International Construction Specialty Conference of the Canadian Society for Civil Engineering (ICSC) (5th : 2015)

Reducing heat island effect by using recycled glass cullet in asphalt shingles Kiletico, Micah J.; Hassan, Marwa M.; Mohammad, Louay N. Jun 30, 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52660-Kiletico_M_et_al_ICSC15_005_Reducing.pdf [ 458.08kB ]
52660-Kiletico_M_et_al_ICSC15_005_Reducing_Heat_Island_slides.pdf [ 888.74kB ]
Metadata
JSON: 52660-1.0076476.json
JSON-LD: 52660-1.0076476-ld.json
RDF/XML (Pretty): 52660-1.0076476-rdf.xml
RDF/JSON: 52660-1.0076476-rdf.json
Turtle: 52660-1.0076476-turtle.txt
N-Triples: 52660-1.0076476-rdf-ntriples.txt
Original Record: 52660-1.0076476-source.json
Full Text
52660-1.0076476-fulltext.txt
Citation
52660-1.0076476.ris

Full Text

5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   REDUCING HEAT ISLAND EFFECT BY USING RECYCLED GLASS CULLET IN ASPHALT SHINGLES   Micah J. Kiletico1, Marwa M. Hassan2, and Louay N. Mohammad3 1 Graduate Research Assistant, Louisiana State University, USA 2 Construction Education Trust Fund (CETF) Ethics Associate Professor, Department of Construction Management, Louisiana State University, Baton Rouge, LA 70803-6419, USA 3 Construction Education Trust Fund (CETF) Ethics Associate Professor, Department of Construction Management, Louisiana State University, Baton Rouge, LA 70803-6419, USA Abstract: As an approach to mitigate the harmful effects of Urban Heat Island (UHI), the use of glass cullet in the production of asphalt roof shingles has the potential to be employed as a cool roof strategy.  The objective of this study was to test the hypothesis that the use of recycled glass cullet increases the solar reflectance index (SRI) without affecting the performance of asphalt roof shingles.  In order to evaluate the feasibility of using recycled glass cullet in this new application, the engineering properties of glass cullet were investigated and compared to conventional aggregates used in the production of asphalt roof shingles.  Laboratory shingle specimens were then prepared in order to measure solar reflectance properties and strength characteristics of conventional and recycled glass roof shingles.  Results show that while the use of recycled glass cullet as a replacement to standard ceramic coated black roofing granules on the top surface of asphalt shingles increased the SRI, the addition of white pigment powder (anatase ultra-fine titanium dioxide [TiO2] particles passing mesh #320) to the surface granules greatly improves the reflectance properties of the roof to a level that meets the cool roof threshold. 1 INTRODUCTION With the continuous consumption of energy from non-renewable sources, many cities worldwide with a population that equals or exceeds 1 million people experience an increase in annual mean air temperature of 1-3°C when compared to its surroundings.  Further, evenings can experience a difference as high as 12°C (USEPA 2014).  This phenomenon is known as heat island effect.  It is becoming increasingly intense as summertime temperatures rise due to global warming.  The objective of this study was to evaluate the feasibility of integrating recycled glass cullet in the manufacturing of fiberglass roof shingles and to investigate the interactions that occur when recycled glass cullet is blended with traditional roofing materials.  The impact of diverting materials from disposal and reusing Construction and Demolition (C&D) materials includes reduced extraction and consumption of virgin resources.  Glass food and beverage containers can be recycled endlessly, and economic benefits of the reuse of glass include: cullet cost less than raw materials, reduces energy demand, prolongs furnace life, and creates 8 jobs for every 1,000 tons recycled (Glass Packaging Institute 2013).  Environmentally, benefits include the reduction of emissions, energy consumption, consumption of raw materials, and waste ending in landfills.  For every six tons of recycled container glass used, a ton of carbon dioxide, a greenhouse gas, is reduced (Glass Packaging Institute 2013). 005-1 2 BACKGROUND Recycling of glass cullet can be categorized into two main groups: (1) new glass and bottle containers; and (2) other applications.  Secondary recycling applications are categorized into eight main groups (Reindl 2003): (1) building materials; (2) concrete production; (3) construction aggregates; (4) industrial mineral uses; (5) building insulation; (6) asphalt paving; (7) remelt; and (8) others.  