Image Image Image Image Image Image Image Image Image Image

Bartlett School of Architecture, UCL

Scroll to top


No Comments

Skin Colouration in Silicone Wearables

Skin Colouration in Silicone Wearables
  • On September 19, 2017

According to Witzany (Witzany, 2014) communication between organisms in all domains, from bacteria and fungi, to mammals, including humans, is necessary in order to coexist. In all the kingdoms of life, no coordination or organisation is to be found that does not depend on communication. This communication can be in the form of vocalisation or sound, tactile behaviour, visual gesture, or combinations of these. This means that the variety of ways in which organisms can communicate is huge.

The skin, as the surface organ which surrounds all other internal organs, is that which interacts most immediately, both with the external environment, and with other organisms, and for this reason generates new forms of communication, which have been enormously enhanced by new technologies.

More recently, wearable technologies have played an ever-growing role in human communication, increasing the number of ways in which we can communicate. As stated by Suh et al. (Suh et al., 2016) they have changed our perception and our awareness of our own bodies, as information provided by them can give us deeper insight into the states of our bodies. As Nunez-Pacheco et al. (Nunez-Pacheco et al., 2014) suggest, they have also created new and alternative opportunities for humans to experience their position in their environment, and new ways of communicating with others. According to Ferraro et al. (Ferraro et al., 2011) they have “mediated the ways in which an individual is perceived by others, interacts with others, and manages his/her own physical space”.

The Aposema project investigated these assertions further, taking one particular case as an example. The aim was to present to the world, through a silicone mask, a new identity for the wearer; an identity constructed by using elements of both the atmosphere, and the wearer’s body. The challenge arising here was how to construct a new identity for the individual through the mask. Visual elements contained within the mask were chosen as a means of establishing this new identity, and colour played a fundamental role in its construction.

This research will focus on analysing different available techniques of colouration in silicone, and select one of these techniques for use on the silicone mask proposed by Aposema, through an experimental study. Communication between the external and the internal will also be a keystone of this thesis, and techniques that permit the mask to sense both the environment and the body will also be investigated and evaluated, as well as the integration between the perception and colouration mechanisms.

Sources of inspiration for this project include the mechanisms used by some animals for concealment, communication, predation and reproduction; mechanisms which are based on perception of the environment and change of coloration (Yua et al., 2014). For example, chameleons change the colours and patterns of their skin to communicate with the outside environment, and with others of their species. They can attract a mate, or defend their territory by flashing bright colours to each other, or can blend into their environment to evade predators, by altering the shade of their skins (Casselman, 2008).

Video 1: Chameleons Are Amazing | National Geographic. Available:

From the inspiration provided by the mechanisms of nature, such as those mentioned above, has emerged the field of biomimetics, which seeks to solve complex human problems using these natural mechanisms. The core concept is not outright imitation of the mechanisms, but rather using only their functionalities. One area that has exploited these ideas is soft robotics.

Soft robotics is a new field which explores the building of ‘soft’ robots, made of fluids, gels, and elastomers that match the elastic and rheological properties of biological tissue and organs. These elements combine to increase flexibility, and to allow adaptations for accomplishing complex tasks that rigid robots could not perform. One of the issues faced by researchers in the field is the colouring of these soft robots.

Video 2: Introducing the Octobot | Harvard University. Available:

Several proposals have been made in this respect, following the ideas of biomimetics. One proposal is to build robots that can match background colouration (Morin et al., 2012); another is to develop skins that react to touch by colouring (Chou et al., 2015). These proposals are relevant to this dessertation, because they suggest ways of colouring robots built with materials similar to those used in the mask proposed by Aposema, and therefore could have some applicability in this project.

Colouration of Aposema’s proposed mask raises several challenges that have not been addressed by other projects, such as the need for the mechanisms of colouring and change of form to work simultaneously, the necessity of a certain level of portability, the potential requirement to react to signals invisible to the human eye (such as UV light), and/or the necessity for the colouration to be changed dynamically, according to the conditions of the environment or the body of the wearer.

This report will present the experiments performed to solve these problems, in order to build a mask that has the above-mentioned functionalities, as well as the results obtained.

The report will take the following form; initially the concept of biomimetics will be presented, using as an illustration the ability to suddenly change colour found in some animals, and how this has inspired some researchers to imitate this functionality in soft robots. There then follow outlines of different methods that have been proposed for colouring soft robots, and the advantages and disadvantages for using each of these methods in the Aposema project will be described. Subsequently, the experiments carried out to develop a mask that reacts to the environment, and to different facial expressions, depending on the atmosphere and body motions of the wearer, will be explained; emphasis here will be placed on the colouring process. The results of the application of the above techniques will then be detailed, and finally some general conclusions will be presented.


Biomimetics refers to human-made processes or systems that imitate nature, as illustrated by the following example.

According to Yua et al. (Yua et al., 2014), some animals, such as chameleons, and some cephalopod species, have the remarkable ability to change their skin colour, for the purposes of concealment, communication, predation, and/or reproduction. This unique characteristic has long inspired scientists to construct artificial systems which mimic such a function. One of the areas in which this idea has been employed is in soft robotics. Robots have been built that mimic their environment, or that are able to change their colour, dependent on certain circumstances. Different techniques for constructing such robots and managing such colour change have been proposed, and these will be described in the following section. These may then serve as inspiration for project Aposema, which seeks, specifically, to find appropriate methods of implementing colour change in a silicone skin, dependent on different environmental aspects. In this section, the mechanisms used by some animals for changing their skin colour will be presented.

2.1 Colour change in animals

As stated by Morin et al. (Morin et al., 2012) cephalopods, such as squid and cuttlefish, exhibit remarkable control over their appearance (colour, contrast, pattern, and shape). These animals use dynamic body patterns for disguise, for protection, and for warning. Other animals, such as chameleons and many insect species, are also able to actively change their colouration for camouflage or display. Others still, such as some jellyfish species and fireflies, use bioluminescence to communicate (Morin et al., 2012). As two examples of these possibilities, the cases of the cuttlefish and the chameleon will be presented.

