Content uploaded by Ana Goidea
Author content
All content in this area was uploaded by Ana Goidea on Apr 09, 2020
Content may be subject to copyright.
The world is currently facing an ecological crisis of
unprecedented scale and urgency and, as the building
sector is a significant contributor to the current state, it
must look towards radical change to achieve a sustainable
practice. The most destructive environmental impact is
found in material extraction, processing and discharge.
This paper presents an alternative to industrially mined
and synthesised materials by utilising biological growth
processes as passive engines for the transformation of
renewable materials. This is achieved through fungal-
lignocellulosic composites which have been developed
along with the design and fabrication processes that are
necessary for their application in the construction
industry.
Plant-derived materials are abundantly and sustainably
available on both local and global scales, particularly in
the form of by-products and recycled waste. Additive
fabrication provides an opportunity to create high value
products from this material, but comes with its own
challenges. In particular, most of the strength of the wood
is lost as fibres are ground down so that the material can
pass through the extrusion nozzle. Rather than relying on
thermosetting plastics or synthetic binders, this project
explores the controlled growth of fungal mycelium within
PULP FACTION
3D PRINTED MATERIAL ASSEMBLIES THROUGH
MICROBIAL BIOTRANSFORMATION
ANA GOIDEA / DIMITRIOS FLOUDAS / DAVID ANDRÉEN
LUND UNIVERSITY
1
the printed material post-extrusion as a binder of
lignocellulosic biomass.
Fungal-lignocellulosic materials inherit properties from
both wood and mycelium, resulting in lightweight and
strong bio-composites. Generally, they exhibit good
insulative performance for both heat and sound, are
hydrophobic, and have good tension and compression
resistance (Yang et al., 2017; Elsacker, 2019). In addition,
the raw materials for such composites are low in cost,
locally sourced, renewable, and able to capture and store
carbon dioxide.
Mycelium Bio-Composites
The main components of mycelium composites are the
biopolymers cellulose and chitin, followed by lignin and
hemicellulose. Mycelium is the vegetative part of a
fungus, made up from a dense network of long, branching
filamentous structures termed hyphae. The cell wall of the
hyphae is made of chitin – a tough, resilient, inert and
non-water-soluble modified polysaccharide that has
promising potential in biotechnology (Latgé and
Calderon, 2006). When the fungus colonises a substrate,
it first grows on the surface and gradually, depending on
4342
the properties of the material, it spreads its mycelium
throughout it in a complex three-dimensional binding
matrix (Boyce and Andrianopoulos, 2006) (Fig. 4).
During growth, the fungus secretes extracellular
polymeric substances (EPS), which are mainly composed
of polysaccharides and proteins (Gazzé et al., 2013).
Their role is to facilitate growth and allow the anchoring
of the cells on the substrate, acting as a glue between the
hyphae and the substrate. Moreover, EPS allow for the
conglomeration of particles around the hyphae, resulting
in an irreversible fusing of the material (Fig. 2).
3D Bio-Printing
Most precedents using lignocellulosic substrate and
mycelial growth for creating bio-composites use casting
as the means of production, for example: The Living’s
Hy-Fi Tower (Nagy et al., 2015); Block Research Group’s
MycoTree (Heisel et al., 2018); Mogu panels (Appels et al.,
2019). Such methods are relatively straightforward and
therefore well-suited for industrial mass-production.
However, the casting process limits the customisation
of the products as well as geometrical complexity that
can be employed for functional performance. In addition,
the strength of the material is markedly determined by
the extent of the mycelium coverage (Yang et al., 2017).
As this is dependent on oxygen, growth is limited to the
material surface. When cast in solid volume, the mycelium
covers a smaller percentage of the total volume, limiting
the potential strength of the composite.
These limitations can be overcome through the use of
digital additive fabrication which allows for a complex
meso-scale structure, radically increasing the surface
area within a given volume and thus ensuring maximum
distribution of hyphae within the composite.
The strategy of additively fabricating mycelium
composites is not unprecedented in nature. Mound-
building macrotermites have evolved to a symbiotic
existence together with fungus of the genus Termitomyces.
