• Authors
  • Dipartimento di Ingegneria Chimica Materiali Ambiente Sapienza, Università di Roma Via Eudossiana, Rome, Italy
  • Dipartimento di Ingegneria Chimica Materiali Ambiente Sapienza, Università di Roma Via Eudossiana, Rome, Italy
FREE ARTICLE

Recycling Technologies

Glossary Ceramic glass Transparent ceramic products with an appearance similar to that of glass. They are characterized by an amorphous phase and one or more crystalline phases. - Classification Set of mechanical actions carried out in dry or wet conditions, designed to perform a "classification" of particles systems according to their morphometrical (e.g., size-shape) attributes. - Comminution Set of mechanical actions carried out to reduce waste materials in particles of suitable size and shape to be properly handled and processed in order to liberate/remove contaminants. - Cullet Particulate solid product resulting from collection-comminution of waste glass. - De-inking Mechanical process that removes "ink particles" and "stickies" from waste paper. - Ferrous metal Magnetic metals mainly composed of iron. - Flotation Mechanical process that selectively separates hydrophobic from hydrophilic materials. Hydrophobic materials are forced to adhere to bubbles and float. - Fluff Fine fractions resulting from automotive shredder residue (ASR). Fluff is constituted by materials characterized by intrinsic low specific gravity (e.g., plastics, rubber, synthetic foams, textiles, etc.). - Municipal solid waste (MSW) All non-hazardous waste resulting from the collection of household, commercial, and institutional waste materials. - Non-ferrous metal Metals that contain no iron (e.g., aluminum, copper, brass, bronze, etc.) - Separation Set of mechanical actions carried out in dry or wet conditions, designed to perform a "separation" of particles systems according to their physical attributes (e.g., density, surface properties, electrostatic properties, magnetic properties, color, etc.) - Sorting Waste particle separation, usually carried out with optical-electronic recognition devices and logics.

Definition of the Subject and Its Importance

Recycling technologies can be defined as the whole of procedures designed to set up physical-chemical actions, at an industrial scale, that perform the recovery of materials and end-use products resulting from the collection of household or industrial wastes. The materials to be recovered and recycled, obviously, influence both processing technologies and plant layouts. In this section an in-depth analysis of the problems arising when suitable recycling technologies must be designed, implemented, and set up is presented with particular reference to paper, glass, metals, plastics, and textiles (not organics or C onstruction and D emolition (C&D) waste ). Recycling technologies must be approached from a processing perspective, that is, by defining a sequence of steps and actions where the waste flow stream feed, and the different products resulting from the different sequential processing steps, are handled in order to produce one or more outputs of materials to reuse. Obviously, processing strategies and equipment must be selected with both low environmental impact and positive economic perspectives in mind. Dealing with waste often means dealing with complex products, that is, products constituted of one or more materials of interest but also of polluting material. The economic value, per unit of weight, of the materials to recover is usually low: recycling technologies thus must assure high production, while minimizing plant investments and management costs. From this perspective, a full characterization of the input waste streams and complete control of the different phases of the recycling process are a key issue when recycling technologies are selected and applied. In this section, for each of the different materials, methodologies, procedures, and logics are presented to preliminarily identify and quantitatively assess recycling technologies according to the characteristics of the materials to be recovered.

Introduction

The concept of recycling is an intrinsic part of nature. The different forms of life, at the end of their life cycle, decompose in the soil and become "compost," which helps plants grow. Furthermore, the organic materials resulting from decomposition of vegetation are food for bacteria and fungi. Bacteria and fungi are a food for earthworms, and earthworms are food for ants, beetles, and so on. This recycling chain can be framed in the theory of conservation of mass, originally stated by Heraclitus (530-470 BCE) [1]:

πάντα χωρεĩκαὶ οὐδὲν μένει: everything changes and nothing remains still

and after by Nasir ad-Din Tusi (1201-1274) [2]:

a body of matter cannot disappear completely. It only changes its form, condition, composition, color and other properties and turns into a different complex or elementary matter

and finally clearly outlined and formalized by M. Lomonosov (1711-1765) and A. Lavoisier (1743-1794) [3]:

the mass of a closed system (in the sense of a completely isolated system) will remain constant over time.

Recycling can thus be considered as something related to nature, and as consequence, to humans. Since the beginning of time people have needed to find a way to dispose of and/or to recycle waste. Obviously, technology influenced and continues to influence recycling strategies. In early pre-industrial times waste was mainly constituted of combustion residues, wood, bones, bodies, and vegetable waste. A simple recycling approach, mimicking what happens in nature, was to dispose of these wastes in the ground. Wastes thus became compost, helping to improve soil. Nearly 4,000 years ago there was a recovery and reuse system of bronze scrap in Europe. Composting is known to have been a part of life in China (2000 BCE). Ancient rubbish dumps excavated in archeological digs reveal only tiny amounts of ash, broken tools, and pottery. In Knossos (Crete, Greece), traces of landfill sites exist, dating from 3000 BCE, where waste was placed in large pits and covered with earth at various levels. The evolution of humans from nomadic hunter-gatherers to farmers increased waste production. Waste could no longer be left behind, and it soon became a growing problem.

The First World War, the Great Depression, and the Second World War each contributed to recycling. Recycling, in fact, was a necessity for many people to survive or for a nation to support the war effort. Nylon, rubber, and many metals began to be recycled during this period. The practice of recycling continued in many countries after the Second World War came to an end, especially those nations with a high dependence on resources, such as Japan. However, in other countries, with the post-war years' economic boom, consciousness about the importance of applying policies addressed to recycling and recovering materials and products at the end of their life cycle rapidly decreased. In the 1960s and 1970s, the importance of recycling returned with the Environmental Movement (e.g., the first celebration of Earth Day was in 1970), and a constant and continuous growth followed.

The mass production of the Industrial Age is the main cause of the past low level of consciousness about the importance of recycling: when products can be produced, and/or purchased, at low cost, at least if compared with the income, it often seems more economical to just throw away old items and purchase new ones. This "disposable goods" mentality and the corresponding problems of waste disposal have created environmental problems that today all countries face today when adopting recycling technologies.

What are the benefits related to a wider use of recycling technologies?

The first benefit is obviously related to the reduction of the environmental impact of human activities. The possibility of significantly decreasing the industrial use of non-renewable resources, such as primary raw materials and fossil fuels, by utilizing products resulting from recycling represents an important step forward toward environmental protection and energy savings; both aspects strongly contribute to a reduced exploitation of natural resources. The second benefit is the reduction of the environmental impact related to waste dumping . Less waste disposed of means less natural sites to select and manage for waste storage and less risk to the environment in terms of soil contamination and surface water and groundwater pollution. The third benefit is related to better design and manufacturing of products, from the simplest (e.g., paper, glass, metals, or plastic containers) to more complex ones (e.g., household appliances, cars, etc.) from the perspective of recycling (e.g., ease in dismantling) at the end of their life cycle.

What problems are to be faced in increasing the use of recycling technologies both quantitatively and qualitatively?

The lack of the public acceptance toward recycling and the subsequent low growth of the related W aste-D erived-P roduct s (WDPs) market are the main factors negatively affecting recycling in quantitative terms [4]. Such problems have progressively decreased thanks to (1) the definition of a clear traceability route of the WDPs, (2) the technical-economical improvement of recycling products, and (3) new legislation at the national and/or international level stimulating recycling and utilization of recycled products. In qualitative terms, an obstacle to wider "up-to-date" utilization of recycling technologies is related to concerns about the application of innovative technologies inside existing recycling layouts. Today, a great amount of equipment (e.g., comminution, classification, and separation units), and related operative technologies, is easily available on the market. Sometimes, lack of knowledge of specific control tools, necessary for correct handling and control of the equipment, can affect final recycling plant layouts and overall quality of the plant itself in terms of capacity to adapt to new possible future market requirements. From this perspective, the utilization of equipment and inspection tools for waste products quality control are fundamental for a modern, efficient, and profitable recycling technologies implementation. Both equipment and control device systems must be selected and fully integrated according to: (1) waste feed attributes, (2) possible feed variation, and (3) required final products characteristics. These three aspects must always be taken into account in the definition and analysis of the recycling technologies utilized for all the materials described in this section.

For paper, glass, metals, plastic, and textiles, the above-mentioned aspects play a different role and assume a different importance. In paper recycling feed characteristics are relatively easy to control. Final product characteristics, that is, the quality of the recovered fibers, is preeminent. In glass recycling, the main problem is feed quality, especially for the recycling of glass collected from M unicipal S olid W aste (MSW) , where the presence of "ceramic glass" can strongly impact the further processing and final quality of recovered glass fragments. With metal recycling, the aim is to maximize the "correct" recovery of non-ferrous metal alloys. The achievement of this goal strongly influences the development of innovative sorting/detection logic in order to assure the requested final products' characteristics. In plastic recycling, both feed and final products influence the selection of the separation devices and strategies, as well as the sensing technologies required for quality assessment during the different processing stages. Finally, dealing with fiber recycling requires maximizing source material identification and its preliminary separation. Process and technologies are quite different for textiles and carpets; carpets are much more difficult to recycle than textiles. Here, we have briefly outlined the different problems that must be taken into account in the recycling and recovery of the most common waste materials. In the following, such procedures will be analyzed and the related processing/control devices and actions illustrated.

Recycling and Materials to Recycle

Recycling Technologies : Paper

Paper is usually made from raw material wood pulp and fiber. Vegetable fibers are mixed and "cooked" until the fibers are sufficiently softened, chemicals (e.g., lye) are added to enhance and accelerate softening. The pulp is then "screened" over a screening media. Water is dropped off and/or evaporated. The material is then pressed for further water removal in order to obtain the "paper sheet." The quality and arrangement of the fibers affects the overall quality of the final manufactured paper material [5]. With this in mind, recycling technologies applied to waste paper are primarily aimed at maximizing recovery of the fibers.

Paper production dates back to the ancient Egyptians (e.g., papyrus paper). Around 200 BCE Cai Lun, a Chinese court official, made paper from tree bark and old fish netting. Its production was considered as a remarkable secret and only 500 years later were the Japanese able to acquire the secret. Papermaking spread to the West when some Chinese paper makers were captured by Arabs after the defeat of the Tang troops in the Battle of Talas River (751 AD). The first European paper mill was built at Jativa (Valencia, Spain) around 1150. From that time to the fifteenth century, paper mills were located mainly in Italy, France, Germany, and England; by the end of sixteenth century they were located all over Europe. In 1719, Rene de Reaumur, a French scientist, observed wasps chewing slivers of wood and building their nest starting from such a fiber paste. The use of wood fibers for papermaking started from this observation.

One of the key points in papermaking is an appropriate handling of connected fibers. They can come from a number of sources including cloth rags and cellulose fibers from plants and trees. The use of cloth in the papermaking has always produced high-quality paper. The presence of cotton and linen fibers in the mix creates papers for special uses. From this perspective, cotton and linen rags can be profitably re-utilized for fine-grade papermaking (e.g., bank notes, certificates, letterhead, resume paper, etc.). Rags are usually cuttings and wastes from textile and garment mills. Also, the paper itself can be profitably recycled. The constituting fibers can be reused five to seven times before they become too short and brittle. Many paper-based-products can be manufactured from recycled waste paper: fruit trays, corrugated cardboard, egg cartons, ceiling tiles, plasterboard, sound insulation panels, etc. Waste paper collection and recycling produces a number of positive environmental effects:
  • Less timber is used for wood pulp production, which has a positive impact on biodiversity, that is preservation of valuable wildlife habitats and ecosystems, such as old-growth forests that are not replaced by managed plantations, very often constituted by allochthonous species, usually fast-growing conifers
  • Waste disposal reduction
  • Energy and water savings, as there is no need for pulping to turn wood into paper. Such a savings depends on paper grade, processing, mill operation, and proximity to a waste paper source and markets
  • Considerable reduction of emissions into the air and water (no bleached is usually required in recycled paper)
  • Lower greenhouse gas production; the larger the amount of waste paper re-used, the lower the emissions will be
The main problems faced in waste paper recycling are as follows:
  • Collection criteria addressed to simplify waste paper handling and further processing
  • Identification of polluting elements and suitable processing strategies in place to remove them
  • Quantity and the quality of pollutants (e.g., effluents) discharged to water

Waste Paper Characteristics

Waste paper mainly originates from pre-sorting and collection of consumer waste and/or industrial waste. The following attention will be primarily addresses consumer waste, which is mainly constituted by:
  • Newspapers, magazines, telephone directories, and pamphlets
  • Cardboard
  • Mixed or colored paper
  • White office paper
  • Computer printout paper

Waste paper is usually subjected to sorting according to its origin and characteristics. Such characteristics have been quantified at in Europe with the definition of the European List of Standard Grades of Recovered Paper and Board [6]. Waste graded papers are then pressed and handled as bales. Bales can be thus assumed as the secondary raw materials fed to waste paper recycling plants.

Different waste paper processing strategies will be thus adopted according to the presence of contaminants, to collected paper grade, and to their final re-use. Waste papers contaminants are usually constituted by:
  • Materials and/or products not directly utilized in paper manufacturing, such as metals (e.g., nuts, screws, foil, cans), plastics (e.g., films, bags, envelopes), cloth, yard waste, leather, and dirt;
  • Materials and/or products directly utilized in paper manufacturing, such as:
    • Inks and toners.
    • "Stickies" (e.g., adhesives, coatings, pitch, resins, etc.): these tend to deposit inside paper manufacturing equipments (e.g., wires, press felts, dryer fabrics, calendar rolls, etc.) causing problems, mainly machine down-time. Furthermore, they are difficult to remove due to their neutral density and resulting particles flow characteristics.
    • Coatings: these are usually constituted by inorganic fillers (e.g., CaCO3, TiO2, clay, etc.) and polymeric binders. Fillers have to be removed from the pulp and lower the overall yield of the recycling process. The presence of wax-treated papers (e.g., cardboard) negatively affects recycled paper quality in terms of weak and slippery properties. Furthermore, wax tends to deposit on equipment.
    • Fillers (e.g., CaCO3, TiO2, clay, etc.): their removal is compulsory when specific paper product, as tissue paper, have to be produced.
    • Papermaking additives (e.g., starch, gums, dyes, etc.): among the most difficult to handle are dyes. Their incorrect removal can affect recycled paper new coloring. In some cases, wet strength additives can prejudice the further waste paper re-use.

As a result, waste paper characterized by high quality grades (e.g., paper-mill production scrap and office waste) requires simpler processing and can be profitably applied as a primary paper pulp substitute in applications such as paper printing and tissues. Waste paper of intermediate grades (e.g., newspapers) must be subjected to a stronger processing, mainly for de-inking, and can be used again by the newspaper industry. Finally, waste paper of lower grade is utilized for packaging and board.

Waste Paper Recycling Technologies

In waste paper recycling the selection and the sequence of the processing units is strongly influenced, as previously outlined, by the characteristics of the waste paper (e.g., grade) and the presence and typologies of contaminants [7]. The latter influence the process, not only by their composition, but also by their chemical-physical attributes (e.g., size, shape, density, surface properties, solubility, strength, etc.). In the following, the main units, related actions, and/or potential problems are reported and described with reference to waste paper processing:
  • Re-pulping, "to pulp" waste paper
  • Screening, for fine contaminant removal
  • Cleaning, for contaminant removal
  • De-inking
  • Water and solid waste treatment
Re-pulping
The re-pulping operation is the first and one of the most important processing stages in paper recycling. Correct re-pulping is a pre-requirement for efficient downstream operations (e.g., cleaning, screening, flotation, etc.). Incorrect re-pulping can damage fibers, preventing their "correct" re-use [8]. A re-pulper is thus a device that converts recovered paper into a slurry of well-separated fibers and other waste paper components by performing mechanical, chemical, and thermal actions [9]. In order to fulfill this goal, a re-pulper has to satisfy specific conditions, that is:
  • Contaminant detachment from fibers, without performing comminution (the larger the contaminant, the easier its removal)
  • Correct mixing between waste paper, H2O, and chemicals to liberate fibers, limiting at the same time their cutting
  • Contribution to large debris removal

Re-pulping can be carried out in batch or continuous conditions. In batch conditions waste paper, H2O, and chemicals are all charged at the beginning of the process and are removed, all at once, at the end of the process, then the process starts again. In continuous conditions waste paper, H2O, and chemicals are continuously added to the pulper, as the pulped product is continuously removed.

Screening
Screening is usually performed by forcing the pulp to pass through a sieve. The sieving surface is characterized by holes and slots of different sizes and shapes. The main goal of this phase is removing contaminants (e.g., bits of plastic, glue globs, etc.) and at the same time to realize a first separation of "short" fibers. Screening performances are influenced by many variables, the most important being:
  • Feed pulp characteristics: fiber size and shape, quantity and quality of debris
  • Screening device characteristics and operative conditions: screen surface (e.g., flat or cylindrical), screen hole size and shape, screen surface cleaning mechanism, fed pulp flow rate, solid-water ratio, stock temperature, etc.
Cleaning
Cleaning is mainly applied to remove heavy contaminants. Separation is usually based on centrifugal forces, frequently using hydrocyclones. These devices are constituted by a cone-shaped (e.g., tapered) cylinder. Pulp is fed under pressure to the device, the rotational movement produces, inside the hydrocyclone, two vortexes create the separation: heavier particles are thus recovered in the bottom part and the lighter ones in the upper part. During this stage, metals, inks, sand, and dirt, are usually removed. The cleaning stage is influenced by many variables, including:
  • Fiber and contaminant characteristics (e.g., size, shape, density, and quantity);
  • Selected device architecture and setting: cylinder/cone size, inlet and/or outlet geometrical characteristics, vortex finder diameter and length, cylindrical section height, cone angle, feeding pressure, pulp dilution, etc.
Deinking

Pulp deinking removes printing ink and "stickies" (sticky materials like glue residue and adhesives). Deinking is usually performed in two steps: (1) washing and (2) flotation. Small particles of ink are thus preliminarily rinsed from the pulp with water by washing, then large particles and "stickies" are removed with the help of chemicals and air bubbles by flotation.

Froth flotation technology has been developed and used for many decades in the mineral processing industry before the technology was adopted by the pulp and paper industry for the deinking of waste papers at the beginning of the 1960s [10]. A flotation process is based on the surface properties of particulate solids systems when suspended in a fluid. Particles according to their, natural or caused, hydrophobic or hydrophilic characteristic tend to adhere to bubbles and float. During flotation deinking, pulp is thus fed to one, or to a bank of, flotation cells, where air (e.g., bubbles) and chemicals (e.g., surfactants) are also present. The surfactants cause flotation of the ink and sticky materials. Air bubbles carry the ink particles to the top of the cell/s, where the foam is continuously removed, realizing the required pulp deinking.

The most significant differences between deinking flotation and mineral flotation are specifically linked to the particular characteristics of waste paper pulp suspensions [10], that is:
  • The large size class distribution and shape of the particles to float (e.g., ink particles), as well as their surface properties. Ink particles, in fact, can vary from about 1 μm-1 mm, they are generally hydrophobic, except for water-based inks. Large particles are usually flat shaped, and other techniques, as previously outlined, such as screening, with slots down to 0.1 mm and centrifugal cleaning, are also used to remove the various impurities of waste paper pulp suspensions (e.g., pressure sensitive adhesives, hot melt glues, plastic films, etc.)
  • The low density of the particles to be removed from the deinking pulp: polymeric particles with specific gravity close to that of the water. Mineral particles (e.g., fillers, kaolin, and CaCO3, utilized for paper coating), in the size range of about 1 mm, should generally not be removed
  • The presence of flocs or networks (e.g., cellulose fibers typically of 1-3 mm in length and 10-30 μm in diameter according to wood essences originally utilized) that tend to flocculate up to constitute about 1% of the volume, in the separation zone of deinking cells, as the turbulence level is decreased
  • The need to add chemicals to the re-pulped waste papers, both to realize a better release of the ink particles from the fibers, at the same time enhancing the flotation process (e.g., calcium soap and caustic soda or other deinking chemicals to be used under alkaline or neutral conditions) and the various chemicals introduced with the waste papers (e.g., surfactants used in the coating color)

After flotation, if necessary, pulp is further beaten, or "refined," in order to separate, as much as possible, fibers, avoiding fibers bundles. When white recycled paper is required, pulp is bleached with hydrogen peroxide or chlorine dioxide.

Water and Solid Waste Treatment

Production of both virgin and recycled paper gives rise to pollutants that are discharged to water (e.g., effluents). Studies providing comprehensive comparative evaluation of the environmental impact linked to the effluents generated from recycling plants and those from paper mills demonstrated the environmental impact of the former is lower than that of the latter. In any case, environmental problems related to paper waste recycling are, with reference to the other recycling technologies, further described in this chapter, those presenting a higher impact [7]. The different waste paper processing stages, and related utilized technologies, are, in fact, always carried out in wet conditions and with a large quantities of water and chemicals.

Water. Four key parameters have to be fully monitored in the waste water resulting from waste paper processing : total suspended solids (TSS), biological oxygen demand (BOD), chemical oxygen demand (COD), and chlorinated organic compounds (AOX). De-inking is the main cause of TSS and BOD, and sometimes these parameters are comparable with the same produced processing virgin pulp. On the other hand, COD and AOX are always lower in effluents resulting from waste paper processing. Waste water must be properly processed before it can be re-utilized or before release in the environment. The significant decrease, in recent years, of Cu, Cr, Pb, Ni, and Cd in printing inks dramatically contributed to reduce heavy metal presence in waste water, sludge, and final recycled-paper-based-products.

Solid wastes. The sludge resulting from waste paper processing contains a solid fraction ranging between 30% and 50%. It is mainly constituted by short fibers, fillers, and ink from the de-inking process. Their relative proportion depends on waste paper source characteristics and processing strategies applied to obtain a final product of the required characteristics. Usually the wastes are sent to dumps. In recent years, different attempts have been made to further process and/or re-use them: composting [11], and removal of clay [12] and other fillers [7] for re-use or their utilization for energy production [13].

Emissions to air. Direct emissions from the process of making recycled paper itself are minimal and considered to be relatively insignificant, although little research has been done in this field. Gaseous and particulate emissions to air are produced when the thermal utilization of sludge generated by the pulp and papermaking processes is carried out. Combustion presents many advantages [13], including reduction of the disposed solid mass and volume leading to lower disposal costs, destruction or reduction of the organic matter present in the sludge, and energy recovery. Critical points related to the adoption of this solution are:
  • The L ow H eating V alue (LHV) characterizing wet sludge, requiring preliminary dewatering and/or drying treatments to bring solids content above 30-35% in order to enable a self-sustained combustion and
  • The presence of potentially hazardous elements (e.g., sulfur, chlorine, cadmium, and fluorine), that requires a complete gas cleaning

At the end of all the above-described processing stages recycled pulp fiber finally enters the machine for manufacturing recycled paper sheets. Waste-paper-recovered-fibers can be used alone, or blended with virgin ones to achieve better strength, or smoothness, of the final paper product.

