Today’s consumption of textiles generates a large volume of textile waste. Therefore, it is needed to find solutions to re-use the textile waste rather recycling fibers into new fibers. Research using pre- and post-consumer textiles in composites is ongoing and an interesting direction. This thesis reports the recycling of discharged cellulose containing textiles by production of all-cellulose composites (ACCs). ACCs are composites consisting entirely of cellulose. ACCs from discarded denim and polycotton hospital sheets (PCO) were successfully produced with partial dissolution and two-step method. Discarded denim fabrics with 100% cotton (CO) and blend material (BCO) was sorted and shredded into fibers, then made into nonwovens by needle punching. The produced nonwoven laminates and intact hospital sheets were used as the reinforcement in the composites while dissolved cellulose in an ionic liquid (IL), 1-butyl-3-methyl imidazolium acetate ([BMIM][Ac]), was used as the matrix phase. The matrix is then regenerated by removal of the [BMIM][Ac] by washing to form the composite. The washed-out [BMIM][Ac] was collected and recycled in order to study the effect of its reusing as recycled cellulose solvent on mechanical properties of ACCs. Combinations of the different methods, materials and solvents were studied. The mechanical properties – tensile, flexural and impact properties and the void content of the produced composites were deter-mined. Microscopic analysis was done to study the cross-section of the composites. It could be concluded that the achieved values for the mechanical properties can be compared with ACCs found in literature. The results from the research implies that it is possible to find a new purpose for recycled textiles in the form of composites. With this approach, it is possible to avoid unnecessary disposal of textiles containing cellulose.

With the growing demand for sustainable options, the existing sorting capacities are limiting the potential for fiber-to-fiber recycling. With the help of near-infrared spectroscopy (NIRS), automated sorting of textiles with high accuracy is possible due to the easy access for polymer identification. Despite the effectiveness of NIRS, some limitations of the process still limit its full potential. Possible disruptors may interfere with and disturb the identification of polymer identities and compositions in different ways. In the following thesis, additives, treatments, and other environmental factors that may hinder fiber identification are further acknowledged. The key results of the thesis state that stains and factors due to wear and tear are the most common possible disruptors that could be identified from pre-sorted post-consumer end-of-life textiles. Further on, stains of ketchup, deodorant, and oil affect the polymer recognition by lowering the recognized fiber content. Water-repellent coatings on 100 % polyamide woven fabric were not detected correctly according to the NIR scanner, as the stated polymer composition was >90 %. Even though some investigated factors, e.g., material structures, were correctly identified by the NIR scanner, the internal deviation of the knitted polyester structure indicates that porous and loose structures hold the ability to interfere with the detection of polymers. To what extent the operating software has been developed is highly relevant to the outcome of how accurate textile sorting may be.