bark.
Only at the site in Köpmannebro it is estimated to be more than 6500 ton contaminated bark
(Kemakta, 2012). The most common combustion techniques are grate combustion or
fluidized-bed combustion. The facility at Sävenäs has four furnaces of the fixed bed
combustion-type, which is the method that will be simulated in this study.
In grate combustion furnace, such as those at Sävenäs, the fuel is carried into the furnace on
moving grates supplying a homogeneous and even amount of fuel to assure a complete and
smooth combustion. A primary air supply is introduced from below with a low flow avoiding
turbulence that would lead to a release of fly ash and unburned particles. The flue gases from
the primary combustion rises to a secondary combustion chamber where it is mixed with fresh
air and often recirculated flue gas, so called secondary air, for a complete combustion of
hazardous gases such as NO (van Loo, 2008).
x
The next step is the cleaning procedure. This is not of importance concerning the combustion
of the bark but since it does concern the process water, thus is still of interest in this study. In
the cleaning procedure the fly ash in the flue gas, from the second combustion, are removed
with an e.g. electrostatic filter. This filter is an electric field, where the fly ash can be removed
due to the ions it contains. The flue gas then passes through wet scrubbers, which consists of
several water curtains that dissolves dust, acidic gases (mostly hydrochloric and hydrofluoric
acid), mercury and other heavy metals from the gas. This solution is the process water that
will be used in the leaching experiments and its characteristics vary with what is being
incinerated (Renova, 2010, van Loo, 2008).
10 |
Chalmers University of Technology | 2.5. Copper
Copper exist, naturally in the environment, the average content is about 50g/ton in the earth’s
crust. Copper commonly occurs as sulfide ores, e.g. CuFeS (90%), but also as oxide ores, e.g.
2
Cu O, (9%) and as pure copper (1%). In the primary copper producing industry it is mainly
2
the sulfide ores that is used, although a large part of the produced copper comes from recycled
materials (Elding et al., 2012).
Half of the amount of produced copper is used in the electric component industry where its
excellent conductivity is highly valued. Other industries that use copper are engineering
industry (21%), building industry (11%), household articles (10%) and transport industry
(8%) (Elding et al., 2012). New materials have started to compete with copper in many of the
common usages, this have accelerated the development of new copper materials with
improved qualities (Elding et al., 2012).
Copper is essential for probably all living organisms, but it can also be toxic with elevated
copper concentrations for many organisms. Vascular plants can be afflicted with shortage of
chlorophyll and many funguses’ microbial digestion cease when copper concentrations are
elevated. Animals are sensitive for copper concentrations both above and below normal. A
lack of copper can cause diarrhea and anemia while an excess of copper causes cramps and
hepatitis B (Elding et al., 2012).
11 |
Chalmers University of Technology | 3. Method
The work process of the project was divided into two parts; first a literature study of the latest
progress in the field of soil washing, as well as on other treatment techniques, and second a
laboratory part, where leaching experiments were performed. The initial part of the project
emphasized on the literature study; what methods had been used earlier and what were their
advantages and disadvantages.
The laboratory part began when a suitable experimental setup could be established based on
earlier research studied in the literature part. The analyses were done with eg
spectrophotometer, ICP-MS (inductively coupled plasma mass spectrometry) and ICP-AES
(inductively coupled plasma atomic emission spectroscopy).
3.1. Leaching experiments
The experimental part consisted of the experiments performed to optimize the leaching
process. The soil and bark samples collected from Köpmannebro and Björkhult were dried
and, in the case of the bark, incinerated to ash before the leaching trials. To optimize the
leaching procedure different L/S-ratios, time of leaching and step-wise leaching were
evaluated. A schematic overview of the experimental procedure can be seen in Figure 3.1.
3.1.1. Sampling
The soil and bark samples were collected from Köpmannebro and Björkhult, which both have
been heavily contaminated with Cu due to earlier wood processing in the area (Kemakta,
2012, SEPA, 2009a). Samples were collected at the same spots previously was identified as
Cu hot-spots (e.g. Kemakta, 2012, Arnér, 2011) and at specific depths with shovels in
stainless steel. The samples were stored in PP-bottles at 4oC before preparation.
In Köpmannebro the bark layer reached from the surface to as deep as 1-1.5 m under which
the clay layer could be found. At Björkhult the soil profile was different and consisted of
additional layers: 0-10 cm sandy soil, 10-30 cm bark layer and beneath 30 cm depth there
were a red soil more fine grained than the sandy soil. Other differences between the sites were
the total absence of vegetation in Köpmannebro, while it grew both grass and trees in the
contaminated areas in Björkhult.
3.1.2. Sample preparation
The sample preparation involved a drying step where the bark and soil samples were dried in
an oven (Memmert U15) at 80oC until their weights were stabilized, approximately 1.5 day
for the soil samples and 2-3 days for the bark samples. During the first 2 hours of the drying
step the soil was mixed a couple of times to prevent it from becoming a stiff solid cake that
would need grounding prior to the leaching experiments. After drying the samples were kept
dry in desiccators until leaching tests or, in the case of the bark samples, until the incineration
step.
12 |
Chalmers University of Technology | Sampling
Clay soil Bark
Drying at 80oC until
weight is stabilized
Drying at 80oC until Incinerate to ashes at
weight is stabilized
Measure pH 850oC
Triplicates of Triplicates of
4.0 or 5.0 g 0.5 g sample
sample
Add 10 ml Add 25 ml Add 50 ml Add 2.5 ml Add 5.0 ml
process water process water process water process water process water
(L/S 2) (L/S 5) (L/S 10) (L/S 5) (L/S 10)
Shake for 30 min Shake for 60 min Shake for 90 min
at 140 rpm at 140 rpm at 140 rpm
Centrifuge for 15
min at 3000 G
Precipitate Supernatant
Add 10 ml(soil)/1 ml(ash) Shake for 5 min
Milli-Q at 140 rpm
Filtration and
spectroscopy
Precipitate Centrifuge for 15 Supernatant
min at 3000 G
Drying at 80o C until Leaching test
weight is stabilized SS-EN-12457-3
Figure 3.1. Flowchart for the laboratory work for the one-step-leaching
The incineration step was performed due to the low availability of copper, high absorption of
leachate and high organic content in the bark samples. The organic content is of importance
due to regulations regarding landfill of organic matter. According to the Environmental Code
it is illegal to deposit organic material (SFS 2001:512). In addition earlier studies have shown
13 |
Chalmers University of Technology | that the copper is easier to leach from ash than from bark. This might be due to strong
interactions between organic compounds and copper (Karlfeldt Fedje et al., 2013, Tateda,
2011).
Two different furnaces were used for the incineration process: a Carbolite Furnaces CSF 1200
and a destruction furnace typ-D 200. The Carbolite Furnace CSF 1200 is an ordinary furnace
where natural convection heats the sample. This oven was available in the lab and used for
incineration both with reducing and oxidizing conditions. Both of the incineration processes
began with grounding the bark so the largest particles were >0.5cm. To reach reducing
conditions the grounded bark was placed in crucibles with caps, to minimize the access to air,
while the incineration with oxidizing conditions the bark was spread in a thin layer (max 4mm
thick) on a plate. In both cases the samples were then incinerated at 850°C for 6 hours and
afterwards stored in desiccators until further tests. The temperature for incineration was set to
850°C due to earlier studies and that large scale biomass furnaces often operate at this
temperature (van Loo, 2008).
Early analysis showed that the copper were much more accessible (see section 4.2.) when the
bark was incinerated in oxidizing conditions but then the bark to ash ratio was very low.
Therefore, as well as to mimic the real process conditions, bark from Köpmannebro was
incinerated at Renova in their destruction furnace typ-D 200. The temperature was the same
as with the Carbolite Furnace CSF 1200. The difference between the ovens is generally that
the destruction furnace applies heat by blowing hot air on the bark which resembles a large
scale furnace where a steady air flow is injected from below to ensure oxygen supply but also
increases the amount of fly ash. This showed to decrease the ash to bark ratio even more, but
since it probably mimic a large scale process better than the lab oven, ash mixed in a 50/50
ratio from both ovens were used for further analysis.
3.1.3. Leaching procedure
The Cu leaching is the principal part of this project and the process was to be optimized. To
extract copper from the soil and ash samples, process water from Renova’s waste-to-energy
incineration plant in Sävenäs was used as leachate. More specifically the process water is a
byproduct from the washing step of the flue gas and has acidic properties (pH≈0.5) that makes
it a promising leachate both from a chemical and economical perspective. The process water
was analyzed with ICP-AES according to section 3.2.3.1.
The dried soil, 4 or 5 g, and bark, 0.5 g, samples were weighed in 50 ml respectively 15 ml
PP-bottles. The process water was then added to the test tubes according to the specific L/S-
ratio, 2, 5 and 10 ml per gram. The test tubes were then kept in a reciprocating shaker (Julabo
SW-20C) for the allotted time of the leaching procedure: 30, 60 and 120 min. The soil from
Köpmannebro was also tested with longer leaching times, 18 and 24 h, due to earlier studies
suggested that soil had a slower release of metals than ash (Yip, 2008, van Benschoten, 1997).
Each soil sample was done in triplicates, while the ash samples were, to some extent, done in
duplicates due to shortage of sample.
The leachates were separated from the soil or ash through centrifugation, which thereby
terminated the leaching process. The centrifugation was done in a Sigma 4-16 at 3000 G for
15 minutes. The supernatants were decanted and filtered using paper filters, pore size 6 μm
and a funnel (soil samples) or, due to the small amount of sample, filtered with a syringe and
glass microfiber filter, pore size 1.6 μm (ash samples). The pH-values of the filtered
14 |
Chalmers University of Technology | supernatants were measured with Universal indicator from Merck to determine if acidification
was necessary. As none of the samples had a higher pH than 2, no acidification was made.
The supernatants were stored at 4°C pending further analysis.
The solid residues were washed with Milli-Q after the centrifugation. At first an L/S-ratio of
5ml per gram solid sample was used but it was later, after the one-step optimization, changed
to 2ml per gram in order to minimize the volume of contaminated water. The samples were
washed for 5 minutes in the reciprocal shaker (Julabo SW-20C), and then centrifuged at 3000
G for 15 minutes to terminate the washing step. The washing supernatant was filtered and its
pH was measured with the same procedure as with the leaching supernatant. The remaining
solids from the washing step were dried at 80oC, until its weight had stabilized, and was then
stored at 4oC awaiting further analysis. The final weight of the dried solids was noted to be
able to approximate the matrix degradation of the ash and soil.
3.1.4. Step-wise leaching
After evaluating the one-step leaching the optimization continued with a two-step leaching to
investigate if this further improved the leaching. The leaching experiments showed that a high
L/S-ratio was most effective; thus L/S-ratio 10 was used for all the two-step leaching
experiments. The leaching time had no significant importance according to earlier
experiments (see section 4.2.1.); therefore short leaching times were chosen: 15+15min,
15+30min and 30+30min. The leaching method was the same as in the one-step leaching,
except that after decanting of the leaching supernatant from the first step, the leaching was
repeated once more before the washing step.
3.1.5. Batch leaching
The optimal leaching parameters were used for a larger sample amount, 20-30g depending on
available sample. These batch experiments where performed in the same way as the previous
experiments.
3.1.6. Leaching test for depositing
To determine if leached ash and soil samples could be used as a resource instead of being
landfilled, a downscaled SS-EN-12457-3 leaching test was performed. The dried pre- and
post-leaching soil and ash samples were leached with Milli-Q, first for 6 h with L/S-ratio 2
followed by 18 h with L/S-ratio 8. During the leaching the samples were continuously shaken
with a reciprocal shaker (Edmund Bühler 7400 Tübingen SM25) and then centrifuged to
separate the leachate from the soil/ash. The volume of the decanted leachates were measured
and filtered; then stored at 4°C awaiting analysis with ICP-AES.
3.2. Methods for metal analysis
The analysis of the leachates was the principal indicator if the leaching of the samples had
succeeded or not. Selected leachates were sent to a commercial and certified laboratory for
external analysis of metal concentrations, as was also done with the original soil and ash
samples. To select the significant samples, not having to send all of them for external analysis
due to high costs and delay of results, a spectrophotometric measurement of the Cu2+
concentration in the filtered supernatants from the leaching and washing steps were made (see
section 3.2.1).
15 |
Chalmers University of Technology | 3.2.1. Analysis of Cu content in leachates and washing water samples
To get a fast estimation of the degree of success of the copper leaching tests, a semi
quantitative spectrophotometric analysis of the leachates and the washing waters were done.
This analysis measures the absorption at 610 nm, where the [Cu(NH ) ]2+-complex has an
3 4
absorption maximum. The conversion of all present Cu2+-ions to [Cu(NH ) ]2+-complexes
3 4
was made by adding NH in excess according to the method in Norin 2000.
3
To quantify the amount of [Cu(NH ) ]2+ in the samples, a standard curve was made. For this
3 4
9.99 g CuSO *5H O was dissolved in Milli-Q and diluted to 100.0 ml. From this solution 5.0
4 2
ml was further diluted with Milli-Q to 100.0 ml. From this solution five standard samples
were made with 5.00, 10.0, 15.0, 20.0 and 25.0 ml of the copper solution. Next 5.0 ml of 5 M
NH was added to each sample as well as to a reference sample without any copper solution.
3
These standard and reference samples were diluted to 50ml with Milli-Q, corresponding to 0,
2, 4, 6, 8 and 10 mM. After analysis of these samples, a standard curve for the absorption of
[Cu (NH ) ]2+ concentrations ranging 0-10 mM could be made according to Lambert-Beer
3 4
law.
To prepare the samples from the leaching experiments for the spectrophotometric analysis
V ml (specific values can be found in Table 3.1) from the leachate samples was mixed
sample
with V ml of 5 M NH and diluted to V ml with milli-Q. The differences in volumes
NH3 3 tot
when diluting different samples were due to the great variance in copper content between for
example the leachate from the ash and the one from the soil. The turbidity of these diluted
samples were then measured and, if necessary, diluted even further if the absorbance was
higher than 600 or precipitation was found. The concentration of NH was kept at 0.5 M
3
independent of V and V for all samples to be comparable with the reference sample.
sample tot
Table 3.1. Volumes for the dilution of leachate and wash water as preparation for spectrophotometric analysis.
Soil Ash
Leachate Leachate Washing
ml
& washing water
water
V 4.0 1.0 0.4
sample
V 2.5 2.5 1.0
NH3
V 25 25 10
tot
3.2.2. Analysis of metals in solid soil, bark and ash samples
The soil, bark and ash samples were sent for external analysis to determine the total elemental
content (see appendix II for results). The external lab prepared the dried samples for analysis
according to standardized methods where the elements were dissolved using different
methods depending on the material of the sample. The ash samples were dissolved according
to the standardized methods ASTM D3683 and ASTM D3682 before analysis. For the bark
and soil the same two methods were used. The samples was dissolved in Teflon containers
using concentrated HNO and H O for analysis of As, Cd, Cu, Co, Hg, Ni, Pb, B, S, Se and
3 2 2
Zn, or melted with LiBO and then dissolved in HNO for Ba, Be, Cr, Mo, Nb, Sc, Sr, V, W,
2 3
Y and Zr. The exception was for analyzing tin (Sn) in soil samples where Aqua regia in
16 |
Chalmers University of Technology | reversed proportion was used for dissolution. The metal concentrations in the corresponding
liquids were analyzed using ICP-MS.
3.2.2.1. ICP-MS
A very common analytical method is ICP-MS which was used to analyze the solids for
elemental composition. The principle of an ICP-MS (Figure 3.2) is that the sample is
converted into an aerosol by either a nebulizer or a laser, depending on whether the sample is
a solution or a solid (Thomas, 2004).
The aerosol is then injected into the ICP-torch that consists of argon plasma controlled by an
electromagnetic field created by a RF-generator. In the ICP-torch the aerosol is evaporated
giving very small solid droplets of sample. These are in turn vaporized into a gas, and finally
through collision with argon electrons, atomized and ionized (PerkinElmer, 2004).
After the sample is converted into single atom ions they are lead through two metal plates,
called the sampler and the skimmer cone, in what is called the interface region. These cones
have centered holes and thereby block the ionized beam that is not centered. The cones also
facilitate the pressure drop, from 101.3 kPa at the plasma torch to 200 Pa in the interface
region, and finally as low as 10-4 Pa in the analyzer region.
Figure 3.2. Schematic view of ICP-MS (Thomas, 2004)
In the analyzer region ions are first focused by ion optics, i.e. electromagnetic fields, before
reaching an analyzer such as quadrupole or Time-of-Flight depending on what instrument is
being used. In the analyzer the atom ions are detected depending on their M/Z-ratio and give
both qualitative and quantitative measurement of the atoms present in the sample (Thomas,
2004).
3.2.3. Analysis of metals in original process water and selected
leachate and wash water samples
The metals in the process water, leachates and wash water were quantified using ICP-AES
(see appendix I for results). The samples were prepared by digesting in 7 M HNO . Even
3
though the sample already is a solution, this is to break any complexes present. This
preparation procedure is according to the standard SS 028150-2 while the analysis is done
according to SS-EN ISO 17294-2:2005.
17 |
Chalmers University of Technology | 3.2.3.1. ICP-AES
Another analytical method similar to ICP-MS is ICP-AES, which was used to analyze the
liquid samples. The method relies on the fact that atoms emits energy at specific wavelength
when returning to ground state. The sample has to be a solution to be analyzed with ICP-AES
and due to the low detection limit often diluted as well. The first steps of the ICP-AES are
very similar to ICP-MS (see section 3.2.2.1.) where the sample is sprayed into an argon gas
flow to create an aerosol. This aerosol is then injected into the plasma torch were the sample
is vaporized, atomized and ionized using a radio frequency generator. In this part it is of
importance that the whole sample is converted to plasma since atoms in ground state would
absorb wavelength from excited atoms of the same elements and thereby lowering the
sensitivity of the method (Levenson, 2001).
Figure 3.3. A schematic view of an ICP-AES, (Levenson, 2001)
Thereafter the similarities end since it is the light emitted from the plasma torch and not the
individual ions, as is the case for ICP-MS that is analyzed. The light from the plasma torch is,
through diffraction grating, refracted in different wavelengths and detected by photomultiplier
tubes. The specific wavelength of different elements makes it possible to detect up to 40
elements simultaneously (Levenson, 2001).
3.2.4. Measurement of pH of soil samples
The pH of the soil and ash samples was measured according to the method in Bergil and
Bydén 1995. The soil samples were prepared by air-drying 15 g until the weight had
stabilized. Then 100 ml Milli-Q was added and the samples were mixed on a reciprocating
shaker for 1 h. The samples were then stored over night for sedimentation of heavier particles.
The next day pH was measured using a WTW pH-electrode SenTix 81 with a WTW Multi
350i.
3.3. Experimental Design
Due to the large amount of results from the leaching experiments, the experimental setup was
designed according to a factorial design with two factors; leaching time and L/S-ratio, with 3
respectively 2-3 levels. Each leaching parameter was performed in triplicates (some
exceptions for ash samples due to low samples amount) to assure a more robust design.
18 |
Chalmers University of Technology | 4. Results and discussion
This study was conducted with the intention to optimize the leaching of copper from soil and
bark ash. Many of the results are promising although the heterogeneity of the samples
sometimes makes it rather hard to conclude the success of the process. As for example more
than 100% copper has been extracted in some of the leaching experiments even though
accumulation from the process water and weight loss is included. Moreover the metal amount
is higher in some ash samples than the initial content of the bark. The heterogeneity of the
samples is probably a major source of error. Although the very small sample amount for the
experiments as well as the small samples sent for total amount analysis.
4.1. Initial metal concentrations in soil, bark and ash
The metal concentrations from the solid samples from Köpmannebro, according to the ICP-
MS analysis, where compared with earlier studies (Kemakta, 2012), Table 4.1. Some of the
values are rather consistent but most differ by at least 50%, which stresses the fact that the
samples are far from homogeneous. However given the large variance in earlier studies,
where a tenfold difference or more is not unusual between the lowest and highest amount, the
difference between earlier studies and this one is not that remarkable. More importantly the
copper amounts are equivalent.
Table 4.1. Comparison of this study's measured metal concentrations with an average from earlier studies.
mg/kg As Ba Cd Co Cr Cu Hg Ni Pb V Zn
Average 0.7 79 0.10 4.8 8.6 1250 0.20 6.2 11 19.8 26
Soil
This study 0.6 497 0.03 2.2 41.5 1090 <0.04 3.1 8 44.4 14
Average 1.8 47 0.30 1.1 2.8 13700 0.30 3.3 57 3.3 74
Bark
This Study 2.0 109 0.27 1.2 3.9 11300 0.06 3.5 31 6.7 44
In Table 4.2 the complete results from the total amounts analyses are presented. The yellow
and red marked values are those that exceed the values for KM (sensitive land use) and MKM
(lesser sensitive land use) respectively, for further information concerning MK and MKM (see
section 2.3.1.). The copper is, as expected, the major contamination although barium has high
enough amounts to be a problem as well. Other exceeding values are those of the ash samples,
for which excessive accumulation of metals during incineration is expected, further discussed
in chapter 4.1.1. Due to the high degree of contamination it is not likely that these samples
will be below the MKM-limit even after leaching.
20 |
Chalmers University of Technology | Table 4.2. Metal concentrations in all solid samples compared with MK and MKM limits. All concentrations are from
the ICP-MS analysis with an uncertainty of 20-25%. Levels above KM is marked in yellow while those above MKM is
marked in red.
Soil Bark Ash
mg/kg Köpmannebro KM MKM
Köpmannebro Björkhult Köpmannebro Björkhult Björkhult
Renova Oxidizing Reducing
As 0.59 0.28 2.0 7.8 <3 11 5.2 22 10 25
Ba 500 860 110 620 430 930 520 2300 200 300
Be 1.4 1.9 0.12 0.67 <0.5 1.5 1.2 1.5
Cd 0.032 0.014 0.27 0.095 0.15 1.3 <0.1 0.27 0.5 15
Co 2.2 0.25 1.2 0.94 2.3 7.5 3.9 2.2 15 35
Cr 42 25 3.9 9.0 71 50 23 24 80 150
Cu 1100 720 11000 15000 19000 130000 43000 110000 80 200
Hg <0.04 0.055 0.062 0.37 <0.01 <0.01 <0.01 <0.01 0.25 2.5
Mo 0.29 0.22 0.25 0.30 3.1 2.4 1.9 1.4 40 100
Nb 9.0 5.9 0.42 7.0 5.4 8.8 <5 6.4
Ni 3.1 0.25 3.5 2.6 21 26 15 7.9 40 120
Pb 7.9 3.2 31 39 61 360 91 150 50 400
S 76 <50 570 510 2700 4900 1900 2400
Sc 7.7 1.9 0.56 1.2 <1 5.9 3.4 2.9
Sr 190 200 30 95 130 230 110 260
V 44 9.7 6.7 7.0 12 38 22 15 100 200
W 1.2 0.73 0.58 0.58 <50 <50 <50 <50
Y 17 5.4 2.3 6.9 2.9 19 33 8.0
Zn 14 3.5 44 36 1500 260 26 160 250 500
Zr 230 110 5.7 42 13 150 22 95
4.1.1. Effects on metal concentrations from the incineration
As mentioned earlier, the fact that the ash samples have high metal concentration is not
surprising. What is striking though is that some of the metals present in high concentrations
are considered volatile and would have been more likely to accumulate in the fly ash. The
metals in question are e.g. cadmium (Cd), lead (Pb), and zinc (Zn), where lead exceeds the
KM-limit for all samples (Table 4.2). However, as mentioned earlier the ratio between bottom
ash and fly ash depends on several factors and with this incineration method these metals have
clearly accumulate in the bottom ash. In Table 4.3 the percentage of the metals in the bark that
is staying in the bottom ash is presented.
