University
stringclasses 19
values | Text
stringlengths 458
20.7k
|
|---|---|
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Figure 44: Graph showing the global warming potential for the reference filter and the
concept filter
The largest impact caused by the reference is the process of extracting, producing and working the
containing steel. The largest impact caused by the concept is instead the extraction and manufacturing
of aluminium components. The emissions are in both cases carbon dioxide, tied to energy production
to feed energy into material extraction processes. Since most impact in both cases comes from the
production phase, it indicates it was right to focus on decreasing the material consumption per
functional unit. The graph also shows that the material reduction gave the desired, lowered impact.
ACIDIFICATION POTENTIAL
Figure 45 shows that the acidification potential is much smaller for the concept filter. Largest impact
has the production phase in both filters cases.
Figure 45: Graph showing the acidification potential for the reference and concept filter
It is somewhat surprising that the concept filter has high acidification potential compared to the
reference filter in the production phase despite the much lower weight. But aluminium production is
resource heavy and it is especially Sulphur dioxide from the production of the canister that shows in
the result. For the reference filter, it is emissions of the same kind that dominates and they are also
tied to material and production of the canister. The acidification potential from the use phase in the
92 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
reference filter case stem from emissions of nitrogen oxide caused by combustion of diesel due to
transportation.
EUTROPHICATION POTENTIAL
Figure 46 shows that production phase and use phase have similar amount of emissions that cause
eutrophication for the reference filter. It is, again, the steel production that causes emissions but also
the large amount of transportation that is required for the reference filter. The shape of the
eutrophication graph and the acidification graph are similar since they are results of the same sources
of pollutions. However, the numbers are relatively very small in both cases.
Figure 46: Graph comparing the eutrophication potential of the concept and reference filter
What makes the reference filter dominate is the lager amount of transportation by truck that emits
nitrogen monoxide. This causes emission in the concept fuel filter case too but the amount of
transportation is much lower. What also causes emissions for the concept filter is the detergent
production in the form of ammonium.
ABIOTIC RESOURCE DEPLETION
shows the difference in abiotic resource use. The concept filter uses very few resources compared to
the reference concept which is visual in this graph.
The use of resources is very spread over all kind of resources considered in this impact category. What
stands out is the use of borax, indium and colemanite in the process estimated for the cellulose
cartridge in the reference filter. However, this is an approximated process and if those materials are
used in reality is not known.
For the concept filter, it was the production of the canister that was most resource consuming and
the materials standing out was bauxite and tantalum. For the use-phase it was the production of
detergent used for remanufacturing that stood out using indium. In this impact category are fossil
fuels needed for production of aluminium and steel included.
93 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
Abiotic Resource depletion
0,3500
0,3000
0,2500
.v
0,2000
iu
q
E 0,1500
-
b
S
g
0,1000
k
0,0500
0,0000
-0,0500
Total Production Use End of life
Reference filter 0,3095 0,1508 3,79E-02 0,1208
Concept filter 0,0002 0,0004 0,0000 -0,0002
Figure 47: Graph showing abiotic resource depletion for the concept and reference filter
9.3.2.4 IMPACT ASSESSMENT CONCLUSION
Over all it can be said that the LCA has shown how the material reduction has led to reduction of
environmental impact. The new way of handling the filters has also led to a big reduction in transports.
What is worth paying attention to is the use of aluminium for the canister in the concept filter;
aluminium was chosen since it is a light material and it was desired to keep the mass low not to cause
unnecessary impact during transport. However, other impact categories show how aluminium causes
the biggest impact for the concept. In the refinement of the concept it is suggested that a throughout
stress test should be done on the canister to be sure how big the load on it will be and after that look
into if other materials are possible and what effects that would have on the transported mass.
The GaBi software, and associated database with the dataset available for the project, could not
provide a good number for water use. It would be desirable to be able to compare the reference filter
with its water consuming cellulose cartridge solution with the washable concept filter. However, the
water estimated to be needed for the concept filter is a very rough estimation and the water may also
be recyclable within the process. In Sweden, where the LCA was placed is water consumption not a
big problem. If the facility is placed in a country with not so abundant fresh water resources, this factor
must be better researched and could in fact have a big impact on the local area. This is somewhat a
result gap and it is something worth considering when moving forward with the concept development.
9.3.2.5 ROBUSTNESS TESTING
In the following section are the results from the sensitivity analysis presented. A sensitivity analysis
was made on the remanufacturing procedure and in the form of a breakeven analysis between the
concept and the reference filter.
SENSITIVITY ANALYSIS
In this part of the sensitivity analysis the idea was to increase consumption of all resources needed
during the remanufacturing process, including transports, since the very nature of the concept builds
on remanufacturing. The remanufacturing procedure was not described in detail in the concept
development section and the amount of ingoing resources was considered needed to be tested to
understand how an increase of resource consumption could play out if the procedure would be
developed. This was done by multiplying the remanufacturing process with 2 and 10. This way a
94 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
doubling of needed resources was represented with 80 washings instead of 40 and an increase with
10 times the amount was represented by 400 washes. Since transportation was also included it would
represent that the distance to the remanufacturing facility would double in the first case and tenfold
in the second.
Global warming potential
Figure 48 shows that even if all the factors in the remanufacturing process were multiplied by 10, with
energy consumption, transport, water use, spare parts and soap included, the concept filter is still
better for the environment, considering global warming potential. This is a good indication of that the
remanufacturing process is a better solution than disposable filters.
Figure 48: Graph showing how the global warming potential is affected when
the parameters of remanufacturing are changed
The biggest change between times two and ten is that the rubber production passes the production
of steel spare parts in terms of emissions. This tells us two things, the remanufacturing procedure
itself is not a large source of emission and the amount of spare parts needs to be kept low.
Sensitivity analysis and conclusion
The remanufacturing process was the biggest area of uncertainty in the model over the concept due
to the readiness of the concept. Values for what was thought to be needed were estimated and
because of that, it was interesting to see how the results developed when they were increased.
However, some of the factors multiplied by 400 are not likely to be that big, such as water use and
spare parts. On the other hand, it gave a good indication of what is most important to think about
when the remanufacturing process is developed.
It is worth refining the filter design until it requires as little spare parts as possible to keep the
environmental impact at a minimum. There is a lot of potential to develop a remanufacturing facility
which can allow for low impact if principles of reuse are implemented, for example could heat used
to wash and dry filters be used in the factory.
95 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
BREAKEVEN
A breakeven analysis was made to see how many times the concept filter needed to be re-used before
it was a better option than the reference filter in terms of greenhouse gases emitted. Figure 49 shows
Figure 49: Showing breakeven point between the concept filter, concept filter
with resources for remanufacturing multiplied with 10 and the reference filter.
the result from the breakeven analysis. The concept filter has a much higher amount of initial
emissions (5.9 kg) but the remanufacturing process emits very little. Due to this, the concept filter has
soon paid off its initial emissions and the reference filter, which emits 2.64 kg each, is the less good
option already after 2.25 replacements. To make the comparison more interesting the high resource
consuming remanufacturing concept was also included in the breakeven analysis (green line). If the
remanufacturing process is that expensive, resource wise, it will still take only four reference filters to
breakeven.
It might be surprising that the breakeven point is after so few exchanges. But considering that the
concept filter only weights slightly more than the reference, it is more understandable. It has already
been shown that the impact for both filters is in the production phase and that even if a concept filter
is remanufactured 400 times during 20 000 h it is still a better option. This indicates that the
remanufacturing process is not very resource requiring.
96 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
10 DISCUSSION
As a response to the project aim, the following chapter will discuss the use and sustainability centred
development process defined beforehand and utilised for the particular case. First, the outcome from
implementing the process on the case will be discussed, that is the final concept solution. Thereafter,
the implementation of process in the particular project is elaborated on. And lastly, the more general
learning outcomes are discussed, in the meta perspective section.
10.1 THE CASE AND THE FINAL CONCEPT
In this section, the final concept’s potentials and weaknesses are discussed. Furthermore, the
concept’s expansion possibilities and recommended next steps are covered.
The final concept is developed to suit a wheel loader in functional sales model, inspired by circular
economy. The main environmental advantage from such a model, compared to today’s situation, is
that the producing company is incentivized to optimize the service and maintenance. The concept
offers a part of the solution to reduce cost and environmental effects from service and the use of
consumables. In addition, the new design also gives incentive to take advantage of other principles,
inspired by circular economy. First, the concept opens up for more users to use the same product, it
allows for reuse, remanufacturing, repair and extended use. In the products end-of-life phase, it also
allows for easier recycling compared to the reference, since it is quite easy to disassemble and
separate materials. These are also materials for which there is a market to resell the materials after
recycling. All these factors might contribute to resource saving beyond what was shown with the LCA.
To add to this equation is that the 40 reuse cycles for the concept filter might be a conservative
assumption. A product life of 40 cycles was assumed based on that today’s suppliers of similar
solutions guarantee their products to last the lifetime of the vehicle. Thus, for the L150, and with
preserved replacement intervals, resulted in the 40 use cycles as the minimum requirement. Though,
it is reasonable to believe that the filters can be designed to endure more than 40 cycles and that the
intervals can be extended, hence giving raise to even larger economic and environmental savings.
A weakness with the concept is that more could probably have been done to resolve the underlying
problem to why fuel filters are at all required; referred to as: “eliminate or reduce the basic need”. The
underlying problems were identified as the fuel being contaminated and the injectors being very
sensitive. Though, these were not questions which suited the particular project very well, they are
interesting aspects to investigate in other projects. In this project, the scope somewhat directed the
process towards resolving symptoms rather than the actual problems. If the objective would have
been less focused towards generating a tangible product, it is believed that the problem could have
been addressed on a higher level of substitution. Thus, higher impact can be obtained in accordance
with the theories about levels of substitution by Holmberg (1998). As the project was carried out, the
final concept might contribute to a lock-in towards fossil fuel powered engines. Though, the solution
is at the same time providing guidelines for how environmentally better products can be developed
already today.
To go from product concept to final product, further analysis is required to determine the exact
structural properties of the product through dynamic fatigue tests, pressure analysis and possibly
optimizing the flow through fluid mechanic simulations of the system. Filtration capacity seems to
require empirical tests to guarantee sufficient performance. Thus, another step could be to carry out
these ISO standardized tests on filter prototypes. Next, complementary real life testing would be
appropriate, to ensure satisfactory performance in field. Including prototyping and trying out the
washable filters small scale, for instance together with individual customers with a large machine fleet.
97 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
It would be important in such a test situation to verify the solution with users to make sure that
requirements about easy use are met.
Making the reman solution economically and environmentally justifiable, including setting up a reman
facility, redesigning the product and establishing new business relations etc., will probably require
high quantities of filters. The concept allows expansion to include all sizes of fuel and oil filters in the
construction equipment machine catalogue. The solution, is also believed to be compatible with old
technology, for machines already on the market. Thereto, expansion to all diesel engines in the whole
Volvo Group is believed to be a possible next step to obtain high quantities; even though this will
include extensive reassessment of these new systems and the related requirements. Regarding trucks
for instance, other requirements are important, such as product weight and the accessibility in the
compressed engine room.
A few perks have been identified connected to the circular business model and the studied system,
from which the company can benefit. First, the model might facilitate in keeping customers loyal to
the brand, which is mentioned in theory by Tukker (2015) as a typical benefit from introducing service
system schemes. In addition, the close customer relation makes it easier to understand the customer,
and develop products, sufficiently adapted to the customer needs. The environmental responsibility
shown through reuse of products and a revised company view on consumables can also be used for
green marketing purposes to attract new and existing customers.
There is obviously a bit of a way to travel before the solution can be implemented. The next step could
be to specify the functional sales model more in detail and establish it internally in the company. In
addition, the business case should be explored in relation to the functional sales model. How can the
solution create revenue for the company and what is a reasonable time horizon? After that, a next
step could be to reach out to companies owning the filter technology. However, even more important
is to strengthen the relation to today’s service provider, which possesses the real customer insights
and understanding of the service systems. To ensure that the user considerations, including physical
and cognitive ergonomic aspects, are being satisfactory met, the solution should be evaluated with
physical user evaluations with a physical product representation; as described in the later sections of
the product development framework, in chapter 3.
Even though results are promising, the concept is just a small piece in the transition to a circular
business model. Even so, the solution is a bridging technology towards truly sustainable ways of
making business and as a good case study in how it can be achieved. The solution can be implemented
quite easily already in today’s system, since much of the required infrastructure for the
remanufacturing is already in place and the technology behind the concept is available.
10.2 THE DEVELOPMENT PROCESS – IMPLEMENTED
Under this headline, the implementation of the process framework will be discussed. The section will
elaborate on which aspects were successful and which were less useful in the particular case, and the
reasons to why this was the case.
The demand for a pre-planned project forced the design process to be quite explicitly defined already
before the prestudy. This might have caused a problem due to the unknowns regarding what the
reference product could be. It showed that a user centred process was not entirely suitable for the
type of product selected, with a quite limited user contact throughout its operating life, and some
process alterations were necessary, for instance extensive user interviews and user tests were
excluded.
98 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
The idea of using criteria was to guide the process in a desired direction, considering all stakeholders
and thereafter help in selecting paths and solutions to ensure the project would have a clear end-
result, a tangible concept which would allow for evaluation, without constraining the process, as can
be a risk when using product requirements. However, in this case also the criteria seem to have
directed the process in a slightly undesirable direction, towards a too narrow solution space. The
weakness in the method seem to be related to the chosen criteria and the weighting, rather than the
principle itself. The constraining criteria regarding the company compliance, completion level and
technical feasibility could probably better be introduced later in the process to maintain an open
solution space in the early ideation stages.
Because the criteria were used in the design process the new design was naturally anticipated to be
better, when evaluated against the very same criteria. Using the criteria as guides in the process has
led to the desired outcomes most times; however, it makes the criteria very influential and if aspects
are missed out in the criteria, these aspects might be entirely lost in the process. This stresses the
importance of carefully considering the criteria, the need to anticipate their contribution and possibly
reassessing them. An example of such missed aspects might be economic feasibility; which has led to
the fact that no conclusions about the business case and the economic sustainability can be made.
When it comes to the scenario, it is believed that it must be better rooted in the organization and the
company’s long term objectives and strategies, in order for the method to be truly successful. This
conclusion is somewhat in line with what is suggested in Bocken et al. (2016); referred to as the
“integrative perspective”, saying that companies need to have a clear overall vision before product
and business model can be designed to become more circular. In this particular case, this could not be
done due to that concrete visions and strategies could not be communicated in detail by the company.
The scenario is defined as ten years from today and the result of this was a context and product
description very close to the one of today. This scenario was considered reasonable at the time.
However, in hindsight it could have been better to try looking beyond the anticipated technical horizon
and try to foresee what would actually be established on the market in ten years from now. This would
have been an advantage when considering that a product in its design stage today, will have a lead-
time of a few years before it reaches the market. A possible definition of the wheel loader then could
have looked very different; involving the concepts of electric drivelines and autonomous or remote
driving for instance. Followed by a result, more in line with that technical development. More insights
about the company’s current research engagements could have pushed the project in such direction.
The initial idea was that the development of the product concept and the scenario were to be done
simultaneously, a method supported in Bocken et al. (2016), claiming that development of business
model and design must go hand in hand. Although, the project showed that there might be a risk that,
by doing so, an ideal case is built; in which the product and the scenario converges, resulting in a
seemingly successful concept, even though the scenario, and thus the concept, might have lost its
connection to reality during the process. As for the criteria, it seems that the scenario had a major
influence over the process and the end-result in the particular case, again stressing the need for
careful considerations and trying to forecast the consequences when such methods are utilized.
Another problem with designing for a scenario is that reality might develop in entirely different
directions. For this case, for instance if the functional sales scenario is never introduced full scale, the
final concept will fall short. On the other hand, learning outcomes can still be obtained regarding life
cycle thinking and important aspects to consider in relation to this. And despite the scenario might
never occur as described in the case, the concept can still work for alternative scenarios; even though
the machine is not part of a functional sales model, the company can still offer filter take-back and
99 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
remanufacturing of used filters to their customers as a complement to the existing use-dispose
solution.
Another relevant question raised during the projects was how close the scenario should be to the
current situation. This is a typical trade-off between implementability and innovation i.e. how soon
can the solutions be implemented, versus how big of an improvement can be obtained. Even though
improvements have been proven short term a lock-in is hindering the concept from being entirely
justifiable from an environmental point of view. That is because the final concept will induce
investments, both monetary and in resource usage for a setting up a remanufacturing facility which is
a draw-back, considering the aim to achieve an entirely sustainable solution over time.
One very important factor leading to the extensive resource reductions was the decision to establish
an environmental objective early in the product redesign process; a strategy which is also
recommended by van Nes and Cramer (2006). In this case, the objective was to focus and prioritise
resource efficiency. Since the studied reference product was found to be a passive product, not
resource consuming when used, the material resource efficiency focus was appropriate. However, an
absence of tools was experienced, when it came to supporting the designers in which decisions are
extra important to consider in relation to the different product types. Especially, relevant guidelines
or recommendations would have been very helpful, since it was quite difficult to pinpoint some of the
less intuitive contributions from the use-phase during the development. Such as in the studied case,
additional transports and maintenance.
Another very influential factor behind the satisfactory result was the level of abstraction on which
improvements where suggested. Major impact could be obtained by starting on the highest level, and
redesigning the goals of the system, in this case the incentive for the company to reconsider their view
on consumables. Followed by the design of the system level flows, through the introduction of
remanufacturing of used products in our case. And further on to the lower levels, designing the
functionality of a product in such system and lastly the tangible product attributes to fulfil the
corresponding functionality requirements. Thus, major improvements could be obtained, even though
only minor alterations were made to the product; to summarise, after redesigning the system and
introducing the remanufacturing scheme, the functionality level design was focused on supporting the
alternative system and the functionality was in in turn supported by the product level design
suggestions. The reason to why solutions are considered on different hierarchies is the previously
presented theories about levels of substitution (Holmberg 1998); discussing the effectiveness of a
change in relation to the level on which it is implemented.
The ecodesign, or Design for environment mindset (DfE), discussed in Tingström (2007) was believed
to be the most important factor in achieving environmental impact reductions. The mindset included
constantly tweaking tools towards an environmental focus. Thereto, the mindset promoted a
continuous consideration of the life cycle perspective in the decision making which, amongst other
benefits, gave the silent stakeholders room to influence. Also, the fact that the scope of the project
quite clearly stated resource efficiency as the prioritized environmental aspect made an important
contribution to the mindset and had a major influence on the final result.
Identified ecodesign tools were used to the best of our abilities and are believed to have affected the
result positively. However, the focus of the tools used were on ideation and the final evaluation of a
well-defined concept. The process was lacking tools to support continuous decision making; for
instance, quickly weighing use-phase aspects to contribution from production for a number of impact
categories. Such tools do not have to be numeric, but rather just guide the decision-making towards
more sustainable alternatives and help the designer in prioritizing actions. What is requested, is a “Life
100 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
cycle based environmental impact assessment for the functionality in product systems”. The absence
of such tool box is identified and problematized around by a number of authors (Tingström 2007,
Bocken et al. 2016). However, since user centred design and LCA, both revolve around functionality,
the authors of this thesis point out that the idea of such a tool coming to existence is not unrealistic.
Assessing a concept with LCA comes with a lot of challenges and might at first sight seem contradictory
to the very nature of an LCA. A concept is ambiguous and uncertain, not much is decided and even
less is known compared to what is generally needed to be in an LCA where every process, material
and matter of transportation needs to be known in detail. However, involving LCA in the product
development process gave some benefits; since it addressed aspects such as materials choice,
material thicknesses, and transportations. However, an experienced LCA analysist could probably
better have decided at what stages and in what areas the LCA could be allowed to be less accurate;
that is, only at the spots where the system knowledge and available data can support this
accurateness.
Conducting an LCA on the final concept in this project could be compared to conducting an LCA on
circular economy, on small, conceptualized scale. The conducted assessment in this project clearly
indicates that circular economy has advantages over linear economy when it comes to environmental
impact, which is mostly tied to the major decrease in materials use over the life of the wheel loader.
However, the reason to why the concept uses less materials is tightly bound to principles of circular
economy being used in the development process.
Even though the final concept showed major material resource saving, the end-result can be
questioned; not as much in terms of the project aim and scope, as when it comes to long lasting
sustainability improvements. One reason to why the target was somewhat missed might have been
due to the process being constrained by a few factors. First, the limited access to information and the
studied system, due to the distance between Gothenburg and Eskilstuna; hence the possibility to
obtain a complete picture of the system, related problems and the direction of the technical and
technological advancements. Secondly, the type of internal knowledge that was obtained from the
interviews was quite limited to certain areas, due to proximity to people with these types of
competences and knowledge; which in turn forced the project towards these areas. Thereto, the
access to real users to observe and interview was quite limited, which is believed to have affected the
outcome negatively. Lastly, the project experienced a limitation from the process and the predefined
end-result being quite explicitly defined, even before the project commenced; the predefined process
and end-result offered little possibility to change directions when it was discovered that the project
was turning towards designing lock-ins.
10.3 THE DEVELOPMENT PROCESS – META PERSEPECTIVE
Under this headline, the project result is discussed from a more holistic and general point of view and
the findings are placed in an extended context.
The framework was developed from material from a number of different sources, which is considered
a strength. However, no contradictory theories were found, criticising the used theories. This might
suggest that the referenced sources make up a too narrow base for conclusions around whether the
framework is supported in theory. It can only be concluded that the framework was quite successful
in the very specific case of this study.
It seems that much of the successfulness in the case is somewhat a result of the very good match
between ecodesign and a user centred design. First, the system perspective promoted by the use
centred process, especially when complemented with the life cycle perspective, seem to have had an
101 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
important contribution to the resource savings. And just like our hypothesis predicted, the idea of
prolonging product life or sharing a product leads to a higher use-rate per individual product. This
means that, even though the initial system has no or weak links to a user, a maintained user
perspective is very important and potential users must be anticipated; such as shared-use users,
secondary user, maintenance personnel, remanufacturing mechanics etc. Since these new users might
have a major influence over the product’s life cycle performance and environmental impact, user
centred product design is crucial. First, to ensure they are actually being used at all; and hence,
utilization being optimized. And secondly, that the products are used correctly and responsibly, to
optimize use efficiency. Otherwise, trade-off effects can quite easily weigh up for the expected
resource savings. A system perspective is argued in Bocken et al. (2016) to be an important key to
avoid such undesirable effects.
In alignment with Tingström (2007) the project has acknowledged the difficulties in using LCA early in
the product development process. One of the problems, which was touched upon previously, is the
fact that early concepts are surrounded by many uncertainties; the tangible attributes are not yet fully
determined, to the level which is required by the LCA methodology as it is described in ISO 14 040.
Another reason to why LCA is not appropriate in the early product development stages, addressed in
theory and confirmed by the project, is that a vital parameter in supporting innovations is for methods
to yield quick results, and to promote fast decision-making (Tingström 2007). The project also
experienced that, in order to make relevant estimations, handle data gaps and to find the right level
of the assessment, a quite high level of knowledge was required; both in the studied domain and about
the use of LCA. The way LCA was used in this project, the method contributed to a focus on lower level
improvements when forcing to define product attributes quite explicitly; while the process, still
promoted a focus on higher level substitutions.
Even though the solution is better than the reference, the solution can be argued to have slightly
missed the goal of being entirely resource efficient over an extended time horizon. The solution is
creating a potential revenue stream from a system built around a fossil fuel dependent product and
which in addition build on a system relying on transportations to a high extent. This obviously raised
some thoughts about the designer’s role and ethics:
- What responsibility do the designers have in ensuring the environmental aspects remain a
priority also after the concepts are handed over to the company for further development and
launch.