Recycling Technologies: Glass

Glass is made of three relatively simple raw materials, silica sand, limestone, and sodium carbonate, which are melted together at high temperatures (about 1,500°C). Additives can be included to modify some properties, such as color, refractive index, durability, etc. [14].

Examples of glass manufactured goods can be found from several thousand years BCE, when such material was used for ornaments. In the Renaissance period, glass use increased. Vessels, bottles, and other glass containers started to be produced and utilized for both decorative and everyday use. At that time glass manufactured goods were expensive to produce. Large-scale production started with the Industrial Revolution and mass production of glass containers began at the onset of the twentieth century. Together with the increase in production and larger use came the problem of handling glass waste. Glass manufacturers produce a large quantity of products of different characteristics that are addressed to different uses. Glass's physical properties, at high temperature, are close to that of a viscous fluid, and as a consequence it can be worked, by craftsmen or on an industrial scale, to obtain final products of practically nearly infinite number of shapes and characteristics. For this reason glass can be found in, according to its composition and use, several products, ranging from those commonly used at home (e.g., bottles, vases, jars, mirrors, etc.), to those utilized in the automotive sectors (e.g., windscreen) and in industry (e.g., fiberglass for the production of Glass Reinforced Plastics (GRP), Glass Reinforced Cement (GRC), special thermal and/or acoustic insulating panels, X-Ray and cathode tubes, etc.). It is thus easy to understand that waste glass production, and its recycling to produce mainly "new" glass containers, assumes great importance.

Glass is one of the materials that is most often recycled. It presents a series of positive characteristics: it is non-absorbent and does not confer flavors and odors; it resists high temperatures, such as those required for cleaning after its use; its strength and mechanical resistance are indispensable for multiple fillings and reuse. These characteristics make glass containers suitable to be used over and over again. Selective waste glass collection and recycling provide great benefits:
  • Reduction of environmental impact related to its disposal
  • Conservation of the non- renewable raw materials (quartz sands) required for its production
  • Energy savings
  • Reduction in the quantity of solid urban waste
The problems to face in glass recycling can be summarized as follows:
  • The definition of collecting criteria able to simplify the further processing
  • The identification of polluting materials and the set up of suitable processing strategies to remove them
  • The separation of "broken glasses" (cullet) according to their color

Recycled glass mainly comes from the selective collection of solid urban waste (bottles, jars, various containers, etc.), usually done by citizens, and only partially from the refuse of glass goods manufacturing and/or glass-based products at the end of their life-cycle. As a consequence, waste glass collection represents one of the most critical steps of the entire recycling process, and the following recycling technologies and separation strategies are strongly conditioned by the criteria and the methods followed during collection. The quality of the collected materials can be quite different, according to the level of knowledge and, more generally, the "education" of the people involved. As a matter of fact, the quality of the glass collected for recycling can strongly differ from region to region or, in the same city, from district to district.

The following discussion is based on urban waste collection as the source of the glass. It is important to consider the final destination of the recycled glass, which can be identified by the classical market categories where glass in commonly utilized, that is: (1) container production, (2) construction industry, (3) special concrete production (e.g., partial substitution of aggregates by glasses), (4) road pavement (e.g., special asphalt where the coarse fraction is partially substituted by glass), (5) abrasive products, (6) wool glass, etc.

The recycling technologies described for glass recycling will be primarily addressed to producing an economically valuable cullet to use to make new containers. Recycled glass is not equally re-utilized in all the above-mentioned market sectors. Only a small fraction is, in fact, re-utilized in fiberglass, bricks, concrete, and asphalt production. This is mainly for two reasons: (1) cullet quality sometimes does not fit well with the quality standards required in some of the these sectors and (2) the glass container industry is the most interested in waste glass reuse (due to the high cost of primary raw materials versus the relatively lost cost of each single glass container).

Cullet characteristics must satisfy strict conditions to be re-utilized for container production. These characteristics are primarily related to both presence of polluting elements and color of the fragments. Furthermore cullet size class distribution is another important parameter to control. Usually particles around 1 or 2 cm are preferred both for handling and quality control purposes.

Contaminant removal and cullet color sorting are the main goals when recycling glass. Furthermore, such goals must be reached using a process that does not produce too fine particles. As a consequence recycling technologies must be designed to fulfill these goals.

Cullet Contaminants Definition

There are two classes of contaminants: materials not constituted by glass (e.g., ceramics, stones, masonry, organics, and heavy metals) and glass fragments of the wrong color, that is cullets whose color characteristics are different from that of the class they belong to (cross-contamination).

Non-glass materials . Ceramics and stones, which have melting points higher than that of the glass, remain un-melted inside the vitrified matter and as a consequence, even if present in a small amounts, degrade the mechanical characteristics (resistance) of the manufactured products (bottles, jars, etc.). Furthermore, they can seriously damage glass processing equipment, increasing maintenance costs. Lead and heavy metals, according to their high volume weight, settling on the bottom of the fusion crucibles and have a corrosive effect on the refractory material, causing, in some conditions, the perforation of the refractory material itself. Optical sorting devices are commonly used to identify and automatically remove non-glass materials. Among polluting materials special attention has been addressed, in recent years, to ceramic glass. This material rapidly increased its presence in waste glass products, mainly due to the introduction on the market of a large amount of ceramic glass manufactured goods, such as dishware, cookware, etc. [15]. Such material, even if seems quite similar to classic glass, is characterized by a different behavior (i.e., higher fusion point) when melted inside glass furnaces, where cullets are usually fed together with natural raw materials (quartz sands) [16]. As a consequence, the presence of ceramic glass reduces the production rates of the furnace, which needs to be shut down to be cleaned more frequently, and sometimes causes damage that requires the furnace to be rebuilt or replaced. Classical optical sorting devices are practically "blind" to ceramic glass, as its physical-chemical characteristics are similar to those of glasses.

Cullet cross-contamination by color. Glass has, according to its color, a different destination of use and, as a consequence, different market value. The use affects the value of the glass containers; as a consequence, white glasses have higher value than the so-called half-white or colored glass (brown, yellow, green). Cullets, that are collected without distinction of color can be primarily used for the production of green glass and only in part for the production of yellow glass. The production of white glass requires that only cullet of that color be employed. Cross-contamination can thus represent a problem because it always contributes to depreciate the cullet's value. For this reason cullet optical sorting by color is extensively utilized.

Cullet Recycling Technologies

On the basis of the previously mentioned washing goals, specific processing layouts have to be defined. Each of them will be constituted by one or more units, in series and/or in parallel, and each one specialized to perform a specific action. In the following the main units, and related actions, are reported and described with reference to waste glass fragments (cullet) processing:
  • Hoppers and conveyor belts, to perform cullet handling
  • Manual sorting of macro-contaminants (e.g., ceramics, metals, plastics, stones, etc.)
  • Crushing and screening, to reduce glass fragments in two or three size fractions, avoiding the production of fines, for a better handling of the glass waste, and to perform, in some cases, also a separation of contaminants
  • Ferrous and non-ferrous metals separation, respectively, by electro-magnets and eddy-current separators, for metals and non-metals not manually sorted
  • Light contaminant (e.g., paper and plastic) removal
  • Cullet fine contaminants and color sorting by magnetic and/or optical devices, to define different cullet quality classes according to market specifications
Hoppers and Conveyor Belts
Storage-feeding and conveying equipment constitutes the skeleton of the plant providing glass waste transportation for the action performed. Conveyors thus assume an important role because they have to operate continuously to assure a constant feed to the different units of the recycling plant. Cullets are strongly abrasive, so conveying, and related mechanical devices, are subject to strong abrasive actions. Furthermore, cullet handling produces fine particles and dusts. These aspects, if not fully controlled (e.g., sealed bearings, enclosed gear boxes and other moving parts protection, suction units for dusts collection) can produce severe damage to conveying units, as well as severe environmental working conditions in the plant. Table 1 are details some of the characteristics that a conveyor belt has to satisfy to be correctly utilized inside a glass recycling plant.
Recycling Technologies. Table 1 Conveyor belt design principles and/or maintenance good practices with reference to waste glasses handling

Cleats

Conveyors without cleats are preferred, because they can represent an obstacle to conveyor surface cleaning by scrapper bar. Their presence is negative if moisture is present.

Bearings

Bearings and other revolving parts (e.g., pulleys) have to be regularly cleaned or, better, should be sealed.

Drivers

Drivers should be selected in order to assure flexibility according to end markets variability.

Surface

Belt constituting materials have to be selected taking into account cullets' abrasive characteristics and related wear effects. Rubber belt are the most commonly utilized but wearing effects are severe, for this reason vibratory conveyors are more and more utilized. They are less subject to wearing but can represent a further source of pollution (e.g., presence of fine metallic parts in the final cullet concentrate).

Moisture and/or organic residues

Moisture can severely affect cullet handling, especially with reference to sorting operations.

Manual Sorting of Macro-Contaminants

Human-based sorting continues to be commonly applied on the coarser fraction. Manual sorting allows removing the larger pieces of contaminants, easily detectable and removable by trained personnel. Today, this phase is often performed by automatic sorting equipment, utilizing different principles, mainly optical, for contaminant detection. Among macro-contaminants, in recent years, ceramic glass has become a problem. As its machine-based identification is difficult, manual sorting continues to be utilized to perform its removal.

Crushing

Waste glass comminution, usually performed by crushing equipment, is carried out to reduce the collected waste glass materials to particles of suitable size and shape in order to be properly handled and processed to remove contaminants and to separate glass into suitable classes of colors to produce new containers or other products. After crushing a sieving stage is usually carried out. Cullets are thus "sorted" according to their size.

Glass, different from aggregates, when subjected to comminution (e.g., crushing) tends to produce, for its chemical-physical characteristics, a lot of fines (pieces smaller than 1-2 mm); such a behavior increases if comminution equipment is not properly selected, that is, if the comminution unit and its operative conditions do not take into account glass mechanical characteristics. A comminution unit usually develops its action through the application of four forces on the materials: impact, shear, abrasion, and compression. The quantitative relationships among them, in terms of cause-effect, strongly differ according to the equipment. For glass comminution, impact forces are those to be primarily applied, allowing a low production of fines. The production of fines causes several problems: (1) loss of product (e.g., recycled glass fragments below 2-3 mm are usually rejected from the market), (2) a double loss of money (e.g., unsold material and rejected fines have to be disposed of), and (3) higher costs for comminution equipment maintenance. Glass fines are extremely abrasive; they can enter the gears and bearings of traditional comminution equipment and cause serious damage. Furthermore, the presence of fines produces the same mechanical problems in the other equipment constituting the processing layout. When selecting a comminution unit it is thus important to evaluate not only the capacity, but also the degree of flexibility characterizing the crusher, that is, the ability to vary power and speed. These two parameters can be modified according to changes in required crushed glass characteristics in respect of the initial feed and possible requirements of the end market. Table 2 describes some of the impact crushing equipment most commonly used to perform waste glass comminution .
Recycling Technologies. Table 2 Characteristics of the comminution units mainly utilized to perform waste glass crushing

Comminution units

Characteristics

Hammer Mills (HM)

Movable hammers mounted on a rotating shaft hit and/or throw glass against mill chamber or other glass. As a result comminution is realized. Cullets are recycled inside the hammer until they do not reach a size lower than the aperture of a grid installed at the exit of the mill chamber. Comminution is efficient, that is, a relatively small size glass grain can be achieved in just one comminution stage. The wear rate of the hammer is high.

Rotating Drums (RD)

A spinning rotor with bars attached to the outside is responsible for the comminution. Glass particles breakage is mainly due to: (1) the impact of the rotating bars on glass and (2) the glass projection against "special plates" installed on the inside the chamber. Crushed material is finally discharged when particles size is lower than the space, adjustable, between the rotating bar and the plates. The wear rate is lower in comparison to hammer mills.

Vertical Shaft Impactor (VSI)

Crushing is performed by a revolving vertical rotor. The material is fed from the top. The mechanical actions resulting for the interaction of the revolving rotor with the glass and with the chamber walls produce the requested glass size reduction. A screening system, in a closed loop with the impactor, allows obtaining the desired cullet size. VSI allows obtaining large productions, but a pre-broken feedstock, usually obtained utilizing as primary crusher an HM, is required. The wear rate is minimized; the resulting cullets are round shaped.

Impact Crusher (IC)

IC utilizes a continuous breaker bar which is mounted horizontally in the rotor. Glass is fed and blown against adjustable aprons. Replaceable liners are installed inside unit. The broken product passes through an open discharge. As for VSI, IC allows to efficiently handle a large volume of throughput of product below 10 mm. Cullets, according to IC comminution actions, are characterized by an angular or sub-angular shape.

Screening and Air Classification

Screening and air classification are usually performed for two purposes: (1) particle size control and (2) contamination removal. The proper use of screening and air classification is to cut cullets into particles belonging to different size class ranges. Such a division can be performed both to facilitate cullets' further processing and to obtain final marketable products characterized by a population of specific and well-defined particle size. Screens and air classifiers can be profitably utilized as "separators" when polluting elements (e.g., wood, paper, iron, steel, aluminum, plastics, etc.) are characterized by a size and/or a shape different from the cullet population to classify. In this case, the setting of a "threshold," for the geometrical attributes, allows separating pollutants from glass. Furthermore, when air classifiers are used, the different densities of the polluting materials, compared to those of glass, represent an important factor in performing an improper "air classifier based separation." In both cases, moisture is a restricting factor for both classification and separation efficiency.

In the selection of screening and/or air classification devices some important factors have to be taken into account: (1) particle size distribution of the feed, (2) particle size distribution after the different classification stages, and (3) market tolerance, that is, precision required with respect to end-user tolerance versus possible presence of under-/over-size in the final classified product. Table 3 describes some of the screening and air classification devices commonly utilized in the waste glass recycling sector.
Recycling Technologies. Table 3 Characteristics of the classification units mainly utilized in the waste glass recycling sector

Classification units

Characteristics

Trommels (TRO)

A TRO is a rotating cylinder whose surface is characterized by apertures of a specific size and shape where cullets can pass through according to their size. For its characteristics TRO can play both as a classifier and a separation unit. Contaminants as plastic bottles or bags are not able to pass through trommel surface apertures and can thus be easily removed. A trommel can be designed to fit practically any size glass processing operation. In some cases hot air is blown to facilitate waste glass feed drying, helping the screening and the following processing stages.

Vibratory Screens (VS)

VS usually work in a closed loop with crushing unit. As for TRO cullets can pass through, or not, the sieving surface, according to their size, usually larger particles are fed again (closed loop) to the comminution unit (crusher) or rejected if recognized as pollutants. According to the requested cullets size class characteristics, vibratory screens can allow to classify particles population up to - 75 μm. When such dimensional grades are requested obviously throughputs are lower. Usually multi-deck VS, up to five, are utilized when the final cullet product has to be divided in several dimensional classes.

Air Classifiers (AC)

These types of classifiers are based on the utilization of air flow to classify particles. Different from sieving, particles are thus classified/separated according to their size, shape, and density. An AC is a vessel where an air flow is generated. Wastes are usually fed from the top, coarse (larger or heavier) and fine (smaller or light) particles follow a different path according to flow characteristics, that can be set to achieve the required classification/separation. Classification/separation can be achieved simply utilizing gravity or/and free or forced vortex generated by static vanes or a dynamic classifier wheel. The equilibrium between the forces determines the "cut point." ACs represent a good alternative to screening, especially when the materials to classify are characterized by the presence of large fraction of fines (particles below 250 μm). When a well classified coarse fraction is requested, air classifiers are utilized prior to screening.

Ferrous and Non-ferrous Metal Separation

Ferrous (e.g., metals containing iron) and non-ferrous metals (e.g., aluminum, copper, lead, zinc, tin) are among the most common contaminants in waste glass. Their origin can be related to the presence of caps, lids, special labels, bottle neck wrap, etc. The presence of these materials even at low level can produce severe damage to the production process (e.g., deposits and chemical reactions in the furnaces, presence of hot spots, clogging or jamming of the injection devices, etc.) and can compromise the final product (container) mechanical characteristics and quality.

Ferrous contaminants are easily removed by magnetic separation. Usually magnetic separation is performed in two stages. A first stage, at the beginning of the recycling process, in order to remove ferrous contaminant before the further processing stages (comminution, classification, sorting by color, etc.) and a later stage, where usually it is applied to remove the finer particles of metals not separated during the previous processing stages. The selection of a magnetic separator is strongly influenced by: (1) waste feed rate, (2) amount of ferrous contaminants, (3) cullet stream depth, (4) contaminants size, and (5) value of the magnetic field to apply to perform separation. Non-ferrous metal contaminants are not affected by the presence of a magnetic field; for this reason eddy current separators are usually utilized for their collection [17] [18]. Eddy current separation devices and related separation strategies are strongly influenced by non-ferrous metal particles size, shape, and conductivity. In this latter case, different from magnetic separation, particles' attributes play a preeminent role. Table 4 details the equipment commonly utilized to perform ferrous and non-ferrous metal separation.
Recycling Technologies. Table 4 Characteristics of the separation units commonly utilized to perform ferrous and non ferrous contaminants separation in the waste glass recycling sector

Separation units

Characteristics

Overhead/Cross Belt Magnets (O/CBM)

The magnet is usually installed above the glass waste flow stream (e.g., conveyor belt). The overhead magnetic field has a belt moving across its surface at approximately a 90° angle to the materials flow. Metals particles are thus attracted, removed from cullets and discharged as the moving belt of the separator turns away from the magnetic field the metals particles. Sometimes, especially when O/CBM strategy is applied to control the possible presence of ferrous particles in the final product, "simple" high intensity magnets are utilized. Collected particles are thus discontinuously removed.

Magnetic Head Pulleys (MHP)

MHP is usually installed at the end of a conveyor belt, beneath the belt. Ferrous particles are thus held to the belt while cullets can be discharged. With the decrease of the magnetic field ferrous particles leave the belt and can be properly recovered.

Magnetic Drums (MD)

MD commonly installed inside feeder chutes, between chutes and conveyors. Their behavior is similar to MHP. Glass waste stream passes over the magnetic drum, ferrous metals are held by the drum, as non-magnetic materials continue their flow. Ferrous materials hold to the drum until a divider provides to its discharge.

Eddy Currents Separators (ECS)

ECS is realized by spinning a magnetic rotor with alternating polarity at high speed. The magnetic drum of the ECS induces electric currents (eddy currents) within the volume of each particle flowing in the proximity of the drum. The effect is that non-ferrous metals passing over the drum are subjected to an ejecting force that throws away non-ferrous metals pollutants from the waste glass stream. The main difference from ordinary magnetic separation is that the magnetization of particles is not induced by the alignment of the internal magnetic domain with the external field. The efficiency of the eddy current separation process is highly dependent on the size of the feed particles.

Light Contaminants Removal

Light contaminants such as plastic, paper, wood, corks, etc. do not constitute, at least in principle, a problem in glass recycling. They burn and volatize at the temperatures of the glass furnaces. Today, new applications requiring granular products are sensitive even to the presence of small amounts of "organics"; for this reason the recycling processes have to take into account the removal of these kinds of contaminants. Such a goal is easily reached by simple screening or by air-screening, that is creating a sort of fluidized bed where heavier particles, cullet, remain on the site and lighter or finer particles can be easily removed by air flow.

Fine Contaminants and Color
Originally cullet optical sorting was performed following an optical-based analogical approach. It was developed and implemented to perform control-separation actions:
  • To remove all contaminants not removed in the previous physical processing stages (e.g., venting, classification, eddy current separation, magnetic separation, etc.)
  • To perform a separation of cullet by color

Normally, ceramics, stones, and opaque particles are sorted before color sorting is applied. In recent years, the increasing presence, inside waste glass, of ceramic glass, which is almost impossible to separate via classical recycling technologies, has resulted in research related to optical sorting to solve this problem with a new class of devices based on X-rays and/or HyperSpectral Imaging (HSI) technology .

Sorting was originally performed manually. Such an approach is labor-intensive. Being based on human senses it is strongly affected by human sorter experience, by the size of the materials to sort, and by time: the level of attention of the worker is strongly affected by time. With technological developments, cullet sorting was, and in some cases continues to be, adopted as detection unit, laser beam technology based devices, and scan line cameras. Table 5 describes the devices, and related architectures, commonly utilized to perform cullets sorting .
Recycling Technologies. Table 5 Characteristics of the detectors, and related logics, commonly utilized to perform cullet sorting

Sorting units

Characteristics

Laser Beam Technology Based (LBTB)

Detection is based on the evaluation of the "characteristics" of the energy and the spectra received by a detector after the cullets, and/or polluting particles, were crossed by a suitable laser beam light. Such an approach presents two technological limits, related to: (1) the constructive characteristics of the equipment and (2) the material characteristics. The sorting logic is mainly analogical. An on-off logic is applied.

Image Analyzers (IA)

IA allow to perform sorting on the basis of the cullets' detected colors. CCD (C harged C oupled D evice) scan line cameras are utilized. They present the advantage, in comparison with LBTB, that practically do not have any detection limitation in terms of geometrical resolution, being the investigated scan line dimension is the only function of the lens. Furthermore, colors are better detected.

Sorting device characteristics. These are related to the optic detector/s size and arrangement. The final sensing architectures, in fact, are influenced by optical information acquisition and its further handling: both factors dramatically influence the sorting architectures, usually based on a pneumatic blast, enabling the modification of the cullet's, and/or polluting particle's, trajectory, after recognition. Furthermore, flow characteristics can influence the selection. To realize an optical sorting (Table 5), the flow has to be, at least in principle, constituted by particles forming a mono-layer (e.g., cullet fed to the color-sorting unit by a vibrating conveyor belt, which keeps the glass in a thin layer). In these conditions glass fragments can be analyzed by the laser beams. Recent equipments are so fast that they can test, according to its dimension, the same cullet several times. As a consequence the larger the glass fragment, the better its control can be carried out. The influence of the "anomalies" can be thus reduced when each cullet is analyzed by more than one detector and more than one time. This is not possible with smaller pieces: they can pass, for their dimension and for the flowing conditions, unsorted, or diffraction/refraction effects (e.g., presence of marked cleavage or surface anomalies) can be so strong that detectors are practically unable to analyze them. As a consequence, such a technique cannot be profitably used with the entire size range of cullets. Materials of smaller dimensions, less than 2 mm, resulting from the processing-cleaning stages, not being correctly investigated, are usually disposed of, with the resulting financial loss and negative environmental impact. For this reason, in recent years, a new class of sorting devices based on imaging has been developed and implemented (Table 5). Imaging allows breaking down the investigation limits, with detection resolution linked to array detector resolution and optics magnification. Obviously the geometrical constraints, linked to the presence of the pneumatic devices, for particle removal remain; however, the imaging approach constitutes a big step forward allowing: (1) a better detection and (2) the potential to perform on-line certification of products.