2.1.1 Mechanisms used by the cuttlefish for camouflage

Buresch et al. (Buresch et al., 2011) state that “cuttlefish are one of few animal groups with the ability to camouflage themselves on a wide variety of backgrounds, from open sandy plains to complex coral and rock reef habitats. Because the colour, contrast, patterning and physical texture of their skin are under direct neural control, camouflage is almost instantaneous”. According to the authors, cuttlefish use camouflage to avoid detection or recognition by predators. This is accomplished using two tactics: background matching, and resembling an object in the immediate area (known as masquerade).

Figure 1. A camouflaged cuttlefish.

Figure 1: A camouflaged cuttlefish.

Brooks (Brooks, 2008) points out that cephalopod species, such as the cuttlefish, have “such remarkable camouflage primarily because of their chromatophores — sacs of red, yellow or brown pigment in the skin made visible (or invisible) by muscles around their circumference. These muscles are under the direct control of neurons in the motor centres of the brain, which is why they can blend into the background so quickly. Another aid to camouflage is the changeable texture of cuttlefish skin, which contains papillae — bundles of muscles able to alter the surface of the animal from smooth to spiky. This comes in pretty useful if it needs to hide next to a barnacle-encrusted rock, for instance”. According to the author, the above mechanisms are complemented by leucophores, cell structures that sit underneath the chromatophores, and which reflect light across a wide range of wavelengths. This means that they can reflect any available light; for instance, white light in shallow water and light blue light in the depths.

2.1.2 Chameleons colour changing to communicate

Casselman (Casselman, 2008) describes how chameleons change the colours and patterns of their skin to communicate with the outside environment. They do not use vocalisation or pheromones, such as some other organisms do. Distinct colours and patterns mean specific things. For instance, a male chameleon may be perceived by rival males as more or less dominant, depending on the brightness of the colours he displays. In this way, male chameleons can attract a mate, or defend their territory by flashing bright colours to each other. When a male displays drab browns and greys, this indicates submission or surrender. Furthermore, to blend into their environment, chameleons can rapidly alter the shade of their skins. All these transformations constitute a language of communication. Figure 3 illustrates the above.

Figure 2: Close-up of the skin of a chameleon.

Figure 2: Close-up of the skin of a chameleon.

Figure 3: Chameleon changing colour in an excited state.

Figure 3: Chameleon changing colour in an excited state.

The transformations described above are produced through a combination of pigments and structural colours. The Ask Nature Team (Ask Nature Team, 2016) explain that “the chameleon skin contains different types of chromatophore (colour-bearing) cells organized in layers within the skin. The upper layer of skin contains cells with yellow and red pigments, while lower layers contain cells with dark melanin pigment, which appears black or brown. Just below the layer of yellow and red chromatophores is a layer of cells called iridophores (iridescent chromatophores) that produce structural colour. Rather than containing pigment, iridophores contain an organized array of transparent, nano-sized crystals that reflect specific wavelengths of light. The reflected light is perceived as colour”.

As described above, this mechanism for colour changing is based on actively adjusting the spacing between nanocrystals, reflecting different wavelengths. In this way, the crystal structures and pigments together produce the overall colour of skin. For instance, blue light over yellow pigment generates a green colour. This is done, according to Yua et al. (Yua et al., 2014), by rapidly retracting or expanding the chromatophores, via direct control of muscles that are in turn controlled by nerves originating in the brain. For the reader interested in this subject, Teyssier et al. (Teyssier et al., 2015) present this topic in greater depth.


According to Pawlyn (Pawlyn, 2011) “bio-inspired design or bio-design emerged as a term partly in the medical world (investigating and implementing new biomedical technologies), partly in robotics, and partly as a broad definition encompassing a range of design disciplines based on biology”.

The field of robotics involves enormous challenges; one of the most recent challenges is the design of soft robots, inspired by nature and biology. These robots are constructed with soft and deformable materials, such as silicone, plastic, fabric, and rubber. Ordinarily, they have no rigid body, but rather a large capacity for deformation and durability. These qualities make these robotics suitable for being placed on the body as a second skin, and allow them to blend easily with the user him- or herself. One of the core challenges pertaining to these robotics, as seen in the case of the Aposema project, is how to give them adequate colouring, and one which is suitable for the specific needs the robot fulfils. It is imperative that a suitable range of hues can be obtained, and also that these hues are compatible with the required durability and deformation of the robot. Different techniques of realising this have been proposed, and these will be presented below.

3.1 Chameleon Robot

Chameleon is a soft robot that was developed by Morin et al. (Morin et al., 2012) from the Department of Chemistry and Chemical Biology and the Wyss Institute of Biologically Inspired Engineering of Harvard University. It contains simple microfluidic networks that can change its colour, contrast, pattern, apparent shape, luminescence, and surface temperature for camouflage/display. The colour can be changed simultaneously in the visible light spectrum and the infrared spectrum, which offers an advantage that is not shared by any living organism.

 Figure 4: Design and operation of a colour layer.

Figure 4: Design of a colour layer.

Camouflage or display colouration is produced by pumping coloured or temperature-controlled fluids through a network of microfluidic channels, contained in thin silicone sheets, and referred to as ‘colour layers’.

The liquid can be coloured with dyes or pigments, or can be heated, or cooled, to alter the colour, thus producing a wide spectrum of available colours, ranging from visible hues to infrared. The operation of colouration is reversible. Unlike other systems that use technologies like electrowetting or electrofluidics, the colour layers in this system, once filled, require no power, and have low requirements for volume of fluid. Chameleon also possesses mobile and elastic capabilities, which are completely compatible with the colour layers.

Figure 4 shows various patterns of colouration generated by filling (or by not filling) the channels of a completed colour layer with solutions of dye (the top four images), and pigment dispersions (the bottom two images).

The idea of using micro-channels which allow changes in colour (and apparent shape) through the introduction of coloured or temperature-controlled fluids, is a promising one for the Aposema project, as this would allow simultaneous activation of the mechanisms of coloration and change of form. It also allows dynamic changes in colouration, according to the given circumstances of the environment or the body.