- How can designers make sound decisions about when investments in additional use of
resources are justifiable; considering trade-offs, technical advances, and lock-in effects?
Example: Pose the revenue from remanufactured fuel filters and the investments in the
suggested remanufacturing chain will be one, even though a small, contribution holding back
the transition to more environmentally sustainable drive line. A possible consequence of such
lock-in effect would be a substantially negative influence on the environment.
- A related question will be, when we can allow ourselves to wait for anticipated technical
advances and when we are better doing something to reduce environmental impact in the
current system? Even though the changes might risk creating lock-in effects? Perhaps
prognostic LCA’s can be utilized to support in this kind of decision making?
- Another question raised during the project, on the topic of substantial and long term
improvements, is how we can incorporate the business model as a factor in the development
process, to ensure the results will be truly resource efficient systems? To close in on the
answers to these questions, further research is requested.
102 |
Chalmers University of Technology | Preparing for tomorrow: Exploring design adaptations of a wheel loader for a circular business model
Bergstrand, H & Jönsson, C
11 CONCLUSION
As a response to the project aim: “to explore the use and sustainability centred product development
process”, the case stressed the potential to obtain real environmental impact reduction, quantitatively
verified with LCA, from a development process guided by a pronounced focus on use and sustainability
aspects. Over the machine’s operable life time, the final washable filter solution is expected to save
approximately 95 % of the material usage, 60 % of the transportation distance and about 90 % of the
total impacts on global warming.
The use of circular economy as inspiration and guidance, resulted in a concept with verified
environmental impact reductions, compared to the reference product in the linear business model.
This fact brings credibility to that the concept of circular economy and in particular, the principle about
material resource efficiency, has the potential to lower the environmental impact from consumption.
As anticipated, the circular business model functional sale, in this case led to an increased use rate per
individual product; which stresses the importance of user considerations in the product development
process, including both existing and anticipated, new users. In addition, use centred design is essential,
in order to ensure that the increased usage per products is as resource efficient as can be.
The final concept’s successfulness is also believed to be a result of the fact that the solution considers
the entire system. Since, according to the theories about levels of substitution, changes are more
efficient the higher up in the system hierarchy they are implemented; from the overall system goal
and flows, to the functionality and lastly to the tangible product attribute level. The systemic
perspective is promoted from both the user design and circular design perspectives, which further
emphasises the strengths of a combined user and sustainability centred perspective.
In the particular project, it was deemed useful to define an environmental objective early in the
process in order to manage contradictory environmental aspects. In this case, the overall
environmental objective focused on material resource efficiency. The case could also conclude that
both the scenario, used to manage the situation of designing for a not yet existing future, and the
evaluation criteria, used to maintain relevancy were very influential over the process; hence needs to
be carefully considered and well aligned with the company vision and strategy, in order not to steer
the project off in undesirable directions.
In addition, the use of what is referred to as the ecodesign mindset seem to have contributed to the
positive case result; as it stresses environmental aspects in every decision. However, the absence of
easy-to-use design tools, to make use of the mindset and promote sustainable alternatives was
acknowledged. The project has, in alignment with product development theories, also concluded that
using LCA according to the ISO-standard, is not efficient as a concurrent design support tool. However,
the project can recommend LCA to be used on existing product systems to address current issues and
raise awareness of the life cycle perspective. Results from such LCAs are recommended to be part of
design, or re-design, processes to reduce impact.
It can be concluded that by introducing the final concept, major reductions can be obtained in all
impact categories in the analysed scenario. Although, some concerns remain: the environmental
impact reduction risk to be negatively affected, depending on how the company decide to proceed
with the concept. This is because the concept risks creating lock-in effects from an environmental
point of view, due to that the generation of revenue is dependent on a fossil fuel based system.
103 |
Chalmers University of Technology | APPENDIX II - STAKEHOLDER ANALYSIS
Stakehol Motivation Method Constraints Risk Relation to product Influence on machine
der design.
Linear Circular Linear Circular Linear Circular Linear Circular Linear Circular Linear Circular
e n w O /r e y u Br e n il p e e K .g n in n u r e n il p e e K .g n in n u r /e m itp u h g iHy tiv itc u d o rp /e m itp u h g iH y tiv itc u d o rp ,s n w o d k a e rB e c n a n e tn ia ms ts o c ,s n w o d k a e rB h g iH w o L m u ie dn eo Mn - m u ie dn eo Mn - m u idh eg Mih - m u idh eg Mih -
r o ta r e p o /r e s U .k s a t e te lp m o C e. lk ps ma t o e Ct /s llik s g n iv irD tc u d o rpe c n a m ro fre p /s llik s g n iv irD tc u d o rpe c n a m ro fre p s n w o d k a e rB.g n it dia nw a s n w o d k a e rB.g n itia w d n a w o L w o L e s o lC e s o lC m u i wd e oM L- m u i wd e oM L-
/o v lo V n a lp s s e n is u B .e n ih c a m lle S e n ih c a m tu o tn e R y tila u q h g ih p o le v e D . ss etr na ip h cd an ma e n ih c a m fo n o itc n u f lle S ro f ts o c e c u d e r d n a .e c n a n e tn ia m d n a e c iv re s s n w og dn ki ad ea ro e Blt fo k c a L .n o ita tu p e r d a b y fs ita s o t n o ita m ro fn i.s d e e n ’s re m o ts u c o t s d a e l ta h t s n w o d k a e rB fo k c a L .n o ita tu p e r d a b y fs ita s o t n o ita m ro fn id e e n ’s re m o ts u. cs m u id e M. .h g Hi m u ie ds eo Mlc - m u ie ds eo Mlc - .g iB .g iB
a
m
e ld d im /r e la e Dn
d
n a e n ih c a m lle S.e c iv re s th g ir e h t d n iF e h t ro f e n ih c a
m.n
o ita c ilp p a th g ir d n a ts re d n U.s d e e n re m o ts u c d n a ts re d n U.s d e e n re m o ts u c fo k c a L.n o ita m ro fn i fo k c a L.n o ita m ro fn i m u id e M m u id e M e s o lC e s o lC m u id e M m u id e M
e c iv r e Sle n n o s r e p
n o ita
tu p e r p e e K h g ih h tiw.e m itp u
n o ita
tu p e r p e e K h g ih h tiw.e m itp u
e c iv
re s ra lu g e R.s la v re tn i e ru tru N .s n o ita le r
ta
e b y lk c iu Q n e h w e tis.d e d e e n .tx e tn o
C.n o ita
c in u m m o C .tx e tn o
C.n o ita
c in u m m o C m u id e M w o L e s o lC e s o lC
m
u i wd e oM L-
m
u i wd e oM L-
y
s r o ts e v n I y lla c im o n o c
E.y
n a p m o c e lb a n ia ts u s y lla c im o n o c E
.y
n a p m o c e lb a n ia ts u s
e
n
o m n ra e ,ts o c e s a h C te k ra m re tfa n o.s tc u d o rp g n o l d n a e lb a ru D o t s tc u d o rp g n its a ln w
od
dn a d e iom vi a-t .s ria p e r tn e m p o le v e d h g iH.s ts o c e v t en de m h gp io Hl .s ts o c h g iH h g iH e n o N e n o N .g iB .g iB
s s
k k
tn e m
n
o
fo
tn u o
m
an is
d n a
s e
c
fo
tn u o
m
an is
d n a
s e
c
mth g ih
-
r iv n E - - - - tim iLru o s e r tim iLru o s e r e n o N e n o N h g iH h g iH u id e m w o L |
Chalmers University of Technology | APPENDIX IV - COMPONENTS AND MATERIALS DECLARATION
Ref. Component name Primary function Secondary function (s)
Filter body
9 canister contain filter contain diesel
prevent particles from entering
withstand pressure
support filter
act as handle/grip when detaching/attaching
protect filter
provide surface for information/instructions
8 spring allow over pressure keep filter in place (centre)
bypass
2 retainer sheet keep housing and seal housing and machine interface
interface together lead diesel
provide attachment of gasket
Interface
3 interface plate connect filter to machine seal fuel system downstream
prevent copying
spread fuel
provide structural integrity to design
1 gasket seal fuel system put "pressure" on threads to lock
(upstream-outside)
- small gasket seal fuel system -
(upstream-downstream)
Filter cartridge
6 Cellulose filter material mechanically filter out -
particles from fuel
- Glue/ "sealing foam" seal between up- and glues metal capping discs to cellulose cartridge
downstream
5 Inner tube support filter Transfer force to spring in case of over pressurized
container
4 Top cartridge capping disc support filter keep flanges apart
prevent leaking of contaminated fuel to clan side
of filter
transfer pressure to inner tube in case of over
pressurized container
7 Bottom cartridge capping support filter keep flanges apart
disc prevent leaking of contaminated fuel to clean side
of filter
Related components
- Dust cap prevent particles from stay on place during distribution
entering before be recyclable
installation inform about recyclability
- Transportation packaging Protect filter during Inform user
transport
Coating/Paint
- White paint protect casing express Volvo’s graphic identity
express Volvo genuine component
- Blue inscriptions inform user about use inform about after life treatment
and installation express Volvo’s graphic identity
express Volvo genuine component
inform about manufacturing
ensure product tracing by printed serial no. and
bar-code |
Chalmers University of Technology | APPENDIX VI - PERSONA AND USE-SCENARIO
Jonas,
a 39 year old service technician from Uddevalla, currently
employed at Swecon, Mölndal. Right after his graduation from high
school as a heavy vehicle mechanic he got a job at a small workshop in
his hometown. A couple of years later he met his wife, Sandra, and soon
went to live with her in an apartment just outside Gothenburg city. Jonas
decided to take the opportunity to get the most out of his education in
heavy machinery and applied for a new job as a mechanic at Volvo BM,
heavy, off-road equipment centre in Mölndal, where he has stayed ever
since, however moved on to become an on-call service mechanic for the
same Volvo service center. In their free time, Jonas and his wife spend a
lot of time with projects in an old villa, which they bought and moved into
almost ten years ago. Over the last couple of years, most weekend for
more than half the year are spent in their camper van, travelling
Scandinavia, for their two daughters, Nellie and Eleonore, 11 and 15 to
compete in the motocross series.
The reason to why Jonas has remained at Volvo during all these years has
much to do with the quality of genuine Volvo equipment; the quality
makes his life easier as a technician and the meetings with proud Volvo Figure: Persona illustration
customers is a rewarding part of his job. Also, he appreciates the variety
in his work, with different machines, different tasks and different clients; he seldom knows what the next day
has to offer. The varying tasks and contexts also comes with some down sides; the working conditions are
sometimes not ideal. An October morning in a muddy quarry, cold, dark and wet and with rain just pouring down
is obviously not the most pleasant conditions for a service. Neither is it easy to keep the delicate filters clean
when changing them a hot, dry day in June, where the dust whipped up by the wind, from conveyors and from
off-road equipment moving around the site is almost impregnable. If you are lucky though, the customers have
made room for and parked the machine in their workshops, but that’s more of an exception than rule. Sometimes
the filters leak too and everything gets messy; the clothes, the machine, the toolbox and tools, spare part and
even the back of the car becomes contaminated with fuel and oil. The washing machine back in the workshop
really saves the day. When thinking about it, Jonas is surprised nothing has been done to improve the handling
of the filters during all these years; it seems filters have remained more or less the same for as long as he can
remember.
Unlike some of his older colleagues he can really see the advantages from the recent shift towards more
sophisticated components and the use of software to determine the machine’s status and search for problems.
Jonas believes it has made most jobs faster and with pleased clients as a result. He is really looking forward to
the remote failure detection system, which he has read is on its way; it could help him be even more prepared
when arriving to the Volvo owners in the area. After working with heavy equipment for almost 20 years his body
is not what it used to be. Nowadays he rather work on machines where it is really easy to reach; climbing around
on a two-story high wheel loader is something he leaves to his younger colleagues. Jonas is not afraid to get his
hands dirty for the right reasons. However, to risk shortening his life due to diesel fumes and diesel and oil spill
on his skin is definitely not something he is prepared to do. Even though the working conditions have improved
by miles during the last decade, he still feels improvements could be made if only the equipment designers would
listen to the expertise from the field. Though, he is proud about how far his workshop has come in terms of
environmental responsibility. Separating hazardous materials such as filters, spray cans and paint buckets, used
oils and other liquids etc. in different containers for recycling feels really good. What happens with the filters
after the containers have been sent away on the other hand, is not something he has thought much about. |
Chalmers University of Technology | APPENDIX VII - RAPID ECO ASSESSMENT, COMPLETE RESULTS
Eco strategy wheel consideration Rapid assessment Comment
1 Design for: Innovation
1,1 Rethink how to provide the benefit Not yet considered More can be done
1,2 Design flexibility for technological Not yet considered More can be done
change
1,3 Provide product as service Not yet considered More can be done
1,4 Serve need provided by associated Not yet considered More can be done
products
1,5 Share among multiple users Not yet considered More can be done
1,6 Mimic biological systems Not yet considered Major system changes and/or
new innovations required
before improvements are
possible
1,7 Use living organisms in product Not yet considered Major system changes and/or
system new innovations required
before improvements are
possible
1,8 Create opportunity for local supply Has been considered, to some More can be done
change extent
2 Design for: Reduced material impact
2,1 Avoid materials that damage human Has been considered, to the Major system changes and/or
or ecological health best of abilities new innovations required
before improvements are
possible
2,2 Avoid materials that deplete natural Has been considered, to the Major system changes and/or
resources best of abilities new innovations required
before improvements are
possible
2,3 Minimize quantity of material Not yet considered More can be done
2,4 Use recycled or reclaimed materials Has been considered, to some Considered due to economic
extent incentive. More can be done
2,5 Use renewable resources Has been considered, to some More can be done
extent
2,6 Use materials from reliable certifiers Has been considered, to the -
best of abilities
2,7 Use waste byproducts Assessment pending More information required
3 Design for: Manufacturing innovation
3,1 Minimize manufacturing waste Has been considered, to the Considered due to economic
best of abilities incentive. Major system
changes and/or new
innovations required before
improvements are possible
3,2 Design for production quality control Assessment pending More information required
3,3 Minimize energy use in production Assessment pending More information required
3,4 Use carbon-neutral or renewable Assessment pending More information required
energy sources
3,5 Minimize number of production steps Assessment pending More information required
3,6 Minimize number of Not yet considered More can be done
components/materials
3,7 Seek to eliminate toxic emissions Assessment pending More information required
4 Design for: Reduced distribution
impact
4,1 Reduce product and packaging Has been considered, to the Considered due to economic
weight best of abilities incentive.
4,2 Reduce product and packaging Has been considered, to the Considered due to economic
volume best of abilities incentive. |
Chalmers University of Technology | 4,3 Develop reusable packing systems Not yet considered More can be done
4,4 Use lowest-impact transport systems Not yet considered More can be done
4,5 Source or use local materials and Not yet considered More can be done
production
5 Design for: Reduced behavior and
use impacts
5,1 Design to encourage low- Not yet considered More can be done
consumption user behavior
5,2 Reduce energy consumption during Not yet considered More can be done
use
5,3 Reduce material consumption during Not yet considered More can be done
use
5,4 Reduce water consumption during Not yet considered More can be done
use
5,5 Seek to eliminate toxic emissions Not yet considered More can be done
during use
5,6 Design for carbon-neutral or Has been considered, to some More can be done
renewable energy extent
6 Design for: System longevity
6,1 Design for durability Not yet considered More can be done
6,2 Design for maintenance and easy Not yet considered Not regarding the filter
repair design, More can be done
6,3 Design for re-use and exchange of Not yet considered Not regarding the filter
components design, More can be done
6,4 Create a timeless aesthetic Not yet considered More can be done
6,5 Foster emotional connection to Not yet considered More can be done
product
7 Design for: Transitional systems
7,1 Design upgradeable product Not yet considered More can be done
7,2 Design for second life with different Not yet considered More can be done
function
7,3 Design for reuse of components Not yet considered More can be done
8 Design for: Optimized End of Life
8,1 Integrate methods for used product Not yet considered More can be done
collection
8,2 Design for fast manual or automated Not yet considered Not regarding the filter
disassembly design, More can be done
8,3 Design recycling business model Has been considered, to some More can be done
extent
8,4 Use recyclable non-toxic materials Has been considered, to some More can be done
extent
8,5 Provide ability to biodegrade Not yet considered More can be done
8,6 Design for safe disposal Not yet considered More can be done |
Chalmers University of Technology | APPENDIX IX - INITIAL 8 FUNCTIONALITY CONCEPTS DESCRIPTION
Concept 1 - Replaceable Cartridge
General description
The concept is based on that only the filter media itself, the cartridge (Figure 1),
is replaced during the service procedure. To allow cartridge replacement, filter
design changes are required. The contaminated cartridge in sent for incineration
and energy recovery.
Advantages
• All materials are separable.
• Less material use, only filter material is replaced.
• High realization level, been done before.
• The disposed part can be used for heat recovery. Figure 1: Replaceable
cartridge concept
Limitations
• Replacement must be done on site with changing and challenging working environments.
• Might be difficult to maintain cleanness during exchange of filtering media.
• Increased risk for contamination of technician from hazardous liquids and fumes
• Require additional manual work.
Unique elements
• The replacement of the filter material allows a good insight in the performance of the filter
thanks to that the old cartridge can be scanned while replaced.
• Contains a disposable unit.
• Require no changes to fuel system.
• Allows for personalization of the casing; since it stays on the same machine.
Concept 2 - Twin-filters
General description
The idea behind the concept is to save material by having several product sharing
material; i.e. less material to be scrapped for each service. For this concept, the
suggestion is for the pre-filter and the main fuel filter to be integrated in the same
canister (Figure 2 ) by redirecting the flow from out-in, out-in to in-out, out-in. The
solution also ought to reduce service time since both filters are changed in one
replacement.
Advantages
• Has the potential to be somewhat material saving.
• It might reduce service times.
Limitations Figure 2: Twin-filter
• The principle is not used today, which might indicate that the technology is concept
non-existent or inefficient.
• There is a risk for a two-in-one solution it is important to be more complicated to install or
heavier to lift.
• Combined functionality might cause reversed performance synergy. (1+ 1 < 2)
• The solution only has limited material saving potential.
• Difficult to implement the filter type on existing machines.
• Both filters must be replaced at the same time. |
Chalmers University of Technology | Unique elements
• It combines two functions in one.
Concept 3 - Docking station
General description
The machine is docked to a station overnight (Figure 3),
where filters are cleaned, hence the machine requires.
Thereto, the machine is refuelled with pre-filtered diesel
from the docking station, to avoid contamination.
Advantages
• Filters does not have to be replaced.
• Diesel thefts can be avoided since the tank is
always empty when the machine is not in use.
• The unit can be shared between all machines on Figure 3: Docking station concept
site.
• It's a Swiss-army knife device, which has the potential to handle multiple issues in future
applications.
Limitations
• The docking station needs to have more functions in order for it to be economically
justifiable.
• There must be many machines in the fleet, sharing the same docking station for the solution
to be efficient.
• The technological feasibility for such solution is yet uncertain.
Unique elements
• The solution can work as a bridging technology for electrical wheel loaders.
• It's a technology that is stationary instead of carried by the machine.
• It is based on shared function principles.
• It is ought to treat the underlying issue with poor refuelling hygiene and diesel quality.
Concept 4 - Recyclable Design
General description
The solution ought to improve both the recyclability and the actual recycling
rate, by considering the design for recyclability principles to the full and
innovate solutions to improve return rate (Figure 4).
Advantages
• Less contamination in the recycling process.
• Less material waste.
• The process of changing filters remains and no additional training of
the technicians is required. Figure 4: Recyclable design
concept
• Principles are relatively easy to implement on current design.
Limitations
• The amount of filters, and the amount of materials disposed is still the same.
• The principle counteracts reuse or repurposing. |
Chalmers University of Technology | • Existing filter design and development is owned by suppliers.
Unique elements
• This concept requires the least changes in system and product design, technology and
logistics.
• Concept also aims to change behaviour; making it more likely that the filter is actually
recycled.
Concept 5 - Filter remanufacturing
General description
Used filters are collected and sent off to a central
remanufacturing facility where the filters are brought
back to as good as new standard, before sent back to
the service centres (Figure 5).
Advantages
• Most of the material is reused
• Filters are cleaned in a secure environment.
• Can be made profitable and expanded to all
Figure 5: Remanufacturing concept
filters.
• Does not require additional work for technician
• The company has experience in the field of remanufacturing.
• Filters are similar which allows for an automated process.
Limitations
• The solution is transportation dependent.
• An unknown buffer of filters is required to keep the loop running.
• Setting up the reman-facility and the logistics chain requires major initial investments.
Unique elements
• A testing procedure can guarantee performance since the remanufacturing is done at a
plant.
• Almost the entire filter is reused without any extra job for the technician.
Concept 6 - Condition based replacement
General description
The fuel filter is changed when the filter performance is
insufficient, instead of within a predetermined time interval.
The need for replacement based on monitoring data from
either an external analysis equipment or existing differential
pressure readings (Figure 6).
Advantages
• Longer useful life reduces overall material usage.
• Less filter replacements are required; which implies Figure 6: Condition based replacement concept
reduced cost for labor and material and reduced total
down-time.
• Solution also provide a status report of the machines condition. |
Chalmers University of Technology | Limitations
• The solution relies on additional technology.
• Needs to be implemented in multiple devices within the machine to be fully effective.
• Replacement should be done by operator to avoid extra traveling for technician.
• The condition determining technology needs to be accurate and reliable.
Unique elements
• Filters are only changed when actually needed.
• Solution might provide additional information about the condition of the machine.
• Have potential to help avoid breakdowns before they occur.
Concept 7 - Local filter cleaning
General description
The contaminated filter is brought back to the service centre where
it is disassembled and put in an industrial dishwasher (Figure 7).
Filters are then reassembled and stored until next service.
Advantages
• A washing machine already exists in the workshop which
could be expanded/redesigned to allow for filter washing.
• No disposable parts which is material saving.
• Exchange procedure will be similar to today’s procedure.
• Substantially reduced transportations.
Figure 7: Local filter cleaning concept
Limitations
• Buffer filters are required, since the number of machines in service centre cover area is quite
small.
• The workshop needs to be able to treat the wastewater.
• Uncertainty whether the technology is available and affordable.
• Additional devices might need to be used to determine the cleanness and guarantee the
performance of the washed filters.
• The number of filters in each workshop might not be enough to support a fully automated
procedure, which will in turn increase the need for manual labour.
Unique elements
• The filter is made reusable locally.
• It might take advantage of technology that already exists in the workshops.
• Could make the service centres self-sustaining when it comes to filters. |
Chalmers University of Technology | APPENDIX XI - SWOT; THREE FUNCTIONALITY CONCEPT
SWOT; Cartridge concept
Strengths
• All work is done on site
• Only transportation of cartridge necessary, i.e less weight needs to be transported
• Result in material savings, since only the cartridge is disposed and replaced
• Cartridge can be incinerated and heat can be recovered
• The concept builds on existing and working technology
• No changes have to be done to the engine or fuel system to implement the filter solution
• Require skilled labour, which today is used by the company as a marketing argument for societal
sustainability.
Weaknesses
• Replacement induces a messy procedure
• Solution requires warehousing of cartridges, compared to other concepts.
• Workshop needs to keep handling hazardous waste (as done today), which is costly.
• The procedure risk exposing the technician for hazardous substances.