The termites harvest plant-based material and carry it
back to the mound where the regulated internal climate is
suitable for fungi. The fungus processes the plant matter,
turning it into nutrients that both the termites and fungus
live on (Turner, 2005). The fungal combs (Fig. 3) have a
particular geometry which, on the one hand provides
access to the termites for managing the comb, and on the
other enable a convective flow of air and respiratory gases
near the comb surfaces. This flow is facilitated by vertical
channels and assisted by the thermal buoyancy generated
by the metabolic heat of the fungus. The combs are
constructed as an intricately folded and interconnected
sheet with an even thickness of approximately 4mm, likely
corresponding to the depth at which the mycelium can
eectively grow while maintaining access to oxygen. The
fungus comb provided an initial set of assumptions for a
design that could provide a suitable balance of parameters
in the project.
Integrated Research Protocols
The research presented in this article concerns the finding
of a set of processes for additive fabrication of fungal-
lignocellulosic materials and the evaluation of their
suitability for architectural fabrication. The primary
intent was to address the questions that arise from the
interdependencies between these processes through a
transdisciplinary approach. Focus has been on testing
feasibility, building a protocol, and establishing a
foundation for informed speculation.
The research was guided by the following questions:
How can a process of bio-fabrication best be structured to
achieve desirable artefacts? How does the introduction of
fungal mycelium aect the material properties? And how
could the developed processes be utilised for fabrication
at architectural scales?
To answer these questions, the presented work explored
the interconnections between (1) the living system,
(2) the digital fabrication, and (3) the computational
design strategy. Subsequently, a number of material
performance tests were carried out on the resulting
samples. The protocol presented here led to the most
successful outcomes with regards to rate of growth,
extrudability, stability and resulting material properties.
Live Pulp
The pulp consists of a substrate that has been inoculated
with fungus. The substrate was developed to comply with
two primary criteria: its ability to support the growth and
development of the fungus, and its suitability for
fabrication which includes both extrudability and the
stability of the material in the print and growth phases.
The main components of the substrate are fine woodchips,
paper pulp, and kaolin clay, which are mixed with water.
Wood and paper pulp compose the bulk of the material
and provide the nutrients for the fungus; during
incubation, these are partially transformed into fungal
biomass. As the substrate doesn’t have an immediate
bonding agent, it remains unstable during printing.
Therefore, clay was added to the mixture to provide
stability during the fabrication and incubation phases.
The substrate also contains a thickening agent which
allows the solid and liquid components to form a coherent
aggregate (Fig. 10).
Two fungal strains were used in the experiments, a strain
of Byssomerulius corium and a strain of Gloeophyllum sp.
They are both wood decomposers, but follow dierent
strategies of wood decomposition termed white rot and
brown rot, respectively. Both fungal strains were
propagated on a malt-yeast medium. When the mycelial
growth was sucient, the fungus was introduced into the
autoclaved substrate. The inoculated substrate was left for
an incubation period of one week, in which the mycelium
propagated through the substrate and adapted to the new
environment, enabling it to resist contaminants introduced
when sterile conditions were no longer maintained.
2
1. Sectio n of a column
showing an assembly of
the fungal-lignoc ellulosic
components. Bonding
between th e segments is
proposed to b e achieved
by extrusion of a con nective
tissue consis ting of a
modified vers ion of the
live pulp.
2. Substrate un der
microscope. The different
magnifications showing:
(1) The print layer s
covered in mycelium .
(2) The fusio n of mycelium
and substrate. (3 ) The partial
decomposition of the
cellulose and li gnin fibres
by the fungus .
3. Termitomyces fungus
comb. This symbiotic
structure is ad ditively
assembled by m acrotermites
from dead plant m atter
inoculated wi th fungal
spores. The fungus slowly
digests the pl ant matter
into componen ts which the
termites can inge st as food,
while it simultan eously acts
to regulate the humi dity of
the mound’s intern al climate.
4. Living printe d composite
after two we eks of
incubation.