3.2 Octobot

Octobot is a soft robotics project developed by Wehner et al. (Wehner et al., 2016), researchers from the School of Engineering and Applied Sciences and the Wyss Institute for Biologically Inspired Engineering of Harvard University, the Dalio Institute and the New York Presbyterian Hospital and the Department of Radiology of the Weill Cornell Medicine and the Department of Chemistry and Chemical Biology of Harvard University.

The Octobot, unlike conventional robots that are made with rigid materials, has the ability to adapt to different environments. It represents an outstanding innovation in technology by creating a small, autonomous, soft robot, which has an actuation system in which channels are interconnected to produce pneumatic motion that is triggered by fluids or gases, supplied via tethered pressure sources. This robot has the ability to crawl and swim, without being connected to any mechanical device.

“The robot is controlled with microfluidic logic that autonomously regulates fluid flow and, hence, catalytic decomposition of an on-board monopropellant fuel supply. Gas generated from the fuel decomposition inflates fluidic networks downstream of the reaction sites, resulting in actuation… [This] rapid decomposition into gas upon exposure to a catalyst offers a strategy for powering soft robotic systems that obviates the need for batteries or external power source[s]” (Wehner et al., 2016).” (Wehner et al., 2016).

Figure 6: Fully soft, autonomous robot assembly.

Figure 5: Fully soft, autonomous robot assembly.

In order to colourise the Octobot, a microfluidic logic is used as a soft controller, and a multi-material printing method is employed to fabricate a pneumatic coloured elastomeric robot body. The process of fabrication is shown in Figure 5. Image (a) depicts the fabrication of the micro-mould (the soft controller that contains the microfluids which trigger fuel decomposition). Image (b) shows matrix materials which are poured into the mould. Images (c) and (-d) show fugitive and catalytic inks, which are EMB 3D printed into the mould matrix. Image (e) shows a detail of how the matrix material is cross-linked. Image (f) shows a detail of the elevated temperature process, whereby water can evaporate, to leave behind an open network of channels inside the robot. Image (g) shows the completed Octobot design.

Figure 8: Inflation of tentacles in the Octobot.

Figure 6: Inflation in the Octobot.

“To achieve the desired autonomous function, we incorporated a soft, microfluidic controller is within the Octobot. The control system is divided into four sections: upstream (liquid fuel storage), oscillator (liquid fuel regulation), reaction chamber (decomposition into pressurised gas) and downstream (gas distribution for actuation and venting)” (Wehner et al., 2016).

Although in the proposed method the robot is composed only of soft materials and thus has the advantage that it does not require its connection to any mechanical device, which would facilitate its integration into wearable technology, the complexity of the chemical processes involved in its construction and operation (microfluidic logic, catalytic decomposition of a monopropellant fuel) would realistically render its use in the Aposema design very complicated. It is also not clear how a wide range of colours can be obtained, nor how they can be dynamically changed.

3.3 E-skin

E-Skin is a project that was developed by Chou et al. (Chou et al., 2015), from the departments of Chemical Engineering, Electrical Engineering and Materials Science and Engineering of Stanford University.

This skin is a “chameleon-inspired stretchable electronic skin (e-skin), in which the e-skin colour can easily be controlled through varying the applied pressure along with the applied pressure duration. As such, the e-skin’s colour change can also be in turn utilized to distinguish the pressure applied” (Chou et al., 2015). This touch-responsive new technology has a wide range of applications, such as interactive wearable devices, prosthetics, and smart robots.

Figure 9: Illustration of the concept of a chameleon-inspired e-skin.

Figure 7: Illustration of the concept of a chameleon-inspired e-skin.

E-skin is stretchable and has interactive colour-changing and touch-sensing properties. These are achieved through a resistive pressure sensor that can be highly tuned, and a stretchable organic electrochromic device. The latter is activated by applying a certain level of voltage to an electrochromic polymer (P3HT) (Chou et al., 2015).

Despite some advantages of this method used by e-skin, such as low energy consumption, which would certainly facilitate its integration into wearable technology, there are some drawbacks for its application to Aposema: firstly, the complexity of the chemical processes involved, and secondly, the difficulty of using an integrated mechanism for colouring and changing the apparent shape.

3.4 Artificial emotion expression by dynamic colour change

According to Nakajima et al. (Nakajima et al., 2017), the perception of facial expression is influenced by facial colour. For example, fear and sadness could be linked to a bluish cast to the face, and both anger and happiness could be linked to a reddish face.

In a project developed by Terada et al. (Terada et al., 2012), from the Department of Information Science of the Faculty of Engineering of Gifu University, the emotional expression of a robot was generated by dynamically changing the colour luminosity of its body. The idea was to generate a colour-based language, aimed at, and comprehensible by, humans, to communicate emotions non-verbally. Although the project worked with traditional robots, the underlying ideas can also be implemented in soft robots.

The expression of emotions here was performed through colours. Colouring occurred through LEDs. Dynamic change in the colour luminosity of an LED was implemented by changing the frequencies and waveforms of blinking. The coloration system consisted of 70 LEDs grouped in 20 units that were autonomous (that is, each could be illuminated in different colours), and a controller. Each unit had two or three adjacent LEDs. The overall, composite colours were obtained through the combination of the three individual colours of the LEDs, red, green and blue, emitted by the LEDs in different brightness.

The following figure shows colours and waveforms for 8 emotions:

Figure 10: Colours and waveforms for 8 emotions.

Figure 8: Colours and waveforms for 8 emotions.


The main objective of the Aposema project is to present a new identity for an individual to the external world, through a silicone mask, based on some elements of the atmosphere, and the wearer’s body.

The main challenges inherent in this are how to capture the required elements of the environment and the body, and how to give a suitable expression to the mask. Techniques must be developed to obtain silicone with the required characteristics for the mask, and to allow the mask to connect to the outside world and the
body, and respond in terms of colour and shape. This section will present the different experiments performed to these ends, and the most important conclusions regarding different methods of colourising the silicone, and the advantages and disadvantages of each one.