• Requires handling of hazardous and flammable waste
• Value recycling impossible due to contaminated material (incineration and heat recovery
possible)
• Performance can not be ensured
• More steps in the exchange procedure
• Performance is dependent on the quality of the installation performed by the technician
• Solution build on disposing principles rather than reuse
Opportunities
• Organic/Bio-based materials can be used for the disposable content.
• The solution has been used before, hence compatible with known technology
• The solution does not require big volumes to still gain environmental benefits
• Allow for a more expensive product; with more features, such as sensors and better design for
easier disassembly and longer exchange intervals
• Could expand to oil filters and fuel filters on all construction equipment (and trucks and busses
etc.)
• Technicians have full control over the condition of the fuel filter and can acknowledge
performance variations.
• On-site recycling might be possible to further reduce transported weight.
• (Since major investments are not required, there is little risk to build a system around a solution
which might not be relevant for future businesses.)
Threats
• Hard to ensure cleanliness during change due to challenging work environment.
• Risk of harmful diesel and grease contamination of equipment, tools and person.
• No performance testing after replacement of the filter is possible.
• Suggested technology goes transverse what is done today.
• The principle is not commonly used today for heavy equipment, which might indicate there are
problems connected to the solution.
• Alternative, future fuels might require different solutions |
Chalmers University of Technology | SWOT; Centralized remanufacturing concept
Strengths
• The filters are reused in “the first cycle”; the multiple use leads to major material savings
over the life cycle
• Exchange process is known to the technician and requires no additional work compared to
reference.
• Reman is a well-known and appreciated solution within the company
• Solution is in line with what customers are taught today; the importance of exchange
hygiene and quality tested genuine filters.
• No inseparable mixing of materials
• No or only small changes needs to be done to the existing engine or fuel system to
implement the suggested solution
• No disposables to handle through expensive waste management programs.
• Company remain in control over filters and materials over the full useful life
Weaknesses
• The location of the reman facility determines the extent of the environmental benefits.
• Large investment to build up the reman chain (competences, facility and equipment, logistics
etc.)
• Every machine needs more than one filter; i.e. an idle buffer of filters is required.
• The reference product lack some of the characteristics of “remanufacturable products”, since
it does not require skilled labour, is relatively cheap, does not contain especially scarce
materials.
Opportunities
• Allow for a more expensive product; with more features, such as sensors and better design
for easier disassembly and longer exchange intervals
• Can guarantee cleanliness; performance testing possible
• Possible to incorporate an automated process can, which may increase profitability.
• Possibility to have good waste water treatment in the reman facility
• Could expand to oil filters and fuel filters on all construction equipment (and trucks and
busses)
• Multiple-use transport packaging solution is possible
• High value recycling is possible due to the facts that materials are washed before disposed
and the exact alloys are known
Threats
• May require quite large volumes of filters to be profitable and environmentally sustainable
• The energy consumption for reman risk to limit the environmental benefits.
• Risk of losing the old supplier and expertise
• The required remanufacturing procedures are unknown today
• No known filter remanufacturer in Europe.
• Risk of a technology lock-in; electric powertrain requires no fuel filters, hybrid technologies
require less filtration and alternative fuels might require different solutions.
• High initial cost for washable filters; i.e. quantity and multiple use is essential.
• Solution implies a substantial risk to build a system around a solution which might not be
relevant for future business directions.
• No known large scale fuel filter washing facility exists today, which might indicate that there
are profitability and/or technology problems connected to the solution.
• Solution requires new competences and take little advantage of existing business
cooperation. |
Chalmers University of Technology | SWOT; Local filter cleaning concept
Strengths
• The filters are reused in “the first cycle”; the multiple use leads to major material savings
over the life cycle
• Exchange process is known to the technician
• No inseparable mixing of materials
• No or only small changes needs to be done to the existing engine or fuel system to
implement the suggested solution
• No disposables (except wastewater) to handle through expensive waste management
programs.
• Company remain in control over filters and materials over the full useful life
• Washing procedure is carried out locally; reduction of heavy transportations
Weaknesses
• Additional work for technician/workshop personnel
• Every facility needs washing equipment
• Every machine needs more than one filter; i.e an idle buffer of filters is required
• Requires local storage space for buffer filters
Opportunities
• Allow for a more expensive product; with more features, such as sensors and better design
for easier disassembly and longer exchange intervals
• Can use the existing washing machine that is in use in the workshops today
• Possible to design a portable washing unit
• High value recycling is possible due to the facts that materials are washed before disposed
and the exact alloys are known
• Could expand to oil filters and fuel filters on all construction equipment (and trucks and
busses)
• Multiple-use transport packaging solution is possible
Threats
• Difficult to guarantee cleanliness
• Might be hard to ensure quality of the process, since there are many local sites with varying
ability, regarding equipment, personnel, knowledge etc.
• There are quite small quantities at every local unit and it is uncertain if larger volumes of
filters are required for solution to be profitable and environmentally sustainable.
• No known manufacturer of washable filters in Europe.
• Solution is not commonly used for the application.
• A possible risk of a technology lock-in; electric powertrain requires no fuel filters, hybrid
technologies require less filtration and alternative fuels might require different solutions.
• High initial cost for washable filters; i.e. quantity and multiple use seem to be essential.
• Solution requires new competences and take little advantage of existing business
cooperation. |
Chalmers University of Technology | APPENDIX XIV - REMAINING CONCEPT UNCERTAINTIES, EXPANDED
Cartridge
Artefact design
How well sealed does it have to be between up and downstream side of the filter/cartridge?
To make the outer casing durable enough to endure wear over the machine’s entire life.
To make the filter fastener durable enough to endure repeated change.
System design
Educate customers about a new principle for fuel filter change, which is somewhat
contradictory to the existing solution, which claim a strength in the encapsulated design.
Handling
Not to contaminate the filter material during transportation and mounting.
How to handle remaining diesel in filter canister?
Not to contaminate the technician during cartridge replacement
Not to contaminate technician or surrounding during transportation of contaminated filter
Is it necessary and in that case, how to clean the canister at cartridge replacement?
How to know which cartridge should be used for each application (separate them)
How to handle contaminated cartridges in workshop
Which aspects need to be considered regarding the context during cartridge replacement?
Snow/rain, dust, cold/heat?
End of life
To make the cartridge in a single material, which can still fulfill the performance requirements
of today’s filter. (Is it necessary to have one material if it is going to incineration?)
Do we have to separate cellulose and diesel? Centrifuge, press, straight to incineration
Remanufacturing
Artefact design
To make the outer casing durable enough to endure wear over the machine’s entire life.
To make the filter fastener durable enough to endure repeated change.
Type of filter media appropriate for remanufacturing?
Do we require additional warehousing?
How can components be tracked?
Are special tools required for filter assembly/disassembly on machine and in that case, how
are they going to look like?
How can we influence the technicians’ behavior to take good care of the filters, in order to
obtain longer product life?
System design
How is the remanufacturing line going to work?
Overall procedure? Automatic handling by robot?
Inspection?
Performance testing?
How to keep track of part during reman?
Do we have to replace some parts?
How to decide which parts to scrap? |
Chalmers University of Technology | Is repair of damaged or worn parts possible?
Not to contaminate the filter material during transportation
Not to contaminate the filter material during mounting.
Handling
How is the remanufacturing logistic chain going to look like?
How to handle remaining diesel in filter canister.
Not to contaminate the technician during filter replacement
Not to contaminate technician or surrounding during transportation of contaminated filter
How to handle contaminated filters before reman in the workshop
End of life
How to enable material identification when end of life is reached
Local washing
Artefact Design
Is the workshop dishwasher clean enough to clean the main filter? Or is it necessary to design
a complementary dishwasher?
What is required to ensure cleanliness?
Enclose clean side?
Clean with purified diesel?
Test afterwards?
Are there other issues preventing use of the existing washing machine? Regulations?
How can the parts be attached and the filter opened?
How can the filter be attached to the machine interface?
What type of filter media can resist a large amount of washings and still live up to the
performance requirements?
How long will the washable filter media last?
Is a magnet required to handle metal residue/particles?
Washable filters cannot handle more than 20% biodiesel (pressure issue). (The reference filter
cannot either)
System design
How can the transportation container be designed?
Can the required level of cleanliness be ensured throughout the entire process, and how?
How is the washing procedure going to be like? Is a special rack needed?
What detergent is going to be used and how is the waste water going to be handled?
How are washed filters going to be dried?
How can customers be convinced of maintained quality of the cleaned filter.
Handling
Who is going to assemble the clean filters?
How can we protect humans from contamination during disassembly before washing? |
Chalmers University of Technology | APPENDIX XVI - FINAL CONCEPT EFFECT GOALS WITH EXPLANATION
Effect goal
The solution is required to…
...prevent damage during handling and transport
If the filter ought to be reused for more than 40 cycles, the filter and/or the system needs to be designed
accordingly.
...prevent filter contamination during handling and transport
The downstream side of the filter must remain totally clean during transportation from remanufacturing facility, to
the client until attached to the machine.
...resist multiple mounting/dismounting
In order for the solution to withstand at least 40 mounting/dismounting; the attachment system must be designed
accordingly.
...allow for washing
As specified in the requirements list, the filter needs to withstand a minimum of 40 cleaning cycles, hence the filter
must be designed accordingly.
...fulfil relevant performance requirements
The design is required to deliver the corresponding performance as the reference; i.e sufficiently respond to the
technical/functional requirements as defined for the existing fuel filter regarding fuel pressure, filtration capacity
and flow rate.
...avoid transportation of excessive matter/materials
Since remanufacturing rely on transportation of the components, the possibility to reduce the material being
transported must be considered.
...avoid diesel spillage during dismounting
Since the filter canister will be filled with diesel, spillage must be prevented when used filter is detached.
...avoid diesel leakage during handling and transport
The filter cannot cause contamination of people or the surrounding while being handled and transported.
…provide adequate information
The information provided on the components must be adequate for the user and the use.
...allow tracking of the product
Tracking of the complete product and possibly even the individual components is deemed required, in order to
optimize distribution, handling and procedures.
...allow for inspection/testing
After remanufacturing, the performance of the filter needs to be ensured by inspection and quality testing.
...allow for high value recycling
The design must allow high value recycling through considering the design for recycling aspects.
The solution is required/desired to…
…maximize life expectancy in general
The solution should consider how the expected life of the filter can be maximized in general; other aspects affecting
the filter life than what is presented above.
...avoid introduction of new tools
If not justifiable in relation more prioritized demands, introduction of new tools is undesirable due to the risk of
additional material usage.
...consider relevant ergonomic aspects
The attachment system and the overall handling must consider relevant physical and cognitive ergonomic aspects;
i.e. allowing for ergonomic use and to ensure adequate information presentation
...consider the demanding environment
The suggested solution must consider the demanding context, in regards to dirt, grease, etc.
...achieve the desired expression
The solution’s form, surface treatment, material, use etc. must comply with the overall expression defined in the
expression board.
...contribute to better service operation
It is considered positive if the new concept in any way can contribute to a service operation which is better from
the technician’s perspective. Faster and/or with less risks involved. |
Chalmers University of Technology | APPENDIX XVII - FINAL CONCEPT EXPRESSION ANALYSIS
Functionality
The concept expresses the use with a different material lid and a handle. Also, the transportation tray offers clear indications
to how the filter is going to be mounted and so is the attachment system on the machine. The resemblance with
Form and Material
The form has been inspired by the stacked cylinders in the expression board; the overall cylindrical shapes and the smaller
rounds, but also the robustness expressed by the setup and material thickness seen in the picture of the cylinders.
The materials used can to some extent be represented by the form and material images; however, since plastic products has
been added to support the functions, the product does not fully resemble the board expression. However, the colour and
shapes for the plastic products have considered the overall expression in terms of the shapes and also the colouring. The dark
grey plastic parts are considered to bear the desired resemblance with metal objects. The two lower pictures are depicted in
the clean and fine, stainless steel filtration mesh inside the canister.
Expression
The overall sustainable expression symbolized by the plants in cellulose cylinders is responded to by the product through the
functionality; focusing on maintaining material value, through reuse. (the focus on sustainable design decisions) maintaining
the materials in the closed value-loop
The desired expression symbolized by a metal recycling process can be said to have been achieved though utilizing the design
for recycling aspects and by the circular model ensuring a 100% recycling rate.
Metaphor
The Metaphor is of course a very subjective measurement; however, it can be argued that the overall expression from the
cylindrical shape and the chamfers below the threads and in the bottom, as well as from the rough and cold surface, indicates
the product contain something valuable, in resemblance with for instance a vault. |
Chalmers University of Technology | APPENDIX XIX - EVALUATION WITH RAPID ECO ASSESSMENT
Eco strategy wheel Rapid assessment Comment
consideration
1 Design for: Innovation
1,1 Rethink how to provide the Not yet considered More can be done
benefit
1,2 Design flexibility for Not yet considered More can be done
technological change
1,3 Provide product as service Not yet considered More can be done
1,4 Serve need provided by Not yet considered More can be done
associated products
1,5 Share among multiple users Not yet considered More can be done
1,6 Mimic biological systems Not yet considered Major system changes and/or new innovations
required before improvements are possible
1,7 Use living organisms in product Not yet considered Major system changes and/or new innovations
system required before improvements are possible
1,8 Create opportunity for local Has been considered, to More can be done
supply change some extent
2 Design for: Reduced material
impact
2,1 Avoid materials that damage Has been considered, to the Major system changes and/or new innovations
human or ecological health best of abilities required before improvements are possible
2,2 Avoid materials that deplete Has been considered, to the Major system changes and/or new innovations
natural resources best of abilities required before improvements are possible
2,3 Minimize quantity of material Not yet considered More can be done
2,4 Use recycled or reclaimed Has been considered, to Considered due to economic incentive. More can be
materials some extent done
2,5 Use renewable resources Has been considered, to More can be done
some extent
2,6 Use materials from reliable Has been considered, to the -
certifiers best of abilities
2,7 Use waste byproducts Assessment pending More information required
3 Design for: Manufacturing
innovation
3,1 Minimize manufacturing waste Has been considered, to the Considered due to economic incentive. Major system
best of abilities changes and/or new innovations required before
improvements are possible
3,2 Design for production quality Assessment pending More information required
control
3,3 Minimize energy use in Assessment pending More information required
production
3,4 Use carbon-neutral or Assessment pending More information required
renewable energy sources
3,5 Minimize number of Assessment pending More information required
production steps
3,6 Minimize number of Not yet considered More can be done
components/materials
3,7 Seek to eliminate toxic Assessment pending More information required
emissions
4 Design for: Reduced
distribution impact
4,1 Reduce product and packaging Has been considered, to the Considered due to economic incentive.
weight best of abilities
4,2 Reduce product and packaging Has been considered, to the Considered due to economic incentive.
volume best of abilities
4,3 Develop reusable packing Not yet considered More can be done
systems |
Chalmers University of Technology | 4,4 Use lowest-impact transport Not yet considered More can be done
systems
4,5 Source or use local materials Not yet considered More can be done
and production
5 Design for: Reduced behavior
and use impacts
5,1 Design to encourage low- Not yet considered More can be done
consumption user behavior
5,2 Reduce energy consumption Not yet considered More can be done
during use
5,3 Reduce material consumption Not yet considered More can be done
during use
5,4 Reduce water consumption Not yet considered More can be done
during use
5,5 Seek to eliminate toxic Not yet considered More can be done
emissions during use
5,6 Design for carbon-neutral or Has been considered, to More can be done
renewable energy some extent
6 Design for: System longevity
6,1 Design for durability Not yet considered More can be done
6,2 Design for maintenance and Not yet considered Not regarding the filter design, more can be done
easy repair
6,3 Design for re-use and exchange Not yet considered Not regarding the filter design, more can be done
of components
6,4 Create a timeless aesthetic Not yet considered More can be done
6,5 Foster emotional connection to Not yet considered More can be done
product
7 Design for: Transitional systems
7,1 Design upgradeable product Not yet considered More can be done
7,2 Design for second life with Not yet considered More can be done
different function
7,3 Design for reuse of Not yet considered More can be done
components
8 Design for: Optimized End of
Life
8,1 Integrate methods for used Not yet considered More can be done
product collection
8,2 Design for fast manual or Not yet considered Not regarding the filter design, more can be done
automated disassembly
8,3 Design recycling business Has been considered, to More can be done
model some extent
8,4 Use recyclable non-toxic Has been considered, to More can be done
materials some extent
8,5 Provide ability to biodegrade Not yet considered More can be done
8,6 Design for safe disposal Not yet considered More can be done |
Chalmers University of Technology | 1. Introduction
The circular economy builds on circular flows of biotic and abiotic resources. Where used
materials are fed back into the production system and no resources are wasted, to compare
with the traditional linear model where materials are used and thereafter disposed. A
challenge in a circular model is to collect materials after the product has been used. To
facilitate this process, new business models may need to be developed (Mentink B., 2014).
One possible way to go is to redesign the ownership structure for a product and sell, not the
product itself, but rather its function, so called functional sales. The environmental gain of
such business model is that more users can share fewer products; this decreases the total
number of products on the market and in turn the amount of materials extracted from the
planet. The secondary gain from such business model is maintained value of materials, since
the materials in the products are more likely to be used longer (Semples, 2014).
Going from one business model to another puts pressure on product developers and producers
since the revenue no longer comes from selling large quantities. What will generate profit is
instead to sell the product as many times as possible with as little maintenance as possible
(Mentink B., 2014). 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
comprehensively research circular economy and resource efficiency to create a
knowledgebase that can be useful when resource efficient and circular solutions are
developed. (MistraREES, 2015). This LCA report is part of the thesis “Preparing for tomorrow:
exploring design adaptations on a wheel loader in a circular business model” which in turn is
part of the Mistra REES program and will contribute with knowledge about how the
environmental impact is affected when a product is designed to fit in a functional sales
business model. The sustainability and use centered product development process, which was
used in the thesis and developed by the authors, can hopefully help contributing to a more
sophisticated process to develop products for a circular business models. At the end of the
program, the Mistra REES group is hoping to have created an information exchange between
industry and academia.
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 is carried out in
collaboration with Volvo Construction Equipment (VCE) located in Eskilstuna, Sweden. VCE
is developing off-road machinery, which includes dumper trucks, and wheel loaders amongst
other heavy road work equipment. Perhaps the most versatile of the products in the VCE
machine fleet are their different wheel loaders. The main objective for wheel loaders is to
move material; possible use scenarios however stretches from loading dumper trucks deep
down in Swedish mineral mines, to transportation of logs between storage and processing
facilities in Brazilian lumber industry. Volvo aims for their construction equipment to
maximize productivity by offering machines for each individual need; their wheel loaders
come in many different variations and there are an array of different attachments and
accessories available. (Volvo CE, 2015).
1 |
Chalmers University of Technology | In order to manage the environmental life cycle impact from their wheel loaders, Volvo CE
has been working with a range of innovations aiming to improve fuel efficiency. A more
recent initiative is to also extend the products useful life, by implementing customer tailored
maintenance agreements and remanufacturing of the components with a high remaining
relative value (Volvo CE, 2015). In line with the current trend of many other environmentally
aware companies, Volvo CE is exploring the opportunity to find new arrangements with their
customers; offering functionality and a life-cycle responsibility as opposed to the traditional
ownership. VCE’s aim is to create change and be one of the companies in Sweden that drives
the transformation towards a sustainable market, with less pollution (Gustavsson N. et al.,
2015).
Part of the purpose for the master thesis project was to investigate important design aspects
for a Volvo CE wheel loader to fit in a functional sales business model. More about this
process can be found in the report: “Preparing for tomorrow: exploring design adaptations on
a wheel loader in a circular business model”. However, to evaluate the outcome, and the
process used, life cycle assessment (LCA) was utilized. This report will describe the
procedure and results from the LCAs.
2 |
Chalmers University of Technology | 1.1 Goal
The goal of the study was to preform one LCA on the existing fuel filter and one on the new
concept to compare their environmental impacts.
The purpose of this study was to use LCA as an assessment tool in a product development
process to see if the product, developed for a circular business model, is better for the
environment than the existing solution. The product development process was part of the
master thesis “Preparing for tomorrow: exploring design adaptations on a wheel loader in a
circular business model”, which aimed to feed data into a research project mentioned in the
introduction. The intended audiences are the representatives at the company where the project
was carried out and the participants in the research project.
1.2 Scope
The LCAs carried out were change oriented rather than attributional since not every process
was fully accurately described. The focus was on the impact a change of filter model would
have.
1.2.1 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 fig. 1 and 2.
Figure 1. Showing a first flow chart of the reference filter life cycle. Figure 2. Showing a first flow chart of the concept filter life cycle.
1.2.2 Functional unit
The functional unit was set to 20 000 hours of fuel filtration in one wheel loader L150. This
timespan was used since it is the estimated lifetime of the wheel loader. A lifetime approach
3 |
Chalmers University of Technology | was needed to illustrate the difference in the two filter solutions; in the existing solution 40
main filters would be used over the lifetime of the wheel loader, and under the concept
solution, only one filter is needed to meet the filtration standard.
1.2.3 System boundaries
Geographical boundaries were set to Europe for parts production and assembly, and Sweden
for the use phase and end-of-life treatment. Extraction of materials and manufacturing of parts
were included in the model as were transports of the assembled fuel filter. For this study the
use phase is defined as from when the filter leaves the factory door until it enters the recycling
facility, that way all transports are included in the use phase.
1.2.4 Limitations
The study was limited to the main fuel filter installed in a Volvo wheel loader L150H.
Excluded from the study were:
The glue and coting of the reference concept since the material was unknown and the
amount was small.
Transportation of the parts, before assembly.
The set up and production of the remanufacturing facility.
1.2.5 Software
The modelling process was done using GaBi (thinkstep AG (ts)). The processes and material
approximations within GaBi can be seen in tables 3, 7 and 10.
1.2.6 End-of-life plans
For the modeling of end-of-life for the metal parts of the filters, datasets developed by Volvo
Group, were used for aluminum and steel. Volvo has currently approximated the process as if
a part of the steel and aluminum used (20-50%) are made from scrap metal. The amount is
dependent on the requirements of the material (aluminum or steel). This means that a
percentage of the metals do not go to end-of-life processes but are instead modelled as if they
were going straight back into the production. In that way, the material stays within the system
boundaries and is not causing impact from extraction or end-of-life processes. The remaining
metal is recycled and credits are given in end of life. Figure 3 outlines this process.
Figure 3. Aluminum and steel scrap are treated as raw materials going into the production phase and
are subtracted from the initial amount of used resources.
For end-of-life of paper, plastic and rubber, data sets in GaBi were used, where the energy
that was generated from the recycling process was used as energy credits in the model. The
chosen processes are found in table 1 together with the chosen process for steel.
4 |
Chalmers University of Technology | Table 1. Showing how the end of life processes were modelled in GaBi.
Material GaBi recycling dataset Reference
Paper EU-27: End of lite of paper/cardboard Based on Eurostat data
(landfill/incineration (2012)
Plastic/rubber EU-27: End of life plastics Based on Eurostat data
(landfill/incineration) (2012)
Steel Steel EOL plan Volvo
1.2.7 Impact assessment
The result thought to be most important in terms of the focus of the study is the material use
and global warming potential. The impact category materials use showed if the new design of
fuel filter was successful or not when it came to materials use. Kilometer traveled showed if
the new concept lead to more transports and if it had impact on global warming potential. A
summary of the impact categories can be found in table 2. The impact categories chosen were
calculated based on impacts ILCD/PEF recommendation where ILCD stands for International
Reference Life Cycle Data System and PEF product ecological footprint. Both are derived by
the European commission and for details please see ILCD handbook (European Commission
(EU), 2011). Since both LCAs used the same categories for the impact assessment and the
difference in impact was most important, details about how the calculations were made are
left out of this report. The impact of ozone was too small to show and was chosen not to
include in this report.
