Supply: Glass Constructions & Engineering

Glass Struct Eng 5, 445–487 (2020).


Telesilla Bristogianni
Faidra Oikonomopoulou
Rong Yu
Fred A. Veer
Rob Nijsse


Presently, tons of top of the range business glass are down-cycled or landfilled resulting from contaminants that stop close-loop recycling. But, this glass is probably a invaluable useful resource for casting sturdy and aesthetically distinctive constructing elements. Exploring the potential of this concept, several types of non-recyclable silicate glasses are kiln-cast into 30 × 30 × 240 mm beams, at comparatively low temperatures (820–1120 ◦C). The defects occurring within the glass specimens resulting from cullet contamination and the excessive viscosity of the glass soften, are documented and correlated to the casting parameters. Then, the kiln-cast specimens and industrially manufactured reference beams are examined in four-point bending, acquiring a flexural power vary of 9–72MPa.

The outcomes are analysed in keeping with the position of the chemical composition, stage of contamination and adopted casting parameters, in figuring out the flexural power, the Younger’s modulus and the prevailing strength-limiting flaw. Chemical compositions of beneficial efficiency are highlighted, in order essential flaws answerable for a dramatic lower in power, as much as 75%. The defects located within the glass bulk, nonetheless, are tolerated by the glass community and have minor impression on flexural power and Younger’s modulus. The conditions for good high quality recycled solid glass constructing elements are recognized and a top level view for future analysis is offered.

1 Introduction

The nice potential of glass casting know-how for the constructing trade is thus far little explored by structural engineers and designers, however are steadily getting found after the success of all cast-glass load bearing constructions such because the Crystal Homes façade in Amsterdam (Oikonomopoulou et al. 2018c). The 3D-shaping potentialities offered by casting can provide sturdy glass elements of bigger cross-sections and a greater diversity of varieties and hues than at the moment obtainable by different glass processing strategies. Parallel to the popularity of the structural and aesthetical strengths of solid glass elements, questions come up concerning their environmental impression and life-cycle.

Using – at the moment not recyclable-disposed glass as a uncooked supply for glass casting at decrease temperatures, is a promising concept that addresses each the urgent drawback of glass waste, and the urgency to decrease the carbon footprint of glass constructing elements (Bristogianni et al. 2018; Oikonomopoulou et al. 2018b). To specify the time period “at the moment not recyclable glass”, other than the profitable recycling of soda lime glass meals and beverage containers, the remainder of the discarded-often excessive qualitycommercial glass hardly ever meets the strict requirements of the producers resulting from contamination from coatings and/or adhesives.

The dearth of infrastructure for assortment, product disassembly and cullet separation regarding these several types of glass, originates from the hesitation of the producers to simply accept this cullet, and thus limits or prevents its recycling. Due to this fact, as this glass can’t circulation again into the unique product system (close-loop recycling), it will get down-cycled to functions similar to combination, ceramic-based merchandise, foam insulation, abrasives (Silva et al. 2017), or is disposed of in landfills. As the necessity of discovering various routes, markets and end-users for the upcycling of the tons of high-quality discarded glass is crucial, the partial diversion of this waste into the constructing trade by casting structural glass elements is price exploring.

The above developments reveal a spot within the literature in regards to the mechanical properties of solid glass elements and the suggestion of a design power for his or her structural use. That is linked with the absence of established manufacturing procedures and high quality management requirements, and thus the nice variability within the power of the solid glass merchandise in keeping with every producer and the corresponding glass composition and casting course of utilized. Using waste glass cullet is an added complication to this subject, giving rise to a collection of conventional and new forms of defects (Bartuška 2008;Bristogianni et al. 2019), which can compromise the power of the glass product.

This paper explores the flexural power of recycled solid glass—a property related to the engineering observe. Purpose is to present perception into the impact of the casting parameters on the power, and to evaluate the plausibility of using waste glass for the manufacturing of protected structural elements. Thus, on this work, a wide range of business glass waste silicates is examined and evaluated for his or her potential to be kiln-cast into structural elements at comparatively low temperatures (820– 1120 ◦C). The occurring defects are documented and correlated with the stage of manufacturing throughout which they’re precipitated.

Thereafter, two collection of four-point bending experiments are performed in kiln-cast glass beams of 30 × 30 × 240 mm dimensions. The outcomes are analysed in keeping with the position of the chemical composition, stage of contamination and adopted casting parameters, in figuring out the flexural power and the origin of fracture. The testing of a restricted variety of industrially manufactured elements serves as some extent of reference.

Desk 1 Specimen preparation, cullet categorization and kiln-casting settings – Full dimension desk

Table 1 Specimen preparation, cullet categorization and kiln-casting settings

2 Experimental work

2.1 Glass cullet categorization and specimen preparation

This work research a collection of attribute business glasses, used for the manufacturing of widespread glass merchandise similar to float glass, glass fibers, cookware and laboratory glassware, solid glass bricks, crystal ware and CRT TV screens.1 The selection of glasses is made in alignment with the forms of waste glass cullet offered by numerous glass manufacturing and recycling firms, so as to handle the recycling of available waste glass sources and thus sort out a sensible drawback.

X-ray fluorescent (XRF) analyses are performed with a Panalytical Axios Max WDXRF spectrometer to outline the chemical composition of the chosen glasses. The offered cullet is completely cleaned with isopropanol, and alien materials (metallic, plastic, cork) is manually eliminated when doable. The recognized contaminants within the given cullet, nonetheless current in traces after the cleansing course of, are listed in Desk 1 in keeping with the next categorization:

i. Coatings (gentle, exhausting, mirror, enamel, frit)
ii. Variations in composition of the identical glass kind (completely different producer, tints)
iii. Exterior contaminants throughout sorting: a. Organics (e.g. plastic, textiles), b. Non-glass inorganics (e.g. ceramics, stones, porcelain, glass ceramics), c. Metals, d. Completely different glass sorts (e.g. borosilicate, lead glass)

The cullet is then used for kiln-casting the 30 × 30 × 240 mm glass beams required for the four-point bending exams. This specific beam dimension is chosen because it supplies a considerable thickness of solid materials in order that the affect of the defects within the bulk will be evaluated, whereas preserving the mass under 1kg, and due to this fact lowering the annealing time. For every glass cullet, a minimum of 3 samples are produced for statistical functions.2 The cullet is positioned inside disposable silica/plaster funding moulds produced from Crystalcast M248,3 in a structured or random method. The moulds are then positioned in a ROHDE ELS 200S or ELS 1000S electrical kiln (Fig. 1) and kiln-cast, which means that just one kiln is employed for the entire casting course of (heating up, forming, annealing and cooling).

Fig. 1 Arrangement of the investment moulds inside the ROHDE ELS 1000S kiln
Fig. 1 Association of the funding moulds contained in the ROHDE ELS 1000S kiln – Full dimension picture

The glass samples are shaped at viscosities between 106–103.5 dPa s and at prime temperatures starting from 820 to 1120 ◦C which might be chosen in keeping with the chemical composition of every glass. The viscosity (η) vary chosen is greater than the 104–101.5 dPa s forming and melting vary adopted by the glass trade, considering the danger of inhomogeneity of the ultimate product. The strategy of glass forming at decrease temperatures is chosen on the one hand to cut back the required power and corresponding CO2 emissions, and however to accentuate the prevalence of defects and consider if their existence is suitable for a structural glass product.

Thus for a number of samples (e.g. float glass, borosilicate rods), 2–3 completely different prime temperatures are examined, equivalent to viscosity ranges of 105–6 dPa s and 103–4 dPa s, to additional research the affect of the defects on the flexural power. All specimens are stored at prime temperature for 10 h,4 quenched at a − 160 ◦C/h charge all the way down to their annealing level, heat-soaked for 10 h and cooled all the way down to their pressure level with a − 4 ◦C/h ramp, earlier than controllably cooled all the way down to room temperature at a quicker charge. This conservative annealing schedule ensures stress-free specimens, as seen by cross-polarized gentle.


1 Cathode Ray Tube (CRT) display screen manufacturing has ceased, but there’s nonetheless a substantial quantity of CRT glass cullet ensuing from the separation of disposed screens (Andreola et al. 2005).
2 In case much less samples are reported, the cullet obtainable was not ample for the manufacturing of three samples. These specimens are nonetheless offered on this research to show the failure mode of the precise kind of glass, slightly than derive an absolute flexural power worth.
3 Crystalcast M248 is an funding powder consisting of 73% silica content material (cristobalite, quartz), 23% calcium sulphate (gypsum) and 1% organics (Goodwin Refractory Providers 2003; Gold Star 2019). The selection of the mould materials is said to the kilncasting method adopted on this work versus the generally reported casting by melt-quenching. As comparatively low forming temperatures are chosen, the corresponding excessive viscosity of the heated glass doesn’t permit its immediate pouring from a melting (platinum or high-alumina) crucible to a preheated (metal or graphite) mould for annealing. Thus, the entire casting course of has to happen in a single mould that may face up to temperatures as much as 1150 ◦C, doesn’t connect to the glass and doesn’t trigger fracture to the specimen throughout cooling.
4 Given the excessive viscosity at prime temperature and the dimensions of the samples, a ten h dwell is empirically discovered appropriate for the elimination of enormous bubbles (> 1 mm) and the incorporation of the coatings to the glass community.


The specimens are produced at a 40 mm part top, after which minimize to dimension with a water-cooled rotary diamond wheel cutter, to take away the highest floor that always incorporates a excessive quantity of flaws (e.g. floor crystallization, bubbles, depletion of alkali within the composition, wrinkling, crazing). Then, the specimens are floor and polished with a Provetro flat grinder and diamond abrasive discs in sequence of 60, 120, 200, 400 and 6005 grit and their ensuing dimensions are documented.

The inhomogeneities within the glass specimens are noticed by bare eye and with using a Keyence VHX-5000 or VHX-7000 Digital Microscope. A qualitative evaluation of the inner residual stresses within the glass specimens is achieved through the use of crossed-polarized filters. Lastly, the beams are ready for the Digital Picture Correlation (DIC) measurement by making a speckle sample on one of many longitudinal surfaces with elastic white and black spray paints.

The preparation course of required for the manufacturing of the kiln-cast specimens is described intimately in Desk 1. Aside from the kiln-cast specimens, the next industrially manufactured specimens are ready and used as a reference:

  • Beams minimize out from commonplace Poesia6 solid glass bricks, floor and polished to a 30×30×240 mm dimension
  • Beams from 8/10 mm thick float glass plies, adhesively bonded with Delo Photobond 4468,7 floor and polished to a 30 × 30 × 240 mm dimension
  • Beams minimize out from commonplace Poesia6 solid glass bricks, floor and polished to a 30×30×240 mm dimension
  • Beams from 8/10 mm thick float glass plies, adhesively bonded with Delo Photobond 4468,7 floor and polished to a 30 × 30 × 240 mm dimension

The grinding and sharpening process adopted for the preparation of the above specimens is similar to the one described for the kiln-cast samples. Nevertheless, the underside and prime floor of the float glass specimens (single and bonded) is stored in its as obtained situation (optically nice polished) and solely the minimize edges are processed.