As shown in Table 4.3, the percentages of metals left in the ash has quite reasonable values
for cadmium (Cd), lead (Pb) and zinc (Zn) as the amount often is around 50% or below,
which proves that most of the metals are enriched in the fly ash and the high amount in Table
4.2 is mostly due to the high amounts in the bark. The exceptions are the lead content in the
ash incinerated at oxidizing conditions and the zinc content in the ash incinerated at Renova,
which probably are due to heterogeneity in the samples. The ash from Björkhult tends to have
higher content for all metals, which probably is because of the higher content of sand in the
bark which is to high extent unaffected by the incineration and then releases adsorbed metals
from its surface in the preparation step for the ICP-MS-analysis and thereby raises the levels
for this sample.
21 |
Chalmers University of Technology | bark from Köpmannebro. In order to minimize the usage of process water L/S-ratios of 2 and
5 for both soil and ash were used. However, initial experiments indicated that an L/S-ratio of
2 was too low for ash samples since the entire leaching agent volume was adsorbed by the
sample. Instead L/S-ratio 10 was included in the experimental set-up. As the results later will
show L/S-ratio 10 was preferable; therefore the experimental set-up was extended to include
L/S-ratio 10 for the soil samples as well.
Each sample was made in triplicate and the amount of copper measured with
spectrophotometry. Measured copper content in the leachate was adjusted by subtracting the
initial copper content in the process water: in order to get a better approximation of the
amount of contaminant leached. In Figure 4.1 the results from these optimization experiments
are presented; the results are normalized on the basis of the highest amount of leached copper
to get an easy overview of the different parameters’ effects. As seen in Figure 4.1 the L/S-
ratio 10 leached 50% more copper than the L/S-ratio 5, while trends connected with the
leaching time are less pronounced.
Comparision of leached Cu from Köpmannebro ash,
optimization experiments
1,0
0,9
0,8
0,7
d
e 0,6
h
c
a 0,5
e
l
u 0,4
C
0,3
0,2
0,1
0,0
L/S 5 30min L/S 5 60min L/S 5 90min L/S 10 30minL/S 10 60minL/S 10 90min
Figure 4.1. Amount of Cu leached with different parameter settings with ash from Köpmannebro. Normalized results
with basis on the largest amount of leached Cu. All results mean of triplicate samples analyzed with
spectrophotometry. Note that the results are corrected for the amount of copper present in the process water.
The washing water from the optimization experiments was analyzed by spectrophotometry as
well. These results were not corrected regarding the initial copper content in the process water
because in contrast to the leachate analysis, it is interesting to know the total amount of
copper released, in the washing water. The washing water is interesting in order to determine
how much weakly bound metal the sample can leach out and by this estimate how the sample
would behave at a landfill. As can be observed in Figure 4.2 no obvious trend is evident.
23 |
Chalmers University of Technology | Comparision of leached Cu in washing water from
Köpmannebro ash, optimization experiments
1,0
0,9
0,8
0,7
d 0,6
e
h
c a 0,5
e
l
u 0,4
C
0,3
0,2
0,1
0,0
L/S 5 30min L/S 5 60min L/S 5 90min L/S 10 30minL/S 10 60minL/S 10 90min
Figure 4.2. Amount of Cu in the washing water from the different leaching experiments with ash from Köpmannebro.
Normalized results with basis on the largest amount leached Cu. All results mean of triplicate samples analyzed with
spectrophotometry. Note that the results are not corrected with the amount of copper present in the process water
used as leaching agent pre washing.
Figure 4.1 and 4.2 show the comparison between the results of the different parameter
settings, but the actual amounts of copper removed are important.
In Table 4.4 the amount of copper per kg solid content ash is presented as well as the
remaining amounts of copper in samples after leaching. As seen the amount of copper added
by the process water is insignificant compared to the initial amounts in the samples, but is
important to consider since the process water could vary in future trials. An interesting result
is the amount of copper left in the sample that has been leached for 30min with L/S 10, as it is
negative. This is due to the measuring insecurity of the spectrophotometric analysis as well as
the external lab’s measuring insecurity with the initial copper content. The heterogeneity of
the samples is also a source of error, since only one solid ash sample, from each type of
incineration, was sent for external analysis and the ash used in the experiments could vary in
copper content compared to these samples.
The values of remaining copper are low, if you calculate the percentage of copper removed it
is close to 100%, however the remaining content still has to be evaluated to determine if it is
low enough for non-hazardous or hazardous landfill or if more treatment is needed (see
section 4.3.).
24 |
Chalmers University of Technology | Table 4.4. Amount of copper per mass unit of solid content ash in the different steps of the leaching process.
m Cu leached total amount of Cu amount of
per m ash, initial Cu content, added via process m Cu in washing Cu left in
spectroscopy extern analysis water, external water per m ash the samples
[mg/kg TS] [mg/kg TS] analysis [mg/kg TS] [mg/kg TS] [mg/kg TS]
L/S 5 30min 14900 79400 15 1630 62900
L/S 5 60min 36700 78200 15 570 40900
L/S 5 90min 33700 77800 15 1050 43100
L/S 10 30min 78100 80200 31 2330 -209
L/S 10 60min 73300 78500 30 1084 4140
L/S 10 90min 71900 79200 30 741 6560
To empower our conclusions on which were the optimal parameters, an ANOVA (statistical
analysis of variance) was made of the leachates’ copper content according to the
spectrophotometric analysis. The results from the ANOVA (Table 4.5) clearly indicate that
the only parameter with significant effect (p-value < 0.05) is the L/S ratio as suggested by the
plot in Figure 4.1. Concluding the significance of L/S-ratio, the optimal setting of L/S-ratio 10
was used for the succeeding optimization process for ash samples. When determining which
leaching time to proceed with the reasoning was, since it was insignificant for the amount of
copper leached; shorter time is preferable, especially in large-scale processes, therefore
leaching time 30min was chosen.
Table 4.5. ANOVA of the optimization experiments of ash from Köpmannebro.
ANOVA SS df MS F F p-value
table
Leaching time 482 2 241 0.60 3.89 0.56
L/S 14900 1 14900 37.25 4.75 0.000053
Interaction 978 2 489 1.22 3.89 0.33
Error 4810 12 401
Total 21200 17
4.2.2. Optimization of leaching from soil samples from Köpmannebro
Similar to the ash samples, optimization experiments were made with the soil samples from
Köpmannebro. The initial experimental set-up was extended with L/S-ratio 10 since it was
successful with the ash samples; hence both L/S-ratio and leaching time had three levels. As
in the case of the ash samples, all leachates were evaluated by spectrophotometry. The
spectrophotometric analysis was unfortunately non suitable for the soil samples because soil
particles dyed the leachates resulting in a too high absorption. However it was assumed that
the error due to the coloring remaining after filtration was equal for all of the samples and
therefore the results were still accurate for conclusions regarding optimization. The
spectrophotometric results were compensated regarding the initial copper content of the
process water before evaluation. In Figure 4.3, where the results have been normalized on the
basis of the highest amount of copper removed, it is evident that the L/S 10 is superior to
lower ratios while as for the ash no obvious trend for leaching time could be deduced.
25 |
Chalmers University of Technology | Comparison of leached Cu from Köpmannebro
soil, optimization experiments
1,0
0,9
0,8
0,7
d
e 0,6
h
c a 0,5
e
l
u
0,4
C
0,3
0,2
0,1
0,0
L/S 2 L/S 2 L/S 2 L/S 5 L/S 5 L/S 5 L/S 10 L/S 10 L/S 10
30min 60min 90min 30min 60min 90min 30min 60min 90min
Figure 4.3. Amount of Cu in leachates from the different leaching experiments with soil from Köpmannebro.
Normalized results with basis on the largest amount leached Cu. All results mean of triplicate samples analyzed with
spectrophotometry. Note that the results are corrected with the amount of copper present in the process water used as
leaching agent.
The washing water from the soil samples was also analyzed with spectrophotometry. The
results shown in Figure 4.4. are not compensated regarding the initial copper content in the
process water. In difference to the corresponding results for the ash samples a trend can be
seen for the soil’s washing water. It is clear that a higher L/S ratio during leaching results in a
less contaminated washing water.
Comparision of leached Cu in washing water
from Köpmannebro soil
1,0
0,9
0,8
0,7
d
e 0,6
h
c a 0,5
e
l
u
0,4
C
0,3
0,2
0,1
0,0
L/S 2 L/S 2 L/S 2 L/S 5 L/S 5 L/S 5 L/S 10 L/S 10 L/S 10
30min 60min 90min 30min 60min 90min 30min 60min 90min
Figure 4.4. Amount of Cu in washing water from the different leaching experiments with soil from Köpmannebro.
Normalized results with basis on the largest amount leached Cu. All results mean of triplicate samples analyzed with
spectrophotometry. Note that the results are not corrected with the amount of copper present in the process water
used as leaching agent pre washing.
A similar table as Table 4.4 for the ash was made for the soil, here (Table 4.6) the
disadvantages with using spectrophotometric analysis is obvious because almost all of the
results are showing a more than 100% removal of copper. In addition to the unsuited method
of analysis the heterogeneity issue is the same as for the ash.
26 |
Chalmers University of Technology | Table 4.6. Amount of copper in all of the leaching process steps for soil from Köpmannebro.
amount of Cu in
amount of Cu total amount of Cu washing water per
leached per mass of initial Cu content, added via process mass of soil, amount of Cu left in
soil, spectroscopy extern analysis water, external spectrophotometry the treated samples
[mg/kg TS] [mg/kg TS] analysis [mg/kg TS] [mg/kg TS] [mg/kg TS]
L/S 2 30min 1010 1090 6 212 -123
L/S 2 60min 1250 1090 6 183 -335
L/S 2 90min 759 1090 6 231 106
L/S 5 30min 1050 1090 15 93 -34
L/S 5 60min 1840 1090 14 74 -812
L/S 5 90min 1300 1090 14 147 -340
L/S 10 30min 8480 1090 29 100 -7460
L/S 10 60min 3510 1090 29 110 -2500
L/S 10 90min 7120 1090 29 92 -6090
Even though the amounts of copper removed probably are deceptive concerning actual copper
contents, the results are still assumed to be comparable and therefore an ANOVA was made
for the results. The statistical analysis (Table 4.7) of soil leachates concluded the same as for
the ash leachates; the only significant parameter is L/S-ratio (p-value < 0.05). Since the L/S
10 had the superior leaching ability it was chosen as the L/S-ratio to move forward with, and
as for the leaching time the same reasoning as with the ash samples was made; shorter time is
better and 30min was chosen.
Table 4.7. ANOVA of the optimization experiments of soil from Köpmannebro.
ANOVA SS df MS F F p-value
table
Leaching time 67100 2 33500 1.27 3.89 0.316
L/S 1350000 2 677000 25.62 4.75 0.0000054
Interaction 277000 4 69300 2.62 3.89 0.069
Error 476000 18 26400
Total 2170000 26
4.2.3. Two-step leaching
The washing water of the ash samples still contained much copper and therefore the
optimization process continued with studying if a two-step leaching could further improve the
amount copper removed in the leaching process. Three different combinations were tested for
the ash samples; 15+15min, 30+15min and 30+30min all with an L/S-ratio of 10. All the
samples consisted of an approximate 50/50 mixture of ash incinerated at Renova and ash
incinerated at the lab under oxidizing conditions and was made in duplicates. The leachates
were then analyzed with spectrophotometry; the results are presented in Figure 4.5 where the
results once again are normalized on the basis of the highest amount of copper removed. As
seen in the figure the 30+30min seems to be the best option. Another interesting observation
of the results is that a considerable amount of copper is leached in the second step, which
indicates an advantage with a two-step process.
27 |
Chalmers University of Technology | Comparision of Leached Cu from two-step
leaching experiments, Köpmannebro ash
1,00
0,90
0,80
0,70
d
e 0,60
h
c a 0,50 Step 2
e
l u 0,40 Step 1
C
0,30
0,20
0,10
0,00
L/S 10 15+15min L/S 10 30+15min L/S 10 30+30min
Figure 4.5. Amount of Cu in leachates from the different two-step leaching experiments with ash from Köpmannebro.
Normalized results with basis on the largest amount leached Cu. All results mean of duplicate samples analyzed with
spectrophotometry. Note that the results are corrected with the amount of copper present in the process water used as
leaching agent.
The washing water from the two-step leaching experiments was also analyzed with
spectrophotometry. As can be seen in Figure 4.6 the two-step leaching with at least one step
of 30min show lower amounts of copper than the experiment with only two 15min steps.
Comparision of leached Cu in washing water
from two-step leaching experiments,
Köpmannebro ash
1,0
0,9
0,8
0,7
d 0,6
e
h
c a 0,5
e
l
u 0,4
C
0,3
0,2
0,1
0,0
L/S 10 15+15min L/S 10 30+15min L/S 10 30+30min
Figure 4.6. Amount of Cu in washing water from the different two-step leaching experiments with ash from
Köpmannebro. Normalized results with basis on the largest amount leached Cu. All results mean of duplicate samples
analyzed with spectrophotometry. Note that the results are corrected with the amount of copper present in the process
water used as leaching agent pre washing.
When the actual amounts of copper removed are calculated the problems with the
spectrophotometric analysis is obvious since all of the experiments show a more than 100%
removal of copper, as seen in Table 4.8. However we believe that the results are still
28 |
Chalmers University of Technology | comparable relative each other, which gives an indication of 30+30min being the optimal
leaching time.
Table 4.8. Amount of Cu in the different steps in the two-step leaching.
amount of Cu total amount of
leached per mass Cu added via amount of Cu in amount of Cu
of ash, initial Cu content, process water, washing water left in the
spectroscopy extern analysis external analysis per mass of ash treated samples
[mg/kg TS] [mg/kg TS] [mg/kg TS] [mg/kg TS] [mg/kg TS]
L/S 10 15+15min 85200 78900 30 71 -6390
L/S 10 30+15min 85400 78200 50 17 -7170
L/S 10 30+30min 103000 79600 30 40 -23900
As earlier mentioned 30+30min leaching seems like the best option by the results presented in
Figure 4.5. An ANOVA was made on the results to see if this conclusion was statistically
significant as well (Table 4.9). This was not the case, which could be the result of too few
replicates, but due to the limited amount of sample no further experiments were made.
Concluding the results from the two-step leaching 30+30min was determined to be best and to
use this setting for further experiments with ash samples.
Table 4.9. ANOVA of two-step leaching results
ANOVA SS df MS F F p-value
table
Leaching time 332 2 166 0.222 3.89 0.804
Error 2240 3 746
Total 2570 5
4.2.4. Batch experiments
After the two-step leaching experiments it was determined to continue with 30+30min
leaching with an L/S-ratio of 10 for ash samples, while only one step leaching of 30min for
soil samples. The next step was to try these optimal conditions on a larger sample size, a batch
sample. The spectrophotometric analysis was only used to compare the results, and its
qualitative accuracy was doubted. In addition to this it was only suitable for analyzing the
copper content while other metal content also is of interest in investigating the success of the
method. Unfortunately the leaching exceeded 100% for some samples (Table 4.8) even with
the external analyses. Explanations to this are probably the heterogeneity of the materials in
addition to the uncertainty of the external analysis.
In Figure 4.7 the result from the batch experiments of ash is presented. The leachates contain
much copper than both the process water and the washing water from both Köpmannebro and
Björkhult ash. This means that the copper content in the process water is negligible as well as
that most of the removable copper is actually removed by the leaching. A difference can be
seen between the two sampling points; the second leaching step is of much more importance
for Köpmannebro ash compared to Björkhult ash.
29 |
Chalmers University of Technology | Comparison of amount Cu in the different steps
of the process in batch experiments
90000
80000
]
S
T 70000
g
k
/ 60000
g
m
step 1 [mg/kg TS]
[ 50000
u
C f 40000 step 2 [mg/kg TS]
o
t n 30000 washing water [mg/kg TS]
u
o
m 20000 process water [mg/kg TS]
a
10000
0
Köpmannebro
Björkhult
Figure 4.7. Results from two-step leaching batch experiments of ash from Köpmannebro and Björkhult. Both batches
being leached for 30+30min with L/S-ratio 10. The copper content for both steps of leaching, the washing water and
the leaching agent being presented in mg of Cu leached per solid mass of ash.
When comparing the leached amount of copper with the initial content in the ash a problem
appear and for the Köpmannebro ash the removal is more than 120%. This could, as described
earlier be due to the heterogeneity of the ash. Interesting to see is that the Björkhult ash,
which only was incinerated by one method, has a more believable result, which indicates that
future research should use only one method, which is more efficient for consistent results.
Another way of handling this problem could be to send both untreated and treated ash for
external analysis.
Table 4.10. Amounts of Cu in the different steps in the batch leaching experiments of ash from Köpmannebro and
Björkhult.
initial Cu
step 1 step 2 washing water process water content Amount of Cu left in
[mg/kg TS] [mg/kg TS] [mg/kg TS] [mg/kg TS] [mg/kg TS] the samples [mg/kg TS]
Köpmannebro 36700 25900 939 63 51600 -11800
Björkhult 84000 15000 820 58 107000 7300
In Figure 4.8, the result of batch experiments of the soil is presented. As seen for both soils
most of the removable copper is extracted in the leaching step and as with the ash, the process
water’s copper content is insignificant compared to the amount in the leachate.
30 |
Chalmers University of Technology | Comparison of amount Cu in the different step of the
process in batch experiments of soil
1400
amount leached Cu
[mg/kg TS]
]
1200
S
T
g 1000 washing water
k
/ g [mg/kg TS]
m
800
[
u process water
C
f o 600 [mg/kg TS]
t
n
u 400
o
m
a 200
0
Köpmannebro Björkhult
Figure 4.8. Batch leaching of soil, 30min with L/S 10. Results based on external analysis of both soil and leachate and
are corrected regarding the initial copper content in process water.
When looking into the actual contents in the leachates the same problem as with the
Köpmannebro ash occurred, i.e. more than a 100% removal. This could be due to the
insecurity of the analysis or that the external lab analyzed only the fine-grained particles of
the soil while the leaching experiments leached the whole particle size distribution.
Table 4.11. Amounts of Cu in the different steps of batch leaching experiments of soil.
Amount of
amount washing process initial Cu Cu left in
soil batch leached Cu water water content the samples
experiments [mg/kg TS] [mg/kg TS] [mg/kg TS] [mg/kg TS] [mg/kg TS]
Köpmannebro 1170 62 29 1090 -117
Björkhult 1380 58 29 719 -686
4.2.5. Long leaching time
Some earlier research suggests a much longer leaching time to efficiently release metals from
soil (Udovic, 2012). Therefore 18h and 24h-leaching time were tested as well. The results
presented in Figure 4.9 illustrate that the percentage copper leached decreases with longer
leaching time if you compare 18h with 24h. The difference between 18h and 30min is
negligible if you look at the scale of the axis since if you put the difference in numbers it is
less than 100mg/kg TS. With this in mind the advantages with a 30min treatment time
compared to 18h is obvious and the decision for 30min is easy. As seen later, when
comparing the content of other metals in the leachates, the longer leaching time have the
disadvantage of other metals from the process water accumulating in the soil, something that
is avoided to a greater extent with 30min leaching.
31 |
Chalmers University of Technology | 4.2.7. Washing water
The same error in the spectrophotometric analysis due to coloring occurred for the washing
water samples, but the same assumption of the error being the equal for all samples was
drawn. When converting the amounts of copper in the washing waters to percentage of initial
content soil it is easy to believe that since the percentage is higher for soil than ash so is the
copper content. This is a faulty conclusion since the initial copper content is so much lower in
the soil hence the actual copper content in the washing water of soil samples is actually lower
than the washing water of ash samples. In Figure 4.11 the amount copper per mass unit of dry
content sample is presented and it is easy to see that all of the soils washing waters have less
copper than the ash washing waters. The improvement in cleaner washing water for two-step
leaching compared to one-step for the ash samples is also easy to see.
Figure 4.11. Amount of Cu per kg of sample in washing waters of soil and ash samples.
4.3. Handling and after-treatment of the leached soil and ash
The main aim with this master thesis was to extract as much copper from the soils and ashes
as possible. However the handling of the leached soil and ash is of great importance as well
since disposal of these could be costly depending on how polluted they are. As described in
chapter 2.3.1 there are mainly two different criteria (KM/MKM and leaching test SS-
EN12457-3) on how to handle the soil and ash dependent the degree of contamination.
For the KM/MKM-criteria the total amounts in the material are measured. Since no
quantification of metals in the leached soil and ashes were made the values used are
calculated from the initial and the leached amounts. This leads to some negative
concentrations due to the heterogeneity of the samples but still gives a hint of the actual
amount remaining.
The other criteria is leaching e.g. the SS-EN12457-3 leaching test where the water-leachable
amount from the material is measured, to predict how much that will leach to the
33 |
Chalmers University of Technology | surroundings at a landfill. Both of these analyses are based on the quantification made with
ICP-MS and ICP-AES.
4.3.1. The KM/MKM-criteria
As been discussed earlier the leaching of copper has succeeded to a great extent, which also
can be seen in Table 4.11 and 4.12. In regards of the KM/MKM-criteria for copper all
samples can be considered safe for reuse except the ash from Björkhult. However 93 % of the
copper has been removed from this sample, which still is a success considering reuse of the
copper. For some of the samples more than 100% of the copper was extracted which is mostly
due to the heterogeneity of the sample. More sample would have been needed for a more
precise result but the trend support the success of the copper extraction.
Table 4.11, Calculated metal concentrations in the leftover of the soil/ash samples compared to the KM/MKM-limits.
Values marked with yellow is above KM and red is above MKM. These values can be compared with Table 4.1 from
before leaching. The values are compensated with the weight loss from the leaching.
Soil Ash
mg/kg Köpmannebro Björkhult Köpmannebro Björkhult KM MKM
30 min 18 h 24 h 30 min 30+30 min 30+30 min
As 0,5 1,3 1,2 0,1 0,4 <1 10 25
Ba 494 491 492 872 1051 2583 200 300
Be 1,4 1,4 1,4 1,8 1,3 1,6
Cd <0,5 <0,5 <0,5 <0,5 <0,5 0,10 0,5 15
Co 1,9 1,1 1,3 0,1 4,8 2,3 15 35
Cr 42 41 41 25 115 27 80 150
Cu <1 7,31 <1 <1 <1 8249 80 200
Mo 0,5 0,5 0,5 0,3 1,6 0,8 40 100
Ni 2,6 1,4 1,8 0,1 <1 <1 40 120
Pb <1 15,2 8,1 7,2 4,3 138,3 50 400
Sr 194 194 194 206 15 240
V 44 41 42 9 23 11 100 200
Zn <1 90 15 38 <1 <1 250 500
A longer leaching time for soil leads to uptake of arsenic from the process water, which can
be seen in Table 4.12 where the total percentage extracted is negative. In this case, the levels
of arsenic (As) is far below the limits but if the process water would have higher arsenic
concentration this could be a problem. In earlier studies the arsenic levels of the process water
has been twice as high and due to the fluctuation in the material incinerated it could be higher.
Unlike for the soil the process water extracts almost all arsenic from the ash samples. This is
of importance since the initial levels in the ash were above the guidelines (see section 4.1.)
and indicates that the process water could be used to clean ashes from arsenic in addition to
copper.