While the recycling of glass cullet is beneficial in most applications, the performance of the product should not be compromised as compared to products prepared with virgin materials.  Numerous advantages can result from the recycling of glass cullet: (1) reduced emissions and energy consumption during processing and manufacturing of virgin materials; (2) reduced consumption of virgin materials; (3) diminished consternation of public concerning emissions; (4) improved economic competitiveness of construction and manufacturing; and (5) reduced glass cullet disposed in landfills. The most common roofing product is asphalt shingles, which account for approximately 85% of roofing products used in the residential sector in the US (Leavell 2006).  A fiberglass roofing shingle is comprised of four major components: a substrate, asphaltic coating with mineral fillers, surface granules, and backdust.  Conventional materials used to produce shingles include fiberglass matting as the substrate, air-blown asphalt as the coating material, and aggregates.  Aggregates are classified by particle size as top surface granules, filler material, or backdust particles.  Typical particle sizes are listed in Table 1.  Materials most widely used as top surface granules include crushed slate, basalt, and trap rock (Pagen et al. 1986).  Finer silt-sized granules are used as backdust, and ideal materials are non-cementitious minerals, such as mica flakes, talc, or sand (Bondoc et al. 1988).  Examples of filler material include fly ash, recycled rubber, recycled shingles, fine grained carbonate rock, dolomite, trap rock, sand, stone dust, and limestone.  Limestone has dominated the market due to its naturally occurring abundance, satisfactory performance, and positive reactions with asphalt as it does not make it brittle or loose granules (Leavell 2006). Table 1.  Typical Particles Size Usage on Roofing Shingles Shingle Component Typical Particle Sizes Mineral Filler Material  45µm - 150µm  Backdust Particles 75µm - 595µm Top Surface Granules 595µm - 2,360µm 3 EXPERIMENTAL PROGRAM The objective of the experimental program was to evaluate the suitability of incorporating recycled glass cullet into the manufacturing process of asphalt roofing shingles and to test the hypothesis that the use of recycled glass cullet will increase solar reflective properties without affecting performance of asphalt roof shingles.  Table 2 provides an overview of the experimental conditions, variables, properties testing, and ASTM specifications.  The experimental program was divided in two phases.  The first phase of the experimental program was to characterize the engineering properties of glass cullet in comparison to conventional materials used in producing fiberglass shingles.  The engineering properties of interest included particle size distribution, specific gravity, absorption, void content, and soundness.  The second phase consisted of preparing laboratory shingles in order to measure and compare the solar reflective properties and strength performance of conventional and recycled glass roof shingles.  Conventional and recycled-glass modified fiberglass reinforced asphalt roofing shingles were prepared in the laboratory by varying the amount of glass cullet and conventional materials used in asphalt roof shingles, and acceptability was based upon the standard specifications described in ASTM D 3462.   3.1 Materials Description and Processing Three sources of recycled glass cullet were collected from C&D processing plants.  Each source was presorted by color and included one source of green glass and two sources of clear glass.  Approximately 005-2 85% of the glass fragments received were larger than the maximum size aggregate (2.36mm) used in producing asphalt roof shingles and needed to be processed into a smaller size.  With the use of a high performance mixer, the fragments were reduced, sieved, and fractionated into sizes used in the production of conventional shingle particle sizes, see Table 1.  The ground glass cullet was then utilized in lieu of conventional mineral aggregates as surface granules, filler material, and backdust.  Table 2.  Description of the Experimental Program and ASTM Test Methods Variable Number of Levels Description Type of Aggregate  Location: Top Surface Granules 4 Green Glass Clear Glass #1 Clear Glass #2 Conventional Materials Usage of Pigment 2 Included with Surface Granules Not included with Surface Granules Phase I: Testing of Aggregate and Asphalt Binder N/A  Particle Size Distribution (ASTM C136 & D422) Specific Gravity (ASTM C128) Absorption (ASTM C128) Void Content (ASTM C1252) Soundness (ASTM C88) Rotational Viscosity (ASTM D4402) Phase II: Testing of Shingle Performance N/A Reflectance and Emittance (ASTM E903 & E408) Tear Strength (ASTM D1922) Conventional shingle aggregates were collected and included ceramic coated roofing granules for the top surface, high calcium limestone filler material, and crushed limestone (stone dust) as the backdust.  