Bioluminescent micro-architectures: planning design in time, an eco-metabolistic approach to biodesign

Abstract

By hybridising the principles of the living with the methods of design, the emerging field of biodesign is exploring how to radically transform the ecological imprint of contemporary material culture while questioning the creative opportunities induced by the appropriation of metabolic processes. This new bio-based foundation challenges architects and designers to rethink the way in which architecture is imagined, represented and materialised. This paper presents developments in the speculative collaborative project Imprimer la lumière examining living bioluminescent bacterial substrates as an architectural building material. In order to appropriate the light performance of these living organisms, the paper asks how to characterise and control these within an architectural design context and reports on efforts to develop computational models for simulating the behaviour, growth rates and life span of living materials and interface these with architectural representational framework. Within nature, bioluminescence is predominantly produced by marine organisms. In this context, the emitted light is a chemical reaction, part of a metabolic system that needs to be sustained. Working with bioluminescence therefore implies taking into consideration the ecosystem in which the light-emitting metabolisms take place as much as their limited lifespan. As a consequence, time must be understood as a key dimension of the architectural design process and wet lab tools and critically implemented into the palette of architectural design instruments and protocols.This paper reports on the examination of how living materials and their environment can be represented, simulated and predicted as part of an eco-metabolistic model developing mechanisms of functionalising and steering a living architectural material.

Introduction

By placing living organisms at the centre of the fabrication process, biodesign asks how the organism—whether in a synthetic or existing biological system and no matter its size—can be connected to steer material performance [1, 2]. This creates a radically new departure from which materials are traditionally conceived, fabricated and appropriated in architecture and design as an inert reality [3] and asks new questions not only of its design and crafting processes but also of its ethical and ecological roles [2, 4]. In ‘Imprimer la Lumière’ we use the bioluminescent bacteria Vibrio fischeri [5] as a model organism to probe principles for a living architecture, functionalising its life as an architectural light source [6]. We question how designing a built environment as a host for other species can perform as a living technology, and what this co-living means in terms of environment control and care-taking. Through 3D-printing of the agar-media used to grow bacteria in labs, we build up a model to observe, understand and predict how their life and performance can be steered by designing the host media typology as well as its nutritional recipe.

In order to understand and in turn control and instrumentalise the organisms, a key focus has been to develop first stepping stones enabling a formal representation of living behaviour. This paper reports on the efforts to simulate the dynamics of the living organism and its dependence on environmental factors (Fig. 1). We implement an agent based system operating in voxel matrix reactive to key parameters such as substrate topology, access to oxygen, nutrition level and humidity level. The aim for these simulations is both to steer performance as well as interface with design and fabrication. Representing living systems is complicated for several reasons. Biological processes are interconnected and sensitive to changes in their environment. The life cycles (the time between cycles of propagation) and life stages (changes in performance of the living organism across its life span) necessitate dynamic representations that both extend in time as a fourth dimension and can correlate their interactions.

Fig. 1

Comparison: 3D printed substrates as scaffold for bioluminescent organisms and their time based simulation

Designing with bioluminescence

Bioluminescence is a chemical form of light predominantly emitted by marine organisms but also some mushrooms and insects [7]. Within architecture and design communities, bioluminescence has been examined as an alternative to artificial lighting and is largely induced by micro-algae [8,9,10,11]. Rare physical experimentations with bacterial bioluminescence in the biodesign field have focussed on garments and textiles [12, 13] and bioreactors for urban furniture perspectives [10]. In ‘Imprimer la Lumière’, we have chosen to work with Vibrio fischeri as it is a model organism commonly used as a marker in biology and medicine. It is therefore well documented and accessible allowing us to engage the dynamics of living organisms in an intuitive and immediate manner. The study develops methods for architectural design and fabrication with living systems and the metabolic systems of nutrient absorption and environmental sensitivity. This overarching methodological ambition of the project questions how architectural design can be tuned to engage with living material systems [14, 15] and how architecture itself could become living [16].

In prior work, we have reported on the design of agar substrate and its 3D printing (Fig. 2) using a collaborative robotic setup [17]. Here, key findings that lay to ground for our simulation set up is a strong understanding of the life-spans of the bacterial colonies, the way that they live on the surface and not the depth of the agar–agar medium (medium) and how colony growth rates are intimately connected to light performance. Their growth rates, the speed at which the number of organisms in a population increases, is influenced by the quality of their living conditions and impact performance directly as colonies need to reach a certain size before they start emitting light. In preliminary studies we have examined the methodological principles and found that Vibrio fischeri colonies live on the surface of the medium and that enlarging the surface-to-volume ratio has a direct impact on the intensity of light emission as well as their lifespan [17].

