Content uploaded by Asya Ilgun
Author content
All content in this area was uploaded by Asya Ilgun on Sep 29, 2022
Content may be subject to copyright.
1 INTRODUCTION
Biofabrication had revolutionized medicine and engineering long before entering the domain of
architecture. Biofabricated living materials consist of, and manifest aspects of, a living system
during their life cycle, in use and after.
Biological systems are composed of animate components whose self-organization into com-
plex adaptive systems spontaneously produce adaptability, self-healing, and metabolization.
This is a result of their ongoing interaction with each other and their environment. Consequent-
ly, the lifetime of living materials is linked to the natural cycle within the organism's ecosystem,
and the flow of biological and chemical resources. Using living systems in design is challenging
since they must be removed from niches and habitats. Yet, as humans, we're used to interfering
with lifeforms to create artificial habitats where they can adapt and evolve. Precision agricul-
ture, medicine, IOT (Internet of Things) gardening, genetic modification of plants, smart bee-
hives, and robot-plant interactions are just a few instances of how we incorporate human tech-
nology into biological systems for primarily human purposes (Gebbers & Adamchuk 2010,
Rhodes 2017, Schmickl et al. 2021).
The biosynthesis of chitin, found in hyphae of mycelial fungi, recently sparked scientific and
commercial interest. Fungi spread through dead plant matter by releasing enzymes that decom-
pose complex plant compounds into simpler ones, which fungi reassemble to form mycelia: an
entanglement of branching cells, known as hyphae. High-end research labs as well as home
kitchens have both experimented with solid state fermentation, in order to make mycelial com-
posites by moulding organic waste and fungi into shapes (Meyer et al. 2020). Mycelium’s natu-
ral habitat conditions are also receptive to other organisms which might harm humans. There-
fore, mycelium material applications and production methodologies, include desiccation of the
mycelium biomass before use: desiccating fungal biomass stops the growth of all organisms
and potentially kills them. Fungal biofabrication in architecture is both a challenge, as well as an
opportunity, to combine the knowledge and skills of all biohybrid systems research and devel-
A Study Model for Reconstructing Urban Ecological
Niches
A. Ilgün
University of Graz, Graz, Austria
R. Mills
École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland
F. Mondada
École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland
T. Schmickl
University of Graz, Graz, Austria
ABSTRACT: The Living Arch is a study model that was designed to showcase our ongoing
research into bioactive architectural systems to be inhabited by honeybees. The arch is a syner-
gistic combination of 3D printing and active mycelia growth, created for a museum exhibition.
It presents (and tests) a subset of our research and development work toward creating living
mycelium beehives. The Living Arch exhibits numerous physical, chemical, and biological
states in a transparent custom-made box fitted with environmental sensors. It facilitates a visual
and olfactory experience of an undergoing slow fungal transformation, which museum visitors
could witness and participate in throughout the 6-month exhibition. In this paper, we present
the Living Arch as an experimental exhibition piece and a study model of a fungal architecture,
as well as the problems that led to the exhibition's early termination.
opment. The main goal of developing the Living Arch is to situate a generally in-lab activity
within a long-term exhibition, in a well-visited art museum where the designer/maker/engineer
of the installation can continue working on it, together with the general visitors of the museum.
We take inspiration from Spyropoulos (2017): “Attention here is placed on behavioural features
that afford conversational-rich exchanges between participants and systems, participants with
other participants and/or systems with other systems.” We installed a system to facilitate the in-
teraction between human participants and a form of architectural routine – an arch geometry –
augmented with biological and technological elements.
1.1 Living Fungal Materials and Artificial Disturbing Regimes
Only recently has the function of fungi in living materials research shifted from simply shaping
and harvesting of their mycelia to the incorporation of their metabolic processes into responsive
design cases (Adamatzky et al. 2021). This shift implies that established processes, technolo-
gies, and concepts are advancing to keep mycelia alive during their' functional material use.
Morphology engineering studies the roles of macromorphologies of filamentous fungi from
genetic to hyphal levels, benefiting the food, medicine and agriculture industries (Meyer et al.
2021). Further, these insights might be helpful for developing methods to create functional my-
celium forms. What seems more of a paradigm shift is that they provide new design criteria for
architects, as well as new modelling and testing platforms for biohybrid architectures (Ozkan et
al. 2021).