The #11 grade top surface igneous roofing granules were coated with a black ceramic pigment and were comprised of pulaskite, which is variation of syenite with little or no quartz.  The limestone filler material was a ground Calcium Carbonate product (96.7% CaCO3) with 98% passing the U.S. Standard Sieve Mesh #50 (0.300mm) and 70% passing the #200 mesh (0.075mm).  For the backdust, crushed limestone #10 was obtained, reduced, and then sieved to the required size and amount. Materials used to produce both conventional and glass modified shingles included air blown asphalt binder and fiberglass matting.  The binder consisted of oxidized air-blown asphalt that is suitable for use as shingle coating.  The fiberglass mat consisted of a non-woven web of glass fibers and served as the substrate for the shingle. A white pigment powder was used in the fabrication of the asphalt shingles.  To this end, the powder was mixed, saturated, and oven-dried with the glass surface granules prior to placement on the surface of the asphaltic coating.  The powder consisted of 98% pure titanium dioxide (TiO2) and was added at 8% by weight of the top surface granules.  The anatase-based ultra-fine powder had a specific gravity of 3.9.  The cool roof attributes of the modified shingle samples were evaluated with and without the addition of pigment powder. 3.2 Asphalt Shingle Preparation In order to prepare fiberglass reinforced asphalt shingles in the laboratory, the process was divided into three steps: preparation of the formwork, preparation of the asphalt blends, and aggregate preparation.  The preparation of the formwork included the thorough cleaning of an ample working surface for the 005-3 placement of a fiberglass mat substrate and then securing metal forms to ensure uniform dimension of the prepared samples.  The prototype sample dimensions were 76mm x 76mm x 3mm. Prior to pouring the asphalt binder in the prepared forms, all aggregates were weighted and fractionated to produce the samples described in Table 3.  For the specimens that did not include pigments, the following weights were used per sample: 25 grams of top surface granules, 5 grams of backdust, and 23.4 grams of filler materials.  The second embodiment consisted of the addition of the white (titanium dioxide) pigment powder to the surface granules.  To include pigment powder, two grams of powder were mixed with 23 grams of surface granules, saturated with water, and then oven-dried.   The filler material was then uniformly mixed with 12.6 grams of liquefied asphalt binder for each sample, which was heated to 204°C.  After mixing the filler material, 36 grams of asphalt coating mixture was reheated to become liquefied at 230°C.  The asphalt coating mixture was then poured in the formwork of each shingle and then placed into the oven at 204°C for impregnation of the fiberglass.  After one hour, the samples were removed from the oven and firm pressure was applied on the top surface granules in order to achieve 100% surface coverage.  Upon cooling of the shingles, the forms were removed and the shingles were separated from the working surface by applying heat using a forced air heat gun.  Finally, the backdust particles were applied to the back surface while keeping the underside heated.   Table 3.  Description of the Shingle Specimens Sample ID Top Surface Material Filler Material X1 Control 1: Ceramic Coated Granules Limestone X2 Control 2: Ceramic Coated Granules Clear Glass 1 A Green Glass Limestone B Clear Glass 1 Limestone C Green Glass Green Glass D Clear Glass 1 Clear Glass 1 C1 Green Glass and Pigments Green Glass D1 Clear Glass 1 and Pigments Clear Glass 1 G1 Clear Glass 2 and Pigments Clear Glass 2 3.3 Reflectance and Emittance Testing   The main purpose of using recycling of broken and waste glass cullet in the manufacturing of asphalt roofing shingle is to alleviate heating and cooling loads in buildings by reducing solar heat flux on the roof and heat island effects by increasing the reflectivity of the roof.  To assess this ability, samples were exposed to a controlled light source from a spectrophotometer in order to measure solar reflective properties in accordance with the Standard Test Method for Solar Absorptance, Reflectance, and Transmittance of Materials Using Integrating Spheres (ASTM E903).  