Material characteristics versus their optical recognition. Cullet attributes, surface status (e.g., dirty or clean) and characteristics (e.g., fragments of bottle neck or jar with or without thread, bottle or jar bottom, etc.) can strongly interfere with the measurements. Measurements are, in fact, based on the evaluation of the transmitted energy received by a detector after the cullets are crossed by an energizing source (e.g., standard or laser beam light). In these measures surface characteristics and the status of each cullet play a decisive role in the further response of the detector (Fig. 1).
MediaObjects/978-1-4419-0851-3_16_Part_Fig1-116_HTML.gif
Recycling Technologies. Figure 1 The spectral response of the cullets, suitably energized, is based on the evaluation of the transmitted energy received by the detectors. The detected energy is thus influenced by the status (dirty or clean) and the characteristics (fragments of bottle neck or vase with or without thread, bottle or vase bottom etc.) of the cullet surface. (a) 16 × 16 mm image field of dirty light green cullet. (b) 16 × 16 mm image field of dark green cullets (bottle bottom). (c) 2 × 2 mm image field of white cullets (threaded bottle neck). (d) 2 × 2 mm image field of half-white cullets (bottle bottom) and (e) 2 × 2 mm image field of brown cullets (bottle bottom)

Ceramic glass recognition . As previously outlined, the frontier in cullet optical sorting is represented by ceramic glass recognition. The only two strategies extensively used and designed to reduce the presence of ceramic glass contaminants are source reduction and manual sorting. The problem with reduction at the source is that usually citizens, in spite of public education campaigns, confuse transparent-glass-like contaminants with normal glass, invalidating both curbside and door-by-door collection. As a consequence, some steps of the sorting process are still carried out manually by "trained personnel" who try to recognize ceramic glass fragments prior to crushing, looking at their shape or evaluating their reflective characteristics. Such an approach is expensive, unreliable, and represents a real and important problem for the whole glass recycling sector. Infrared sensors are sometimes used to detect opaque cullet within the "dirty" recyclable glass in order to separate ceramic/ceramic glass and stone fragments from glass cullet. No entirely effective and low-cost solution has been found to date for ceramic glass on-line automatic sorting in recycling plants. X-ray sorting techniques have been recently proposed as a solution for ceramic glass identification [19]. Anyway, it must be considered that the use of X-ray equipment in a plant requires appropriate shielding and must follow strict rules to protect workers from exposure, with an increase in costs and environmental and safety problems. Glass sorting, originally based on analog devices, utilizing laser beam technology, has moved towards digital image techniques [20, 21]. Scan line color cameras are thus seeing greater use in this sector to implement selection strategies addressed to identify opaque objects inside the flow stream and/or to separate cullets according to their color [22]. Technology in this field, although sophisticated, remains practically "blind" with regard to the identification of ceramic glass materials. Spectrophotometers should be able, at least in principle, to identify these contaminants; however, they are usually only able to work on a point-by-point basis and are not able to cope with real-time sampling/sorting architectures such as those required in glass recycling plants [23]. They are used, in several industrial fields, mainly at the laboratory scale. A new class of sensing devices based on HSI has recently created new potential for the on-line recognition of glass and ceramic glass fragments inside glass recycling plants [24]. A more detailed description of such a technique is reported in the later section Future Directions.

Recycling Technologies: Metals

Human history and progress are linked to the discovery and utilization of metals [25]. Thanks to these materials, humans have been able to interact and modify the environment, performing advances in agriculture, warfare, transport, etc. The Industrial Revolution, from steam to electricity, was conditioned by metals. Even what, and how, people eat was and is strongly influenced by metals.

The first use of metals dates back to about 7000 BCE in Anatolia (Turkey), where some Neolithic communities started to replace handmade stone knives and sickles with "hammering" native copper. The tools worked as well as their stone equivalents and lasted far longer. The first examples of extractive activities belongs to 4000 BCE. Deep shafts were cut into the hillside at Rudna Glava, in the Balkans, to excavate copper ore. Mining was considered a sort of ritual activity; as thank for the exploited metals, fine pots, bearing produce from the daylight world, were placed in the mines as a form of recompense to propitiate the spirits of the dark interior of the Earth [26]. Thousands of years later, humans started again to understand the importance of preserving the environment and, together with more stringent economic reasons, started to apply more metal recycling .

Metals present many advantages; they can easily be recycled because a specific material can melted several times without losing its properties. Metals commonly utilized in the recycling sectors mainly derive: (1) from the collection and processing of post-consumer metal products and (2) from metal industrial wastes (e.g., working residues, metallic scraps, etc.). Recycling strategies are strongly affected by the previously mentioned origins of the metals. Furthermore, metals' value is strongly affected by several costs, that is, the quality of recovered products (e.g., composition, contaminants residues, etc.), (3) recycled product market, and (4) metal market value. Together with these costs, their processing (e.g., collection, transport, sorting, etc.) and waste disposal settlements also play an important role. Metals to recycle can be divided in two different families, that is: ferrous and non-ferrous metals. Such a distinction is important because it dramatically influences recycling strategies and related adopted technologies. Ferrous metals scraps are mainly constituted of iron and steel scraps, primarily obtained from automotive dismantling and household appliances (e.g., large kitchen appliances, washers and dryers, etc.). Such waste is usually collected and preliminarily sorted in different classes of products of different grade before being sent to a recycling plant or directly to metal refiners. Wastes resulting from their processing (e.g., wood, plastics, fibers, etc.) are, according to their quantity and physical characteristics, totally or partially recovered; the unrecovered fraction is sent to a landfill. Another lesser source of ferrous metals results from the processing of the bottom ashes produced by incinerators.

Non-ferrous metals scraps are mainly constituted by aluminum, copper, zinc, and lead. Aluminum is the main scrap deriving from household waste (e.g., cans, containers, etc.); the others primarily result from waste from industrial and commercial activities.

Ferrous Metals

Iron and steel are the main materials utilized in many industrial sectors: building and construction, automotive, chemical, operative equipments, etc. These materials are so common in use for several reasons: (1) relatively low costs, (2) high availability, (3) good mechanical attributes, (4) easiness in working, and (5) because they can also, at least in principle, be easily recycled, the main reason being linked to the ease in recovering them due to their magnetic properties.

Iron and steel are obtained from raw materials (e.g., iron ores) and/or 2) from recycling. Different production methods are thus utilized: B last F urnace (BF) and B asic O xygen F urnace (BOF) , when primary raw materials are utilized, and E lectric A rc F urnace (EAF) when recycled products are employed. Metal scrap recycling allows reduction of both energy-production costs (e.g., less energy is required for produced unit of weight when scraps are re-melted instead of using iron ore) and environmental impact (e.g., reduced exploitation of primary raw materials such as iron ores, limestone, and coal necessary when primary metals are produced by BF or BOF). Furthermore, a corresponding decrease of CO2 emission is achieved, with a further environmental benefit.

Metal recycling is a well-established practice and it will continue to grow with the increased availability of automotive-derived scraps. In the future, however, this source will probably be reduced as more motor vehicles are designed with plastics and/or polymeric-based composites. In a quantitative way, steel represents the most recycled product, more than aluminum, paper, and glass together and greater than all other metals combined (e.g., aluminum, copper, nickel, chromium, zinc).

Metal Scraps' Sources and Characteristics
Metal scraps can be obtained from many sources:
  • Home scrap, that is, the scrap (e.g., working production waste, defective parts, etc.) derived from a manufacturing process. In this case, waste is directly re-melted. There are no problems related to metal scrap quality, as the material is constituted only of the metal to recycle
  • Industrial scrap usually consists of the wastes produced in iron- and/or steel-manufacturing plants, mainly leftover product resulting from specific manufacturing actions. Such wastes are usually sold and re-used in foundries
  • Post-consumer scrap is metal waste derived from products that have reached the end of their life-cycle (e.g., industrial equipment, cars, metals structures, home appliances, etc.). In this class of scrap, contamination, mainly the presence of non-ferrous metals (e.g., aluminum, copper, zinc, and lead), can represent, in some circumstances, an important problem to face and solve

Other iron and steel intermediary products, and/or waste, play an important role as recycled products: steel-making slag and flue dust (resulting from BF, BOF, and EAF), waste sludge and filter cake (resulting from BF and BOF), spent pickle liquor, and mill scale.

One of the most important steps in the development and set-up of metals recycling systems is creating appropriate strategies to identify and sort metals into groups presenting similar characteristics. Such a "grouping" must be carried out according to specific rules defined by steel-makers and market requirements.

Ferrous Metals Recycling Technologies

On the basis of what has been previously outlined, specific processing layouts have to be defined. Each is constituted by one or more units, in series and/or in parallel, and each is specialized to perform a specific action. The main actions, and related units, are reported and described below with reference to ferrous metal wastes processing:
  • Manual sorting (e.g., non-metal miscellaneous material detachment, large non-ferrous metal separation) and preparation (e.g., cutting of large metal manufactured goods, removal and dismantling of specific metal unit, etc.)
  • Crushing and screening, to reduce metal scraps to easy-to-handle pieces to be directly fed into furnaces or subjected to further classification-separation actions
  • Separation of the different metal fractions into groups characterized by similar composition attributes
  • Testing and/or sorting of the different resulting ferrous metals end-life-goods-derived-material, to characterize and certify different classes of products according to market requirements
Manual Preparation and Sorting

Metal manufactured goods to recycle are usually constituted of units of large dimensions (e.g., automobiles, metal structures, etc.); for this reason they must first be reduced to easy-to-handle pieces, both for further processing and/or for direct re-use inside furnaces. For these reasons, shears, hand-held cutting torches, crushers, or shredders are commonly utilized. After this preliminary stage, manual sorting, if required according to ferrous metal waste to recycle, is carried out. At this early stage of the process large contaminants that are easily detectable with human senses (e.g., car batteries, plastics, foams, wood, non-metallic elements, etc.) are removed.

Materials Handling

Conveying units are mainly usually conveyor belts. For the characteristics of the feed, especially at the early stages of ferrous metals handling (e.g., relatively large pieces of materials), handling equipment is of rough construction. Ferrous wastes are usually stockpiled and the primary feeding, after the preliminary operation outlined in the previous paragraph, is realized by cranes. Ferrous metals are abrasive; as a consequence, all the different parts of the equipment utilized for conveyors are subject to strong abrasive actions; also, the presence of fine particles and dusts must be carefully checked and reduced to avoid mechanical problems and to assure good environmental working conditions.

Comminution

After manual preparation and sorting, further size reduction actions are applied to scraps. Comminution actions are different according to destination of the materials, that is: (1) direct feeding to the furnaces or (2) further processing. In the first case, large scrap materials are milled utilizing shears, flatteners, and torch-cutting and turning crushers. The resulting pieces are then compacted by baling or briquetting in order to increase the apparent density of the scrap aggregates to re-melt, trying to avoid their possible floating in the mold. In the second case, crushing actions are usually applied, the goal being to reduce scraps to suitable dimensions to allow their processing (further crushing stages, sieving, separation, sorting, etc.).

A comminution unit usually develops its action through the application on the materials of four forces: impact, shear, abrasion, and compression. The quantitative relationships among them, in terms of cause-effect, strongly differ according to the equipment. For ferrous metals comminution, impact, and shear forces are those to be primarily applied, in order to optimize metal scraps and minimize fine particle production. Coarser fractions are mainly constituted by iron and steel, finer fractions usually contain the residues. Finer fractions can be divided in heavy and light (fluff) fractions. They present different characteristics in composition according to constituting particles weight. A utomotive S hredder R esidue (ASR) can be considered as the main source of metals. ASR heavy fraction mainly contains aluminum, stainless steel, copper, zinc, and lead. ASR fluff, for quantities and characteristics, represents an important class of ferrous metals end-life-goods-derived material of particular interest for the recycling sector. Fluff represents about 25% of the weight of a car. It is usually constituted by materials characterized by intrinsic low specific gravity (e.g., plastics, rubber, synthetic foams, textiles, etc.). When processed to perform their recovery, they pollute the materials presenting higher specific gravity (i.e., copper, aluminum, brass, iron, etc.), constituting parts of the electrical devices of the vehicle that, for their shape, size (e.g., wires, metal straps, slip rings, wipers, etc.) and utilization remain concentrated in the lighter products. Such "polluting agents ," for their intrinsic characteristics, are not well removed by classical separation techniques. The development and application of efficient washing strategies for fluff could dramatically reduce waste and environmental pollution, allowing, at the same time, an increase in energy recovery through pure sorted polymer re-use. Furthermore, the potential to use finer fluff fractions to produce energy could contribute to increasing the full recovery of such kinds of products. To reach this goal the quantity and the quality of the metal contaminants have to be strongly controlled in order to not prejudice the quality of the final fluff-based fuel. Always with reference to car dismantling, ASR heavy fractions contain large quantities of both ferrous and non-ferrous metals. Their recovery is usually performed adopting recycling technologies based on heavy media and eddy current separation.

Shredding is usually the main comminution action applied. Processing layouts, embedding this phase, are usually applied to automobile hulks and to the so-called white goods, that is, stoves, refrigerators, washing machines, etc. The most utilized class of shredders are those based on the use of S wing -H ammer S hredders (SHS) , subordinately, R otating D rums (RD) , are also employed. The main characteristics of both types of equipments are described in Table 6.
Recycling Technologies. Table 6 Characteristics of the comminution units mainly utilized to perform ferrous metals shredding

Comminution units

Characteristics

Swing -Hammer Shredders (SHS)

The input material is fed from the side. Fed material flow is controlled according to the energy required for comminution, usually greater of one order of magnitude in respect of minerals. Material is transported into the relatively narrow gap between the impacting tools and the lower part of housing, where it is subject to an intense deformation and comminution. The material, which has become sufficiently small, is discharged from the chamber of comminution by means of grates. The configuration of the discharge grates can vary. Normally grate is placed above the rotor and in some cases a second one below it. According to different comminution chamber size and shape, rotating of the rotor in respect of the feed, grates position and configuration, different comminution actions can be thus performed. In the last decade a lot of efforts have been addressed to investigate to utilize SHS with vertical mounted rotors. Such comminution units are actually mainly utilized in the processing of metallic cuttings, waste wood, paper waste, etc. Their possible full and systematic use in ferrous metal recycling sector could embed several advantages, that is: (1) a lower residence time of the metal particles inside the comminution chamber (e.g., higher flow rate and less energy consumption), (2) a better liberation and (3) a lower compaction of the liberated thin-walled metal pieces. A more systematic utilization of SHS, with vertical mounted rotors, is strongly linked to a further development of shredders' mechanical architectures and utilized materials characteristics.

Rotating Drums (RD)

A spinning rotor with chains or bars attached to the outside is responsible for the comminution. Ferrous metals fragmentation is mainly due to: (1) the impact of the rotating chains or bars on metals and (2) metals projection against "special plates" installed on the inside the chamber. Crushed material is finally discharged when particles size is lower than the space, that can be properly set, between the "rotating unit" and the wall. This equipment is commonly utilized as the primary crusher. For their characteristics do not allow the degree of flexibility in terms of operative conditions as those allowed by SHS.

Separation
Separation technologies are applied when the shredded materials to recover are composed of different families of particles characterized by different physical-chemical attributes and different relative composition, texture, and shape. For these reasons automatic and in-series handling-separation strategies are required, as simple manual sorting or separation are unable to recover in an efficient and economically profitable way the different materials. Table 7 lists the equipment commonly utilized to perform separation of end-life-goods-derived products; those resulting from car dismantling represent the main source of complex ferrous metal waste to recover. Because they are constituted of different particles of different magnetic properties, specific weight, color, chemical composition, etc., they require different separation strategies [27-29].
Recycling Technologies. Table 7 Characteristics of the separation units commonly utilized to perform separation in the ferrous metal recycling sector, with particular reference to ASR

Separation units

Characteristics

Belt Magnets (BM) and Drum Magnets (DM)

BM and DM are usually utilized in the first stage of processing, which is on coarser fractions as they result from primary crushing. Permanent and/or electromagnets are usually utilized. Separation is realized adopting BM and DM. In the first case the magnet is located between pulleys around which a continuous belt travels. In the second case the magnet is installed inside the rotating shell, metals particles are attracted, removed from the other non magnetic fractions and discharged as the moving belt of the separator turns away from the magnetic field the metals particles. Following this approach iron and steel cannot be separated from nickel and magnetic stainless steels. An improper separation can negatively influence the further melting stage. For this reason, hand sorting, to reduce the contamination of the ferrous products, is usually performed after this stage.

Eddy Currents Separators (ECS)

ECS is realized passing the waste products to separate into magnetic field, as a result, eddy current induced in the non-ferrous metals, produces ejecting forces that throw away non-ferrous metals the waste feed flow. ECS is commonly applied after the first magnetic separation stage (e.g., DM or DM). The most utilized ECS architecture is based on an inclined ramp. The material is thus fed to the ramp. The ramp surface is usually constituted by stainless steel. Under the ramp surface a series of magnets is positioned. Due to the eddy current non-ferrous metals are deflected sideways. Separation is realized according to the trajectory followed by the different classes of materials. Other separation architectures are based on the use of a rotating cylinder or a conveyor belt: magnets are positioned around the rotating axis of the cylinder or fitted inside the head pulley, respectively. As in the previous case separation is realized according to materials trajectory variations.

Heavy Media Separators (HMS)

HMS are based on the utilization of a medium constituted by a finely milled solid (e.g., magnetite or ferrosilicon) and water. According to the solid/water ratio the density of the medium can vary. Usually such a value is between the value of the specific gravity of the two classes of materials to separate, so that a sink and a float product is obtained. In this process the recovered materials are then washed and dried. The fine particles of the heavy media are recovered, by magnetic separation, from the slurry resulting from product washing and re-utilized inside the process. Decreasing the size of the particles to separate, also separation efficiency decreases for the increasing effect of viscosity, in this case cycloning is utilized.

Testing

Different from other recycling -derived materials, the possibility of performing a rapid test on samples collected from the different recycled ferrous metal flow streams is particularly important, especially with reference to alloys. Alloys of similar grades and composition are usually difficult to discriminate. Specific attributes are thus evaluated both by expert personnel and by specific tests. Recognition, via human senses and analytical equipment, is thus performed to evaluate characteristics such as color (e.g., copper and brass distinction), apparent density and hardness (e.g., lead distinction from copper and brass), magnetic properties (e.g., iron and stainless steel), presence and attributes of spark patterns as they results from abrasion test, chemical reaction to reagents, chemical and X-ray spectrographic analysis (e.g., alloys composition), thermal behavior (e.g., melting point), etc. All the above-mentioned approaches are expensive and are difficult to implement on-line.

Sorting
As outlined in a previous paragraph, sorting is important because it allows removal of contaminants from the different ferrous metal flow streams handled in the recycling plant. Different sorting strategies addressed to recognize different metal scrap constituents have been developed and used. Table 8 lists the devices, and related architectures, commonly used to perform metal scrap sorting.
Recycling Technologies. Table 8 Characteristics of the sorting units commonly utilized to perform sorting in the ferrous metal recycling sector, with particular reference to ASR

Detection units utilized for sorting

Characteristics

Portable Optical Emission Spectrometers (POES)

POES can be utilized to perform the on-site sorting and identification of metals. Such an approach, even if not reaching the precision of the corresponding laboratory device, is quite useful to perform fast quality control, usually well satisfying recovered products grade requirements. POES is able to detect up to 90% of the currently produced grades of steel.

Image Analyzers (IA)

IA allow to perform sorting on the basis of the detected color. IA belongs to the first class of devices utilized for metal scraps sorting. Adopting this approach, zinc, copper, brass, and stainless steel are commonly well sorted. Even with technological improvements, both in terms of speed of processing (the same pieces can be checked several times), sensor quality (better discrimination in terms of recognizable colors), and resolution (minimum identifiable scrap piece), IA is not efficient when slightly different alloys have to be recognized.

Laser-Induced Breakdown Spectroscopy (LIBS)

The detection architecture is based on the analysis of the optical spectrum, or fingerprint, of a small spot on each metal particle that is evaporated using powerful laser pulses. Even though very powerful, the techniques show some limitations. The most important thing is that it is particularly sensitive to the status of the scrap surface. Presence can negatively influence measurements because laser pulses can penetrate for just few Å in the surface.

X- ray Fluorescence Spectroscopy (XRF)

XRF is based on the emission of X-rays emission inside an XRF unit. The fluorescence radiation generated by the atoms when they release the energy, after the excitation stage is collected and analyzed. Both emitted wavelengths, and the energy released are functions of the elements constituting the waste sample. Correlating the emission intensity it is thus possible to evaluate the content of a specific element within the sample.

Non-ferrous Metals

Aluminum is the main non-ferrous metals being recovered and recycled. The main source of aluminum scraps comes from the packaging, transport and homeware industries. Aluminum can be "infinitively" recycled. Its re-melting achieves several environmental goals: considerable energy savings (about 95%) in comparison with the energy required to produce it from bauxite ore, a consequent reduction of primary non-renewable raw materials exploitation, and a reduction of overall emissions (airborne dusts and CO2). The automotive sector significantly increased, in recent years, its use of aluminum; for this reason scraps from end-of-life vehicles represent the main source of aluminum for recovery and reuse. Other non-ferrous metals commonly recycled are copper, zinc, and lead, but as outlined in the previous paragraph the recycled quantities are not comparable with aluminum and they are also relatively easy to recover. For this reason, the following recycling technologies will be described with reference to automotive aluminum scraps.

Aluminum from Automotive Scraps

Aluminum scrap is usually classified into two categories: cast and wrought aluminum alloys. Recycling of aluminum scrap introduces several technical problems to the secondary ingot market. Because of its high reactivity, aluminum cannot be refined pyrometallurgically, as with copper or iron scrap. Therefore, the aluminum scrap can be recycled only by blending and dilution in order to obtain a specific alloy. Wrought aluminum alloys contain low percentages of alloying elements (e.g., silicon, magnesium, copper, and zinc), less than about 4% of the total. Casting aluminum alloys contain the same elements as wrought, but in greater amounts (the silicon content in cast alloys can range up to 22%).