Other metals that were enriched, i.e. got a negative value in Table 4.12, in the soil by the
process water were molybdenum (Mo), lead (Pb) and zinc (Zn). Molybdenum was enriched in
all of the soil samples although, as with arsenic, far below the guidelines. When it comes to
lead and zinc it is hard to draw any conclusion, as the sample from Köpmannbro leached for
18 hours and the one from Björkhult enriched these metals, while the sample from
Köpmannebro that has been leached for 30 minutes and 24 hours have very different results.
The enrichment of metal contaminations in the soil was as mentioned noticed at low levels
which due to the insecurity of the results makes it hard to prove but it is still an important
34 |
Chalmers University of Technology | trend to be aware of in further studies. Worth mentioning is that the ash samples do not have
any uptake from the process water from any of the quantified metals.
Table 4.12. The percentage of metal removed from the samples in the leaching experiments. A negative value means
that there has been an uptake from the process water to the sample.
Soil Ash
% metal
Köpmannebro Björkhult Köpmannebro Björkhult
removed
30 min 18 h 24 h 30 min 30+30 min 30+30 min
As 10 -116 -102 56 96 108
Ba 3 3 3 1 11 1
Be 2 3 3 6 20 6
Cd 676 424 312 2077 131 69
Co 15 49 40 48 38 8
Cr 1 4 4 2 14 2
Cu 113 99 116 199 123 93
Mo -57 -60 -58 -38 74 49
Ni 17 55 44 47 431 1269
Pb 169 -87 0 -118 99 19
Sr 1 1 1 0 95 17
V 4 9 8 13 43 37
Zn 282 -520 0 -972 391 140
Even though the leaching succeeded in extracting all or most of the major contaminant, i.e.
copper, the barium (Ba) levels are still too high for all samples to be regarded as KM or MKM
(see Table 4.11). This is much due to the process waters inefficiency to extract barium
(generally around 3.0 %). This is not that of a problem for the ashes since these got other
contamination that prevents them from this usage but as Ba is the only element in soil residue
exceeding the guidelines it would have been a major success if it could be extracted and the
soil could be returned to the site and thereby removes the need of other filling material (for
comparison with initial concentration see Table 4.1).
Other results of interest from Table 4.11 and 4.12 are the successful extraction of cadmium
(Cd) and lead (Pb) from the ash from Köpmannebro. These levels where initially over the
MKM-limit (see section 4.1.) but was reduced below the KM-limit after leaching. This was
not the case for the ash from Björkhult where the process water only managed to extract 18%
lead compared to 99% for the ash from Köpmannebro. This might be due to that the lead is
adsorbed on the sandy soil existing in the ash sample from Björkhult (the layers were not as
divided as for Köpmannebro) since lead otherwise is rather easy to extract from ash with
acidic leachate (Ohtsubo et al., 2004).
Finally the chromium (Cr) amount is initially below the KM-limit but has increased after the
leaching. This is not due to any uptake from the process water but rather due to the ashes
weight loss during leaching. The weight loss (presented in Table 4.13) for the Köpmannebro
ash is so high that metal levels that are initially close to the limits risk breaking them if the
leaching was not successful of the particular metal. In this study however this only applies to
chromium for the ash sample from Köpmannebro. This loss of solid material is also one
reason why the ash from Köpmannebro still have high amount of contaminations compared to
total mass after leaching since it loses more than half of its weight.
35 |
Chalmers University of Technology | Table 4.13, The weight percentage of how much solid material that is lost in the leaching and washing step.
Köpmannebro Björkhult
Ash Soil Ash Soil
51,4 2,1 11,9 4,1
4.3.2. The results from SS-EN12457-3 leaching test
The result from the EN-leaching test (Table 4.14) gives an indication of how to handle the
disposal of soil and ash. However not all of the contamination (can be found in NFS 2004:10)
was analyzed so the recommendation of methods of disposal are only made regarding the
metals presented in Table 4.14. This discussion will focus on the copper content since it is the
main substance in this thesis eventhough there are also other metals (e.g. Zinc (Zn) and
selenium (Se)) contributing to exceeding the guidelines the copper concentration raises the
risk grade even higher than the zinc or selenium concentration alone.
Table 4.14. Metal concentration in the water leachates after shake test (SS-EN12457-3). Green marked values is those
within the non-hazardous waste limit, yellow marked is those within the hazardous waste limit and the red marked is
those above the hazardous waste limit.
mg metal/kg soil or ash As Ba Cd Cr Cu Hg Mo Ni Pb Sb Se Zn
Köpmannebro 30+30 min 0,04 2,9 0,11 0,04 1019 0,04 0,76 0,02 0,07 50,2
A Köpmannebro 30+30 min, no wash 0,04 3,5 0,24 0,07 2440 0,03 1,84 0,12 0,14 94,7
S Köpmannebro 30 min 0,04 2,2 0,00 0,11 11 0,18 0,11 0,04 0,22 1,1
H Björkhult 30+30 min 0,01 7,9 0,02 0,03 752 0,03 0,08 0,22 0,05 3,7
Björkhult 30 min 0,11 10,6 0,05 0,28 3043 0,28 0,28 0,24 0,56 7,1
Köpmannebro 30 min 0,01 1,4 0,03 0,12 78 0,03 0,55 0,07 0,05 7,6
S Köpmannebro 18h 0,01 1,3 0,03 0,10 51 0,03 0,51 0,07 0,05 7,2
O Köpmannebro 24h 0,01 1,1 0,02 0,09 39 0,03 0,40 0,06 0,05 5,6
I Köpmannebro untreated 0,03 0,5 0,00 0,03 52 0,03 0,03 0,15 0,05 0,4
L Björkhult 30 min 0,01 1,6 0,02 0,03 66 0,03 0,10 0,08 0,05 3,3
Björkhult untreated 0,02 0,5 0,00 0,03 41 0,03 0,03 0,06 0,05 0,3
Upper concentration limits
Inert waste 0,5 20 0,04 0,5 2 0,01 0,5 0,4 0,1 0,06 0,1 4
Non-hazardous waste 2 100 1 10 50 0,2 10 10 10 0,7 0,5 50
Hazardous waste 25 300 5 70 100 2 30 40 50 5 7 200
Above hazardous waste >25 >300 >5 >70 >100 >2 >30 >40 >50 >5 >7 >200
4.3.2.1. Results of the EN-leaching of the ash samples
All of the ashes, except for the one leached in one step for 30 minutes, are well above the
limits for needing be handled at sites for hazardous waste. The fact that all of the samples
exceed limits for hazardous waste is not that surprising due to the extremely high initial
amount of copper. However, it is somewhat suprising that the Köpmannebro ash leached once
for 30 minutes has so low copper values compared to the other samples especially as the same
sample has a lower copper yield in the actual leaching test. Logically it should have more Cu
left in the ash complex than other more successfully leached samples (chapter 4.2). This is
mainly due to that a higher L/S-ratio was used during the washing step for the first
experimental trials. At first L/S 5 was used but was later lowered to L/S 2 due to wanting to
minimize the water amount. A washing step with L/S 2 was used for all samples in Table 4.14
except for the Köpmannebro ash leached for 30 minutes. Another probable explanation is that
the process water has not been given enough time to destroy the ash matrix and thereby
weakened the metal-to-ash bonds. The heterogeneity of the samples is also a likely
contributor to the low copper concentration although a far smaller source of error.
36 |
Chalmers University of Technology | Another important remark in the results from the EN-leaching is the importance of the
washing step. The ash sample from Köpmannebro leached in two steps; 30+30min, was tested
both pre and post washing. Comparing the results it is noted that the pre washing sample
leached twice the metal amount or higher for most of the metals analyzed compared to the
post washing sample. Even though the washing step seems to be of importance in the leaching
process, the copper values are still ten times the limit for being treated as hazardous waste.
However, comparing these results with the one for Köpmannebro ash leached for 30 minutes
has only 1% of the earlier mentioned. This is below the non-hazardous waste limit and is
probably most due to the more extensive washing step for this sample (L/S 5 instead of L/S 2
as mentioned earlier). This suggest clearly that the washing step is insufficient and needs to be
improved by a higher L/S-ratio and perhaps a longer washing time.
The results of the ash samples from Björkhult indicates that the two-step leaching
substantially improves the leaching since all of the analyzed metal concentration has been
lowered compared to one step leaching. However, similar to the ash samples from
Köpmannebro, the copper concentration after washing is still far from being acceptable as
hazardous waste which once again might be due to an insufficient washing step.
4.3.2.2. Results of the EN-leaching of the soil samples
The soil samples have much lower metal concentrations compared to the ash samples. All of
them fulfill the requirements, regarding the analyzed metals, to be deposited at landfills for
hazardous waste. Two of the samples even comply with the limits of non-hazardous waste.
One of them is the soil sample from Köpmannebro, which has been leached for 24 hours.
Compared with the other Köpmannebro soil samples, leached for 30 minutes and 18 hours the
amount of all quantified metals leached in the EN-test decrease with longer leaching time.
Since the amount of copper removed in the process water leaching is approximately equal for
these samples (chapter 4.2) this suggest that longer leaching time fixate the metals left in the
soil matrix stronger with time. This is consistent with the results in Table 4.12 where fixation
of metals to the soil can be seen as accumulation from the process water to the soil.
The untreated soil samples from both sites had lower release of metals than the samples
leached with process water for 30 minutes. This indicates that the process water, as intended,
weakens the metal to soil bonds. Although the washing step has to be improved to remove
more of the weakly adsorbed metals left in the soil for the method to be even more efficient.
This could be done by extending the washing step, use a higher L/S-ratio for this step or using
a sequential/multiple steps washing.
Another conclusion from Table 4.2 and 4.14 is that the process water increases the amount of
zinc in the soil. This might be due to ion exchange between the process water and the soil/ash
matrix. While the untreated soil samples have 0.3 respectively 0.4 mg zinc per kg soil the
treated soil samples has zinc contents above 3 mg per kg soil. The amount of zinc in the soil is
not alarming, but indicates that process water can increase the toxicity of the soil. This is an
important effect to be aware of since the metal content of the process water differs depending
on what is being incinerated at Sävenäs.
In continuous trials other contaminants, which might accumulate even more in the soil, could
be present in a higher concentration which would result in a treated soil that need a more
expensive disposal method than the untreated one. However it is important to remember that
this is only at very low concentrations and only looking at the limits it is still better to treat the
37 |
Chalmers University of Technology | soil than to send it for landfill untreated even though some metals accumulate and some are
left more easily leachable after treatment.
4.3.3. The KM/MKM-criteria versus SS-EN12457-3 leaching test
These results are somewhat contradictory since it is barium (Ba) that is the major concern
regarding the KM/MKM-criteria while it got very low values in the leaching test. This
however is probably due to it is hard fixated to the matrix and will not be a problem leaching
to the surroundings even though it is present in very high amounts.
Even more contradictory is however the high levels of copper (Cu) that is extracted in the
leaching test while the levels should be very low, according to the KM/MKM-calculations.
This is probably due to that the KM/MKM-values are not measured but calculated from the
initial amounts in the solids and the concentrations in the leachates and washing waters.
Unfortunately this makes the results from the leaching test more reliable than those for the
KM/MKM-guidelines which, except for the barium levels, yields positive results concerning
reuse of the soil.
38 |
Chalmers University of Technology | 5. Conclusions
The overall result from the copper leaching with acidic process water is promising for both
sites in the regards of extracting copper for reuse purposes. However when it comes to the
amount still left in the leached samples the ash from Björkhult still has too high levels of
copper after leaching to be below the MKM-limit. This is a problem even though the process
water has managed to extract more than 90% of the copper.
The copper levels after leaching is only a problem for the ash from Björkhult, but all of the
samples have far too high levels of barium to be acceptable for reused as KM and even as
MKM. The barium levels are high but it seems as it still very stable in the sample. This has
been proven both in the leaching trials with process water and in the SS-EN12457-3 leaching
test.
The ash on the other hand has too high levels and leach too much, not only regarding barium
but for other metals as well, which might lead to a handling somewhat more costly. This
however might be solved using a more extensive washing step, which is proven to be of great
importance in the SS-EN12457-3 leaching test. As of now the amount of easily leachable
metals has increase in many samples but with a better washing step it could prove to be a
good decontamination technique.
Another important remark is the accumulation of metals from the process water to the soil. In
this study it did not lead to any hazardous levels of the accumulated metals. However, due to
the variations in the content of the process water this might prove to be a problem for further
studies.
39 |
Chalmers University of Technology | 6. Further research
For continued study within the this field of research it would be of major interest to improve
the washing step. This has been concluded to be insufficient in the regards of removing easy
leachable metal contaminations still left after leaching with process water. A longer washing
step with a higher L/S-ratio might give rise to cheaper ways of disposal or even reuse.
Another point of interest is the up scaling of the process. This especially applies to the mixing
and separation part where shaking and centrifugation is used in this study. Both of these
methods are impractical and uneconomical if not impossible to use for a large-scale process.
The mixing could be solved by for example stirring but the separation could prove more
difficult since it is of importance to remove as much process water as possible as this increase
the levels of contaminations in the soil or ash if not removed.
To solve the up-scaling cooperation with companies working with full scale soil washing can
be of interest since there already exists mobile solution for soil wash. These method utilizes
primarily water as leachate but the possibility to include a step for process water could be
interesting to look into due to the high leaching yield achieved in this study (Svevia).
Finally the results from extracting arsenic from ash with process waters suggest another field
of interest for the method. This is very interesting since arsenic is not only a wide spread
contamination but is often used in similar wood processing methods instead of blue vitrol.
This vastly increases the number of sites of interest.
40 |
Chalmers University of Technology | Preparing For Tomorrow
Exploring design adaptations of a wheel loader for a circular business model
Master’s Thesis within the Industrial Design Engineering and Industrial Ecology Programs
HAMPUS BERGSTRAND
CORNELIA JÖNSSON
Department of Energy and Environment
Division of Environmental System Analysis
Chalmers University of Technology
ABSTRACT
One suggested way of dealing with the ever-increasing demand for services and products from a
continuously growing population is through circular economy. The concept of circular economy
means, among other things, optimizing products´ use and useful life and making fewer products
accessible for more users.
The purpose of this thesis project is to explore the potential of a use and sustainability centred
product development process. The process is aiming to explore how a product system can be
redesigned to become more resource efficient. To explore the potential of the process, it is applied
on a case study. The targeted product is the Volvo wheel loader, L150H, presumptively involved in a
functional sales business model, inspired by circular economy.
Since there seems to be no consensus about what the process should look like, the project first
defines a suitable framework, based on the use(r) centred development process. The inclusion of
ecodesign and a product life cycle perspective is used to emphasize the sustainability aspects.
Inspired by backcasting, a future sustainable scenario is developed and key areas for reaching the
scenario are identified. Consumables, and more specifically, the main fuel filter is selected for the
redesign phase. Requirements for a new solution are identified and solution paths are presented for
reducing the environmental impact. An iterative design phase results in a remanufacturing concept,
and the corresponding re-design of the fuel filter.
The new filter concept is designed to last the entire life of the machine through a robust design and
by being remanufactured, instead of disposed, after each service. A conducted LCA shows that by
using one filter over the expected machine life, instead of an estimated 40 filters, major material
savings can be achieved. In addition, the assessment show that the concept also induces a reduction
in global warming potential by almost 90 % over the life cycle.
The project result indicates that the use and sustainability centred product development process not
only has the potential to generate a material resource efficient solution, but also, a solution which is
sufficiently responding to the users’ needs. Despite the promising result in this case, the process still
requires further development, and to be tested on more cases before its potential can be verified in
general terms.
Keywords: EcoDesign, Circular economy, EcoDesign framework, Functional sales, Remanufacturing,
Life Cycle Assessment (LCA), Design for environment, Backcasting, Design for circular business model,
Sustainable design |
Chalmers University of Technology | ACKNOWLEDGEMENTS
There are several people which we would like to thank for their contribution to this thesis. First of all,
our academic supervisor from the department of environmental systems analysis at Chalmers
University of Technology, Siri Willskytt, who has offered a lot of most appreciated support throughout
the project. Secondly, we would also like to thank the research project Mistra REES, which initiated
and made this thesis possible.
We would also like to send our warmest thanks to our industrial supervisors. Here in Gothenburg,
from the advanced technology and research department at Volvo Group, environmental researcher
Lisbeth Dahllöf. Lisbeth has supplied us with the best possible preconditions to carry out this project.
Our warmest gratitude also goes to our second industrial supervisor, Peter Eriksson, technology
planning manager at Volvo Construction Equipment in Eskilstuna; for the insights regarding the
company and the technological advancements, and also, for his hospitality during our two visits in
Eskilstuna.
Furthermore, we would like to direct gratitude to Helena Strömberg, director of the Industrial design
engineering master’s program at Chalmers University of Technology, for her most valuable support in
the finishing stages of the project. We would also want to thank our committed examiner, Professor
Anne-Marie Tillman, for the inspiration she shared, and for her most appreciated contribution to the
finishing touches on the report. Next, our thanks go to environmental researcher, Mia Romare, at
Volvo Group, for her support in conducting the final LCA.
Finally, we would like to kindly thank all interviewees for their much valuable contribution. Especially,
the people at the service provider, who really enriched the project with very interesting and important
insights to the user perspective.
Thank you!
Hampus Bergstrand Cornelia Jönsson
Gothenburg, January 2017 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
1 INTRODUCTION
Humanity has only one Earth and it has limited resources. If resources are not used wisely, they will
be more difficult to access in the future and the diversity of products we are used to have on the
market will require more energy to maintain (Smuk L. 2015). One suggested way to tackle this problem
is to decouple resource use from meeting the needs of the users by “closing the loops” and reuse
materials, to reduce extraction. This is often called a circular economy.
An emerging area of interest today is what implications the design phase can have on the
environmental impacts of a product. Today, the design phase only consumes about 15% of the
economic resources during a products life, but the decisions made to the function and design
determines the future environmental impact of the product (Tischner 2000). Despite that no clearly
defined process has yet been explicitly defined to guide designers through pitfalls in the early stages
of product development. However, general tools and new perspectives on design exist, most of them
based on a life cycle perspective. Some of them are the eco strategy wheel (Brezet and van Hemel
1997, Okala 2012), design checklists and ecodesign (Brezet and van Hemel 1997, Tingström 2007). All
of them offers new and useful perspectives on design for the environment, but none of them offers a
well-defined development process.
Vezzoli et al. (2015) calls for more knowledge within the area of customer satisfaction in sustainable
product service systems (SPSS) to understand more about the design factors that makes costumers
likely to start, and continue to use SPSS. Moreover, if we are about to introduce new products and
habits on the market, it is useful to start with the users. With them in the centre it is possible to
understand the implications of a change, and design a product which encourage sustainable
satisfaction of needs. Therefore, the process used as a foundation in this project was the use centred
product development process which emphasizes the needs of the users and affirms the importance
of these throughout the process. This report will describe how the challenge of fitting ecodesign tools
in the use centred product development process was handled and how the process was tried on a real
case. In the end of the project the new concept was evaluated with life cycle assessment to assess the
potential in the process, the methodology used and the product concept itself.
1.1 PURPOSE
The purpose of this master thesis is to explore the potential behind the suggested use and
sustainability centred product development process when redesigning an existing product. The
potential is decided based on obtained product improvements, in terms of user benefits and
environmental impact reduction, assessed from a life cycle perspective.
1.2 AIM
The project aims to apply the use and sustainability centred product development process on a given
product system, through a case study. In the case study, an appropriate component, or system of
components, will be selected for redesign. The redesign will be focused on improvements which
promotes user and environmental benefits. In addition, the aim is to evaluate the product concept
with life cycle assessment, to verify the improvements.
1 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
1.3 DESIGN BRIEF
The case, on which the process could be tested was defined as follows: “Explore design adaptations
of the wheel loader, L150, for a circular business model and assess the change in environmental impact
related to the design adaptions, with a life cycle assessment”. The L150 was chosen since it had been
part of a life cycle assessment study before and some of that work was believed to have value in this
project.
The timeframe for the project was 20 weeks and two master students comprised the project group.
The deliverables were an illustrated concept description and two reports; the complete master thesis,
describing the entire project and a complementary report describing the conducted life cycle
assessment.
The only predefined condition for the circular business model was that the wheel loader should
partake in a functional sales business model. However, the intended outcome of the circular economy
business model from an environmental perspective was to lower resource consumption of any
resource except fuel1. Thus, the focus for this project has been in the pillar of environmental
sustainability rather than economic or social. The concept sustainability will represent environmental
sustainability throughout the report.
The case was given, and made possible, by two parts: Mistra REES and Volvo Group. Mistra REES
(Resource-Efficient and Effective Solutions) is a 4-year program run by a consortium of Swedish
companies, universities and social actors with the vision to hasten the transition towards circular
economy. The program aims to make a comprehensive study of circular economy to create a
knowledge base that can be useful when resource efficient and circular solutions are being developed
(MistraREES 2015). This thesis plays a part of the Mistra REES program and will contribute with
knowledge about how the environmental impact is affected when a studied product is designed to fit
in a function sales business model. At the end of the program the Mistra REES group is hoping to have
created an information exchange between industry and academia; feed relevant knowledge into the
process of designing for resource efficient business models and understand more about in which
sectors the gain from a transition is bigger, for the company and the environment.
Volvo Group, with the divisions: Buses, Trucks, PENTA and Construction Equipment, is one of the
industrial partners in the Mistra REES project. This thesis project was carried out in collaboration with
Volvo Construction Equipment (Volvo CE or VCE) located in Eskilstuna, Sweden. Volvo CE is developing
off-road machinery, which includes everything from dumper trucks and wheel loaders to heavy road
work equipment. Perhaps the most versatile of the products in the Volvo CE portfolio are their
different wheel loaders. The design work at Volvo CE is guided by the corporate core values: Safety,
Environmental sustainability and Equipment quality (Volvo 2015).
1.4 REPORT STRUCTURE
The thesis starts with an introduction (Chapter 1), including a background to the project, the purpose
and aim, followed by the definition of the project brief. The theoretical framework is presented in
Chapter 2 and 3; where the first chapter introduces circular economy and the later presents the
development framework, which will then be utilized in the case study. The case study, which is
constituting the main body of the thesis is presented in Chapter 4 through to 9. The case study contains
the overall Process (Chapter 4), and three stages: Prestudy (Chapter 5), Concept development (Chapter
6-8) and Evaluation (Chapter 9). The main body of the thesis, the case study, is then followed up by a
1 A lot of the company’s environmental work so far has been mitigation of emissions from fuel and this project was a way to
focus on new areas.
2 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
2 AN INTRODUCTION TO CIRCULAR ECONOMY
As was mentioned in the introduction, circular economy is the suggested way of closing the flows of
materials promoted by Ellen MacArthur foundation (Ellen MacArthur Foundation 2015). The
foundation describes how the society of today is involved in a linear economy where goods are
bought, used and disposed. Typically, purchasing decisions spring from different sources including
new or altered needs, new models being released, a worn or broken product, or simply a hunger for
a new item. The replaced products often end up in quite non-constructive end-of-life processes, such
as landfills or in incineration. Ellen MacArthur foundation points out that nothing is disposed in nature;
what is considered waste by one species is valuable matter for another. This is the idea the chemist
Michael Braungart and the architect Bill McDonough used when they developed the concept cradle
to cradle (C2C). C2C utilizes the
same idea, but transfer it to
materials in industrial processes
which they argue, should be
looked upon as nutrients, either
biological or technical. The
fundamental rules governing their
idea are that: waste equals food,
processes should be powered with
renewable energy and human and
natural systems should be
respected by celebrating diversity.