Preliminary investigations looked at how to mix the silicone, in order to alter its normal transparent state. These experiments were carried out through the addition of different pigments to the silicone, which added a layer of skin colouration with a realistic output. The second investigation aimed to identify appropriate mechanisms for sensing the atmosphere and the wearer’s body. In terms of atmosphere, the decision was made to focus on sensing the colour of the environment around the wearer. The challenge was to find a method that would be able to quickly detect changes in environmental colour, and that would take into account also the movements of the user. In terms of the wearer’s body, the decision was to sense the wearer’s face. It was important that the expressions of the wearer were reflected accurately.

As a final stage, the objective was to find a way to generate expression in the mask by changing colour and shape. These expressions must reflect the aspects sensed both in the environment and the wearer’s face. The challenge in this case was to precisely identify techniques that would achieve this goal.

4.1 Mix and colouration in silicone

Silicone is a material that is not easy to manage. It exists as a liquid, and is obtained after mixing two different constituent liquids. The investigations carried out for this report used silicone Smooth-On EcoFlex® 00-30, 00-50 00-35 (for fast prototyping), Dragon Skin® 10 FAST and Mold Star® 20T. The resulting type of silicone is recommended for use on human skin, and was chosen as a fit for the design of the mask. The properties of each of these constituent materials influence the hardness of the resulting silicone after casting. The lower the viscosity of these liquids, the easier it is to mix and pour them, and in this way the final material can be made harder and more resistant.

The constituent materials come in two parts, A and B. Part A is mixed with part B in a desired proportion called the mix ratio (calculated either by volume or weight), and the mixture is then either poured, brushed or sprayed onto a mould. Mix ratios vary from product to product and are always listed on the accompanying technical information for that product. An accurate gram scale or a triple beam balance must be used if the components are to be mixed by weight. If the scales are not accurate, the silicone will not cure, even if it is left to dry for a longer period of time than that specified in the technical information.

In the design process for these experiments, moulds were 3D printed, and cast manually.

4.1.1 Pigmentation in silicone

Aposema encountered enormous challenges because the designs were being made for wearable technology that had to represent and convey information from the most detailed and expressive part of the body, the face. Skin has a huge range of colours, especially facial skin, because it is the part of the body which is the most directly exposed to the environment. As external factors, such as potentially harmful UV light are constantly coming into contact with the face, the coloration also has a lot of fine detail. Freckles, spots, sunburn, blushes, as well as hairs and many other layers or marks may be present on the skin of the face, and this is extremely challenging to match.

Attempts were made to mimic skin coloration by using Silc Pig® pigments in a base layer mixed with coloured flock. To make the silicone look lifelike required a great deal of time and experience. It would normally require a professional brush paint to paint in all the details and colours that human faces have, where the imitation skin to be subjected to close scrutiny. However, Aposema research is not concerned with the area of FX makeup design, so this technique was necessarily explored in a monochromatic way.

 Figure 11: Skin colour mixing

Figure 9: Skin colour mixing & Figure 10: Skin colour mask

The first step was to mix the Silc Pig® pigments with part A of the silicone (Figure 9). As these pigments are very concentrated, a very small amount of pigment will colour a proportionally large amount of silicone. The more pigment added in proportion to the volume of liquid silicone, the more dramatic the resulting colour effect. After obtaining the right match, part B of the silicone is mixed for casting. Figure 10 shows the results obtained.

This method of pigmentation in silicone has some disadvantages. Firstly, it does not admit of dynamic colouration, that changes over time, for example, depending on the conditions of the environment, or of the body. Secondly, it does allow complex colouring as would be needed in the case of the realistic representation of facial skin, but it is difficult to achieve.

In order to allow for more flexibility, Aposema was interested in effecting colour change by implementing channels, as has been done in some of soft robots described above. It was therefore important to have transparency in the silicone, to make it possible to see the design of the air channels and the air pockets. For this reason, pigmentation in silicone was not the best option, because it would hide everything beneath the upper layer of silicone.

Therefore, it was decided to implement this technique by applying a monochromatic color palette with a lower density of SILC Pig ® pigments in the outer layer of design, so that the silicone has transparency and contrasts with the inflation layer containing the color change, in order to make it more noticeable.

4.1.2 Fluorescent paint and air pockets

The question of how humans might be able to extend their sight and see beyond the visible light spectrum was also taken into consideration. Technology could possibly allow humans to see more frequencies of light, such as some other animal species already do; honeybees, for instance, are capable of seeing ultraviolet light. To this end, an experiment was devised which would show how paint mixed with silicone part A would look if it were exposed to UV light.

Figure 13: Fluorescent paint with colour mixed in silicone body and air pocket

Figure 11: Fluorescent paint with colour mixed in silicone body and air pocket

This experiment involved the mixing of neon nights Ultraviolet|UV|Black Light| Fluorescent Glow wall paint with silicone EcoFlex® 00-30. Two different methods were employed (Figure 11). The colouring in dotted rectangle A of this figure is made by mixing the wall paint in part A of the silicone. Because the paint is not made for silicone, some white gaps between the coloration appear in the mould after casting. The result looks messy and unpolished.

The experiment shown in dotted rectangle B involved painting the silicone after casting. In this case, the air pockets, which were already designed in the 3D printed mould, could be easily coloured by hand (Figure 11).

Video 3: Aposema: Fluorescent paint with UV light | Aposema. Available:

In summary, the fluorescent paint method is interesting because it reacts to UV light, and allows individuals to perceive signals that normally cannot be detected.

Aposema decided to use fluorescent paint inside the air pockets, because even though the process of painting the air pocket manually is not that accurate, when the design is exposed to UV light, these small details are not evident (Video 3).

4.1.3 Thermochromic paint

Aposema’s objective is to explore different ways of communicating something that is otherwise hard to see or perceive, as is the case with UV light, detailed in the previous section. This next experiment looked at generating colour changes by an increase or decrease of temperature, something that is not always immediately detectable by humans.

SFXC thermochromic paint changes colour when the surface to which it is applied reaches 31ºC or above. It was selected for use in this experiment, precisely because the temperature required for the change to happen is low, compared with other SFXC thermochromic paints.