Actual scrap sorting technologies produce mixtures of cast and wrought aluminum not suitable for recycling in wrought alloy production.

The wrought fraction of these mixtures has a higher value; if selectively collected its reuse as wrought aluminum alloys would prevent unnecessary downgrading. Moreover, aluminum alloys, used in vehicle manufacturing, are increasing and the recovery of aluminum alloys is quite interesting. Considerable efforts have been devoted to substitution of steel and cast iron with lighter materials such as aluminum alloys and polymers. Steel replacement by aluminum alloys could have a counterbalancing effect in the case of success in selection of cast and wrought aluminum.

Non-ferrous Metals Recycling Technologies

The same processing steps and strategies previously described for ferrous metals can be also applied for non-ferrous metals scraps. Such phases, for scrap originating from end-of-life vehicles (one of the main non-ferrous metals sources, in particular aluminum) can be synthetically identified in:
  • Collection, dismantling (e.g., removal of re-usable vehicle parts as engines, doors, glass, seats, etc. and hazardous parts as batteries, fluids, etc.) and/or manual sorting (e.g., metal miscellaneous material detachment, large ferrous metal separation)
  • Crushing and screening, to reduce non-ferrous metal scraps in pieces easy to handle for further processing
  • Separation of the different non-ferrous materials in classes of products characterized by composition attributes
  • Testing and/or sorting
Collection, Dismantling, and/or Manual Sorting

Non-ferrous metal manufactured goods to recycle can be constituted by pieces of different dimensions, ranging from aluminum cans up to a "jet airliner," thus, according to their size, they can be directly handled or must be cut to be properly handled. Often, non-ferrous metal parts are linked (e.g., bolted, welded, etc.) with other materials and have to be liberated. Hand dismantling and/or sorting in some cases is necessary, but in many cases it is time consuming and inefficient for the dimension and the degree of locking of the different constituting materials. For these reasons comminution/separation actions must be applied adopting specific processing layouts.

Materials Handling

Conveying units are mainly conveyor belts of rough construction. Non-ferrous wastes are usually stockpiled and the primary feeding, after collection, dismantling, and/or manual sorting is done by cranes. Ferrous metals are abrasive, and as a consequence all the parts of the equipment used for conveyors are subjected to strong abrasive actions. The presence of fine particles and dusts also has to be carefully monitored and reduced to avoid mechanical problems and to assure good environmental working conditions. Furthermore, problems related to explosive characteristics of aluminum dust have to be taken into account.

Comminution and Screening
Comminution and screening are currently applied to reduce metal scrap to different size classes that are sent for proper processing and recovery of the different materials constituting the feed. The considerations developed with reference to ASR metal fraction comminution, in terms of equipment and characteristics (Table 9), can also be directly applied to ASR non-ferrous metals.
Recycling Technologies. Table 9 Characteristics of the comminution units mainly utilized to perform non-ferrous metals size reduction

Comminution units

Characteristics

Alligator (AS) and/or Guillotine Shears (GS)

AS mimics, to perform cutting, the behavior of an alligator mouth. The device is constituted by two jaws, one fixed and the other mobile. The main advantage in the use of AS is its versatility in terms of different aluminum scraps it is (e.g., ship, aircraft, automotive vehicles and in general large objects) able to cut. Furthermore better detachment is allowed of different metals constituting a specific piece to dismantle. When a GS is utilized, aluminum scrap is placed underneath a cutting blade, which drops down onto the scrap creating the cut. GS are characterized by higher power ratings and productivity than AS.

Impact Shredders (IS)

IS as hammer mills are among the most used devices for aluminum scrap size reduction. Fragmentation is realized for two co-occurring effects: (1) hammers impact being the main one, and (2) scraps projection the secondary one, against mill chamber internal surface. Impact crushers are often also utilized. They use the same milling actions but invert the relative effects.

Rotary Shredders (RS)

RS utilized, as main milling actions: (1) cutting and (2) impact for their architecture are normally utilized to process light metal scraps (e.g., foil and beverage cans and containers).

Separation
Aluminum-based material separation technologies are mainly based on magnetic [27], eddy current [17], air [28], and sink-float separation [29]. The main characteristics of the different separation units, based on the previously mentioned physical principles, are reported in Table 10.
Recycling Technologies. Table 10 Characteristics of the separation units commonly utilized to perform separation in the non-ferrous metal recycling sector

Separation units

Characteristics

Magnetics Drum Separators (MDS)

MDS is usually constituted by a stationary drum with half of its surfaced lined with NdFeB magnets installed inside a rotating cylinder that is set up as a conveyor belt. Ferromagnetic particles are attracted, removed from the other non magnetic fractions and discharged as the moving belt of the separator turns away from the magnetic field ferromagnetic particles.

Eddy Currents Separators (ECS)

ECS is realized passing the waste products to separate into magnetic field, as a result, eddy current induced in the non-ferrous metals, producing a forward thrust (F) and torque (T) on the particles resulting in their ejection from the stream of non-metallic materials. Separation is realized according to the trajectory followed by the different classes of materials. ECS is commonly applied after the first magnetic separation stage (e.g., MDS). Other eddy current based separation architectures are based on the use of a rotating cylinder or a conveyor belt: magnets are positioned around the rotating axis of the cylinder or fitted inside the head pulley, respectively. As in the previous case separation is realized according to materials trajectory variations.

Air Separators (AS)

AS is usually applied to preliminary recover light fractions contained in the feed or on non-magnetic-fractions as they result from previous MDS and ECS. Light fractions (e.g., plastics, rubbers, foams, fibers, etc.) are usually sucked by a nozzle positioned above the conveyor.

Sink and Float Separators (S&FS)

S&FS are based on the utilization of a medium constituted by a finely milled solid (e.g., magnetite or ferrosilicon) and water. According to the solid/water ratio the density of the medium can vary. Usually such a value is between the value of the specific gravity of the two classes of materials to separate, so that a sink and a float product is obtained. In this process, the recovered materials are then washed and dried. The fine particles of the heavy media are recovered, by magnetic separation, from the slurry resulting from product washing and re-utilized inside the process. Decreasing the size of the particles to separate, separation efficiency also decreases for the increasing effect of viscosity, in this case cycloning is utilized.

Among the different separation approaches, sink-float separation (S&FS) plays an important role when non-ferrous metals products have to be separated and/or refined before the final re-melting stages. Such products, in fact, are usually composed of a wide range of materials characterized by different density, shape, and size class distribution, strongly affecting sink-float separation. For this reason separation is performed in different stages. After a preliminary removal of non-magnetic-fine fractions, a two- or three-stage density separation is carried out. With a typical three-stage density separation, low-density plastics, foam, and wood are usually preliminary removed (cut-off density: 1 g/cm3), then, utilizing a cut-off density, 2.5 g/cm3, high-density plastics, magnesium, and hollow aluminum alloys are recovered in the floating fraction and then processed again utilizing ECS. The "remaining" sink fraction is constituted by brass, zinc, lead, copper, and so on. One of the main limits of S&FS is that it is not capable of performing a good separation of cast from aluminum alloys or differentiating among the different alloy groups. Furthermore, S&FS present other limitations, namely:
  • Cost, the process is expensive both at technical (e.g., heavy media costs, complexity of the processing circuit, etc.) and environmental levels (e.g., strict control of the heavy media with reference to possible environmental pollution, water recovery and cleaning, etc.)
  • Media recovery, separated scraps are obviously contaminated by heavy media, thus they have to be cleaned and the media recovered (e.g., utilization of specific processing circuits)
  • Separation is strongly influenced by particulate solids' morphological and morphometrical attributes

To attempt to totally or partially solve address these issues, in recent years innovative separation/sorting technologies have been successfully proposed and adopted in many recycling plants.

Sorting
The main role of sorting strategies is to perform a sort of further refining of non-ferrous metals as they result from the previous separation stages. From this perspective, sorting is mainly addressed at realizing greater discrimination between different families of alloys and different classes inside the families. The technology utilized to fulfill this goal is based on two different approaches: color sorting (I mage A nalysis: IA) and L aser-I nduced B reakdown S pectroscopy (LIBS) . For IA, a new process applied to aluminum scrap allows enhancement of IA performance. It consists of a selective etching of the scraps in different solutions that produces, as a result, the coloring of the scrap according to the presence and quantities of specific alloys agent. LIBS allows determination of the chemical composition of each scrap in a reliable and cost-effective way. The main limitation of this approach is linked to the characteristics of the investigated surface (e.g., presence of paints, lubricants, adhesives, or other polluting substances) since the laser pulse laser can only penetrate to a depth of 30 Å or less on the surface of the aluminum. Table 11 lists the devices and related architectures commonly utilized to perform non-ferrous metal scrap sorting.
Recycling Technologies. Table 11 Characteristics of the sorting units commonly utilized to perform sorting in the non-ferrous metal recycling sector, with particular reference to products resulting from S&FS

Detection units utilized for sorting

Characteristics

Image Analyzers (IA)

IA allow to perform sorting on the basis of the detected color. IA belongs to the first class of devices utilized for non-ferrous metal scrap color sorting. Adopting this approach, zinc, copper, brass, and stainless steel are commonly well sorted. Thanks to great technological improvements, both in terms of speed of processing (the same pieces can be checked several times), sensor quality (better discrimination in terms of recognizable colors) and resolution (minimum identifiable scrap piece), recent studies demonstrated that IA allows good separation of magnesium alloys from hollow aluminum products.

Laser-Induced Breakdown Spectroscopy (LIBS)

The detection architecture is based on the analysis of the optical spectrum, or fingerprint, of a small spot on each metal particle that is evaporated using powerful laser pulses. Even if, at least in principle, very powerful, the techniques show some limitations. The most important is that it is particularly sensitive to the status of the scrap surface. Presence can negatively influence measurements because laser pulses can penetrate for just few Å in the surface.

A new class of sensing devices based on H yper S pectral I maging (HSI) has recently opened new interesting scenarios for the on-line recognition of the different products resulting from both ferrous and non-ferrous metal waste processing. A more detailed description of this technique is given in the section Future Directions.

Recycling Technologies: Plastics

Re-utilization of waste plastics has increased with "new" plastic polymers; such re-utilization has increased not only quantitatively but also qualitatively (e.g., a larger range of recycled polymers). Plastics can be roughly divided in two types: thermoplastics , which soften when heated and harden again when cooled, and thermosets, which harden by curing and cannot be re-molded. Thermoplastics are by far the most common types of plastic, comprising almost 80% of the plastics used in Europe, and they are also the most easily recyclable. It is easy to understand that when collection-recycling strategies are set up, mixing between thermoplastics and thermosets has to be strictly avoided.

Plastic materials can be considered relatively modern, but some "natural" polymers exist in nature (e.g., amber, tortoiseshell, and horn) that behave very similarly to "modern" manufactured plastics and were used in the past in similar ways. For example, horn, which becomes transparent and pale yellow when heated, was used in the eighteenth century to replace glass.

The first plastic material produced for industrial-scale applications was the Parkesine, later named Xylonite . This material was invented by Alexander Parkes, who exhibited it as the world's first plastic in 1862. It was used for such objects as ornaments, knife handles, and boxes, and for flexible products such as cuffs and collars. Since that time great steps forward have been made and today plastics are widely produced and utilized, creating massive problems with litter and waste disposal.

Plastics are continuously replacing other materials in a number of applications. From greenhouses, mulches, coating, and wiring, to packaging, films, covers, bags, and containers. It is only reasonable expect to find a considerable amount of P lastic S olid W aste (PSW) in the final stream of M unicipal S olid W aste (MSW) . In the EU countries, over 250 × 106 t of MSW are produced each year, with an annual growth of 3%. In 1990, each individual in the world produced an average of 250 kg of MSW generating in total 1.3 × 109 t of MSW [1]. Ten years later, this amount almost doubled at 2.3 × 109 t. In the United States, PSW found in MSW has increased from 11% in 2002 [2] to 12.1% in 2007 [3]. Increasing cost and decreasing space in landfills have forced considerations of alternative options for PSW disposal [4]. Years of research, study, and testing have resulted in a number of treatment, recycling , and recovery methods for PSW that can be economically and environmentally viable [5]. The plastic industry has successfully identified workable technologies for recovering, treating, and recycling of waste from discarded products. In 2002, 388.000 t of polyethylene (PE) were used to produce various parts of textiles, of which 378.000 t were made from PE discarded articles [6].

The better solution "to recover" plastic goods should be, when possible, to re-use them as they are. Such a choice can be adopted, for example, for crates or other plastic-manufactured containers, which can be used several times for products and/or materials transportation and handling. Re-using is better than recycling because less energy and resources are required (JCR, 2006). In any case both re-use and/or recycling present several important advantages, related to:
  • Reduced use of fossil fuels
  • Energy savings
  • Reduced emissions of CO2, SO2, and NOx

The main problems in plastic recycling are mainly related to the difficulties of developing and setting up reliable automatic sorting architectures and systems able to perform an efficient selection of the different polymers. These problems can synthetically be divided in two classes, that is: single and multiple type and color plastic separation. In the first case (e.g., bottle, container, etc.) separation is relatively easy, the main problem being related to correct polymer recognition independent of the presence of fillers and other chemical additives. In the second case (e.g., cellular phones, electrical and electronic devices, automotive parts, etc.) separation is more complex, as the objects are constituted by different types of plastics characterized by the presence of different fillers and chemical additives. In the latter case, both separation and recognition strategies have to be sequentially applied.

Waste Plastics Sources and Characteristics

Used plastic packaging and other plastic items can be valuable resources in the manufacture of new products and in the generation of energy. It is important that a society aims to make the best affordable use of these valuable plastic resources. This is good for the environment, for the economy, and for the international community. Analyses by the European Community (EU) indicate that besides their ecologic importance, raw materials and energy are also the most important competitiveness factors for EU industries. Therefore, the need to increase recycling, improving at the same time the quality and homogeneity of recycled materials to minimize environmental pollution and usage of resources, is thus an important topic the EU. There is a strong drive to recycle polymers from end-of-life products and avoid their ending up in landfills and waste incinerators because plastics recycling reduces CO2 emission and saves resources. The worldwide production of plastics was 230 million tons in 2005 [30]. In Europe, 53.5 million tons were produced in total. Out of 22 million tons of post-consumer plastic waste in Europe in 2005, 53% was disposed, 29% was used for energy recovery, and 18% was recycled [30]. According to the last EU Directive 2004/12/EC on packaging and packaging waste, the recycling level of plastics should dramatically increase in the next years. New, more cost-effective separation technology can thus provide an important incentive to increase recycling rates. The recycling of polymers that are present in relatively pure streams such as post-industrial waste and separately collected containers of food and beverage is generally well-developed in Europe. The situation is very different for the large and complex stream of post-consumer waste, including wastes such as W aste E lectrical and E lectronic E quipment (WEEE) , household waste , and A utomotive S hredder R esidue (ASR) . Effective recycling of these wastes is possible, as has been demonstrated at some places in Europe, by large investments in logistics and dismantling (cars, electronic equipment) or hand-sorting (household waste). Such strategies are expensive, however, and they are therefore not widely applied.

There are about 50 different groups of plastics, with hundreds of different varieties [31]. All types of plastic are recyclable. To make sorting and thus recycling easier, the American Society of the Plastics Industry developed a standard marking code to help consumers to identify and sort the main types of plastic. An example of these types and their most common uses are reported in Table 12.
Recycling Technologies. Table 12 Example of most common types of plastics and use

MediaObjects/978-1-4419-0851-3_16_Part_Fig9-116_HTML.gif

PET

Polyethylene terephthalate - Fizzy drink bottles and oven-ready meal trays.

MediaObjects/978-1-4419-0851-3_16_Part_Fig10-116_HTML.gif

HDPE

High-density polyethylene - Bottles for milk and cleaning liquids.

MediaObjects/978-1-4419-0851-3_16_Part_Fig11-116_HTML.gif

PVC

Polyvinyl chloride - Food trays, cling film, bottles for squash, mineral water and shampoo.

MediaObjects/978-1-4419-0851-3_16_Part_Fig12-116_HTML.gif

LDPE

Low density polyethylene - Carrier bags and bin liners.

MediaObjects/978-1-4419-0851-3_16_Part_Fig13-116_HTML.gif

PP

Polypropylene - Margarine tubs, microwaveable meal trays.

MediaObjects/978-1-4419-0851-3_16_Part_Fig14-116_HTML.gif

PS

Polystyrene - Yogurt containers, foam meat or fish trays, hamburger boxes and egg cartons, vending cups, plastic cutlery, protective packaging for electronic goods and toys.

MediaObjects/978-1-4419-0851-3_16_Part_Fig15-116_HTML.gif

OTHER

Any other plastics that do not fall into any of the above categories. An example is melamine, which is often used in plastic plates and cups.

All separation-sorting techniques are based on the identification of one or more physical properties to utilize to discriminate the materials to process in order to establish classes of physical attributes and to set up appropriate technologies to address materials inside these classes. One of the most utilized properties is the density. Unfortunately, such a parameter is not particularly useful when plastics have to be separated because this value is similar for the different polymers to recycle.

Technologies that address these resources need to be extremely powerful, as they must be relatively simple in order to be cost-effective, but also accurate enough to create high-purity products and able to valorize a substantial fraction of the materials, present in the waste, into useful products of consistent quality in order to be economical. On the other hand, the potential market for such technologies is large, and environmental regulations along with oil price increases have increased the interest in many industries both in waste sorting technologies, for the production of high quality secondary polymers, as well as in developing automatic sensors for quality assessment of waste-derived secondary polymers. This latter aspect is particularly crucial because: "no matter how efficient the recycling scheme is, sorting is the most important step in recycling loop." Fast, accurate, and reliable identification of the primary plastics and the polluting materials in the feed is thus essential to set up suitable mechanical actions on PSW and optimal sorting strategies on the resulting products. The attainment of this goal is thus fundamental for companies that buy recycled plastics, because they obviously want those recycled plastics have the same characteristics as virgin ones. Otherwise, it is not efficient, and sometimes dangerous, to use recycled plastics materials. A simple example is the case of polyethylene theraphalate (PET) and polyvinylchloride (PVC) , which are sometimes indistinguishable by sight. These two resins are contaminants to each other. Combinations of PVC and PET resins can result in the release of hydrochloric gases. The PET resin will be ruined even with only a few parts per million of PVC resin.

Waste Plastics Recycling Technologies

From what has been previously reported, it is clear that correct plastic recycling is not an easy task. Among all the different wastes materials and products analyzed in this section, plastics are probably the most difficult to separate and recover. Preliminary and efficient plastics sorting, as well as a continuous monitoring of the different waste plastics flow streams, are both key issues to develop optimal PSW products recycling strategies. From this perspective plastics recycling technologies can be divided into four main categories [32]:
  • Re-extrusion, that is, the re-introduction inside an extrusion cycle of plastics presenting the same characteristics
  • Mechanical, developed to recover different plastic products by a physical processing
  • Chemical, addressed to produce feedstock chemicals for the chemical industry
  • Energy recovery, that is, complete or partial waste plastics materials oxidation to produce heat, power, and/or gaseous fuels, oils, and/or materials to be disposed of (e.g., ashes)

In the following, particular attention will be addressed to mechanical recycling, one among the four mentioned approaches that maximizes "waste plastics recovery," producing lower environmental impact.

Re-extrusion

The main assumption with re-extrusion is that the utilized waste scraps have to be constituted by polymers presenting the same characteristics as the original product. Manufactured products resulting from this process that do not satisfy quality composition constraints are usually addressed to a use where mechanical properties are more important than compositional ones (e.g., crates).

Mechanical Recycling
When mechanical recycling strategies are applied, sorting, at the different stages of the processing, represents an important issue. Despite great technological developments, most current plastic sorting continues to be done by hand. Manual sorting is a simple process that needs very little technology. It is a labor-intensive, costly, and inefficient method for sorting materials and more specifically plastics. For this reason, as previously outlined, the Society of the Plastics Industry instituted a voluntary labelling system. The system created a set of codes (Table 12) for each of the six most commonly used resin types. Even with this labeling system, it is still difficult to manually distinguish polymer types due to the condition of the plastics as they reach the separation facility. Plastic containers, in fact, may be crushed, cracked, or covered, rendering the resin label practically useless. In any case, systematic and extensive manual sorting of plastic parts, bottles, and the like is counterproductive since accurate, high-speed flake-sorting technology exists to separate one plastic from another; in fact, a number of automated sorting strategies have been investigated, developed, and implemented in recent years. They can be divided in two categories according to the size of plastics objects to sort:
  • Macro-sorting deals with the separation of bottles or containers, as a whole. Such an approach has the advantage that it does not require any specific preparation of the sample before sorting. Specific polymers' attributes have been detected and according to their characteristics further separated, usually following air-blow-based strategies.
  • Micro-sorting is applied after the plastic materials have been milled into pieces. This system has the advantages of lower handling costs and larger volume processing. A more sophisticated technology is required: a real mechanical processing sequence (e.g., comminution, classification, separation, etc.) has to be set up and applied.
Macro-sorting
Plastic macro-sorting is addressed to separate plastics manufactured goods as recovered after their use. Strategies have to be thus addressed to recognize big targets. The main problem to face, following this approach, is to set up a suitable processing line that is able to handle large pieces and consequently large and cumbersome stocks. Different techniques have been investigated in the past years and are currently utilized, some of them are outlined below:
  • Near-infrared spectroscopy
  • X-rays analysis
  • Laser aid identification
  • Marker systems

Near-infrared spectroscopy (NIR)480432 . This techniques is one of the most utilized to perform an automated sorting of post-consumer plastic containers. NIR has the advantage that direct or close contact between the detector and the sample is not necessary. NIR instruments are also compatible with flexible fiber-optic probes. It is based on the energizing of the unsorted, unidentified plastic with near-infrared waves (600-2,500 nm). When the infrared light reflects off the surface of the plastic, the different resins' characteristic infrared absorption bands can be measured. The detected bands are then compared to known polymer spectral bands response, in the same wavelength range, to determine the plastic type. Such an approach is characterized by many advantages. The most significant one is the detection/identification speed. Because of the great scanning speed allowed by spectroscopic devices, many readings of one sample can be taken in short periods of time; a multiple check of the same object is thus possible, allowing set up of proper and reliable identification strategies. Detection speed also allows an increased volume of plastics sorted in smaller amounts of time. The second advantage is the lack of specimen preparation. Labels or other obstructions like dirt do not significantly interfere with readings thanks to the option of performing multiple checks. Finally this detection architecture presents another advantage: color does not interfere with proper resin identification. Except for black, the readings are independent of the color of the resin. Black containers represent a problem, because their color is a strong absorber in the near-infrared region. As a result, black plastic produces a featureless spectrum that, in many cases, does not allow proper identification [33].