Table 2 Showing impact categories chosen for the study.
Impact category Unit
Climate change kg CO -eq.
2
Eutrophication Mole of N-eq.
Acidification Mole of H+ eq.
Abiotic resource depletion kg Sb-eq.
1.2.8 Sensitivity analysis
A sensitivity analysis was carried out from two aspects, first to test the concept by increasing
the amount of resources used during the remanufacturing process to see if the concept was
still a good option, second to make a breakeven analysis between the reusable filter and the
disposable.
5 |
Chalmers University of Technology | 2. Technical overview
This chapter starts with an introduction about why fuel filters are needed and continue with
describing the procedure for first the reference concept, the concept that is in use today and
move on to describe the new, developed concept; the concept filter.
2.1 Fuel filters
For a diesel engine to run, diesel has to be delivered through the injectors into the cylinders
where it is ignited. The injectors have a crucial role to make the engine more effective, by
delivering the right amount of diesel at the right time. To be able to have an effective engine
and meet the increasingly high requirements on exhaust gases the injectors have been made
more and more complex to be able to deliver fuel at the exactly right moment. This has made
them sensitive to particles in the fuel that cannot pass through the outlets of the injectors and
risk to clog them. (Interviewee 1, 2016) A fuel filter is a physical barrier between the fuel
tank and the injectors that separates particles from fuel.
It is not only particles that causes problem for injectors, water does as well. Diesel fuel always
contains water which needs to be separated from the fuel to not cause problems in the
injectors or in the combustion. The L150H wheel loader has two fuel filters; one pre-filter and
one main filter. The pre-filter takes care of the bigger particles and the water which ends up in
a container under the filter and is emptied manually when an indicator tells it is full. The main
filter sits just next to the pre filter and takes care of smaller particles. Fueling a wheel loader
is different from fueling a car. Off-road equipment is sometimes fueled from private diesel
tanks which might not hold the same cleanness standards as do commercial gas stations. This,
together with a dirty environment where the fuel-up takes place adds to the impurities
(Interviewee 2, 2016).
2.2 Architecture of the reference fuel filter
The ingoing parts in the reference fuel filter are (see figure 4): a steel casing (9) which
contains a cellulose based filter media (6) and is merged with a cap (2+3) which allows the
filter to be spun-on the machine. In the bottom of the canister is a spring (8) which allows fuel
to by-pass if the filter is clogged. The fuel flows in from the outside, it enters on the outside of
the filter medium and leaves from the middle.
Figure 4. Showing all ingoing parts of the reference fuel filter. 1-gasket, 2- element cap, 3-interface plate, 4 and 7- filter
media cap, 5- inner tube, 6- cellulose cartridge, 8-spring, 9- canister.
6 |
Chalmers University of Technology | 2.3 Reference filter in use
Filter exchange needs to be done on a regular basis to stay within the prescribed requirements
for the fuel filters. When in use, the filter media gets clogged and prevents fuel from passing.
Today filter replacements are part of the service agreements and is done every 500 hours. The
interval is set with empirical testing to avoid clogging. If the filter is clogged, the fuel pump
can fail and not enough fuel is delivered to the injectors causing the engine to lose power.
Three oil filters and two fuel filters are replaced at this interval, which gives a total of 200
filters if the life length is 20 000 hours (Prosis, 2016)
Filters are ordered to the local workshop and sorted according to the service needed. The
replacing procedure starts with the service technician bringing the right filters to the machine
on site, turning the machine on service mode, detaching the existing filter with a filter wrench
and putting the used filter in a container to bring back to the workshop. The gasket of the new
filter is greased to facilitate the tightening and spun on the machine. When attached, the filter
needs to be “bled” which is done with a hand pump, it needs 200-300 pumps to build up
pressure and get rid of air in the system. The entire procedure takes about 42 minutes. Used
filters are stored in a container for hazardous waste together with other fuel and lube filters.
The container is picked up when full, by a recycling company. The procedure is illustrated in
figure 5. (Interviewee 3, 2016, Prosis. 2016).
Figure 5. Showing an overview of the reference filters use phase.
7 |
Chalmers University of Technology | 2.4 Architecture of concept fuel filter
The developed concept was composed of an aluminum canister (9) with a cartridge (2-6)
made from stainless steel. A gasket (8) helped to avoid leakage between the machine and the
filter. The spring (7) from the reference concept was kept to allow for bypass of fuel. A big
difference was that the interface cap (see component 3 in figure 4) was replaced by a different
interface design. As can be seen in figure 6 is the attachment component no longer part of the
filter but will stay on the machine and reduce the material and weight of the filter.
Figure 6. Showing the exploded view of the concept filter.1- transportation lid, 2 and 6- filter element cap, 3- supporting
grid, 4- stainless steel filter media, 5- supporting inner tube, 7- spring, 8- gasket, 9- canister.
2.5 Concept filter in use
The new filter was assumed to be produced by the same supplier as today but the biggest
difference was that the filter material in the cartridge was washable and the filter will no
longer be disposable. The procedure was thought to be as follows (see fig. 7): the technician
brings a clean fuel filter to the wheel loader that needs replacement, removes it from its
transport container which also protects it from contamination. Unscrews the filter on the
machine and put in the transport container, remove transportation lid and attach the new filter
and take the used back to the workshop. Instead of disposing the filter it is sent to a
remanufacturing facility to be washed before it comes back to be used again.
8 |
Chalmers University of Technology | Figure 7. These are the steps the concept filter is going through during its use.
2.6 Remanufacturing of concept filter
The remanufacturing facility was placed in Flen, Sweden, for this study since the company
already has a remanufacturing process in place at this location (Interviewee 2, 2016). The idea
is that the filter arrives at the facility in its transport container and is disassembled. The
different parts are then washed and dried. To make sure that the filter can perform as well as a
new one a testing procedure is expected to be needed before it is put back in its container and
sent back to the service facility to be mounted on another wheel loader.
In this scenario, the filter is replaced immediately; the machine is therefore not expected to
wait for its filter to be washed to be able to run normally. Due to this slightly more filters
were expected to be needed than machines that were under service contract at a certain service
facility. For a facility that every year services 50 machines with this filter it was assumed that
8 additional filters were needed if the return time was 5 days. Fifty filters will sit on the
machines, 4 are on remanufacturing and 4 are available for replacement which gives a total of
1.16 filters per machine.
9 |
Chalmers University of Technology | 3. Method
This chapter describes the methodology that was used to perform the study.
3.1 Methodology
Life cycle assessment is, in short, a way to describe and calculate the resources a product uses
and the pollutions it causes during its life cycle. To be considered a LCA, the description must
follow a standard of how the assessment should be conducted developed by ISO (ISO 140 40,
1997). ISO suggest that the procedure should be carried out in four steps:
Goal and Scope definition, where the product subject of the study should be defined, the
intended application declared, the reason for carrying out the study and who the intended
audience is. It is also recommended to clarify the system boundaries and what types of
environmental impacts are considered. During this step a functional unit should be decided
around which the analysis will focus, and to which all resource use and emissions will be
related.
Inventory Analysis, in this step a system model should be developed according to the goal and
scope definition. The model is usually in the form of a flowchart and comes with a list of
inputs and outputs for all processes. The step also includes a calculation procedure where
outputs and inputs are related to the functional unit.
Impact assessment aims to present useful results of the study. For doing so it uses two steps:
classification where the different emissions and resource use are grouped into categories and,
characterization where the emissions and resource use are summarized and their actual impact
is calculated.
Interpretation, which has the goal to report and explain the results in an informative way and
in this part, could also include opportunities and recommendations for how to decrease the
impact of the product of service studied.
There are two general types of LCAs: consequential and attributional. In the consequential
LCA an impact of a change to the system is analyzed, where-as in an attributional the impact
of the system is investigated. The second type requires more accurate data to involve all
aspects of how the system is affected by every process and material.
To test the robustness of the model and LCA, a sensitivity analysis can be performed. It is
most useful if it is made on data that has been difficult to get correct or certain parts of the
system that are more sensitive to change. (Bauman & Tillman, 2004)
10 |
Chalmers University of Technology | 4. Inventory analysis: reference filter
In this chapter, the reference filter will be described in detail. All information needed for the
modelling and impact assessment is found here.
4.1 Flowchart
A more detailed flow chart was constructed (fig. 8) to facilitate the modelling of the life cycle.
It is still a simplified model and the manufacturing/assembly/packaging- step was developed
in the GaBi-model. The material for incineration that goes out from the use-process is
packaging material and the dust cap. The end of life process is described in section 4.7.
Figure 8. The life cycle of the reference filter modelled.
4.2 Production
Production of the reference fuel filter is done in Germany (Interviewee 1, 2016). All metal
components are made at the same site as the assembly process. Raw material in the form of
steel sheet comes to the factory and is deep drawn, or stamped and bended at the site. The
different components surface treatments, if needed, are also made on site. The cellulose based
filter medium comes to the factory as big rolls of paper and is both cut and folded at site.
Most processes are automated but man labor is required in some processes. The only
components that arrive ready made to the factory are the rubber gaskets and packaging
materials. (HowiItsMade, 2014). Table 3 shows a materials and manufacturing processes list
where the material in each component can be found together with how they are translated to
GaBi data sets.
11 |
Chalmers University of Technology | 4.3 Transport
Transports were estimated for the existing filter based on available data (see table 4). The
assembly was known to be done in Germany (Interviewee 1, 2016). However, the
transportation of parts to the assembling facility was excluded due to lack of data and that the
volumes and weights of each part were small so the effect on the end result was thought to be
minor.
Volvo CE has a European warehouse in Gent (Underhallsnyheter, 2016) where the filters are
transported before they continue to Sweden, in this case Göteborg. Moreover, the distance the
technician traveled was approximated and for the end-of-life treatment filters were assumed to
go to Halmstad.
Table 4 showing travel distances for one reference filter. The number needs to be multiplied with 40 to get the total transport
for the functional unit.
Part Geography Kilometer Vehicle Reference
Fuel filter Germany-Gent 450 Truck Google maps
Fuel filter Gent-Göteborg 1 240 Truck Google maps
Fuel filter Göteborg-Client 2x30 Truck Google maps
Fuel filter Client-Halmstad 170 Truck Google maps
Total 1 920
The transport was modelled with GaBi process GLO:Truck (Thinkstep) which is a combustion
engine truck running on diesel with a Euro 0-5 mix emission standard. The truck has a total
weight of maximum 5 ton and has a filling rate of 85%. With the diesel process EU-27:
Diesel mix at filling station (Thinkstep).
16 |
Chalmers University of Technology | 4.4 Use
The fuel filter is a passive product and does not consume resources when mounted on the
wheel loader. The considered impact during its use phase was the diesel needed by the
machine to carry the filter. The amount of diesel used was determined with weight allocation
according to table 5.
Table 5 showing how allocation of fuel was made to decide impact during use phase (during the entire life length) of
reference filter. The total consumption of diesel allocated to the filter was 20 l
Calculation of use phase diesel consumption
Wheel loader weight 23 405 [kg]
Fuel consumption per hour 18 l/h
Weight fuel filter 1.3 [kg]
Weight percentage 5.4 e-5%
Fuel consumption allocated to filter during 20 000h 20 [l]
4.5 End of life
Fuel and oil filters in Sweden are recycled by Rang-Sells in Halmstad. They have an
automated procedure called ReUseOil where the filters are shredded in the first step. In the
second step the material is centrifuged to recycle any oil which can be reused as motor oil.
Metal is then separated from other materials to be melted and recycled and the residue is
taken to incineration where the heat is recovered for district heating or heating of cement
ovens. The plant recycles about 6 000 000 filters every year (Rangsells, 2011).
4.6 Energy consumption
Table 6 shows the energy consumption over the life of the reference filter. It is divided in
non-renewable and renewable energy. The largest amount of energy, electricity, is needed
during the production phase and the energy comes from non-renewable sources.
Table 6 the energy requirements from each stage in the life of the reference filter.
Reference filter
Energy source Renewable [MJ] Non-renewable [MJ]
Production 184 1586
Use 16 1450
EOL 40 -722
Total 241 2314
17 |
Chalmers University of Technology | 5. Inventory analysis: concept filter
In this chapter the concept filter will be described in detail. All information needed for the
modelling and impact assessment is found here.
5.1 Flowchart
As can be seen (fig. 9) the production of the remanufacturing facility was left outside the
borders of the system. Waste products in the use phase come from spare parts, made from
metal or rubber that needs replacement in the remanufacturing process.
Figure 9. Flow chart of concept filter life cycle.
5.2 Production
As in the case of the reference filter, all production of metal parts is carried out in the same
factory. Since the material and dimension of the canister is difference for the concept filter,
another production process is required. The canister in the concept filter is made from
aluminum and milled instead of the process of deep drawing. The filter medium is not made
from cellulose but from woven steel wire which requires a weaving procedure instead of
cutting and folding paper. The gaskets are taken from the same place but the packaging
material this time is a plastic container. All materials, weights and processes are found in
table 7.
18 |
Chalmers University of Technology | 5.3 Transport
It was assumed that the company could have cooperation with their current supplier in the
development of the washable filters and the supply chain would therefore look the same. The
difference between the concept filter and the reference is the 4th post in table 8 where the
transport back and forth to the remanufacturing facility is presented.
Table 8.Ttransports of the concept filter.
Part Geography Kilometer Vehicle Reference
Fuel Germany-Gent 450 Truck Google maps
filter
Fuel Gent-Göteborg 1 240 Truck Google maps
filter
Fuel Göteborg-Client 2x30 Truck Google maps
filter
Fuel Client-reman-client 30 5765 Truck Google maps
filter
Fuel Client-Halmstad 170 Truck Google maps
filter
Total 32 496
The transport was modelled with GaBi process GLO:Truck (Thinkstep) which is a combustion
engine truck running on diesel with a Euro 0-5 mix emission standard. The truck has a total
weight of maximum 5 ton and has a filling rate of 85%. With the diesel process EU-27:
Diesel mix at filling station (Thinkstep).
5 392 km from client to Flen*2 (back and forth)*39 for amounts of replacement=30 576
22 |
Chalmers University of Technology | A simplification, concentrated on material flows, was made for the GaBi modeling. The
processes and materials are found in table 10.
Table 10 Processes used to model the remanufacturing in Gabi. Amounts of materials per wash of one filter.
Inflow material Quantity/one filter GaBi Process Reference
Detergent 20 [g] RoW: soap production Eco invent
Electricity 229 [kJ] SE: Electricity grid mix6 Thinkstep
Water 4 [kg] EU-27: Process water Thinkstep
Steel parts 12.8 [kg] RER: Steel plate Worldsteel
Rubber parts 10 [g] GLO: Vulcanization of Thinkstep
synthetic rubber7
Steel and rubber parts were estimated to need replacement in 10% of the cases when the filter
was in for remanufacturing. The numbers for water and detergent were simply assumed, but
to estimate electricity consumption of an industrial washing machine was used as a reference
(Wexiödisk, 2016) where 8 filters were thought to go in each of the baskets that could go into
the machine.
5.6 End-of-life
The end of-life-for the concept filter will be determined by the remanufacturing facility which
will replace parts when needed. The parts that are not viable for reuse are sorted out in either
the inspection or the testing step and will be washed and discarded to steel or aluminum
recycling. No shredding is needed. GaBi built in data sets were used for paper and plastic
recycling and Volvo data sets for metals.
5.7 Energy consumption
In table 11 the energy consumption for each stage in the concept filter´s life cycle is shown. A
large amount of energy is needed during manufacturing in which extraction of for example
aluminum is included.
Table 11. Energy required for each step in the concept filter´s life cycle. The renewable energy is dominated by hydro and
wind power.
Concept filter
Energy source Non-renewable [MJ] Renewable [MJ]
Production 263 68
Use 36 21
EOL -121 -45
Total 178 44
6 A mix where nuclear and hydro power energy sources dominates.
7 For ingoing processes see table 7
25 |
Chalmers University of Technology | 6.2 Transport
Table 13 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 due to the fact that each
reference has to be transported from Germany to Sweden, whereas the concept only
completes that journey 1.16 times and then it is only taken back and forth to Flen in Sweden.
Table 13. Comparison of the transports needed to provide the wheel loader with main fuel filtration for 20 000 h.
Product Kilometers
Reference 76 800
Concept 32 496
6.3 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.
6.3.1 Global warming potential
As can be seen in figure 12 the concept filter has minimal global warming impact compared
to the reference solution. Both have biggest impact during the production step, thus tied to
material extraction.
Global Warming Potential
200
150
.v 100
iu
q
E
- 50
2
O
C
g 0
k
-50
-100
Total Production Use End of life
Reference filter 118 157 26 -66
Concept filter 12 18 2 -8
Figure 12.Graph showing the global warming potential for the reference filter and the concept filter.
The largest impact caused by the reference is the process of extracting and working the steel.
The largest impact caused by the concept is instead the extraction and manufacturing of
aluminum components. The emissions are in both cases carbon dioxide, tied to electricity
production to feed energy into material extraction processes. Since the greatest impact in both
cases comes from the production phase, it was believed that the approach to reduce the
27 |
Chalmers University of Technology | material consumption per functional unit was the most effective method of impact reduction.
The graph also shows that the material reduction gave the desired, lowered impact.
6.3.2 Acidification potential
In figure 13 it is shown that the acidification potential is much smaller for the concept filter.
Biggest impact has the production for both filters.
Acidification Potential
0,50
0,40
0,30
.q
e
+ 0,20
H
f
o
e 0,10
lo
M
0,00
-0,10
-0,20
Total Production Use End of life
Reference filter 0,44 0,41 0,19 -0,16
Concept filter 0,05 0,09 0,01 -0,05
Figure 13. Graph showing the acidification potential for the reference and concept filter.
It is somewhat surprising that the concept filter has a relatively high acidification potential
compared to the reference filter in the production phase despite the much lower weight.
However, aluminum production is resource heavy and contributes Sulphur dioxide from the
production of the canister that shows in the blue bar. For the reference filter, it is emissions of
the same kind that dominates and they are also tied to material and production of the canister.
The acidification potential from the use phase in the case of the reference filter is in the form
of emissions of nitrogen oxide caused by combustion of diesel due to transportation.
6.3.4 Eutrophication potential
Figure 14 shows that production phase and use phase have similar amount of emissions that
cause eutrophication for the reference filter. It is, again, the steel production that causes
emissions but also the big amount of transportation that is required for the reference filter. All
those numbers are small relative the 18 000 tons of NOx that was emitted in the region of
Västra Götaland 2013 (Miljömål, 2016).
28 |
Chalmers University of Technology | Eutrophication Potential
1,4000
1,2000
1,0000
.q 0,8000
e
N 0,6000
f
o
e
0,4000
lo
M 0,2000
0,0000
-0,2000
-0,4000
Total Production Use End of life
Reference filter 1,3064 0,8367 7,86E-01 -0,3165
Concept filter 0,0083 0,0080 0,0003 0,0000
Figure 14 Graph comparing the eutrophication potential of the concept and reference filter.
What makes the reference filter dominate is the lager amount of transportation by truck that
emits nitrogen monoxide. That is what causes emission by the concept fuel filter too but the
amount of transportation is much lower. What also causes emissions for the concept is the
soap production in the form of ammonium.
6.3.3 Abiotic resource depletion
Figure 15 shows the difference in abiotic resource use. The concept filter uses very few
resources compared to the reference concept which can be seen in this graph.
Abiotic Resource depletion
0,3500
0,3000
0,2500
.v
0,2000
iu
q
E 0,1500
-
b
S
g
0,1000
k
0,0500
0,0000
-0,0500
Total Production Use End of life
Reference filter 0,3095 0,1508 3,79E-02 0,1208
Concept filter 0,0002 0,0004 0,0000 -0,0002
Figure 15.Graph showing abiotic resource depletion for the concept and reference filter.
The use of resources is very spread over all kind of resources taken into account in this impact
category. What stands out is the use of borax, indium and colemanite in the process estimated
for the cellulose cartridge in the reference filter. However, this is an approximated process
and whether those materials are used in reality is not known.
29 |
Chalmers University of Technology | For the concept filter it was the production of the canister that was most resource consuming
and the materials standing out was bauxite and tantalum. For the use phase, it was the
production of detergent used for remanufacturing that stood out using indium.
6.4 Analysis
Over all it can be said that the LCA has shown that the material reduction has led to reduction
of environmental impact. The new way of handling the filters has also led to a big reduction
in transports. What is worth paying attention to is the use of aluminum for the canister in the
concept filter; aluminum was chosen since it is a light material and it was desired to keep the
kilos down to not cause unnecessary impact during transport. However, other impact
categories show how aluminum causes the biggest impact for the concept. In the refinement
of the concept it is suggested that a thorough stress test should be done on the canister to be
sure how big the load on it will be and after that look into if other materials are possible and
what effects that would have on the transported mass.
6.3.5 Water use
GaBi software, with the dataset available for the project, could not provide a good number for
water use for the model. It would be desirable to be able to compare the reference filter with
its water consuming cellulose cartridge solution with the washable concept filter. However,
the water estimated to be needed for the concept filter is a very rough estimation and the water
may also be recyclable within the process. This is somewhat a result gap and it is something
worth considering when moving forward with the concept development.
30 |
Chalmers University of Technology | 7. Sensitivity Analysis
In the following section, the results from the sensitivity analysis are presented. A sensitivity
analysis was made on the remanufacturing procedure and in the form of a breakeven analysis
between the concept and the reference filter.
7.1 Remanufacturing
The idea in this part of the sensitivity analysis was to increase consumption of all resources
needed during the remanufacturing process, including transports, since the very nature of the
concept builds on remanufacturing. The remanufacturing procedure was not described in
detail in the report “Preparing for tomorrow: exploring design adaptations on a wheel loader
in a circular business model” and the amount of ingoing resources was considered needed to
be tested to understand where the limit was if the procedure would be developed. This was
done by multiplying the remanufacturing process with 2 and 10. This way a doubling of
needed resources was represented with 80 washings instead of 40 and an increase with 10
times the amount was represented by 400 washes. Since transportation was also included it
would represent that the distance to the remanufacturing facility would double in the first case
and tenfold in the second.
7.1.1 Global warming potential
Figure 16 shows that even if all the factors in the remanufacturing process were multiplied by
10, which includes energy consumption, transport, water use, spare parts and soap, the
concept filter is still better for the environment in terms of global warming potential. This is a
good indication for that the remanufacturing process is a better solution than disposable
filters.
Global Warming Potential
140
120
100
.v
iu
q 80
E
-
2
O 60
C
g
k
40
20
0
Reference filter Concept filter + 40 Concept filter + 80 Concept filter + 400
(standard)
Figure 16. Graph showing how the global warming potential is affected when the parameters of
remanufacturing are changed.
The biggest change between times two and ten is that the rubber production passes the
production of steel spare parts in terms of emissions. This indicates two things, the
31 |
Chalmers University of Technology | remanufacturing procedure itself is not a large source of pollution and the amount of spare
parts needs to be kept low.
7.1.2 Acidification potential
When it comes to acidification, as shown in figure 17, the relative difference is not as big as
for global warming potential.
Acidification Potential
0,500
0,450
0,400
.q 0,350
e
+ 0,300
H
f 0,250
o
e 0,200
lo
M 0,150
0,100
0,050
0,000
Reference filter Concept filter + 40 Concept filter + 80 Concept filter + 400
(standard)
Figure 17. Graph showing how the acidification potential is affected when the parameters of remanufacturing are changed.