Fig. 2

From slicing preview to printed agar to glowing

Understanding living conditions

Understanding the living conditions of the bacteria is therefore fundamental in our project. Despite Vibrio fischeri being well studied and documented, it is not a straightforward organism to work with. In studies that seek to understand the symbiosis between the squid and the Vibrio fischeri bacteria, wild bacteria in squid host organisms have been seen to glow 1000 times brighter than in lab culture [18, 19], demonstrating the insufficient understanding of its original habitat, the forming of colonies and their dynamics. What is understood is that the environment itself is dynamic; the squid changes the environment of its light organ where the bacteria live, to control the light emission, and thereby the camouflage that it provides, by actively controlling the living conditions of the bacteria [20].

Physical experimentation

Fabrication

As a way of fabricating self-illuminating micro architectures (small scale structure for self illumination) we developed a bespoke 3D printing system, as described in our two previous papers [6, 17]. The medium is extruded through a high precision micro-dispensing unit (ViscoTec ecoPen700) on a collaborative robot (UR5e). Self regulating heating elements keeps the medium fluid throughout the printing process. The medium is based on recipes for lab agar that include nutrients. To improve the viscosity and plasticity for controlling print resolution and flow rate, we add gelatine and glycerol.

After inoculation, the probes are placed in an incubator at 25 degrees. To record the living performance of the bacteria we set up a camera within the incubator to take photos every 30 min. In this way we can evaluate the living organisms through the light they emit. Time-lapse animations compiled from the images visualise the growth and movement of the bacteria over time and enables us to evaluate the success of the designed substrate and how the shaping of it impacts their life-cycles and life-stages.

Learnings

Our experiments show that the Vibrio fischeri are not able to move or propagate noticeably on the surface nor within the medium. Instead they spread easily in liquid broth. This awareness has led us to formulate new design criteria, which have the aim to maintain high humidity levels. This is done by integrating three dimensional basins in which the fluid living broth can be poured after the printing process to form fluid pools. The pools form better living environments for the bacteria and while the pools themselves are dim and diffuse; at the edges of the wet zones the glow is highly intensified. This indicates that this zone offers a good balance of humidity and oxygen for the bacteria. An intensification of bioluminescence is also observed in the process of the pools drying out. Our assumption is that this brightness is caused by a film of humidity which constitutes the preferred living condition. In this way we observe that despite 3D printing the medium in a fixed form, the environment changes during the life span of the bacteria, as oxygen and nutrition is depleted and as humidity levels change.

Simulation

Our simulation of ‘Imprimer la Lumière’ seeks to formalise and represent the dynamic interactions observed in the living experiments. The simulations learn from existing simulation methods, but develop a distinctive multi-agent approach. This allows us to simulate the development of the bacterial colonies and their interactions with their environment as they metabolise nutrition and oxygen. This approach places time at the centre of the simulation. In the simulation, time is abstracted as successive time steps and not represented as measured time. The simulation also abstracts the light performance of the bacteria. Vibrio fischeri’s signal sensitivity has been simulated [21], but does not allow for an understanding of the overall light performance. Capillary effect and surface tension have proved central in our investigations, but these physical phenomena, caused by forces on a molecular level, are too complex to simulate. This paper develops a simplified agent based system operating in voxel matrix correlating the growth rate with the key parameters observed in living experiments: topology, access to oxygen, nutrition level and humidity level.

Voxel simulation for living systems

The simulation is based on a voxel logic. Voxels, often referred to as 3D pixels, are a tool used for abstract spatial representation. Stripped down, they are numerical values arranged in a 3D grid [22]. They provide a simple and intuitive framework, easy to visualise in 3D, where complexity can be gradually added. Our simulation builds a voxelised representation of the host environment; the 3D printed form of the agar media, the integrated pools and the oxygen within the host environment (Fig. 3). As in the physical reference experiment, we start by inoculating everywhere.

Fig. 3

Stages of the simulation: printed geometry, adding liquid where the broth was poured, voxelising, simulating, post production blurring in photoshop

The voxels’ position in space is implicit by their placement in the 3D array storing them. Each voxel has variables describing its physical state, including; humidity, nutrient level and oxygen which are calculated in respect to the physical properties; gravity, diffusion and evaporation that produce their interactions.

Each voxel has a number describing the bacteria count of the single voxel, which is a function of the physical state. The simulation operates through a principle of neighbourship. In ‘Imprimer la Lumière’, the voxels interact with each other through its cubic 26 neighbours. For each time step of the simulation the physical conditions and bacteria count of each voxel are updated by defined equations using their previous conditions and their neighbours’ conditions as input. This creates a method of simulating physical and bacterial states in the 3D model over time.