X-rays analysis . This sorting approach is based on the study of the transmitted or reflected wavelengths in the X-ray region. This technology is mainly applied for PVC sorting. Chlorine atoms in PVC give a unique peak in the X-ray spectrum that is readily detectable. X-ray fluorescence (XRF), based on energy level variations of core electrons of atoms, can be used to detect elements in plastics, except for H, C, N, and O, that are usually detectable utilizing infrared spectroscopy. XRF presents many advantages: ease of use, rapid preparation and analysis of the sample, a large range of element detection, etc. It can be thus utilized on-line. Furthermore, it allows one to quantitatively determine the presence and characteristics of fillers, pigments, and flame retardants.

Laser aid identification . With this approach the detection architecture identifies plastics by shining a laser beam onto the surface to be identified and then analyzing the material's response. Utilizing an infrared thermographic system, various material properties including absorption coefficient, thermal conductivity, thermal capacity, and surface temperature distribution can be thus determined. The detected properties can then be analyzed to identify plastic type. The resulting system is suitable for quick analysis and identification of various plastics. The approach presents some advantages, that is: (1) different thickness, forms, and surface structures of plastics containers do not play any role in the identification and (2) printing and different additives (softeners) also do not play any role. The limits are related to (1) the presence, in terms of quantity and quality (particularly of carbon), of fillers, (2) difficulty in classification of plastics due to the evaluation of the maximum temperatures directly after laser radiation and, as a consequence of (1) and (2), a lower identification speed, in comparison with spectroscopy and X-ray, wherein checking of plastic containers can be carried out within only 1/10 s.

Marker systems. This approach entails marking either the container or the resin itself with something readily detectable. There are no barriers standing in the way of an automated sorting system that would read a hidden marker and identify resin type. Many studies and attempts were carried out in the 1990s, but with low success, mainly due to problems arising both at the production and recycling levels. Every packaging production line would have to install a marking system on their line. Also, each recycler would need to install a machine to scan for the marking on the containers.

Micro-sorting
Plastic micro-sorting separate post-consumer plastics after a combination of comminution/separation processes specifically addressed to remove contaminants (e.g., non-plastic materials) and to increase bulk density, lowering storage requirements and shipping/transport costs, ease material handling and conveying, and liberating materials. Comminution is thus a fundamental and critical step when complex plastic waste streams have to be processed. Waste feed size reduction, in fact, has to satisfy comminution-liberation requirements and at the same time not produce too much fine fractions, which represent a problem in the further separation stages. According to the recycling plant "input" feedstock, different comminution strategies (e.g., number of comminution stages, utilized equipment, and operative conditions: dry or wet) have to be selected. Table 13 describes some of the comminution units commonly used to perform waste plastics shredding.
Recycling Technologies. Table 13 Characteristics of the comminution units mainly utilized to perform waste plastics shredding

Comminution units

Characteristics

Hammer Mills (HM)

Movable hammers mounted on a rotating shaft hit and/or throw plastic against mill chamber or the other waste fed material. As a result comminution is realized. Particles are recycled inside the hammer until they do not reach a size lower than the aperture of a grid installed at the exit of the mill chamber. HM can handle without problems metal contaminants, high energy is required, milled particles are not uniform and the process produces a lot of noise.

Ring Mills (RM)

A RM is usually constituted by a steel rolling blade. This blade chops and grinds the plastic that is placed inside the roller. After it has been ground up to the desired size, it falls through the small holes located beneath the rolling blade.

Shear Shredders (SS)

This machine uses one or more rotating shafts, each with a set of cutting disks or knives mounted closely together on the shaft(s) that sits in a chamber at the bottom of a feed hopper. As the shaft rotates, the cutting devices pull the material down through the small spaces between the cutting disks/knives and the surrounding chamber.

Two or Four Shaft Shear Shredders (TSSS or FSSS)

The equipment can be composed by two- or four-shafts shredder with rotary blades (e.g., sharp-corners disks provided with hooks) and spacer combs, that keep the tools clean and make material unloading easy. Once the material goes into the hopper, the shredder catches the material and begins to cut it grossly. Thanks to the high cutting torque and the different conformation of the cutters group it is possible to shred pieces made of different materials. FSSS can handle metal contaminants, relatively low energy is required, particles are well liberated, good size control, throughput rate is lower in comparison with conventional shredding machine without screen, high maintenance cost.

Granulators (Gn)

The main feature of Gn is represented by the rotor conformation provided with short blades with staggered arrangement. During rotation every tool scratches the material and makes the final shredding. Gn are particularly efficient when material characterized by high thickness and resistance has to be cut. They realize a good liberation of the materials, a high throughput, they cannot handle metals (e.g., contaminants) and are characterized by high maintenance costs.

Cryogenic Comminution Units (CCU)

CCU realize a fine grinding by using liquid nitrogen, usually the material is blended with liquid nitrogen to provide sub-zero temperature level up to −150°C, to cool the material in a grinding mill. The cryogenic process produces fairly smooth fracture surfaces. Little or no heat is generated in the process. This results in less degradation of the rubber. Even if the price of liquid nitrogen has come down significantly, recently the process is always characterized by high operative costs. CCU are thus ideal for fine pulverizing of thermo plastic and heat sensitive materials, they also allow to reach an excellent liberation of the materials.

After comminution, plastic wastes are reduced in dimension but obviously maintain their original composition, that is plastics and contaminants (e.g., ferrous and non-metals, nonferrous, foams, film, rubber, labels, paint and coatings, metallic foils, glass, rocks, sand, dirt, etc.). Contaminant removal has thus to be carried out adopting different classification/separation strategies strictly linked to the size class distribution of the flow streams and to the contaminants' characteristics, with respect of the polymer/s to recover. For waste plastic recycling, different from what is usually carried out in the recycling sector, some separation stages are carried out in wet conditions, that is, using a fluid, usually water and sometimes heavy-media, to enhance separation devices' efficiency. In some cases, when the feedstock is particularly contaminated (e.g., automotive shredder residue), the recycling process starts with a water or heavy media-based separation stages in order to remove as much contaminants as possible, as metal, rocks, glass, and sand that could damage size reduction equipment could negatively affect the further recycling stages.

To fulfil the previously mentioned classification/separation goals different techniques are currently used:
  • Air classification
  • Magnetic and eddy current separation
  • Density-based separation processes:
    • Sink-float separation (wet process)
    • Jigging (wet process)
    • Hydrocycloning (wet process)
    • Centrifuge-based separation (wet process)
    • Air table classifiers and gravity table separators (dry process)
  • Surface-based separation processes:
    • Electrostatic separation (dry process)
    • Flotation (wet process)
  • Selective dissolution

Air classification . Air classification is commonly used to remove, in dry conditions, light contaminants such as dust, small foam particles, paper, glass powders, etc. Usually aspirators or air-cyclones are used [28]. Air classifiers are, in principle, simple devices, their control can vary according to feed characteristics. Equipment has to be correctly set for each stream of material. Separation of material is based on differences in terminal velocities in an airstream and is dependent on particle density as well as morphological and morphometrical attributes. An air-cyclone provides a simple and economical means for most medium to coarse and/or heavier particle collection applications. The centrifugal action and gravitational forces are the operative principles of cycloning. The air flow containing the particles to classify/separate goes through a high-velocity inlet, forcing particles to the collector wall in spiral motion. This, together with gravitational pull, forces the heavier particles downward, while the lighter ones travel upward via the inner vortex and out the air outlet on the top side. Air-based classification represents a fundamental step in any plastics recycling facility, handling complex plastic-rich parts from end-of-life durables (e.g., automotive derived parts, electronic and electrical devices, and appliances, etc.).

Magnetic and eddy current separations . These separation techniques are utilized both at the beginning of the recycling process, as well as after different handling stages. Usually ferrous (e.g., low-grade stainless steel, nickel alloys, etc.) and non-ferrous metals (e.g., aluminum) are removed using magnets [27] and eddy current [17] and/or electrostatic separators [29], respectively. The characteristics of these separation devices are described in the section dealing with metal recycling. The magnetic separators commonly utilized are belt magnets, magnetic pulleys, and drum magnets. The refining/control of the final products is usually carried out by high-intensity permanent magnets.

Density separation processes . Density separation is the most frequently applied technique to recover different plastics from a mixed plastics-pollutants streams. Such an approach can be also be profitably used to separate polymers belonging to the same family but containing different additives. Density-based separation techniques are more reliable than those "only" based on plastics surface characteristics. Bulk plastic properties, in fact, are less sensitive to possible alteration linked to specific environmental conditions (e.g., lighting, oxidation, etc.) or contaminant presence (e.g., oil, dirt, various costing, etc.). When density separation is applied, waste materials to be separated are placed in a medium characterized by a density that is intermediate between two or more densities of the particles constituting the waste. Following this approach, the fluid and/or recovered solid fractions have to be further processed for environmental and cleaning purposes, respectively.

Sink and float process . Sink and float separation systems are very common. They represent a simple and robust approach to separate materials characterized by different densities. The method simply involves depositing the materials in a tank filled with water or other liquid at a specific density. The lighter materials float and the heavier ones sink. For a sink and float system to work efficiently the materials' densities must differ greatly from one another (e.g., polypropylene, PP: 0.96 g/cm3, high-density polyethylene, HDPE: 0.94 g/cm3, medium-density polyethylene, MDPE: 0.926-0.940 g/cm3, low-density polyethylene, LDPE: 0.915-0.925 g/cm3, linear low-density polyethylene, LLDP: 0.91-0.94 g/cm3). Furthermore, even when applied the process is difficult to handle because chemicals have to be added to water, to modify density, or specific heavy liquids [34], as bromoform (CHBr3) (2.87 g/cm3), TBE: 1,1,2,2-tetrabromoethane C2H2Br4 (2.95 g/cm3) and methylene iodide (3.31 g/cm3), have to be utilized. Such liquids are highly toxic, require stringent conditions to minimize exposure to workers and create plastics contaminated fraction that have to be further cleaned [34, 35]. Furthermore, the presence of possible contaminants and bubbles on the plastic surface, plastics particle size and shape, and characteristics of fillers and additives also strongly affect separation.

Jigging . Jigging is based on the application of repetitive pulsation actions to a particle bed by a current of water in stratification of plastic waste particles of different specific gravity. A jig operates in a cyclic manner where one cycle consists of four stages, namely, inlet, expansion, exhaust, and compression. In the inlet stage the bed lifts up en-masse. Near the end of the lift stroke the particles at the bottom of the bed start falling resulting in loosening of the bed which, in turn, causes its expansion or dilation. During the third and fourth stages of the jig cycle, the particles resettle through the fluid, and the bed collapses back to its original volume. The pulsation and suction is repeated to bring about stratification with respect to specific gravity across the bed height.

Hydrocyclones. Hydrocyclones are an economical and effective tool for separating mixed plastics and for removing many contaminants from a target plastic. A hydrocyclone transfers fluid pressure energy into rotational fluid motion. This rotational motion causes relative movement of materials suspended in the fluid thus permitting separation of the materials from one another [36]. The mixed fluid enters tangentially at the inlet, which causes the material to rotate within the vessel and ultimately to form a vortex. As this vortex of fluid spirals within the cyclone, heavier materials are forced outward by centrifugal force and down from the barrel section into the cone section. The materials more dense than the fluid flow down the inner wall and exit through the apex and out the underflow port with a portion of the fluid. Lighter materials are swept into the center vortex by inward fluid motion, and are carried vertically up through. Different from classical applications (e.g., mineral processing, food industry, pharmaceutical industry, etc.), when hydrocyclones are utilized in plastics waste recycling, two factors have to be carefully taken into account: (1) the tendency of recycled plastic particles to assume a plate-like shape and (2) the low differences usually existing between different plastics. Hydrocyclones can be considered intermediate density based separation units. For their characteristics, in fact, can be placed between a classical sink-float and a centrifugal process. Hydrocyclones present several advantages: they require very little space, are quite efficient, and outputs can be high; on the other hand, they require a more complex fluid-dynamic circuit (e.g., presence of pumps) and stricter control of feed characteristics (e.g., water solids ratio), etc.

Centrifuges . These equipments are very efficient; they balance optimal separation performances with a reasonably high separation rate. Morphological and morphometrical particle attributes affect in a limited way this separation, because of the applied centrifugal fields characterized by high values. This technique is particularly efficient when fibers and/or film-like particles have to be recovered.

Air table classifiers and gravity table separators. These devices come from mineral processing and metal recycling industries (e.g., automotive derived waste containing plastics). Their application is quite limited.

Surface based separation processes: Electrostatic separation . When this separation is applied usually the particle charging method is based on the triboelectric effect. Such an effect is based on a simple principle: when dissimilar materials, for example, particles of two different plastics, are rubbed together they transfer electrical charge and the resulting surface electrical charge differences can be used to separate the two plastics in an electric field; usually charged plastics fall down freely in the area between two electrodes. The particles are drawn to either positive or negative electrode according to the polarity of the charge. According to their trajectory they are thus collected and separated. Many plastics can be separated with this technique: ABS (Acrylonitrile Butadiene Styrene) and HIPS (High Impact Polystyrene) from end-of-life electronic devices, ABS and PMMA (Polymethyl Methacrylate) from automotive waste, PE (Polyethylene) and PP (Polypropylene), PET, (Polyethylene Terephthalate) and nylon, PVC (Polyvinylchloride) and PE from cable scrap, PVC and PC (polycarbonate) from bottles, etc. Electrostatic separation has two main advantages: it can be carried out in dry conditions and the separation architectures and equipments are relatively simple. The main disadvantages are related to the shape of the particles, which influences their surface charge and separation effect. Furthermore when electrostatic separation is applied particles' humidity and moisture have to be strictly controlled.

Flotation . Froth-flotation is another possible method to perform plastics micro-sorting. Flotation works similarly to sink and float systems. Froth-flotation is based on the plastic particles surfaces' chemical-physical attributes; for this reason it is particularly suitable for when plastics of similar densities but different surface properties have to be separated. As outlined in [31] a specific plastic can be separated from a complex waste stream by flotation after treating the waste in alkaline solution [32, 33]. Separation of mixed plastic, according to different plastic typologies, can be achieved utilizing appropriate collectors. A large literature exits on this topic [31, 37, 38]. Furthermore, specific wetting agents can be also used to prepare a hydrophobic property [36, 39]. Due to the conditioning some plastics that normally sink (hydrophilic behavior), adhere, according to their composition, to air bubbles generated by a controlled air flow pumped into the system. As a result such particles float to the surface. Materials that are not affected by the bubbles sink to the bottom. Collection systems at the top and bottom of the system can then recover the separated fractions. Other parameters affecting froth flotation are particle size and shape. The main advantages linked to the utilization of froth flotation in plastic recycling are that the technique is well known and settled from a technological point of view and that it is quite flexible in terms of application possibilities. The limitations are primarily related to plastic particle surface status (e.g., dirtiness and/or pollutants) and to the difficulty in defining precise control logics because of plastic waste variability and important factors that have to be taken into account.

Selective dissolution . Selective dissolution is a plastic sorting option that was investigated in depth, as the markers for macro-sorting purposes, in the early 1990s. The process separates mixed or commingled plastics waste into nearly pure reusable polymers without any mechanical pre-sorting techniques. The selective dissolution is based on two different principles: temperature-dependent solubility of different plastics in a single solvent and solvent-dependent solubility of different plastics at a specified temperature. These technologies are not cost-effective for commodity polymers but are sometimes the only solution for the liberation of different coatings associated with PP, such as paint or skin.

Chemical Recycling

Chemical recycling , using a depolymerization process, is applied to convert waste plastics, utilizing heat or heat and catalyst, in smaller molecules (e.g., gases, liquids, solid waxes, etc.), that can be used as a feedstock to produce new plastics or other chemical products. The term chemical is thus used, because an alteration is bound to occur in the chemical structure of the polymer. In recent years, a lot of attention has been addressed to this recycling approach (e.g., non-catalytic thermal cracking, catalytic cracking, and steam degradation) in order to produce different fuel fraction from plastic waste products. Several polymers can be profitably processed adopting this approach. Polyethylene terephthalate (PET), certain polyamides (nylon 6 and 6.6), and polyurethanes (PURs) can be efficiently depolymerized. The resulting chemicals can then be used to make new plastics that can be indistinguishable from the initial virgin polymers [40]. Polyethylene (PE) has been targeted as a potential feedstock for fuel (gasoline) producing technologies [41]. The two cited cases are just an example of chemical recycling potentialities. A great deal of literature exists on this topic, because a lot of research efforts and technologies development have been addressed to improve the utilization of this recycling technology. Chemical recycling, in fact, presents, at least in principle, a big advantage with the possibility of treating heterogeneous and contaminated polymers with limited use of pre-treatment [42]. An excellent review and analysis of this technique is reported in Al-Salem et al. [41].

Energy Recovery

Energy recovery is based on using waste plastics to produce energy in the form of heat, steam, and electricity. Deriving from crude oil, waste plastics when burned generate a high calorific value. Furthermore, producing water and carbon dioxide upon combustion, plastics behave similarly to other petroleum-based fuels [43]. Such a solution can be considered technically and economically correct when the other recycling strategies (e.g., sorting, mechanical, chemical, etc.) cannot be profitably applied. A typical example is represented by fluff, which is the light fine fraction resulting from car dismantling. Fluff is constituted of plastics, rubber, synthetic foams, etc. and well fulfils the concept of waste-to-energy product . This material, after a "washing" stage, (e.g., polluting materials removal: copper, aluminum, brass, iron, etc.), can be profitably utilized as fuel. Energy production can thus dramatically contribute to increase the full recovery of such a kind of secondary waste with a lower environmental impact (e.g., landfill reduction). Recent studies demonstrated that when plastic waste energy recovery is performed, foams and granules contribute to destroy CFCs and other harmful blowing agents present [44]. However, several environmental problems related to the emissions have to be faced when such an approach is followed, mainly the presence of (1) volatile organic compounds (VOCs), (2) particulate solids, (3) particulate-bound heavy metals, (4) polycyclic aromatic hydrocarbons (PAHs), (5) polychlorinated dibenzofurans (PCDFs), and (6) dioxins. Finally, the presence of flame-retardants (FRs) can influence the combustion process.

All the considerations previously outlined refer to thermoplastics. When thermoset plastics have to be recycled several problems arise. They, in fact, cannot be readily dissolved, melted, re-compounded, and reshaped like thermoplastics. Specific recycling strategies have to be set up. Mechanical recycling is thus primarily addressed to fine grind them for a further re-use as fillers in new thermoset resins or thermoplastic compositions or to recover natural filler or fibers originally utilized in the original thermoset product. In any case, chemical recycling as well as energy recovery processes can be applied to large range of thermoset materials.

Recycling Technologies: Fibers (Textiles and Carpets)

Fibers , both natural and artificial, are commonly utilized in daily life, as well as in technical applications. Also with reference to fiber-based apparels, legislation fixed severe constraints about their disposal at the end of their life-cycle; as a consequence proper recycling systems must be adopted. Recycling can be carried out via two different approaches: (1) to recover energy or (2) to recover fiber materials for their further reuse.

The energetic utilization of end-of-life fibers is not particularly efficient, because the energy generated from burning is less than the energy required for fiber manufacturing. This approach makes sense, from an ecological point of view, since proper combustion results in energy gains without significant air pollution and the consumption of resources. However, the production of synthetic fibers is more expensive compared to non-fibrous plastics. Also the production of natural fibers, like cotton, requires a large use of resources (e.g., water). Hence, product recycling of fibers will increase the sustainability of products and processes. With reference to cotton its caloric value is about 17 MJ kg [45], on the other hand, the energy demand for producing 1 kg of raw cotton is between 38 and 46 MJ kg [45] considering an oil consumption of 1 kg. As a consequence, any recycling process is more convenient than thermal utilization. If the same considerations are applied to man-made fibers, a different fiber-related-energetic-balance can be drawn. The water consumption for the production of synthetic fibers is significantly lower (about 1/10) compared to cotton. For acrylic fibers, for example, the demand ranges between 0.3 and 15 l H2O per 1 kg of fibers [46]. Energy consumption for polymerization, spinning and finishing is between 369 and 432 MJ kg [45] [46]. Given a caloric value of crude oil between 38 and 46 MJ kg it can be concluded that for 1 kg of fiber about 11 kg of crude oil is necessary. However, during thermal utilization only the caloric value can be used that is about the same or slightly lower compared to crude oil. From this it is obvious that thermal utilization should be replaced by any other process.

Technologies actually available for textile and carpet recycling do not offer satisfying solutions in terms of economic and ecologic demands, this fact is mainly linked to the difficulty of developing a correct separation, first, and a full characterization, after, of fibers, and other polluting materials. Fiber composition as well as their morphological and morphometrical attributes represent important factors to develop optimal re-utilization strategies, this last aspect being particularly relevant in carpet recycling . Recycling technologies have been developed, in terms of logics and complexity, dealing with textiles and carpets, respectively.

Textiles

Systematic textile recycling originated in the Yorkshire Dales (Great Britain) about 200 years ago, and the rag and bone men of days past were the predecessors of the actual "textile recycling businesses." They collected not only clothing, but also handbags, shoes, bedding, and curtains for re-use. These materials were then often sold abroad, as second hand clothing, but also to provide raw materials to the "wiping" and "flocking" manufacturers and for fibers reclamation to make new garments. Furthermore, it is well known that textiles were and are recycled for papermaking. Textiles recycling can follow different rules according to "recycled textile function":

According to their reutilization, recycled textiles can be up-cycled or down-cycled. In the first case they are used for more technically demanding application (higher value); in the second case, they are utilized for less demanding application (lower value). In these two cases, original textiles have to be mechanically processed, adopting specific comminution, classification, and separation strategies in order to recover the constituting fibers from the other materials (contaminants). Products resulting from these approaches are usually: shoddy (e.g., fabric made from the recycling of knitted products), mungo (e.g., fabric made from the recycling of woven products), cotton rag paper made from recycled cellulosic fabrics, etc.