How the concept of cradle to
cradle is used in circular economy
principles is developed in Ellen
MacArthur Foundation (2015) and
Figure 1: Circular economy system diagram, showing examples of
illustrated in Figure 1, the system
how products can be used longer by actions in the use phase by
diagram of circular economy. both user and producer. (EllenMacArthurFoundation 2015)
It is crucial in circular economy that materials circulate, in one biological and one technical circle, to
become new products. Another principle in circular economy is that resource yields should be
optimized, meaning that the use of extracted materials should be maximized before recycling. Figure
1, the circular economy system diagram, suggests ways of doing so. A rule of thumb in the system
diagram is that products should be used in the inner circles for as long as possible, to reduce the
amount of resources required per use (Ellen MacArthur Foundation 2015). Since the product of
concern in this project was mainly consisting of technical materials, principles on the right side of the
diagram were applied.
The linear business model has made our society expert at delivering products and services to
customers. Supply chains are in place and materials are efficiently shipped, one way, from extraction
to the user and further to disposed. In a circular economy, reversed cycles are suggested; the flow of
materials needs to be expanded with a take back system simultaneously with that we start looking at
disposed products as raw material for new products. However, more knowledge is needed on how to
get the materials back into the factories in the most useful and efficient way (Ellen MacArthur
Foundation 2015).
A product, machine or component often contains multiple parts and even if the life is over for the
entity the different parts might still have lots of potential left. One suggested way not to lose this
4 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
potential is to reuse parts, hence saving material and energy. Another way is to repurpose parts, which
means that a component changes contexts after its first use and serve another, or similar, purpose in
another context. Ideally, the entire functionality of the product is still utilized in the new context;
otherwise there is a risk that the product’s value and material is not fully utilized. If no other option is
available, or if there is no way the product can any longer meet the user needs, the product should be
recycled. Recycling includes the in going materials being collected, treated and redistributed to
manufactures of new products. According to the principles of circular economy, the recycling process
must strive to maintain as much of the material value as possible; i.e. a minimum amount of resources
should be required to retain the same material value as the initial product (Ellen MacArthur
Foundation 2015, Bocken et al. 2016).
There are multiple ways for users and producers to keep or bring back the initial value of products. As
mentioned earlier many of them are included in the circular economy diagram (Figure 1). One way for
companies to do so, is through remanufacturing. Remanufacturing is a process where products are
taken back from the market by the company instead of being disposed, and through a series of
processes, the products are brought back to as-good-as new performance and sold again. The
processes needed may involve cleaning, repairing, upgrading and exchange of worn out parts. The
companies sometimes give the same warranty on remanufactured products as for new products
(Sundin 2004). Remanufacturing is thought to be more environmentally friendly than new production,
since up to 85 % of the weight of remanufactured products comes from used components and the
process only require 20-50 % of the energy compared to conventional production. In addition, 20-80
% of costs can be saved (Sundin 2011).
Another opportunity for a company to utilize the materials better in their products is by providing
their products as services in product service systems (PSS). A (PSS) is described by Baines et al. (2007)
as: “... an integrated product and service offering that delivers value in use. A PSS offers the opportunity
to decouple economic success from material consumption and hence reduce the environmental impact
of economic activity.” A PSS can range from selling access to the machine all the way to only sell the
results of the use of the product (Tukker 2015). According to Baines et al. (2007) and Tukker (2015),
the environmental benefit with the concept product service system is that a PSS, per definition, offers
the opportunity to decouple economic success from material consumption. In addition, PSS facilitate
product longevity activities and waste management.
The adaption of the wheel loader to circular economy in this project has been built around the concept
of functional sales; which, in this project is defined as the company selling operable machine hours.
Functional sales is based in theory about use-oriented PSS (Baines et al. 2007, Tukker 2015), since it
focuses on selling the use of the product. The principle is tied to circular economy primarily in how it
is aiming to reduce resource use. In the wheel loader case, the idea with functional sales is that the
company receive payment per hour of operation, fuel and operator excluded. The company is
responsible for keeping the machine in operable condition, including repairs, service and
maintenance etc. Ideally, this has two positive implications for the environment. First, the company
wants to maximize the income from one machine by keeping it on the market with high hours of
operation per year, as long as possible (product utilization and durability). And since maintenance,
which today is revenue, will turn into a cost for the company, the second implication would be
optimized maintenance schemes; fewer occasions and less resource use, to maximize profit. However,
a functional sales model is not per definition more sustainable than a linear business model (Tukker
2015). Therefore, this project has focused on design for closing loops and design for less material use
in addition to the functional sales model.
5 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
3 A USE AND SUSTAINABILITY CENTRED PRODUCT
DEVELOPMENT PROCESS
As stated in the introduction, the purpose of this project is: ”to explore the potential behind a use and
sustainability centred product development process” and as mentioned in the introduction; no
univocal such process seems to exist today. Instead there is a wide range of design for environment
tools available, but the task of sorting out which of them are suitable for a certain problem
formulation, is often perceived as difficult (Knight and Jenkins 2009). This fact states a lack of working
strategies or defined development processes when designing for the environment. Therefore, a
process was compiled which could give recommendations about the way forward in every step of the
product development process. The user centred product development process described in seven
steps by (Bligård 2015, Wallgren 2016) is used as the foundation for the framework. This process is
selected since it promotes a holistic view and focus on the use and function rather than the tangible
product itself; which according to Tingström (2007) are essential aspects also for sustainability centred
product design. The seven steps in the process are:
1. Identification of target group and available users within the group who can contribute with
valuable information.
2. Collect information.
3. Analyse information to reach understanding about the user’s situation and needs.
4. Ideation with the goal to meet user needs through a new product, system or service design.
5. Requirements formulation as a concurrent process together with ideation to remember the
user’s situation. The requirements are gonging to transform throughout the development
process and translate from user quotes to technical requirements that describe the solution.
6. Concept selection where the concept which best meet user needs is chosen.
7. Validation of concept. Where the concept is tested against user needs and requirements.
In each of the seven steps, the environmental sustainability aspect is acknowledged by emphasizing
the concept of ecodesign, to form the use and sustainability centred process. Ecodesign is described
by NRC (2003) as a strategy to systematically incorporate environmental consideration in the process
of product and process design. However, since the concept has been developed, expanded and used
for different applications it is now rather seen as a “way of thinking and analysing” rather than “…a
specific method or tool” (Lindahl 2006).
As discussed in Chapter 2, circular economy suggests multiple ways of closing material loops and
finding ways to more sustainable consumption. In order to ensure that the result will lead to resource
savings, as stipulated by the project scope, the use and sustainability centred product development
process is going to be the core of the project and, while working with the process, resource efficiency
will to be the guiding strategy. The process is split up in three parts: prestudy, product development
and evaluation where the prestudy is aiming to narrow down a problem statement to something
manageable for the product development process to work with. The product development process
aims to create a product with less environmental impact and the evaluation aims to assess if less
impact was achieved.
3.1 PRESTUDY
To guide the prestudy towards not only a manageable product to design, but also a product with
potentials in resource savings has Holmberg´s (1998) way of somewhat defining the problems, been
helpful. He states that: “In order for a society to be sustainable, nature’s functions and diversity must
not be to systematically subjected to…
6 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
… increasing concentrations of substances extracted from the earth’s crust.
…increasing concentrations of substances produced by society.”
With those two recommendations as guides, the choice of methods for narrowing the project scope
is helped to be tightly tied to material savings. However, to increase the efficiency of any change,
Holmberg stresses the importance of being aware about on which level of the system a change is most
efficiently implemented from a resource saving point of view. He suggests seven levels where a
substitution can be done, here presented in increasing order of efficiency: the raw material level, the
material level, the component level, the sub-system level, the system level, the strategic level or the
value level (Holmberg 1998). Changes on all levels can be positive, although, more radical changes
occur higher up in the system; hence, are potentially being more beneficial for the environment.
A system focus is also suggested by (Bligård 2015) in his description of the product development
process. His focus is not primary environmental sustainability, but he concludes that a system focus
during the development work facilitates the development of relevant requirements and criteria. It is
chosen here to expand his view to include environmental sustainability and the importance of a
system perspective is valid also in ecodesign to address environmental issues concerning the product.
Such an expansion of the concept is promoted by several authors (Holmberg 1998, Tingström 2007,
Ellen MacArthur Foundation 2015, Bocken et al. 2016).
By defining the system, the product is part of, both on a high level and on a zoomed in level, it can be
ensured that most aspects and needs of the system are acknowledged and taken into account.
However, if the system the product will be part of does not yet exist, a system model might be
challenging to develop, which could have the consequences that not the right problems are addressed.
The suggested way for the use and sustainability centred product development process to find
relevant problem formulations is to use the backcasting methodology outlined by Robinson (1982). It
should preferably start by describing the current situation through a system model; this will provide
useful knowledge about existing products and systems, the system model should then function as the
baseline scenario needed in the first step of the backcasting process. Secondly, a future scenario
should be visualized, where the most desired future context and situation for the (future) product are
described. The fundamental idea of backcasting, also used here, is to then work out ways to achieve
the desired situation by starting from the future scenario and work backwards to connect the two
scenarios; this reverse approach is used in order not to be locked-in to already existing solutions and
systems. In the use and sustainability centred product development process the connections between
the two scenarios could be considered key areas for development work. If the topic of each key area
is compared to what resources and competences are available for a certain project, a suitable problem
formulation or design opportunity can be identified.
A simplified system analysis described in Bligård (2015) can be used for developing the system model
needed in the first step of the backcasting process. Information can be obtained by literature scanning,
focusing on the overall strategies, structures and goals of the company which the product is produced
by. The system model could be further developed with observation studies, interviews, literature
studies and workshops. To gain more information about conducted environmental work within the
company and the current strategy, the RDAP-scale, by Clarkson (1995), could be used. Depending on
the nature of the action, the RDAP-scale sorts sustainability activities according to the categories:
reactive, defensive, accommodative and pro-active. It is recommended to also use any existing life
cycle assessments of existing products of similar characteristics to address the life cycle view and
support the environmental problem identification in this stage. The end goal of developing a system
model is thereby threefold: using it in the earlier described backcasting process, to help highlight
7 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
connections and try to identify the seven levels where substitution could be done and define which
level is most impactful and feasible for the actual project.
When visualizing the future situation, an environmental objective should be decided to help guide the
development in the desirable direction. Ideally, the objective is outlined with support in theory and is
aiming guide the development of the scenario and to work as a mindset throughout the project. When
the future situation is visualized as a scenario, it fulfils two purposes; first, to be part of the backcasting
process and secondly to be central in the product development process. The scenario will be the
expected reality for which the product is going to be designed; hence, it must be well enough defined
to serve the purpose as a system description. Such description will include a reasonable estimation of
the context, user(s), task(s) and product(s); although, since part of the objective is to redesign some
of these parameters, the scenario must be developed concurrently with the improvements as new
information arise, and design parameters are set. The idea is supported by Holt (1989) and Carroll
(2000), who both agrees on that a scenario helps a design team in their process. Erickson (1995),
Carroll (2000) and BØDKER (2000) also state that a scenario facilitates discussions within the team,
but also in the communication with stakeholders.
The developed environmental objective is suggested to be extended beyond the prestudy. Aspects
learned could be formulated as guiding criteria for the decision of a suitable problem formulation.
Furthermore, the criteria could also be used in the following product development process, to make
sure the objective is acknowledged in every step of the product development.
3.2 PRODUCT DEVELOPMENT
As emphasized in previous sections; an important part in understanding the user needs is to examine
the system. However, as the system only exists as a future scenario, much of the system description
is based on the current situation; users, tasks, contexts and products; although, some changes in
regards to these system components in a future circular scenario was done in the backcasting process.
In accordance with the statements from Gould (1995), Margolin (1997) and Preece (2002): the
developers need to know the target group they are designing for if the developed product is going to
be successful. The use and sustainability centred product development process has added a
requirement for successful products; low environmental impact. The target group for this success
could be described as future users2 since they do not yet exist. However, from revisiting the user
centred product development process in the beginning of the chapter, it is emphasized in step number
1 that users within the important target groups, who can contribute with information should be picked
out. To contribute information is for obvious reasons impossible for the future users. However, they
need to be taken into consideration, and the resource management act (New Zealand Parliament
1991) offers a version of what could function as their requirement on how a product should be
produced by saying that it should be done in a way so it is: “sustaining the potential of natural and
physical resources (excluding minerals) to meet the reasonably foreseeable needs of future
generations” (Part 2, purpose and principles, RMA). Since no one from the use group future
generations can be interviewed for obvious reasons, instead they should be treated as silent
stakeholders (Sharma and Starik 2004). The same goes for every part of the ecosystem that cannot
communicate in a way sufficient for a product development process. Their stakes still must be
accounted for as discussed by the same authors. A sustainability centred product development
process should have the environment and future generations as part of the core, since these will have
to take the future consequences of material extraction and dispose. The strategy when moving
2 If the whole product life cycle is considered the future generation will be a user of the recycled materials in the product as
well as the quarry where the metal is extracted from.
8 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
forward in this phase could be to pay attention to resource efficient solutions and be aware of all users
and stakeholders in every part of the product life cycle. A concrete way of achieving this can be to use
any kind of environmental impact screening method or tool.
Including the idea of resource efficiency in the product development process could probably be done
in more than one way. For this process a product life cycle perspective is suggested for all stages of
the use and sustainability centred product development process. The definition of the product´s life
cycle by Sundin (2004), includes: raw material extraction, production, use and end of life can be used
for the development process. Also in cases where the most resource efficient solutions can be
obtained through a product service system, the life cycle perspective is a valuable contribution. Sundin
(2004) continues by saying: “Having a life-cycle perspective on combined services and goods means
that lifecycle considerations must be considered for both physical products used in the PSS and the
services used during and between the contract times”. This stresses the importance of that all impact,
from every part of the product’s life cycle, should be accounted for, in every stage of the product
development process described in this chapter.
A life cycle perspective throughout the processes could make the way of analyzing and ideating slightly
different in the use and sustainability centred product development process compared to the
traditional process. The extended user group and the extended view on the product life cycle results
in product requirements not only based on use and functionality, but are also extended to include
manufacturing (process and material), distribution and recycling. To assist in the altered analysis and
ideation processes, the following tools are suggested to be used as a complement to the common user
centred methods and tools, to ensure the development process arrive at a solution with less
environmental impact.
ECO STRATEGY WHEEL AND ECODESIGN CHECKLIST
The ecodesign checklist and eco strategy wheel, introduced by Brezet and van Hemel (1997) are life
cycle based tools. The checklist consists of a number of different questions to be answered by the
designer, addressing the environmental impact from the product during its different life cycle stages;
whereas the wheel suggests aspects and new perspectives for the product system to consider.
TEN GOLDEN RULES
The ten golden rules of ecodesign presented in Luttropp et al. (2006) can be used to ensure that
environmental aspects are considered:
1. Don’t use toxic substances and arrange closed loops for necessary but toxic ones.
2. Minimize energy and resource consumption in production and transport through
HOUSEKEEPING
3. MINIMISE energy and resource consumption in the usage phase, especially for products with
most significant environmental aspects in the usage phase.
4. Promote repair and upgrading, especially for SYSTEM dependent products.
5. Promote LONG LIFE, especially for products with most significant environmental aspects OUT
of usage phase
6. Use structural features and high quality materials to minimize WEIGHT not interfering with
necessary flexibility, impact strength or functional priorities
7. Use better materials, surface treatments or structural arrangements to PROTECT products for
dirt, corrosion and wear
8. PREARRANGE upgrading, repair and recycling through access ability, labelling, modules,
breaking points, manuals
9 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
9. Promote upgrading, repair and recycling by using few, SIMPLE, recycled, not blended
materials and no alloys
10. Use as FEW joining elements as possible and use screws, adhesives, welding, snap fits,
geometric locking etc. according to the life cycle scenario.
DESIGN FOR RECYCLING CHECKLIST
In addition, a recycling company has established a guide of design guidelines for design engineers,
which are supposed to favour recyclability (Domini 2001). The principles suggested for this process
are summarized as in the following list.
Design for recycling principles:
- Reduce the number of different types of materials
- Use materials for which recycling is possible, and for which there is a demand for the recycled
material.
- Different materials must be separable, preferably through disassembly. Otherwise through
material fragmentation and material separation.
- Mark parts with material type.
Together, the methods create an environmental foundation for the analysis and generation of ideas
in the different stages of the development process. However, to make sure that the product lives up
to the requirements, continuous evaluations should be carried out during the process (Johannesson,
Persson et al. 2004, Bligård 2015, Wikberg Nilsson, Ericson et al. 2015). When an evaluation is made
and potentials for improvements are identified, ideally, designers should take a step back in the
process and look for alternative solutions; the process is iterative with continuous evaluations and
design improvements. A suggested way for continuous evaluation in the use and sustainability centred
product development process is evaluation against a set of criteria; similar to what was mentioned in
the framework described by Bligård and mentioned previously, when defining the prestudy process.
3.3 EVALUATION AND VALIDATION
To make sure that the end result lives up to the requirements and to assess the environmental
sustainability, the following tools and methods are suggested to emphasize the sustainability focus
and complement the more commonly used products development evaluation methods.
RAPID ECO ASSESSMENT
The eco strategy wheel (Brezet and van Hemel 1997) could be utilized as a rapid eco assessment tool
as a qualitative complement to the quantitative LCA. It has similar characteristic as a checklist and it
is quicker, less comprehensive, and partly covers other aspects than an LCA. Thus, it can be utilized
multiple times during the process and in general earlier than quantitative analysis. In addition, rapid
eco assessment is suggested as a good way of evaluating a reference product to identify existing
problems for the product development process to resolve.
LIFE CYCLE ASSESSMENT (LCA)
Life cycle assessment is a way to describe what resources a product uses and what pollutions it causes
during its life cycle. By deciding on which processes are part of the product lifecycle, and by modelling
all in and outflows over the system boundary, the environmental impact of a product can be
quantified. Different LCA’s can have different approaches and the focus is decided by the functional
unit, which is directing what is included in the LCA. LCA is a very quantitative and ambitious tool and
might be more suitable for the later stages of the product development process, when specific product
10 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
attributes such as material is decided. Ideally, the LCA can be conducted concurrently with the later
stages of the product development process to also provide new insights.
USE AND USER EVALUATION
To ensure that the use and user aspect of the product development process are sufficiently
acknowledged, user tests should be carried out. Verifying sufficiency in this area is important for the
use and sustainability centred development process primarily in regards to two aspects. First, a
product which is easy and effective to use could provide better preconditions for utilization. Secondly,
a truly useful product is likely to promote product longevity effects. A design process with a use
perspective should evaluate the use to verify improvements and for this, the following methods can
be utilized.
Concept user evaluation, as described in Wallgren (2016), is a test for evaluating an early stage concept
or, in later stage, a product prototype. It can be done by letting the user have opinions about the
design and perceived functionality of a concept or let the user try a prototype. A prototype can be
tested on site or in a laboratory by letting the user solve tasks with help of the product representation.
The test is usually observed and can be supported by a, more or less, structured interview.
11 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
4.1.2 SCENARIO
The product concept expected to come out of the concept development phase was required to be
better for the environment and be part of a functional sales business model. To be able to design an
appropriate solution for a future reality, a scenario was needed in which all the affecting factors from
a functional sales model were included. The process for arriving at the scenario started with a
brainstorming session where concepts of the future wheel loader product system were developed and
concretized as various functionality concepts; including short product descriptions, accompanied with
brief descriptions of the system it would perform in. The functionality concepts were placed on a
timeline as an attempt to sort them by how far off into the future the realization of the concept could
be expected. With concrete concepts for alternative functionality focused product systems, it was
easier to decide on the aim of the case study in terms of compatibility with current technologies,
concept readiness and other factors that together defined a reasonable time perspective. The timeline
was used to communicate these aspects with the stakeholders, with the aim to agree on an
appropriate level for the particular project. The scenario was used as the vision, or future scenario, in
the backcasting process. When the work with outlining and concretizing the scenario continued three
considerations guided the process:
- In order to benefit the company the most, a scenario that is not too far off in the future should
be used as a base for the design process.
- The product concept should ideally be able to serve a purpose in today’s context as well as in
the future scenario.
- What is considered most valuable is a case study showing the potential of implementing a use
and sustainability centred development process on a system with today’s preconditions,
rather than a utopic science fiction scenario, where everything is different.
The development of the scenario continued over the whole prestudy process, with new facts
continuously refining the scenario definition, all the time guided by the environmental objective:
reduce resource consumption and close loops. In the following concept development process, the
scenario was concurrently evolved with the product design, since the scenario, and the product
system, including user, use, product and context were shown to be highly dependent components.
4.1.3 KEY AREAS
In a parallel process with the scenario development, interviews and literature studies were carried out
to find out how different eco strategy wheel principles were used in the product development process
in the company. The goal was to find areas where the environmental work could be developed. The
results from the interviews were therefore compiled to areas with a Kawakita Jiro -analysis (KJ-
analysis) and resulted in ten key areas. The key areas were identified in the analysis as important areas
to address in a transition to a circular business model. The eco strategy principles aim to cover every
change that can be made at all levels of Holmberg’s (1998) levels of substitution. And since the
interviews were based on the eco strategy principles, it resulted in that the key areas were addressed
on different system levels.
4.1.4 CHOICE OF COMPONENT FOR DEVELOPMENT
To stay focused on what was important to consider in the project, a set of criteria was developed to
help narrow the project (left side of figure 2), simultaneously with the scenario development and the
identification of key areas. Some of the criteria followed naturally from the choice of business model,
while others developed during the prestudy in dialogue with different stakeholders in the project. The
stakeholders were allowed to communicate their prioritizing through a common weighting of the
evaluation criteria. Ultimately, the criteria were used to decide about which key area to further
15 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
develop. The decision was made through a weighted evaluation, similar to a concept selection method
described in Wikberg Nilsson et al. (2015). Points were given to each key area, based on the estimated
score for the respective evaluation criteria.
The key area which pointed out as having most potential, repair and maintenance, was analysed
further; the interview results from the area were organized and expressed as a number of problems.
As a complement to the interviews, an LCA, conducted on a similar wheel loader was consulted.
Information about what activities had an impact in this key area was considered. The LCA offered a
life cycle perspective in the aspect of resource consumption and emissions. These problems went
through a similar evaluation process, using refined evaluation criteria, with the purpose to establish
each problem's relative potential to be sufficiently solved during the project. To narrow down the area
further, towards a tangible reference component, or system of components, the existing LCA and the
supervisors were consulted; which in turn led to that an appropriate component could be selected for
the following concept development. The selected component was the main fuel filter.
4.2 CONCEPT DEVELOPMENT
The primary goals of the second main process, the concept development, were; first, to define the
specific system and identify relevant needs and requirements for the concept development, and also
to further refine the evaluation criteria to suit the concept evaluation. Thereafter, to develop
conceptual design solutions for alternative functionality through a use and sustainability focused and
iterative design process. The iterative process also includes continuous evaluation against the
previously defined criteria, in order to ensure the most appropriate concept is select for further
refinement in the, final concept refinement stage. As can be seen in Figure 2, the concept
development process was divided into the three parts: Basic needs and requirements study,
functionality concept development and final concept refinement.
4.2.1 BASIC NEEDS AND REQUIREMENTS STUDY
The first part of the product development stage included a system analysis and a complementary
problem analysis (see figure 2). With the base in the identified system and the problems, the outcome
of the needs and requirements stage were initial needs and requirements and evaluation criteria for
concept assessment in later stages. These outcomes were the foundation for the following concept
development.