Solar absorptance and reflectance were measured by using a spectroreflectometer with an absolute integrating sphere, 15°/h (Model: LPSR 200IR, AZ Technology).  Measurements of emittance were produced by a spectrafire with an absolute ellipsoidal cavity, 15°/h (Model: TESA 2000, AZ Technology).  Prepared samples for reflectance testing are presented in Table 3; sample dimensions were 76mm x 76mm x 3mm.  To consider variability in reflectance and emittance measurements, two to six readings of three replicates were conducted.   Calculations of Solar Reflectance Index (SRI) under standard solar and ambient conditions are based on Approach II of the Standard Practice for Calculating Solar Reflectance Index of Horizontal and Low-Sloped Opaque Surfaces (ASTM E 1980). Based upon measurements of solar absorptance  and thermal emissivity   of the shingle samples, the SRI was calculated for three convective coefficients   that correspond to low, medium, and high wind conditions at 5, 12, 30 W•m–2•K–1, respectively: 005-4  [1]   [2]   3.4 Tear Strength Testing To assess roof shingle tear strength, prepared asphalt shingles were tested using a pendulum tear strength tester according to the Standard Test Method for Propagation Tear Resistance of Plastic Film and Thin Sheeting by Pendulum Method (ASTM D 1922).  Conventional and glass-modified shingles were compared to industry minimum strength performance standards as specified in Standard Specification for Asphalt Shingles Made from Glass Felt and Surfaced with Mineral Granules (ASTM D 3462).  Tear strength was determined using a Thwing-Albert ProTear Electronic Elmendorf Tear Tester.  Initial tear strength testing included sample types X1, C1, and D1 shown in Table 3.  These samples were made using 36 grams of asphalt coating mixture as outlined in the aforementioned asphalt shingle preparation process.  However, results were inconclusive due to the thickness of the shingle, which increased strength values beyond the testing capabilities of the Elmendorf Tearing device.  To achieve measurable results, the procedure was modified to reduce the thickness of the shingle by using 20 grams of asphalt coating mixture.   4 RESULTS AND ANALYSIS 4.1 Particle Size Analysis Table 4 illustrates the particle size characteristics of the three sources of glass cullet.  Based on these characteristics, collected glass cullet sources appear to be well-graded.  However, glass cullet sources are too coarse to be directly used in glass-modified asphalt shingles without processing.  The glass cullet particle sizes were reduced with the use of a commercial grade blender with 101.6mm blades and were then fractionated into sizes suitable for use in roof shingles as top surface granules (2.36mm – 595µm), backdust particles (595µm – 75µm), and filler material (150µm – 45µm).  The cumulative particle size distribution curves of the ground glass cullet are shown in Figure 1 (a to c) and are compared to the particle sizes of conventional aggregate used as top surface granules, backdust particles, and filler material.  Results show that the laboratory grinding process effectively reduced the size particles of the recycled glass cullet and the nominal maximum aggregate size of the glass cullet, rendering it similar to conventional ceramic coated top surface granules at 2.36mm with 7.4 to 7.7% fines.  The fineness modulus was calculated for each aggregate type, and glass sources and ranged from 2.59 to 2.78 whereas the markedly smaller conventional aggregate fineness modulus was 1.93.   (a) 005-5  (b)  Figure 1: Cumulative Particle Size Distribution Curve of (a) Top Surface Granules (b) Backdust Particles, and (c) Filler Material Table 4.  (a) Particle Size Characteristics of Recycled Glass Cullet Sources and (b) Comparison of Specific Gravity, Absorption, and Soundness of Conventional and Recycled Glass Cullet (a) Characteristics Green Glass Clear Glass 1 Clear Glass 2 Coefficient of Uniformity (Cu) 4.01 4.52 4.21 Coefficient of Curvature  1.43 1.07 1.41 Fineness Modulus (FM) 5.63 5.34 5.63  4.2 Reflectance, Emittance, and Solar Reflective Index (SRI) The mean solar reflectance and thermal emittance results are summarized in Table 5.  In general, reflectance results were influenced only by the top surface material employed and were unaffected by the type of filler material used.  