2.2 4-point bending take a look at arrange

1st collection of experiments (12 kiln-cast, 6 reference specimens)

The first collection of experiments is performed so as to present a basic overview concerning the flexural conduct of the completely different glass specimens. The specimens are examined utilizing a Zwick Z10 displacement managed common testing machine in a laboratory air atmosphere and at a charge of 0.2 mm/min. The fourpoint bending fixtures have a 110 mm span for the loading rollers and a 220 mm span for the help rollers, with 10 mm diameter fastened loading pins, and are loosely linked to the testing machine to permit some hinging (Fig. 2a).

2nd collection of experiments (53 kiln-cast, 5 reference specimens)

The 2nd set of experiments includes the repeated testing of every glass class and supplies correct displacement information. The variety of examined specimens per glass class is ready to a few, which is restricted for testing a brittle materials whose power is by default statistical because of the randomness of the occurring flaws within the glass (Quinn et al. 2009). This research, nonetheless, goals to cowl a broad number of glass sorts and examine them in keeping with their flexural conduct, so as to discover which recycled glass merchandise have additional potential for structural use.

For these exams, a Schenck 100KN displacement managed hydraulic common testing machine is employed, and the specimens are examined in a laboratory air atmosphere utilizing a 0.3 mm/min displacement charge, which roughly corresponds to a 0.5MPa/s charge.8 The four-point bending fixtures have a span of 100 mm for the loading rollers and 200 mm for the help rollers, with 20 mm diameter fastened loading pins (Fig. 2b). To permit for minor changes and rotational actions, the help fixture is positioned on a semi-circular pin, whereas the loading fixture is loosely linked to the testing machine. As well as, a 1 mm thick silicone rubber strip is positioned between every loading pin and the specimen.

Fig 2
Fig. 2 a Fixture and set-up of 1st collection of four-point experiments. b Fixture and set-up of 2nd collection of four-point experiments. An LVDT sensor is positioned on the center of the span. The entrance floor of the specimen is roofed with a speckle sample for the DIC measurement. The metallic strips positioned subsequent to the help pins are cushioning the specimen upon fracture and defend the LVDT sensor from injury. No contact happens between the specimen and the strips throughout the bending take a look at – Full dimension picture

To measure the displacement of the beam resulting from bending, two strategies are employed: 1) a Linear Variable Differential Transformer (LVDT) displacement sensor (Solartron AX 2.5 Spring Push Probe calibrated to a 0.5μm accuracy) is positioned below the center level of the decrease floor of the beam (measuring the purpose of most displacement), and a couple of) a 2D-DIC measurement, utilizing a high-resolution (50.6MP) Canon EOS 5Ds digital camera that takes one image per second of the speckled floor of the beam. The images of the 2DDIC measurement are analysed utilizing the GOM Correlate software program. One picture pixel corresponds to 31.5μm, due to this fact given the software program accuracy of 0.05 pixel, any displacement above 1.57μm will be captured. Flexural power and Younger’s modulus calculation The flexural power (σ) is computed from the equation under:


the place F the utmost load, L the help span, Li the load span, b the beam’s width and d the beam’s top.9

The calculation of Younger’s (E) modulus is carried out by correlating the drive information obtained from the Schenck machine with (1) the utmost displacement from the LVDT sensor and (2) the utmost displacement from the DIC evaluation (Fig. 3).


5 The 600 grit ending is ready in keeping with ASTM C1161-13. As well as, Quinn et al. (2005) observe of their research on Machining Cracks in Floor Ceramics that sintered response bonded silicon nitride flexural specimens with 600 grit grinding fail resulting from materials flaws slightly than machining injury. This commentary will be prolonged to glass specimens.
6 Poesia is the producer of the solid glass bricks for the Crystal Homes façade (Oikonomopoulou et al. 2018a).
7 This UV-curing acrylate is chosen as a result of it varieties a powerful bond with the glass surfaces that results in the monolithic behaviour of the glued pattern (Oikonomopoulou et al. 2018a). Below four-point bending, the bonded glass pattern is anticipated to indicate cohesive failure within the substrate (glass ply) and never delamination.
8 A barely quicker displacement charge was chosen for the 2nd collection, with the intention to cut back the full variety of DIC photos per experiment and thus confine the dimensions of the recordsdata produced by the picture processing software program GOM Correlate to a most of 25 Gigabytes. Each the first and 2nd collection displacement charges are under the speed of stress enhance of 1.1±0.2 MPa/s indicated by ASTM C158-02. A displacement managed charge is favoured over drive managed, to keep away from the crashing of the specimen upon failure, but additionally to permit for potential pop-ins (slight crack arrests) at most drive, when the crack entrance interacts with an interface encountered within the glass mesostructure.
9 It needs to be famous that because of the fastened loading pins, a scientific optimistic error might happen resulting from a frictional constraint of μ · F/2 occurring at every pin, with μ being the coefficient of friction (Quinn et al. 2009). This drive creates a counteracting second of μ·F·d/2, thus the above equation needs to be rewritten as:


Assuming a average μ = 0, 3, the systematic error could possibly be of magnitude 8.2% for the first collection of experiments and 9% for the 2nd. Nevertheless, resulting from inadequate information concerning the μ worth, the flexural power isn’t corrected on this research, and the reader ought to consider the potential of an error of roughly the aforementioned magnitude.


Fig. 3 Analysis of the displacement in y axis, using GOM Correlate software. The maximum displacement due to bending at point A is calculated by subtracting the total displacement at point A from the average displacement at point B and C
Fig. 3 Evaluation of the displacement in y axis, utilizing GOM Correlate software program. The utmost displacement resulting from bending at level A is calculated by subtracting the full displacement at level A from the common displacement at level B and C – Full dimension picture

On condition that the cross part of the beam in relation to the fixture spans leads to a comparatively stiff structural factor, a shear deflection needs to be accounted to the full vertical deflection. The bending and shear deflection at mid-span with respect to the beam level above the help pins, and for a 1:2 four-point bending fixture ratio, are outlined by the formulation under:

F 3,4

the place

F 5

Including the 2 segments of vertical deflection and fixing in the direction of the Younger’s modulus, it’s concluded:10

F 6

3 Outcomes

3.1 Defect analysis for kiln-cast specimens

The failings occurring within the floor and bulk of the produced glass specimens are qualitatively11 documented in keeping with kind and trigger. Purpose is to correlate the defects discovered to the glass supply used and adopted casting and post-processing process, and to subsequently assess their contribution to the specimens’ flexural power. The casting associated defects are categorized12 in:

1. Crystalline Inclusions
2. Glassy inhomogeneities (twine/ream)
3. Gaseous inhomogeneities (bubbles)

An outline of the defect classes and their causes is present in Fig. 4, primarily based on which a documentation of the noticed flaws per glass kind is offered in Desk 2.

Fig. 4 Categorization and causes of the defects encountered in the kiln-cast glass specimens
Fig. 4 Categorization and causes of the defects encountered within the kiln-cast glass specimens – Full dimension picture

Desk 2 Analysis of kiln-cast specimens – Full dimension desk

Table 2


10 For the Younger’s modulus calculation, the Poisson ratio of v = 0.22 of soda lime silica glass is used. Though among the many examined glasses there could also be a ± 0.02 deviation to this worth, this has a negligible impact on the outcomes.
11 The quantitative evaluation of the extent of inhomogeneities in solid glass specimens of appreciable cross section- and thus a number of layers of defects versus a thin-walled glass—is a posh course of that includes a number of completely different testing strategies (e.g. Computed Tomography Scanning to detect and measure density differentials, three-dimensional Imaging Actual-Time Polarimetry to outline the placement and form of twine, and many others). This evaluation is stored out of the scope of this research as the principle intention is to firstly establish the sort and site of essential flaws that require future consideration, and thus quantitative documentation.
12 Categorization primarily based on Bartuška (2008).


In additional element, the reason for these defects is related to a number of of the next manufacturing stages13,14:

I. Uncooked Materials.

A. Contamination.

i. Coatings.

A number of “flat” defects are noticed in kilncast specimens from float glass cullet coated with enamel paint or ceramic frit, because of the inadequate melting of the coatings (Fig. 5). The XRF analyses of two attribute coatings (Desk 3) present compositions wealthy in excessive meltingpoint metallic oxides and particularly in chromium(III) oxide (melting level of Cr2O3 is 2435 ◦C, NIH Database). The Xray diffraction (XRD)15 evaluation of kilncast glass samples (Fig. 6) reveals in these instances the presence of eskolaite (mineral identify of chromic oxide).

ii. Minor compositional variations.

Minor compositional variations result in glassy inhomogeneities similar to twine and color streaks. Some examples with heavy striation are recognized within the “Float combo” and “Lead CRT” (Fig. 7) samples.

iii. Exterior contaminants.

On this class, the presence of glass ceramics or chemically completely different households of glass within the cullet (not detectable by eye, e.g. aluminosilicate shards in borosilicate or soda lime silica cullet)), is essentially the most essential, resulting in specimens which fracture upon cooling, resulting from strains brought on by thermal growth variations. That is skilled within the “Float combo” (Fig. 8a) and “ Borosilicate combine Maltha” specimens. Extra particularly, the “Float combo” specimens had been solid by using a compilation of flat glass shards (of approx. 20– 50 mm width) offered by Maltha Recycling. This flat glass compilation is rejected from the recycling stream because the misguided deposition of glass ceramic plates (e.g. cooktops) within the flat glass assortment container—an usually encountered phenomenon—renders the complete container unsuitable for recycling.

The XRF and XRD analyses of attribute items from the flat glass compilation pattern (Desk 4; Figs. 8b, 9) place the contaminants within the commercially relevant lithium aluminosilicate glass ceramics system, which is characterised by the near zero thermal growth coefficient (Höland and Beall 2020). The very low thermal growth coefficient (CTE) contrasts with the standard 9.5 × 10−6/Okay (at 20–300 ◦C) of float glass (Shelby 2005), resulting in unavoidable cracking. Nevertheless, the discount of the flat-glass compilation pattern’s particle dimension (nice cullet or powder), may decrease the strains within the remaining solid product, and due to this fact this technique requires additional investigation.