4.2.1.1 SYSTEM ANALYSIS
For the conceptual solutions to consider a significant part of relevant influencing factors, a holistic
approach, obtained by system analysis was used. The system analysis considered the individual
contributions from the four system components: product/artefact, task, user and the surrounding
context; and described how the aspects influence each other and overall system performance (Bligård
2015). In the particular project, the system analysis again served as a baseline; defining a reference
case which allows for the comparative evaluation in the end. The holistic system and life cycle
perspective was important for the environmental impact screenings, since it allowed for connections
between impact, system components and the different life cycle stages. The system analysis aimed to
complement the system overview carried out in the prestudy; however, focusing on the selected
reference product.
To obtain a general understanding of the system before moving into depth, a system overview was
constructed. Conducted interviews seeking to find evidence of current problems helped refining the
overview. The system boundaries helped ensure the problem was limited to a manageable level to
keep solutions relevant for the system. Key factors, similar to what is described as “external factors”
16 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
for sustained system performance in Bligård (2015) were identified and described for the respective
sub-system.
ARTEFACT
The reference artefact was described in terms of its design and functions based on information from
the company and manufacturer. However, the most important contribution to the understanding of
the current artefact design came from a dissection of one new and several used filters of different
models. To collect the information about the reference product, a table of the component content
was made, and a representative CAD model was constructed to facilitate communication and
development of graphics to explain new concepts. In addition, the collected information was used to
conduct the LCA of the reference filter.
In order to identify and assess environmental impact from the product and to identify possible actions
for improvement, the product’s life was investigated and described from raw material production to
disposal and evaluated with rapid eco assessment, as the tool is described in the development
framework (Chapter 3.2). The life cycle overview was also aiming to support the product development,
since it indicated where changes could be most efficiently implemented, and where the suggested
improvements had potential to optimise the positive impact.
The functionality of the reference product was determined from interviews with in-house product
engineers, from manufacturer information and from conducting market research of various
corresponding solutions. The artefact functionality was structured with the help of a functions
analysis; defining the reference object's primary, sub- and support functions according to the theories
described by Österlin (2010). Procedures were assumed based on information from the company and
complemented with data from similar manufacturing, presented in online video clips and in
manufacturing techniques literature. The artefact could not be observed in use in this study due to
project constraints.
TASK
Compared to the description of the user and the context, the task description is to a larger extent
based on the service instructions; i.e. how the operations are supposed to work according to the
company manuals. This data was then, if possible, verified and complemented with data from the
interviews.
The task was specified with help from a hierarchical task analysis as presented by Bohgard (2011). The
task was based on the entire service scenario, to cover all the relevant occasions where the product
was handled by a person. Another reason to why it was deemed important to cover the entire service
operation was to highlight other relevant product requirements and possible design solutions
incorporating combinations of needs.
USER
Several complementary methods were used to describe the identified user of the chosen reference
product. First, the different user types were addressed and categorized in terms of their influence over
and the contact with the product; primary user, secondary user, side user and co-user were
established, according to the theories and definitions presented in Janhager (2005). A simplified
stakeholder analysis was performed in which the motives and the product design influence from the
most important users were outlined.
The primary user was categorized according to the level of experience and skill described in Janhager’s
theories (2005). The primary user was further elaborated through a persona and a complementary
17 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
scenario. The combination of the methods served as important tools in communicating the use and
user profile.
The persona was constructed from the template referred to as “user character” in Janhager (2005)
and the format for the scenario was also picked up from the same reference. The user analysis was
based on data obtained through interviews and observations at the service provider and to some
extent on information from within the company.
The reason to the quite extensive user analysis for a product with its main functionality entirely
disconnected from a user, was that the situation was anticipated to change and the use to increase
due to focus on redefining the functionality and extending the product life.
CONTEXT
The context description was also constructed from descriptions of the surroundings, communicated
through interviews, literature review and to some extent observations. Another important source of
information in defining and analysing the user and the context were reports describing previous work
carried out in quarry environments from the department of product and production development at
Chalmers (Wernberger Jonsson et al. 2011, Bergstrand et al. 2014). The context description presented
in the system analysis of the machine, in the prestudy, was important also for the selected reference
product and the related service operation. As a complement to the context description, the external
factors which were deemed relevant for each sub-system were addressed according to what is
prescribed by Bligård (2015).
4.2.1.2 PROBLEM AND NEEDS IDENTIFICATION
Problems connected to the studied system were compiled in problem identification documents and
categorized according to area of relevance. In addition, aspects in the current system which might
cause problems in a future functional sales business model were addressed. At this stage the learnings
from the rapid eco assessment were used and complemented with a recyclability screening.
Needs and requirements engineering, described in Bligård (2015), was used to define relevant needs
and requirements in the areas: functionality and use requirements. In addition to these traditional
requirements, the silent stakeholder, the environment, was also considered through the requirements
category: sustainability. Thereto, a set of desired concept properties were defined; i.e. aspects which
were considered positive if the concept could fulfil. The purpose of the requirements list was to focus
the solution generation process on only relevant aspects for the particular system and the case study
scope; without being very explicitly defined. The requirements were then continuously updated and
more explicitly defined when the level of abstraction was further narrowed. To communicate that the
level of abstraction had been narrowed down further, the overall process effect goal was redefined.
Criteria for the evaluation of the concept solutions were considered necessary for a structured
selection process with many, and to some extent contradictory, demands. The criteria had to be a mix
of objective and subjective criteria in order to cover all the relevant aspects. The criteria used in this
stage evolved from the criteria used in the prestudy; though, compared to the previous criteria, they
could now be more explicitly defined due to the more explicitly defined problem. This fact also made
the evaluation more accurate. The old criteria from the prestudy, had an important role by
acknowledging the different project stakeholders’ priorities in a new weighting process. The expanded
set of criteria was weighted by a pairwise criteria comparison, inspired by Johannesson et al. (2004).
18 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
4.2.2 FUNCTIONALITY CONCEPT DEVELOPMENT
4.2.2.1 INITIAL CONCEPT DEVELOPMENT
The purpose of the first concept development was to generate several conceptual designs for the main
functionality. The idea of producing several concepts was to maintain a broad solution spectrum. In
this first development stage, the concepts were described quite inexplicitly and in terms of their
functions rather than product attributes. This was because it was believed that a change in system or
functionality level would have a much bigger positive environmental impact than product design
changes alone; an assumption based on the levels of substitution presented in (Holmberg 1998).
The ideation process described in the suggested sustainability centred design process (Chapter 3.2),
was used to explore how the effect goal: “Reduce environmental impact from main fuel filter”, could
be attacked. The different strategies presented in the ecodesign ideation methodology, compiled by
Okala (2012), were assessed according to their appropriateness for the particular problem. The most
relevant strategies were combined, resulting in the four solution categories; "reduce basic need",
"extend useful life", "reduce impact from materials" and "reduce indirect use of resources". The
different categories were used for ideation of conceptual solutions together with the needs and
requirements list with the goal to achieve the effect goal. Mind maps were used to develop solutions
to the identified problems. The technical solutions and the solution paths were combined to form
eight independent, self-supporting concepts.
It was considered important to clearly separate the concepts in order to make the evaluation explicit.
For the same reason, the aim was also to present the concepts on the same level of detail. The
separation was obtained by focusing the descriptions on the differences within the concept range and
by stressing the unique features of each individual concept. To facilitate communication, the concept
description and analysis were complemented with a concept illustration and a numeric, weighted
evaluation for each of the eight concepts.
The criteria which were defined and weighted in the previous needs and requirements study were
used in a concept selection matrix, as described in Wikberg Nilsson et al. (2015), to assess the eight
concepts. In addition to evaluating the concepts against each other, a comparison against the
reference with a Pugh’s comparison matrix (Wikberg Nilsson et al. 2015) was conducted. However,
since Pugh’s evaluation method is different from the evaluation matrix, the criteria had to be revised
and the criteria were weighted slightly differently. The reason to the alterations is connected to the
fact that Pugh’s evaluation method compares concepts to the reference; hence factors such as
technological feasibility, company compliance and completion level were obviously not considered
relevant for comparison against an already existing solution.
4.2.2.2 SECOND CONCEPT DEVELOPMENT
Three concepts were selected for further refinement, however all concepts served the process with
valuable learning outcomes and sub-solutions which were brought to the next development stage. In
this stage of the process, it was deemed necessary to also include technical solutions in order to assess
feasibility; although, still concentrated to the functionality and the principles of the technical solutions
rather than tangible product concept properties.
After the refinement process, the three concepts were described in detail. Additional research
regarding each of them was outlined and the tasks were described in more detail. In addition, a SWOT
analysis (Johannesson et al. 2004) was carried out; in which the assumed strengths, weaknesses,
opportunities and potential threats, were identified and addressed. In addition, remaining problems
and uncertainties were outlined, as important complements to the descriptions, in order to
19 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
5 PRESTUDY
In the following chapter, the result from the prestudy is presented. From the initial system to an
unambiguously defined problem formulation for the use and sustainability centred process to work
with in the following chapters.
5.1 WHEEL LOADERS
Rather than being specialized for a certain application the wheel loader, seen in Figure 3, is a very
flexible machine, used in a wide range of different environments and purposes. Wheel loaders come
in many sizes and with an array of customization equipment and attachments, to allow versatility and
fit each individual customer’s needs.
Figure 3: Wheel loader L150, schematic image (Prosis 2016)
The operational weight of Volvo construction equipment’s wheel loaders (compact wheel loaders
excluded) varies between 11 tons for the smallest L60 model, up to 50 tons for the largest L350 model.
The initial price varies greatly with customizations and special equipment; general estimations
however, suggest a basic price somewhere around 100 SEK per kilogram. A wide range of special
equipment and attachments are offered by Volvo construction equipment, while others are provided
by the dealers’ suppliers and mounted on the new machine according to the customers’ requests. By
default, the machine comes with a one year producer’s warranty, which sometimes can be extended
by the dealer. When new, the machine is most frequently sold with additional service arrangements
(Interviewee 3 2016). The service arrangements include preventive maintenance according to
intervals prescribed in the service manual. An overview of the estimated general target groups can be
viewed in Figure 4.
23 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
5.2 SIMPLIFIED SYSTEM DESCRIPTION
Figure 6 shows a simplified system model of L150. The successfulness of the machine’s daily
operations is decided by a combination of the system components, represented on different levels in
the model. The model’s inner circle illustrates the fact that the wheel loader is actually a set of
products that forms an entity, more valuable than the individual parts alone. The machine is in turn
affected by other products and information in the context, captured by the second circle. The model
also shows that there are external factors affecting the system performance on the different levels.
These factors can be governed by for example business plan, policies, the market etc. and are not
direct relevant to the object; though, will still have a major impact on the system performance and
the machine design. The system components which were identified as extra important for the system
performance and the result of this project are described more in detail in the following sections.
Figure 6: Simplified System Model for L150, as used in a quarry. The inner circle contains the systems
of components needed for the machine to function. The bigger circle contains factors important for
the operation of the wheel loader in the quarry. Outside the circles are factors which affects the
wheel loader, but are not specific for the machine or the context. Boxes on the borders of the circles
indicates the factors can be influential in both systems.
5.2.1 VOLVO CONSTRUCTION EQUIPMENT (VCE)
The manufacturer of the L150 wheel loader, Volvo construction equipment is part of Volvo Group and
delivers under three brands: Volvo, SDLG and Terex Trucks. The brands compile a broad range of
products including excavators, wheel loaders and other heavy duty machinery. The company employs
about 15 000 people worldwide, with a majority in EU and Asia. The headquarter is located in
Gothenburg.
25 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Since VCE is part of Volvo group and thereby also part of their achievements and investments, the
main focus of environmental work has been to reduce CO and NO . But VCE has also worked to make
2 X
their production sites more energy efficiently and to educate operators to use their equipment more
energy efficient. Volvo CE is also parts of the construction climate challenge (CCC) program, which is
aiming to increase knowledge about how construction equipment can reduce its environmental
impact.
According to Hart (1995), two of the three environmental activities could be classified as pollution
prevention, rather than sustainable development or product stewardship. The CCC endeavour
however, shows signs of the latter.
Table 1: Reactive, Defensive, Accommodative, Proactive-Evaluation of actions. Table shows most
actions carried out by the company are either defensive or accommodative.
Reactive Defensive Accommodative Proactive
Lower CO Constructed an Hybrid and electric. CCC initiative.
2
emissions excavator on diesel Lowering energy CO -neutral site
2
Lower NO with lower emission consumptions at sites. initiative.
X
emissions levels than hybrids. Remanufacturing.
Looking at the Reactive Defensive Accommodative Proactive scale, Table 1, the sustainability activities
are somewhat spread over the entire scale. However, the activities which are applied on the products
in present time are more to the reactive and defensive side. The pro-active and accommodative
technologies are all still in the research phase. The dominating strategy however, is a mix of defensive
and accommodative actions.
The interviews revealed that capital and energy has been focused on reducing emissions from diesel
so intensively that other initiatives have been put aside due to lack of time and resources. Here might
a lot of possibilities to improve the sustainability work be found, in areas such as materials choice,
weight reduction, utilization and more.
5.2.2 PRODUCT SYSTEM: L150
The L150 is among the smaller wheel loaders in the large loader platform category, with an operational
weight of approximately 25 tonnes. Buckets of size 3.4 - 14 m3 can be attached to the loader and the
static tipping load at full turn is almost 16 tonnes. The machine has a 13-liter, 6-cylinder diesel engine,
engineered to live up to the emission standards in the close future. The company is aiming to be a
premium brand and deliver wheel loaders of high quality, with high operator comfort and good
serviceability. Some of the features are a cabin that is tiltable, to facilitate maintenance and service.
In addition, design changes have been made to the machine to achieve better access to frequently
used service points (Equipment. 2015, Interviewee 2 2016). Figure 6 shows some of the ingoing
components of the system and in which level of the system they are found.
The selling of the machines is done through a dealer. The salesman tries to understand the need of
the customer to find a machine that can help satisfying it in the best possible way. A mismatch can
lead to the machine being used incorrectly, which in turn reduces up-time and in the end, gives the
company a reputation of poor quality products (Interviewee 9 2016). The dealer orders the machine
from to the producing company and additional equipment is mounted in the dealer workshop before
the machine is delivered. The extra equipment might come from the producing company; however,
sometimes also from competitors (Interviewee 3 2016).
26 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Due to its size, the L150 rarely comes to a workshop for service; instead, the technician comes to the
machine. Every machine is connected to a data system which tells the technician when a certain
machine requires service. The technician then calls the operator and decide on a suitable date and
time. Prior to the service, the technician loads his car with the required equipment and drive out to
the site where the machine is operating. A build-in GPS in the machine tells the technician where it
can be found. That might be in a static location as a quarry, or at a more dynamic site as a road
construction site. Often the operator is not present when the service takes place, since that conflicts
operating efficiency. An interviewee told that, sometimes a note is left from the operator in the wheel
loader with more instructions to the technician about what needs to be done in addition to the
standard service (Interviewee 3 2016).
A L150 wheel loader require certain service components to be replaced at every 500 hours
(Interviewee 4 2016, Prosis 2016). The components are often disposables and include for instance oils
and filters which are replaced to guarantee that the machine will run for another 500 hours. Over the
lifetime, service and maintenance results in a consumption of materials weighing more than the
machine itself. Most this weight comes from tire replacement, although liquids and filters make an
extensive contribution too. Service and maintenance also generates additional costs, adding to the
cost of operator, fuel and the initial investment. The service parts are an important source of income
for the producing company.
Basically, the technician has two main assignments: planned, preventive maintenance and unplanned
operations (Interviewee 9 2016). For planned service, the technician conducts service according to set
intervals, whereas unplanned operation is when something fails and must be repaired or replaced.
The latter can happen during operating hours and since the wheel loader can be crucial in a production
line, a break down quickly leads to loss of income for the customer (Interviewee 2 2016). The
replacement parts can be bought in three different qualities and price classes. New parts are, as the
name indicates, factory new parts. A remanufactured part on the other hand is a part for which
company has developed a process to take back and remanufacture them to become as-good-as-new,
and the to sell them with a warranty; although as a cheaper option compared to new parts. The
cheapest alternative though, is used parts bought from disassembled machines. In the latter case, the
guarantee is limited (Interviewee 4 2016). To make it possible to have a working supply chain with
reused and remanufactured parts, certain actions need to be done already at the development stage
of the wheel loader. Components needs to be easy to remove and remanufacture, so that the
procedure can be both profitable and doable (Interviewee 7 2016).
Also, there are some aspects to consider in a wheel loader’s end of life phase. First, the dealer and the
company often lose track of the machine already after the first or second change of ownership. In this
way, it is difficult to make the customers continue buying certified company parts, which is lost income
for the company. For the same reason, it is also difficult to make sure that hazardous matter and
components are taken care of in a proper way.
Instead of investing in a new machine, the owner can choose to make a certified rebuild. This is a
relatively new program meaning that parts are refurbished or exchanged, so that the machine can be
brought back to its original condition in terms of performance. This process is quite long today,
including interviews and investigations to be sure about what needs to be replaced. Then, parts are
remanufactured or ordered as new or used. Through certified rebuild, the company manages to build
machines that are as good as new, with a large percentage of already used parts. Today the objectives
for the process is not to save the environment, but to get the machine back in the loop of consuming
service parts. (Interviewee 7 2016). However, it is identified as a good way for the company to use less
virgin materials and the method has good potential for expansion.
27 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
In the end, depending on if a certified rebuild was made or not, the wheel loader is often sold abroad.
Due to this, components and materials are getting completely out of reach for the company and the
machine might end up in a context where it cannot be recycled properly. This in turn means that not
only materials are lost, but also that the wheel loader might contaminate the environment in its end
of life phase.
5.2.3 CONTEXT
The L150 handles mostly virgin materials such as hard virgin bank, shot rock and lose virgin bank.
However, those tasks can be found in a range of environments and contexts as can be seen in Figure
7. For the case of this study, the focus has been on an environment similar to the bottom right image
in Figure 7 (Equipment 2015); a quarry or similar site where the wheel loader handles dirt, gravel or
rocks. This is a very dusty and dirty environment, where machines are exposed to weather and heavy
wear.
Figure 7: Examples of contexts in which the wheel loaders operate, forestry, infrastructure, construction
and quarry
5.2.4 BUSINESS MODEL
Generally, a company buys a wheel loader to be used in the everyday business operations. The
ownership of the machine lies with this customer. As mentioned, it is very common that a service
agreement is signed along with the purchase. A service agreement might look slightly different,
depending on by how much of the responsibility for the machine’s reliability will be on the service
provider. Generally, a service agreement guarantees up-time and gives a possibility for the customers
to pay smaller amounts more frequently for this service. Even though not very common, leasing
agreements exists; these are however signed between the customer and today’s dealer, and thus
means little difference for the equipment manufacturer compared to normal purchase.
When the price of the machine is discussed, it is often discussed in terms of “total cost of ownership”
(TCO) and it aims to help the buyer to compare the machine with other options. TCO includes all the
costs per hour it takes to keep the machine running, including fuel, operator, initial investment, service
28 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
and expected repair and spare parts. Since TCO is calculated per operating hour, it gives the investor
a possibility to compare the cost of the machine with what profit is possible to make hourly; and
thereby assist in determining whether if the machine is a good investment or not (Interviewee 2 2016).
In general, machines which are operating in high performing production lines stay with its first owner
until the cost for maintenance and the level of unreliability reaches a point where the machine is no
longer profitable. At that point, the machine is sold to an application with lower requirements, and
where a new machine is not affordable. The procedure is repeated as long as there is remaining value
in the machine to be extracted.
Three kinds of users were identified in the case of the wheel loader, each with a different connection
to the product. The users identified through interviews were: owner, operator and technician. All
three users are interested in different part of the machine and its performance and at different times.
The owner is obviously the company owning the machine. It is in the owner´s interest that the machine
is reliable.
The operator uses the machine as a tool for performing the tasks in the production. Important factors
for the operator are that the machine has the right characteristics for the job and provides a good
workplace. To facilitate the work, the operator must be able to monitor what goes on around and in
the machine. Information from the surrounding is presented through the windows with support from
the displays in the cabin, whereas machine information is presented on displays and through
interpreting movement and sounds from the machine and the controls. Based on that information,
the operator makes decisions and executes tasks through the controls in close proximity.
The service technician interacts with the wheel loader when performing the service. It is important
that the technician can conduct the service in the different contexts without being constrained by the
machine design. The technician gets information about what service needs to be done from multiple
sources. By monitoring how many hours the machine has been running, the right kind of service can
be provided and the right parts changed. In addition, information from the operator or owner is an
important complement for the technician to learn if the wheel loader shows any abnormalities in its
everyday performance, to help assessing possible issues. Finally, the technician can receive
information through a visual scanning of the machine. The technician interacts with the machine
through the different service points, mainly in or around the engine compartment.
5.3 FUTURE SCENARIO
In Figure 8, the result of the first concept generation is presented in the form of a timeline. The idea
was to have a blue-sky ideation concerning what the future of a wheel loader could look like and then
try to sort the ideas based on how far off in the future the idea could be expected to be (if ever)
realized. The ideas, which are partly a result of a screening of trends and the company’s current
commitments, aim to find a reasonable time span for when the scenario should take place, the
scenario to design for. A time for the scenario was sought, which was not too close, nor too far away
from the present state. The ellipse, to some extent symbolizes an estimation of where in time the
Figure 8: Results from an ideation session, in the form of concepts, presented on a constructed
timeline based on estimated technological readiness and feasibility.
29 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
scenario takes place, in terms of technical and technological reediness based on the ideas that came
out of the ideation.
In this section, a description of the scenario used for this project is presented. As can be seen in the
time line, it is not a scenario representing the ultimate goal of circular economy, but is rather a leap
towards it. It is reasonable to expect that the use of wheel loaders will not change much over the next
10 to 15 years. Partly because a lot of infrastructure projects are planned for which wheel loaders are
used today and partly because it is a machine with a lot of different applications; hence quite hard to
replace. Infrastructure projects require material from quarries which too is a context where wheel
loaders successfully perform today. Both contexts accounted for in this project are thereby thought
to be running in the next ten to fifteen years; even though some changes are of course expected to
occur.
5.3.1 PRODUCT: L 150
The average life length of L150 before major rebuilds is approximately 20 000 hours, which
corresponds to approximately 10 years if the machine runs 8 hours a day. This means that it is not
reasonable to believe that VCE will rely on a wheel loader that is completely different from today’s
machine within next decade.
A lot of investment has been done to adapt the engine to tough regulations concerning low emissions
of particulate matter and carbon dioxide. The adaptations made are thought to be sufficient for the
machine to stay within the regulated limits for the next 8 years (Interviewee 3 2016). This gives the
company time to explore other alternatives, but also less incitements to release new models. In
addition, the machine is built on common platforms which might make alterations slower; which is
yet another argument in favour for the assumption that the machine in the future scenario will look
quite similar to today's machine.
5.3.2 BUSINESS MODEL
As previously mentioned, the business model in the future scenario will be a functional sales model,
in which the ownership of the machine stays with the producer and the access to an operable machine
is sold to the customers. In PSS terminology, the company will take the role as both enabler and
provider (Tukker 2015). The customer (previously the owner) and the operator only need to worry
about planning the tasks and operating the wheel loader, since all service and maintenance is included
in the contract. This will require that a close dialogue is kept between customer and the service
provider and that monitoring of the machine is done to avoid unnecessary down-time. Down-time will
be compensated for by the provider, and thus will induce major costs in the case of breakdowns.