Hyphae grow vegetatively by bonding solid substances, or aerially extending in the air to
reach nutrients. When the necessary conditions are met, the aerial growth gives rise to self-
healing properties such as the ability to bridge gaps between separate mycelial masses, resulting
in the formation of fungal monoliths (Elsacker et al. 2021). Local differentiation in mycelial tis-
sue is possible because of environmental gradients. For example, hyphal density increases with
the O2 levels in the medium. However, using sealed formworks to grow large volumes results in
a lack of control over local conditions. Thus, it is hard to ensure the homogeneity of biomass or
its functional gradience. The micro and macro morphologies of mycelial bodies are connected to
many other growth parameters such as pH, nutrition, medium type and size, temperature, and af-
fect other beneficial biological activities. For example, fungi create metabolites and enzymes
that break down or transform hazardous contaminants (Krull et al. 2013). Other living species
can also benefit from therapeutic volatiles secreted during mycelial growth (Rathore et al.
2019).
In the case of the Living Arch, clay units provide a layer of enclosure and internal stay-in
scaffolding, representing a departure from formworks which are impermeable and temporary.
We add sensors and actuation mechanisms to enable visitors to interact with the flow of the re-
sources and maintenance as well as to allow exploration of conditions ideal for the continuing
mycelial formation. In the meantime, the aerial mycelia can bridge the gaps caused by the or-
ganic deformation of the bricks and naturally glue them together. The goal of this experiment is
to encourage regular human interaction with a structure under slow fungal construction. We ex-
pect to see whether this human interaction can be registered in both living and nonliving parts of
the fungal structure, while the changes – occurring naturally or as a response to stimulation of
the structure’s environment – can help us understand the agency of mycelium in designed forms
in the human built environment. Evidence of this assembly can also establish new production
and use-case criteria for mycelium materials and architectures, which can, in turn, contribute to
open-source bio-fabrication and research communities (biofab forum).
1.2 A Study Case for Cohabitation: Social Insects and Fungi
In an increasingly urban world, we are replacing natural habitats of key species, such as hon-
eybees, with artificial and bio-incompatible ecosystems. We hypothesize that functional diversi-
ty is missing and/or inhibiting a healthy lifestyle of honeybees. Materials that support honeybee
health and natural microflora, such as living mycelial materials, could be seen as a breakthrough
in hive design. We approach this ecological design issue from both a practical and theoretical
standpoint, assuming social insects as co-occupants of the built environment. In the EU-H2020
project HIVEOPOLIS, we want to create a beehive that re-assembles a possibly symbiotic rela-
tionship between honeybees and specific mycelial fungi species (Ilgün et al 2021). This case
study's major purpose is to combine the structural properties of living mycelium with the posi-
tive bioactive capabilities. The self-supporting habitat becomes a living agent for supporting the
honeybee health by inhibiting the growth of harmful bacteria, producing immune-boosting me-
tabolites that the bees consume, biomonitoring pesticides and other toxic substances brought to
the hive from polluted areas.
2 A STUDY MODEL FOR HUMAN PARTICIPATION
The Living Arch is designed to be a self-supporting structure assembled with corbelling tech-
nique. In a corbel arch, each stone protrudes more than the one below it. We chose this tech-
nique to use mycelial hyphae as a bioadhesive on horizontal joints, which reduces the need for
temporary scaffolding during the assembly phase, thus making the process simpler, faster, and
more suitable for out-of-lab conditions. Nomadic animals provide good examples for rapid con-
struction methods, using locally available natural materials while minimizing labour. For exam-
ple, a pair of swallows builds their nest using their beaks and their saliva to mix locally sourced
mud over one or two weeks (Jung et al. 2021). Earth-based materials such as clay have been
used around the world as a building material for thousands of years for its ecological and aes-
thetic benefits. Mesopotamian nomadic people adopted building forms that can be built and
dismantled quickly with local materials, which are nevertheless resistant against heat and cold.