The experiment was carried out by pumping water at a high temperature into a cast silicone mould mixed with thermochromic paint. The mould had air pockets of different sizes, so that the change of coloration could be analysed. Video 4 shows the results obtained.

Video 4: Aposema: Thermochromic Paint | Aposema. Available:

One limitation of this method is apparent in the fact that because the room temperature is relatively high, the differentiation of colour change is so subtle, that it is almost imperceptible. One solution for this would be to use another SFXC thermochromic paint that triggers the colour change at a higher temperature. A more significant limitation lies in the fact that the air pocket needs to be of a diameter of at least 0.15 mm for the colour change to be appreciable. This limitation is important with regard to Aposema, as in this case the number of air pockets would of necessity be reduced considerably, and the parametric design process could not be accomplished.

The use of thermochromic paint is an interesting possibility, because it may allow an opportunity for humans to apprehend environmental factors (such as light spectrums or small variations in temperature) that are not always otherwise detectable by the human body, and would in this way increase the individual’s perception of the world. The drawback remains, however, that this method requires specific thermochromic material that activates at a certain temperature, which depends on the particular situation.

Aposema decided not to use this technique primarily because it is necessary to heat or cool the liquid that would be pumped into the mask for the colour change to occur, and this would add a layer of difficulty that can be resolved in a simpler way by using other methods.

4.1.4 Food paint and air channels

In order for the mask to accommodate a degree of mobility, it is important that the design factors in that the colour change can be effected while any deformation occurs. The use of liquids could help facilitate this. Rainbow Dust Color Flo Liquid Food Concentrate was used for this purpose. Pumping this into several channels, which can be activated independently, also served as an experiment to try to achieve the optimum colour change. The results obtained are shown in Video 5. This method is similar to that used in the Chameleon system referenced in section 3.1.

Video 5: Aposema: Actuation with Liquid | Aposema. Available:

One of the advantages of using liquids for coloration is that the gradient of colour can change subtly, depending on the amount of liquid pumped inside the air pockets. As the silicone in this design is transparent, the more liquid, and the bigger the air pocket, the more intense the coloration produced (Figure 12).


Figure 12: Change in gradient of colour.

This process required a design which included a liquid pump that allowed liquids to flow along the mask, and different containers for carrying the different liquids that were to be pumped inside the chambers. The liquid flowed along the chambers and could produce the inflation of the air pockets.

The advantage of using food paint and channels to colour the silicone in the case of the mask (and of soft robots) is that change of form and colour can be achieved simultaneously. In addition, colour changes can be made dynamically, as is required in the case of the mask that must react to certain conditions of the environment and body.

4.2 Colour change by sensing the environment

It is impossible, or at the very least unhelpful, to conceive of a body or object without considering the links between it and the space, environment or atmosphere around it. At the same time, the physical attributes of said body or object must be considered, and one of the chief aspects, in regard to this report is, of course, colour. In this context it is vital to consider the individual’s perceptions of his or her environment or atmosphere, and how colour might factor into this equation, in the same way that some animal species (detailed above) change their own external appearance (in form or colour) depending on the colour of the environment. So, if the mask is to react to the surroundings, it must connect with them.

4.2.1 RGB sensors with Flora RGB Smart NeoPixel

This experiment consisted of reading the colours that surround the wearer, in order for the mask to mimic them. For this purpose, an Adafruit Colour Sensor was used, as well as Flora RGB Smart NeoPixels, that are already designed to be worn on the body.

The Flora RGB Smart NeoPixels have full 24-bit colour ability, and the controller chip takes care of the pulse width modulation (PWM). Since the LED is so bright, less current is required to power the output light. The NeoPixels were connected to a micro Arduino board using conductive thread.

blog post 2

Figure 13: Elements for building the Adafruit Colour Sensor circuit: a) Adafruit Colour Sensor b) Flora RGB Smart NeoPixels c) Arduino Micro d) External battery. & Figure 14: Diagram of circuit colour Sensor.

To sense the colour of the space close to the person, an RGB sensor was used. This was also connected to a micro Arduino board to light up each of the NeoPixels, which would mimic the colour that the sensor detected (Figure 13). An Adafruit Colour Sensor was used. The ultimate intention would be to produce a wearable version with a Flora RGB sensor, for easy implementation in the mask. For the prototype, however, the sensor was still connected to an Arduino UNO card (Figure 14).

blog post 3

Figure 15: RGB sensors with Flora RGB Smart NeoPixel

The experiment allowed to verify that the actuator was correctly connected to the sensor to mimic the colour that surrounds the wearer (Figure 15).

One limitation discovered in this experiment was that in order to obtain an accurate reading, it was necessary to direct the sensor towards the object for a long time. This renders the interaction less fluid, and less realistic.

Another drawback is that in order to see the colour change, it is necessary to be in a darkened space, since the NeoPixels do not shine strongly enough to be seen in full daylight, restricting the use of the mask to only specific times of the day.

4.3 Colour change using liquids by sensing the body

For this, the sensing of the body was oriented toward touch and muscle movements, given that they were the most easily apprehended by the technology, to induce changes in the mask. In order for the mask to have more mobility, a requirement was that the colour change be possible whilst deformation occurred. For this, food paint was employed (see section 4.1.4 above).

4.3.1 Actuated by muscle sensors

The MyoWare muscle sensor allows reading of muscle activity in different parts of the body, since biomedical sensor pads can be attached directly to the wearable board itself. This sensor, traditionally used for medical research, measures muscle activity by detecting its electrical potential, referred to as electromyography (EMG).

Video 6: Aposema: Actuation by Muscle Sensors | Aposema. Available:

This sensor was placed on the face in order to read the individual’s expressions. In this example, an air pump was attached to the sensor through an Arduino board, and in this way the wearer could manipulate the inflation of a soft robotic prototype (Video 6).

The experiment was successful because the air pump responded accurately to the movements of the muscles. It reacted immediately and precisely to the motions of breathing. However, the simple version of the motion sensor can only read one muscle at a time. This is a limitation because it would necessitate that the variety of human expressions, which are responsible for the actuation of the air pump, depended only on the movement of a single muscle, which is, evidently, a gross oversimplification of the complexity of human expression.