Traces of metallic, clay or stone result in crystalline inclusions of a most of two mm dimension, however these are tolerated by the glass community (Figs. 10, 11). Nevertheless additional research is required to establish the crystalline inclusions (using scanning electron microscopy) and to check if their position stays impartial when the glass is subjected to temperature gradients.

B. Cullet dimension and form.

Within the addressed glass viscosity vary, the geometry of the cullet is usually mirrored in striations and/or three-dimensional bubble veils within the remaining glass part. In instances of very nice cullet (e.g. “Automotive Windshields” samples) this geometry isn’t distinguishable, and a slightly excessive content material miniscule bubbles prevails (Fig. 12).

Fig. 5
Fig. 5 a Microscope picture of a “Oven doorways, 1120 ◦C” kiln-cast glass with flat crystalline inclusions, twine, color streak (resulting from partially molten coating materials) and bubbles. b Microscope picture of a “Automotive windshields, 1120 ◦C” kiln-cast glass with crystalline inclusions and bubbles – Full dimension picture

Desk 3 Coating composition – Full dimension desk

Table 3

Fig. 6 XRD pattern of kiln-cast “AGC Float with black enamel” glass at 1120 ◦C. The sample is at a large extent amorphous (black curve) yet it presents some sharp crystalline peaks (coloured sticks)
Fig. 6 XRD sample of kiln-cast “AGC Float with black enamel” glass at 1120 ◦C. The pattern is at a big extent amorphous (black curve) but it presents some sharp crystalline peaks (colored sticks) – Full dimension picture
Fig. 7 “Lead CRT, 870 ◦C” specimen containing intense cord (seen as wavy lines) and bubbles
Fig. 7 “Lead CRT, 870 ◦C” specimen containing intense twine (seen as wavy traces) and bubbles – Full dimension picture
Fig. 8
Fig. 8 a Fractured “Float combo, 1120 ◦C” specimen resulting from glass ceramics contamination. b Glass ceramic shards encountered within the flat glass compilation pattern. The left column reveals the shards within the “as-received” clear situation, whereas the best column reveals their opaque model after heat-treatment at 1120 ◦C for 10 h. This conduct suggests a lithium aluminosilicate β-quartz stable resolution section within the clear situation that transforms to β-spodumene throughout heat-treatment at temperatures above 1000 ◦C. The bigger crystals within the later situation scatter the sunshine and result in opacity (Shelby 2005) – Full dimension picture
Fig 9
Fig. 9 XRD sample of the yellow clear glass ceramic depicted in Fig. 8b, within the as obtained situation (a) and after heat-treatment at 1120 ◦C for 10 h (b). The crystal construction in b is much like β-spodumene, but the fabric presents a number of phases – Full dimension picture

Desk 4 Chemical composition, crystal section and CTE of typical lithium aluminosilicate glass ceramics, in comparison with the examined glass ceramic samples, the solid “Float Combo Maltha” specimen and a typical window glass – Full dimension desk

Table 4

*Lithium is a light-weight factor that can not be detected by the XRF evaluation and due to this fact the proportion equivalent to lithium oxide is mirrored to a better content material of silica dioxide. In line with the bibliography, the offered composition ought to have a 2-3% lithium oxide content material and a decrease silica dioxide content material by 2-3%
**A decrease than 3% lithium oxide content material is anticipated within the chemical composition
[1] XRD measurements performed by Ruud Hendrikx (TU Delft, 3me) utilizing a Bruker D8 Advance diffractometer, Bragg–Brentano geometry and Lynxeye place delicate detector
[2] XRF measurements performed with a Panalytical Axios Max WD-XRF spectrometer by Ruud Hendrikx (TU Delft, 3me);
[3] Montazerian et al. (2015); [4] Songhan Plastic Know-how Co., Ltd.; [5] Schott (2015b); [6] Höland and Beall (2020); [7] Shelby (2005); [8] Chyung (1977); [9] Brennan (1979); [10] Campbell and Hagy (1975)

Fig. 10 a Microscope image of “Lead CRT, 870 ◦C” specimen, containing undissolved blue particles of—most probably—cobalt oxide. b The variable inclusions in the “Borosilicate mix Coolrec, 1120 ◦C” specimen are tolerated by the glass network
Fig. 10 a Microscope picture of “Lead CRT, 870 ◦C” specimen, containing undissolved blue particles of—most likely—cobalt oxide. b The variable inclusions within the “Borosilicate combine Coolrec, 1120 ◦C” specimen are tolerated by the glass community – Full dimension picture
Fig. 11 Crystalline inclusions and bubbles detected in the bulk of a “Schott DURAN tubes 1120 ◦C” specimen, viewed through cross-polarized light. Although some inclusions e.g. the depicted 62.5μm stone, induce stress to the surrounding glass, this is well tolerated within the 30 × 30 mm glass cross section
Fig. 11 Crystalline inclusions and bubbles detected within the bulk of a “Schott DURAN tubes 1120 ◦C” specimen, considered by cross-polarized gentle. Though some inclusions e.g. the depicted 62.5μm stone, induce stress to the encompassing glass, that is nicely tolerated throughout the 30 × 30 mm glass cross part – Full dimension picture
Fig. 12
Fig. 12 The scale, form, and association of the cullet, together with the forming temperature, result in organized (a), random seen (b) and random non-traceable (c) meso-structures within the glass part. a “AGC float with black enamel, 1120 ◦C” (picture width ≈ 30 mm), b “Wertheim, 820 ◦C” (picture top ≈ 30 mm), c “Automotive windshields, 1120 ◦C” (picture top ≈ 30 mm) – Full dimension picture


13 In Desk 2, flaws precipitated throughout levels III.B. and IV (postprocessing and dealing with flaws) should not talked about as they don’t seem to be linked to the fabric and its casting technique, however are slightly arbitrary and solely related to the fracture evaluation of every particular specimen.
14 The next microscope photos had been made utilizing a Keyence VHX-5000 or VHX-7000 Digital Microscope.
15 All XRD analyses on this work had been performed utilizing a Bruker D8 Advance diffractometer, Bragg-Brentano geometry and Lynxeye place delicate detector.


II. Glass forming.

A. Cullet association within the mould.

That is related with the geometry of the cullet (I. B) together with the firing schedule and corresponding viscosities of the shaped glass (II. B, C). An outlined cullet form and excessive viscosity can result in organized meso-structures composed of bubble veils (Figs. 13b, 14), twine or crystallized interfaces which lead to a extra predictable failure sample (Fig. 13). Such organized constructions additionally assist in distinguishing the position of those defects when current on the glass floor or within the bulk.

B. Forming temperature and corresponding viscosity in relation to dwell time.

The highest temperature impacts the extent of homogenization and the content material of air-bubbles. All samples current miniscule bubbles because of the comparatively low forming temperatures. As well as, the “cage” precept describing the blending of dense liquids is relevant on this case, which means that many of the molecules equivalent to an preliminary cullet piece will stay in the identical place in relation to their neighboring cluster of molecules (cullet piece). The extent of diffusion is elevated when a viscosity of 103.5 dPa s magnitude is reached, but it surely doesn’t in any case result in a totally combined glass within the given dwell time (see Figs. 12, 13).

Fig. 13 Kiln-cast experiments with Schott DURAN borosilicate rods of 24mm diameter forming 50mm cubic samples. a Crystallized hexagon structure, engineered at 970 ◦C. b Bubble-veil hexagon structure engineered at 1120 ◦C
Fig. 13 Kiln-cast experiments with Schott DURAN borosilicate rods of 24mm diameter forming 50mm cubic samples. a Crystallized hexagon construction, engineered at 970 ◦C. b Bubble-veil hexagon construction engineered at 1120 ◦C – Full dimension picture
Fig. 14 Bubble veil observed in a “Schott DURAN tubes 1120 ◦C” specimen. The maximum bubble diameter is less than 1mm, while the majority of the bubbles has a diameter below 0.2mm
Fig. 14 Bubble veil noticed in a “Schott DURAN tubes 1120 ◦C” specimen. The utmost bubble diameter is lower than 1mm, whereas the vast majority of the bubbles has a diameter under 0.2mm – Full dimension picture

C. Firing schedule together with temperature differentials within the kiln that promote crystallization.

That is notably relevant for the float and borosilicate glass samples shaped at 970 ◦C. In these samples, the entire interface between every cullet piece is crystallized. In line with the XRD evaluation (Fig. 15), the borosilicate samples develop b-cristobalite crystals, whereas the float glass samples wollastonite 2M, b-cristobalite and devitrite (Figs. 16, 17). Crystallization is favoured as a result of the samples are shaped under their liquidus level (TL is round 1080 ◦C for the precise float glass, and round 1200 ◦C for the precise borosilicate16) but reaching a low sufficient viscosity that kinetically permits nucleation. Nucleation begins on the interfaces, as there, a neighborhood compositional variation happens because of the volatilization of alkali and boron (within the case of the borosilicate glass).

Nevertheless, depletion of such components might result in unstable native compositions, as noticed within the crystallized layer of the borosilicate samples, which proves porous and water-absorbing (Fig. 17). Aside from the “engineered” crystallized constructions described above, the temperature circumstances and fluctuations throughout the kiln may also provoke native and random crystallization within the type of stones, at places of compositional alteration. Native variations within the composition will be brought on by contaminants within the uncooked materials, contact with the mould materials, volatilization of compounds, and gasoline bubbles. Due to this fact, such stones should not solely present in specimens produced from evidently contaminated cullet (e.g. “Automotive windshields” samples), but additionally in additional pure specimens (e.g. “Absolutely tempered (FT)17 float” samples).

Fig. 15 XRD patterns of float glass (left) and Schott DURAN borosilicate rods (right) fused at 970 ◦C
Fig. 15 XRD patterns of float glass (left) and Schott DURAN borosilicate rods (proper) fused at 970 ◦C – Full dimension picture
Fig. 16 Microscope images of the crystallized interface of the “Float 10mm fused 970 ◦C” samples (fractured surface). The parallel needle-like form of the crystals refers to devitrite
Fig. 16 Microscope photos of the crystallized interface of the “Float 10mm fused 970 ◦C” samples (fractured floor). The parallel needle-like type of the crystals refers to devitrite – Full dimension picture
Fig. 17 Crystallized interface of the “Schott DURAN borosilicate rods, fused at 970 ◦C” samples. a Microscope image showing a split interface due to fracture. b Water permeability of the crystallized interface (image height ≈ 30 mm)
Fig. 17 Crystallized interface of the “Schott DURAN borosilicate rods, fused at 970 ◦C” samples. a Microscope picture displaying a break up interface resulting from fracture. b Water permeability of the crystallized interface (picture top ≈ 30 mm) – Full dimension picture

D. Response with mould floor.

Through the kiln-casting course of (on the studied viscosity vary), the glass in touch with the silica/plaster funding mould, varieties a skinny crystallized interface, that may be simply eliminated by the described post-processing methodology (Fig. 18). Nevertheless, of specific curiosity are defects brought on by the interplay of the mould with the glass which might be deep sufficient to stay upon grinding (Fig. 19). These will be, for instance, stones of approx. ø 1–2 mm created from unfastened mould materials that by accident acquired integrated within the glass soften. One other attribute flaw happens because of the friction of the mould floor that obstructs the entire fusion between the cullet items. Because of this, localized or networks of infolds seem on the glass surfaces, which may additionally encapsulate mould materials. Upon grinding, the tip of those flaws might stay on the glass floor, and is noticed in depths as much as 5 mm. Lastly, just one case is noticed the place the glass bonds to the mould floor and breaks throughout cooling resulting from thermal growth variations (pattern “Borosilicate combine Maltha”).