Figure 2 shows the ceramic-coated roofing granules on the top surface of the conventional shingles and compares it to the shingles prepared with top surface glass granules, with and without the addition of white pigment powder.  To measure the effectiveness of glass cullet as a top surface material, mean reflectance values were compared for samples with the same filler material.  Samples A, B, C, and D include glass cullet as a top surface material, which resulted in an increased reflectance as compared to Samples E and F.  The utilization of glass cullet as top surface granules 005-6 resulted in a mean reflectance range increase from 0.029 to 0.050.  T-test statistical analyses with 95% confidence level were performed and showed significance for Samples E vs. A (P-value = 0.004), Samples E vs. B (P-value = 0.015), and Samples F vs. D (P-value = 0.0005). Table 5.  Results of Reflectance, Emittance, and Solar Relfective Index (SRI) ID Material Composition Solar Reflectance Thermal Emittance at 300K Convection Coefficient Top Surface Filler Low Med High 5 W/m²K 12 W/m²K 30 W/m²K X1 Control 1: Ceramic Coated Granules Limestone 0.040 0.917 -0.1 0.1 0.2 X2 Control 2: Ceramic Coated Granules Clear Glass 1 0.036 0.917 -0.6 -0.4 -0.2 A Green Glass Limestone 0.069 0.928 4.4 4.3 4.2 B Clear Glass 1 Limestone 0.090 0.906 5.2 5.6 6.1 C  Green Glass Green Glass 0.069 0.918 3.6 3.7 3.9 D Clear Glass 1 Clear Glass 1 0.086 0.911 5.1 5.5 5.8 C1 Green Glass & Pigments Green Glass 0.263 0.917 28.0 28.1 28.3 D1 Clear Glass 1 & Pigments Clear Glass 1 0.254 0.921 27.1 27.2 27.2 G1 Clear Glass 2 & Pigments Clear Glass 2 0.275 0.933 30.6 30.4 30.2                                                  (a)                                   (b)                                       (c)                                                    (d)                                    (e)                                       (f) 005-7 Figure 2: Qualitative Comparisons of the Top Surface of Shingle Samples Prepared with and without Glass Cullet  Although the increase was statistically significant, glass granules on the top surface alone do not meet the standards to be classified as a cool roof material.  Samples C1, D1, and G1 represent the case in which a white pigment powder was added to enhance the reflectance performance of the glass-modified shingles.  The addition of white pigment powder to the top surface granules resulted in a mean reflectance increase from 0.168 to 0.194 for the samples with glass cullet top surface granules.  T-test statistical analyses with 95% confidence level were conducted and showed significance differences for Samples C vs. C1 (P-value < 0.001) and Samples D vs. D1 (P-value < 0.001).  The overall mean increase in reflectance from conventional materials to top surface glass granules with pigments was 0.218.  A t-test statistical analyses with 95% confidence level was performed and showed significant differences for Samples F vs. D1 (P-value < 0.001).   4.3 Tear Strength Initial testing involved Samples X1, C1, and D1 so that ten constant radius measurements could be tested.  However, the results of initial tear strength testing were inconclusive as no readings were obtained.  Despite using the heaviest pendulum (6400gf) available for the Elmendorf device, the pendulum was unable to swing freely to tear the specimen, as the mass of the pendulum was supported by the sample.  In order to decrease the strength of the shingle and to obtain useable readings, this test method was repeated by modifying the procedure to produce thin laboratory shingles.  The amount of coating was decreased from 36 to 20 grams, which was the thinnest specimen achievable using the procedures outlined in the experimental plan.  Results showed that the average tear strength for both conventional and glass-modified shingles was substantially greater than the minimum shear strength of 1,700gf recommended by ASTM D3462, and results of a one-way ANOVA showed no significant difference between means at a level of significance of 0.0885.  It is recommended to develop assembly methods to decrease the thickness of the coating layer of the laboratory samples and to pursue the use of industrial manufacturing processes in order to further evaluate the effects of glass fillers on actual production samples 4.4 Economic Evaluation of Glass Modified Asphalt Shingles To perform a cost analysis of glass modified asphalt roofing shingles compared to conventional shingles, material costs for conventional aggregates and glass cullet as top surface granules and as filler material were estimated.  