The thick mud bricks of Beehive houses – human homes named after their form – in Harran,
Turkey retain the cool or hot air, and can be laid by one person in one day, using the bricks from
their neighbouring ruins (Ozdeniz et al. 1998). And now, in the digital age, the micro and mac-
ro-level behaviour of clay artifacts can be customized simultaneously, thanks to recent advances
in and widespread dissemination of 3D modelling and printing knowledge in industrial, artistic,
and educational contexts (Habib 2019, Conti, Dubor et al 2019). When it comes to digital clay
design and manufacturing, our work employs parametric drawing techniques to minimise the
weight and production time of clay parts with the main goal of making mycelium-clay biocom-
posites.
Figure 1. Assembly of the Living Arch, April 2020, Kunsthaus Graz.
2.1 Design and Making of the Living Arch
The overall form of the structure was created using a parametric curve (in this case a quartic
Bézier curve) arranged around a polar axis. These curves' division points define the 3D printing
layer heights. Three more curves specify the parameters for thickness, width, and interior struc-
ture. First, we subtract selected points from the dome-like geometry to get an arch-like geome-
try. The remaining contour lines define the structure's outside perimeter. To disperse a specific
number of tween curves within this distance, consider the spaces for mycelium inoculate. Then
we drew a single polyline for each side of the arch by weaving the control points of hexagonal
cells together (fig. 2a). The G-code, translating this single polyline into the clay deposition path
of the 3D clay printer, is produced directly from design data within the same digital model. This
allows access to other 3D printer control parameters such as speed and flow rate, instead of rely-
ing on proprietary software translators that tend to contain assumptions about build strategy.
Recent research indicates successful mycelial development through up to 9.5mm thick clay,
despite the clay's lack of nutritional value for fungi (Jauk et al. 2021). We evaluated 3D printing
factors such as material viscosity, speed, and extrusion rate to create single extruded wall struc-
tures using our hexagonal weaving method, as thin as possible suitable for hyphae penetration
and reach. In preliminary trials, we showed microscopic scanning and visualisation of Trametes
Versicolor L. (TV) hyphae development on two polymeric 3D printing materials. As expected,
hyphae spread across the porous specimen infected with cellulose particles, but only superficial-
ly across the non-porous printed specimen (Ilgün et al., in press). Therefore, we used a com-
mercially available fibre composite clay: stoneware silica sand mixed with 10% wastepaper fi-
bre (Keramik Kraft, Austria). Our goal of printing thin walls demanded increased material
connectivity across facing walls, which influences the internal topologies and hence on the de-
sign model itself (Fig. 2a, Fig. 3a).
Figure 3. a) 3D model of the living arch with the left side showing the longitudinal section, b) Arrows
showing the centre of weights of each unit and two assembly phases, c) A collage to depict the physical
construction of the arch and its digital toolpath representation.
After 3D printing, the units were air dried for two days loosely covered with plastic
sheets. When they were stabilised enough for transportation, we sterilised them by autoclaving
in 121°C for 20 minutes, in heat resistant microfilter bags. After they had cooled down, we
filled the vertical channels with pre-grown mycelium spawn, ground particles of entire flax
plants which had been inoculated with the mycelium of TV (Fig. 3b). Prior to this, we applied a
common mushroom spawning method, an incremental solid state fungal fermentation process
that allowed us to grow large amounts of mycelial biomass from only 10ml liquid mother cul-
ture. A total of 15 pieces were incubated separately, in room temperatures at approx. 23°C, until
the mycelium infill developed varying hyphae densities. 6 units (3 on each side of the arch)
were grown for 12 days, while the rest were grown for 6 days.
Figure 2. Photographs of a unit of the arch, a) during 3D printing, b) after filling it with mycelium spawn,
c) on the 8th day of incubation, d) embedded in soil after 6 months.
On the day of assembly, we transported the units to the Kunsthaus Graz and assembled plas-
tic sheets around our work area in front of the growth chamber to decrease contamination risk
(Fig 1). We first filled the bottom part with 10L of unsterile peat soil, stacked five pieces on
each side (Fig. 2b). According to our plan, these five initial pieces would grow to form two
monolithic pieces, not allowing the units to slide against each other when the next units are add-
ed on top. This would allow us to separate these two complete units and add the remaining
components on top to complete the structure. During this first phase, our sensor tag was freely
hanging in the box registering the changes in the ambient, described in more detail below. We
had planned to place it in the upper most cuplike unit of the structure, if this construction pro-
cess had been successful. Additionally, we would place the mist maker in the middle of the
chamber for adding the humidity to the environment in the following months (Fig 4c).