In order to solve the above problem a MyoWare cable shield was attached, which allows three biomedical sensor pads to be added directly to the Arduino board. Given that the eyes and the mouth are the most expressive parts of the face, the muscle structure of a human face was analysed (Figure 16) and two of the sensors placed on the forehead, where they could detect the movement of both eyebrows, and the other sensor was placed on one of the cheek bones, which move as the mouth moves in, for instance, a smile. In this way the range of muscle movements was expanded, in order to realise a more complex interaction.

muscles myo sensor

Figure 16: Face muscles connected to Myo Sensor.

The most recent version of the sensor which was used to detect muscle movement is specifically designed to be implemented as wearable technology. It does not require an extra battery to power it, and its electronics have been built to a smaller scale, so that it can be easily hidden and incorporated inside a flexible material, such as silicone.

Experimentation with sensors was very useful in terms of design because it allowed an exploration of how muscles in the face move, and how they generate different expressions. The main concept of the design discussed in this report is the reassignment of identity; an exploration of facial muscles and how expressions are created can feed into the discussion of how people might interact in the near future, where possible technological advances will mean that individual’s faces can display augmented, more nuanced, or even deliberately restricted, expressions, or react in different ways to hitherto imperceptible environmental factors.

Video 7: Aposema: Reading Facial Expressions with Muscle Sensor | Aposema. Available:

Bearing this central concept in mind, another experiment was conducted in which two muscles of the face were connected to sensors and these in turn to a water pump through an Arduino board (Figure 17 & Figure 18), and by combining a series of different facial expressions, the mask was inflated and coloured simultaneously. The more extreme the expression, the more liquid was pumped inside the air pockets, and the more intense was the colour displayed.

blog post 5

Figure 17: Elements for building the Myo sensor circuit: a) Power supplier b) sensor pads c) Micro water pump d) Myo Muscle Sensor. e) MyoWare cable shield f)Mosfet. g) Resistor. h) Arduino & Figure 18: Diagram of water pump circuit with two colours pumped.

This experiment was very successful; as changes in facial expression occurred, different colours were pumped through the air channels, and the actuator responded correctly to the expressions (Figure 19). One of the main concerns, however, is that technology has not yet reached a point at which all the electronics and mechanisms are designed at a sufficiently small scale to make these applications practicable. This means that they cannot be embedded in wearable technology. They could, however, be placed in a backpack that the wearer could carry with him or her. These experiments nevertheless allow the Aposema group to understand these limitations and to work in a speculative way.

blog post 4

Figure 19: Colours of the mask reflecting face expressions.


Video 8: Aposema: Final Mask | Aposema. Available:


A major achievement of the project is to have built a mask that has the ability to react to atmospheric factors, and the wearer’s body, by generating visual signals, such as changes in shape and colouration, which will allow the user to assume a new identity. The experiments carried out show the feasibility of this, and allowed an assessment of the most appropriate techniques of building such a mask.

It was established, through experimentation, that the technique of colouration by means of liquids (for example food paint), air channels, and air pockets, was the most appropriate technique to apply in this case. This is primarily because this method allows greater flexibility, since the colours are able to change dynamically. It also permits the strength of colour to be altered, depending on the amount of liquid that is pumped inside the air pockets. In addition the mechanisms of colouring and inflation can be activated simultaneously, which, again, makes the mask more dynamic and expressive. A limitation that this method has is that it is difficult to embody the required elements (a pump and containers for the liquids) in a wearable.

During the experimentation process emerged the idea of detecting signals that cannot be perceived by humans and then “display” them. This is the case of UV lights that could be visualized through fluorescent paints.

Sensors and actuators could be identified for reading the colours that surround the wearer, in order for the mask to mimic them. These elements can be easily integrated into the mask.

One limitation of the aforementioned sensors, however, is that, in order to obtain an accurate reading, they must be directed towards the relevant object for some considerable time, rendering the interaction less fluid, and less realistic.

A further issue that arises is that, in order to see the colour change clearly, it is necessary to be in a darkened area, since the actuator do not shine strongly enough to be seen in daylight. This evidently narrows down the functionality of the mask to specific times of day or to specific spaces.

It was possible to construct the mask so that it conveyed changes in the muscles of the face in its colouration, responding to the intensity of the muscular movement. Again, however, one significant limitation lies in the fact that the elements involved in this have not yet been designed on a sufficiently small enough scale to allow them to be fully integrated into wearable technology.

One important achievement of the project was the analysis of different methods of colourising silicone, their respective advantages and disadvantages, and the feasibility of using them in the construction of wearables. Although this analysis was carried out for this project with the specific aim of colourising the mask, the results obtained may be of use for other applications, allowing researchers in other fields to choose the most appropriate technique for their particular cases.


monda 2

Figure 20: Concept Diagram


Biomimetics has been a useful concept for application in a number of different areas, for example in terms of colouring soft robots; it has also proved a valuable source of inspiration for this report, and the experiments detailed within it.

Although projects related to the colouring of soft robots have been undertaken, no proposals have been made to date for the colouring of silicone wearable technology, such as the mask proposed by the Aposema project. There may be inherent challenges in such a proposal, including the need for the mechanisms of colouring and change of form to work simultaneously, the potential requirement to react to signals invisible to the human eye (such as UV light), and the necessity for the colouration to change dynamically and responsively, according to the changing conditions of the environment, or the body of the wearer. All of these challenges were addressed in this project, with encouraging results.

In addition to this achievement, an analysis of the different methods of colouring silicone, their respective advantages and disadvantages, and their applicability in the case of wearable technology was obtained as an additional outcome. This would allow the most appropriate colouring technique to be selected for any given project, depending on the characteristics and needs of the wearable technology under development; such needs could include portability of the components, the necessity of producing colouration that changes over time, the desire for a more complex colouration, the requirement to operate in environments with high luminosity or the need to react to signals invisible to the human eye (such as UV light).