Figure 10 shows an estimation of how the revenue is split on the areas machine sales, spare parts and
soft offers in today’s model whereas Figure 9 shows the estimated, anticipated sources of income in
the circular business model. The diagrams are principle figures to illustrate the difference between
the different business models and emphasize that the idea of a functional sales model is that the cost
of repair and maintenance is carried by the producing company, and profit can be increased by
designing better products to decrease this expenditure post. This implies that the rent should cover
investment costs, spare parts, maintenance, service technicians, repairs, soft offers etc. Despite the
pressure to cover all these cost, there is quite an opportunity minimize costs by using
remanufacturing, rebuild, optimizing procedures and designing longer lasting machines and
consumables etc.
30 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
5.3.4 RESEARCH AND PRODUCT DEVELOPMENT
Research and development will focus on decreasing the number of consumables, and the amount of
service the machine require. A focus will also be on machine modularity, allowing for easier repair,
service and remanufacturing to promote product longevity. Thereto, the business model is going to
need further development. A new question to answer is how to involve innovation and updates in the
machines, to allow it to stay attractive for the customer over a longer period of years. Part of the
development will also be to educate employees about environmental issues and make sure they
understand what impact their products have.
In addition, the process of additive remanufacturing can be explored to allow for a higher number of
parts to be suitable for remanufacturing, and even on-site remanufacturing. This, together with the
development of design for remanufacturing practices, will make sure that the majority of parts will
come from remanufacturing. Consumables can be designed for the machine life length or for reuse,
to limit waste and costs.
Product development is focused on design for extended life and all projects could start with three
questions: Which cores already in production can we use? How can innovation be applied? How do
we remanufacture/recycle the component? All new design solutions are carefully compared to, and
evaluated against, the data gathered by the machines on site. Engineers are not only measured on
cost, but also on environmental impact performance to increase the importance of an environmentally
friendly product development process.
5.4 IDENTIFIED KEY AREAS
In the following chapter identified key areas are presented. A key area, is a certain subject which is
identified to require consideration, if the circular business model, as described in the scenario, is to
be implemented. The areas can be a mix of methods, actions, theories etc., clustered together and
categorized according to common aspects.
5.4.1 REUSE
The producing company does not commonly reuse components today. Sometimes, as described
previously, used parts are sold by the dealer to fill the demand from owners not wanting to invest in
new parts. A suggested reason to why reuse is not utilized to the maximum of its capacity could be
that the company has no possibility to warehouse or keep track of possible components to reuse
(Interviewee 6 2016). In the scenario, this is different since the company stays owner of the product,
thus a range of possibilities opens to reuse parts from the wheel loader.
What has been identified as a challenge is that it must be acknowledged that parts are going to be
used in different machines and adapt both the machine and the part for this already in the
development stage. Designing for reuse includes the aspects of durability, adaptivity and possibility to
detach from the machine, and more.
Reuse is not necessarily the ultimate solution in all situations; some problems and possible obstacles
were identified. Reuse might hinder radical improvements; old and potentially inefficient components
and technology is preserved to allow for the reuse. Another problem is to ensure the quality of a used
component; which in turn might lead to either expensive quality testing, or the risk for premature
breakdowns. Another criticism towards reuse in this case is that it might induce large costs for
warehousing, administration and logistic chain, since parts and customers are widely dispersed,
geographically.
32 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Another obstacle for the company's reused parts business seem to be that the company structure is
designed for developing, producing and selling machines; while the reuse business also requires the
purchasing department to be in close contact with the market. Today, this contact is primarily
maintained by dealers. For the company to take part of potential revenue from reuse, establishing
such relationships and developing the required knowledge within the own organization is seemingly
essential actions.
5.4.2 REMANUFACTURING
Remanufacturing is usually labour intensive, but on the other hand, it saves virgin materials to a high
extent. The logistics and system around the process is still under development within the company
and representatives from remanufacturing strives to give design for remanufacturing higher
importance in the product development process (Interviewee 7 2016). Remanufacturing is already
good business today, generating income on used products. In the future scenario, remanufacturing
will not be a source of income, but rather a way to reduce costs, since components required for repair,
rebuild and maintenance can be remanufactured instead of replaced with a new one.
Today, only parts which are profitable in small numbers are involved in the remanufacturing schemes.
In the future, the number of remanufactured parts might need to be higher. This can probably be
achieved by using design for remanufacturing principles more frequently in the product development
process. Another problem today is that the company does not always own the rights to remanufacture
the components in their machines.
Whether a component is remanufactured today or not, is decided based on the economic value of the
product; i.e. expensive products such as the engine and the transmission are included in
remanufacturing schemes whereas, some products are designed as disposable components or
consumables, such as filters, oils and other liquids. In a circular business model, every consumable is
a cost and hence needs to be reduced. Another aspect is that by extending the product’s life with
design, the component might get more expensive, which in turn gives an incentive for
remanufacturing.
5.4.3 REPAIR AND MAINTENANCE
It can be concluded that a lot of research is required to find the best way of handling repair and
maintenance in a circular business model. But what was found after a brief literature review, including
the existing LCA, was that repair and maintenance consumes a lot of parts and materials, which
contributes to the exhaustion of resources on Earth.
As mentioned in the system analysis, most services are made on site, which means all equipment
needed for the service must be transported to the site by the technician. If a more environmentally
friendly approach is desired, it might be of interest to reduce the amount of service occasions.
A lot of environmental benefits comes with reducing service; and in a functional sales model, many
benefits are also economical. Ways of reducing service can be to develop better solutions from the
beginning, construct systems that do not need service, or design a condition based service model,
preferably together with longer lasting solutions.
33 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
5.4.4 RECYCLING
Another identified key area is recycling; since the company today take little responsibility for the
recycling of their wheel loaders, since the machines are often sold to countries outside Europe in the
later stages of their life. On top of that, the country where the wheel loader ends its life might not
have a good system for handling scrap or materials for recycling (Interviewee 10 2016).
In the case of consumables and service parts, such as oils, fuel and lube filters, the situation looks a
bit different. In Sweden, filters are classified as hazardous waste and thus needs to be taken care of
properly. Service stations have agreements with recycling firms such as Stena, which take care of, and
recycle, filters (Interviewee 9 2016).
Wheel loaders, as many other products, are getting more complex and rely on more computer
governed processes. Often, components that can control processes contain scarce metals, which
further stresses the need for recycling. If the producing company stays owner of the wheel loader, it
will mean that knowledge exists about what materials the machine contains and will be available in
the end of life phase. This fact implies a good opportunity to turn used materials into revenue.
Recycling methods and strategies are system problems; although, a lot can probably be done about
the wheel loader design to promote recycling. Action which can be done is to investigate if all design
for recycling principles are used for each component and to make sure that only materials are used,
for which there is an actual demand.
5.4.5 REPURPOSING
A replaced component is in general replaced for a reason, but it has not necessarily lost its entire
value. Parts of the machine might be unharmed and sometimes even in the same condition as when
it was put in place. There is today no system in place to track and take care of most of these
components. It is possible that new and innovative applications can be found to elongate the life of
components in new applications, within or outside the organization. This sums up the key area:
repurposing.
Few case studies on repurposing could be identified during the research phase. However, one
identified case was about repurposing of car, bus and truck batteries, by using them as energy storage
units in factories or apartment complexes (BMW-Group 2015). To make repurposing successful, it
should be considered already in in the design process in the future, since the early phase of a design
process is most suitable and less costly to make changes to a component (Österlin 2010). If the area
is considered early in the design process, repurposing is believed to have potential to offer an
alternative approach to problems of wasted capacity.
5.4.6 UTILIZATION
Another identified key area is utilization, especially for wheel loaders since the tasks that the machines
perform many times involve waiting. Also, the owner might have a fluctuating demand, which is
preventing maximum utilization. Service and reparation are actions which require the machine to
stand still and thus prevent maximum utilization. In a function sales model, a wheel loader will be paid
for when it is utilized, which could work as an incentive for the company to increase utilization of their
products through design. One interviewee mentioned that many wheel loaders are used 12 hours a
day and then the operator takes Fridays off, which would mean the wheel loader stays still for 3 days
(Interviewee 3 2016). Shared use can be one solution; although, L150 is a big machine and is not easy
to move around between users. The problem in the future scenario is that the customer, who has the
real influence over utilization, will no longer be accountable for wasted capacity. Clever solutions and
user studies are probably needed to increase utilization. Possible soft offers to affect customers to
34 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
maximize utilization can include combined fixed and floating rates in the contracts, or guaranteed
operating hours, or bonus systems for high utilization rates etc.
5.4.7 USE EFFICIENCY
Interviews showed that the behaviour of the driver and the application in which the wheel loader is
used greatly affect the life length of the wheel loader (Interviewee 2 2016). Every hour of use is an
hour of wear and reduction of life length; thus, use efficiency appears to be another key area. To get
the most out of the invested knowledge, material and capital, every hour should be used to its
maximum. Since the service provider offers education in operation efficiency, it can be concluded that
improvements are considered necessary in that field (Interviewee 3 2016). More studies and
observations are required to understand where most can be done to optimize the use of wheel
loaders.
In a circular economy, with a functional sales model, it is in the interest of the provider and owner of
the machines that they are operated in the best possible way, and are effectively protected against
excessive wear. The draw-back on the other hand, is that customers have less economic incentive to
treat the machine well. An optimization of those factors would increase the life length and the hours
possible to sell, and at the same time, potentially decrease maintenance. A challenge is to find the
right actions for saving resources; not only adding technology, hence risk to make the operator's task
more demanding. Suggested ways are monitoring technology to help the operator improve the driving
habits. Another improvement is applications to help plan tasks in an effective way. Thereto,
autonomous driving will probably make a very important contribution to this area. However, the most
important objective will be to analyse what operating behaviour has negative impact and work the
way from there.
5.4.8 MATERIAL SELECTION
Another key area is material selection, which has major impact on a variety of factors tied to functional
sales. The choice of material affects recycling rates, reusability, weight and hence fuel consumption,
remanufacturability as well as impact on ecosystems. So obviously, material selection is part of all
other key areas presented; despite that, the area is presented as a stand-alone to stress the challenges
and advantages.
Today, the company works with a red list of materials, forbidden to use in any of their products for
different reasons. A suggestion could be to develop a list of preferred materials that should actively
be used in products due to better recycling options and therefore less environmental impact.
5.4.9 BUSINESS MODEL
Today, the whole organization, the goals and the visions are built around delivering wheel loaders to
customers. The employees are measured on efficiency indicators connected to the amount of sales
(Interviewee 2 2016, Interviewee 5 2016). The business model key area contains a broad set of
challenges over an equally broad set of areas. Some aspects related to design are to keep the feeling
of a premium and quality product, even if it is used for longer and multiple times. Another aspect is to
keep up the customers’ interest for the machine and make it possible to incorporate new technology
and user benefits in the older machines. Related to this is to keep the machine within regulatory
boundaries over a longer time period. Until now, regulations have been focusing on emissions; though,
what demands will need to be stressed in the future might be hard to anticipate. Having machines
that run for longer, will mean that they must be able to adapt to new regulation during the period.
35 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
The company has not yet developed a concrete plan for what a circular business model would look
like. It will require time and competence to come up with a model that is profitable and suits
customers, users and the company (Interviewee 2 2016). If a transition towards a functional sales
business model is done, in simple terms, it could mean that the focus ultimately shifts from:
“manufacturing a wheel loader” to “solve the customers’ tasks more efficiently”. A use centred
development process will be a key to meet the demands of a functional sales business model, with an
inherent user focus.
5.4.10 ORGANIZATIONAL STRUCTURE
The last key area concerns the organizational structure and strategies of the company. During
interviews at product development departments, sales departments, at site with dealers,
remanufacturing representatives and with advanced technology department it was concluded that
information sometimes has trouble to find its way between departments. In some cases, there seems
to be a shortage of communication between sales, remanufacturing and product development, which
leads to that some possible opportunities for the company, the customers and the environment might
be overlooked.
Making important information easier to share between departments would benefit the company and
make it easier for new ideas to grow. If the environmental issues were better recognized, they could
be developed and lead the way into a more environmentally sustainable product. An improved
structure to facilitate communication and cooperation can have huge innovation potential and be
beneficial for most stakeholders.
5.5 KEY AREA EVALUATION
Table 2 presents the evaluation criteria which evolved during the prestudy, and worked as sounding
board to stress the relevant aspects in the project. The first two criteria are based in the thoughts from
circular economy resource efficiency. Other aspects are relevant for the project to stay within the
scope, time limit and make best use of competences. Other aspects assisted in the communication
with the company, about how close to current work the final concept should end up.
Table 2: Evaluation criteria with explanation
Evaluation criteria Explanation
The concept’s potential to offer a reduction of environmental impact compared to the
Material resource existing solution; in regards to extraction of new raw material according to Volvo’s
efficiency internal “sustainable materials report”. If applicable, quantitative figures from LCA
should be used to back up the result.
The concept’s potential to offer a reduction of environmental impact compared to the
Energy efficiency existing solution; in regards to fuel efficiency and energy consumption in production
etc. If applicable, quantitative figures from LCA should be used to back up the result.
How much the concept is expected to allow for, stimulate, and/or require innovative
Innovation potential
solutions.
The suggested concept’s expected ability to generate revenue, goodwill and developed
Volvo benefit
customer relations for the Volvo company.
The extent to which the concept is expected to create customer value by offering a
reasonable price over the life span (TCO) and/or satisfying customer needs to a higher
Customer benefit
extent than current solutions. For instance, equipment efficiency, reliability, employee
work environment etc.
To what extent the suggested concept is expected to satisfy user needs; ergonomically
User benefit (physical and cognitive), manoeuvrability, overall impression (quality, esthetical),
maintainability etc.
36 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
efficient. The objective will be to reduce the life cycle environmental impact from consumables, which
has been identified as a quite substantial impact contributor in the current LCA. The problem
formulation is defined as follows:
“There is an extensive environmental life cycle impact during the product’s use phase connected to
the consumables used for maintenance, such as filters, oils, coolants, tires etc.”
Consumables in the case of this study are defined as: “a part, or component that is added to the
machine with certain intervals for maintained performance and extended machine life”. The
consumable is often replaced, sometimes based on wear, and sometimes based on a set interval.
When replaced, a brand-new piece is installed and the used part is disposed. Examples of consumables
on a wheel loader are tires, hydraulic oil-, oil- and fuel filters, batteries, lamps and additives (Prosis
2016).
5.7 SELECTION OF REFERENCE COMPONENT
The main fuel filter was selected among the consumables as an appropriate reference product to focus
the development process on. The decision to continue to work with the main fuel filter was based on
aspects ranging from substantial environmental impact to project practical issues. The main fuel filter
is a disposable, which is replaced at almost every preventive service occasions, and is disposed as
hazardous material after use. Other aspects making the filter a good component to redesign and try
out the development process on, is the fact that it is a component with an unnecessarily high
environmental impact, and which allows for proper product investigation prior to redesign, remotely
from Gothenburg. Ideally, an appropriate product should have a mechanical function rather than a
process or chemical function. It was also important that the rights to the product better belongs to
the company for it to work as the reference product. Such product fulfils the competency criteria and
is possible to redesign without changing too much of the surrounding system. The main fuel filter lives
up to all these criteria and is therefore used as the reference product in the second stage of the L150
case study.
38 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
The concept development stage of the case study is aiming to take the learning outcomes from the
pre-study, and by using the use and sustainability centred development process, ultimately come up
with a final product concept. The concept is supposed to be relevant in a well-defined and credible
future scenario, which is realistic and explicit enough to allow for a quite detailed environmental
impact assessment. Furthermore, the suggested solutions ought to sufficiently respond to the pre-
defined effect goal defined as:
“To reduce environmental impact from fuel filtration (corresponding to the task of today’s main fuel
filter) and ensuring that the solution fulfils the adequate requirements from a functional sales
business model”
The concept development is made up by a structured process which can be described in three parts;
the first part aiming to establish the theoretical foundation for the development process through a
holistic system understanding; whereas the second and third parts, respectively, are aiming to
generate and refine adequate solutions to the problem, by considering the different aspects of the
project foundation described in the prestudy.
The use and sustainability centred development process aims to ensure that the final result can be
considered better from an environmental perspective and for the specified user. The credibility of the
final concept relies on the solution to be relevant for the defined system and the scope, which is
obtained by a structured process and verified by a final concept evaluation in the last step of the case
study.
39 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
6 BASIC NEEDS AND REQUIREMENTS STUDY
6.1 PRODUCT SPECIFIC SYSTEM ANALYSIS
The structure which is used to obtain the development baseline was system analysis. The existing
reference system; the reference context, product, task, the user(s) are all described for in the following
sections.
6.1.1 SYSTEM OVERVIEW
The Artefact, also known as the reference object for the study: the main fuel filter, is considered a
replacement article or consumable by the company (Interviewee 1 2016). During the fuel filter’s useful
life, it is static on the machine, filtering fuel with very little human contact. In fact, the filter is only
handled by the specified user(s) for a very short period, during mounting and dismounting and when
transported and disposed. A consequence of this fact is that the focus of the system analysis will be
shifted slightly away from the user and the task and focus more on the context and the product’s
lifecycle. However, since a part of the objective is to extend the product’s life, tasks and users are
likely to have a more central role in the suggested concepts, hence are still included in the system
analysis.
The main fuel filter is attached to the machine’s fuel system. The main fuel filter’s objective is to
prevent particles to reach the sensitive fuel injectors. Even small particles can cause damage to the
injectors, due to the high flow rates in the fuel system. Replacement of injectors is very expensive and
tedious (Interviewee 1 2016, Prosis 2016).
When the machine is started, the fuel pump (marked as 1 in Figure 11), feeds diesel with high pressure
from the machine’s fuel tank up in the fuel system. Before the diesel reaches the main fuel filter (2) it
passes through the primary fuel filter (3), for removal of larger particles and water. The primary fuel
filter is similar to the main filter, however equipped with an additional water trap screwed on to the
bottom of the primary fuel filter canister (4). Both filters are screwed on to a separate fuel filter
housing unit (5). The housing is feeding the filters with fuel through several inlets spread around the
edge of the filter and the housing then receives the filtered fuel through a hole in its centre. After the
diesel has been filtered by the main fuel filter, it is pumped onwards to the six fuel injectors (6) where
it is sprayed into the engine cylinders with high accuracy (Prosis 2016).
40 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Figure 11: Fuel system overview (Prosis 2016). (1) Fuel pump, (2) Pre-filter, (3) Main filter, (4) Water
trap, (5) Filter housing, (6) Fuel injector(s).
Particles enter the fuel system mainly through contaminated diesel, however possibly also due to fuel
tank corrosions and during service and repair operations on the system. Contamination of diesel
outside the fuel tank is often caused by “poor fuel hygiene”, i.e. deficient diesel storage at the local
sites or careless refuelling (Interviewee 1 2016). In some countries, insufficient diesel quality with
elevated levels of particles and water in the fuel, is believed to be a consequence of less serious
providers, suppliers or distributors (Interviewee 11 2016). Water in the fuel can cause internal
corrosion and less efficient combustion (Interviewee 1 2016). Water in the fuel is due to many of the
same factors as for particle contamination; however, water in the diesel can also be a consequence of
condensate air in storage tanks or the machine’s fuel tank (Interviewee 11 2016). Particles can be both
organic and inorganic matter; metal residue from corrosion, sand and dust from the environment etc.
(Interviewee 1 2016). In trucks, clogging due to build-up of organic matter on the filter surface is a
known phenomenon, caused by bacteria thriving in fuels with high water content (Interviewee 11
2016).
For trucks, the company researchers also have determined that the filter contamination rate is
primarily dependent on the mileage, and to some extent also the fuel quality (Interviewee 11 2016).
However, a wheel loader, in comparison to a car or truck, consumes diesel for other tasks than only
moving forward. Actual diesel consumption is detached from how may kilometres the machine runs;
hence, a better indicator for filter change is hours of operation rather than mileage. It is also
reasonable to assume that construction equipment in general have more problems with insufficient
diesel quality than on road equipment, since these machines are more commonly refuelled from local
storage tanks, sometimes of lower quality and with less circulation of fuel. These aspects might in turn
cause organic matter to grow and the storage tank to corrode etc. (Interviewee 1 2016). The filters
have been developed through empirical testing and the filter life is ensured by ISO-standard specified
41 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
quality tests (Interviewee 1 2016). The very same filter model is used on machines up to the biggest
machine in the company product portfolio, the L350 (Prosis 2016). These facts indicate that today’s
preventive maintenance concept, based on operable hours might not be optimal, if the fuel filters
alone were to decide the service system design.
6.1.2 SYSTEM MODEL
The overall system goal is to filter fuel in order to reduce the risk of particles reaching and inflicting
damage on the engine fuel injectors. To understand how this goal is fulfilled, the system is broken
down into sub-systems and system components and illustrated in the system models in Figure 12 .
The system model describes the connections between the system components within the system and
sub-system boundaries. The model also describes how the subsystems interact over the system
boundaries. Furthermore, the model illustrates some system entities which might affect the system
from outside the system boundaries, and in which ways. The focus is on flow of matter, information
and energy; however, it also includes the different external factors for the different sub-systems which
are thought to have impact on the system performance in some way. The most important system
components: user and artefact are marked out in the illustration. The flow of energy and matter in
the artefact sub-system is quite complex, however it is a quite good illustration of how the
components are attached and interact.
Figure 12: System model for the main fuel filter. Illustrating connections and flows between system
components and over the system boundaries.
42 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
The main system boundaries, marked with s black dashed line in Figure 12, is framing the system
elements which make up the fuel filter and contribute to its function. The elements which are enclosed
by the sub-system boundaries (dashed grey) are considered to have a strong connection to the
artefact and influences its function to a high extent. The outer system boundaries contain the service
elements; these are considered to influence the main task, i.e. the filter exchange. From outside the
system boundaries, contextual factors contribute to the system performance; these are further
elaborated in the context description in the following chapter.
6.1.3 CONTEXT
Much of the environment description in the context paragraph from the system analysis of the wheel
loader, presented in the prestudy (Chapter 5.2), is also relevant for the fuel filter system described in
this chapter. In addition, with the complementary scenario description in later chapter (6.1.6.3), the
overall context and related problems is believed to be covered on a general level. Although, a
complementary description of the influencing factors from the context, which are identified to
influence the system performance follows in this section. The key factors, visible in the system model
in Figure 12, are presented according to the different sub-systems: fuel filter, service components,
surrounding and the subsystem, fuel system. The most important aspect from the individual system
layers are presented next.
SUBSYSTEM: MAIN FUEL FILTER
Within the inner system boundaries, the filter replacement is dependent on the technician’s skills and
expertise. These aspects are assumed to be related to the background and training of the technician.
However, the replacement hygiene can also be connected to the user personality. The replacement
aside, the main fuel filter performance is as previously mentioned affected by diesel quality; which in
turn is dependent on suppliers, storage tanks and refuelling hygiene.
MAIN SYSTEM: SERVICE COMPONENTS
The success of the service and the possibility for the technician to ensure a high level of replacement
hygiene, is believed to be affected by external factors such as the weather, light conditions and dust.
The user’s ability to perform the designated task, is also affected by the character of the different tasks
(type and workload), the machine’s serviceability and the service policy and instructions from the
machine manufacturer.
EXTENDED CONTEXT: SURROUNDING
In the extended context or the surrounding, several different aspects have been identified to influence
the service operation. It might be quite static factors such as roads and facilities, the quality of
replacement parts, the customer’s demands and policies and the employer’s equipment, policies,
procedures and schedule. However, it can also be more dynamic factors such as the weather again,
workers in the site, client location, and unexpected additional repair etc.
6.1.4 ARTEFACT
As a support to understand the system, challenges and potential problems in the existing artefact, the
existing main fuel filter is described in this chapter. The described artefact is also the subject of
comparison in future concept evaluation. For the sake of the sustainability focus, a complementary
description/analysis of the reference artefact’s life cycle is also presented in this chapter.