As previously outlined, textile fibers can be classified into natural (e.g., cotton and wool) and synthetic. Recycled fibers demand is strongly influenced by several factors as:
  • Fibers composition and characteristics. The presence of fiber blends (e.g., elastic polyurethane) that make recycling more difficult, or the presence of polymers not commonly recycled (e.g., acrylics and polyesters) negatively impact on fiber recycling process
  • The possibility to identify new industrial sectors where recycled fiber products can be utilized
Textiles Source and Characteristics

Textile wastes can be originated by industry and/or consumers.

Textiles industrial wastes are originated during the processing, production, and/or the manufacturing phase. Such wastes are easy to recycle, the fiber composition and characteristics being known. Contaminants are usually not present.

Textiles consumer wastes are more difficult to recycle. They are usually constituted by fiber mixtures and contain "contaminants" (e.g., non-fibrous materials such as buttons, buckles, or other metal parts).

Waste textiles are usually collected by charitable organization. End-of-life apparel is then sorted according to a possible re-use, that is:
  • As re-wearable, cleaning and wiping clothes, short cut for nonwovens as well as for the paper and cardboard industry [45]
  • As recovered fibers, in this latter case a mechanical processing has to be applied in order to produce fibers of desired length for their further re-use

Textiles Recycling Technologies

According to the considerations previously outlined recycling technologies are applied when "recycled fibers" have to be produced. Waste textiles processing is usually carried out in dry conditions. Such an approach presents two advantages: (1) low energy consumption (e.g., drying is not required) and (2) no water treatment. In this perspective the main actions and the related equipments are reported in the following:
  • Human based sorting (e.g., separation of the different apparels typologies)
  • Milling and classification, to obtain fibers of the requested morphological and morphometrical attributes
  • Separation of the different fibrous and non-fibrous material according to their physical-chemical attributes
  • Fiber tailoring and characterization
Manual Preparation and Sorting
A human based sensing approach is commonly followed to recognize and preliminary separate the different apparels according to the different types of fibers. After this phase of sorting/grading, clothes are then packed as bales. Each bale is obtained by pressing and identified in terms of average fiber composition and weight. Each bale is thus assumed as the minimal identifiable raw material unit to address to different possible recycling phases, that is:
  • Second-hand clothing
  • Wiping and polishing cloths for industry
  • New products in the reclamation sector (e.g., component for new high-quality paper, upholstery, insulation, even building materials, etc.)
  • Filling materials (e.g., car insulation, seat stuffing, etc.)
Milling and Classification

Textiles milling is usually carried out utilizing equipment where cutting actions are maximized and the possible comminution effects on hard components (e.g., buttons, zippers, etc.) are minimized to reduce the presence of fine particles contaminants. Classification is usually carried out to recover/remove fine fractions before the further tailoring stage/s. Usually zig-zag classifier and/or pneumatic tables are utilized. As a result of this processing non-ferrous metal and plastics, if present, can also be recovered for further recycling .

Separation

Separation is usually addressed to recover the non-textiles materials inside the milled textiles products. Separation is usually carried out adopting magnetic separators (e.g., hump magnet, magnetic pulley, etc.). Metallic fractions (e.g., button, zippers, etc.) are thus recovered.

Fibers Tailoring and Characterization
Tailoring is a process specifically addressed to produce individual fibers and to disintegrate all residual textiles and yarns. At this stage of the process the main target is thus to develop a processing sequence able to progressively produce fibers from fabric. A three-in-one process, finalized to separate fibers from fabric, is usually applied, that is: picking, pulling, and tearing. Such a goal is usually reached adopting a series of drums with spiked surfaces characterized by an increasing number of finer spikes. Fibers classification is then carried out adopting air classifiers, their characteristics can vary according to classification goals. Tailoring can produce rather long (greater than 2 mm, e.g., new "long fiber" textiles making, nonwovens, etc.) or short (less than 1 mm, e.g., viscosity modification, composite reinforcement, concrete, mortars, adhesives, etc.), fibers according to their re-use. Tailoring process has to be quantitatively and qualitatively assessed, performing a morphometrical and morphological fibers characterization. Fibers morphological and morphometrical attributes (e.g., fiber length, width, and profile structure) thus represent important factors to develop optimal recycled fibers re-utilization strategies, allowing correct identification and new potential applications. Most of the literature describing the recycling of fibers does not provide any details about fiber characterization. Recently, a procedure (MorFi), originally developed for pulp characterization [47], was successfully applied for short fiber characterization [48]. Following this procedure, the length of fibers (FL) was measured automatically adopting an imaging-based approach: a suspension flowing through a flat cell observed by a digital CCD video-camera. The analysis of morphological properties of fibers performed by MorFi provided arithmetical average length of fiber (the value most sensitive to the effect of shortening of degraded fibers during their mechanical treatment), expressed by the equation
where z i is the number of fibers in a given class of length, and l i is the mean length of fibers in the given class.

Carpets

Differently from textiles, carpets represent a more difficult product to recycle, the reason being linked to its compositional characteristics. A carpet, in fact, is usually constituted by a two-layers backing of polypropylene. In between the layers styrene-butadiene latex rubber (SBR) is joined by calcium carbonate (CaCO3) and the fibers are tufted into the rubber (the majority being nylon 6 and nylon 6.6 textured yarns). The SBR adhesive is a thermoset material, which cannot be re-melted or re-shaped. Nylon generally performs the best among all synthetic fibers as carpet face yarn, but it is also the most expensive. This also explains why most of the recycling effort is on nylon recovery. Due to the "complexity" of carpet, at least in comparison with textiles, if fibers have to be recovered, more complex comminution-classification-separation strategies have to be applied. It is possible to recover fibers, mainly polypropylene, from the backing that range from 3 to 25 mm in length. On the other hand, a fraction originating from the pile yarn is obtained. The latter fibers are between 12 to 25 mm in length [49-53]. The composition of this fraction is reported to have 36% PP (fibers from the backing), 18% nylon (fibers from the pile yarn), and 46% (non-fibrous) SBR and CaCO3 [50, 51]. An extensive literature exists on different possible positive re-use of recycled carpet fibers in concrete and soil reinforcement and several other applications [49-53].

Carpet Source and Characteristics

Carpet wastes can be originated by industry and/or consumers. Different from textiles, post-consumer-waste carpets represent the larger source. Carpet can be considered a sophisticated product. It is, as previously described, constituted by many materials assembled to assure the durability of final manufactured product, the consequence is that its disassembly and recovery is difficult and requires complex technology and suitable separation control actions to properly recover and certify the different constituents. Carpet-derived products, both fibers and polymeric materials, can be thus originate lower or higher value recycled products, according to the adopted recycling strategies.

Carpets Recycling Technologies

As for waste textiles, carpets recycling technologies are carried out in dry conditions. The main processing steps are outlined in the following:
  • Preliminary fiber identification and sorting
  • Comminution and classification
  • Separation
  • Solvent extraction of nylon
  • Nylon depolymerization
  • Melt processing
Fiber Identification and Sorting

Carpet recycling strategies have to be set up according to upper surface fiber characteristics. Fiber identification thus represents a key issue to properly address carpet to different downstream recycling tracks [54]. Usually portable infra-red (IR) spectrophotometers are utilized; they usually allow a fast and reliable recognition of nylon 6, nylon 6.6, polypropylene, polyester, and wool fibers. Sorting is usually applied in the collection point; sometimes such a control is centralized in the stocking facilities of the recycling plant. This last solution is usually more efficient.

Comminution

Shredding and/or grinding are the commonly applied size reduction actions [45, 55]. These actions are usually performed by a mill with rotary drums equipped with hardened blades. The material after shredding is then sieved by a grid installed at the exit of the mill chamber. The material, which has become sufficiently small, is thus discharged from the mill chamber, the other remains in the mill chamber for re-cutting. Comminution usually produces an increment of the temperature, such a fact can negatively affect milled materials characteristics, for this reason comminution equipments are usually designed to realize a high torque and a low rotational speed. Studies have been carried out to perform carpet cryogenic milling [56]. Following this strategy comminution results are particularly efficient because the freezing action of liquid nitrogen, or CO2, changing the mechanical behavior of the different carpet components allows their better milling and liberation of the different constituents. Costs are usually higher than classical blade based milling and as a consequence the use of this technology is limited. In some cases milling is also realized utilizing water jet [57].

Separation

Carpets constituents separation is usually carried out adopting different processing strategies based on a series of combined comminution-classification stages and a further density based separation utilizing both the "simple" gravitational and/or centrifugal field. The second approach is usually adopted to enhance the differences existing between the different materials to separate. Following these strategies filler, nylon and polypropylene can be recovered. A detailed description of this approach is reported in two papers from J. Herlihy [58] and H. P. Kasserra [59]. In some cases the separation of the different carpet components is realized without comminution [60, 61]. Waste carpet is thus subjected to a preliminary clipping of the exposed fiber, a further bombardment of air and steam of the carbonate-filler latex backing and a combination of peeling and picking. The combination of these actions allows recovery of up to 95% of the face fibers.

Solvent Extraction of Nylon

Such an approach is utilized when high value nylon has to be recovered from carpet at the end of its life-cycle. Solvents commonly utilized are aliphatic alcohol [62], alkyl phenols [63], and hydrochloric acid [64]. A preliminary comminution of the carpet is always required. The optimal size-class ranges between 1 and 5 cm. Solvents utilization presents both advantages and disadvantages. The advantages are that the use of solvents allows to good recovery of nylon (yield about 90%) characterized by a relatively low degradation. The disadvantages are those related to the use of chemicals and their further recovery and/or re-use, when possible. Furthermore, process temperature and the time required for nylon extraction are other constraints to take into account. Their values change according to the dissolution process adopted. Sometimes S uper C ritical F luids (SCF) are utilized [65, 66]. The process is commonly carried out in batch conditions at high or low pressure and temperatures according to the fluid utilized: CO2 [65] or formic acid [66], respectively.

Nylon Depolymerization

Depolymerization is usually applied to recover nylon, from nylon carpets, because nylon resin has a remarkable higher value than the other polymers commonly utilized in carpet manufacturing [67]. A typical depolymerization process [68] is based on a preliminary carpet sorting and a further mechanical shredding, the recovered nylon six face fibers are sent to a depolymerization reactor and treated with superheated steam in the presence of a catalyst to produce a distillate containing caprolactam, that is the single monomer that after polymerization originates nylon 6. The crude caprolactam is then distilled and re-polymerized to produce again nylon 6 [69]. Another process to obtain caprolactam is based on the utilization of a two-stage pyrolysis process. The ground nylon carpet, without separation, is dissolved with high-pressure steam and then continuously hydrolyzed with super-heated steam to form caprolactam [70, 67].

Melt Processing
Melting and compounding are two other processes currently carried out for carpet recycling. Both require a preliminary size-reduction process. By melting, a thermoplastic polymer is converted by extrusion in resin pellets [45, 55], if more polymers are blended together and then extruded a compounding process is applied. For its characteristics, the products resulting from comminution have to be "compacted" before the extrusion process; they, in fact, are quite bulky. Specific equipment (crammer-compactor feeder) is thus utilized. Extruders can vary, ranging from single screw, twin-screw co-rotating, or twin-screw counter rotating architectures according to feed and required products characteristics. The melted-extruded products are then cooled and cut to obtain pellets. Usually ring, strand or underwater pelletizers are utilized. In Table 14 are synthetically reported some of the characteristics of pelletizing devices commonly utilized for waste-carpets-based melted-extruded products chopping.
Recycling Technologies. Table 14 Characteristics of the pelletizing equipment utilized for waste-carpets-based melted-extruded products chopping

Equipment

Characteristics

Water Ring Pelletizer (WRP)

A WRP is commonly utilized for liquid polymers in which pellets fall from cutter knives into an annulus on the surface of a body of cooling liquid. The velocity and trajectory of the pellets are controlled by the projection of a spray of cooling liquid radially across the dies from which the pellets are severed. Band heaters in proximity to the dies maintain the polymer in a liquid state prior to extrusion. WRP is particularly suitable to be applied for polymers characterized by a low melt flow index (e.g., polyethylene).

Strand Pelletizer (SP)

In a SP polymer strands discharging from the die head are sent to cooling water-based-device. The water "wetting" the polymer strands is eliminated utilizing an "air knife." The dry, solidified polymer strands are then delivered towards a strand pelletizer where cutting is applied. One of the main disadvantages of SP is that a large floor space and particular care has to be addressed to control possible strands breakage. Polymers as nylon, polyester terephthalate, and polypropylene are commonly processed by SP.

UnderWater Pelletizer (UWP)

UWP is usually constituted by an extruder that conveys the polymer melt to the die plate through the start-up valve. The melt stream is then divided into a ring of strands that flow through the annular die into a cutting chamber flooded with process water. A rotating cutter head in the water stream cuts the polymer strands into pellets, which are immediately conveyed out of the cutting chamber. The pellets are cooled and transported in a slurry to the centrifugal dryer. Pellets are then separated by water through rotating paddles. Polymers as nylon, polyester terephthalate, and polypropylene are commonly processed by UWP.

Melted-extruded products present different characteristics and market values according to the characteristics of the original feed-stocks [71]. When carpets are constituted by polymers as plastic and polypropylene [72], the resulting pellets (e.g., compounds) are of low quality due to the fact that they are thermodynamically unstable when melt-mixed [73, 74]. They must be thus stabilized to prevent coalescence during melt processing [75]. This process of stabilizing polymer blends is commonly called compatibilization. It usually consists of the addition of a premade block copolymer composed of blocks that are each miscible with one of the homopolymers [76].

A melting plant is relatively simple to utilize and maintain. Its main limits are primarily related to its operating principle, that is, low flexibility in terms of possible modification of final products characteristics. Furthermore, the presence of water represents a further environmental processing problem (e.g., H2O filtering, temperature control and re-circulation to face when melting is applied).

Pellets resulting from the previously described melt-based process are commonly utilized in the molding process, as they result from recycling or are blended with virgin polymers. Other common applications are in glass fiber-reinforced composites, where they are utilized as matrices [77].

Future Directions: Innovative Control/Sorting Devices/Logics Integration in Recycling Plants

The different recycling technologies described and analyzed for the different waste materials (e.g., paper, glass, metals, plastics and textiles) clearly demonstrate that improvements in comminution-classification strategies, and further separation technologies (e.g., magnetic, electrostatic, sink-float separation, flotation, etc.) can be mainly carried out by introducing innovative control devices and architectures, as equipment technology and characteristics have reached a very high level of quality and reliability. Such classes of innovative devices can be also utilized as detection systems to realize innovative sorting architectures. Single and/or combined control/sorting actions can be thus developed, taking into account different aspects: (1) waste streams physical-chemical characteristics, (2) market requirements for concentrates of specific quality, and (3) related innovative control/sorting device/logics operating at different scale, that is, at single equipment and/or plant scale.

Waste Feed Streams Characteristics

Processing actions applied to different particular solids waste streams, constituted by different materials (e.g., paper, glass, metals, plastics and textiles), usually perform a change of some physical attributes of the wastes, these changes depend on the intrinsic characteristics of the constituting materials and the actions applied. For example, comminution produces a reduction of the size class distribution of waste, originating smaller particles of different morphological and morphometrical characteristics and producing the liberation of the different materials originally locked (e.g., mixed particles). On the other hand, separation actions allow grouping together of particles according to a specific property (e.g., density, conductivity, magnetic, color, texture, etc.), originating a concentrate.

What are the strategies to apply to make improvements in terms of a correct utilization of waste streams characteristics versus adopted recycling technologies finalized to a higher recovery of concentrates?

The answer is in principle very simple, being related to the correct application of recycling technologies taking fully into account waste streams' physical-chemical characteristics. For example, the definition of proper comminution strategies addressed to obtain adequate size class distribution, minimizing the presence of fines and ultrafine particles, is obviously related to a correct knowledge of wastes. With reference to plastics, milling actions based on cutting represent the best solution; on the other hand, impact actions better realize comminution for glasses. Lack of knowledge of the composition of the waste materials to process, in terms of constituents and their time variation, can strongly impact the quality and quantity of the final recovered products. Waste materials present a high degree of compositional variability. Batch sampling, as usually performed on feeds and processed flow streams, does not allow performance of a full and continuous monitoring of the materials flow. Low-cost, reliable, and robust waste streams' physical-chemical characteristics detection devices, realizing continuous monitoring, could represent the solution to performing a full-time-independent-evaluation of waste materials streams handled in the plant.

Waste Derived Concentrates Recovery and Quality Assessment

Concentrate quality affects its value and, as consequence, the economic revenue of the recycling process. The possibility to realize a continuous full monitoring of concentrate characteristics is important: (1) to apply correct waste processing strategies to quantitatively and qualitatively maximize recovery, (2) to build a production data base embedding produced product characteristics, and (3) to perform a full products certification.

Recovery maximization is one of the key issues when recycling technologies are applied. Waste products are usually constituted by different materials of different characteristics. An optimal target could be represented to set up recycling actions addressed to separate and recover all the different waste constituents. In this case "the zero-waste" target should be fully reached. If successful, this strategy could allow re-integrating all the wastes in new production cycles. Such a goal is almost impossible to reach, but the strong scientific development, the related technological innovation and the larger attention of new generations towards environmental problems has and will continue to contribute to introduce new technologies for a more efficient recovery. To maintain a trace of when and what is produced in terms of WDPs is another key issue. The achievement of this goal allows establishment of a time correlation between waste feed and resulting concentrate. Correlations are useful not only in technical terms, for a better understanding of plant behavior with respect to waste feed variations, but also because they provide useful information about consumption and related waste production, contributing this way to better waste collection and handling strategies prior to the application of recycling technologies. Finally, product certification in the recycling sector sometimes represents a negative point. The proposed approach could strongly contribute to solving this problem, allowing at the same time a big step forward in product quality detection performed continuously and not on a batch basis.

Control Actions and Logics

As outlined in the previous paragraphs, one of the key points related to a systematic introduction of control logics, inside processing layouts, is to define simple, reliable, robust, efficient, and low-cost architectures able to perform a full characterization of the different flow streams in terms of waste feed and/or resulting product composition, that is, grade of the recovered materials and presence and characteristics of pollutants. The two aspects are intimately linked to the definition of suitable process-control-strategies and to the subsequent certification of the recovered materials.

How can control actions and logics be fully introduced and widely utilized in waste recycling?

Each material is characterized by specific attributes; these attributes are usually detected, with sensing devices able to collect one or more piece of information related to the characteristics of handled materials. Information is then processed following logics oriented to maximize the positive effects of the single and/or or group of actions. Control actions can be commonly categorized in four groups:
  • Feedback control: a control system that monitors its effect on the resulting product. On the basis of the collected information, tuning actions are applied on equipment, recycling plant sections, and operative variables in order to modify the output (product) accordingly.
  • Feedforward control: a control in which changes are detected at the process input (waste feed) and an anticipated correction signal is applied before process output (concentrate) is affected.
  • Cascade control: an automatic control system in which various control units are linked in sequence, each control unit regulating the operation of the next control unit in line.
  • Ratio control: a control procedure in which a predetermined ratio between two or more variables is maintained.

Feedback and feedforward controls are most commonly utilized in recycling.

Waste recycling flow streams are usually constituted by complex particulate solids systems. A particle is thus the minimum portion of material that can be processed. Each particle is characterized by different attributes: size, shape, composition, texture, etc. To define a control logic means identifying one or more rules to handle one or more of the previous mentioned attributes as they have been collected inside a particle flow stream, with the specific aim of verifying whether processing has produced an output satisfying the expected requirements of these attributes, both qualitatively and quantitatively, in the concentrate. In plastic recycling, for example, if a recovery of PP and/or PE has to be carried out, the presence of other plastics materials as PET or PVC, or other pollutants, as metals, glass, etc., must be avoided. In this case, the attribute composition plays a pre-eminent role in control logic definition. Actions to perform are related to: (1) the characteristics of equipment/s generating the PP or PE concentrate and (2) the parameters allowing their operation. Relationships existing between equipment/s operative conditions (processing parameters set up) and quality of the output, for a specific feed, have to be clearly investigated and formalized, constituting the basis for control logic implementation. A quantitative evaluation of the composition of PP and PE concentrate is thus fundamental to set up the logic in a quantitative way. In this case, as when recycling technologies have to be applied to other waste products, the quantitative collection of the attributes qualifying, or certifying, an intermediate and/or a concentrate product is not easy in terms of economically acceptable devices. This problem is common in waste recycling, where the value per unit of weight of product is usually low. To be economical, a process requires the processing of large quantities of waste and the production of a corresponding high amount of waste-derived concentrate. Often this condition does not match with the adoption of sensing devices that, to fulfill the previous requirements, are usually very expensive. A different approach has thus to be followed. If composition can be correlated with other parameter/s, easily detectable, a transposition logic can be applied: that is, the correlated property can be assumed as the control parameter and new control logics applied. Such an approach is particularly meaningful when on-line control logics have to be defined. The possibility to apply transposition logics sometimes produces a strong simplification in control. An example of what is described is reported in the following where different examples of sensing architectures based on H yper S pectral I maging (HSI) techniques are reported with reference to some of the materials taken into account in this section.

Plant Scale Control

The application of control actions and logics, on a recycling plant scale, can be considered a common practice. Originally performed, and in some cases also today applied, following a human senses-based approach, with the technological development "humans" have been and are being replaced by sensing devices. The first step to implement a plant scale control consists of the identification of some key points inside the plant where product characteristic detection can give useful information about the process. According to the values of the detected parameters, and utilizing pre-assigned rules, the equipment operative variables are modified accordingly. These actions can usually be performed with a feedback or a feedforward approach, more rarely adopting a cascade control. Approaches vary according to waste materials characteristics, control objectives, and implementation modalities: (1) introduction of control logics inside an existing plant or (2) in new ones. In this latter case, a large flexibility in control architecture design can be performed, not existing predefined plant architectural constraints.

Single Equipment Control

Following this approach, control actions and logics play at single equipment scale. Process and product parameters detection is carried out before (feedforward) and/or after (feedback) the single processing unit and control logics act accordingly. Such an approach is very promising, especially for future improvements in the recycling sector. The "intelligent processing machine," is equipment able to "understand" how it works, according to feed flow stream and resulting product characteristics, and able to modify its "behavior" accordingly, adopting pre-assigned and/or time-dependent learning rules. This is particularly relevant in recycling, where very often handled materials do not show constant compositional characteristics, affecting equipment performance.