2.2 Wireless Sensing and Disturbance Regime
In ethology and in applications such as precision agriculture, measuring the state of plants, ani-
mals and their environment is key for understanding the biological dynamics. Measuring the
state and response of a group of organisms is even more central in biohybrid systems that aim to
not only monitor but also steer their decision making (Halloy et al. 2007, Schmickl et al. 2021).
What should be measured and modulated depends on the organisms involved, such as environ-
mental properties (e.g. ambient temperature), or behavioral responses (e.g. growth or move-
ment). In the case of interacting with living mycelia, relative humidity (RH) is essential to
maintain hyphal development throughout the life cycle. Persistent mycelial colonization re-
quires CO2 levels more than 5000 PPM, while lower CO2 levels may, depending on the fungus
species, lead to fruit formation. These parameters are also important in our longer-term applica-
tion of growing beehives with this method, as honeybees actively regulate temperature (Jones &
Oldroyd 2006), humidity (Human et al. 2006) and CO2 (Seeley 1974).
Our installation intentionally enabled visitors to engage with fans and humidifiers as actua-
tors, so to change the environmental conditions for the biohybrid artifact. Thus, our system in-
cluded sensor nodes that continuously measured the environmental state and a central minicom-
puter that collected the data as well as presented/ visualized it to the visitors. These sensor nodes
are custom-designed boards using a nRF52840 microcontroller, equipped with Bluetooth (BLE)
communication, that samples the HDC2010 humidity sensor and Sensirion SCD-30 differential
infrared CO2 sensor. We used a Raspberry Pi 4B mini-computer as the BLE central device, re-
ceiving the sampled data, and injecting this into a time-series database (influxDB), which is vis-
ualized through Grafana, a tool that enables rapid preparation of intuitive and colorful dash-
boards. A specifically designed dashboard was displayed on a monitor in the exhibition. For
maintenance and monitoring throughout the exhibition we used a remote desktop tool
(Anydesk) and an offsite database mirror – both particularly useful during the uncertain condi-
tions of the Covid-19 pandemic.
Figure 4 describes the envisioned interaction scenario in which the sensor nodes were placed
within the arch's uppermost clay part – which is effectively a porous and empty cup – in order to
gain a better understanding of the environmental conditions within the 3d printed elements, ra-
ther than just the ambient air (Fig. 4.1). Although causal relationships between variables such as
RH, temperature, and CO2 are known, the effects on mycelium activity are difficult to predict.
This is complicated by the fact that the measurements are taken from a single spot in a 210L
box. Because RH is a function of temperature in a confined space with a breathing organism and
forced air exchange causes sudden disruptions in homeostatic environments, it is difficult to
draw any firm conclusions. In consequence, it is critical to incorporate human senses and per-
ception into our study and development of living mycelium structures at this point. In our envi-
sioned interaction scenario, if the measurements stray outside of the healthy domains for myce-
lia, a text alert would direct the human observer to a visual and/or olfactory evaluation of the
structure and give them the opportunity to bring the environment back into a sustaining state.
Figure 3. Overview of the installation and systems to monitor and modulate its environment.
3 CONCLUSIONS
In this study, we were unable to address the contribution that regular human interaction which is
mediated by environmental sensors brings to our biodesign strategy. The first reason is that on
day 14 of the exhibition, a highly invasive fungus – green mould – contamination occurred in
the base peat soil which was not pasteurised beforehand, and the spores dispersed through the
air, colonizing various areas of the arch structure over the following weeks. We believe it was a
Trichoderma species, which is a major source of contamination in the mushroom industry,
caused by a variety of factors including oxygen deprivation. We used 80% ethanol on the areas
which had visible green mould growth, but this also meant that we slowed down or even
stopped the mycelium growth. Secondly, due to the fixed opening dates and hours of the exhibi-
tion, we were not able to proceed with the alerting configuration according to our planned inter-
action regime. This could be avoided by improving the design, the fabrication process and
monitoring design as follows:
An improved structural design for a self-supporting overall form of the structure, which may
include a material-aware physics model, can ensure a smoother assembly process of the
pregrown units on site. This would require an extended design study, a flexible digital model al-
lowing the feedback from physical prototypes’ performance to digitally produced morphologies.