Implementing rules

In our multi agent system we have three agent definitions; medium, liquid broth and air. We chose humidity and nutrient level to be the physical variables describing the condition in each voxel. Since the liquid broth and the agar medium contain different amounts of nutrients and water, and since the bacteria act differently in the two mediums, we split them into two voxel types. The third voxel type, air voxels, are used to define which voxels are on the surface of the media. Bacteria can live in broth voxels and in the agar voxels which are on the surface of the geometry.

The number of bacteria in a medium or liquid broth voxel is set as a relative number between 0 and 1. As we observe a dimmer light in the liquid broth than on the medium, the maximum amount of bacteria is here set to 0.1 instead of 1, representing the stationary phase, where the rate of growth and death are equal [23]. To model propagation in the medium, the bacteria number will double at each time step, as long as there is humidity and nutrition in a voxel. This continues until the population reaches the maximum level. In liquid broth, this propagation doubling is further transferred to each neighbouring broth voxel. In the medium, the bacteria population must reach its maximum before transferring bacteria to neighbours. The bacteria count in each voxel affects and changes the available nutrition state as bacteria delete nutrition across their lifespans. The nutrition state is calculated as the previous amount of nutrient subtracted with the amount of bacteria.

Humidity is the most complicated state to calculate as it is fluid and subject to evaporation, diffusion and gravity. Gravity plays an important role as all our prototypes dry from the highest point and down. We calculate humidity as dependent on the saturation state of each voxel across the time steps. As humidity falls through the medium in time, it passes on humidity to its lower neighbour voxels as long as their state is not at full saturation. The combination of diffusion and gravity is particularly important for simulating the fluid behaviour of the broth. If two neighbouring broth voxels have different humidity levels, humidity will be transferred between them to even out the difference. The rate of evaporation is proportional to the voxels' exposure to air. When a broth voxel dries out it is replaced with an air-voxel and all the nutrients and bacteria are transferred to the voxel underneath.

A second humidity based rule mimics the capillary effect at the edges of the pool. Liquid broth voxels that have at least one horizontal medium voxel neighbour are set to have a lower evaporation rate and no diffusion rate. The voxel will still transfer humidity to lower voxels by gravity. This should ensure that the liquid voxels at the edges of the pools are one higher than the rest of the water surface.

Tuning

Since we are not working with exact physical values and forces, tuning the simulation to real world observations is essential. As the rules and parameters are interconnected, we cannot tune the parameters independently. The observations used for reference for tuning are based on the photographic registration images taken within the incubator (Fig. 4). Here, we can see that the bacterial colonies are largest and strongest 24 h after incubation as nutrition and oxygen levels have not yet been depleted. As these observations are taken from the top view, we have rendered the 3 dimensional voxel model from the same view.

Fig. 4

Frames every 12 h from the experiment used as reference for the simulation

Resolution

During the tuning of the simulation we found that the resolution of the model fundamentally impacts results. The resolution is important both for a recognisable representation of the geometry and the accuracy of the simulation. Running-time is a concern when deciding on the resolution. Since we are working with 3D arrays, the time complexity of the code is O(n^3), which means that doubling the resolution will slow down the runtime 8 times. On the other hand, a higher resolution is important to get realistic results, otherwise small geometrical details might be neglected. During our experiments we found that for our geometries, the resolution of the voxel framework should be high enough for every 3D-printed wall to be minimum 3 voxels thick so that the outer voxels representing the surface are not directly neighbouring the surface on the other side of the wall.

Our initial intention was to prototype and test at low resolution so as to tune and validate the model in respect to observed results, and then increase resolution for verification. This quickly proved ineffective as the system did not scale the same performances. Impediments for scaling were also found in the code basis of our simulations. Initial tests undertaken in the graphical programming environment Rhino/Grasshopper™ were replaced by a new infrastructure using Python with NumPy [24] for both calculation and visualisation. This allowed us to speed up the iterations through the 3D array drastically.

Our designs are informed by the smooth continuous lines that the 3D printing of medium performs best with. This base geometry is incompatible with the orthogonal logic of the voxels, creating a pixelated look. The incompatibility also gives practical issues. Neighbouring surface voxels on a curved wall can have drastically different exposure to air, (Fig. 5), and therefore change humidity level at different timerates. This is evident when looking at the simulation in perspective view where non-verified vertical lines of darkness become observable on an otherwise light surface. To mitigate this we implemented a diffusion rule set, but this complicates the tuning and slows down runtime.

Fig. 5

The pixelated curve gives the medium an uneven exposure to air and therefor an uneven drying

Diffusion

In order to tune the interaction between the liquid pools and their drying, we use several small steps of gravity and diffusion with small coefficients. Diffusion enables us to even out humidity and its drying, but if it is not implemented and tuned correctly, it can give the opposite effect. When using larger coefficients, the system becomes chaotic (Fig. 6). Diffusion at 2% of the difference between the humidity in neighbouring cells has empirically proven to replicate results well. This coefficient has to be tuned with the number of diffusions per timestep. When there are too many diffusion sets in a row without other influences, it creates a chaotic behaviour of self-resonance. We observe similar behaviour when diffusion is applied in areas with low differences between neighbouring voxels. We find that these chaotic behaviours can be mitigated by adding a gravity-update between every diffusion state change.

