From by-product to design material: working silk floss waste into a new paper

The work began as a degree thesis.

Since then it has travelled the whole length of via Ignazio Persico: the paper was born at Spazio Chirale and then reached the machines of Punto and Tratto Chirale.

Yuang Yu defended her master’s thesis in Product and Service Design at Sapienza University of Rome, Designing with Silk Waste — A Sustainable Packaging Exploration. The research investigated the possibility of turning a waste stream of the silk supply chain into a packaging material.

The waste in question is spelaia, the silk floss waste: the short, soft, irregular fibre left at the margins of silk processing.

The floss waste used in the trials is supplied by Nido di Seta, an agricultural cooperative in San Floro, in the province of Catanzaro. The cooperative was founded in 2014, when Miriam Pugliese, Domenico Vivino and Giovanna Bagnato took over an abandoned mulberry grove and the running of the Silk Museum from the municipality. Today it follows the entire supply chain, from growing the mulberry to reeling and weaving, with organic certification and natural dyes.

The thesis came to an end. The research did not.

Between a thesis defended and a material a company can put into production there is a distance. This article is about how we are covering it.

An internship as a thesis laboratory

For some years now Sapienza has entrusted me with the Waste-Materials Manufacturing module, within the second-year Sustainability Design Studio of the master’s in Product and Service Design. The module is built on direct experimentation: students produce a biomaterial starting from the processes developed in our labs, test some of its properties and study its characteristics.

Yuang Yu came out of those classes.

At the end of the course we select a few students interested in taking the work further through an internship at Chirale.

Yuang Yu therefore worked on industrial-grade machines in a production environment, alongside the technicians who use those same machines every day for commissioned work. In that context research is measured against the criteria of production from the very beginning: manufacturability, costs, available technologies, repeatability of the result.

It is the same principle as our Test Before Invest, brought forward to a stage in which the project does not yet belong to any company.

The material comes before the process

When we started, the problem was not working out how to process the paper.

The paper did not exist yet.

The formulation work took place between Spazio and Tratto Chirale. At Spazio Chirale we study sustainable materials and develop new formulations, starting from natural matrices and from the by-products of other supply chains.

In the first phase the research explored several families of materials, held together by alginate: a film, a textile-like material and a first paper. Ten different formulations in all.

Today we work on the paper, formed with the traditional mould-and-deckle technique from cellulose fibre and silk floss waste.

The goal was not to obtain something that looked like paper, but to make a short, irregular fibre cohesive enough to be handled and processed, developing a procedure capable of delivering repeatable results.

Not a sheet identical to the previous one: the floss waste always changes a little, and every sheet comes out different from the others. What we make repeatable is not a standardised product, but a process capable of delivering similar — not identical — results.

The result is the paper we use today: a thickness between 0.6 and 0.8 mm, an untreated surface, fibres in plain view.

At that point the problem changed. An interesting material and a workable material are two different things, and only the second one leads to a product.

Punto Chirale: printing on a surface that is not industrial

The direct UV printing trials took place at Punto Chirale, the lab devoted to industrial-grade fabrication technologies and the source of our commissioned work. The machine is a large-format flatbed plotter.

Our paper does not have the regularity of an industrial substrate: it has fibres coming to the surface, local variations in density and differences in thickness that can be measured across a single sheet. These are exactly the characteristics that give it its identity and, at the same time, make printing on it far from straightforward.

During the university research an initial test had shown that a surface treatment made the sheet more uniform and improved the print quality. At this stage we chose the opposite path: adapt the digital process to the material and leave the surface as it is. Making it uniform would have removed precisely the characteristics we are interested in.

The problem showed up in the very first trials. The same file that produced crisp detail on an industrial substrate lost definition on the thinnest marks on our paper.

It was not the file’s fault. Local variation in thickness changes the distance between the surface and the print heads: fractions of a millimetre are enough for the droplet to reach the substrate out of focus, with a wider footprint. Thin strokes lose resolution and the narrowest reversed-out areas close up.

The solution did not come from a single setting, but from controlling four variables at once: flatness of the sheet, positioning on the vacuum bed, printing distance and image preparation. Several test sessions were needed to define a sufficiently stable process window.

Colour is decided before the RIP

Colour management also called for a dedicated round of trials.

The Raster Image Processor we use is Roland VersaWorks, the same environment in which we control printing processes on unconventional materials in our labs. This is not new ground: in an earlier Test Before Invest devoted to the restoration of artistic stained glass with Studio Silice we had built an entire colour process in VersaWorks for printing on glass, assessing the result both in transmitted and in reflected light.

With silk floss waste the problem is a different one, and it is solved before the RIP, while preparing the files. We worked on the colour separations: neutrals were kept on black alone and we limited contamination in the coloured areas.

In the RIP we then worked on preserving the primaries and on managing the profile, preventing the processing from reintroducing unintended colour components.

It sounds like a technicians’ detail. On artwork made of thin marks it is not: a touch of magenta along the edge of a black line, on an ivory-coloured sheet, reads to the naked eye as a halo.

Tratto Chirale: engraving and cutting a heat-sensitive material

The second process is, at least on the face of it, the simpler one.

The engraving and cutting trials took place at Tratto Chirale, on a hundred-watt CO₂ laser system.

The system has far more power than is needed to engrave or cut such a thin sheet: the difficulty therefore lies in precisely metering the energy transferred to a fibrous, non-homogeneous, heat-sensitive material. An overly aggressive setting does not produce a deeper cut: it widens the kerf, blackens the edge, alters the surrounding fibres and wipes out the smallest details.

Nominal power, here, tells you very little. What we characterised is the relationship between the power actually transferred, speed, number of passes, geometry of the drawing and response of the material — with separate process windows for engraving and for cutting, because they are two different problems.

The useful result is not a number. It is that a hundred-watt CO₂ laser can work an extremely delicate material, provided the process is characterised on the real material instead of on the default settings. It is the difference between owning a technology and knowing how to use it, and it is the part of our work that does not show in the finished product.

From characterisation to application

By the spring we will present the samples obtained, together with Nido di Seta, to a number of companies with whom we are assessing possible applications for the material.

We are building a small library of processes — UV-printed surfaces, laser engraving and cutting, different levels of customisation — each with the process window that makes it repeatable in real production. And this is how a waste stream of the silk supply chain gradually stops being a material to be recovered and becomes a material to be designed with.