6.1.4.1 MAIN FUEL FILTER DESCRIPTION
The main fuel filter, seen in Figure 13, is classified as a static product from a human factors perspective,
since it is not actively used by a person throughout most of its operable life. From an environmental
perspective, the filter is classified as a passive product, since a very small part of the product’s
43 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
environmental impact can be attributed to its use-phase (no direct consumption of resources); but is
rather connected to the production and end-of-life stages. Also, within the scope of this analysis, the
filter is not sold as a shelf product, but is included in a service package, which implies it is rather an
industrial product than a consumer product. This fact might have implications on what information
needs to be presented, packaging design, product appearance etc. (Kotler et al. 2011).
Figure 13: CAD rendering, representing the main fuel filter
The exploded view in Figure 14 illustrates how the various components are put together to form the
filter. After the primary filter, diesel enters the main fuel filter housing as described earlier and further
through eight holes in the filter interface plate (3) and into the filter canister (9). The interface is held
in place with the retainer sheet (2), which is also encapsulating the entire construction. The bypass
spring (8) in the bottom of the canister prevent dangerous pressure build-ups caused by for instance
unexpected filter clogging. The cellulose cartridge (6) in the canister is supported by an inner tube (5)
and capping discs on the bottom (7) and top (4). The diesel pass through the cellulose filter media of
the cartridge at a specified nominal flow rate and particles larger than a specified size are filtered out.
Filtered fuel continues through the threaded hole in the interface plate (3), back up in the housing and
onwards to the engine’s diesel injectors. To prevent diesel leakage and mixing of filtered and
unfiltered fuel, the main fuel filter is equipped with a rubber gasket (1). The filter comes with a plastic
dust cap to prevent contamination of the filter prior to assembly. The filter canister is coated with a
protective layer of powder paint.
Figure 14: Main fuel filter, exploded view. (1) Gasket (2) Retainer sheet (3) Interface plate (4) Filter
element cap (5) Inner tube (6). Filter element (7) Filter element cap (8) Bypass spring (9) Canister
44 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
rubber gasket is greased with motor oil, the filter is screwed on and tightened. It needs to be screwed
until the gasket is against the machine interface and then another three quarters to a full lap,
according to instructions on the filter. The technician needs to bleed the filter, which is done with a
hand pump, it requires 2-300 strokes to build up the pressure and get rid of air in the system. The
entire procedure takes about 0.7 hours (Prosis 2016). The used filter is sorted in a container for
hazardous waste together with other fuel and lube filters. It is picked up when full by a recycling
company (Interviewee 3 2016).
6.1.6 USER(S)
As been identified before, the existing reference object is exposed to very little user contact during its
entire operating life of approximately 3 months3. Identified direct contact is more or less only made
at two occasions; at mounting and dismounting (Interviewee 3 2016). However, since one of the main
objectives was to extend the filter life, it has been deemed important to still maintain a user
perspective, including the identification of the potential user characteristics, special considerations
and the actual user needs.
The identified users are described with support in user description and classification theory and the
primary user description is then complemented with a persona and a scenario in the following
paragraphs.
6.1.6.1 USER CLASSIFICATION
The identified user types are divided according to their relation to the artefact.
PRIMARY USER(S)
Since the task in this analysis has been defined as “the standard service operation with focus on the
filter replacement” the primary user is identified as the service technician; the person who will perform
the service hence the filter replacement.
SECONDARY USER(S)
Secondary users of the reference object are service centre personnel (order filter, plan service, pack
service boxes etc.), waste treatment personnel (collect filters), recycling personnel (collect filters and
manage the recycling procedure) and possibly also the after-market retail personnel (selling the
service and service components).
SIDE USER(S)
Side users are consumer (indirectly pays for filter), machine operator (rely on filter functionality),
product designer, service designer (design procedure and overall service system in which filters are a
part), manufacturing personnel (assemble and pack filters) and possibly performance testing
engineers (test filters in protected environment). Side users are in different ways in contact with the
products, but for different reasons than the main functionality. Amongst the examples, the
manufacturing personnel is in most physical contact with the product and hence needs to be
considered.
CO-USER(S)
Co-users might be other service technicians and workshop personnel; performing similar service tasks
and using the same tools, or performing other mechanic work on the machine.
3
500-hour interval/40 h a week = 12,5
48 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
6.1.6.2 USER INFLUENCE ON PRODUCT
The users or stakeholders which have been deemed to affect the needs and requirement for the fuel
filter are outlined in this section. The influence from these four users has also led to the decision to
have them represented in the following concept development criteria. The different stakeholders use
and motives are outlined.
OWNER
According to our study, the customer or owner is primarily concerned about two aspects in regards to
fuel filter: high up-time and low cost (Interviewee 2 2016, Interviewee 5 2016). Up-time is in turn
dependent on two aspects: mitigating the risk for unplanned stops and designing efficient service (few
and short). The customer often buy the filter as part of a service agreement, thus the cost in this case
rather refer to the total cost of ownership (TCO) than the individual filter price. The owner might also
affect the fuel filter design and performance through the diesel quality put in the machine; through
the selection of diesel providers and maintenance of storage tanks etc.
OPERATOR
The operator seems to have little influence over the product design and whether it is bought in the
first place. However, the operator may be concerned by the up-time of the machine and especially
the avoidance of unplanned stops. The operator influences the filter design and performance through
the refuelling hygiene.
SERVICE TECHNICIAN
The technician has some influence over the filter design and later requirement will be focused on
responding to this person’s needs. It is in the technician’s interest that the filter handling is somewhat
streamlined and clean, connected to what is referred to as the serviceability. This might include the
filter being easy to replace, the procedure does not include additional physical strain (ergonomic
handling) and that leakage and spill is prevented during handling etc. (Interviewee 3 2016, Interviewee
4 2016) The technician also influence the filter performance through the replacement hygiene.
SILENT STAKEHOLDERS
The objectives for the silent stakeholders: the environment and future generations, are not as closely
connected directly to the fuel filter; although, the fuel filter system has important indirect implications
for these stakeholders. The stakeholder’s possibility for prosperity will be negatively affected if natural
resources and sinks are exhausted. The environment could be changed beyond recognition if to many
ecosystems are exploited in the hunt for more materials leaving difficult life conditions for surviving
spices. This will impact future generations and their possibility to live acceptable lives. But the material
extraction itself can impact them as well, by exhausting resources today that might be crucial in the
future, we leave them in an unfavourable situation for being able to satisfy their basic needs. Both
silent stakeholders obviously need help to influence product design and today they are given voices
on a couple of different levels in the industry; first, through laws and legislation governing
manufacturing emissions, use-phase emissions and end of life treatment. Thereto, the company's own
environmental policies give the environment a voice trough sustainability strategies and research;
primarily run by the environmental department in this case. A last, but very important, influence is
obtained from individuals; designers seeking sustainable solutions and customers choosing
responsible alternatives.
VOLVO
Volvo, as a stakeholder is primarily concerned about two closely tied aspects: profit and customer
satisfaction (Interviewee 2 2016, Interviewee 5 2016). The first aspect can also include quantities,
prizing, manufacturing cost and more. The latter is about sufficiently responding to the previously
49 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Figure 17: Persona summary, Jonas. Used in the product development process to obtain and communicate
a uniform understanding of the user
6.2 PROBLEM IDENTIFICATION AND ANALYSIS
The system analysis of the fuel filter has led to several conclusions regarding actual and potential
problems, which all must be considered for the conceptual solution to achieve the desired effects. In
this section, problems are identified and together with the system analysis result translated into the
relevant product needs and requirements. The main problem areas, which have also been the primary
subjects for the screening are the sustainability and user related issues. The problems are sorted per
theme and presented in the following paragraphs.
6.2.1 SYSTEM DESIGN PROBLEMS
The first and what seems to be the most prominent problem regarding the fuel filters is the fact that
the entire filter is considered a disposable, even though a major part of the materials in the filter is
not consumed when disposed, according to our filter dissection. As can be seen in Figure 18, the metal
and plastic does not seem to be worn when disposed. The filter cartridge on the other hand has some
visible contamination after use. The recycling procedure of filters in general is quite inefficient in terms
of material recovery (Interviewee 13 2016) and the number of scrapped filters4 are high (Interviewee
3 2016, Interviewee 4 2016, Prosis 2016), which indicates there is potential to benefit from, by
considering design for recycling principles to a higher extent.
4 Fuel and oil filter together is about 20 units per machine and year
51 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
There are indications that the preventive maintenance scheme is not ideal for the Swedish market and
leads to premature disposing of the filters. Some of the indicators pointing in this direction are: the
lack of incentive to change maintenance system due to the high profitability from today’s system, the
fact that few fuel filter breakdowns or quality issues have been reported and lastly, from the drive to
coincide replacement of fuel, oil and breather filters and other service articles. According to the
company fuel filter specialist, the filters could be designed more robust to last longer if the need was
expressed (Interviewee 1 2016).
The fact that the filters are dimensioned to work for several machines in several different applications
and for different sizes suggests that the filters are not optimized to the various applications; hence,
contributes to a loss in potential resource utilisation. The particular main fuel filter is for instance used
in wheel loaders up to the largest machine in the range, the L350; in addition, a range of larger
excavators and articulated haulers share the exact same filter. This indicates that the fuel filter life
could benefit from individual design, condition based replacement or individualized replacement
schemes etc. (Interviewee 4 2016, Prosis 2016).
52 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
As been touched upon before, the main fuel filter can be classified as a passive product. This suggests
that the filter has its most important environmental impact during the production phase and from
end-of-life. Hence these are the phases improvements should be focused towards. However, it can be
argued that the weight of the filter contributes to the fuel consumption of the machine on which is
mounted and therefore the use phase impact also requires some consideration. However, the
allocated contribution5 from the filter alone is obviously very small.
Although, as the rapid eco assessment indicates, little is done in any of the filter’s life cycle stages to
reduce environmental impact. There are only a few aspects for which there are clear and conscious
initiatives to do so. One is material selection; where the company’s internal material selection process
seems to have been considered in the supplier requirements. The manufacturing and detailed design
related aspects have not been possible to assess to the full extent, due to limited access to information
from the filter supplier.
6.2.5 ADDITIONAL CONSIDERATIONS
Additional aspects believed to be contributing to the complete understanding of the problem picture
are presented in the following paragraphs.
FUTURE PROBLEMS AND THREATS
There are constantly increasing demands on companies to consider their sustainability work
(Interviewee 5 2016, Interviewee 8 2016, Interviewee 9 2016, Interviewee 10 2016) and it is only a
matter of time before customers pay attention to the wasted material and demand more sustainable
solutions. It is also reasonable to believe that the system of today induce extensive costs for disposal
of hazardous waste.
Fuel filters are most likely to be needed as long as today’s diesel combustion technology is used, due
to the need for protection of the injectors. However, new powertrain technologies are emerging as
for the truck and car sectors. For hybrid solutions, a much smaller engine is required and in turn,
smaller fuel filters (Interviewee 1 2016, Interviewee 5 2016). And for electrical engines, no fuel filters
are needed. Although, companies anticipating the shift to alternative technologies is not a valid excuse
of not doing anything about the 6 000 000 filters (RagnSells 2016) that are disposed in Sweden every
year; and even less of an excuse for the enormous numbers globally.
Another challenge which involves fuel filters, is the demand for up-time and the cost of manual labour.
These facts open the possibilities for service robots, which might be able to perform full services in
the future (Interviewee 11 2016), and thereby new kinds of requirements are enforced on the filters.
STRENGTHS AND BENEFITS
Along with the problems regarding the existing fuel filtration system, the analysis has also revealed
many of the benefits and strengths of the current solution, which just as well needs to be considered
in future concept development.
First of all, the filters work very well; fulfilling the specified performance requirements with few
reported quality issues (Interviewee 1 2016). Secondly, filters are good business for the company
today and a lot of effort has been put into convincing the customers that the genuine filters are
superior. As a result, many customers are prone to choose the genuine filters (Interviewee 5 2016,
Interviewee 9 2016). The consumables business opportunity has also a lot to thank the service
agreements; since almost all new machines today are sold together with a service agreement
5 Considering a 1.5 kg fuel filter on a + 20 000 kg wheel loader.
56 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
(Interviewee 3 2016), the company remains in control over which service products are used; and
genuine filters are used as standard in all service agreements.
A strength connected to the current design, is that the filters are encapsulated, which help maintain
cleanliness and avoid contamination. Also, the closed filter design is thought to prevent pirate copying.
Lastly, the filters are produced in a controlled manufacturing environment, which ensures sufficient
cleanliness (Interviewee 1 2016).
Another strength is that the service procedure in which the filter has a significant role is quite well
established among technicians and customers. Also, the service design department has done much to
improve the serviceability of the machines and has also improved customer satisfaction by making the
service operation as efficient as possible, thus cutting downtime (Interviewee 1 2016, Interviewee 3
2016, Prosis 2016)
6.2.6 PROBLEM DEFINITION SUMMARY
In order to facilitate the understanding, the problem picture, as identified, is summarized in this
section.
MAIN PROBLEM
The main problem, driving the need for the development of a new solution is defined as:
“Environmental impact from fuel filtration seem to be higher than necessary”
PROBLEM BREAKDOWN
The overall problem is broken down in different levels, presented under in the following bullets.
HIGHER LEVEL PROBLEMS
- The filters are considered consumables
- The filters are not reusable
- The filters are not adapted for functional recycling
- Most of the material in the filters still have major remaining capacity after disposal
LOWER LEVEL PROBLEMS
- There is no feedback when the filter is tightened.
- The sealing gasket needs to be greased with motor oil
- To prevent fuel spill during replacement, a container is put underneath the filter during
replacement today
- The filter needs to be entirely clean downstream
- Tools might be required for loosening of filter – it is sometimes stuck
- It is quite messy for the service technician to change the filters
6.3 CONCEPT NEEDS AND REQUIREMENT DEFINITION
6.3.1 NEEDS AND REQUIREMENTS LIST
From system analysis and problem picture, together with the project scope a list of initial
requirements were created. The needs and requirements are divided in the different areas of:
sustainability, functionality and use as seen in Table 5 below. The list is also complemented with
additional consideration, under the header: other.
57 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Table 5: Initial needs and requirements list
Category Need/Requirement Specifications
1 Sustainability The solution...
over the life cycle, compared to the
...must be less material demanding
1.1 reference
functionally recyclable and allowing
...must be reusable or recyclable
1.2 for 100 % material recycling
...must not lead to human exposure from hazardous
1.3 substances liquid, fumes
1.4 ...must consider distribution optimizations
according to Volvo's Sustainable
...must not contain scarce or hazardous materials
1.5 materials report
...must not lead to contamination of the environment
1.6 by hazardous substances
2 Functionality The solution...
2.1 ...must prevent particles to reach injectors larger than XX micrometre
2.2 ...must comply with existing standards ISO XXX
...must not induce additional risk of contamination of
2.3 the fuel system.
pressure, flow, vibrations and
...must resist the relevant mechanical demands
2.4 external force
2.5 ...must resist the relevant contextual demands corrosive liquids, diesel, oil, dust
...must resist the relevant external contextual
2.6 demands cold, humidity etc.
...must be possible to use without major changes of
2.7 the existing system today's fuel system and housing
3 Use The solution...
3.1 ...must be possible to use by one person
3.2 ...must be possible to use by the specified user experienced technician
4 Other It is considered positive if the solution...
4.1 ...contributes to decreased total down-time
...contributes to a more time efficient service
4.2 procedure
...contributes to a less demanding procedure for the in terms of physical and mental
4.3 user load
4.4 ...contributes to less transports
4.5 ...allows tracking of component
4.6 …expresses quality and company genuine equipment according to expression board
58 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
7 CONCEPT DEVELOPMENT
7.1 SOLUTION PATHS
The first step in the concept generation process: the solution path definition, is presented in the
following paragraph.
BASIC NEED MANAGING
Amongst the solution categories, this one is located on the highest system level, thus solutions within
this category are assumed to have the biggest positive influence on the system’s environmental
impact. The category is mainly derived from the ecodesign strategy: “Design for innovation”. Primarily
two design related actions are considered adequate for the category and the system as defined; these
are:
System/product redesign to…
- Eliminate the basic need
- Reduce the basic need
EXTEND USEFUL LIFE
The category is a derivate of the Eco wheel strategy “Design for system longevity”. It focuses on either
prolonging the product life and/or sub-components. Maintenance programs, refurbishing, reused
products are strategies mentioned to prolong product life and remanufacturing and reused
components are strategies to accomplish a longer sub-component life. Relevant design related actions
which have been identified for this category are:
System/product redesign to…
- Last longer; robust design, durable materials, more efficient (optimize)
- Foster connection with user.
- Allow for reuse of product or components
- Allow for repurposing of product or components
- Allow for refurbish and remanufacturing of product or components
- Incorporate maintenance program to avoid premature scrapping, through monitoring wear
and service/repair.
REDUCE IMPACT FROM MATERIALS
The category is a combination of the aspects presented under the ecodesign strategy: “Design for
reduced material impact” and “Design for optimized End of Life”. The category includes focus areas
such as design for recycling and recycling systems design etc.; and also, responsible materials
selection. Design related actions to reduce environmental impact from materials which have been
deemed relevant are:
System/product redesign…
- For maintained material value; design for recycling
- For increased product/material return rate
- Of efficient recycling system; labour and energy efficient.
- To become more material efficient; optimize design in relation to performance/structural
requirements.
- Responsible material selection
61 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
7.4.1 REPLACEABLE CARTRIDGE
Traditionally, in most filter solution, only the
cellulose filter media has been considered a
disposable, since this is the only part which is
actually consumed. Usually the canister, the bypass
spring and interface etc. remain intact. This concept
takes advantage of the principle of reusing the full
functionally of the intact metal and plastic
components. The concept represented in Figure 21
shows how the product solution could look like; a
replaceable filter cartridge made entirely in
cellulose material. Compared to how cartridge
solutions worked in the past, this solution is also
adapted for the studied application, use situation
and users to counteract the reasons to why the Figure 21: Replaceable cartridge concept,
cartridge replacement nowadays has been second concept development stage. The used
filter to the left is replaced with a new cellulose
discarded as an alternative in many applications.
filter, seen to the right in the figure.
THE PROCESS
How the designed process is anticipated to work is represented in Figure 22 and described as follows:
The filter casing is opened and the cellulose insert is removed, the housing is cleaned and the insert is
replaced with a new one and the filter house is then sealed and reattached to the machine. The used
insert is brought back by the technician and disposed in a certain bin, which can then be sent to a
recycling process were the cartridges are centrifuged to extract remaining diesel, and the cellulose
fibre is then energy recovered at a district heating facility or possibly sent to a biogas production
facility.
Figure 22: Cartridge concept, schematic overview. Showing the replaceable cartridge's life stages in
the service loop, from arriving to the workshop to the recycling.
THE ENVIRONMENTAL BENEFIT
The cartridge solution is sprung out of the circular economy theories concerning how materials should
not be disposed if they are not consumed, which in turn will lead to less material use over the life-
time. Since the outer casing and the supporting structure can be reused, less material is disposed every
filter change and less material is transported. Another benefit regarding the solution is that the
64 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
requires tracking, warehousing, and economic transformation; with the latter including reman
components to be considered inventory, rather than sales articles. Another problem lies in how to
remain in control of the components over the product’s useful life, to ensure they return for
remanufacturing. One of the main issues in this lies in motivating the customers, which is especially
challenging when the value of the product is decreasing and when the product is leaving the primary
customer and possibly ending up on more remote markets. Another limitation with reman is that it is
often not considered profitable for smaller and less expensive products, even though quantities are
big; this is because of the individual and often manual treatment that is necessary. Furthermore,
companies seem to experience problems regarding how to set up the logistic chains (Interviewee 7
2016). Many of the identified issues are resolved in a circular business model, where incentives are
created to increase the products’ and the individual components’ life.
The company has experience in setting up new remanufacturing facilities. Obviously major
investments are necessary to set up an automated remanufacturing line. However, the large amount
of quite undamaged filters, which end up in recycling facilities today offers an opportunity, just waiting
to be exploited. In addition, a disposed filter equals hazardous waste, which needs to be handled
accordingly, through costly procedures. For the concept to be economically and environmentally
justifiable, the idea is to expand the filter remanufacturing to include all filters which are changed with
the same intervals, i.e. the primary fuel filter and the three oil filters.
THE PROCESS
The suggested reman procedure illustrated in Figure 23Figure 23 is aiming to bring the filter back to
as good as new standard. First, the contaminated filter is removed from the machine and exchanged
for a remanufactured filter. Used filters are placed in the sealed transport container in which the
remanufactured filters came. The container is brought back to the workshop and placed on a pallet,
which is sent to a reman facility regularly in return for a pallet of remanufactured filters. In the reman
facility, the filters are opened and brought back to as good as new standard (depending on filter media
the filtering cartridge is replaced with a new one, or washed.) The filter is then resealed and put back
in the transport container waiting to be returned to the service centre. The controlled environment in
the reman facility guarantees a clean product. The disposed filter material and/or washing liquids are
taken care of at the remanufacturing facility and treated in an, for hazardous waste, appropriate way.
Figure 23: Remanufacturing process, schematic overview. Illustrating how the washable filters are
moving within the service loop between the wheel loader, the workshop and the remanufacturing
facility.
66 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
7.4.3 LOCAL FILTER CLEANING
The third, and last, concept is also built on the principle
to reuse non-consumed components. This time, it is
obtained by using a washing machine placed in the local
service centres, to wash the filters. The local filter
cleaning concept in Figure 24 is described in detail in
the following paragraphs. The idea is that if the solution
is designed more robust and slightly more material
demanding, it might be possible to develop much
better and more environmentally friendly filters,
considering the entire life cycle. Benchmarking showed
that there are washable filters in aluminium or stainless
steel, which can be washed instead of being replaced. Figure 24: Local filter cleaning, second
Also, according to the specifications, these filters seem concept development stage. Concept
to last longer before clogging than the cellulose based visualized together with an idea about
alternatives used today. what the washing machine could look like.
THE PROCEDURE/PROCESS
An overview of the concept process is shown in Figure 25. The technician removes the used filter and
replaces it with a clean filter. The contaminated filter is disassembled and the parts are stacked in the
sealed transport container in which the clean filter was transported. The container is brought back to
the workshop where it is placed in the dishwasher. When the machine is filled with filter containers it
is started. After the filters have been washed and dried, they are reassembled and put back in the
clean transportation container and stored until needed.
Figure 25: Local filter cleaning, schematic overview. Illustration how the filter never leaves the loop, it
just travels between the workshop and the wheel loader.
68 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Material usage
Energy consumption
Material selection
Concept completion level
Technological feasibility
Customer benefit
Serviceability
0 1 2 3 4 5 6
Local filter cleaning Filter Remanufacturing Replaceable Cartridge
Figure 26: Unweighted concept comparison. The three concepts pick scores from different areas. The
Filter remanufacturing concept receive the highest unweighted average (3.6), compared to Local filter
cleaning (3.4) and Replaceable cartridge (3.1)
MATERIAL USAGE
Material usage was rated the most important aspect, hence had a major contribution to the concepts’
final, weighted result. Due to the very low amounts of disposed material per use cycle in the two reuse
concepts, these two came out as clear winners in this category.
CUSTOMER BENEFIT
Since the idea for all solutions was that they should to be used in a functional sales business model,
the customers will be in little contact with the solution hence the customer benefit was weighted low.
The most important aspects for the customers will be uptime, and the performance of each solution
will not be compromised, hence the quite small differences between the solutions for this criterion.