Fig. 6

Example result of chaotic, uneven and untuned diffusion

To allow for a more intuitive understanding of the interactions between bacteria growth rate and light emission, we found that examining the state change of the humidity level in isolation (Fig. 7) allowed us to build an understanding of the growth environment. By validating the performance of the humidity level across the runtime of the model, we were able to save calculation time and consolide parts of the model before looking at the interaction between all parameters.

Fig. 7

Looking at how liquid is simulated alone. White is full of humidity, black is dry

Representation

For the representation, we used Matplotlib in Python. The colour displayed for each voxel is given by the number of alive bacteria relative to the maximum alive bacteria. The colour cyan (rgb(0, 255, 255)) is used for the maximum light approximating the colour of the emitted light from the bioluminescent bacteria. The colour is set linearly, e.g. a voxel “half full” of bacteria is assigned the colour rgb(0, 127, 127).

A central challenge in the simulation is to represent the transparency of the voxels and the light they emit through neighbouring voxels. In the top view, this is solved by adding the value of light in the whole vertical column (projection). In 3D this calculation is more complex and we therefore settled on generating 3D-views without transparency. As the top view is used as the central validation data corresponding to our photographic recordings we are able to validate the model. However, we observe that the perspective views enable us a better understanding of the interactions in the model. Further research will investigate maximum intensity projection and volume rendering [25] as used in CT and astrophysical data enabling an implementation of an additive logic of light emission.

To increase the visual similarities with the reference photos we use blur and median filters to reduce the pixelated look (Fig. 8). Using libraries for image processing in python enables us to integrate this step as part of the simulation.

Fig. 8

Using Image filters from Python Pillow library. Blur + median filter, × 0, × 1, × 3, × 5

Results and discussion

When comparing the simulation with the photographs used as reference (Fig. 9) the similarities are pronounced across the time steps of the simulation. The drying of the medium from the top of the geometries reproduces the incremental diminishing of light from the centres of the structures. The intensification of light within the pools as they dry out is also persistent with observed results.

Fig. 9

Comparison of photographs and simulation

The core limitation of the simulation lies with its pairing to the topology of the 3D printed micro-architecture structure. The basic interactions of the multi-agent model have been tuned to correlate the performances of the sample model and will still need to be tested across a series of other topologies to be fully verified. Furthermore, it will be important to challenge the basic topology of the 3D printed micro-architecture structure in order to fully test the models transferability.

Accepting these central limitations, the most significant deviation between simulation and the real observation is seen in the first hours. This can be explained by the lag phase which the simulation does not consider. The lag phase is the stage in bacterial growth when the bacteria are first introduced in a new substrate and need time to get established before they start propagating. The length of this phase depends on the environment and how different it is compared to its previous environment [23]. This means that the conditions of the culture which is used to inoculate is important for predicting when the first light occurs. This difference in time is expected as we did not tune the time parameter of the simulation in this research phase. We would therefore not expect a direct comparison between hours passed and iterations of the simulation. Defining the temporality of the simulation is hard as it necessitates the correlation of the exact control of the environment of the physical experiments. The drying process depends on the specific ventilation of the petri dishes and relative humidity in the incubator.

Another expected reason for differences are the imperfections of the 3D-printed media compared to the 3D-model it is fabricated from. The thickness of the extruded media is not completely consistent and some of the pool walls have holes. These pools will dry out earlier than predicted. Integrating 3D-scanning into the workflow can help bridge the gap between the designed shape and the fabricated result.

Conclusion

The simulation in ‘Imprimer la Lumière’ examines the making of a multi-agent simulation that is able to calculate the complex interactions between a living organism and its heterogeneous and dynamic environment. It builds the foundations for a new way to understand the performances and interdependencies of a new class of living architectural materials. The primary contribution of the study is the conceptualisation of a new representational framework in which the model calculates its dynamic interactions across time. The study finds that the heterogeneity of the environment; its composition of medium, liquid broth and oxygen and the complex dynamics of changing humidity levels as well as gravity and dispersion can be replicated within a voxel based simulation. The study succeeds in validating the model and verifying results between the observed life states and life stages of the living experiments and the simulated model. As such, the model replicates, in time and across material, the light performance of the living organism.

The model builds the conceptual as well as instrumentalised stepping stones for understanding how design can steer the performance of a living architecture. By parametrisising key environmental factors and correlating their behaviour, we are able to build the foundations for an appropriation of their steering into the architectural and design-led domain.