An example of such feedback would be the measured values like individual module weights
while wet, dry, and after mycelium development. Regarding the scaffold material, we can mix
the material in-house – potentially raw clay powder with small sawdust to reach a 3D printable
viscosity – rather than using a commercially available fibre clay for which we cannot obtain re-
liable information on the material composition and which is quite dense in terms of minerals. A
more porous, oxygen-permeable enclosure for the mycelia would allow for aeration of the inner
parts of the units, resulting in faster and healthier mycelium growth and a better chance of con-
taminating microorganisms.
We utilized a targeted medicinal mushroom mycelium species, TV, because our main re-
search goal is to effectively integrate potentially therapeutic compounds into the material char-
acter of the construction. But in the future, Oyster mushroom mycelium will be preferable due
to its rapid growth tendency, in the context of an exhibition display, which makes the system
more resistant to contamination. In terms of mycelium inoculation processes, to physically push
the mycelium substrate through the channels within the units that make up the arch's higher cur-
vature parts proved difficult, if not impossible. To circumvent this, we can fill the channels with
fine, dry lignocellulosic particles that can easily flow through the dry clay, then pour liquid my-
celium culture on top. Furthermore, the first pieces must either rest flat on the ground soil or the
mycelium must develop quickly enough throughout the soil to establish a sturdy foundation.
Perhaps the most critical component of our experiment plan, an uninterrupted flow of sens-
ing, visualisation, alerting, and manually led control behaviour was incomplete and failed to en-
able interaction between the living arch and the public audience. The unsuccessful start of the
fungal construction with contamination caused a chain reaction in the assembly procedures. The
unhealthy mycelial colonisation caused structural instability, while the relative humidity levels
decreased as the small window was left open by the museum visitors and since we could not
place our mist maker which was planned for the control of temperature and humidity without
opening the chamber, the structure kept losing humidity towards the space in the museum. We
have provided the sensing in our system, but we did not have the actuators under computer con-
trol. Such control could have improved the environmental regulation for the mycelia, when the
museum was empty, and to monitor visitor’s choices - but also to override them in case their
chosen action would be harmful to the mycelial development. Humans are able to identify pat-
terns and select appropriate actions in many domains, without needing a formal set of rules to
follow. In future participatory studies we aim to ensure our ability to learn from this rich source
of knowledge.
There are numerous advantages of coupling living organisms with artificial systems in biohy-
brid design, but as we have previously discussed, this poses the challenge of disembodying
them from their natural habitats. Our biohybrid architectural system employs a digital design to
a manufacturing approach involving biological development. The adaptive mycelial growth,
which we utilize in our paradigm, is a unique design method that aims to achieve novel architec-
tural objectives such as material responsiveness for healthy built environments for humans and
other organisms. In this method, it is critical to test the responses of fungal mycelia to designed
forms and to novel environmental factors in human-built settings. With this work we proposed a
combination of human sensory perceptions and technological instruments. Furthermore, by in-
corporating such biologically responsive structures into our daily lives, we can help to advance
the conversation about microbiological biodiversity and the health aspects of the built environ-
ment.
Acknowledgements:
We thank Rafael Barmak and Daniel Hofstadler for their contributions to the sensing hardware
and software for the exhibition. This work was supported by the EU H2020 FET project HIVE-
OPOLIS (no. 824069). AI and TS were supported by the Field of Excellence COLIBRI.
4 REFERENCES
Adamatzky, A., Gandia, A., 2021. Living mycelium composites discern weights via patterns of electrical
activity. J. Bioresources and Bioproducts 9, 53.
BioFabForum [WWW Document], n.d.. BioFabForum. URL https://biofabforum.org/ (accessed.2022).
Dubor, A., Marengo, M., Fernández, P. Experimentation, Prototyping and Digital Technologies towards
1:1 in Architectural Education. JIDA 2019, 606.
Elsacker, E., Søndergaard A., Wylick A.V., Peeters E. & Laet D.L. 2021. Growing Living and Multifunc-
tional Mycelium Composites for Large-Scale Formwork Applications Using Robotic Abrasive Wire-
Cutting. Constr Build Mater 283: 122732.