E. Quenching charge to the annealing level.

On this research a decrease quenching charge of − 160 ◦C/h is adopted compared to the abrupt quenching adopted in industrial glass casting.18 The experimental outcomes present that this charge is ample to stop crystallization. Nevertheless, consideration is raised to the truth that a slower cooling charge might intensify the extent of polymerization of the glass community and result in a denser glass (Ito and Taniguchi 2004). Though this isn’t experimentally confirmed on this research, it stays a chance to be taken into consideration.

F. Annealing scheme.

A conservative annealing scheme has been used, thus the residual stresses detected within the samples utilizing cross-polarized filters are negligible and don’t appear to compromise the flexural power. Concerning the samples minimize out from the usual Poesia glass bricks, these do have minor residual stresses, which can be seen by the perimeter order within the isochromatic sample obtained by an Ilis StrainScope Flex round polariscope (Fig. 20), and likewise advised strongly by the tendency of this glass to chip throughout post-processing.


16 The liquidus level of glasses TL is discovered round a viscosity of 104 dPa s, and is estimated from the chemical composition of the given glasses in keeping with Fluegel (2007a).
17 The designation “Absolutely tempered” refers back to the cullet used for these samples, which originates from shattered absolutely tempered float glass panels. The ultimate kiln-cast elements are annealed and thus not tempered.
18 On this research, quenching might final even 4 h and takes place throughout the kiln, which is inherently completely different from the quenching at atmospheric circumstances throughout hot-pouring of glass that lasts solely a number of minutes.


III. Publish processing.

A. Insufficient elimination of current flaws.

As additionally mentioned at level II.D, not all floor flaws will be fully eliminated by post-processing (Fig. 21a). On this class of defects, the publicity throughout grinding of bubbles trapped within the glass bulk needs to be included. This leads to stress concentrating semi-circular intrusions of sharp edges on the glass floor that scale back the power. As well as, since bubbles can provide beneficial circumstances for the formation of crystals of their inside, the publicity of such gas-pockets on the floor naked the extra threat of stone publicity (see Figs. 19b, 21b).

B. Introduction of recent flaws.

The introduction of recent scratches from “renegade” abrasive grits (Quinn 2016) is principally noticed in glasses with decrease hardness, on this research notably the “Leerdam Lead” samples. Chipping is principally occurring within the cut-out commonplace Poesia samples, as mentioned in II.F. All samples current the danger of micro-cracking throughout coarse grinding that’s not sufficiently eliminated within the later levels of grinding and sharpening.

Fig 18
Fig. 18 Floor response to the mould materials. a Aspect floor of a “Automotive Windshields, 1120 ◦C” pattern, as launched from the mould. Improper fusion of the cullet, inclusions from the mould materials and stone formations are noticed. b Aspect floor of an “Oven doorways, 1120 ◦C” specimen, as launched from the mould. The white zones are crystalline formations from the response of the glass coating to the mould materials. Improper fusion of the cullet can be noticed, in addition to mould materials inclusions – Full dimension picture
Fig 19
Fig. 19 a Infold with stone inclusions from the response of the glass to the mould materials, within the floor floor of an “Oven doorways, 1120 ◦C” specimen (picture top ≈ 30 mm). b Microscope picture of the fracture origin of a “Schott DURAN tubes 1120 ◦C” specimen. Word that the crystalline inclusion from the response to the mould, should not solely located on the backside floor, however prolong to the majority as nicely – Full dimension picture
Fig. 20 Isochromic fringes observed via an Ilis StrainScoep Flex circular polariscope in a standard Poesia cast glass brick. The depth of the depicted sample is 10 cm
Fig. 20 Isochromic fringes noticed through an Ilis StrainScoep Flex round polariscope in a normal Poesia solid glass brick. The depth of the depicted pattern is 10 cm – Full dimension picture
Fig 21
Fig. 21 a Clustering of floor bubbles and stone inclusions on the backside floor of an “Oven doorways, 1120 ◦C” specimen that weren’t eliminated after grinding, type the power limiting flaw. b Microscope picture of a fractured “Schott DURAN tubes 1120 ◦C” specimen, displaying a bubble in proximity to the floor and stone formations originating from the bubble inside – Full dimension picture
Fig. 22 Overview of tested specimens
Fig. 22 Overview of examined specimens – Full dimension picture

IV Dealing with.

A collection of dealing with flaws (chippage, cleavage, percussion cone, level contact) randomly happen in among the specimens. The response of the solid specimens to dealing with injury versus that of industrially produced glass (e.g. float, extruded rods) requires additional investigation, but the extra pure solid specimens should not noticed to be extra prone than commonplace glass merchandise. Nevertheless, consideration needs to be drawn to the extra contaminated glass samples, as occasional massive defects on the floor (> 2 mm) amplify the impact of an impression.

3.2 4-point bending exams

The outcomes of each collection of experiments are offered in Figs. 22, 23, 24, 25 and 26 and Tables 5, 6, 7 and eight. The info from the primary collection is principally used for a primary basic steerage and as a affirmation of the second collection, which is the principle focus of this research. It needs to be burdened that the variety of examined specimens per class is restricted, and thus the offered outcomes are solely indicative and never ample for deriving statistical conclusions.

Fig. 23 Flexural strength results of 1st series of four-point bending experiments
Fig. 23 Flexural power outcomes of 1st collection of four-point bending experiments – Full dimension picture
Fig. 24 Flexural strength results of 2nd series of four-point bending experiments
Fig. 24 Flexural power outcomes of 2nd collection of four-point bending experiments – Full dimension picture
Fig. 25 Force versus Displacement graph. The displacement is measured from the DIC analysis
Fig. 25 Drive versus Displacement graph. The displacement is measured from the DIC evaluation – Full dimension picture
Fig. 26 Comparative graph of the Young’s modulus measured by the LVDT sensor during the 2nd series of four-point bending experiments
Fig. 26 Comparative graph of the Younger’s modulus measured by the LVDT sensor throughout the 2nd collection of four-point bending experiments – Full dimension picture


Desk 5 Outcomes of 1st collection of four-point bending experiments in regards to the kiln-cast beams – Full dimension desk

Tab 5


Desk 6 Outcomes of 1st collection of four-point bending experiments in regards to the reference beams – Full dimension desk

Tab 6


Desk 7 Outcomes of 2nd collection of four-point bending experiments in regards to the kiln-cast beams – Full dimension desk

Tab 7


Desk 8 Outcomes of 2nd collection of four-point bending experiments in regards to the reference beams – Full dimension desk

Tab 8

Though the primary and second collection differ within the fixture set-up (span, curler radius, connection element to common testing machine) and the sensitivity of the testing machine (10KN max. utilized load for the machine utilized in 1st collection versus 100KN for the 2nd), the outcomes of the 2 exams are aligned. Extra particularly, the samples of the first collection which might be solid at 1120 ◦C (“FT Float”, “Schott DURAN 24 mm rods”, “Oven doorways Coolrec”) rating throughout the identical flexural power vary (40–50MPa), the fused samples at 970 ◦C are considerably weaker (10–20MPa) whereas the pure single pane float samples have a barely higher efficiency (common flexural power of 55MPa). This efficiency rating and worth vary coincides with the outcomes of the 2nd collection other than the case of the fused float samples at 970 ◦C, the place a noticeably low flexural power is reported (< 10 MPa). That is attributed to the one-off prevalence of a community of micro-cracks on the floor of those specimens, which couldn’t be simply eliminated by post-processing.

Fig 27
Fig. 27 Aspect view of kiln-cast specimens fractured throughout the 2nd collection of four-point bending exams. Word that the first crack begins perpendicular to the beam’s lengthy axis after which splits within the case of medium/massive accrued elastic power (previous to cracking), or propagates as one crack within the case of low power (e.g. crystallized specimens). On the prime (compressive zone), the crack varieties compression curls – Full dimension picture
Fig. 28 Graph depicting the location of the fracture origin of the 2nd series specimens at the bottom surface. Note that fracture origins found at the two long edges are usually related to machining flaws
Fig. 28 Graph depicting the placement of the fracture origin of the 2nd collection specimens on the backside floor. Word that fracture origins discovered on the two lengthy edges are normally associated to machining flaws – Full dimension picture
Fig. 29 Mirror surfaces of fractured specimens (2nd series of experiments) depicting the main defect categories responsible for catastrophic failure. The reported flexural strength is linked to the type of defect but also to its size
Fig. 29 Mirror surfaces of fractured specimens (2nd collection of experiments) depicting the principle defect classes answerable for catastrophic failure. The reported flexural power is linked to the kind of defect but additionally to its dimension – Full dimension picture
Fig. 30
Fig. 30 Measuring instance of the fracture mirror and defect dimension at origin. All measurements are performed using a Keyence VHX-7000 Digital Microscope and with the fractured floor positioned perpendicularly to the microscope’s optical path. To acquire the mirror radius, the diameter of the mist-hackle boundary on the backside floor line (most tensile stress) is measured after which divided by half. This technique is chosen as not all mirrors are discovered semi-circular. Extra particularly, because of the stress gradient alongside the peak of the pattern (resulting from loading in bending), the mirrors seem elongated on this route, or might even be incomplete. Due to this fact a measurement alongside the underside floor is most well-liked. Furthermore, prolonged flaws on the floor or machining injury could cause the one-sided elongation of the mirror, and thus the measurement of the diameter as a substitute of the radius is opted – Full dimension picture
Fig. 31
Fig. 31 Flexural power versus 1/ √R graph for a collection of glass specimens. Generally, the upper the power, the smaller the mirror dimension. Nevertheless, because the mirror dimension to power relationship of various glass compositions is reported, multiple mirror constants A are relevant, and thus the info should not all corresponding in a single line (e.g. “Wertheim pellets” samples) – Full dimension picture
Fig. 32
Fig. 32 Flexural power versus essential flaw width (at fracture origin) for a collection of glass specimens. The power reduces with the rise of the flaw dimension, and specimens with floor stones are likely to fail at decrease power values than the purer samples that fail from post-processing and dealing with flaws – Full dimension picture

In Fig. 24, depicting the flexural power of the second collection of specimens, three important zones will be noticed: specimens of a flexural power under 30MPa, between 30 and 55MPa—the place most samples are situated, and between 55 and 75MPa. In all specimens, crack initiation begins on the backside floor (or at very shut proximity), on the space between the help pins (zone of most tensile stress, see Figs. 27, 28). As a basic development, glass specimens produced at decrease viscosities and from purer cullet are discovered on the prime zone of the flexural power graph, whereas specimens with apparent strengthlimiting flaws uncovered on the backside floor fail at low values.