The biggest contributors are rubber and steel parts production and soap production. They are
causing emissions in the form of Sulphur dioxide and nitrogen oxides.
7.1.3 Eutrophication potential
Figure 18 shows that even with a ten folding of everything consumed in the remanufacturing
process, the eutrophication potential of the reference concept is still bigger. Nearly all the
damage of eutrophication can be avoided with the concept solution.
Eutrophication Potential
1,400
1,200
1,000
.q
e
N 0,800
f
o
e 0,600
lo
M
0,400
0,200
0,000
Reference filter Concept filter + 40 Concept filter + 80 Concept filter + 400
(standard)
Figure 18. Graph showing how the eutrophication potential is affected when the parameters of
remanufacturing are changed.
32 |
Chalmers University of Technology | 7.1.4 Resource depletion
Figure 19 shows that the concept filter with its remanufacturing process saves resources.
Resource Depletion
0,3500
0,3000
0,2500
.v
iu 0,2000
q
E
-
b S 0,1500
g
k
0,1000
0,0500
0,0000
Reference filter Concept filter + 40 Concept filter + 80 Concept filter +
(standard) 400
Figure 19.Graph showing how the resource potential is affected when the parameters of the
remanufacturing process are changed.
A conclusion from figure 19 is also that the remanufacturing process for each filter consumes
almost no resources compared to the currently used disposable reference filters. This is still
true if the input parameters are ten folded. The remanufacture procedure has little impact on
resource consumption compared to new production of filters.
7.1.6 Analysis
The remanufacturing process was the biggest area of uncertainty in the model over the
concept due to the readiness of the concept. Values for what was thought to be needed were
estimated and because of that, it was interesting to see how the results developed when the
used amount was multiplied. However, some of the factors multiplied by 400 are not likely to
be that big such as water use and spare parts. On the other hand, it gave a good indication of
what is most important to think about when the remanufacturing process is developed.
It is definitely worth refining the filter design until it requires as little spare parts as possible
to keep the environmental impact minimum. There is a lot of potential to develop a
remanufacturing facility which can allow for low impact if principles of reuse are
implemented, for example, could left-over heat from the washing and drying of filters be used
in the factory.
33 |
Chalmers University of Technology | 7.2 Breakeven
A breakeven analysis was made to see how many times the concept filter needs to be reused
before it was a better option than the reference filter in terms of greenhouse gases emitted.
7.2.1 Breakeven
Figure 20 shows the result from the breakeven analysis. The concept filter has a much higher
amount of initial emissions (5.9 kg) but the remanufacturing process emits very little. Due to
this, the concept filter has soon paid off its initial emissions and the reference filter, which
emits 2.64 kg per each, is the less good option already after three replacements. To make the
comparison more interesting, the high resource consuming (times ten from previous figure)
remanufacturing concept was also included in the breakeven analysis (green line). If the
remanufacturing process is that expensive, resource wise, it will take five replacements to
breakeven.
Breakeven
30
25
20
.v
iu
q
E
- 15 Reference
2
O
C Concept
g
k 10
Concept reman x10
5
0
0 1 2 3 4 5 6 7 8 9 10
Number of replacements
Figure 20. Showing breakeven point between the concept filter, concept filter with resources for remanufacturing multiplied
with 10 and the reference filter. It can be seen that the concept breaks even at 3 disposable filters but that it takes 5to break
even with a more resource consuming remanufacturing (reman) process.
7.2.2 Analysis
It might be surprising that the breakeven point is reached after so few exchanges. Though,
considering that the concept filter only weights slightly more, it is more understandable. It has
already been shown that the impact for both filters is in the production phase and that even if
a concept filter is remanufactured 400 times during 20 000 h it is still a better option. This
indicates that the remanufacturing process is not very resource intensive.
34 |
Chalmers University of Technology | 8. Discussion
Assessing a concept with LCA comes with a lot of challenges and might at first sight seem
contradictory to the very nature of an LCA. A concept is lose and sketchy, not much is
decided and even less is known compared to how it is needed to be in an LCA where every
process, material and matter of transportation needs to be known in detail. However,
involving LCA in the product development process also comes with a lot of second thinking
of materials choices, thicknesses and the need of extra details.
8.1 Concept filter
During the product development process of the concept fuel filter some assumptions where
needed to be made only for the sake of the LCA. For example the different manufacturing
processes the location of the remanufacturing facility and the weights of each component.
Normally the product would be prototyped and tested before such parameters would be
decided; manufacturing strategies would be developed by someone with special competence
in such technologies. It was not the case in this study, simply because it was out of scope for
the project. But the result is not useless, quite the opposite. There is a reference and a base to
start from when developing a manufacturing lane for this filter saying how to avoid huge
environmental impacts.
The choices of datasets in GaBi were made based on what seemed to correspond to the
process that was decided to be needed for the concept filter future production, use and end of
life. They might all be different in the final product. But in this LCA an indication is given to
where most attention should be put when further development is used. One example is the fact
that the spare parts contributed so much to the emissions in the remanufacturing process, to
reduce them is therefore important in the development of the concept.
8.2 Reference filter
The modelling of the reference filter was more based on measured and collected data
compared to the concept filter. The data was then approximated with processes in GaBi and
might not give the whole truth. However, the accuracy of the both LCAs are quite similar and
it made them more suitable for comparison. Where the same material and manufacturing
technique were used in both filters the same data sets have been used to make the comparison
more accurate. The biggest reason for that the reference filter stood out as the lesser option in
the comparison was the large amount of materials used. A change in the choice of processes
in GaBi would not have affected that fact and the results are therefore to be seen as useful.
8.3 Transports
The distances and means of transportation have been a data gap for the reference filter. To
come around this issue the same nodes, trucks and transportation routes were estimated for
both the concept filter and the reference to make the comparison as fair as possible. If
additional information had been found, the same changes would have been applied to both
filters and would not have had any significant impact on the end result. The matter of
transportation was estimated to a 5 ton truck with an emission standard of Euro 0-5 mix. This
35 |
Chalmers University of Technology | should be seen as a conservative assumption. Most trucks have a class 5 emission standard
today and the way the filters are transported might not be with such a big truck. This indicates
robustness in the remanufacturing model; it does not have to increase transports and it saves
materials.
The location of and transport to the remanufacturing facility are both unknowns but the values
used here could be seen as a base scenario in which parameters could be modified to find a
good solution.
8.4 Choice of datasets
Since production of both filters was decided to be located in Germany the datasets were
chosen accordingly which in this case meant that EU-27 processes were chosen as often as
possible. The impact from an EU-27 process is an average value of the impact of similar
processes in 27 different EU countries. If an “agg” process was chosen (no input datasets
needed) the European or German datasets were preferred to create a model that corresponded
as well as possible to a real manufacturing scenario in Europe. By doing so, it is likely that the
production in both cases can be considered an average scenario when it comes to emissions.
The processes should not be considered either the best or the worst case.
8.5 The bigger picture
The result from the LCA is in line with the idea of circular economy: by rethinking the system
a product is part of, the product’s or the service’s environmental impact can be decreased.
(Ellen MacArthur Foundation, 2015) The concept in this study was developed with design for
environment strategies and a product development process that was concerned about the
environmental aspects, which paid off in terms of less environmental impact.
The concept development focus was to decrease the amount of materials used and reuse what
was already extracted. This is supported by Miljömålsberedningen (2016) which writes in the
report En klimat- och luftvårdsstrategi för Sverige, Del 1 that raw materials extraction is
accountable for 19% of the total amount of emitted greenhouse gases globally and waste
management is accountable for 3%. In this study, the focus was metal, which is positive since
the extracted amount of ores increased with 133% between 1980 and 2008 according to
OECD (8.2 Gt year 2008).
36 |
Chalmers University of Technology | Acknowledgement
`
IwouldliketoexpressmysinceregratitudetomysupervisorStefan`ıaOskGardarsdo`t-
tir, for always making time for me and my never-ending questions. Stefan`ıas support
and guidance throughout the whole project has been invaluable, whether it was help
with the simulations in Aspen Plus or feedback on the report.
I am also particularly grateful for the assistance given by my supervisor Fredrik
Normann, who helped me bounce ideas and always provided valuable input. Fredrik
has been a guiding light, steering me in the right direction in the decision making
process regarding the content of the thesis.
Many thanks to Henrik Jilvero for sharing his knowledge regarding dimensioning
of process equipment. I also would like to thank Karin Pettersson for helping me
understand the overall balance of the pulp mill and providing essential process data
required in the project. Ragnhild Skagestad and Nils Henrik Eldrup are acknow-
ledged for their fast and well executed investment cost analysis.
Lastly, I wish to acknowledge the much appreciated feedback on the report
provided by my examiner Klas Andersson and my opponent Simon Lindqvist.
G¨oteborg, June 2014
Joakim Hedstr¨om
iii |
Chalmers University of Technology | Abstract
Global warming is a serious threat that could have catastrophic effects if emissions
of greenhouse gases are not reduced drastically. This thesis evaluates the suitability
for pulp mills to help in reducing these emissions with focus on carbon capture and
storage(CCS)technologies. Thisindustryisbasedonarenewablefeedstockthatcan
be used to reduce greenhouse gas emissions by replacing fossil fuels in transportation
or heat and power generation. Furthermore, capturing CO from pulp mills would
2
result in a“negative”net emission that would reduce the global CO emissions.
2
Three future scenarios of the pulp mill are considered. The first scenario assumes
the pulp mill is run as the present day situation and that post-combustion capture of
CO using the MEA processes is applied to the stack gas emissions from the recovery
2
boiler. In the second scenario the recovery boiler is substituted with black liquor
gasification technology and the produced syngas is used for electricity production.
Pre-combustion capture of CO using the Selexol process is applied in this scenario.
2
The third scenario also utilizes black liquor gasification technology for recovery but
the syngas is used to produce DME instead of electricity. Pre-combustion capture
of CO with the Rectisol process is used in this scenario.
2
Each scenario is divided into two cases; one with capture and one without cap-
ture. The cases are simulated using Aspen Plus and the utility consumption of the
processes is determined. Pinch analysis is used to reduce the utility demand. All
process equipment is dimensioned to provide a basis for an investment cost ana-
lysis which is performed by an external partner. The results from the simulations
are used in an overall energy and mass balance of the pulp mill which determines
the additional resource consumption associated with carbon capture. The resource
consumption together with the investment cost is used to calculate a total cost for
carbon capture.
The scenario utilizing the MEA process shows the highest potential to offset
global CO2 emissions with a net reduction of 715ktCO /year at a capture cost of
2
431SEK/tCO . The scenario with black liquor gasification for electricity production
2
has the lowest net reduction with only 318ktCO /year and also the highest capture
2
cost of 453SEK/tCO . The scenario with co-production of DME at the pulp mill
2
has the potential to reduce global emissions with 393ktCO /year and a low capture
2
cost of 88SEK/tCO .
2
Today there exists an over-abundance of low price emission certificates, in Mars
2014 an emission certificate costed 5.0e. Hence, in none of the studied scenarios
carbon capture was profitable. However, if the emission certificate market recovers
then the pulp mill would be a suitable candidate for CCS.
Keywords: CO capture, MEA, Rectisol, Selexol, pulp mill, pre-combustion, post-
2
combustion, Aspen Plus, process simulation, BECCS
v |
Chalmers University of Technology | 1 INTRODUCTION
1 Introduction
This thesis investigates the possibilities of applying carbon capture and storage to
the pulp and paper industry. A brief introduction to this subject is presented in this
chapter, along with the aim and scope of the work.
1.1 Background
There is little doubt among climate scientists that industrial emissions of CO , and
2
other greenhouse gases, are causing global warming and our emissions needs to
rapidly decline, or else the global temperature increase will surpass 2◦C. This rise
in temperature, compared to the pre-industrial average temperature, has long been
considered the limit for which the most serious consequences of global warming can
be avoided (European Council, 1996). However, recent research suggests that even
an increase of only 1◦C may have dangerous effects (J. B. Smith et al., 2009).
Most CO emissions originate from combustion of fossil fuel which today’s en-
2
ergy system is highly dependent on. The global emissions of CO from fossil fuels
2
in 2011 were 31GtCO , this corresponds to an increase by 49% compared to the
2
levels in 1990 and emissions are steadily rising (IEA, 2013). A complete substitution
to renewable energy sources cannot be realized overnight and therefore transitions
technologies are needed; Carbon Capture and Storage (CCS) is one example. By
capturing the CO formed in, for example coal-based power production, the environ-
2
mental impact of such a facility would be greatly reduced. This would buy mankind
some well needed time to find alternative ways to satisfy the ever growing energy
demand. However, everything comes at a price. The capture processes are energy
consumers as well. Hence, the heat or electricity that can be sold is reduced and
inevitably, so are profits. For CSS to be widely implemented, policies that provide
incentive for capturing CO needs to be in place. The IPCC (Intergovernmental
2
Panel on Climate Change) states that such policies can be in the form of economic
instruments, government funding or regulation (IPCC, 2007). One example is the
EU’s emissions trading scheme which aims to reduce CO emissions by gradually
2
decreasing the number of available certificates. A certificate grants the holder the
right to emit one tonne of CO , thus large emitters such as power plants needs to
2
purchase a considerable number of certificates. Consequently, this policy has the
potential of making CCS economically viable. However, the price of a certificate
has steadily decreased during the past years, reaching an all-time low of 2.8e in
April 2013 (Swedish Energy Agency, 2013). Low prices means that the focus of
CCS projects may need to change from aiming to capture bulk amounts of CO to
2
smaller but more easily available CO with less costs associated. Hence, opportun-
2
ities to capture CO in industries which previously have not been associated with
2
CCS should be examined.
An industry branch that has received increased attention with respect to CCS in
the past years is the pulp and paper sector. In a Swedish perspective this industry is
1 |
Chalmers University of Technology | 1 INTRODUCTION
ofgreatimportance. ThetotalCO emissionsfromfossilfuelinSwedenin2011were
2
49MtCO , as a comparison the pulp and paper industry emitted 22MtCO origin-
2 2
ating from biofuel combustion (Swedish Environmental Protection Agency (2013);
European Environment Agency (2011)). Hektor (2008) has evaluated the implica-
tions of applying post combustion capture onto pulp and paper mills whereas Pet-
tersson (2011) examined the pre-combustion alternative. An interesting aspect of
the pulp and paper industry is that it is based on a renewable feedstock, i.e. forest
resources. That implies that a pulp mill can be considered a zero net contributor
to global CO emission, since the trees will reabsorb the CO through photosyn-
2 2
thesis as they grow. This reasoning is only valid if a sustainable management of
the forest resources with sufficient replantation is practised. However, if this is the
case then capturing the CO would actually result in a negative net contribution
2
to the global CO emissions. If CCS technology is applied to emissions originating
2
from biomass it is called BECSS, abbreviation for Bio-Energy with Carbon Capture
and Storage. By implementing BECSS it is possible to compensate for emissions
from other sources, such as transport, since CO actually is removed from the at-
2
mosphere. This also gives the means to recover from an “over-shoot” of the 2◦C
target. Applying BECCS to the pulp and paper industry is not only important in
a Swedish perspective. The pulp and paper industry is responsible for 5.7% of the
global industry energy usage and the industry is in turn accountable for 21% of
global CO emissions (IEA (2007); IEA (2013)).
2
1.2 Aim and scope
The aim of this thesis is to evaluate the possibility for implementing CO capture
2
at a pulp mill given three future scenarios. The first scenario uses the conventional
recovery boiler to regenerate the cooking chemicals. The second scenario replaces
the recovery boiler with black liquor gasification technology to produce additional
electricity. The third scenario also utilizes the black liquor gasification technology
but DME is produced instead. The study investigates if pulp mills have any easily
available CO sources that can be captured at a low cost. Specifically, the study
2
estimates the potential net reduction in global CO emissions, the cost for the re-
2
duction and the different sellable products produced in the pulp mill for the three
scenarios.
Models of the capture processes are built using the simulation tool Aspen Plus.
A rate-based approach in the simulations enables the possibility to design and size
allequipmentandtherebyperformaninvestmentcostanalysis. However, costcalcu-
lations are beyond the scope of this thesis and are instead performed by an external
partner. Theresultsfromthesimulationsareusedtocalculatetheoverallenergyand
mass balance of the pulp mill. By combining the results from the overall balances
and the investment cost analysis a total cost for carbon capture is obtained.
2 |
Chalmers University of Technology | 2 CARBON CAPTURE AND STORAGE
2 Carbon capture and storage
CCS can be divided into three consecutive steps. Naturally the first step is to
capture the CO . Three different approaches can be adopted for this task: post-,
2
pre- or oxy-fuel combustion. The second step is to compress and transport the CO
2
to a suitable storage site. Lastly the CO is pumped down into the carefully selected
2
underground storage site.
2.1 Post-combustion
The concept of post-combustion imply separation of CO from the flue gases after
2
combustion, see Figure 1. This technology has a large advantage due to the fact
that it can be retrofitted to any existing plant.
Energy
Air CO 2
Flue gases
Combustion Separation
Fuel CO 2 lean flue gases
Figure 1: Schematic representation of the post-combustion process.
Chemical absorption is the most commonly used separation method in post-
combustion applications. However other methods such as membranes, adsorption
or cryogenic technology exist as well. In chemical absorption the solvent forms
weak bounds to the CO in an absorber at low temperatures, around 50◦C. These
2
bounds are broken in a stripper at higher temperatures (around 120◦C, the exact
temperature depends on the solvent used) and the CO is released. The solvent is
2
then recycled back to the absorber, see Figure 2. The main disadvantage is that the
regeneration of the absorbent is energy consuming.
Both organic and inorganic absorbents can be used in the process. The most
frequently used organic solvent is monoethanolamine (MEA). Piperazine (Pz), di-
ethanolamine (DEA) and N-methyldiethanolamine (MDEA) are other examples of
aminesused. Someseparationprocessesuseamixtureofseveraldifferentabsorbents
(Bougie & Iliuta, 2012). Examples of inorganic solvents are potassium carbonate,
sodium carbonate and ammonia. (Hektor, 2008)
3 |
Chalmers University of Technology | 2 CARBON CAPTURE AND STORAGE
CO
CO-lean flue gases 2
2
Absorber Stripper
Flue gases
Solvent
recycle
Figure 2: Simplified flowsheet of the post-combustion process.
2.2 Pre-combustion
The principle of pre-combustion capture is shown in Figure 3. In this approach the
fuel is gasified, i.e. partially combusted. The gasification is carried out in a pressure
range of 30-70bar using a deficient supply of oxygen resulting in a gas mixture
consisting mostly of CO, CO and H (Gibbins & Chalmers, 2008). Depending
2 2
on the application of the formed syngas (synthesis gas), it is sometimes desirable
to adjust the ratio between CO and H . This is carried out by adding water and
2
passing the mixture through a number of reactors containing catalyst beds. Inside
the reactors the water-gas shift reaction occurs. The addition of water displaces the
equilibrium towards the right hand side of the reaction:
CO+H O ←−→ CO +H (1)
2 2 2
The CO is then separated from the syngas and thereafter the syngas is either
2
combusted or used for other purposes such as fuel synthesis. The separation of
CO is most often carried out using physical absorption e.g. the Rectisol or Selexol
2
processes. Physical absorption is used when higher concentration of CO is present,
2
compared to the levels present in post-combustion where chemical absorption has to
be used instead. An advantage with physical absorption is that CO is absorbed at
2
high pressures and released at low pressures which reduces the energy consumption
for regeneration of the absorbent. However, there will be a loss of efficiency due to
the shift reaction since less CO reaches the combustion chamber and consequently a
reduced mass flow of CO pass through the turbine. Hence, less power is generated.
2
(Gibbins & Chalmers, 2008)
4 |
Chalmers University of Technology | 2 CARBON CAPTURE AND STORAGE
Water CO 2 Energy
Air
Gasification
Water-gas
Separation
H
2
Combustion/ CO 2-lean flue gases
Fuel shift reaction Gas turbine
Figure 3: Schematic layout of the pre-combustion process.
2.3 Oxy-fuel combustion
Oxy-fuel combustion differs from the other technologies in the aspect that oxygen
is used instead of air in the combustion process, see Figure 4. Oxygen is separated
from the air in an initial step, the dominating method is cryogenic air separation.
The pure oxygen stream is then used in the combustion of the fuel. However, using
pure oxygen results in high peak-flame temperatures and therefore a part of the
flue gases are recycled to meet material constraints. The flue gases have a high
concentration of CO , over 80%, and the remaining part is mostly water which can
2
easily be removed by condensation. An advantage with this process is that the gas
volume is heavily reduced in the absence of nitrogen which makes flue gas treatment
cheaper. A disadvantage with oxy-fuel combustion is that the separation process to
obtain the oxygen is energy demanding. (Davison & Thambimuthu, 2005)
N 2 Energy CO 2
Fuel
Air Separation O 2 Combustion Flue gases Separation H 2O
Figure 4: Schematic overview illustrating the oxy-fuel combustion process.
Chemical looping combustion (CLC) is another version of the oxy-fuel combus-
tion concept. This method uses metal particles as oxygen carriers between two
fluidized bed reactors. In the first reactor the oxygen carriers are exposed to air
and are oxidized, they are then transported to the second reactor. In this reactor
fuel is injected and the oxygen carriers are reduced by the fuel to form water and
CO . The oxygen carriers are transferred back to the first reactor to close the loop.
2
(Moldenhauer et al., 2012)
5 |
Chalmers University of Technology | 2 CARBON CAPTURE AND STORAGE
2.4 Transportation and storage
Once the CO is captured it needs to be transported to a suitable storage site, this
2
can be carried out by either ships or pipelines. If the CO is to be transported using
2
the latter alternative, it needs to be compressed to a supercritical state, i.e. to a
pressure above 80bar. This increases the density of the fluid and thereby facilitates
the pumping, which is less energy demanding than transportation in a gaseous state.
If instead ships are used, the pressure is only raised to 7bar which transforms the
CO into liquid state. (IPCC, 2005)
2
The storage of CO can either be on- or off-shore and three types of geological
2
formations are suggested to be suitable: oil and gas reservoirs, unminable coal beds
or deep saline formations. Storing CO in oil and gas reservoirs would be benefi-
2
cial since this increases or maintains the pressure inside the reservoirs and thereby
enhances the oil recovery. Storage in coal beds, where mining is unlikely, would
give methane as a byproduct, since it is displaced when CO is injected. Deep sa-
2
line formations storage does not facilitates extraction of natural resources, however
global storage capacity in these formations is estimated to be far more extensive
than the previous alternatives. (IPCC, 2005)
6 |
Chalmers University of Technology | 3 THE PULP MILL
3 The pulp mill
Thepurposeofapulpmillistoprocesswoodlogsintopulpwhichcanbeusedtopro-
duce various paper products. Mechanical and chemical pulping are two approaches
used to produce pulp. The former method accomplish this by mechanically grind-
ing the wood logs. The latter method uses chemicals instead. Consequently, the
two processes produce pulps with greatly differing properties. There is also hybrid
processes such as CTMP (chemithermomechanical pulping) which uses chemicals to
soften the wood before it is mechanically grounded into pulp. However, the demand
for chemical pulp is larger and is thus the dominating technique. There exist a
few chemical pulping methods using different chemicals e.g. kraft cooking, sulphite
cooking and soda cooking. SCA O¨strands pulp mill uses the kraft process which is
described in the following three sections. For more information about the pulping
process, see Ek et al. (2009a) and Ek et al. (2009b).