Ceramic coated black roofing granules were quoted at $145-185 per ton and approximately $100 per ton for the uncoated headlap granules by Specialty Granules Inc. (SGI) and 3M Industrial Mineral Division.  Because shipping and freight costs are so varied, the prices were given as free on board (FOB) at the granule manufacturing plant, where the buyer pays for all transportation costs.  High calcium limestone filler material was quoted between $17-40 per ton by Lhoist North America and Great Lakes Calcium.  The estimated price for recycled glass was provided by Strategic Materials Inc. and Dlubak Glass, and roofing granules were approximately $131-135 per ton and $150 per ton for filler material.  However, if the market demand increases, the price for glass roofing aggregates could eventually compete with container cullet pricing and 15.8mm minus plate glass recycling at $78-100 per ton.  The cost of obtaining white pigment powder at a rate of 8% by weight of the surface granules was $3.00 per pound.   According to the United States Department of Energy (USDOE), although cool asphalt shingles currently sell for up to $0.50 per ft2 more than conventional asphalt shingles (USDOE, 2013), the cost increase associated with implementing recycled glass granules with pigments is approximately $0.112 per ft2.  Conventional ceramic coated granules, which are utilized at a rate of 0.50 pounds per ft2, can be replaced by glass granules without the use of pigment powder for a cost reduction of $0.008 per ft2 in material cost alone.  Glass filler material at a rate of 0.46 pounds per ft2 can be utilized at a cost increase of $0.03 per ft2 in place of limestone filler material.  However, since filler material had no effect on reflectance or emittance, it was not included in the economic analysis.  According to the Department of Energy Cool Roof Calculator, the estimated savings associated with changing reflectance to 0.30 from a black roof for residential homes in Baton Rouge, LA is $0.061 per ft2 per year, resulting in $91.50 savings per year for a 005-8 typical 1500 ft2 residence (USDOE, 2014).  With the annual savings achieved from increasing reflectance, the payback period to offset the increased cost of glass granules with pigments is estimated at 1.8 years, which is a relatively short period of time as compared to the service life of a residential roof. 5 CONCLUSIONS Results show that glass cullet received from source recycling plants are generally coarse grained aggregate and therefore must be ground or crushed in order to be applied into the manufacturing process of asphalt roof shingles.  Typical asphalt shingles are characterized by low reflectance, and results show that reflectance values were influenced only by the type of top surface aggregate utilized, and the type of filler material had no effect on reflectance values.  The results show that a typical black ceramic coated asphalt shingle produced reflectance values from of 0.036-0.040.  Replacing the top surface granules with green glass produced reflectance values of 0.069 and clear glass from 0.086-0.090.  In order to achieve cool roof attributes, the addition of a white pigment mixed together with the top surface granules increased reflectance to 0.263 for green glass and 0.254-0.275 for clear glass.   Glass modified shingles were tested for durability by measuring resistance to tearing.  Conventional and glass modified shingles exceeded the minimum standard for tear strength of 1700 grams and showed no significant differences between the means. Compared to a conventional black shingles, initial costs analysis shows only a modest increase in material costs of $0.112 per ft2 to produce glass modified asphalt shingles with pigments, but annual savings of $0.061 per ft2 per year in building energy consumption can be achieved through the increased reflectance. Based on the results of this study, it is concluded that glass cullet can be successfully blended with conventional materials to produce a sustainable asphalt shingle that has a solar reflectance of greater than 25% without compromising performance.   6 ACKNOWLEDGEMENTS The author would like to acknowledge Ghesquiere Plastic Testing Inc. for providing tear strength testing, Lhoist North America for providing limestone filler material, BlendTec for providing a mixer,  Strategic Materials Inc. for providing glass cullet, 3M Industrial Mineral Products for ceramic coated roofing granules, and the Louisiana Transportation Research Center (LTRC) for granting us access to their laboratories. 7 REFERENCES Bondoc, Alfredo A., Duane A. Davis, Stanley P. Frankoski, and Bruno E. Magnus. United States of America Patent 4717614. 