An outline of the principle fracture origins is offered in Fig. 29, summarizing essentially the most essential defect classes: stones, crystalline interfaces, floor bubbles, and machining injury. The scale of the fracture mirror is measured in a collection of specimens (Fig. 30) and plotted towards the flexural power σ (Fig. 31) primarily based on “Orr’s equation” (Quinn 2016):

F 7

the place R corresponds to the mirror radius (on this research the mirror dimension extending to the mist-hackle boundary atthe backside floor of most pressure is measured) and A is the attribute mirror fixed per glass composition.

Sometimes, the bigger the failure stress is, the smaller the encountered fracture mirror will likely be. The rise of the width of the essential flaw is, as anticipated, answerable for the lower of the flexural power (Fig. 32). The upper power specimens appear to fail primarily from machining flaws, whereas stones or crystalline interfaces are answerable for the fracture of the decrease power specimens. But, the sort, dimension, amount and site of flaws alone can’t justify why some glass samples rating decrease than others. The structural efficiency of every glass kind must be reviewed as conjointly depending on the chemical composition of the glass in addition to its inherent defects (see Sect. 4). Additionally, fracture load uncertainty could also be relevant resulting from environmentally assisted gradual crack development (Quinn et al. 2009), because the utilized loading charge is slower (approx. by half) than the advised charge by the ASTM C158-02 guideline. The impact of gradual crack development needs to be additional experimentally investigated in a broader vary of testing speeds.

Concerning the Younger’s modulus, the calculation performed primarily based on the LVDT information leads to values which might be roughly 15% decrease than these present in literature. That is thought of a scientific error and is attributed to the standard of the sensor. Nevertheless, in every triplet of examined glass kind, there are matching E values reported. As well as, the stiffness relationship between the completely different glass households (Fig. 26) is present in accordance with the literature (Corning 1979; Campbell and Hagy 1975), and particularly: 


4 Dialogue

The flexural power of the solid glass specimens is conjointly associated to their chemical composition and inherent defects. To grasp during which instances the issues are the power limiting issue and when the mechanical properties associated to the composition have a figuring out position, the interpretation of the outcomes is structured within the following classes: a. Noncontaminated glass specimens, b. Contaminated vs. non-contaminated glass, c. Non-contaminated homogeneous glass specimens vs. with crystallized interfaces, and d. Reference specimens. On this method, the defects are categorized and remoted so their impact will be studied with extra readability, whereas the absence of overruling flaws (within the case of the pure samples) highlights the impact of the chemical composition.

Fig. 33
Fig. 33 Common flexural power to Younger’s modulus graph in regards to the non-contaminated kiln-cast specimens of the 2nd collection – Full dimension picture
Fig. 34
Fig. 34 Aspect floor of fractured specimens. a First three specimen correspond to “AGC darkish blue float 1120 ◦C” glass, then the subsequent two are “Leerdam 820 ◦C”, adopted by a “Barium CRT 870 ◦C” glass. b From prime to backside: “Wertheim 900 ◦C”, “Poesia 1070 ◦C” and three “FT Float 1120 ◦C” specimens – Full dimension picture

(a) Non-contaminated glass specimens

The purest, most homogeneous samples of every glass household included on this work are chosen for comparability (Figs. 33, 34). As these examples comprise much less imperfections, the impact of their chemical composition on their flexural power is highlighted. Desk 9 lists related calculated and/or measured bodily and mechanical properties of those glasses, together with information present in literature for comparable compositions.

Due to this fact, though the LVDT calculation doesn’t present precise values, it may be reliably used for a comparative evaluation between the completely different glass sorts. The DIC measurement is utilized to offer extra correct information concerning the utmost deformation, and for performing extra exact calculations of the E moduli for a collection of glass samples (see Sect. 4a). Nonetheless, the coupling of the DIC measurement throughout 4- level bending with a non-destructive testing technique for figuring out the E modulus, such because the Impulse Excitation Approach, is suggested in future testing, to confirm the reliability of the outcomes.

As seen in Fig. 33, there is a rise within the flexural power with rising Younger’s modulus,19 within the lead silicate, borosilicate, barium silicate and AGC darkish blue float glass samples. The rise in power is attributed to the rise of the common bond power and atomic packing density of the glass community. That is associated to the Younger’s modulus by the equation under (Makishima and Mackenzie 1973):

F 8

the place Cg is the atomic packing density (additionally talked about as Atomic Packing Issue, APF), and Gt the full dissociation power per unit quantity. Primarily based on the chemical compositions derived by the XRF analyses, the APF20 and Gt 21 are calculated and listed in Desk 9.

Due to this fact, by reviewing Desk 9, it’s anticipated that the lead silicate glass samples, which current the bottom dissociation power and packing density, could have the bottom power as nicely,22 whereas the soda lime silicate (SLS) glasses could have the best power. Additionally in accordance with the literature, the BaO containing silicate glass has a decrease flexural power and Younger’s modulus than the CaO silicates however greater than the lead silicates (Volf 1984; Corning 1979). Nevertheless, the Younger’s modulus alone can’t justify the deviation from the linear E/power relationship that current the Poesia, Wertheim, and FT float glass samples, and additional rationalization is required per glass kind.

Desk 9 Measured and calculated properties of the chosen (pure) glasses (in daring), and reference glasses of comparable composition – Full dimension desk

Table 9

The Poesia glass is a modified soda lime glass with a decreased forming temperature in comparison with typical float glass (TL is at round 980 ◦C, due to this fact 80–100 ◦C decrease than for SLS). It incorporates K2O and B2O3 in small quantities (< 3 wt%), and has a better Na2O/CaO ratio than typical SLS recipes. Regardless of the marginally decrease E modulus23 than the one in all AGC darkish blue glass, it offered the best flexural power amongst all examined specimens. That is attributed to a decrease brittleness of this specific glass. Sehgal and Ito (1998) state {that a} greater molar quantity (Vm) performs a key position within the discount of the brittleness, as a extra open construction permits extra deformation previous to crack initiation. Extra particularly, an rising soda/calcia ratio would lower the brittleness, in addition to the partial substitution of soda for potash. That is in accordance with the compositional variations of Poesia glass to the standard SLS recipe, which contribute to a extra open construction (Fig. 35) that enables for a barely elevated lodging of the stresses across the level/flaw the place the crack will provoke. The “Poesia 1070 ◦C specimens failed resulting from machining injury (Fig. 36).

Fig. 35 Graph of total dissociation energy versus the molar volume
Fig. 35 Graph of whole dissociation power versus the molar quantity – Full dimension picture
Fig. 36
Fig. 36 a Microscope picture of the underside floor and nook, and the fracture floor of a “Poesia 1070 ◦C” specimen. The reason for failure is grinding injury, which is demonstrated by the fracture mirror elongation to the left and the consecutive machining crack hackles alongside the fractured edge. b close-up of the fracture origin – Full dimension picture

The Wertheim glass has the best measured Younger’s modulus and the best calculated whole dissociation power, whereas it’s calculated molar quantity is much like the Poesia glass. The upper stiffness (compared to an SLS glass) will be attributed to the partial substitution of silica with alumina (≈ 5%), that reduces the openness of the community Sehgal and Ito (1998). Just like the Poesia glass, it has a decrease forming temperature than SLS glasses (TL at round 1015 ◦C), which will be linked to the combined alkali effect24 and the presence of boron trioxide in a small amount (Morey 1932). In line with the E/Vm properties of this glass, a a lot greater flexural power ought to have been anticipated.

The cause this glass failed at a decrease stress is linked to its kiln-casting at temperatures (820 ◦C, 900 ◦C) nicely under its liquidus level, which resulted in evident inhomogeneities. These inhomogeneities are concentrated within the interface created between every pellet of glass, and compose a three-dimensional community of planar zones consisting of bubbles and unfastened crystal formations. As well as, the forming temperature favours the prevalence of stones, because of the response of the recent glass with the mould, that are sufficiently sub-surface that they can’t be totally eliminated throughout commonplace postprocessing. These stones appear to weaken the glass floor and contribute to the formation of deeper striations throughout grinding, that are the sources of failure. The above described 3D community is probably not answerable for the crack origin, however provided that the specimens fail from a flaw in shut proximity to the community, it could contribute to zones of concentrated stress alongside the floor (Fig. 37).

Fig. 37
Fig. 37 Fracture floor of the “Wertheim 900 ◦C” specimen. a Microscope picture displaying the underside floor and fracture mirror. The specimen failed from a grinding scratch (crimson arrow) subsequent to the fusion interface (white arrows). b Shut-up of the mirror and the crystalline interface situated slightly below the fractured floor (solely the crystal formations alongside the underside floor are fractured) – Full dimension picture

It’s also attention-grabbing to match the “AGC darkish blue float” to the “FT float” specimens. These two glasses have very comparable compositions and are virtually similar in calculated atomic packing density and whole dissociation power. Nevertheless, the measured Younger’s modulus of the “FT float” glass specimens is decrease, and so is the flexural power. That is most likely linked to the thermal historical past of those two glasses. On the one hand, the “AGC darkish blue” glass has a barely decrease liquidus level (TL is round 1046 ◦C, whereas for the “FT Float” is 1063 ◦C). However, the darkish color of the AGC glass appears to contribute to the standard of the casting. The dense darkish blue color absorbs extra infra-red gentle throughout heating than the clear gentle blue, thus the physique will warmth up quicker. In an identical method, the darkish glass will set quicker throughout cooling because of the higher warmth loss by radiation (Kita˘ıgorodski˘ı and Solomin 1934; Burch and Babcock 1938).

The quicker setting charge can affect the coordination state of the transition metallic oxides included within the composition25 and thus have an effect on the full dissociation power of the community bonds— one thing not accounted for within the calculations. As well as, the decrease liquidus level and elevated warmth absorption promote the complete fusion of the cullet items and the elimination of stone formation, thus resulting in the diminishing of flaws on the glass floor, and to a better flexural power. In antithesis, infolds on the glass floor of the “FT float” specimens are created by inadequate fusion of the cullet positioned subsequent to the mould partitions, and crystalline inclusions resulting from contamination from the mould, are the principle reason for failure of the “FT float” samples, in keeping with the evaluation of the fracture mirrors (Fig. 38). Resulting from these flaws, the “FT float” glass specimens fail at values decrease than anticipated compared to the remainder of the samples.