Gebbers, R., Adamchuk, V.I. 2010. Precision Agriculture and Food Security. Science 32: 828–831.
Habib, A., Khoda, B. Development of Clay Based Novel Hybrid Bio-Ink for 3D Bio-Printing Process.
Journal of Manufacturing Processes 2019, 38, 76–87.
Halloy, J., Sempo, G., Caprari, G., ... Deneubourg, J.L., 2007. Social Integration of Robots into Groups of
Cockroaches to Control Self-Organized Choices. Science 318: 1155–1158.
Human, H., Nicolson, S.W., Dietemann, V., 2006. Do honeybees, Apis mellifera scutellata, regulate hu-
midity in their nest? Naturwissenschaften 93: 397–401.
Hyde K.D., Xu J., Rapior S., …, Stadler, M. 2019. The amazing potential of fungi: 50 ways we can ex-
ploit fungi industrially. Fungal Divers 97: 1–136.
Ilgun, A., Szopek, M., Schmickl, T. 2021a. The Eusocial Cathedral and the Buzzaar: A Novel Synthesis
from De-and Reconstructing the Living and the Artificial. In Procs Naturally Hypernatural V. (In
Press)
Ilgün, A., Angelov, K., Stefanec, M., …, Schmickl, T. 2021b. Biohybrid Systems for Ecosystem Level
Effects In Procs ALIFE pp41-51.
Jauk, J., Vašatko H., Gosch L., Ingolf C., Klaus A, Stavric M., 2021. Digital Fabrication of Growth. In
Procs 26th CAADRIA Conf. 1: 753-762
Jones, J.C., Oldroyd, B.P., 2006. Nest Thermoregulation in Social Insects, Adv Insect Physiol. 33: 153.
Krull, R., Wucherpfennig, T., Esfandabadi, M.E., …, Wittmann, C., 2013. Characterization and control of
fungal morphology for improved production performance in biotechnology. J Biotechnol 163:112.
Jung, Y., Jung, S., Lee, S., Kim, W., Kim, H.-Y., 2021. Avian mud nest architecture by self-secreted sali-
va. Proc Natl Acad Sci USA 118: e2018509118.
Meyer, V., Basenko, E.Y., Benz, J.P., Braus, G.H., …, Wösten, H.A.B., 2020. Growing a circular econo-
my with fungal biotechnology: a white paper. Fungal Biology and Biotechnology 7: 5.
Meyer, V., Cairns, T., Barthel, L., …, Krull, R., 2021. Understanding and controlling filamentous growth
of fungal cell factories: novel tools and opportunities for targeted morphology engineering. Fungal
Biology and Biotechnology 8: 8.
Özdeniz, M.B., Bekleyen, A., Gönül, I.A., Gönül, H., Sarigül, H., Ilter, T., Dalkiliç, N., Yildirim, M.,
1998. Vernacular domed houses of Harran, Turkey. Habitat International 22: 477–485.
Özkan, D., Dade-Robertson, M., Morrow, R., Zhang, M., 2021. Designing a Biomaterial Through Digital
Fabrication. In Procs 39th eCAADe Conf.: 77-84
Rathore, H., Prasad, S., Kapri, M., Tiwari, A., Sharma, S., 2019. Medicinal importance of mushroom my-
celium: Mechanisms and applications. J Func Foods 56: 182–193.
Rhodes, C.J., 2017. The Imperative for Regenerative Agriculture. Sci Prog 100: 80–129.
Romano, D., Donati, E., Benelli, G., & Stefanini, C. 2019. A review on animal–robot interaction: from
bio-hybrid organisms to mixed societies. Biol cyb. 113(3): 201-225.
Schmickl, T., Szopek, M., …, Zahadat, P. 2021. Social integrating robots suggest mitigation strategies for
ecosystem decay. Front Bioeng Biotechnol: 9.
Seeley, T.D., 1974. Atmospheric carbon dioxide regulation in honey-bee (Apis mellifera) colonies. J IN-
SECT PHYSIOL J Insect Physiol 20: 2301–2305.
Spyropoulos, T, 2017. Constructing Participatory Environments: A Behavioural Model for Design, PhD
thesis, University College London.