3 4
4544
Following the initial incubation, the pulp was 3D printed,
after which it went through a second and longer
incubation period. This allowed the mycelium to grow
through the printed artefacts and transform the substrate
into the desired bio-composite. Once the growth had
reached the target state, the printed component was
desiccated to reach its final and stable form, stopping
the decomposition process.
Fabrication Strategy
The live pulp was 3D printed using Vormvrij Lutum v4,
which relies on a combination of pressurised air and a
rotating auger to extrude material. A nozzle diameter of
3.5mm was used in combination with a layer height of
1.5mm, which provided a working balance of resolution,
stability, and print speed.
Several factors influence the stability of the print, and the
ability to produce artefacts with the desired geometric
variation. A larger nozzle and consequently greater wall
thickness make for more stable prints, but have the
drawback of lower resolution and decrease in surface to
volume ratio, which reduces the amount of mycelial growth
on the material. Straight vertical walls are prone to both
deformation and collapse. To reduce this, the curvatures
have been maximised and additional interconnections
between walls were introduced.
During desiccation, the material contracts in volume by
approximately 30%. In order to minimise the resulting
distortion, a set of aluminium meshes with vertical
channels were used as print base and cover. These
secure the position of the first and last layers, thereby
constraining the contraction to the Z-axis. Mesh-print
adhesion was improved by the explorative growth of the
mycelium. The meshes allow vertical airflows through the
print, supporting biological growth by ensuring even
moisture levels and the circulation of respiratory gases,
and eventually facilitate rapid and even desiccation.
Design Strategy
In addition to the architectural scale constraints, the
design of the components had to accommodate both the
biological requirements of the fungus and the mechanical
constraints of the printing process. A reaction-diusion
simulation based on the the Gray-Scott model generated
the basis for the form-finding of the fabricated geometries.
The scale of the pattern was derived from the fungus
comb reference. This generative model has been developed
by increasing the feed rate along the vertical axis. The
boundaries of the geometry have been created at the
transition points between the two simulated substances.
Subsequently, the resulting geometry consists of two
(internal and external) interwoven volumes that never
converge, lending itself to functional use in the
architectural outcome. Similar to the structure found in
fungus combs, the model ensures significant vertical
continuity that is beneficial to flows of both air and
structural forces. (Fig. 5)
The curves that constitute the print layers are taken
through a secondary algorithmic transformation which
connects all curves into a single curve on a per-layer basis.
This transformation allows for a continuous extrusion rate
along an uninterrupted toolpath which improves speed,
stability and precision in the print process. This also
ensures that the entire printed component is cohesive and
that additional stabilising cross-bracings are created
without disturbing the continuity and separation of the two
sets of volumes. In order to maintain thin extrusions while
increasing the print height, the design strategy combined
vertical continuity with recurring interconnections, while
strengthening through double curvature.
Results
Three dierent material samples (designated S0, S1, and
S2) were tested for the resulting material properties and
their suitability for architectural application. S0 was printed
substrate with no fungal inoculation, S1 was pulp with the
fungal species Byssomerulius corium, and S2 was pulp
with a Gloeophyllum sp. (Fig. 6). Three tests were carried
out: a bending test to evaluate stiness, a test for dispersion
in water, and a test for water absorption properties. Since
the results from these tests indicated that the Gloeophyllum
pulp composite has the most desirable properties,
additional samples were produced and further scanned to
characterise the distortion of the material during drying
(Fig. 7). This is notable since most other mycelium-based
materials use white rot fungus, while here is was found that
the brown rot fungus, Gloeophyllum, gave the better result.
Mechanical Performance
The bending test showed that the samples with more
extensive hyphae distribution exhibited significantly
higher stiness than the mycelium-free sample and
aorded a slightly higher force before failure (Fig. 8).
The deformation before failure was twice as high for
S0 as S2, with S1 falling in between. The hardness of
the material as perceived when cutting the samples with
a sharp knife was significantly higher with increased
hyphae coverage (S2 > S1 > S0).