Fig. 38
Fig. 38 Microscope photos of the underside and fracture floor of a “FT Float 1120 ◦C” specimen, displaying the fracture origin (Fig. 38b is a magnification of the fracture origin). The reason for failure is a mixture of grinding scratches (crimson arrows) appearing upon a floor infold with crystalline inclusions (white arrows). The hertzian cones (gray arrows) that discuss with impression injury have an reverse to the crack entrance route and are thought of secondary breaks. Wanting by the hertzian cones, traces of crystalline inclusions will be noticed – Full dimension picture


19 The graph in Fig. 33 relies on the Younger’s modulus calculated from the DIC measurements. The reported E modulus is roughly 5% greater than in literature, which could possibly be partly associated to testing errors and partly to the fabric itself and its casting process.
20 The atomic packing density is calculated utilizing the next formulation:


the place xi is the molar fraction, Vi the ionic quantity of the ith oxide and Vm is the molar quantity of glass, and particularly:




the place NA is the Avogadro’s quantity, rA,B = ionic radii of MxOy oxide, M is the molecular mass and ρ is the density of the glass. The Vi is derived from Makishima and Mackenzie (1973) and Inaba et al. (1999) primarily based on Pauling’s ionic radii. The density is calculated from the chemical composition utilizing the mannequin developed by Fluegel (2007b).
21 The full dissociation power is calculated from the dissociation power of the oxide constituents listed within the work of Inaba et al. (1999).
22 PbO has one of many lowest Gi as reported by Makishima and Mackenzie (1973), and a comparatively excessive molar atomic mass. The elevated mass of the lead ion slows down the chemical reactions throughout quenching, and leads to a much less organized/packed community.
23 It needs to be famous that the calculated EPoesia utilizing the APF and Gt equivalent to the chemical composition is discovered a lot decrease than the EAGC blue. This could possibly be associated to a fallacious estimate of the B2O3 content material, which can’t be decided by the XRF evaluation, and/or a better packing density attributable to the thermal historical past of the kiln-cast elements. The EPoesia derived from the LVDT information is thus thought of extra dependable for additional evaluation.
24 The time period describes anomalies noticed in glasses and melts containing a mix of two or extra alkali oxides. In line with Shelby (2005), the viscosities of such melts are decrease than these containing the identical quantity of a single alkali oxide.
25 The XRF identifies a collection of transition metals on this glass that act as colorants: 0.76% Fe2O3, 0.065% TiO2, 0.029% MnO, 0.023% Cr2O3).


Concerning the fracture evaluation of the glasses studied on this class, essentially the most prevailing causes of failure are discovered to be machining injury and dealing with flaws (see additionally Fig. 28, most “pure” specimens fail at an edge flaw), justifying that the purity of the cullet and the comparatively excessive forming temperatures (compared with different glass samples on this work) remove the amount of power limiting flaws. Exceptions are discovered within the “FT float” collection, as described above, and the “Schott DURAN tubes” specimens. These borosilicate glass samples are in reality shaped at a excessive viscosity (≈ 104.5 dPa s < TL) and are characterised by an elevated quantity of bubbles (primarily concentrated on the interfaces created between every cullet piece throughout forming). These bubbles type clusters for crystal development and if situated on the glass floor or at shut proximity, they turn out to be the power limiting flaw that results in fracture (Fig. 39). The flexural information obtained for these two glasses from the first collection of four-point bending experiments match with the outcomes of the second take a look at.

Fig. 39
Fig. 39 Microscope photos of the fracture origin of a “Schott DURAN tubes 1120 ◦C” specimen. a Incomplete fracture mirror across the origin flaw, which is a crystalline inclusion at 1mm inside from the underside floor. That is the one specimen that did not break at a flaw precisely on the floor. b Magnification of the crystalline inclusion, which is clustered along with air bubbles. Early mist and hackle seem throughout the mirror on account of interplay of the elastic wave with the defect – Full dimension picture

(b) Contaminated vs. non-contaminated glass specimens

This class research glass specimens kiln-cast from contaminated cullet, and compares them to the purer specimens described above. The entire studied specimens, “Float combo”, “Oven doorways”, “Automotive windshields”, and “AGC enamel black” are typical soda lime silicates and have a considerable amount of distinct crystalline inclusions, and/or heavy twine. Their flexural power is barely decrease than the one noticed for the FT Float specimens and their Younger’s moduli are comparable. The “AGC enamel black” collection appears to have the best flexural power on this class, which is attributed to the truth that just one kind of glass is used for the casting of those samples (thus no twine is noticed resulting from minor compositional variations). As well as, the substantial dimension of the glass items permits their thorough cleansing, which isn’t the case within the smaller sized cullet of the “Oven door” and “Automotive windshield” samples.

All specimens fail at decrease values than many of the purer glasses studied above, primarily resulting from crystalline formations on the floor (Fig. 40). These stones are created both from inherent contamination, or from additional response of the contaminants with the mould materials. The a number of defects situated within the bulk of those specimens should not activated throughout the 4-point bending nor do they appear to cut back the Younger’s modulus. Quite the opposite these defects are tolerated throughout the glass community. Nevertheless, the extra the defects within the bulk, the extra the probabilities of such flaws to be uncovered on the floor, and consequently the upper the danger of failure.

Fig. 40
Fig. 40 Microscope photos of crystalline formations that operate because the origin of fracture in contaminated kiln-cast specimens. The response of cullet contaminants (e.g. coatings) with the mould appear to advertise such formations. a The stone within the “AGC Float with black enamel, 1120 ◦C” specimen is adjoining to the enamel interface. b Stone in a “Oven doorways 1120 ◦C” specimen – Full dimension picture

(c) Non-contaminated homogeneous glass specimens vs. with crystallized interfaces

On this class, the soda lime and borosilicate glass samples which might be kiln-formed at 970 ◦C and comprise structured crystallized interfaces (“Float 1cm, 3 horizontal layers”, “Float 1cm, 24 vertical layers”, “Schott DURAN 10 vertical layers”), are studied compared to their extra homogeneous variations, kiln-formed at 1070 ◦C and/or 1120 ◦C (“FT Float 1120 ◦C”, “Schott DURAN 10 vertical layers 1070 ◦C, 1120 ◦C”). The actual side with this class is that the “defects” or zones of compositional/structural variation, are intentionally engineered at specified places and geometrical patterns. Thus, in antithesis with the random prevalence of stones described within the class above, on this part, the dimensions and distribution of the crystalline formations will be anticipated. As a consequence, their impact on the structural efficiency will be straight correlated.

Fig. 41
Fig. 41 Fracture sample and flexural power vary (MPa) of homogeneous (prime) and with crystallized interfaces (backside) specimens. Word that if the crystallized interface is located solely within the bulk (backside proper), the bending power of the specimen is much like a homogeneous one (prime proper). When the crystal lized interface is uncovered on the backside floor, the specimens fail at a decrease drive, from a flaw originating on the glass-crystal interface. Particularly within the case of the borosilicate crystallized specimens, the low elastic power saved leads to a single crack with out forking – Full dimension picture

Due to this fact, it may be noticed that the fused “Float 1cm, 3 horizontal layers” specimens current very comparable flexural power and Younger’s modulus with the extra homogeneous “FT Float” specimens (Fig. 41). It’s because the crystalline interfaces are situated within the bulk, in parallel layers to the underside floor, and thus should not uncovered to the utmost tensile stress zone. They behave due to this fact in an identical method to the homogeneous specimens. This isn’t the case nonetheless with the “Float 1cm, 24 vertical layers” specimens, the place the crystalline interfaces are uncovered on the backside floor, and in reality aligned perpendicularly to the tensile forces. Though the Younger’s modulus stays comparable, the flexural power is diminished by greater than 20%. The fracture origin of those samples is all the time situated at these crystal-glass interfaces and initiates from the glass zone in rapid proximity. The crystalline formations thus appear to behave as stress inducing components, of maybe greater fracture toughness than the encompassing glass matrix, which weaken the glass specimen.

The kind and thickness of the crystalline interface additionally performs a major position. The skinny b-cristobalite layer created in “Schott DURAN 10 vertical layers 970 ◦C” leads to a dramatic drop of 75% of the power, and a lower of the Younger’s modulus. The fracture origin is, in an identical method to the float examples, all the time situated on the crystalline-glass interface.

At this level, figuring out the impact of those crystalline formations and their geometrical association, consideration needs to be raised to the intermediate states between a fused glass specimen produced at viscosities round 106 dPa s and a homogeneous specimen solid at temperatures nicely above the liquidus level (the place the speed of diffusion is way greater). Specimens produced at a 105 or perhaps a 104 dPa s viscosity appear to retain traces of the interface created between every cullet piece throughout heating up, within the type of refined bubble veils, twine and spots of crystalline formation. That is evident for instance within the “Schott DURAN 10 vertical layers 1070 ◦C”, kilncast at a 105 dPa s viscosity (Fig. 42).

These samples contained the above described bubble veils and stones in the identical geometrical association because the “Schott DURAN 10 vertical layers 970 ◦C”. These specimens, though stronger than the fused model, had a 30% decrease flexural power than the specimens kiln-cast at a 50 ◦C greater temperature. All of them failed from both a crystalline flaw or a bubble situated in one in all these veils (Fig. 43). It is a very essential subject, given the truth that each the 1070 ◦C and 1120 ◦C produced specimens look the identical and are clear and never similar to the contaminated samples described within the class above. This highlights how essential a 50 ◦C temperature distinction will be when casting at viscosities simply round and under the liquidus level.

Fig. 42
Fig. 42 “Schott DURAN 10 vertical layers” specimens produced at 1070 ◦C (left column) and 1120 ◦C (proper column). The remnant bubble veils within the specimens produced at a decrease temperature lead to a decrease flexural power. The origin of fracture in these specimens will be present in one in all these bubble veils – Full dimension picture
Fig. 43
Fig. 43 Microscope photos of the fracture mirror of a “Schott DURAN 10 vertical layers, 1070 ◦C” specimen. a Succession of air bubbles near the underside floor, interacting with the elastic wave. The reason for fracture is a stone formation (proper finish of the image) in proximity to the bubble clustering. b Magnification of stone formations extending from the floor to the majority. Their perpendicular to the floor route and the presence of a bubble, counsel that these stones are shaped from the interplay of the mould materials with the bubbles created on the fusion interface between the glass rods throughout forming at a-favourable for crystallization-temperature – Full dimension picture

(d) Reference specimens

The industrially manufactured glass specimens are examined so as to present some extent of reference and comparability with the kiln-cast glass samples. Their structural efficiency is described under per kind.