3.1 Wood as raw material
A tree mainly consists of four compounds: cellulose, lignin, hemicelluloses and ex-
tractives. Pineiscommonlyusedasraw-materialinthepulpingprocessandtypically
has the composition shown in Figure 5.
Figure 5: Approximate composition of pine wood.
Cellulose is the main building block of the wood and is long linear polysac-
charides. Several cellulose chains can aggregate and form microfibrils which are the
building blocks of the fibres. These fibres are main component of the produced pulp.
Lignin is a polymer consisting of a branched network of different aromatic al-
cohols. However the exact structure of this complex molecule is not yet known.
The lignin is present to some extent in the fibres and gives stiffness. The space
between the fibres is mainly occupied by lignin which acts as a glue. The lignin is
the compound which is degraded in chemical pulping to free the fibres.
Hemicelluloses are, just like cellulose, polysacchariedes but they are branched
and have shorter chains. Their most important function is to give strength to the
cell walls in cooperation with lignin. Hemicelluloses are to some degree degraded in
7 |
Chalmers University of Technology | 3 THE PULP MILL
chemical pulping. However, this degradation should be minimized as the pulp yield
increases if hemicelluloses are kept intact.
Extractives are various compounds with low molecular mass and they constitute
a small percentage of the wood. They have various functions such as protecting
the tree against fungus and some are important for the metabolism of the cells.
Extractives are undesired in the pulp as they can cause stains and discolouration of
the paper products.
3.2 The kraft pulping process
A basic flowsheet describing the kraft pulping process, including the recovery system
for the cooking chemicals, is presented in Figure 6.
Steam Water Water Steam Water Steam
Wood Wood Impregnation Pulp Oxygen Pulp
Bleach plant
handling & Cooking washing delignification
Bleach filtrate
Bark Black liquor
Steam
Make-up lime
Washing and
Evaporation Recovery Water
Biofuel boiler reburning of
plant boiler
lime Energy
Make-up
Steam Lime (CaO)
Steam chemicals
Weak white liquor
Lime (CaCO)
3
Pulp Smelt Green White liquor
dissolver liquor preparation
Mass
Energy
White liquor
Figure 6: A schematicoverview illustratingthekraftpulping process. Theboldarrows represent
the flow of the wood or pulp through the process.
The first step in the process is the wood handling, of which the two most im-
portant steps are debarking and chipping. The bark of the tree has a high content
of extractives and a low amount of fibers, it may also contain sand. Thus, the bark
is removed in the first step of the process. The discarded bark is often used as fuel
by producing steam in a biofuel boiler. The debarked logs are then cut into small
chips, typically with dimensions in the following range: (20-30)x(15-30)x(3-8) mm.
By chipping the wood into small pieces of approximately equal size, the cooking
chemicals are distributed more evenly and faster into the wood.
8 |
Chalmers University of Technology | 3 THE PULP MILL
The wood chips pass a steaming vessel where air inside the chips is replaced by
steam. They are then transported to the impregnation vessel where warm impregna-
tionliquidisappliedwiththepurposeofraisingtemperatureandtoevenlydistribute
the cooking chemicals. The impregnation liquid is a mixture of white liquor contain-
ing the cooking chemicals i.e. sodium sulfide (Na S) and sodium hydroxide (NaOH),
2
but also black liquor containing used cooking chemicals. After the impregnation the
chipsenterthedigesterandthetemperatureisraisedto160-170◦Candconsequently
the cooking chemicals start to react. Sodium sulfide reacts with water to form hy-
drogen sulfide ions and hydroxide ions. Hydrogen sulfide ions are responsible for
degrading the lignin whereas the hydroxide ions neutralize acidic groups and keep
the degraded lignin in solution. The cooking chemicals are consumed during cooking
to form sodium carbonate (Na CO ) and sodium sulfate (NaSO ). The mixture of
2 3 4
degraded lignin and cooking chemicals is called black liquor.
After the cooking is terminated the pulp proceeds to the washing step. The
black liquor is displaced by washing water. Several different types of equipment can
be used for the washing step e.g. drum filters and pressurized diffusers.
Oxygen delignification is applied after cooking to further reduce the amount of
lignin present in the pulp. This operation has several benefits compared to prolong-
ing the cooking. The greatest one being that the pulp is bleached, resulting in a
decreased demand of other bleaching chemicals and thereby reduced pollution. In
addition this increases the strength of the pulp compared to prolonging the cooking.
Before the pulp can be shipped to the paper mill it needs to be bleached further
to a desirable brightness. This is conducted using several stages which utilizes
a variety of bleaching chemicals e.g. chlorine (C), chlorine dioxide (D), oxygen
(O), hydrogen peroxide (P) and chelating agent (Q). Bleaching with compounds
containing chlorine was common but has gradually declined since the 1970s and
today many mills uses bleaching sequences without chlorine due to environmental
concerns, OQP(OP)Q(PO) is an example of a chlorine free sequence. At this stage,
the pulp can be dewatered and transported to a paper mill.
3.3 The recovery system for cooking chemicals
The black liquor containing dissolved lignin and used cooking chemicals is removed
in the washing step and pumped to the evaporation plant. By evaporating a major
part of the water, the dry content is increased from 15-20% to 70-80%. This raises
the heat value of the black liquor substantially. The evaporation plant consists of
six or seven separate evaporators, which are coupled in such a way that the steam
formed in an evaporator is used to heat the next one. This reduces the energy
demand.
Toregeneratethecookingchemicalsandalsoutilizethelatentheatoftheorganic
components, the strong black liquor from the evaporation plant is combusted. This
is performed in a recovery boiler which is a large furnace with a height of 60-70m,
see Figure 7.
9 |
Chalmers University of Technology | 3 THE PULP MILL
Boiler Super-heater
Flue gases Economizer
bank bank
Electrostatic
presipitator
Ash recovery
Furnace
Ash recovery
Mixing
Strong black liquor
tank Tertiary air
Make-up chemicals
Liquor guns
Secondary air
Primary air
Smelt Smelt spouts
Figure 7: Cross-sectional view of the recovery boiler and its components.
The strong black liquor is sprayed into the furnace along with air. The air is
injected at several different heights which make it possible to form an reductive
environment close to the bottom where the combustion is incomplete, while higher
up in the furnace the atmosphere is oxidizing. After the strong black liquor is
injected it forms small drops which undergo drying, pyrolysis and char combustion
before reaching the bottom of the recovery boiler where a smelt is formed. In the
top layer of the smelt carbon reacts with sodium sulfate to regenerate the sodium
sulfide, see Reaction 2.
4C(s)+Na SO (l) −−→ Na S(l)+4CO(g) (2)
2 4 2
The heat released by combustion is absorbed by the walls, consisting of boiler
tubes. Heat is also extracted from the flue gases as they pass a super-heater bank,
boiler bank and economizer before being treated in a electrostatic precipitator to
remove fly ash. The steam produced in the recovery boiler is of high pressure (over
100bar) and is expanded in a back pressure steam turbine to produce electricity.
MP-steam (approximately 12bar) and LP-steam (approximately 4bar) is extracted
from the turbine to cover the need of the pulping process. The steam generation
efficiency, based on lower heating value, is usually around 75% for a recovery boiler
(Anderberg (2010); Vakkilainen & Ahtila (2011)). If a steam surplus is present,
which often is the case for a modern pulp mill with a high degree of heat integration,
10 |
Chalmers University of Technology | 3 THE PULP MILL
the steam is further expanded in a condensing turbine. In such cases the pulp
mill can be a major net exporter of electricity (Pettersson & Harvey, 2012). The
recovery boiler typically have an electrical efficiency around 12% (Pettersson, 2014).
In addition, the pulp mill also has a potential of providing district heating as there
often exist low temperature waste heat suitable for this purpose (Hektor, 2008).
The smelt is drained through smelt spouts into a smelt dissolver, basically a
large tank filled with weak white wash, which is a water solution containing a small
amount of cooking chemicals from other parts of the process. This results in a
solution called green liquor. In the white liquor preparation plant the green liquor
enters a slaker where reburned lime (CaO) is added which leads to the formation
of calcium hydroxide (Ca(OH) ), see Reaction 3. The solution then proceeds to the
2
causticization vessels where sodium hydroxide is formed by Reaction 4. Now that
all cooking chemicals are regenerated, the white liquor is filtrated and recycled back
to the cooking step. Calcium carbonate (CaCO ) is separated by the filtration and
3
washed before it enters the lime kiln. Here it is dried and heated to over 850◦ which
results in Reaction 5. The reburned lime can once again be used in the slaker and
the cycle is complete.
CaO(s)+H O −−→ Ca(OH) (s) (3)
2 2
Ca(OH) (s)+Na CO (aq) ←−→ 2NaOH(aq)+CaCO (s) (4)
2 2 3 3
CaCO (s) ←−→ CaO(s)+CO (g) (5)
3 2
11 |
Chalmers University of Technology | 4 BLACK LIQUOR GASIFICATION
4 Black liquor gasification
The recovery boiler suffers from several drawbacks such as corrosion, fouling and
smelt-water explosions (Pettersson, 2011). The latter occurs when water comes in
contact with the molten smelt and thereby instantaneously evaporates, causing a
pressure wave due to the increased volume. These explosions have been costly with
respect to both production losses and human lives (Anderson (1969); Grace (2008)).
ImprovementshavebeenmadeovertheyearsbuttherecoveryboileranditsRankine
steam cycle have some inherent limitations such as low thermal efficiency and low
power-to-heat ratio. In addition, modern pulp mills have a steam surplus which
motivates a better use of the energy contained in the black liquor, such as electricity
generation or biofuel production. All the facts stated above have been incentives for
examining other alternatives to recover the cooking chemicals. (Stigsson & Berglin,
1999)
Black liquor gasification is an alternative technology that has the potential to
solvemanyofthestatedproblems. Gasificationofspentcookingliquorwasexamined
as early as in the 1950s and 1960s. Since then, a number of research projects and
pilot plants have examined this process. However, most were discontinued since they
did not show any significant improvement compared to the conventional Tomlinson
recoveryboilerorotherreasonssuchasshiftingdevelopmentprioritiesbycompanies.
(Gebart et al. (2005); Consonni et al. (1998); Bajpai (2014))
Today two gasification technologies are competing; fluidized bed technology de-
veloped by ThermoChem Recovery International (TRI) and entrained flow techno-
logy supplied by Chemrec. The former process operates at a relatively low temper-
ature around 600◦C and near atmospheric pressure. The black liquor is injected
in the bottom of a fluidized bed and steam is used as gasifying agent. Part of the
produced syngas is combusted in heat tubes integrated in the fluidized bed to ensure
that high enough temperatures are reached (Dahlquist, 2013). TRI has built two
facilities using their technology, whereas Chemrec has installed one plant using their
air-blown atmospheric-pressure version and one demonstration plant utilizing their
oxygen-blown pressurized gasification technology (TRI (2012); Chemrec (2011)).
However, the Chemrec technology is considered to be the most commercially ad-
vanced alternative as the TRI technology has had problems reaching the needed
operating temperature to obtain a sufficient conversion (Dahlquist (2013); Bajpai
(2014)). Hence, only the Chemrec technology will be considered from here on and
a detailed description is provided in the following section.
4.1 Chemrec black liquor gasification
As previously mentioned Chemrec has developed two different technologies for black
liquor gasification: air-blown atmospheric-pressure gasification and oxygen-blown
pressurized gasification. The former technology is used as a booster to relieve over-
loaded recovery boilers by providing additional capacity. The latter technology is
13 |
Chalmers University of Technology | 4 BLACK LIQUOR GASIFICATION
designed to fully replace the recovery boiler. A schematic description of the oxygen-
blown pressurized gasification is presented in Figure 8. Here follows a brief descrip-
tion of the gasification process, for a more detailed description, see Ekbom et al.
(2005).
Figure 8: The pressurized, oxygen-blown, high-temperature black liquor gasification technology
developed by Chemrec. Adapted from Lindblom & Land¨alv (2007).
Before the black liquor enters the gasification process it is filtered to remove solid
impurities and the pressure is elevated to 32bar. After the pressure increase, the
black liquor is preheated to 120◦C to adjust the viscosity so that a good atomization
is achieved in the burner i.e. disintegration into a spray of fine droplets. Black liquor
and oxygen is injected into the entrained flow reactor through the burner where it is
atomized. The diameter of the droplets and the size distribution is adjusted so that
a high conversion of carbon is achieved, but equally important is the regeneration of
the cooking chemicals; therefore a high reduction rate of inorganic salts is needed as
well. Animportantdesignfeatureistheuseofoxygeninsteadofair,asoxidant. This
requires an on-site air separation unit such as a cryogenic air fractionation process.
Because of the use of oxygen as gasifying agent the temperature in the reactor
reaches values of 950-1000◦C, which is well above the melting point of the inorganic
salts. The melted inorganic salts flow downward into the quenching section, where
condensate from the gas cooling unit is injected and as a result the smelt droplets
are cooled and solidified. The droplets fall down into the condensate in the lower
section of the vessel where they are dissolved to form green liquor. To increase the
14 |
Chalmers University of Technology | 4 BLACK LIQUOR GASIFICATION
dissolutionrate, partoftheexitinggreenliquorisrecycledbacktothequenchvessel.
Theexitinggreenliquorenterstwoparallelheatexchangerswherethetemperatureis
reduced from 220◦C to 90◦C. The heat exchangers are used to preheat weak white
wash liquor and evaporator condensate to 205◦C, which then enters the quench
vessel and the bottom section of the gas cooler, respectively. The produced green
liquor will contain some dissolved gases that will be released when the pressure is
reduced. This gas is rejoined with the gas out of the gas cooler.
The gas formed in the gasification process is a mixture mainly composed of
CO, H , CO , H O and H S but it also contains minor amounts of N , CH and
2 2 2 2 2 4
COS. After the quench section the produced gas enters the counter-current gas
cooler which is a cylindrical pressure vessel with four cooling sections. In the first,
second and third section the gas releases heat to generate MP-steam and LP-steam.
To achieve the final cooling in the top section cooling water is utilized. As the
temperature is reduced thorough the cooler, water will condensate and fall down to
the bottom. This serves as a cleaning step as particulate material is removed from
the produced syngas by the falling condensate.
The syngas produced is primarily used in two different applications. One altern-
ative is to combust the syngas in a gas turbine to produce electricity and steam,
which is denoted black liquor gasification combined cycle (BLGCC). The other al-
ternative uses syngas as a feedstock for production of fuel and is denoted black
liquor gasification with motor fuel production (BLGMF). The concepts are further
explained in the following two sections.
4.2 Black liquor gasification combined cycle
The process schematics for the black liquor gasification combined cycle (BLGCC)
is illustrated in Figure 9. The first three operations; air separation, gasification and
gas cooling are explained in section 4.1.
The syngas produced in the gasification contains a considerable amount of H S,
2
since approximately half of the sulfur leaves with the gas phase, resulting in a con-
centrationofaround2mole-%onadrybasis(Ekbometal.,2005). Allsodiumleaves
the gasification dissolved in the green liquor. Hence, a split between sodium and
sulfur is achieved. This split of the cooking chemicals is one of the benefits with the
gasification process, since this presents opportunities to produce polysulfide cooking
liquors. The pulp yield can be increased by a few percent by using polysulfide cook-
ing liquors and consequently less wood is needed to produce a certain amount of
pulp. However, the split of sodium and sulphur comes at the price of an increase in
the causticizing load which leads to increased fuel consumption in the lime kiln. The
reasonis that less sulphur is present in the green liquor andtherefore the sodium will
form more Na CO instead of Na S, consequently a larger causticizing load follows,
2 3 2
see Reaction 4. (Lindstr¨om et al., 2006)
ToproducethealternativecookingliquorstheH Sneedstobecapturedfromthe
2
syngas. Another reason to capture the H S is due to emission regulations for SO ,
2 x
15 |
Chalmers University of Technology | 4 BLACK LIQUOR GASIFICATION
Weak
w wh ai st he G liqr uee on
r
Steam (CO2) Air Steam
† † † †
la)
m
lB iqla uc ok r n o it a c ifis a G Syngas g n ilo o c s a G Syngas n o it a z ir u h p lu s e DO C la nv oo im
t
pe r 2 Syngas e n ib r u t s a G Flue gas a e t s y r e v o c e r t a
e
H
r o t aG rS e nR eH g() HP-Steam e n ib r u t m a e t S
o
(
Flue
O2 Steam H2S Electricity
gas
Electricity Steam
s
tin
tin
u
Air
u
n o it N2
T
O C S
a S
r
a d
p n
e s
r iA
a
s u a
lC
Figure 9: Process schematic illustrating the black liquor gasification combined cycle. † These
processes are simulated in this work.
which are formed from H S in the combustion zone of the gas turbine. To avoid such
2
problems the sulphur levels in the syngas need to be reduced below 20ppmv before
entering the gas turbine (Korens et al., 2002). A commercial process that can reduce
sulphur levels to meet the requirements is the Selexol process. The solvent used in
this process is a mixture of dimethyl ethers of polyethylene glycol. The chemical
formula is CH O(CH CH O) CH , where n is between 2 and 9. The Selexol solvent
3 2 2 n 3
utilizesphysicalabsorptiontoremoveacidgases. Hence, nochemicalreactionoccurs
which is an advantage compared to chemical solvents, such as MEA, which degrades
over time due to formation of heat-stable salts (Breckenridge et al., 2000). The
Selexol process also has a high selectivity of H S over CO , which makes is suitable
2 2
for gas turbine applications, since the co-absorption of CO can be minimized. One
2
downside with the Selexol solvent is its limited ability to absorb COS. A solution
is to add a COS hydrolysis unit before the Selexol process. By addition of steam,
COS reacts and forms CO and H S in the hydrolysis unit, and consequently the
2 2
concentration of COS is reduced from over hundred ppmv down to 10ppmv. (Kubek
et al., 2000)
In the Selexol process (Figure 10) the solvent and the syngas are contacted in a
counter-current packed absorber operating at high pressures, 20-140bar (Maxwell,
2004). The sulfur compounds are absorbed and some CO is inevitably also co-
2
absorbed. The H S-rich solvent is then passed through a second column where N is
2 2
16 |
Chalmers University of Technology | 4 BLACK LIQUOR GASIFICATION
Clean syngas
CO2 Selexol recycle
Absorber
Selexol recycle
CO2
H2S
AbH so2 rS ber StH rip2S per
H2S
Concentrator
N2
Syngas
Figure 10: Simplified flowsheet of the Selexol process with CO capture, based on the work of
2
Field & Brasington (2011a).
added as stripping gas. Most of the absorbed CO is stripped from the solvent and
2
recycled back to the absorber, whereas the H S-rich solvent enters a second stripper.
2
In this packed stripper heat is applied in the reboiler to regenerate the solvent and
to produce a gas stream rich in H S. The solvent is recycled back to the absorber
2
and the H S-rich gas stream is sent to a Claus unit for further processing. If CO
2 2
capture is desired the Selexol process can be modified to meet such requirements as
well. By letting the sulfur free syngas pass a second absorber the CO is absorbed.
2
The solvent is regenerated by lowering the pressure and the flashed CO can be
2
compressed and transported to storage. (Field & Brasington, 2011a)
The H S-stream sent to the Claus unit should have a concentration of at least
2
40mole-%. Ifthisisfulfilledthestraight-throughClausprocesscanbeused, whichis
the least complex alternative. Otherwise alternatives such as direct oxidation can be
used for lower concentrations. In the Claus process the H S gas stream is introduced
2
in a reaction furnace together with air. The reaction occurring in the combustion
zone converts a large portion of the H S into elemental sulphur. To convert the
2
17 |
Chalmers University of Technology | 4 BLACK LIQUOR GASIFICATION
remaining part of the H S, the gas is passed over a number of catalytic beds. To
2
recover essentially all sulphur, which is important to reduce raw material costs, a
Shell Claus Off-Gas Treating (SCOT) unit can be added after the Claus unit. With
this unit over 99.7% of the sulphur is recovered (Linde Process Plants Inc., 2012).
As previously mentioned the elemental sulphur can be used to prepare polysulphide
cooking liquor by dissolving it in the green liquor formed in the gasifier:
3S(aq)+Na S(aq) −−→ Na −S −S(aq) (6)
2 2 3
The clean syngas is combusted in a gas turbine to produce electricity. The flue
gases then pass through a heat recovery steam generator where HP-, MP-, and LP-
steam is produced. The MP- and LP-steam is used where it is needed in the pulping
process. The HP-steam is passed through a steam turbine to produce additional
electricity. If a steam surplus is present in the pulp mill, the steam can be expanded
further in a condensing turbine to increase electricity production.
4.3 Black liquor gasification with motor fuel production
The process schematic for motor fuel production is shown in Figure 11. DME has
been chosen as a product but several other alternatives exist, such as methanol,
hydrogen or FT-diesel. DME is chosen since this fuel can be used in modified diesel
engines and is considered a fairly realistic alternative to be implemented in the years
before 2020 (Pettersson & Harvey, 2012). This perception was further strengthened
by the European BioDME project which was successfully concluded in 2012. In
this project DME produced in Chemrecs pilot plant showed a great well to wheel
efficiency, compared to other biofuels, after a driving distance of 450000km by the
test fleet of Volvo trucks. The energy consumption was 270MJ/100 km compared
to e.g. 440MJ/100 km for ethanol from wheat straws (Salomonsson, 2013).
The gasification process is identical to that of the BLGCC alternative. However,
the gas cleaning process differs substantially as the DME production has a more
stringent requirement of the sulphur level in the syngas. To avoid deactivation
of the catalyst in the DME synthesis a sulphur concentration below 0.1ppmv is
required. The Selexol process is unable to meet this specification and hence another
gas cleaning process must be used (Korens et al., 2002). The DME production is
also inhibited by the presence of CO . The reason is that CO is a reaction product
2 2
in the formation of DME and therefore shifts the equilibrium towards the reactants.
Hence, the CO concentration should be reduced to below 3mole-%. The Rectisol
2
process is capable of producing a clean synthesis gas with the requirements specified
above. (Ju et al., 2009)
The Rectisol process was developed in the 1950s and is today widely used in
sour gas treatment. Methanol is utilized as solvent in this process and it separates
the unwanted species, i.e. CO , H S and COS from the remaining gas by physical
2 2
absorption (Sun & Smith, 2013). Unlike the Selexol process, the Rectisol process
is capable of removing COS to satisfying levels and hence a COS hydrolysis unit is
18 |
Chalmers University of Technology | 4 BLACK LIQUOR GASIFICATION
Weak
w wh ai st he G liqr uee on
r
Steam CO2 Steam
†
lB iqla uc ok r
n
o it a c ifis
a G
Syngas
g
n ilo o c
s a G
Syngas
n o it
a z ir u h
p lu s e
Dl Oa v Co
m
d
ne ar 2 Syngas
sise
h t n y s
E M D
DME
n o it
a llits id
E M D
Purge
O2 Steam H2S Steam
gas
DME
s
tin
tin
u
Air
u
n o it N2
T
O C S
a S
r
a d
p n
e s
r iA
a
s u a
lC
Figure11: ProcessschematicofblackliquorgasificationwithDMEproduction. †Theseprocesses
are simulated in this work.
redundant. The solubility of CO and H S differs to such an extent that selective
2 2
removal is possible. Also the solubility of sour gases, and thereby the absorption
rate, is enhanced at high pressures and low temperatures. Hence, Rectisol processes
are operated at temperatures between −20 to −60◦C and pressures in the range of
30-80bar (Weiss, 1988).