Dispersion in Water
The resilience of the material bond when wet was tested
by submerging the samples in water for a period of 10
hours. After this period, the water and samples were
stirred (Fig. 9). The sample without mycelium (S0) quickly
swelled and completely disintegrated upon agitation.
S1 and S2 remained intact during stirring.
Hydrophobicity
A droplet of water was placed on each of the surfaces of
the three samples, and the subsequent absorption was
observed. The droplets on S0 and S1 were quickly
absorbed, while the droplet placed on S2 did not absorb
but maintained its shape, indicating strong hydrophobicity
on the material surface (Fig. 11). The samples’ capacity for
buering water in vapour form was also measured, and
remained equally high in all three samples.
5. Column assemblage:
design to fabrication.
6. Dessic ated printed
sample comparison.
From left: S1 , S0, S2.
7. Detail of the 3D scanne d
prototype. Analysis of
deviation from the
toolpath sent for fabrication,
after printin g, growth and
desiccation. Although there
is considerable distortion,
it is locally dis tributed
throughout t he height of
each module an d therefore
not global, the tolerances
not penalisin g the current
design application.
8. Mechanical performance:
bending test .
7 8
5
6
4746
Bio-Integrated Design:
Architecture as Multi-Scalar Interfaces
The printed prototypes and tests conducted on the
resulting bio-composite demonstrate some of the
advantages of the proposed approach. The resulting
components were highly hydrophobic with a retained
capacity of moisture buering, and remained stable even
when exposed to prolonged submersion in water. The
transformation of the material by the fungus resulted in
improved stiness and hardness, and eliminated the
tendency of the samples to delaminate between printed
layers. The surface hardness of the resulting material was
markedly dierent from many other reported mycelium-
based materials. This may be due to the use of a brown rot
fungus instead of white rot, and this strain’s interaction
with the substrate. However, further studies are required
to investigate these relationships.
The ability to fabricate larger scale elements relies on
navigating the requirements in the design space, and
the stability and predictability of the printed components
in the growth and desiccation phases. The robustness
of the process was improved by the inclusion of clay to
the substrate as well as the use of stabilising meshes.
Equally, the component design is of critical significance,
both in terms of enabling the growth of the mycelium
and stabilising the material during and after fabrication.
This resulted in a requirement for a high surface area,
high curvature form.
When engaging with the agency of microorganisms
as well as with highly responsive and interdependent
materials, significant constraints are placed on the
design. These constraints require integration between
the multiple scales of the project, from the microscopic
scale of microbial behaviour, through the material
arrangement at the centimetre scale, all the way to
the component and assembly, and eventually human
scales. Rather than considering these constraints as
limitations, they present an opportunity for responsive
and functional architectures.
The demonstrated components (Fig. 1) assemble into a
column that retains several of the properties that allow
the fungus to thrive: it has a high surface area ratio, the
vertical interstitial spaces allow for convective flow, and
the material exhibits an active interaction with air and
water vapour. These properties remain after the element
is constructed, and can be utilised to aect and modulate
the environment in direct proximity to the column. Rather
than a passive load-carrying element, such a structure
should be considered a part of a building’s vascular
system, mediating and enabling flows that drive an
active modulation of the micro-climates which the
occupants inhabit.
Conclusion
The project demonstrates both the challenges and the
potential of additive fabrication of mycelium composites.
The introduction of fungus improves the properties of the
resulting material in multiple ways, resolving diculties
associated with wood printing through improved water
resistance and increased stiness and hardness.
Compared to previous fungus composites which are
typically fabricated through casting, additive fabrication
can improve the conditions for fungus growth, enabling
faster growth rates and more complete coverage. This can
result in better material performance and more ecient
manufacturing. The process enables complex and
customised form beyond what can be achieved through
casting, opening up new functional potential in the
resulting products.
The biologically active process adds constraints, such as
the need for sterile material processing and the prolonged
wet state. However, it was demonstrated how a
combination of material composition, design integration
and fabrication processes can be used to overcome these
challenges, potentially enabling the use of such materials
in the construction industry. If implemented at large
scales, such a shift could radically reduce the building
industry’s ecological footprint by lessening the need for
extraction of non-renewable minerals and for energy
intense chemical processing, while ensuring
environmentally safe and biodegradable properties.