The beams minimize out from commonplace Poesia solid glass bricks (initially scorching poured at round 1200 ◦C) are extra homogeneous than the kiln-cast specimens produced within the lab. Aside from some minor striae and few bubbles, they don’t comprise essential defects similar to stones, because the purity of the uncooked supply, the above liquidus level forming temperature, the abrupt quenching at atmospheric circumstances, and the stainless-steel moulds used for his or her casting, stop their formation. Nonetheless, these specimens current a ten% decrease flexural power than the much less homogeneous, recast specimens at 1070 ◦C. That is attributed to the quicker annealing scheme adopted for these elements, which causes residual stresses frozen within the glass, and makes it extra prone to wreck. As a consequence, throughout the chopping and grinding of the part in dimension, a number of chips and ensuing cleavage injury are precipitated resulting from inadequate annealing, which aren’t totally eliminated throughout sharpening. The added stress and machining defects are the reason for fracture, at a decrease power.

Contemplating the only float glass pane specimens, these are essentially the most homogeneous of all studied samples, with a pristine polished backside and prime floor. Since these specimens are minimize out from bigger float panels, their edges are floor and polished as described in Sect. 2.1. All single pane samples from the primary collection of four-point bending experiments failed from a machining flaw on the edge, throughout the backside zone of most tensile stress. The typical flexural power for the ten mm panes is 55MPa, which is 20% greater than the “FT Float 1120 ◦C” specimens however 20% decrease than the best scoring specimens “AGC darkish blue 1120 ◦C” and the “Poesia 1070 ◦C”. Undoubtedly, the standard of the underside edges can dramatically have an effect on the flexural power of the float glass pattern in bending.

In line with the dimensions and the sharpening high quality of the samples, and the take a look at settings, a variety of flexural strengths in 4 point-bending are reported in literature, from 35 to 170MPa (Veer and Rodichev 2011), 51–71.5MPa (Veer 2007), 53–129MPa (Yankelevsky et al. 2016) to call a number of. The 55MPa power of the examined samples on this research is on the low finish of this vary, and consistent with the literature, given the comparatively tough edge finishes. A a lot greater power could possibly be anticipated with finer sharpening. In that sense and considering that the kiln-cast specimens exhibiting greater tensile strengths failed as nicely from machining flaws, it may be derived {that a} a lot greater power is feasible with the commercial nice sharpening of the kiln-cast specimens.

The beams produced from adhesively bonded (Delo Photobond 4468) 8/10 mm thick float glass plies, and examined with their plies parallel to the underside floor, have a median flexural power of 48MPa (1st and 2nd 4 level bending collection), which is 10% greater than the kiln-formed “FT Float 1120 ◦C” specimens. Not one of the specimens failed from an edge flaw; the reason for failure is attributed to minor dealing with injury on the backside floor. The Younger’s modulus of the adhesively bonded beams is decrease than that of the monolithic, kiln-cast SLS specimens, because of the adhesive layers.

Total, the flexural power values obtained from the industrially produced reference samples are on the prime finish of the 30–55MPa (second) zone, and don’t exceed the efficiency of the purest kiln-cast samples (discovered within the first zone). That is an encouraging consequence, given the truth that all of the kiln-cast specimens produced for this research have some stage of inhomogeneities.

5 Conclusions

A wide range of business glass waste sorts is examined for the power to be kiln-cast into structural elements at comparatively low temperatures (820–1120 ◦C), and the flexural power of the kiln-cast specimens is evaluated.

The kiln-casting experiments present that meticulous separation of cullet on the recycling amenities ensures a profitable casting. Coatings and traces of exterior contaminants similar to organics and metals are tolerated by the glass community but result in defects and low flexural power, whereas contamination by glass ceramics and glasses with important compositional variations causes the fracture of the specimens throughout cooling. Glass compositions with a decrease liquidus level facilitate low temperature kiln-casting which ends up in extra homogeneous glass surfaces, because the decrease viscosity throughout forming minimizes the prevalence of sintering flaws, surfaces bubbles and stone formation from mould contamination.

Concerning the four-point bending experiments, though the variety of examined specimens per glass kind isn’t ample for deriving statistical information, they do present a very good overview and affordable estimate of the structural efficiency of every specimen kind, in keeping with the chemical composition, stage of contamination, and adopted casting parameters.

The impact of the chemical composition on the power is distinctly noticed within the specimens produced from purer cullet and at greater forming temperatures. Amongst these samples, a transparent enhance within the power and Younger’s modulus is noticed, consecutively from the lead silicate, to the borosilicate, barium silicate and as much as the soda lime silicate household. The purer, extra homogeneous samples predominantly fail from exterior defects induced by machining and dealing with injury. The impact of the composition is nonetheless blurred within the extra contaminated samples, the place crystalline formations shaped on the backside floor throughout the zone of most tensile stress, are the prevailing reason for fracture resulting in a considerably decrease power.

Inside the soda lime silica household, notably promising are the marginally modified recipes containing small quantities of K2O and B2O3 and a better Na2O to CaO ratio. The decrease viscosity of those glass melts facilitates the casting course of, whereas their extra open construction (greater molar quantity) presents a much less brittle various for the same Younger’s modulus to that of SLS glasses, main finally to a better flexural power.

Glass households of a fair decrease liquidus level, such because the studied lead silicate and barium silicate samples, are enticing for decrease power manufacturing. Nevertheless, for structural functions demanding greater power, the barium silicate possibility is rather more promising because of the greater E modulus and fewer susceptibility to scratching.26

Concerning the extra inhomogeneous specimens, produced from contaminated cullet at temperatures across the liquidus level, they nonetheless current a very good flexural power and are appropriate for structural functions demanding decrease tensile power, similar to bricks. The failings occurring within the bulk should not activated throughout the four-point bending take a look at and have a minor and even negligible contribution to the power and E modulus. Nevertheless, an elevated density of defects within the bulk ought to suggest a better density of flaws on the floor as nicely, which ought to result in a median power discount. A better forming temperature (above the liquidus level) would considerably assist in diminishing the quantity of flaws, however contemplating the financial and environmental benefits of decrease temperature processing, such an act could be solely significant if greater design strengths had been required per particular case.

Crystallized geometrical constructions are induced inside soda-lime-silica and borosilicate specimens produced at greater viscosities (106–105 dPa s). If these constructions are situated within the bulk, the flexural power of the specimen is the same as that of a extra homogeneous casting at a better temperature (near the liquidus level). Nevertheless, the publicity of such constructions on the floor subjected in pressure can result in a dramatic lower of power of even 75% in keeping with the character of the produced crystalline formations. On this case, the origin of fracture all the time happens within the glass/crystal interface. Particular consideration needs to be given to castings shaped at between 105–104 dPa s viscosities, because the glass merchandise might seem homogeneous however retain important inhomogeneous zones of miniscule bubbles and stones on the former interface created between every cullet items throughout heating up. Such formations uncovered on the tensile floor are essential for the specimen’s power.

Industrially SLS manufactured glass samples, postprocessed within the lab amenities to match the studied specimen dimension, current comparable flexural power to that of the float glass kiln-cast specimens (at 1120 ◦C), but rating on the decrease finish of power values reported within the literature. Machining flaws from the processing to dimension, and inadequate annealing within the case of the solid bricks, are the components answerable for the decrease power. A finer sharpening would considerably enhance the power, not solely of those samples, but additionally of the purer kiln-cast specimens. Nevertheless, provided that the bottom power specimens could be much less affected by a finer sharpening high quality, the statistical power wouldn’t be elevated that a lot, as it’s dominated by these decrease outliers.


26 Yamane and Mackenzie (1974) show of their mannequin the proportional relationship of Vicker’s hardness to the Younger’s modulus and bond power. As some extent of reference, Ainsworth’s (1954) measurement of Vicker’s hardness for a 18Na20·10BaO· 72SiO2 (mol%) glass is 522 kg/mm2 and for a 18Na20 · 10PbO· 72SiO2 (mol%) glass 445 kg/mm2.


6 Suggestions

The outcomes of this research present the potential of recycling waste glass into solid structural constructing elements. Nevertheless, for the protected software of such merchandise, additional validation is required and an elevated variety of examined specimens per class (≥ 30, Quinn et al. 2009) is required to derive statistical predictions. Specifically, the repetition of testing is of essential significance within the case of the contaminated samples, the place a better diploma of variability is anticipated within the mechanical properties. The systematic testing of such samples needs to be linked with a quantified documentation of the sort and stage of inhomogeneities within the glass previous to testing. Cautious and intensive fracture evaluation of the examined specimens can be essential to establish essentially the most essential defects, and the connection of the flaw dimension to the flexural power.

The physiochemical identification of the crystalline formations on the glass floor by scanning electron microscopy is required for categorizing such essential flaws. Additional testing is important, as nicely, to find out the affect of scale issue, and of static fatigue in moist environments (impact of gradual crack development). As well as, the learning of the behaviour of crystalline inclusions within the bulk glass below thermal gradients related to constructing functions is vital to remove the danger of thermal cracking. Extra, non-destructive testing for figuring out the Younger’s modulus and the extent of inhomogeneities within the solid glass can be advised, implementing the Impulse Excitation Approach. Investigation of whether or not such a quick and cheap non-destructive method may function a top quality management technique for solid glass merchandise is price exploring.

Concerning the extra contaminated elements, consideration needs to be given in bettering the standard of the stone-containing floor of the recycled glass. Chemical strengthening of the floor by ion trade could possibly be a—excessive value—resolution though this isn’t seemingly to assist with deep defects. One other less complicated resolution relevant for top viscosity castings (the place the diffusion charge is low), is the structuring of two (appropriate) cullet qualities contained in the mould: a purer alongside the demanding zones, and a decrease extra contaminated high quality within the bulk (Fig. 44). Such a composite glass would allow using contaminated, undesirable cullet with out essentially compromising the power of the ultimate product.

Lastly, the engineering of crystalline or bubble veil geometrical constructions throughout the glass is price additional exploration, as they will result in fractures inside a predictable power vary and site. Additionally they can result in constructing elements with non-standard look and thus greater architectural attraction.