A modified version of the Rectisol process, provided by Gatti et al. (2013), is
presented in Figure 12. The syngas produced in the gasifier enters the bottom
of a packed absorber divided into two sections. Refrigerated methanol enters the
absorber and after passing through the top section, about 45% of the methanol,
containing almost exclusively absorbed CO , is withdrawn. The remaining solvent
2
flows through the bottom section and exits the absorber, rich in both CO and
2
H S. Both streams are flashed to recirculate co-absorbed CO and H . After further
2 2
pressure reduction the streams enters two stripper columns to desorb the CO . The
2
CO rich stream enters the top of the strippers whereas the stream containing H S
2 2
enters in the bottom, this is to ensure that only a small amount of H S ends up in
2
the CO stream for sequestration. A stripper, equipped with a reboiler and partial
2
condenser, regenerates the methanol which is recycled to the absorber. The top
product from the stripper is a gas stream with high H S-concentration which is to
2
be further processed in the Claus unit.
The Claus and the SCOT unit is identical to those in the BLGCC case, for
19 |
Chalmers University of Technology | 4 BLACK LIQUOR GASIFICATION
Clean syngas CO2 H2S
Methanol recycle
Absorber StC riO pp2 er StC riO pp2 er StrH ip2S per
Syngas
Figure 12: Simplified flowsheet of the Rectisol process, based on Gatti et al. (2013).
more information see Section 4.2. The clean syngas out from the Rectisol process
enters the DME synthesis. The conventional manufacturing process of DME is an
indirecttwostepsynthesisroute, wheremethanolisproducedviamethanolsynthesis
and then converted to DME by dehydration. A new technology has recently been
developed by Ohno et al. (2006). The new process is a direct synthesis where DME
is produced from the syngas according to this overall reaction:
3CO+3H −−→ CH OCH +CO (7)
2 3 3 2
The new process is operated at 50bar and 260◦C in a slurry bed reactor contain-
ing a catalyst (Ohno et al., 2006). A syngas composition of 1:1 molar ratio of H
2
and CO gives highest conversion. The conventional process requires a molar ratio
of 2:1 (Ju et al., 2009). The syngas out of the gasifier has a molar ratio of 1.03
which means that the new process eliminates the need of a water-gas shift reactor to
adjust the ratio (Ekbom et al., 2005). However, the major advantage with this new
process is that it overcomes the equilibrium limitations of the methanol synthesis
and thereby gives higher conversion (Marchionna et al., 2008). The gas mixture
produced in the DME reactor is separated in a series of distillation columns. Un-
reacted gas and formed methanol are recycled back to the reactor. As can be seen
in Reaction 7, CO is formed in the DME production. However, the possibility to
2
capture this carbon dioxide has not been considered in this work.
20 |
Chalmers University of Technology | 5 METHOD
5 Method
In an initial literature review the pulp mill was studied to determine the most
probable development of the pulping process in the nearest future. Three scenarios
were identified. The first one is the“business as usual”case where the conventional
recovery boiler continues to be the technique used for regeneration of the cooking
chemicals. In the second scenario the recovery boiler is replaced with a gasifier
which generates syngas that is used to produce electricity in a gas turbine. The
third scenario also utilizes a gasifier but instead of producing electricity the syngas
isusedasrawmaterialforDMEproduction. Inthisworkthepossibilityofcapturing
CO in these three scenario are examined, see Table 1. In the first scenario post-
2
combustion capture using the MEA process is examined. For the second scenario
pre-combustion capture with the Selexol process has been determined to be the most
suitable. Pre-combustion capture using the Rectisol process is applied to the third
scenario.
Table 1: Three possible scenarios for the future development of the pulp mill. The last letter in
the abbreviation indicates which capture process is applied to that scenario.
Recovery Additional Capture Capture
Scenario
system product technology process
Recovery
RB-M N/A Post-combustion MEA
boiler
Black liquor
BLGCC-S Electricity Pre-combustion Selexol
gasification
Black liquor
BLGMF-R DME Pre-combustion Rectisol
gasification
5.1 Process simulation
ToevaluatethesuitabilityandthespecificcostofcapturingCO ,processsimulations
2
of the capture processes are performed. The software used is Aspen Plus 8.0, section
5.5 gives a brief account of the capabilities of this modelling tool.
ThecostofcapturingCO canbeestimatedifeachofthescenariosisdividedinto
2
one case with CO -capture and one without capture. These cases will henceforth
2
be denoted ”CCS”and ”NC”, respectively. For the RB-M-scenario, the NC-case is
simply the present day situation where the flue gases are vented to the atmosphere
and hence will not need any modelling. A remark worth noticing is that only the
flue gases from the recovery boiler have been considered, but it is also possible to
captureCO fromthebiofuelboiler. Duetolackofprocessdataregardingthebiofuel
2
boiler and its flue gases, this possibility has been omitted in this work. However,
21 |
Chalmers University of Technology | 6 MODELLING
6 Modelling
This chapter presents the input data used in the modelling work and then gives a
thorough review of the modelling approach. Assumptions regarding the modelled
components and their design specifications are also presented.
6.1 Input data
This thesis is a case study on SCAs pulp mill in O¨strand, hence the simulations are
based on data obtained from their engineering department. However, only the most
essential data has been provided, i.e. the mass flow of black liquor and the oxygen
concentration and volume flow of flue gases. Remaining data has been extracted
from other sources and where applicable scaled to match the pulp production rate
of SCAs pulp mill. The flue gas stream entering the RB-M-scenario is based on
Hektor (2008) in addition to the process data provided by SCA. The specifications
for the syngas stream entering both the BLGCC-S- and BLGMF-R-scenarios has
been determined from Gebart et al. (2011), Ekbom et al. (2005) and Kubek et
al. (2000), as well as process data from SCA. From the aforementioned sources,
the composition, temperature, pressure and mole-flow of the streams entering the
capture process have been determined, see Table 2. For all three scenarios, the
simulated processes are adapted to capture 85% of the ingoing CO .
2
Table 2: Process stream data used in the simulations. The composition is specified as mole-%
except the COS-concentration where ppmv is used.
n˙ P T
Scenario H CO CO H S CH O N H O COS
[mol/s][bar] [◦C] 2 2 2 4 2 2 2
RB-M 5014 1.0 110 0 0 13.3 0 0 4.4 63.3 19.0 0
BLGCC-S 865 31.5 30 39.2 38.1 19.0 1.9 1.3 0 0.2 0.2 10
BLGMF-R 865 31.5 30 39.2 38.1 19.0 1.9 1.3 0 0.2 0.2 122
6.2 RB-M-scenario
For the RB-M-scenario where post combustion capture with MEA is applied, the
property method Electrolyte NRTL is used for the simulations in Aspen Plus. This
method can handle aqueous and mixed solvent systems for a wide range of con-
centrations and hence should give an accurate representation of the behaviour of
the compounds involved in this process. The physical property data is provided in
the Aspen Plus rate-based MEA template, which has been used as a starting point
for the simulations. This data contains reaction kinetics which has been validated
25 |
Chalmers University of Technology | 6 MODELLING
against industrial applications. A number of reactions occur in the system at hand
butthesevenreactionsdefinedbelowareconsideredtobemostinfluential. Reaction
8-10 are assumed to be in equilibrium, whereas Reaction 11-14 are rate-controlled
reactions (Aspen Technology, 2012b).
H O+MEAH+ ←−→ MEA+H O+ (8)
2 3
2H O ←−→ H O+ +OH− (9)
2 3
HCO − +H O ←−→ CO 2− +H O+ (10)
3 2 3 3
CO +OH− −−→ HCO − (11)
2 3
HCO − −−→ CO +OH− (12)
3 2
MEA+CO +H O −−→ MEACOO− +H O+ (13)
2 2 3
MEACOO− +H O+ −−→ MEA+CO +H O (14)
3 2 2
6.2.1 RB-M-CCS-case
The flowsheet from Aspen Plus is presented in Figure 13 in which all components
andstreamsarenamed. Thissectionwillexplainthegeneralassumptionsanddesign
parameters chosen for the simulation. All auxiliary equipment such as coolers, heat-
ers, pumps, flashes etc. are explained in more detail regarding modelling approach
and assumptions in Section 6.5, as they are identical for the three different scenarios.
However the modellingof columns willbe explainedin detail for eachprocess as they
differ from case to case.
Figure 13: Aspen Plus flowsheet of the MEA capture process for the RB-M-CCS-case.
26 |
Chalmers University of Technology | 6 MODELLING
The flue gases from the recovery boiler enters the capture process at ambient
pressure and a temperature of 110◦C. Before the flue gases is fed to the absorber
the temperature is reduced to 40◦C in COOL-1, but since a considerable amount
of water is present some of it will condensate. FLASH-1 separates the condensed
water and the remaining flue gases, which leaves in the top of the flash and enters
the bottom stage of the absorber, ABS-1. In the top a solution of MEA and water
enters the absorber at 40◦C. A weight ratio of 30% MEA is commonly used and
this corresponds to a mole ratio of 0.126. The mole ratio is used as a measurement
to maintain the specified proportions between water and MEA. Since the solution
entering the absorber is a recycle stream it will contain a certain amount of CO
2
bound by MEA in the form of MEACOO–. Therefore the weight ratio between
all MEA species and water would give an incorrect value as the molar weight of
MEACOO– is higher than that of MEA.
The absorber is modelled as a packed column filled with Mellapak 250Y, which
is a structured packing. A rate-based approach using the ”RadFrac”block has been
applied for the simulations meaning that transport resistance is included in the
model. For the liquid phase, diffusional film resistance is assumed with reactions
occurring in the film. The film is discretized using 10 points, see Table 3. The
reactions are fast and most CO will react with MEA as soon as it enters the
2
liquid (Kothandaraman, 2010). The discretization points are therefore concentrated
towards the liquid-vapour interface, where the gradients will be the largest. No
reactionsareassumedtotakeplaceinthevapourphaseandthereforeonlydiffusional
resistance is taken into account in the vapour film.
Table 3: Locations of the discretization points in the liquid film. The liquid-vapour interface is
located at 0 and 1 corresponds to the bulk liquid. (Kothandaraman, 2010)
Point 1 2 3 4 5 6 7 8 9 10
Distance 0 0.001 0.005 0.010 0.050 0.100 0.150 0.200 0.300 1
Theabsorberisdividedinto20stages. Incontrasttoequilibriumcalculations, an
increased number of stages do not increase the separation performance of a column,
instead more accurate calculations are obtained as the column is divided into more
calculation steps. However, using more stages increases the number of equations
solved and hence there is a trade-off between accuracy and computational time. The
height of the column is set manually to achieve a certain capture rate. The diameter
on the other hand is calculated using the flooding approach in Aspen Plus. In this
approach the stage with the highest vapour and liquid flows are chosen as a base
stage. The column diameter is then adjusted so that fluxes on this stage are 70%
of that corresponding to flooding, a state where a too high vapour flux prevents the
liquid from flowing downwards. The most important design parameters and results
from the absorber is presented in Table 4.
27 |
Chalmers University of Technology | 6 MODELLING
Table 4: Design parameters and dimensions of ABS-1.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
ABS-1 Mellapak-250Y 20 1 9.8 15
Two product streams exits the absorber. The top product is the flue gases now
CO lean. A fraction of the stream is MEA-slip which has to be recovered from
2
the flue gas to reduce the cost of make-up MEA. A packed wash column, with the
dimensions presented in Table 5, is added to accomplish this task. In the column,
water from FLASH-1 and some make-up water flows counter-current to the flue
gas flow and removes MEA. The clean flue gases leaving the wash column contains
2.2mol-% CO .
2
Table 5: Design parameters and dimensions of WASH.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
WASH Mellapak-250Y 10 1 6.3 3
The CO absorbed in the liquid solvent exits the absorber at the bottom stage
2
mainly as MEACOO–. The amount of CO in a stream is expressed in the form
2
of loading, i.e. the ratio of all species containing CO and all MEA-species. In this
2
study the optimal loading for the rich solvent exiting the absorber was found to
be 0.536. PUMP-1 elevates the pressure of the bottom product to 1.8bar before it
enters a heat exchanger (HE-1) which raises the temperature to 101◦C. To model
HE-1, the ”MHeatX”block is used as this block is suitable when recycle streams is
involved, sincethisblocktearstheenergystreamthusfacilitatingeasierconvergence.
Theheatedstreamentersthestripper(STR-1)onthesecondstageoutof20intotal.
The stripper is modelled with the same assumptions regarding mass transfer and
discretization of the liquid film. In contrast to the absorber the stripper is equipped
with both a partial condenser and a reboiler. However, the partial condenser is
modelled as a cooler and a flash to ease convergence. Table 6 summarizes the design
parameters of the stripper.
Table 6: Design parameters and dimensions of STR-1.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
STR-1 Mellapak-250Y 20 1.8 6.4 15
The bottom stream from the reboiler exits STR-1 with a lean loading of 0.284
and a temperature of 118◦C. This lean loading was found to be the optimal value
to obtain a low reboiler duty. To minimize the utility demand of the process this
stream is heat exchanged in HE-1 with the inlet stream to the stripper. VALVE-1
28 |
Chalmers University of Technology | 6 MODELLING
then lowers the pressure to 1 bar. Small amounts of MEA are lost in both the
flue gases and the CO -stream and hence a make-up stream is added. In addition,
2
the water from the wash column is added to adjust the ratio between MEA and
water. The amount of make-up MEA added is determined by a balance block in
Aspen Plus, which performs a component balance over all inlet and outlet streams.
Before the regenerated absorbent is fed to the absorber once again, the temperature
is reduced to 40◦C by COOL-3.
The gas stream leaving the partial condenser of STR-1 has a CO -concentration
2
of 99.0mol-% and is compressed to 80bar in a multistage compressor consisting of 4-
stageswithintermediatecoolingto25◦C. Atsuchhighpressuresthegasmixturewill
be in a supercritical state and therefore a pump can be used to increase the pressure
to the 150bar required for transportation and storage (Field & Brasington, 2011b).
However, the Electrolyte NRTL model only gives accurate results up to medium
pressures i.e. tens of atmospheres. Hence, this section of the flowsheet is modelled
separatelyusingamodelmoresuitableforhighpressuresnamely,thePSRKproperty
method which is based on the Predictive Soave-Redlich-Kwong equation-of-state
model (Aspen Technology, 2010). Figure 14 shows the flowsheet of the compression
section. The streams out from the multistage compressor labelled L1, L2 and L3
are condensed water.
Figure 14: Aspen Plus flowsheet of the CO compression section.
2
6.2.2 RB-M-NC-case
The RB-M-NC-case is equivalent to the present situation at a market pulp mill.
Hence, no modelling in Aspen Plus is necessary. This case is however important for
the study as it will be compared with the previously described capture case.
6.3 BLGCC-S-scenario
For the BLGCC-S-scenario, the Selexol process will be used to capture CO . The
2
Aspen Plus rate-based DEPG template is used as a starting point for the simula-
tions (Aspen Technology, 2012a). The solvent trademarked under the name Sel-
exol is, as explained in Section 4.2, a mixture of dimethyl ethers of polyethylene
29 |
Chalmers University of Technology | 6 MODELLING
glycol (DEPG) with a chemical formula of CH O(CH CH O) CH . The weight-
3 2 2 n 3
distribution between DPEG of varying chain length is provided by the template and
is presented in Table 7.
Table 7: Weight-distribution of DEPG-compounds.
Chain length (n) 2 3 4 5 6 7 8 9
Weight-% 0 6 23 26 21 14 7 3
The Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT) is used
as property method. This method determines the intermolecular forces by dividing
them into repulsive and attractive forces and calculate contributions from different
segmentsofthemolecules(Field&Brasington,2011b). Hence, thismodelissuitable
for polymers or other compounds with repeating units such as DEPG. Physical
data used in the template has been regressed against vapour-liquid equilibrium data
(Aspen Technology, 2012a).
6.3.1 BLGCC-S-CCS-case
The model of the Selexol process for carbon capture is based on the model construc-
ted by Field & Brasington (2011a). Figure 15 presents a flowsheet of the process,
including numbering of units and streams.
Figure 15: Aspen Plus flowsheet of the Selexol capture process for the BLGCC-S-CCS-case.
Thegasifierproducessyngasatapressureof31.5bar. Beforeenteringthebottom
stage of the absorber (ABS-1), the syngas is compressed and cooled to 52bar and
30 |
Chalmers University of Technology | 6 MODELLING
35◦C. At the top stage a Selexol stream enters, which is partially loaded with
CO . The purpose of ABS-1 is to absorb essentially all H S and COS. A rate-
2 2
based modelling approach using the ”RadFrac” block is applied for all columns in
this simulation. Film diffusion resistance has been applied, however no reactions are
modelledsinceitisaphysicalabsorptionprocess. Toeaseconvergenceofallcolumns
inthissimulationtheconvergencemethodissettoSum-Rateswhichisrecommended
for wide-boiling mixtures. The design parameters for ABS-1 are presented in Table
8.
Table 8: Design parameters and dimensions of ABS-1.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
ABS-1 IMTP 50mm 20 52 1.9 20
The H S rich solvent stream exits the absorber below the bottom stage and is
2
heated to 105◦C before entering the H2S-concentrator column (H2S-CONC). The
column uses nitrogen as stripping gas with the purpose of releasing most of the
absorbed CO and recycle it to ABS-1. See Table 9 for design specifications of
2
H2S-CONC.
Table 9: Design parameters and dimensions of H2S-CONC.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
H2S-C. IMTP 50mm 20 52 1.3 26
The bottom stream from H2S-CONC is expanded in VALVE-1 to the operating
pressure of the H S-stripper (STR-1) which is 2bar. STR-1 is equipped with both
2
a reboiler and a partial condenser. However the partial condenser is modelled as
a cooler and a flash for two reasons. Firstly to ease convergence and secondly to
enable the possibility to extract a bleed stream, mostly consisting of water, from the
recycle stream entering the top stage of the stripper. The simulations predicts the
temperature in the reboiler to be 206◦C which is an unreasonably high value. Field
& Brasington (2011b) states that the Selexol process can be modelled accurately
in Aspen Plus with the exception of the stripper, where the actual temperature
should be around 130◦C. One possible reason for this deviation could be that the
vapour-liquid equilibrium data used in the regression of physical properties was
inadequate for DEPG6, DEPG7, DEPG8 and DEPG9 (Aspen Technology, 2012a).
Consequentlythemodelpredictsatoohightemperatureforevaporatingthemixture
present in the reboiler. Since the temperature is over 186◦C, MP-steam cannot be
used to provide the needed energy, but MP-steam has been assumed anyway as this
would be used in reality. Another consequence of the high temperature is that the
cooler (COOL-4) after the reboiler will have a larger duty. Otherwise this should
not affect any other results.
31 |
Chalmers University of Technology | 6 MODELLING
Table 10: Design parameters and dimensions of STR-1.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
STR-1 IMTP 50mm 20 2 1.7 5
The top product from STR-1 has a H S-concentration well above the required
2
40mol-%, see Section 4.2. The lean solvent exits the bottom of the reboiler and is
cooled to a temperature of 15◦C by COOL-4. After an pressure increase to 52bar
by PUMP-2, the lean solvent enters the top stage of the second absorber (ABS-2).
The purpose of this absorber is to reduce the CO -concentration of the syngas below
2
3mol-%. See Table 11 for design specifications.
Table 11: Design parameters and dimensions of ABS-2.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
ABS-2 IMTP 50mm 30 52 3.4 32
A Selexol stream, loaded with CO and some co-absorbed H and CO, is extrac-
2 2
ted from the bottom of ABS-2. This stream is divided into two by SPLIT-2, where
a fraction of 0.91 enters the flash section of the process and the remaining part is
recycled back to ABS-1. FLASH-2 and FLASH-3 operates at 14bar and 6.2bar
respectively, and by lowering the pressure the co-absorbed H and CO is desorbed
2
along with some CO . After re-compression and cooling to 52bar and 15◦C this
2
gas recycle is joined with the flue gases entering ABS-2. The bottom product from
FLASH-3 is further expanded to 1.5bar in FLASH-4, where almost all CO is re-
2
leased. From FLASH-4 the lean solvent is recycled back to ABS-2 and enters stage
8 at 52bar and 15◦C. Small amount of solvent is present in all of the outlet streams
and to compensate for this, a balance block calculates the component balance for
each DEPG fraction and add the correct amount to the ”SELEX-MU”-stream.
A CO -concentration of 95mol-% is obtained in the stream sent to the compres-
2
sion section which is modelled in a separate flowsheet, as the ”PC-SAFT”model is
not accurate for critical conditions. The compression section is identical to the one
presented for the RB-M-CCS-case, see Section 6.2.1 for details. The only difference
is that the stream that is to be compressed contains essentially no water, hence no
water is condensed during compression.
6.3.2 BLGCC-S-NC-case
As explained in Section 4.2 the Selexol process is still needed even if carbon capture
is not applied since the H S has to be removed from the syngas. However, if CO
2 2
removal is not required the process can be greatly simplified as the second absorber
and the associated flash section is redundant. Figure 16 presents the flowsheet of
the Selexol process without carbon capture.
32 |
Chalmers University of Technology | 6 MODELLING
Figure 16: Aspen Plus flowsheet of the Selexol process for the BLGCC-S-NC-case.
From the flowsheet it can be observed that the layout of the bottom section is
identical to the case with capture, the only exception being that the recycle of lean
solvent from the stripper now enters ABS-1 instead of ABS-2. However sizes of the
equipment will vary, see Table 12.
Table 12: Design parameters and dimensions of columns.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
ABS-1 IMTP 50mm 30 52 2.0 30
H2S-C. IMTP 50mm 20 52 1.2 12
STR-1 IMTP 50mm 20 2 1.7 5
6.4 BLGMF-R-scenario
For the BLGMF-R-scenario the Rectisol process is used to capture CO and H S.
2 2
TheAspenPlusrate-basedMEOHtemplateisusedasastartingpointsincetheRec-
tisol process utilizes methanol as solvent (Aspen Technology, 2012c). This template
also uses the PC-SAFT as property method. Physical properties in the template
have been regressed against experimental data.
6.4.1 BLGMF-R-CCS-case
The design of the process is based on the single stage Rectisol process found in Gatti
et al. (2013) and Munder et al. (2010). Figure 17 shows the finalized Aspen Plus
flowsheet of the process.
The syngas produced in the gasifier contains a considerable amount of water. It
has to be removed or else it will freeze and block the pipes. Water removal can be
achievedbycoolingthegasandtherebycondensingthewater. Hence,isitpossibleto
33 |
Chalmers University of Technology | 6 MODELLING
Figure 17: Aspen Plus flowsheet of the Rectisol capture process for the BLGMF-R-CCS-case.
perform this in the gas cooler in the gasification section of the process and therefore,
this operation is considered to be outside the boundary of this simulation. Because
of this, the water is removed by a ”Sep”block that has no utility consumption. This
component separator is an artificial unit, without a real counterpart, that separates
the flow by specified component split fractions.
The absorber (ABS-1) operates at a pressure of 60bar and therefore the syngas
is compressed and subsequently cooled to a temperature of −34◦C before entering
the column at the bottom stage. All columns in this case are modelled using a
rate-based approach with a ”RadFrac” block. Diffusional resistances are assumed
for both liquid and vapor phase. See Table 13 for design specifications of ABS-1.
The purpose of the column is to absorb CO and H S to such an extent that the
2 2
syngas exiting the column fulfils the requirements to be used as feedstock for DME
production, see Section 4.3. The syngas leaving the column has a CO concentration
2
of 0.2mole-% and contains 0.092ppmv of total H S and COS. Methanol enters the
2
top stage of the column at 60bar and −50◦C. The low temperatures present in
the column favours the absorption of CO and H S over the other gases present.