Acknowledgements
This research has b een funded by The Crafoor d Foundation and Boverket,
the Swedish Nati onal Board of Housing, B uilding and Planning. We would also
like to extend our grati tude to Eva Frühwald Hansson and Maria Fred riksson at
Lund Universit y for their assistance with th e material testing.
References
Appels, F.V.W., Camere, S ., Montalti, M., Kara na, E., Jansen, K. M.B.,
Dijksterhuis, J ., Krijgsheld, P. and Wösten, H. A. 2019. ‘Fab rication factors
influencing m echanical, moisture- a nd water-related propertie s of
mycelium-based composites’, M aterials & Design, 161 , pp. 64-71.
Boyce, K. J. and Andrianopoulo s, A. 2006. ‘ Morphogenesis: C ontrol of Cell
Types and Shape’, in Kües U., Fisch er R. (eds), Growth, Dif ferentiation and
Sexualit y vol 1: The Mycota (A Com prehensive Treatise on Fungi as
Experimental Sys tems for Basic and Applied R esearch), Berlin, Heidelberg:
Springer, pp. 3 -20.
Elsacker, E., Vandeloo k, S., Brancart , J., Peeters, E. and D e Laet, L. 2019.
‘Mechanical, physical and chemical characterisation of mycelium-based
composites wi th different types of lig nocellulosic substrates’, PLo S ONE 14(7),
p. e0213954. (doi: 10.1371/jo urnal.pone.0213954)
Gazzè, S. A., Saccon e, L., Smits, M. M., Duran, A. L., Leake, J.R. , Banwart, S.A .,
Ragnarsd ottir, K.V. and McMas ter, T.J. 2013 . ‘Nanoscale Obser vations of
Extracell ular Polymeric Substanc es Deposition on Phyllosilic ates by an
Ectomycorrhizal Fung us’, Geomicrobiology Journal, 30(8), pp. 721-730.
Heisel, F., Lee, J. , Schlesier, K., Rippmann , M., Saeidi, N., Java dian, A., Nugroho ,
A., Mel e, T., Block , P. an d Hebel, D. 2018. ‘ Design, Cultivation and A pplication
of Load- Bearing Mycelium Compon ents: The MycoTree at the 2017 Se oul
Biennale of Archi tecture and Urbanism’, International Journal of Sustainable
Energy Development, 6(1), pp. 29 6-303.
Latgé, J. P. and Calde rone, R. 2006. ‘ The Fungal Cell Wall’, in Kües U.,
Fischer R . (eds), Growth, Differentiatio n and Sexuality vol 1: The M ycota
(A Comprehen sive Tre atise on Fungi as Experimental System s for Basic
and Applied Research), Berlin, Heid elberg: Springer, pp. 73 -104.
Nagy, D., Loc ke, J. and Benjamin, D. 2015 . ‘Computational B rick Stacking for
Construc ting Free-Form Struct ures’, in Thomsen M ., Tamke M., G engnagel C.,
Faircloth B. and S cheurer F. (eds), Modelling Beh aviour, Cham: Sprin ger, p.
203-212.
Turner, J. S. 2005. ‘ Extended physiolo gy of an insect-built s tructure’, American
Entomologist, 51(1), pp. 36 –38.
Yang, Z., Zhang, F., Still, B ., White, M. and Amstisl avski, P. 2017. ‘Physical and
Mechanical P roperties of Fungal Mycelium -Based Biofoam’, Journ al of
Materials in Civil Engineering, 29(7). (doi: 10.1061/(ASCE)MT.1943-
5533.0001866)
9
10
9. Dispersio n in water.
The fungus-free sample
disintegrates co mpletely,
while the two fun gal
composites exhib it minimal
swelling and remain in tact
after stirring .
10. Substr ate development
prototype, h ere without
fungus. T he substrate was
tested for extru dability,
as well as for the desi gn –
material compatibilit y.
11. Hydrophobicity.
From left: S 0, S1, S2.
11
4948