SERVICEABILITY
As prioritized as the third most important criteria, serviceability has quite an impact for the overall
result. The cartridge scored low in this category since it requires additional work, however even more
so due to the higher risk for the technician to be contaminated by diesel during cartridge replacement.
The other two concepts are considered better due to their similarities with the reference solution,
which has been optimized to minimize manual labour and which can also be considered quite safe.
TECHNOLOGICAL FEASIBILITY
The score for each concept under technological feasibility is quite similar. However, the differences
are a consequence of the uncertainty regarding the possibility to produce a technically and
economically feasible washing solution for each service facility, or in large scale at a centralized
remanufacturing facility.
70 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
CONCEPT COMPLETION LEVEL
The result for concept completion level differs little between concepts. However, the completion level
concerns the filter solution. Less attention will be paid to the system design. Thus, the first concept
would seem more ready since the system is already in place and can be described in more detail.
MATERIAL SELECTION
Material selection was considered the second most important factor. The reason to the differences
between the concepts is due to that the two reusable material concepts are believed to require high
alloy canisters, which might include scarce materials, to endure the performance requirements from
multiple refurbish/washing and a substantial amount of handling and transportations.
ENERGY CONSUMPTION
The energy consumption was considered of less relative importance at this stage, due to the focus on
materials. The differences between the concepts in this category are due to estimated variances in
concepts weight and transportation distances.
7.5.2 CONCEPT SELECTION
Considering the predefined criteria, the filter remanufacturing concept came out of the evaluation as
the one with the highest potential, as can be seen in Table 9; hence was selected as the final concept
for further refinement and a final evaluation against the reference case.
Table 9: Weighted concept evaluation. The remanufacturing concept outscore the other two
concepts, hence proceeds to the final concept refinement stage.
A very important learning from the second phase of the product development process was the
washable stainless steel filter material which came up when researching the “local filter cleaning”
concept. This stainless-steel cartridge was considered appropriate to use for the final version of the
remanufacturing concept for substantial material saving over the product life.
7.6 SPECIFIC NEEDS AND REQUIREMENTS
Before the final concept refinement, the baseline for the following development process was
established through a more specific version of the basic needs and requirements which were
presented earlier. The specific needs and requirements also contain product requirements from the
established remanufacturing procedure.
71 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
7.6.1 LIST OF SPECIFIC NEEDS AND REQUIREMENTS
The needs and requirement for the final concept is presented in Table 106 below.
Table 10: Final concept, needs and requirements list
Category Need/Requirement Specifications
1 Sustainability The solution...
...must be less material demanding over the life cycle, compared to the
1.1
reference
...must be reusable > 40 use cycles, which is
1.2 corresponding to the useful life of
one machine.
...must be recyclable 100 % material recycling possible
after useful life
...must not lead to human exposure from liquid, fumes
1.3
hazardous substances
1.4 ...must consider distribution optimizations
...must not contain scarce or hazardous according to Volvo's Sustainable
1.5
materials materials report
...must not lead to contamination of the by hazardous substances
1.6
environment
2 Functionality The solution...
2.1 ...must prevent particles to reach injectors larger than XX micrometre6
2.2 ...must comply with existing standards ISO XXX6
...must not induce additional risk of
2.3
contamination of the fuel system.
...must resist the relevant mechanical demands pressure, flow, vibrations and
2.4
external force
2.5 ...must resist the relevant contextual demands corrosive liquids, diesel, oil, dust
...must resist the relevant external contextual cold, humidity etc.
2.6
demands
...must be comply with the relevant filter specified in classified filter
2.7
performance requirements requirements6
…must resist (exchange) > 40 of each
2.8
disconnections/connections
2.9 …must resist disassembly/assembly > 40 times
2.10 …must resist washing, filter > 40 times
2.11 …must resist washing, canister > 40 times
2.12 …must resist transportation (2 way) > 40 times
2.13 …must allow for inspection and testing pressure drop testing
…must prevent downstream contamination water and particles
2.14
during disconnection/connection
…must prevent downstream contamination water and particles
2.15
during transportation
3 Use The solution...
3.1 ...must be possible to use by one person
3.2 ...must be possible to use by the specified user experienced technician
...must be possible to remove without
3.3
introduction of new tools
6 Regarding requirement 2.1, 2.2, 2.7; detailed performance requirements are classified, hence left out of the list.
72 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
8.2 THE CONCEPT FILTER
In the following paragraphs, the final filter concept (Figure 29) is presented in terms of its product
attributes.
Figure 29: Explode view, major component. From left: Lid, filter cartridge, by-pass spring and canister
MAIN FUNCTIONALITY AND PERFORMANCE
The relevant performance requirements regarding flow and pressure are fulfilled by preserving much
of the interior surfaces similar to the same as the reference filer; the volume remains a static design
parameter. The filtration capacity requirements however are expected to be fulfilled through adopting
solutions from existing products with similar filtration requirements, further described in the filter
cartridge sub-section. In addition, a similar bypass solution is implemented for the concept.
To resist multiple washing aluminium was selected for the canister. Stainless steel was also
considered, however discarded due to the weight which would have been required and due to the
differences in production possibilities between the materials considering the desired design. In order
for the filter cartridge to endure multiple washes, an existing stainless steel filter mesh solution (Figure
30) was selected.
Figure 30: Photo of product sample of existing Figure 31: Filter concept from
filter solution from below, with QR-code visual
76 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
To save transported amount of material compared to the reference solution, the concept design has
eliminated the interface plate from the transported unit. The main functions of the interface plate are
to attach the filter to the machine, bring structural integrity and distribute incoming fuel. Attaching to
the machine is solved by an alternative attachment system. The structural integrity is obtained by
thicker material and a slightly different shape. The distribution of fuel is done by a static unit, similar
to the interface, which stays on the machine.
Removing the interface also caused problems related to the transportation. First, the new attachment
solution had to be protected in order to endure multiple transportations and also, the interface to
some extent helped keep the filter inside clean before mounting; these issues were resolved through
a protective transportation lid presented under “transportation tray”.
Today, there is a lot of information printed on the canister. Some of which is not easy to understand,
nor adapted for the specialized company technicians. This might be due to the fact that users can vary
after the service program ends. However, in a functional sales model we can expect a more
homogenous user group. Due to their high level of knowledge and experience regarding the service
operation, the information which is necessary to be provided by the filter can be assumed to be
minimal. Therefore, only the logo is printed on the canister. However, if the technician perceives a
need for additional information, this can be obtained by scanning the QR-code which is laser engraved
on the bottom of the casing, seen in Figure 31. The page connected to the QR offers information about
safe handling, hazards and risks, assembly instructions and more, although it must be further
evaluated if such a solution provides enough information to the user. The QR-code is also intended to
function as an identifier in the reman factory.
There is also a marking engraved on each filter component and the lid, telling which filter type they
belong to, in this case: “main fuel filter” is engraved, to allow for separation of the five different filters
changed with the same interval.
8.3 THE ATTACHMENT SYSTEM
The filter-machine attachment system is described in the following paragraphs. To endure the large
number of dismounting and mounting over the expected life, alternative attachment solutions were
considered. Several different systems were conceptualized and evaluated in regards to their individual
potential, together with machine experts from the current service provider.
The selected solution ensures that the force is no longer applied to the canister body when twisting it
during the dismounting, but only to the new attachment solution, called the lock nut. The solution,
referred to as the “meat grinder” attachment is already in use among some filter manufacturers today.
The solution is illustrated in Figure 32 below.
77 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
To ensure the lock nut is used as intended and no force is applied to the canister body, the lock nut
has been designed with the right affordance. In this case, meaning that the grip pattern on the lock
nut has been designed to be recognised by the user and associated with the action that the designer
desires. The second part in ensuring the correct use is to educate the users about the change in
replacement procedure. Different lock nut shapes, grip patterns and sizes from various applications
were assessed, before the final solution was selected. The fact that the filter body is static when the
lock nut is tightened is believed to have the effect that no greasing of the gasket is required; which in
turn saves time, due to one lesser step in the service procedure and also, saves the technician from
additional contact with hazardous substances.
The surrounding contextual aspects were primarily considered in the design of the attachment system;
for instance, by reducing the number of moving parts, avoiding small spaces and horizontal surfaces
where there is a risk for material build-ups. The attachment system ought to prevent spillage during
dismounting through a solution where the filter canister remains still, while the lock nut is the only
part moving. Also, the canister walls are slightly higher in relation to the location of the inlet, which
causes the percentage of the volume which is filled with diesel to be smaller than that for the
reference solution.
8.4 THE TRANSPORTATION SYSTEM
The transportation system seen in Figure 33 is presented in the following section.
THE TRAY
After the possibility to clean the filter, preventing the filter to be in any way damaged during handling
and transportation is the most important aspect in the effort to increase the product life. The
suggested solution ought to do this, first through the design of the filter; a robust casing without sharp
edges as previously accounted for. The more important contribution to dent and scratch prevention
comes from an external product. The transportation tray offers protection to the filters from right
after dismounting to the point when the filter is remounted at the machine. The transportation tray
allows the filters to be securely fastened to prevent them from moving around during transportation.
The upright position in combination with the lid, also helps preventing diesel leakage. In addition, the
bottom plate of the tray is diesel proof, with a wall around the edge which will further prevent diesel
leaking. Since five, similar sized filters are replaced at the same intervals; the suggestion is to have a
combined transportation tray for all five filters.
The shape of the transportation tray (see Figure 33) is considering anthropometric features of the
technician. The distribution of weight was considered when the location of the filters was decided and
the carry shoulder strap was added to facilitate carrying.
THE LID
To prevent contamination of the important clean side of the filter between cleaning and mounting a
transportation lid was designed (see Figure 34). The lid attaches to the set of threads which are used
for mounting the filter to the machine housing. The filter lid is supposed to be removed from the clean
filter just prior to assembly and then screwed onto the used filter to prevent diesel leakage during the
transportation.
79 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Figure 34: Transportation lid overview
A centre-cone integrated to the lid prevents the cartridge from moving around inside the canister;
hence reducing the risk for damage or excessive wear during handling and transport. The edge of the
lid extends down over the full length of the threads to protect them from damage. Also, a more robust
type of threads has been used for the concept compared to the reference.
The grip flange on top is quite small, and the shape has been selected among other alternatives since
it implies gentle handling (since it bears little resemblance to a “power grip”). The reason is to make
the technician generate just the right amount of force, hence not risk damaging or wearing out the
filter prematurely.
The filter lid is designed to allow for attachment of a carabiner, as shown in Figure 35. This will allow
the technician to snap the filter to the belt while climbing up to larger machines. The reason for this
addition is primarily because, it is anticipated that this will lead to the lid staying on the canister until
just prior to replacement, hence reducing the risk for filter material contamination. However, it is also
considered a safety and ergonomics benefit
that the technician’s hands can be freed up
before and after filter replacement.
Two different approaches were considered
for preventing spillage during transport;
either draining the filter from remaining fuel
or to enclose the canister. The latter was
selected, since draining would cause
additional risk of spillage. It would imply an
additional, time consuming step and it would
also be difficult to ensure that the filters are
entirely emptied. The downside of the
containment solution however is that the
remaining fuel in the filters add to the total
weight that has to be transported. On the
other hand, the reman facility will be a better
place to deal with hazardous liquids, than
directly on site, as would be the alternative. Figure 35: Transportation tray, additional functionality
80 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
8.5 ADDITIONAL DESIGN CONSIDERATION
ADDITIONAL REMANUFACTURING CONSIDERATIONS
In order to allow for the washing, all parts are supposed to be
produced in materials which can resist the minimum of 40 washings,
as stipulated by the requirements. This also concern the
transportation lid and tray. In addition, all components are possible
to easily separate from each other, to allow for an efficient washing.
ERGONOMIC USE
The final design has considered design for ergonomic use aspects
described in the theory in Bohgard (2011). Instead of both pressing
the filter upwards and twisting when mounting, as for the reference;
the suggested solution allows the filter to be held in one hand while
using the other to screw on the lock nut the first couple of threads,
before the weight of the filter can be released and the lock nut can
be screwed on. In addition, a carry strap can be attached to the
transportation tray to facilitate carrying, see Figure 36.
Another important ergonomic design improvement was regarding
adequate feedback; the reference solution offered insufficient
feedback to the technicians for when the reference filter was in the
right position. For the purpose of feedback, but also not to risk over
or under tightening the filter, a small step-out the bottom of the Figure 36: Transportation
thread (“stop-edge”) which meets the lock nut edge when tray, carry solution
appropriately tightened was added to the canister.
INSPECTION AND TESTING
A suggestion for testing of the solution after cleaning is to test for instance differential pressures. Too
high differential pressure would indicate the filter is not clean, hence it is still clogged which prevents
the flow and too low differential pressure indicates there is a leak in the filter media, which is letting
fuel pass through with less resistance.
RECYCLABILITY
The concept is designed to be entirely separable into parts (see Figure 37) each containing a single,
documented material mix, i.e. a specific stainless steel alloy (parts 2-7), rubber mixture (8), plastic
mixture (1 and the tray) and aluminium alloy (9). It was considered whether the different parts in the
cartridge were going to be separable as well, to allow replacement of worn sub-components; however,
the solution was discarded, due to the conclusion that such solution might compromise the structural
integrity of the cartridge, hence the main functionality of the filter.
81 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Figure 37: Concept, exploded view. (1) Transportation lid (2) Filter element cap (3) Outer tube (4)
Filter element (5) Inner tube (6) Filter element cap (7) Bypass spring (8) Gasket (9) Canister
Also, all units are sufficiently marked with material content since it is reasonable to assume that some
filters might end up out of reach for the company; thus, the markings help ensure recyclability even
in such cases.
AESTETICS AND EXPRESSION
For or the filters to be used for as long time as possible, it has been concluded that they also need to
look attractive. However, that does not automatically mean that they must look brand new; it is
possible that even though a filter looks as if it is well-worn on the outside but is perfectly fine on the
inside it can still be trusted. However, more testing and research is needed in this area.
However, for this study, it was deemed important that scratches and smaller damage was concealed
as much as possible, which was achieved by a brushed surface (Figure 39). The material and the design
will also allow a thin surface layer to be lathered away at a point in time when scratches and cuts are
compromising the overall appearance of the filter. The expression is also maintained through the
filters being delivered wrapped in paper to the assembly facility. This is to avoid the filter body being
painted when mounted on the engine block.
Figure 39: Material properties close-up Figure 39: Final concept, realistic rendering. With place
rendering for all filters replaced at the same service intervals
82 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Three factors were mainly considered for the design to achieve the desired visual and tactile
expression; the surface quality including material, colour and surface treatment. The second factor is
the visual and tactile expression of the form/shape. Many variations were assessed to find the
satisfactory form expression. The third factor was the functionality, both perceived and actual. The
finalized concept can be seen in Figure 39.
8.6 FINAL CONCEPT EFFECT GOAL FULFILMENT
A qualitative concept verification against the effect goals is presented in Table 11 below. For each
effect goal, fulfilment is justified with concrete examples in the right column.
Table 11: Effect goal fulfilment evaluation
Effect goal Goal fulfilment explanation
The solution is required to…
...prevent damage during handling and
- Robust canister with no sharp edges
transport
- Robust threads
If the filter ought to be reused for more than 40
- Protected and secured by transport tray solution
cycles, the filter and/or the system needs to be
- Threads protected by lid
designed accordingly.
- Canister secured by lid centre cone
...prevent filter contamination during handling
and transport - Lid prevent contamination from outside
The downstream side of the filter must remain - Lid centre cone prevent contamination between up
entirely clean during transportation from and down stream side
remanufacturing facility, to the client until - Possibility to attach carbineer might lead to the lid
attached to the machine. staying on for longer.
...resist multiple mounting/dismounting - Lock nut solution prevent force to be applied directly
In order for the solution to withstand at least 40 to filter canister
mounting/dismounting; the attachment system - Robust threads
must be designed accordingly. - Large threaded surface better resists external force
- Less wear on gaskets since filter body is static when
lock nut is tightened
- A stop-edge on the canister prevent over-tightening
of lock nut.
...allow for washing
- Filter is “disassembleable”
As specified in the requirements list, the filter
- Washing away particles is allowed by the stainless-
needs to withstand a minimum of 40 cleaning
steel filter mesh
cycles, hence the filter must be designed
- Materials for all components enables washing
accordingly.
- Split lines and cavities are avoided when possible
...fulfil relevant performance requirements
- Similar pressure resisting and fluid dynamic
The design is required to deliver the
properties can be expected since the design of the
corresponding performance as the reference; i.e.
inside is maintained relatively intact
sufficiently respond to the technical/functional
- Selected technical solutions are proven to work in
requirements as defined for the existing fuel
similar application (aluminium canister, lock nut and
filter regarding fuel pressure, filtration capacity
stainless steel mesh)
and flow rate.
...avoid transportation of excessive
- Lock nut is static on the machine
matter/materials
- Filter interface is static on the machine, integrated to
Since remanufacturing rely on transportation of
the housing
the components, the possibility to reduce the
material being transported must be considered.
83 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
...avoid diesel spillage during dismounting
Since the filter canister will be filled with diesel,
spillage must be prevented when used filter is - Higher edge, relative to containing diesel volume
detached.
...avoid diesel leakage during handling and
transport
The filter cannot cause contamination of people - Transport tray maintain up-right position
or the surrounding while being handled and - Lid encapsulate the canister
transported.
…provide adequate information - Lock nut offers the right affordance (prevent the use
The information provided on the components of filter body as grip when dismounting)
must be adequate for the user and the use. - QR-code enables additional information to be
obtained
- Any unnecessary information is avoided
...allow tracking of the product
Tracking of the complete product and possibly
- QR-code enables easy tracking on site, during
even the individual components is deemed
transport and in reman facility.
required, to optimize distribution, handling and
- Components are individually marked
procedures.
...allow for inspection/testing - Filter is disassembleable
After remanufacturing, the performance of the - Components can be manually or automatically
filter needs to be ensured by inspection and scanned for defects
quality testing. - Quick connection to a rig after reman for testing is
possible
...allow for high value recycling - Business model ensure materials stay with the same
The design must allow high value recycling owner which provides total control and insight
through considering the design for recycling - Business model allow for 100 % return rate
aspects. - All parts which made from different materials are
separable
- Parts are marked for content
- Clean parts are recycled, since being cleaned in
reman facility.
The solution is required/desired to…
…maximize life expectancy in general - Large threaded surface better resists external force
The solution should consider how the expected - Robust canister
life of the filter can be maximized in general; - Brushed metal canister surface conceal damage and
other aspects affecting the filter life than what is scratches
presented above. - Parts can be replaced
- Canister material and design allow
lathering/grinding/polishing to retain appearance
- Filters are delivered in protective packaging to
assembly line when new.
- Filters are delivered wrapped in paper to prevent
being painted in assembly line.
...avoid introduction of new tools
- Existing tools can be used for dismounting and no
If not justifiable in relation more prioritized
tool is used for mounting
demands, introduction of new tools is
84 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
9.3.1 PROCESS
The LCA was conducted according to the process described by ISO_14040 (2006) and modelled in GaBi
(software Thinkstep AG). According to suggestions from ISO_14040 (2006), a goal and scope was
defined, an inventory assessment conducted and finally an impact assessment done which was made
understandable through an interpretation subsection. The goal and scope should be defined and work
as guiding marks throughout the process. The scope decides the system boundaries of the studied
system, information about system boundaries is crucial to understand the impact assessment.
Because of that, a brief description of the system will follow:
The goal of this particular study was to perform one LCA on the existing fuel filter and one on the new
concept filter to compare their environmental impact.
The purpose was to use LCA as an assessment tool in the product development process to see if the
product, developed for a circular business model, was better for the environment than the existing
solution.
INITIAL FLOWCHART
Both LCAs covered the life cycle form cradle to grave since materials use was of interest. Flow charts
showing all parts of the fuel filters’ lives can be seen in Figure 40. Note that the life cycle used to
describe the flows in the LCA differs from the product life stages used in the product development
terminology in previous chapters.
Figure 40: Initial flowcharts, reference (left) and concept (right)
FUNCTIONAL UNIT
The functional unit was set to 20 000 hours of fuel filtration in one wheel loader L150, which is the
function of the main fuel filter. 20 000 h is the estimated amount of available operating hours in the
wheel loader used in this study.
SYSTEM BOUNDARIES
Geographical boundaries were set to Europe for parts production and assembly, and Sweden for the
use phase and end of life. Extraction of materials and manufacturing of parts were included in the LCA
as well as transports of the assembled fuel filter.
88 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Table 13: Material content in the reference filter and in the concept filter respectively.
Part Material, reference Material, concept
Canister Stainless steel Aluminium
Filter media Cellulose Stainless steel
Inner tube Nylon Stainless steel
Filter media caps Stainless steel Stainless steel
Spring Stainless steel Stainless steel
Gasket Styrene-Butadiene rubber Styrene-Butadiene rubber
Supporting grid Not applicable Stainless steel
Interface plate Stainless steel Not applicable
Element cap Stainless steel Not applicable
Transportation lid Not applicable Plastic
The remanufacturing procedure needed to be somewhat defined for the LCA modelling. Table 14
shows the data used in the model. The production of the facility itself was not included in the LCA.
Table 14: Remanufacturing process data
Inflow material Quantity/one filter
Detergent 20 [g]
Electricity 229 [kJ]
Water 4 [kg]
Steel parts 12.8 [kg]
Rubber parts 10 [g]
9.3.2 RESULT
In the following section the results from the assessment are presented, starting with results based on
comparing one filter against the other. Further down is the impact assessment presented.
9.3.2.1 MATERIALS
Table 15 shows how much material is needed for each filter solution to provide fuel filtration over 20
000 h in a wheel loader L150. It was estimated that one of the concept filters could withstand all wear
and tear for one machine life time. However, the filter might be possible to use even after the assumed
time. The assumption of one filter per machine should be seen as a conservative assumption, in the
following analysis the number 1.16 filters are used. This is because slightly more filters than machines
are needed since the machines are not supposed to stand still and wait for their filter to come back.
The reference filter is lighter but since 40 filters are needed instead of 1.16 it needs over 21 times as
much materials.
Table 15: Material usage comparison; the weight of the reference filter and the concept filter
respectively both based on functional unit and individual filter
Reference filter Concept filter
Materials use for 20 000 h 55.3 kg 2.59 kg
Materials use per filter 1.37 2.23
90 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Figure 43 shows material content in each filter. It is obvious from the graph that the concept filter is
heavier and contains a large amount of aluminium.
Materials in one filter
2500
2000
Plastic
1500
Cellulose
]
g
[ Rubber
1000
Aluminum
Steel
500
Reference filter Concept filter
0
RefeCroenncceep ftilter Concept filter
Figure 43: Materials content in each filter. The reference to the left and the concept to the right
9.3.2.2 TRANSPORTS
Table 16 shows the distance each 20 000h of fuel filtering requires. The reference concept travels
more than double the amount of the concept solution. This is since each reference has to be
transported from Germany to Sweden whereas the concept only has to do that journey 1.16 times
and then it is only taken back and forth to Flen in Sweden.
Table 16: Comparison of the transports needed to provide the wheel loader with main fuel filtration
for 20 000 h.
Product Kilometres
Reference 76 800
Concept 32 496
9.3.2.3 IMPACT ASSESSMENT
The following graphs are showing the difference between the reference filter and the concept filter
based on the functional unit for all chosen impact categories.
GLOBAL WARMING POTENTIAL
As can be seen in Figure 44, the concept filter has minimal global warming impact compared to the
reference solution. Both have biggest impact during the production step, thus tied to materials
extraction and pollution.
91 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.