Although they were analysed in the section related to biomimetics applications in soft robots, techniques that use complex chemical processes were not taken into account in the process of experimentation and implementation. A good field for future research is to study more deeply these methods, which have the advantage of not requiring connection to any mechanical device.

Some components required for colourising silicone wearables have some limitations, which are mainly due to the fact that they were not originally developed for being used in wearables. The search for solutions to this problem is an area of research of great interest.

Colourised silicone wearable technology, like the mask developed in this project, has many possible applications, additional to those proposed in the Aposema project. It may be used in the detection of contaminated environments, or environments with a high level of UV radiation, or, alternatively, it may be of interest to those developing and creating multimedia public performances, among others.

Video 9: Aposema: Final Mask | Aposema. Available:


Aposema Design Iterations, Adi Meyer, Sirou Peng, Silvia

Figure 21: Aposema Design Iterations



Figure 1: A camouflaged cuttlefish. – 7/07/2017

Figure 2: Close-up of the skin of a chameleon. TEM image of nanocrystals changing spacing to change colour in an excited state. Scale bar200 nm Adapted from Teyssier et al 2015.

Figure 3: Chameleon changing colour in an excited state. – 12/04/2017

Figure 4: Stephen A. Morin, Robert F. Shepherd, Sen Wai Kwok, Adam A Stokes, Alex Nemiroski, George M. Whitesides (2012). Camouflage and display for soft machines. Design and operation of a colour layer.

Figure 5: Michael Wehner, Ryan L. Truby, Daniel J. Fitzgerald, Bobak Mosadegh, George M. Whitesides, Jennifer A. Lewis and Robert J. Wood (2016) An integrated design and fabrication strategy for entirely soft, autonomous robots. Fully soft, autonomous robot assembly.

Figure 6: Wehner, Ryan L. Truby, Daniel J. Fitzgerald, Bobak Mosadegh, George M. Whitesides, Jennifer A. Lewis and Robert J. Wood (2016). An integrated design and fabrication strategy for entirely soft, autonomous robots. Inflation of tentacles in the Octobot.

Figure 7: Also shown are (bottom left) the structures of both the neutral and oxidised states of the electrochromic polymer in poly(3-hexylthiophene-2,5-diyl, P3HT), and (bottom right) a schematic of the circuit layout (PS, pressure sensor; ECD, electrochromic device). (Ambanja panther chameleon and young hand images from Illustration of the concept of a chameleon-inspired e-skin.

Figure 8: Kazunori Terada, Atsushi Yamauchi and Akira Ito (2012). Artificial Emotion Expression for a Robot by Dynamic Colour Change. Colour and waveform for 8 emotions.

Figure 9: Rueda, S. Peng, S. Meyer, A. (2017). Skin colour mixing.

Figure 10: Rueda, S. Peng, S. Meyer, A. (2017). Skin colour mask

Figure 11: Rueda, S. Peng, S. Meyer, A. (2017). Fluorescent paint with colour mixed in silicone body and air pocket

Figure 12: Rueda, S. Peng, S. (2017). Change in gradient of colour.

Figure 13: Rueda, S. (2017). Elements for building the Adafruit Colour Sensor circuit

Figure 14:  Rueda, S. (2017). Diagram of circuit colour Sensor.

Figure 15: Rueda, S. (2017). RGB sensors with Flora RGB Smart NeoPixel.

Figure 16: Rueda, S. (2017). Face muscles connected to Myo Sensor.

Figure 17: Rueda, S. (2017). Elements for building the Myo sensor circuit.

Figure 18:  Rueda, S. (2017). Diagram of water pump circuit with two colours pumped.

Figure 19: Rueda, S. Peng, S. Meyer, A. (2017). Colours of the mask reflecting face expressions.

 Figure 20: Rueda, S. Peng, S. Meyer, A. (2017).  Concept Diagram

 Figure 21:  Rueda, S. Peng, S. Meyer, A. (2017).   Aposema Design Iterations.



Video 1: Chameleons Are Amazing | National Geographic. Available: 19/09/2017

Video 2: Introducing the Octobot | Harvard University. Available: 19/09/2017

Video 3: Rueda, S. Peng, S. Meyer, A. (2017). Aposema: Fluorescent paint with UV light | Aposema. Available: 19/09/2017

Video 4: Rueda, S. Peng, S. Meyer, A. (2017). Aposema: Thermochromic Paint | Aposema. Available: 19/09/2017

Video 5: Rueda, S. Peng, S. Meyer, A. (2017). Aposema: Actuation with Liquid | Aposema. Available:

Video 6: Rueda, S. Peng, S. Meyer, A. (2017). Aposema: Actuation by Muscle Sensors | Aposema. Available: 19/09/2017

Video 7: Rueda, S. Peng, S. Meyer, A. (2017). Aposema: Reading Facial Expressions with Muscle Sensor | Aposema. Available: 19/09/2017

Video 8: Rueda, S. Peng, S. Meyer, A. (2017). Aposema: Final Mask | Aposema. Available: 19/09/2017

Video 9: Rueda, S. Peng, S. Meyer, A. (2017). Aposema: Final Mask | Aposema. Available: 19/09/2017






Ask Nature Team (2016). Skin changes colour. Available at – 12/04/2017

Brooks Michael (2008). Do you speak cuttlefish?. New Scientist.

Buresch Kendra C., Mäthger Lydia M., Allen Justine J., Bennice Chelsea, Smith Neal, Schram Jonathan, Chiao Chuan-Chin, Chubb Charles, Hanlon Roger T. (2011). The use of background matching vs. masquerade for camouflage in cuttlefish Sepia officinalis. Elsevier Vision Research.

Casselman Anne (2008). Chameleons evolved colour change to communicate. National Geographic News, available at – 3/06/2017.

Chou Ho-Hsiu, Nguyen Amanda, Chortos Alex, To John W.F., Lu Chien, Mei Jianguo, Kurosawa Tadanori, Bae Won-Gyu, Tok Jeffrey B.-H. and Bao Zhenan, (2015). A chameleon-inspired stretchable electronic skin with interactive colour changing controlled by tactile sensing. Nature Communications.