2 2
Since this physical absorption into methanol is an exothermic process the produced
heat needs to be removed, so that the benefits of a low absorption temperature can
be utilized throughout the whole column. This is achieved by adding cooling loops
inside the absorber. In the flowsheet the cooling loops are modelled by specifying
side duties which extract heat from certain stages of the column, namely 14 and 18.
Two liquid product streams exits the absorber. At stage 18, about 45% of the
methanol is withdrawn. This methanol stream is loaded with CO but essentially
2
free of H S. As H S has a higher solubility than CO in methanol, almost all H S
2 2 2 2
34 |
Chalmers University of Technology | 6 MODELLING
Table 13: Design parameters and dimensions of ABS-1.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
ABS-1 IMTP 75mm 30 60 1.4 21.7
will leave the column in the bottom product. Both of these streams enters a flash
(FLASH-1 and FLASH-2) where the pressure is reduced to release co-absorbed CO
and H which is then recycled to the syngas inlet. From the bottom of FLASH-1
2
the methanol stream loaded with CO exits and is thereafter cooled to −50◦C. This
2
stream is then split into two streams, 90% enters the first stripper (STR-1) and the
remaining part enters the second stripper (STR-2). The bottom product of FLASH-
2 is heated and then also enters STR-1 but on the bottom stage. STR-1 is a packed
column operating at 6bar, see Table 14. The lower operating pressure of STR-1 res-
ults in that absorbed gases will be released. Since the methanol stream only loaded
with CO enters the top of the column, a very pure top product, consisting of CO
2 2
and almost no H S, is obtained. The bottom product still contains a considerable
2
amount of CO , some of which is released and recycled by FLASH-3 that operates
2
at 2bar.
Table 14: Design parameters and dimensions of STR-1.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
STR-1 IMTP 75mm 30 6 1.5 15
TheliquidstreamfromFLASH-3containingbothCO andH Sentersthesecond
2 2
stripper (STR-2) at the bottom stage, see Table 15 for specifications. In the top of
STR-2 the remaining 10% of the absorbent only containing CO enters. A 1-stage
2
Rectisol process normally uses N as stripping gas in the second absorber. However,
2
that is unsatisfactory in this application since an diluted CO -product would be
2
obtained. To reach the same amount of recovered CO a reboiler is added to STR-2.
2
Over the top of STR-2 a stream of almost exclusively CO is obtained.
2
Table 15: Design parameters and dimensions of STR-2.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
STR-2 IMTP 75mm 30 2.7 0.9 20
The CO -streams out from STR-1 and STR-2 are mixed, resulting in a CO -
2 2
concentration of 96 mole-%. This stream is then elevated to a pressure of 150bar in
the compression section. As previously stated the PC-SAFT method is not accurate
in the supercritical range and thus is the compression section modelled in a separate
flowsheet, see Section 6.2.1 for a description.
35 |
Chalmers University of Technology | 6 MODELLING
The bottom product of STR-2 is feed to stage 5 of STR-3 which is a stripper
equipped with both a partial condenser and a reboiler. STR-3 operates at 1.2bar
to enhance the desportion of both CO and H S. The desorbed gases exit at the
2 2
top and consists of 42 mole-% H S and the remaining part is almost exclusively
2
CO . This stream is further processed in a Claus plant to obtain elemental sulfur
2
which can be used to produce polysulfide cooking liquor. An almost pure methanol
stream (total impurities lower than 5ppmv) is obtained from the bottom of STR-3.
This stream is recycled back to the inlet of the absorber. However, before it is fed
to the top stage both pressure and temperature is adjusted to 60bar and −50◦C.
There will also be small losses of methanol in the gas streams leaving the system,
a balance block is therefore added to adjust the make-up methanol injected to the
recycle stream.
Table 16: Design parameters and dimensions of STR-3.
Name Packing material Stages Pressure [bar] Diameter [m] Height [m]
STR-2 IMTP 75mm 30 1.2 2.4 5
6.4.2 BLGMF-R-NC-case
If CCS is not applied to the BLGMF-R-scenario this will have almost no effect on
the Rectisol process, since CO still has to be removed from the syngas stream, see
2
Section 4.3. The only difference is that the CO stream is vented to the atmosphere
2
instead of being compressed. Hence, the compression section is removed for this
case. Otherwise the process is identical to the case with capture both regarding
flowsheet layout and unit sizes. Thus, no additional modelling is needed for this
case.
6.5 Auxiliary equipment modelling
The processes described in the previous sections have several basic unit operations
in common such as pumps and valves. The same modelling approach has been
used for these equipment throughout the simulations and the description in the
following sections is therefore valid for all modelling cases. Information about the
modelling blocks described in this section has been collected from Aspen Plus Help
documentation.
6.5.1 Compressor
Compressorsaremodeledwiththe”Compr”blockwiththeexceptionofthemultistage
compressorpresentinthecaseswithCO -sequestration,whereinsteadthe”MCompr”
2
block is used. The calculation method is set to: ”Polytropic using the ASME
36 |
Chalmers University of Technology | 6 MODELLING
method”. To predict the power consumption with higher accuracy a polytropic effi-
ciency for centrifugal compressors is estimated using Equation 1 which is dependent
on volume flow, F, expressed in m3s−1 (R. Smith, 2005). A mechanical efficiency
for losses in e.g. bearings and seals is also specified, a value of 98% is used for all
compressors (Kurz et al., 2010).
η = 0.017ln(F)+0.7 (1)
6.5.2 Pump
The block ”Pump” is used to model pressure increases. By specifying the outlet
pressure as well as pump and driver efficiencies, Aspen Plus calculates the power
requirement. Forcentrifugalpumps,thepumpefficiencyisdependentonvolumeflow
rate, F expressed as m3h−1 in Equation 2 (R. Smith, 2005). The driver efficiency
compensates for losses in the electrical motor which converts electrical energy to
mechanical energy. A value of 90% is assumed for the driver efficiency and applies
to all pumps (Evans, 2010).
η = −0.01(lnF)2 +0.15ln(F)+0.3 (2)
6.5.3 Valve
Valvesaremodelledusingthe”Valve”block. Themodelassumestheflowisadiabatic
and uses two-phase calculations to determine the outlet condition for the specified
pressure.
6.5.4 Flash
For flashes, both vertical and horizontal, the ”Flash2” block is used. This model
performs rigorous two-phase calculation. The model assumes that sufficient vapour
disengagement space is present, meaning that the gas stream leaving the flash does
not contain any liquid and vice versa.
6.5.5 Mixer and Splitter
The ”Mixer”block combines several streams into one and performs adiabatic phase
equilibrium flash calculation to determine the outlet conditions. The ”FSplit”block
divides an ingoing stream into two or more outlet streams with same compositions
and conditions as the inlet. The sizes of the streams are determined from a specified
split fraction.
6.5.6 Heater and cooler
Both heaters andcoolers are modelledusing the ”Heater”block. By specifying outlet
temperature and pressure, ”Heater”calculates the heating or cooling demand. Two-
37 |
Chalmers University of Technology | 7 RESULTS AND DISCUSSION
Figure 18: Utility consumption of the MEA process.
As illustrated in Figure 19, the BLGCC-S-scenario also has a considerable differ-
ence in both cooling water and electricity consumption. This is due to the removal
of the top-cycle and the compression section in the case without capture. The steam
demand is essentially unchanged as the reboiler duty in bottom-cycle is constant for
both cases.
Figure 19: Utility consumption of the Selexol process.
For the BLGMF-R-scenario presented in Figure 20, the difference between the
two cases is marginal. The difference stems from the removal of the compression
section in the case without capture.
If the capture cases of the BLGCC-S- and BLGMF-R-scenarios are compared, it
can be observed that the utility demands are essentially equal. The Selexol process
has slightly lower electricity demand compared to the Rectisol process, but uses
MP-steam instead of LP-steam. However, the use of MP-steam would outweigh
the lower electricity demand. The reason is that usage of MP-steam implies a
loss in electricity, since this steam cannot be expanded to LP-steam and produce
electricity. Thus, ifonlyutilitydemandisconsideredthentheRectisolprocesswould
be the preferred alternative also for the BLGCC-S-CCS-case, but by a very small
margin. However, literature reports that the Selexol process is the economically
40 |
Chalmers University of Technology | 7 RESULTS AND DISCUSSION
WhenthetwocasespresentedinFigure21and22arecompareditcanbenoticed
that the capture case requires more solid wood fuel. The reason is that the capture
process requires a considerable amount of LP-steam and to satisfy this demand more
solid wood fuel has to be combusted to produce steam. The biofuel boiler produces
HP-steam which is expanded to the needed LP-steam. Consequently the electricity
production increases as well. This is a necessity as the capture case has larger
electricity consumption. However, the increase in production is greater, resulting in
a larger surplus that can be sold.
Bark 48 MW RB-M-NC
Fossil fuel 37 MW District heating 12 MW
Wood 584 MW Pulp (Kraft) 422 kADt/y
Solid wood fuel 78 MW Pulp (CTMP ) 95 kADt/y
Electricity (use) 52 MW
Electricity (sold) 2 MW
Figure 21: Overall balance displaying major in and out streams to the pulp process for the
RB-M-NC-case. For better overview, pulp and CO flows are presented on a yearly basis.
2
Bark 48 MW RB-M-CCS CO2 capture 751 kt/y
Fossil fuel 37 MW District heating 12 MW
Wood 584 MW Pulp (Kraft) 422 kADt/y
Solid wood fuel 216 MW Pulp (CTMP ) 95 kADt/y
Electricity (use) 62 MW
Electricity (sold) 20 MW
Figure 22: Overall balance displaying major in and out streams to the pulp process for the
RB-M-CCS-case. For better overview, pulp and CO flows are presented on a yearly basis.
2
42 |
Chalmers University of Technology | 7 RESULTS AND DISCUSSION
The overall balance of the two BLGCC-S-cases, illustrated in Figure 23 and
24, differs from the balance of today’s pulp mill i.e. the RB-M-NC-case. There
is an increase in fossil fuel usage, since the BLGCC-S-scenario results in higher
caustization load and thereby more fuel is needed in the lime kiln. Also the pulp
production rate increases because of the use of polysulfide cooking liquors.
RegardingdifferencesbetweenthetwoBLGCC-S-cases, onecanseethatthecap-
ture process is electricity demanding and hence, the electricity surplus is decreased.
The capture process changes the composition of the syngas entering the gas turbine
and consequently the produced electricity is slightly lower. Another effect is that
the flue gas flow is smaller and less heat can be extracted in the HRSG resulting in
a slight increased need of solid wood fuel to satisfy the steam demand.
Bark 48 MW BLGCC-S-NC
Fossil fuel 42 MW District heating 12 MW
Wood 584 MW Pulp (Kraft) 439 kADt/y
Solid wood fuel 197 MW Pulp (CTMP ) 95 kADt/y
Electricity (use) 72 MW
Electricity (sold) 93 MW
Figure 23: Overall balance displaying major in and out streams to the pulp process for the
BLGCC-S-NC-case. For better overview, pulp and CO flows are presented on a yearly basis.
2
Bark 48 MW BLGCC-S-CCS CO2 capture 229 kt/y
Fossil fuel 42 MW District heating 12 MW
Wood 584 MW Pulp (Kraft) 439 kADt/y
Solid wood fuel 198 MW Pulp (CTMP ) 95 kADt/y
Electricity (use) 80 MW
Electricity (sold) 83 MW
Figure 24: Overall balance displaying major in and out streams to the pulp process for the
BLGCC-S-CCS-case. For better overview, pulp and CO flows are presented on a yearly basis.
2
43 |
Chalmers University of Technology | 7 RESULTS AND DISCUSSION
The BLGMF-R-cases, presented in Figure 25 and 26, have for the same reason
as the BLGCC-S-cases an increased usage of fossil fuel as well as an increased pulp
production rate. Another major difference for the BLGMF-R-cases compared to
today’s pulp mill, RB-M-NC-case, is a much higher need for solid wood fuel. The
reason is that much of the energy contained in the black liquor is not used to satisfy
the need for either heat or electricity but instead leaves the system in the form of
DME. Hence, additional fuel and electricity has to be purchased from an external
source to satisfy the demand.
IfthetwoBLGMF-R-casesarecompared, itbecomesclearthattheyareidentical
with the exception that the capture case consumes marginally more electricity. That
is the electricity needed for compressing the CO .
2
Bark 48 MW BLGMF-R-NC
Fossil fuel 42 MW District heating 12 MW
Wood 584 MW Pulp (Kraft) 439 kADt/y
Solid wood fuel 330 MW Pulp (CTMP ) 95 kADt/y
Electricity (use) 82 MW DME 161 MW
Electricity (bought) 12 MW
Figure 25: Overall balance displaying major in and out streams to the pulp process for the
BLGMF-R-NC-case. For better overview, pulp and CO flows are presented on a yearly basis.
2
Bark 48 MW BLGMF-R-CCS CO2 capture 229 kt/y
Fossil fuel 42 MW District heating 12 MW
Wood 584 MW Pulp (Kraft) 439 kADt/y
Solid wood fuel 330 MW Pulp (CTMP ) 95 kADt/y
Electricity (use) 82 MW DME 161 MW
Electricity (bought) 13 MW
Figure 26: Overall balance displaying major in and out streams to the pulp process for the
BLGMF-R-CCS-case. For better overview, pulp and CO flows are presented on a yearly basis.
2
44 |
Chalmers University of Technology | 7 RESULTS AND DISCUSSION
7.4 Cost of carbon capture
By comparing the two overall balances for each scenario it is possible to quantify
the cost of carbon capture in the form of additional amount of purchased solid wood
fuel, cooling water and sold or bought electricity. The result of such a comparison
is summarized in Table 19.
Table 19: The additional resource consumption per kilo CO captured for the three scenarios.
2
Utility [kJ/kgCO ] RB-M BLGCC-S BLGMF-R
2
Solid wood fuel 5510 170 0
Cooling water 4460 1130 370
Electricity -700 1360 220
A remark regarding electricity is that both the RB-M- and BLGCC-S-scenario
have a surplus of electricity which is sold whereas the BLGMF-R-scenario has to
purchase electricity to cover a deficit. Thus, the values in Table 19 should be inter-
preted as follows: applying carbon capture to the RB-M-scenario increases the sold
electricity by 700kJ/kgCO , the BLGCC-S-scenario decreases the sold electricity by
2
1360kJ/kgCO and in the BLGMF-R-scenario 220kJ/kgCO additional electricity
2 2
has to be bought. It is important to distinguish between sold and bought electricity
as they are priced differently. All resources listed in Table 19 can be priced and
thereby, it is possible to calculate an operating cost for CO -capture. However, to
2
calculate the total cost of CO -capture the investment cost of the capture processes
2
has to be taken into account. Table 20 presents the investment cost for the six cases.
Table 20: Investment cost for the capture processes, provided by Skagestad & Eldrup (2014).
Investment cost [MSEK] RB-M BLGCC-S BLGMF-R
CCS 659 656 380
NC 0 226 299
Fromthedifferenceininvestmentcostbetweenthecasewithcaptureandthecase
without capture, an annual payback cost is calculated. Table 21 shows the annual
cost for carbon capture divided into different contributions. Costs for maintenance
andsalariesforoperatorsandengineerswereprovidedbySkagestad&Eldrup(2014).
TheRB-M-scenariohasthehighesttotalannualcost, threetimeshigherthanthe
BLGCC-S-scenario. Each of the seven cost categories surpasses the other scenarios.
The reason for this is that the MEA process handles a larger gas flow and hence will
have both higher operating and investment cost. However, this also implies that
the RB-M-scenario captures more CO . Consequently the specific cost of carbon
2
capture is lower than for the BLGCC-S-scenario. It is hardly surprising that the
BLGMF-R-scenario has the lowest capture cost since only the compression section is
47 |
Chalmers University of Technology | 7 RESULTS AND DISCUSSION
Table 21: Thecostofcarboncaptureforthethreescenarios,dividedintocontributionofdifferent
categories. All costs are specified in MSEK/year.
Scenario RB-M BLGCC-S BLGMF-R
Investment cost 60.0 39.1 7.4
Maintenance 23.2 15.2 2.9
Operators 13.2 0 0
Engineers 5.4 0 0
Solid wood fuel 229.8 2.2 0
Cooling water 63.9 4.9 1.6
Electricity -72.0 42.4 8.2
Total cost [MSEK/year] 323.5 103.8 20.0
Specific cost [SEK/tCO ] 431 453 88
2
addedforthecapturecase. InMars2014thepriceofaemissioncertificatewas5.0e.
If the capture cost for the of BLGMF-R-scenario, 88SEK/tCO corresponding to
2
9.7e/tCO , is compared with the price of an emission certificate it can be concluded
2
that carbon capture is not profitable. However, the BLGMF-R-scenario will not
likely be implemented before several years have passed and by then the price level
of emission certificates may have recovered. Hence, the pulp mill has the potential
to be a profitable carbon capture and storage alternative in the future.
7.5 Sensitivity analysis
The largest contribution to the total annual cost in Table 21 is the electricity, with
the exception of the solid wood fuel for the RB-M-scenario. However, the solid
wood fuel price has held a steady level during the last few years and therefore, the
electricity price was varied in a sensitivity analysis to evaluate its impact on the
capture cost in the three scenarios (Swedish Energy Agency, 2014). Another reason
is that the electricity price is harder to predict and even varies over the seasons of
the year. Table 22 presents the electricity prices used in the sensitivity analysis.
Both an increase and decrease of 30% have been considered.
Table 22: Electricity price used in the sensitivity analysis.
Price [SEK/MWh] -30% 0% +30%
Sold 343 490 637
Bought 413 590 767
Figure 30 illustrates the result of the sensitivity analysis. A higher electricity
price means a decreased cost for the RB-M-scenario, as the additional electricity
sold generates a higher income. For the BLGCC-S-scenario, the penalty due to less
48 |
Chalmers University of Technology | 8 CONCLUSIONS
8 Conclusions
This thesis evaluates the possibility of applying carbon capture and storage onto
a pulp mill. Three scenarios regarding the future development of the pulp mill is
examined.
The first scenario assumes that the pulping process remains unaltered and hence
uses the conventional recovery boiler to regenerate the cooking chemicals. For this
scenario, post combustion capture using the MEA process was studied. The simula-
tions resulted in a high utility consumption, especially regarding steam and cooling
water. However, the overall balance showed that when producing the extra required
stream, additional electricity was generated. Consequently more electricity could
be sold and this reduced the cost of carbon capture. The cost of CO -capture was
2
431SEK/tCO . The potential effect of applying BECCS was highest of the three
2
scenarios with a net reduction of global CO emissions by 715ktCO /year.
2 2
In the second scenario black liquor gasification technology is used and the pro-
duced syngas is utilized for electricity production. The Selexol process was determ-
ined to be the most suitable capture process. The calculations resulted in rather
low additional resource consumption, with the exception of electricity usage. Con-
sequently the electricity consumption constitutes the major part of the CO -capture
2
cost of 453SEK/tCO . This was the most expensive of the three scenarios and
2
also the potential effect of applying BECCS had the worst performance with a net
reduction of global CO emissions by only 318ktCO /year.
2 2
Black liquor gasification technology is also used for the third scenario, but the
syngas is instead utilized in DME production. The capture process used in this
scenario is the Rectisol process. The additional resource consumption associated
with CO -capture was shown to be the lowest of the three scenarios, resulting in
2
a cost of only 88SEK/tCO . The potential effect on global CO emission for this
2 2
scenario is a net reduction of 393ktCO /year.
2
A sensitivity analysis was conducted to test the robustness of the results. The
uncertainty of the electricity price greatly affects the results and thus this parameter
wasvaried. Thesensitivityanalysisshowedthatforadecreaseoftheelectricityprice
by 30%, the scenario using the MEA process had a slightly higher capture cost than
the scenario utilising the Selexol process. However, the scenario using the Rectisol
process was superior regardless the pricing of electricity.
To conclude this thesis, the result implies that the pulp and paper industry could
be a suitable future candidate to which BECCS can be applied. The cost penalty
associated with carbon capture proved to be too high in the present day situation;
with an over-abundance of low price emission certificates. However if the emission
certificate market recovers, there would be an incentive for capturing CO in the
2
scenario utilizing the Rectisol process.
51 |
Chalmers University of Technology | A OUTPUT STREAM REQUIREMENTS AND RESULTS
A Output stream requirements and results
CO -outlet stream
2
Table 23: The requirements and the results of the simulations for the CO -stream ready for
2
sequestration. Spices labelled with an ”m” have their composition specified as mole-% and for
”p”ppmw is used. A green colour indicates that the requirements are fulfilled and a red colour
indicates the opposite.
CO CO H CH N H O H S COS O Solvent
Case 2 2 4 2 2 2 2
[m] [m] [m] [m] [m] [p] [p] [p] [p] [p]
Requirements >95 - - - <4 - <1500 - <10 -
RB-M-CCS 99.2 0 0 0 0 2956 0 0 14 0
BLGCC-S-CCS 95.1 3.8 0.3 0.1 0.1 315 360 30 20 35
BLGMF-R-CSS 96.4 2.7 0.2 0.2 0 0 343 11 0 378
H S-outlet stream
2
Table 24: The requirements and the results of the simulations for the H S-stream entering the
2
Claus process. Spices labelled with an ”m” have their composition specified as mole-% and for
”p”ppmv is used. A green colour indicates that the requirements are fulfilled and a red colour
indicates the opposite.
CO CO H CH N H O H S COS O Solvent
Case 2 2 4 2 2 2 2
[m] [p] [p] [p] [m] [m] [m] [p] [m] [p]
Requirements - - - - - - >40 - - -
BLGCC-S-NC 24.1 3 1 1 22.8 2.4 50.5 146 0.2 3
BLGCC-S-CCS 24.3 0 0 0 23.1 2.4 50.0 156 0.2 0
BLGMF-R-NC 57.0 0 0 2 0 0 41.6 2598 0 1415
BLGMF-R-CCS 57.0 0 0 2 0 0 41.6 2598 0 1415
61 |
Chalmers University of Technology | B PINCH ANALYSIS
B Pinch analysis
Pinch analysis is a structured way of improving existing or creating new heat ex-
changer networks with an high degree of heat integration. Individual temperature
differences are chosen for each type of stream. The following individual temperature
differences are used in this work:
• 10◦C for gas streams.
• 5◦C for liquid streams.
• 5◦C for evaporating and condensing streams.
For these condition a pinch point is calculated. The pinch point is where the
lowest allowed temperature difference is present. Over the pinch temperature there
is a deficit of heat and hot utility is therefore needed. Below the pinch, cold utility
is needed as there is an execs of heat. Hence, to minimize utility the three ”golden
rules”of pinch analysis should be abided:
• Do not use external cooling above the pinch.
• Do not transfer heat through the pinch.
• Do not use external heating below the pinch.
If these rules are followed an maximum energy recovery heat exchanger network
is obtained. However, this may not be the economically most advantages alternative
since such an network often requires many units. To obtain an better design the
network can be ”relaxed” i.e. the smallest heat exchangers are removed and are
replaced with heaters and coolers. This increases utility consumption but gives
lower capital cost. Hence, an optimum can be found. In this work no extensive cost
analysis has been done to optimize the number of units. Instead, as a rule of thumb,
heat exchangers smaller than 100kW has been removed. The design of the heat
exchanger networks for the capture processes are illustrated in the following pages.
63 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.