Innovative Sensing Technologies in the Waste Sector

In recent years a lot of innovative sensing technologies and related control logics have been proposed in recycling . A sensing station is usually constituted by a conveying device (e.g., conveyor belt, vibrating channel, etc.) for the separation and steadying of the material, and a detection unit, positioned underneath or above the conveying device or at the material discharge area. Collected information can be then utilized to quantitatively assess material characteristics (certification) or to modify operative conditions of the equipment handling the materials before or after sensing (control) or to develop separation actions by actuators, e.g., valve bank blowing out materials constituents according to specific component characteristics (sorting). In the following, some sensing devices and their operative principles, together with possible application fields, are described.

Electronic Imaging (EI) Visible (VIS) Wavelength Based

EI-based sensing devices and algorithms [78] are the most widely used in recycling. They belong to the first class of control device utilized in this sector. EI application in recycling technologies comes from architectures developed in mineral processing and food control sectors. Such a technology moves from black & white (B&W) to color sensing systems. Actually color is widely applied in many recycling sectors: glass, WEEE, metals scraps, fluff, wood, etc. Color-based sensing is based on the collection and analysis of materials surface characteristics. Such an approach can represent sometimes a limitation because the collected information "only" relates to the surface (e.g., varnished objects in principle cannot be detected in a material-related way). In many cases, for the relatively low cost of this approach, EI will solve many control/sorting/quality assessment problems when the materials to check have passed shredding stages beforehand that remove the existing surface coatings or that break up the material in a way that with utmost probability uncoated fracture faces can be observed [79]. EI-based detection architectures are commonly constituted by an energizing source (lighting system), a CCD, matrix, or single array camera. Single array-based devices are particularly suitable to be utilized in waste material quality control because the detection principle (scan line) and the target of investigation (waste particles) moves towards each other with a constant speed. Waste particles can thus be fully investigated adopting different time-scale-related sampling strategies according to control/quality actions to apply.

Electronic Imaging (EI) near Infrared (NIR) Based

The principle that is at the basis of NIR technology is the measurement of object reflectivity within a wavelength ranging between 1.100 and 2.100 nm [80]. In this wavelength range, materials such as plastics, paper, and textiles are characterized by specific spectral firms allowing their recognition. This range of wavelengths is not visible to the human eye. This is also the reason that optical sorting systems must be used. The new generation of NIR-based detection devices can operate a good plastics distinction (e.g., PP, PS, PET, EPS, PC, or PVC) [81], as well as allow identification of materials such as paper, card, cardboard. or wood and natural fibers. A recognition limit of such a procedure is in the identification of materials that cannot be identified due to a lack of individual stone, porcelain, or dark materials, that, for the low level of reflectance, in the investigated wavelength range, do not allow recognition.

X- Ray Fluorescence Spectroscopy (XRF)

XRF is based on X-ray emission by a radioactive source or X-ray tube inside an XRF unit. X-rays emitted are absorbed by the atoms in the waste sample generating fluorescence as the atoms relax and release energy. The emitted wavelengths, as well as the energy released, are a function of the elements constituting the waste sample. Emission intensity is correlated to the content of a specific element within the sample. Such a technique was and is widely used in many recycling sectors as: metal/alloys, glass, plastics, wood (detection of arsenic, chromium, and copper in treated wood [82] and treated wood waste [83-85]), WEEE, waste-derived-fuels, etc.

Dual Energy X-Ray Transmission (DE-XRT)

DE-XRT is similar to that applied for luggage inspection and medical applications. Waste products are transported by a conveying belt. Waste material is energized, from the bottom, by X-rays. Transmitted radiation is collected by X-ray line detectors [86]. In order to separate the effects of density of the X-rayed object and the material thickness, the radiation intensity is generally measured in two different energy ranges. Following this approach thickness influence is eliminated and the radiation that passes through the material allows performance of materials recognition according to its density. The sensing approach allows classification of the material on volume basis. Different from EI or NIR techniques, surface characteristics (e.g., dust, water, small quantities of pollutants) do not practically influence recognition.

Laser-Induced Breakdown Spectroscopy (LIBS)

The system analyzes the optical spectrum, or fingerprint, of a small spot on each metal particle that is evaporated using powerful laser pulses. LIBS has been widely used as a diagnostic technique for the analysis of both surfaces and gaseous streams for metals [87, 88]. LIBS uses a high-power laser that is directed towards the sample through a series of mirrors and lenses that create a small micro-plasma on the surface of the targeted material. Due to the extremely high temperatures produced within the plasma (greater than 20,000°C) the atoms within the plasma emit light (or energy) characterized by different wavelengths. Certain wavelengths of energy are unique to different elements. The system then analyzes the optical spectrum, or fingerprint, of the micro-plasma collected and re-directed toward a fiber optic cable, which then feeds the signal to a spectrometer. The intensity of the emission is directly proportional to the amount of that element present in the sample. The costs and complexity of the system, as well as limitations in efficiency, are the main reason why LIBS is only applied in a number of very specialized operations, commonly for surface contamination. Applications have been developed with reference to wood waste contaminated with chromated copper arsenate sorting [89] and scrap metals [90].

Hyperspectral Imaging (HSI)

HSI, known also as chemical or spectroscopic imaging, is an emerging technique that combines the imaging properties of a digital camera with the spectroscopic properties of a spectrometer able to detect the spectral attributes of each pixel in an image. Thus, a hyperspectral image, is a three-dimensional dataset with two spatial dimensions and one spectral dimension.

HSI was originally developed for remote sensing applications [91], but has found a large utilization in such diverse fields as astronomy [92, 93, agriculture [94-96], pharmaceuticals [97-99], medicine [100, 101], and in recent years in the recycling sector [24, 102, 103], where important projects were also sustained by the European Union [104, 105].

Among the previous mentioned sensing techniques HSI can be considered one of the most interesting and subject to a wider and wider utilization inside the recycling sector. HSI, in fact, presents some advantages related to its intrinsic characteristics:
  • Continuous monitoring, with HSI-based devices as scan line cameras,
  • Utilization of different time-scale-related sampling strategies, according to the control/quality actions to develop,
  • Implementation of fast and reliable recognition logics, strongly linked to HSI detectors characteristics (e.g., possibility handle spectra as images),
  • Null environmental impact of the device,
  • Relatively low costs of the devices.

Furthermore, HSI devices, and related operative architectures can be easily integrated inside existing recycling plants, or implemented in new ones, with an optimal cost-benefit ratio. HSI, for its intrinsic properties, can thus be profitably utilized, both as a smart detection engine for sorting and as flow stream quality control, that is, certification of recovered materials and/or products. HSI is fast, accurate, affordable, and it can strongly contribute to lowering the economic threshold above which recycling is cost efficient. For its characteristics it can be meaningfully and reasonably developed, and applied, with reference to many solid waste-handling-sectors ranging from inorganic to organic waste sources.

Different cases studies describing the potentialities of HSI integration inside recycling technologies are reported in the following section.

Hyperspectral Imaging (HSI) Based Applications

An HSI system is typically constituted by optics, a spectrograph, a camera, an acquisition system, a translation stage, an energizing source (lighting device), and a control unit (PC). The camera, spectrograph, and illumination conditions determine the spectral range of the detection architecture. The sample/target is usually diffusely illuminated by a tungsten-halogen or LED source. A line of light reflected from the sample enters the objective lens and is separated into its component wavelengths by diffraction optics contained in the spectrograph. A two-dimensional image (spatial versus wavelength dimension) is then formed on the camera and saved on the computer. The sample is moved past the objective lens on a motorized stage and the process is repeated. Two-dimensional line images acquired at adjacent points on the object are stacked to form a three-dimensional hypercube that may be stored on a PC for further analysis.

The applications described in the following are based on sensing devices working in two different wavelength spectral ranges, from 400 to 1,000 nm (VIS-NIR range) and from 1,000 to 1,700 nm (NIR range). The first consists of a CCD camera, a line scan spectrograph (ImSpector™ V10E, SpecIm™, Finland), a lighting architecture, the spectrograph ImSpector™ V10E operates in the spectral range of 400-1,000 nm with a spectral resolution of 2.8 nm. The details of the acquisition architecture are reported in Table 15. The second is a SpecIm NIR spectral camera consisting of an ImSpector N17E imaging spectrograph for the wavelength region 1,000-17,000 nm and a temperature-stabilized InGaAs camera and a lighting architecture (Table 16).
Recycling Technologies. Table 15 Technical characteristics of the ImSpector™ V10E

Sensor

- 2/3″ CCD Array 780 × 580

- Firewire digital output

- Pixel resolution: 12 bit

Spectral range

400-1,000 nm

Spectral resolution

2.8 nm

Smile

< 1.5 μm

Keoneyst

< 1 μm

Entrance slit

30 μm × 14.2 μm

Image size

6.5 mm × 14.2 mm

Numerical aperture

F/2.4

Illuminant

- Anodized aluminum cylinder

- Barium sulfate internal coating

- d/O illumination and viewing conditions

- Adjustable height and distance

- 150 W cooled halogen lamp

- Stabilized power source

Recycling Technologies. Table 16 Technical characteristics of the ImSpector™ N17E

Sensor

- TE-cooled INGaAs photodiode array 640 × 512

- 14 bit, USB2, LVDS, CameraLink

Spectral range

900-1,000 nm ± 10 nm

Spectral resolution

2.6 nm

Spatial resolution

Rms spot radius < 15 μm

Aberrations

Insignificant astigmatism, smile or keystone

Effective slit length

12.8 mm

Numerical aperture

F/2.0

Stray light

< 0.5% (halogen lamp, 1,400 nm notch filter)

Both equipments are installed to perform the inspection of the waste materials on a laboratory scale conveyor belt (Fig. 2). The two devices are fully controlled by a PC unit equipped with the Spectral Scanner™ v.2.3 [106] acquisition/pre-processing software.
MediaObjects/978-1-4419-0851-3_16_Part_Fig2-116_HTML.gif
Recycling Technologies. Figure 2 Particulars of the architecture set-up utilized to perform a progressive and continuous waste samples spectra acquisition based on the ImSpector™ series devices

Sample Set Selection

The waste or derived products investigated belong to some of the classes of materials analyzed in this Section and characterized by different specific attributes, different sorting-selection problems, and quality requirements. From this perspective, waste glass fragments (cullet), light fraction derived from car shredding residues (fluff), and complex plastics-based waste streams have been selected. Results show that the HSI approach allows development and set up of strategies able to reduce analytical costs, improving the speed of the waste streams characteristics detection/analysis, and/or simplifying the procedures in terms of possible on-line implementation of fast and robust classification procedures oriented to develop innovative control strategies "human judgments and error free" as well as innovative certification criteria.

Spectra Acquisition and Detection Logics Implementation

Spectra related to the different investigated can be acquired adopting the acquisition architecture described in Fig. 2. Such a strategy is adopted because it mimics, at laboratory scale, the real behavior of the control architecture at an industrial scale, that is, the progressive and continuous horizontal translation of the sample and the "synchronized" acquisition of the spectra at a pre-established step, allowing a tuning of the detection/inspection frequency of the waste materials according to their characteristics. Analyses can thus be performed to verify the fulfillments of different goals:

First goal: the possibility to identify specific spectral attributes for each of the constituents of the different waste streams according to their intrinsic chemical-physical characteristics. Starting from this information, analysis can be carried out performing a characterization of the "shape" of the entire detected spectra and/or identifying, at specific wavelengths, peaks or valleys characterizing the detected spectral firm.

Second goal: the definition of fast, reliable, and robust recognition-classification procedures, based on different logics as: (1) spectral firms correlation, (2) single band intensity comparison at specific wavelengths, and (3) specific wavelengths intensity ratio analysis, in order to perform the discrimination of the different constituents inside a specific waste stream and to allow to reach a certification of the different products in terms of their composition.

Third goal: the possibility to perform a correlation among detected spectra, sample textural attributes, presence, characteristics, and localization of "pollutants": this latter aspect being of great interest to develop innovative sorting strategies. To validate the efficiency of the HSI-based technique to perform a topological assessment of the different materials an approach based on Principal Component Analysis (PCA) can be also adopted.

Case Studies

Tests reported are referred to different solid waste materials characterized, as previously outlined, by a different nature and physical-chemical attributes and "affected" by different processing-separation and/or control problems. For each investigated waste product, a synthetic overview of the target to reach by the HSI approach (Issues) and current status of the related utilized characterization approaches (State of the art) is reported.

Case Study No. 1: Ceramic Glass Identification Inside Waste Glass Products (cullets ) Issues: The amount of ceramic glass in post consumer glass waste stream is strongly increased in recent years, mainly for the introduction on the market of many ceramic glass manufactured goods [15]. These products constitute a new generation of consumer household goods used for their thermal-shock resistant properties. It can be argued that, considering the typology of ceramic glass products, contamination involves both the main production lines of glass recycling plants , e.g., the flat glass cullet and the container glass cullet. Due to its physical properties, similar to those of glass, ceramic glasses are almost impossible to detect adopting the automated optical technologies, commonly utilized for cullet color sorting [107]. Therefore, identifying and removing ceramic glass from the glass waste stream has long been a challenge for recyclers of glass.
State of the art: The two actions usually carried out to realize transparent ceramic contaminant removal are source reduction and manual sorting. Therefore, there is the need for the development and the implementation of a system able to realize a real-time identification of ceramic glass in the cullet stream. Sorting techniques based on X-ray [79] and FT-IR spectroscopy [24] have been proposed as methods for ceramic glass detection, but both are expensive and difficult to implement for safety and efficiency reasons, respectively.
The HSI approach clearly has potential in developing innovative sorting strategies. In the visible range (400-700) nm the technique presents a high discrimination power for classical cullet separation by color, allowing the possibility to decrease the minimum size of sorted glass particles, on the other hand, in the visible range the recognition of glass from ceramic glass is almost impossible. Moving in the wavelength range between (700-1,000) nm and (1,000-1,700) nm it is possible the architecture allows discrimination between glass (Fig. 3) and ceramic glass fragments (Fig. 4). Such a discrimination is relatively easy for clear samples; darker samples are more difficult to discriminate. Color, in fact, seems to affect the reflectance levels. Amber glasses show reflectance values near zero, green glasses present intermediate reflectance values, whereas clear glasses display higher reflectance values. A two-steps sorting, a first based on HSI acting between (700-1,700) nm and the second in the (700-1,000) nm range, could thus allow good preliminary cullet sorting targeted to ceramic glass contaminant identification/removal and a further cullet color-based separation. Application of this technique, in the waste glass sector, can thus allow development of innovative sorting logics: (1) specific attributes can be associated to particulate solid materials, (2) classification procedures based on wavelengths intensity ratio can be defined to perform sorting, and (3) the elements to sort can be topologically assessed adopting a Principal Component Analysis (PCA) [108]. An example related to this last point is outlined in Fig. 5. The HSI devices for their hardware architecture can be easily installed both to integrate and/or to substitute classical sorting imaging-based devices.
MediaObjects/978-1-4419-0851-3_16_Part_Fig3-116_HTML.gif
Recycling Technologies. Figure 3 Spectral plots in the VIS-NIR field (400-1,000 nm) of glass samples
MediaObjects/978-1-4419-0851-3_16_Part_Fig4-116_HTML.gif
Recycling Technologies. Figure 4 Spectral plots in the VIS-NIR field (400-1,000 nm) of ceramic glass samples
MediaObjects/978-1-4419-0851-3_16_Part_Fig5-116_HTML.gif
Recycling Technologies. Figure 5 Representation of the HSI data set, as acquired (a), of different cullets resulting from a recycling process. The fragment on the right end side of the image is ceramic glass. (b) Corresponding false color image embedding the results of all the three score plots, (image of scores on PC1: 97.46, image of scores on PC2: 1.82 and image of scores on PC3: 0.43) related to PC1, PC2, and PC3 components as resulting from the application of the PCA. Contaminant (ceramic glass) can be easily identified
Case Study No. 2: Light Fraction Resulting from Car Dismantling (Fluff) Characterization Issues: Fluff represents about 25% of the weight of a car and is usually constituted by materials characterized by intrinsic low specific gravity (i.e., plastics, rubber, synthetic foams, etc.). When processed to perform its recovery, fluff is polluted by materials presenting higher specific gravity (i.e., copper, aluminum, brass, iron, etc.), constituting parts of the electrical devices of the vehicle that, for their shape, size (i.e., wires, metal straps, slip rings, wipers, etc.) and utilization remain concentrated in the lighter products. Such "polluting materials," for their intrinsic characteristics, are not well removed by classical separation techniques.
State of the art: Fluff is usually produced after different comminution-classification stages. The final classification is usually carried out by cycloning or venting (air suction or blowing systems), in order to separate the light material. A good separation could contribute to reducing the waste disposal and environmental pollution, and increasing the energy recovery through pure sorted polymer re-use. Furthermore, the possibility to utilize finer fluff fractions to produce energy could dramatically contribute to increase the full reutilization of such products. To reach this goal the quantity and the quality of the metal contaminants have to be strongly controlled in order to not prejudice the quality of the final fluff based fuel. The need to develop both efficient selection and control strategies to obtain contaminant-free, or almost free, fine fluff products thus assumes a fundamental role in all the processing and control stages of the recycling chain.
Results show the HSI proposed architecture allows identification of all fluff constituents and polluting materials (Fig. 6). The discriminating power is high in the region between (400-1,000) nm, this is a quite important result being thus possible to utilize, for fluff inspection, a single device (ImSpector™ V10E) sorting/control unit, with considerable costs reduction. From this perspective, the HSI approach can be profitably applied utilizing "simple" band intensity comparison, at specific wavelengths, among unknown fluff particles and a reference library of spectra, previously built, embedding the different waste materials spectra constituting the light fractions to investigate. The approach is particularly flexible to use because fluff of different origin and/or composition can be sorted/controlled adopting the same logic, but working on different specifically spectra library, built according to new materials characteristics to detect.
MediaObjects/978-1-4419-0851-3_16_Part_Fig6-116_HTML.gif
Recycling Technologies. Figure 6 Spectral plots in the VIS-NIR field (400-1,000 nm) of fluff materials. The HSI approach clearly allows discrimination between the different light materials constituting fluff
Case Study No. 3: Polyolefins Recognition-Separation Issues: Polyolefin recovery from complex waste streams is a challenging issue that has not yet been profitably solved. Furthermore, polypropylene (PP), high density polyethylene (HDPE), and low density polyethylene (LDPE) together are both difficult to separate and chemically incompatible. In order to produce high-purity granulates from these concentrates, of a quality comparable to materials produced from post-industrial waste, the mixture must be sorted very accurately, and in order to be economically and ecologically sound, most of the polyolefins should end up in a useful product. Such accurate and efficient separations exist, but they involve multiple separations. They are therefore expensive, difficult to control, and often do not allow the production of good concentrates. The possibility to develop efficient and low-cost recognition logics to control the process and certify the products thus represents a key issue in polyolefin recovery .
State of the art: Innovative processing plant layouts have been realized to process complex plastic-based wastes by shredding and sink-floating to produce polyolefin concentrates of varying quality. Analysis of such concentrates generally shows a mixture of polyolefins, rubbers, foams, fibers, and wood, next to varying amounts of materials heavier than water. Currently available separation techniques, based on the difference in flotation proprieties in water, can be used to separate lighter types of plastic such as PP, HDPE, and LDPE from the heavier types such as polyethylene terephthalate (PET) and polyvinyl chloride (PVC) . A known method is to separate the mixtures into five fractions using separation media with densities of 880, 920-930, 940, and 970 kg/m3. Such a procedure will create high-purity PP and HDPE products, whereas foams, most of the wood and rubbers, LDPE, filled PP, and residual heavy materials will end up in relatively small residue fractions. Certain expensive liquids have been specially designed to separate at one of the target densities for polyolefin recycling. Other technologies, such as electrostatic separation and thermal adhesion, have been able to create only a single relatively pure product.
Analyses show PP and PE recognition is strongly influenced, in the visible region, by the color of the samples. Based on the analysis of HSI spectra, it is evident the possibility to define specific parameters useful for recognition of the two polyolefins, as, for example, the slope of the spectrum in a selected wavelength range or a band ratio among two different wavelengths. Analyses show PP and PE spectra present significant differences in regions around 750 nm and (900-950) nm (Fig. 7). It is important to note that the best and most precise recognition logic, especially for polluting particles detection and their topological assessment inside plastic waste feed, should be based on more sophisticated and complex statistical analyses, such as (PCA) (Fig. 8), Partial Least Square (PLS), Neural Network (NN), etc., that usually require long computation time. Considering that in most industrial applications the fast response of the detecting/sorting device is one of the main constraints, as, for example, when particles are moving on a conveyor belt and they are sorted on-line, the adoption of simplified logics, working properly, is preferred. This latter approach (multiple bands intensity comparison at specific wavelengths) can be profitably applied for recovered polyolefins quality control.
MediaObjects/978-1-4419-0851-3_16_Part_Fig7-116_HTML.gif
Recycling Technologies. Figure 7 Spectral plots in the VIS-NIR field (400-1,000 nm) of polyolefins: PP (a) and PE (b)
MediaObjects/978-1-4419-0851-3_16_Part_Fig8-116_HTML.gif
Recycling Technologies. Figure 8 Image on scores on PC2:0.81 (b) of the HSI data set of (a). Samples/Scores Plot PC1-PC2 clearly outlines the correspondence existing among typology of contaminants and their "mapping" inside the plot (c)
The examples clearly outline the importance of HSI in recycling technologies as an innovative, flexible, and low-cost tool that, combining imaging and reflectance spectroscopy, can profitably allow performance of both waste feed and recovered products characterization, control, and certification. HSI-based technology through the detection of the spectral signature of waste and waste-derived products, of different nature and composition, allows to extract and quantify those physical-chemical attributes influencing their characteristics and behavior. Results demonstrated as the proposed technique, and the related recognition logics, will have a greater impact on the development of recycling technologies finalized to implement on-line innovative sorting strategies, as well as new control procedures. The possibility to reach a primary goal in recycling, that is a full control, at a low cost, of the quality of the different flow streams inside the plant, according to the different processing stages, can strongly contribute to develop innovative-inside-processing products certification.