Ferraro Venere, Ugur Secil (2011). Designing wearable technologies through a user centered approach. DPPI’11, Milano, IT.

Morin Stephen A., Shepherd Robert F., Wai Sen Kwok,. Stokes Adam A, Nemiroski Alex and Whitesides George M. (2012). Camouflage and Display for Soft Machines. Science 337, No 6096: 828-832.

Nakajima Kae, Minami Tetsuto and Nakauchi Shigeki (2017). Interaction between facial expression and colour. Scientific Reports.

Nunez-Pacheco Claudia, Loke Lian (2014). Crafting the Body-Tool: A Body-Centred Perspective on Wearable Technology. DIS 2014.

Suh Ayoung, Li Ruohan and Liu Lili (2016). The use of wearable technologies and body awareness: a body-tool relationship perspective. Springer International Publishing Switzerland.

Terada Kazunori, Yamauchi Atsushi and Ito Akira (2012). Artificial Emotion Expression for a Robot by Dynamic Colour Change. IEEE International Symposium on Robot and Human Communication.

Teyssier Jérémie, Saenko Suzanne V; Van der Marel Dirk and Milinkovitch Michel C. (2015). Photonic crystals cause active color change in chameleons. Nature Communications, available at 03/07/2017.

Wehner Michael, Truby Ryan L., Fitzgerald Daniel J., Mosadegh Bobak, Whitesides George M., Lewis Jennifer A. and Wood Robert J. (2016). An integrated design and fabrication strategy for entirely soft, autonomous robots. Nature.

Witzany Guenther (2014). Why Biocommunication of Animals. Springer Science+Business. Media Dordrecht.


Carmel Majidi (2013). Soft Robotics: A Perspective–Current Trends and Prospects for the Future. Soft Robotics.

Daniela Rus and Michael T. Tolley (2015). Design, fabrication and control of soft robotics. Nature.

Deepak Trivedi, Christopher D. Rahn, William M. Kier and Ian D. Walker (2008). Soft robotics: Biological inspiration, state of the art, and future research. Applied Bionics and Biomechanics, 5(3), 99-117.

Filip Ilievski, Aaron Mazzeo, Robert F. Shepherd, Xin Chen, and George M. Whitesides (2011). Soft Robotics for Chemists. Angewandte Chemie International Edition 50, no. 8: 1890—1895.

Georges Teyssot (2005). Hybrid Architecture: An Environment for the Prosthetic Body. Sage Journals.

Gregory R. Gossweiler, Cameron L. Brown, Gihan B. Hewage, Eitan Sapiro-Gheiler, William J. Trautman,Garrett W. Welshofer, and Stephen L. Craig (2015). Mechanochemically Active Soft Robots. ACS Applied Materials & Interfaces.

Michael Wihart (2015). The architecture of soft Machines. PhD thesis, Barlett School of Architecture, University College London.

Rolf Pfeifer, Max Lunagarella and Fumiya Lida (2012). The challenges ahead for Bio-Inspired “Soft” Robotics. Communications of the ACM.

Sampath Kumar Karutaa Gnaniar, Rajesh Elara Mohan, Edgar A. Martinez-Garcia and Roberto C. Ambrosio Lazaro (2016). Towards Bio-Inspired Chromatic Behaviours in Surveillance Robots. Robots.
Sangbae Kim,Cecilia Laschi and Barry Trimmer (2013). Soft robotics: a bioinspired evolution in robotics. Trends in Biotechnology.

Qiming Wang, Gregory R. Gossweiler, Stephen L. Craig and Xuanhe Zhao (2014).
Cephalopod-inspired design of electro-mechano-chemically responsive elastomers for on-demand fluorescent patterning. Nature Communications.



Pawlyn Michael (2011). Biomimicry in architecture. RIBA Publishing.



Donna J. Haraway (1991). Simians, Cyborgs, and Women: The reinvention of nature. Free Asociation Books.

Gerald H. Thayer (1909). Concealing –
Colouration in the animal kingdom. The Macmillan Co. New York.

Guenther Witzany (editor) (2014). Biocommunication of Animals. Springer.

Hugh B. Cott (1940). Adaptive Colouration in animals. Methuen, Oxford University Press.

Philip Ball (2016). Patterns in Nature: Why the natural world looks the way it does. The University of Chicago Press Books.

William Burroughs (1994). The Soft Machine. Grove Press.



Artemel, A. (2013). Retrospective: The Incredible Inflatable Architecture Of The 1960s. Retrieved from: retrospective-the-incredible-inflatable-architecture-of-the-1960s/.12/04/2017 12/04/2017 12/04/2017 12/04/2017



Bahramzadeh, Y. and Shahinpoor, M. (2013). A Review of Ionic Polymeric Soft Actuators and Sensors. Soft Robotics, 1(P), 38-52.

Herzog, T. (1977). Pneumatic structures: a handbook for the architect and engineer. Crosby Lockwood Staples.

Khampanya, R. (2014). SOFTBOT: The implementation of evolutionary algorithm in silicone gel robots. Soft Pnaumatic Pavillon, Interactive Architecture Lab, University College London.

Koyac, M. (2013). The Bioinspiration Design Paradigm: A Perspective for Soft Robotics. Soft Robotics, 1(P), 28-37.

Majidi, C. (2013). Soft robotics: A perspective — Current trends and prospects for the future. Soft Robotics, 1(P), 5-11.

Negroponte, N. (1975). Soft architecture machines. Cambridges, MA: MIT press.

Pask, G. (1969). The architectural relevance of cybernetics. Architectural Design, 7(6), 494-496.

Sterk, T. (2003). Using actuated tensegrity structure to produce a responsive architecture. ACADIA.

Stokes, A. A., Shepherd, R. F., Morin, S. A., Llievski, F. and Whitesides, G. M. (2013). A Hybrid Combining Hard and Soft Robots. Soft Robotics, 1(P), 70-74.

Trimmer, B. (2013). A Journal of Soft Robotics: Why Now?. Soft Robotics, 1(P), 1-4.

Submit a Comment