Fig. 44
Fig. 44 Prototypes of composite kiln-cast glass elements, containing a purer glass on the backside and a weaker, extra contaminated glass on the bulk and prime zone. a Gradient from a pure clear soda lime silica glass (backside) to {a partially} crystallized zone produced from waste glass powder (residue from the glass container recycling course of). b Reinforcement of a “FT Float” glass beam by a backside layer comprising the stronger “ACG blue” glass – Full dimension picture

Acknowledgements The authors want to specific their gratitude to Giorgos Stamoulis for his important contribution in getting ready the four-point bending experiment and DIC measurements, in addition to Kees van Beek for his steerage. We’re notably grateful to George Quinn and James Varner for his or her invaluable enter concerning the experimental process and the fracture evaluation of the outcomes. The suggestions obtained from Bert Sluijs, Mauro Overend, Christian Louter and Karl-Heinz Wolf is extremely appreciated. We’d additionally wish to thank Ruud Hendrikx for the XRF and XRD analyses, Mariska van der Velden for helping within the specimen preparation, Sander van Asperen and Wolfgang Gard for offering entry to their labs’ Keyence digital microscopes, and Henning Katte and Daniel Schreinert (Ilis) for sponsoring using StrainScope Flex. The authors are additionally very grateful to Cor Wittekoek (Vlakglasrecycling), Danny Timmers (Maltha Glasrecycling Nederland), Marco Zaccaria and François Boland (AGC Belgium), Brian Wittekoek (Coolrec), Bettina Sommer (Royal Leerdam Crystal), and Klaas Roelfsema (Schott), for the glass cullet contribution, which was of key significance for this work. Lastly, we want to thank Erik Muijsenberg (Glass Service) for the “Glass Defects” (Bartuška 2008) e-book, and Peter de Haan (AGR Delft) for the “Shade Atlas of Container Defects” (Aldinger and de Haan 2019) and “Shade Atlas of Stones in Glass” (Aldinger and Collins 2016) books, which had been very useful within the strategy of defect categorization and identification.

Compliance with moral requirements

Battle of curiosity On behalf of all authors, the corresponding writer states that there is no such thing as a battle of curiosity.


Abrisa Applied sciences: SCHOTT Duran® Lab Glass (Tubed) (2014)
Ainsworth, L.: The diamond pyramid hardness of glass in relation to the power and construction of glass. Soc. Glass Technol. 38, 501–547 (1954)
Aldinger, B.S., Collins, B.Okay.: Shade Atlas of Stones in Glass. American Glass Analysis, Butler (2016)
Aldinger, B.S., de Haan, P.W.: Shade Atlas of Glass Container Defects. American Glass Analysis, Butler (2019)
Andreola, F., Barbieri, L., Corradi, A., Lancellotti, I., Falcone, R., Hreglich, S.: Glass-ceramics obtained by the recycling of finish of life cathode ray tubes glasses.Waste Manag. 25, 183–189 (2005).
ASM Worldwide: Purposes for Glasses Engineered Supplies Handbook Desk Version. ASM Worldwide (1995)
Bartuška, M.: Glass Defects. Glass Service Inc. and Práh, Prague (2008)
Brennan, J.J.: Program to Research SiC Fiber Strengthened Glass Matrix Composites. United Applied sciences Analysis Heart, Connecticut (1979)
Bristogianni, T., Oikonomopoulou, F., Justino de Lima, C.L., Veer, F.A., Nijsse, R.: Structural Solid Glass Elements Manufactured from Waste Glass: Diverting On a regular basis Discarded Glass from the Landfill to the Constructing Business. Heron 63 (1/2 Particular subject: Structural Glass) (2018)
Bristogianni, T., Oikonomopoulou, F., Veer, F.A., Nijsse, R.: The impact of producing flaws within the meso-structure of solid glass on the structural efficiency. In: Zingoni, A. (ed.) Advances in Engineering Supplies, Constructions and Programs: Improvements, Mechanics and Purposes, pp. 1703–1708. CRC Press, Leiden (2019)
Burch, O.G., Babcock, C.L.: Impact of glass coloration on setting charges in manufacture of glass bottles. J. Am. Ceram. Soc. 21(10), 345–351 (1938)
Campbell, D.E., Hagy, H. E.: Glasses and Glass-Ceramics. In: Lynch, C.T. (Ed.) CRC Handbook of Supplies Science, vol. II: Materials Composites and Refractory Supplies. CRC Press, USA (1975)
Chyung, Okay.: Clear Beta-Quartz Glass Ceramics. Corning Glass Works, United States Patent US4018612A (1977)
Corning: Properties of Corning’s Glass and Glass Ceramic Households. In: Supplies for the Design Engineer. USA (1979)
Fluegel, A.: Glass viscosity calculation primarily based on a worldwide statistical modeling strategy. Glass Technol. Eur. J. Glass Sci. Technol. A 48, 13–30 (2007a)
Fluegel, A.: International mannequin for calculating room-temperature glass density from the composition. J. Am. Ceram. Soc. 90, 2622–2625 (2007b).
Friedrich & Dimmock Inc.: Comparative Values of Borosilicate Glasses. In: Simax Glass Properties
Gold Star: Funding Casting Powder Security Knowledge Sheet, (2019)
Goodwin Refractory Providers Ltd: Crystalcast (M248). (2003)
Gregory, C.: Hole Fibers. In: Sanghera, J.S.A., I.D. (ed.) Infrared Fiber Optics. CRC Press LLC, USA (1998)
Heimerl, W.: Chemical resistance and corrosion, and ion launch. In: Bach, H., Krause, D. (eds.) Evaluation of the Composition and Construction of Glass and Glass Ceramics. Springer, Berlin (1999)
Höland, W., Beall, G.H.: Glass—Ceramic Know-how, third edn. Wiley, Hoboken (2020)
Inaba, S., Fujino, S., Morinaga, Okay.: Younger’s modulus and compositional parameters of oxide glasses. J. Am. Ceram. Soc. 82, 3501–3507 (1999).
Ito, S., Taniguchi, T.: Impact of cooling charge on construction and mechanical conduct of glass by MD simulation. J. NonCryst. Solids 349, 173–179 (2004).
Kita˘ıgorodski˘ı, I.I., Solomin, N.W.: Price of setting of glass throughout working. Soc. Glass Technol.J. 18, 323–335 (1934)
Makishima, A., Mackenzie, J.D.: Direct calculation of Younger’s moidulus of glass. J. Non-Cryst. Solids 12(1), 35–45 (1973).
Martienssen, W., Warlimont, H.: Springer Handbook of Condensed Matter and Supplies Knowledge (2005)
MatWeb: C-Glass Fiber.
Montazerian, M., Singh, S.P., Zanotto, E.: An evaluation of glassceramic analysis and commercialization. Am. Ceram. Soc. Bull. 94, 30–35 (2015)
Morey, G.W.: The impact of boric oxide on the devitrification of the soda-lime-silica glass Na2O–CaO–B2O3–SiO21.
Quat. Syst. 15(9), 457–475 (1932).
Mueller, J., Boehm, M., Drummond, C.: Course of CRT waste glass processing: electronics recycling trade communication. Waste Manag. 32, 1560–1565 (2012).
Nationwide Institutes of Well being (NIH): PubChem database.
Oikonomopoulou, F., Bristogianni, T., Veer, F.A., Nijsse, R.: The development of the crystal Homes façade: challenges and improvements. Glass Struct. Eng. 3(1), 87–108 (2018a).
Oikonomopoulou, F., Bristogianni, T., Barou, L., Jacobs, E., Frigo, G., Veer, F.A., Nijsse, R.: Interlocking solid glass elements, exploring a demountable dry-assembly structural glass system. Heron 63, 103–138 (2018b)
Oikonomopoulou, F., Bristogianni, T., Barou, L., Veer, F.A., Nijsse, R.: The potential of solid glass in structural functions. Classes realized from large-scale castings and stateof-the artwork load-bearing solid glass in structure. J. Construct. Eng. 20, 213–234 (2018c).
Priven, A.I.: Analysis of the fraction of fourfold-coordinated boron in oxide glasses from their composition. Glass Phys. Chem. 26(5), 441–454 (2000).
Quinn, G.D.: NIST Really useful Follow Information: Fractography of Ceramics and Glasses, 2nd Ed. (2016)
Quinn, G.D., Swab, J.J.: Fracture toughness of glasses as measured by the SCF and SEPB strategies. J. Eur. Ceram. Soc. 37(14), 4243–4257 (2017).
Quinn, G.D., Ives, L.Okay., Jahanmir, S.: On the character of machining cracks in floor ceramics: part II—comparability to othersilicon nitrides and injury maps. Mach. Sci. Technol. 9, 211–237 (2005).
Quinn, G.D., Sparenberg, B.T., Koshy, P., Ives, L.Okay., Jahanmir, S., Arola, D.D.: Flexural power of ceramic and glass rods. J. Check. Eval. 37, 222–244 (2009).
Schott: Tubular Glass Photobioreactors (2015a)
Schott: NEXTREMA® (2015b)
Schott: DURAN Technical Knowledge (2017)
Sehgal, J., Ito, S.: A New Low-Brittleness Glass within the SodaLime-Silica Glass Household. 81(9), 2485–2488 (1998).
Shelby, J.E.: Introduction to Glass Science and Know-how. The Royal Society of Chemistry, UK (2005)
Silva, R.V., de Brito, J., Lye, C.Q., Dhir, R.Okay.: The position of glass waste within the manufacturing of ceramic-based merchandise and different functions: a evaluation. J. Clear. Prod. 167, 346–364 (2017).
Songhan Plastic Know-how Co., L.: Schott Nextrematextcircled 724-3, 712-3 Glass Ceramic 
Specialty Glass Merchandise: STARPHIRE Extremely-ClearTM Soda Lime Glass
Thompson, D.A.: Low liquidus glasses for tv tube faceplates. US Patent US4331770A (1980)
Veer, F.A.: The power of glass, a nontransparent worth. Heron 52, 87–104 (2007)
Veer, F.A., Rodichev, Y.: The structural power of glass: hidden injury. Power Mater. 43, 302–315 (2011).
Volf, M.B.: Chemical Strategy to Glass. Elsevier, Amsterdam (1984)
Yamane, M., Mackenzie, J.D.: Vicker’s Hardness of glass. J. Non-Cryst. Solids 15(2), 153–164 (1974).
Yankelevsky, D., Spiller, Okay., Packer, J., Seica, M.: Fracture traits of laboratory-tested soda lime glass specimens. Can. J. Civ. Eng. (2016).
Yun, Y.H., Bray, P.J.: Nuclear magnetic resonance research of the glasses within the system Na2O–B2O3–SiO2. J. NonCryst. Solids 27(3), 363–380 (1978).
Zhdanov, S.P., Shmidel’, G.: Coordination state of boron in sodium borosilicate glasses from NMR information. Fiz. Khim. Stekla 1(5), 452–456 (1975)

Supply hyperlink

About Author

Leave a Reply

Leave a Reply

Your email address will not be published.

Translate »