Bibliography

  • Primary Literature
  • 1. Kahn CH (1979) The art and thought of Heraclitus. Cambridge University Press, Cambridge
  • 2. Mamedbeii GD (1959) Muhammed Nasir al-Din al-Tusi on the theory of parallel lines and the theory of ratios. (Azerbaijani), Izdat. Akad. Nauk Azerbaijzansk. SSR (Baku)
  • 3. Dumas M (1955) Lavoisier, théoricien et expérimentateur. Presses Universitaires de France, Paris
  • 4. Heijungs R, Huppes G, Guinée JB (2010) Life cycle assessment and sustainability analysis of products, materials and technologies. Toward a scientific framework for sustainability life cycle analysis. Polym Degrad Stab 95(3):422-428
  • 5. Bartl A, Hackl A, Mihalyi B, Wistuba M, Marini I (2005) Recycling of fibre materials. Process Safety Environ Protect 8(B4):351-358
  • 6. Confederation of European Paper Industries (2009) CEPI sustainability report. Brussels, Belgium
  • 7. Ochoa JAG (2008) Feasibility of recycling pulp and paper mill sludge in the paper and board industries. Resour Conserv Recycl 52(7):965-972
  • 8. Wiegand PS, Unwin JP (1994) Alternative management of pulp and paper industry solid wastes. Tappi J 77:91-97
  • 9. Wolfer EP, Venkat WB, Maroju BV, Martiny A (1997) Method for recovering fiber from effluent streams. U.S. Patent 5, pp 593-542
  • 10. Saint Amand FJ (1999) Hydrodynamics of deinking flotation. Int J Miner Process 56:277-316
  • 11. Tandy S, Healey JR, Nason MA, Williamson JC, Jones DL (2009) Heavy metal fractionation during the co-composting of bio-solids, deinking paper fibre and green waste. Bioresour Technol 100(18):4220-4226
  • 12. Moo-Young HK Jr, Zimmie TF (1997) Waste minimization and re-use of paper sludges in landfill covers: a case study. Waste Manag Res 15(6):593-605
  • 13. Werther J, Ogada T (1999) Sewage sludge combustion. Prog Ener Combust Sci 25(1):55-116
  • 14. Safeglass (Europe) Limited, Nasmyth Building, Nasmyth Avenue. East Kilbride. UK G75 0Q. http://www.breakglass.org/Glass_making.html
  • 15. Höland W, Beall G (2002) Glass-ceramic technology. The American Ceramic Society, Westerville, p 372
  • 16. Pannhorst W (1997) Glass ceramics: state of the art. J Non-Cryst Solids 219:198-204
  • 17. Rem PC (1999) Eddy current separation. Delft University of Technology, Delft
  • 18. Bonifazi G, D'Addetta A, Massacci P (2002) Classification by neural net of a particle stream in an eddy-current drum separator. Int J Part Part Syst Charact 19:96-102
  • 19. Jong TPR, de Dalmijn WL (2002) X-ray transmission imaging for process optimisation of solid resources. In: R02, 6th World congress on integrated resources management, Geneva, Switzerland, CD-Paper 173
  • 20. Bonifazi G (2000) Imaging based sorting logic in solid waste recycling. In: Proceedings of the 16th international conference on solid waste technology and management, vol 6, Philadelphia, USA, pp 14-26
  • 21. Bonifazi G, Massacci P (2000) Cullets (glass fragments) quality control by artificial vision: a textural based approach. In: 4th World Congress R00 - Recovery, recycling, re-integration, Toronto, Canada, CD-Paper 31, pp 723-728
  • 22. Bonifazi G, Massacci P (1998) Cullets (glass fragments) quality control by artificial vision: a color based approach. In: Proceedings of international conference on quality control by artificial vision, Takamatsu, Japan, pp 94-99
  • 23. Serranti S, Bonifazi G, Pohl R (2006) Spectral cullet classification in the mid-infrared field for ceramic glass contaminants detection. Int J Waste Manag Res 24:48-59
  • 24. Bonifazi G, Serranti S (2006) Imaging spectroscopy based strategies for ceramic glass contaminants removal in glass recycling. Int J Waste Manag 26:627-639
  • 25. Cramb AW (1996) A short history of metals. Dept. of Materials Science and Engineering. Carnegie Mellon University. http://neon.mems.cmu.edu/cramb/Processing/history.html
  • 26. Gascoigne B (2001) History of metallurgy. HistoryWorld. Ongoing. http://www.historyworld.net/wrldhis/PlainTextHistories.asp?historyid=ab16
  • 27. Alter H (1977) Magnetic separation - Recovery of salable iron and steel from municipal solid waste. Environmental Protection Agency, Cincinnati
  • 28. Shapiro M, Galperin V (2005) Air classification of solid particles: a review. Chem Engin Process 44:279-285
  • 29. Wills BA (1997) Mineral processing technology, 6th edn. Butterworth-Heinmann, Boston, pp 232
  • 30. Bradley D (1965) The hydrocyclone. Pergamon, New York
  • 31. Takoungsakdakun T, Pongstabodee S (2007) Separation of mixed post-consumer PET-POM-PVC plastic waste using selective flotation. Sep PurifTechnol 54:248-252
  • 32. Buchan R, Yarar B (1995) Recovering plastics for recycling by mineral processing techniques. J Miner Met Mater Soc 47:52-55
  • 33. Drelich J, Kim JH, Payne T, Miller JD, Kobler RW (1999) Purification of polyethylene terephthalate from polyvinyl chloride by froth flotation for the plastics (soft-drink bottle) recycling industry. Sep Purif Technol 15:9-17
  • 34. Kang H, Schoenung JM (2005) Electronic waste recycling: a review of US infrastructure and technology options. Resour Conserv Recycl 45(4):368-400
  • 35. Veit HM, Pereira C, Bernardes AM (2002) Using mechanical processing in recycling printed wiring board. J Miner Met Mater Soc 54(6):45-47
  • 36. Askvik KM, Hetlesæther S, Sjobölm J, Stenius S (2001) Properties of the lignosulfonate-surfactant complex phase. Colloids Surf A Physicochem Eng Aspects 182:178-189
  • 37. Singh BP (1998) Wetting mechanism in the flotation separation of plastics. Filtration Sep 35:525-527
  • 38. Shen H, Pugh RJ, Forssberg E (2002) Floatability, selectivity and flotation separation of plastics by using a surfactant. Colloids Surf A Physicochem Eng Aspects 196:63-70
  • 39. Fraunholcz N (2004) Separation of waste plastics by froth flotation - a review. Part I, Miner Engin 17:261-268
  • 40. Andrady AL (2003) Plastics and the environment. Wiley, Hoboken, pp 792
  • 41. Al-Salem SM, Lettieri P, Baeyens J (2009) Thermal pyrolysis of high density polyethylene (HDPE). In: Proceedings of the 9th European gasification conference: clean energy and chemicals, Düsseldorf, Germany
  • 42. Scheirs J (1998) Polymer recycling: science, technology and application, 1st edn. Wiley-Blackwell, New York
  • 43. Dirks E (1996) Energy recovery from plastic waste in waste incineration plants. In: Brandrup J, Bittner M, Menges G, Michaeli W (eds) Recycling and recovery of plastics, 1st edn. Hanser, Munich, pp 746-769
  • 44. Zia KM, Bhatti HN, Bhatti IA (2007) Methods for polyurethane and polyurethane composites, recycling and recovery: a review. React Funct Polym 67(8):675-692
  • 45. Hawn K (2001) An overview of commercial recycling technologies and textile applications for the products. In: 6th annual conference on recycling of polymer, textile and carpet waste, Dalton, USA
  • 46. Cupit MJ (1996) Opportunities and barriers to textile recycling, AEA Technology, Report 0113, Oxfordshire, UK
  • 47. Passas R, Voillot C, Tarrajat G, Caucal G, Khelifi B, Tourtollet G (2001) Morfi as a novel technology for morphological analysis of fibers. Recents Progres en Genie des Procedes 15:259-264
  • 48. Bartl A, Mihalyi B, Marini I (2004) Applications of renewable fibrous materials. Chem Biochem Engin 18:21-28
  • 49. Wang Y (1995) Reuse of carpet industrial waste for concrete reinforcement. In: RILEM Proceeding (Disposal and recycling of organic and polymeric construction materials), vol 27, London, pp 297-306
  • 50. Wang Y (1997) Properties of concrete reinforced with recycled carpet waste fibers. In: Proceedings of International symposium on brittle matrix composites 5, Warsaw, pp 179-186
  • 51. Wang Y (1999a) Ecotextile'98: sustainable development. In: Proceedings of the Conference, Bolton, pp 165-171
  • 52. Wang Y (1999) Utilization of recycled carpet waste fibers for reinforcement of concrete and soil. Polym Plast Technol Engin 38:533-546
  • 53. Wang Y (2002) Recycling of automotive fibers. In: Proceedings of Joint INDA-TAPPI Conference, Atlanta, pp 160-167
  • 54. Bohnhoff A, Petershans J (2002) De-centralised technology for the sorting of textile floor coverings. In: 7th annual conference on recycling of polymer, textile and carpet waste, Dalton, USA
  • 55. Strzelecki C (2004) Modern solutions for shredding, grinding and re-pelletizing post-industrial fiber, nonwovens and carpet scrap. In: Annual conference on recycling of polymer, textile and carpet waste, Dalton, USA
  • 56. Bacon FC, Holland WR, Holland LH (1998) Method and machine for recycling discarded carpets. U. S. Patent 5, 704, 104
  • 57. Howe MA, White SH, Locklear SG (2001) Method and apparatus for reclaiming carpet components. US Patent 6, 182, 913
  • 58. Herlihy J (1997) Recycling in the carpet industry. Carpet and Rug Industry, pp 17-25
  • 59. Kasserra P (1998) Recycling of polyamide 6.6 and 6. In: Prasad PN et al (eds) Science and technology of polymers and advanced materials. Plenum, New York, pp 629-635
  • 60. Hagguist JAE, Hume RM (1993) Carpet reclaimer, U.S. Patent 5, 230, 473
  • 61. Schut JH (1995) Big plans for carpet. Plast World 53:25
  • 62. Booij M, Hendrix JAJ, Frentzen YH (1997) Process for recycling polyamide-containing carpet waste, European Patent 759, 456
  • 63. Frentzen YH, Thijert MP, Zwart RL (1997) Process for the recovery of caprolactam from waste containing nylon by extraction with alkyl phenol, World Patent 97, 03, 04
  • 64. Sarian AK, Handerman AA, Jones S, Davis EA, Adbye A (1998) Recovery of polyamide from composite articles, U.S. Patent 5, 849, 80
  • 65. Sikorski ME (1993) Recycling of polymeric materials from carpets and other multi-component structures by means of supercritical fluid extraction, U.S. Patent 5, 233, 021
  • 66. Griffith AT, Park Y, Roberts CB (1999) Separation and recovery of nylon form carpet waste using a supercritical fluid antisolvent technique. Polym Plast Technol Engin 38(3):411-432
  • 67. Honeywell Nylon Inc (2005) http://www.infinitynylon.com
  • 68. Elam CC, Evan RJ, Czernik S (1997) An Integrated approach to the recovery of fuels and chemicals from mixed waste carpets through thermocatalytic processing, Preprint papers -American Chemical Society. Div Fuel Chem 42(4):993-997
  • 69. Bajaj P, Sharma ND (1997) In: Gupta VB, Kothari VK (eds) Reuse of polymer and fibre waste in manufactured fibre technology. Chapman & Hall, New York, pp 615
  • 70. Brown T (2001) Infinity nylon - a never-ending cycle of renewal, 6th annual conference on Recycling of polymer, Textile and Carpet Waste, Dalton, GE, USA. http://hdl.handle.net/1853/10385
  • 71. Schut JH (1993) A recycling first: carpets! Plast Technol :22-25
  • 72. Young D, Chlystek S, Malloy R, Rios I (1998) Recycling of carpet scrap, U.S. Patent 5,852,115
  • 73. Hagberg CG, Dickerson JL (1997) Recycling nylon carpet via reactive extrusion. Plast Engin 53:41-43
  • 74. Datta RJ, Polk MB, Kumar S (1995) Reactive compatibilization of polypropylene and nylon. Polym Plast Technol Engin 34(4):551-560
  • 75. Dagli SS, Xanthos M, Biesenberger JA (1992) Blends of nylon 6 and polypropylene with potential applications in recycling, effects of reactive extrusion variables on blend characteristics. ACS Symp Ser 513:241-257
  • 76. David DJ, Dickerson JL, Sincock TF (1994) Thermoplastic composition and method for producing thermoplastic composition by melt blending carpet, U. S. Patent 5, 294, 384
  • 77. Muzzy J, Wang Y, Hagberg C, Patel P, Jin K, Samanta S, Bryson L, Shaw B (2004) Long fiber reinforced post-consumer carpet. In: ANTEC 2004, Annual Technical Conference of the society of plastics engineers, Chicago, USA
  • 78. Jähne B (1993) Digital image processing; concepts, algorithms, and scientific applications, 2nd edn. Springer, Berlin
  • 79. de Kattentidt HUR, Jong TPR, Dalmijn WL (2003) Multi-sensor identification and sorting of bulk solids. Control Engin Pract 11:4147
  • 80. Bearmann GH, Levenson RM, Cabib D (eds) (2002) Spectral imaging: basic principles and prospective applications. Kluwer, Dordrecht
  • 81. Leitner R, Mairer H, Kercek A (2003) Real-time classification of polymers with NIR spectral imaging and blob analysis. Real-Time Imag 9:245-251
  • 82. American Wood Preservers' Association (AWPA) (1999) American wood preservers' association book of standards. American Wood Preservers' Association, Grandbury
  • 83. Blassino M, Solo-Gabriele HM, Townsend T (2002) Pilot scale evaluation of sorting technologies for CCA treated wood waste. Waste Manag Res 20:290-301
  • 84. Kormienko M (1999) Sorting technologies for CCA-treated wood waste. Master of Science Thesis, University of Miami, Coral Gables, USA
  • 85. Solo-Gabriele H, Townsend T, Kormienko M, Stook K, Gary K, Tolaymat T (2000) Alternative chemicals and improved disposal-end management practices for CCA-treated wood. Final Technical Report #00-03. Florida center for solid and hazardous waste management, Gainesville, USA
  • 86. de Jong TPR, Dalmijn WL (2002) X-ray transmission imaging for process optimisation of solid resources. In: Proceedings of R'02 congress, Geneva, pp 1-6
  • 87. Hahn DW, Flower WL, Hencken KR (1997) Discrete particle detection and metal emissions monitoring using laser-induced breakdown spectroscopy. Appl Spectrosc 51:1836-1844
  • 88. Hahn DW (1998) Laser-induced breakdown spectroscopy for sizing and elemental analysis of discrete aerosol particles. Appl Phys Lett 72:2960-2962
  • 89. Radziemski LJ, Cremers DA (1989) Laser-induced plasmas and applications. Marcel Dekker, New York
  • 90. de Mesina MB, Jong TPR, Dalmijn WL (2007) Automatic sorting of scrap metals with a combined electromagnetic and dual energy X-ray transmission sensor. Int J Miner Process 82:222-232
  • 91. Goetz AFH, Vane G, Solomon TE, Rock BN (1985) Imaging spectrometry for earth remote sensing. Science 228:1147-1153
  • 92. Hege E, O'Connell D, Johnson W, Basty S, Dereniak E (2003) Hyperspectral imaging for astronomy and space surveillance. Proceedings of the SPIE 5159:380-391
  • 93. Wood KS, Gulian AM, Fritz GG, Van Vechten D (2002) A QVD detector for focal plane hyperspectral imaging in astronomy. Bull Am Astron Soc 34:1241
  • 94. Monteiro S, Minekawa Y, Kosugi Y, Akazawa T, Oda K (2007) Prediction of sweetness and amino acid content in soybean crops from hyperspectral imagery. ISPRS J Photogram Remote Sens 62(1):2-12
  • 95. Smail V, Fritz A, Wetzel D (2006) Chemical imaging of intact seeds with NIR focal plane array assists plant breeding. Vibrational Spectroscopy 42(2):215-221
  • 96. Lyon RC, Lester DS, Lewis EN, Lee E, Yu LX, Jefferson EH (2002) Near-infrared spectral imaging for quality assurance of pharmaceutical products: analysis of tablets to assess powder blend homogeneity. AAPS Pharm Sci Tech 3(3):17
  • 97. Rodionova O, Houmøller L, Pomerantsev A, Geladi P, Burger J, Dorofeyev V (2005) NIR spectrometry for counterfeit drug detection: a feasibility study. Anal Chim Acta 549(1-2):151-158
  • 98. Roggo Y, Edmond A, Chalus P, Ulmschneider M (2005) Infrared hyperspectral imaging for qualitative analysis of pharmaceutical solid forms. Anal Chim Acta 535(1-2):79-87
  • 99. Ferris D, Lawhead R, Dickman E, Holtzapple N, Miller J, Grogan S (2001) Multimodal hyperspectral imaging for the non invasive diagnosis of cervical neoplasia. J Low Genit Tract Dis 5(2):65-72
  • 100. Kellicut D, Weiswasser J, Arora S, Freeman J, Lew R, Shuman C (2004) Emerging technology: hyperspectral imaging. Perspect Vasc Surg Endovasc Ther 16(1):53-57
  • 101. Zheng G, Chen Y, Intes X, Chance B, Glickson JD (2004) Contrast-enhanced near-infrared (NIR) optical imaging for subsurface cancer detection. J Porphyrins Phthalocyanines 8(9):1106-1117
  • 102. Serranti S, Bonifazi G (2009) Hyperspectral imaging detection architectures for polyethilene (PE) and polypropylene (PP) identification inside plastic waste streams. In: Proceedings of waste-to-resources, III International symposium MBT&MRF. Hanover, Germany, pp 463-474
  • 103. Serranti S, Bonifazi G, Bonoli A, Dall'Ara A (2009) Composting products quality assessment and monitoring by hyperspectral imaging based logics. In: Proceedings of waste-to-resources, III International symposium MBT&MRF. Hanover, Germany, pp 584-597
  • 104. W2Plastics (2008) Collaborative Project 212782 - FP7-ENV-2007-1: magnetic sorting and ultrasound sensor technologies for production of high purity secondary polyolefins from waste
  • 105. HYSPIMGLASS (2002) CRAFT Programme: CRAF-1999-71817: development of a Novel and high speed spectral imaging system to detect glass-like contaminants in the recyclable, cost-effectively increasing glass recycling and avoiding landfilling
  • 106. SSOM (2008) Spectral scanner operative manual (Version 2.0). DV Optics S.r.l., Italy http://www.dvoptic.com/index.html
  • 107. Bonifazi G (2000) Imaging based sorting logic in solid waste recycling. In: The Sixteenth international conference on solid waste technology and management - session 6A: recycling and source reduction. Philadelphia, USA, pp 6.14-26
  • 108. Geladi P, Isaksson H, Lindqvist L, Wold S, Esbensen K (1989) Principal components analysis of multivariate images. Chemometr Intell Lab Syst 5(3):209-220
  • Books and Reviews
  • An good paper presenting an excellent review of recent progress in the recycling and recovery of Plastic Solid Waste (PSW), with particular emphasis "on waste generated from polyolefinic sources, which makes up a great percentage of our daily single-life cycle plastic products" is: Al-Salem SM, Lettieri P, Baeyens J (2009) Recycling and recovery routes of plastic solid waste (PSW): a review. Waste Manag 29:2625-2643
  • Beede DN, Bloom DE (1995) Economics of the generation and management of MSW. NBER Working Papers 5116. National Bureau of Economic Research, Inc, Cambridge, MA
  • Caputo AC, Pelagagge PM (2001) Waste-to-energy plant for paper industry sludges disposal: technical-economic study. J Hazard Mater 81(3):265-283
  • Cofie O, Kone D, Rothenberger S, Moser D, Zubruegg C (2009) Co-composting of faecal sludge and organic solid waste for agriculture: process dynamics. Water Res 43(18):4665-4675
  • El Haggar S (2007) Sustainable industrial design and waste management: cradle-to-cradle for sustainable development. Academic, St. Louis, pp 424
  • Galperin V, Shapiro M (1999) Separation of solid particles in a fluidized bed air classifier. Powder Handling Process 11:2
  • Gesing A, Berry L, Dalton R, Wolanski R (2002) Assuring continued recyclability of automotive aluminium alloys: grouping of wrought alloys by color, X-ray absorption and chemical composition-based sorting. In: Proceedings annual meeting on automotive alloys and aluminium sheet and plate rolling and finishing technology, Seattle, USA
  • Gesing A, Steward C, Wolanski R, Dalton R, Berry R (2000) Scrap preparation for aluminium alloy sorting. In: Proceedings TMS fall extraction and process metallurgy meeting, Pittsburgh, USA
  • Hosokawa Micron Group (2011) http://www.hmicronpowder.com/application/classification
  • Huth-Fehre Th, van den Broek W (1995) NIR-Remote sensing and artificial neural networks for rapid identification of post consumer plastics. J Mol Struct 348:143-146
  • Johansson JE (2007) Plastics - the compelling facts and figures. 6th IdentiPlast Biennial Conference on the Recycling and Recovery of Plastics. Brussels, Belgium
  • Kunii D, Levenspiel O (1991) Fluidization engineering, 2nd edn. Butterworth, Heinmann, pp 233
  • Marques GA, Tenorio JAS (2000) Use of froth flotation to separate PVC/PET mixtures. Waste Manag 20:265-269
  • Méndez A, Fidalgo JM, Guerrero F, Gascó G (2009) Characterization and pyrolysis behaviour of different paper mill waste materials. J Anal Appl Pyrolysis 86(1):66-73
  • Oshitani J, Kiyoshima K, Tanaka Z (2003) Continuous dry material separation from automobile shredder residue. Kagaku Kogaku Ronbunshu 29:8-14
  • Pascoe RD (2005) The use of selective depressants for the separation of ABS and HIPS by froth flotation. Miner Eng 18:233-237
  • Sekito T, Matsuto T, Tanaka N (2006) Application of a gas-solid fluidized bed separator for shredded municipal bulky solid waste separation. Waste Manag 26:1422-1429
  • Sekito T, Tanaka N, Matsuto T (2006) Batch separation of shredded bulky waste by gas-solid fluidized bed at laboratory scale. Waste Manag 26:1246-1252
  • Svoboda J (2004) Magnetic techniques for the treatment of materials, Kluver Academic Publisher, New York, USA. pp. 656. http://www.springer.com/earth+sciences+and+geography/book/978-1-4020-2038-4
  • Van Nieuwenhuijzen A, Van der Graaf J (2010) Handbook on particle separation processes. IWA, London, UK, pp 400
  • World Bank (2007) Environmental, health, and safety guidelines for pulp and paper mills. Draft technical document. Environment and Social Development Department, International Finance Corporation, Washington, DC
  • Yoshida M, Oshitani J, Kaname K, Gotoh K (2006) Fluidized bed medium separation (FBMS) of Cl-containing plastics in home electric appliance shredder residue. Kagaku Kogaku Ronbunshu 32:115-121