1988. Glass Packaging Institute. October 23, 2013. http://www.gpi.org/recycling/glass-recycling-facts. Leavell, Daniel N. "Roofing Materials." In Industrial Rocks and Minerals, 7th Edition, by Metallurgy and Exploration Society for Mining, 1173-1178. 2006. Pagen, Charles A, George Stepien, Jr., and Paul A. Morris. United States of America Patent 4,588,634. 1986. Reindl, John. "Reuse/Recycling of Glass Cullet for Non-Container Uses." Dane County Department of Public Works, 2003. US Department of Energy, U.S. Department of Energy. DOE Cool Roof Calculator. February 18, 2014. http://web.ornl.gov/sci/roofs+walls/facts/CoolCalcEnergy.htm. U.S. Department of Energy, Office of Energy Efficiency & Renewable Energy. Cool roofs are ready to save energy, cool urban heat islands, and help slow global warming. December 2013. https://www1.eere.energy.gov/buildings/pdfs/cool_roof_fact_sheet.pdf. USEPA. "U.S. Environmental Protection Agency." State and Local Climate and Energy Program, Heat Island Effect. February 21, 2014. http://www.epa.gov/hiri/index.htm.    005-9  5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   REDUCING HEAT ISLAND EFFECT BY USING RECYCLED GLASS CULLET IN ASPHALT SHINGLES   Micah J. Kiletico1, Marwa M. Hassan2, and Louay N. Mohammad3 1 Graduate Research Assistant, Louisiana State University, USA 2 Construction Education Trust Fund (CETF) Ethics Associate Professor, Department of Construction Management, Louisiana State University, Baton Rouge, LA 70803-6419, USA 3 Construction Education Trust Fund (CETF) Ethics Associate Professor, Department of Construction Management, Louisiana State University, Baton Rouge, LA 70803-6419, USA Abstract: As an approach to mitigate the harmful effects of Urban Heat Island (UHI), the use of glass cullet in the production of asphalt roof shingles has the potential to be employed as a cool roof strategy.  The objective of this study was to test the hypothesis that the use of recycled glass cullet increases the solar reflectance index (SRI) without affecting the performance of asphalt roof shingles.  In order to evaluate the feasibility of using recycled glass cullet in this new application, the engineering properties of glass cullet were investigated and compared to conventional aggregates used in the production of asphalt roof shingles.  Laboratory shingle specimens were then prepared in order to measure solar reflectance properties and strength characteristics of conventional and recycled glass roof shingles.  Results show that while the use of recycled glass cullet as a replacement to standard ceramic coated black roofing granules on the top surface of asphalt shingles increased the SRI, the addition of white pigment powder (anatase ultra-fine titanium dioxide [TiO2] particles passing mesh #320) to the surface granules greatly improves the reflectance properties of the roof to a level that meets the cool roof threshold. 1 INTRODUCTION With the continuous consumption of energy from non-renewable sources, many cities worldwide with a population that equals or exceeds 1 million people experience an increase in annual mean air temperature of 1-3°C when compared to its surroundings.  Further, evenings can experience a difference as high as 12°C (USEPA 2014).  This phenomenon is known as heat island effect.  It is becoming increasingly intense as summertime temperatures rise due to global warming.  The objective of this study was to evaluate the feasibility of integrating recycled glass cullet in the manufacturing of fiberglass roof shingles and to investigate the interactions that occur when recycled glass cullet is blended with traditional roofing materials.  The impact of diverting materials from disposal and reusing Construction and Demolition (C&D) materials includes reduced extraction and consumption of virgin resources.  Glass food and beverage containers can be recycled endlessly, and economic benefits of the reuse of glass include: cullet cost less than raw materials, reduces energy demand, prolongs furnace life, and creates 8 jobs for every 1,000 tons recycled (Glass Packaging Institute 2013).  Environmentally, benefits include the reduction of emissions, energy consumption, consumption of raw materials, and waste ending in landfills.  For every six tons of recycled container glass used, a ton of carbon dioxide, a greenhouse gas, is reduced (Glass Packaging Institute 2013). 005-1 2 BACKGROUND Recycling of glass cullet can be categorized into two main groups: (1) new glass and bottle containers; and (2) other applications.  Secondary recycling applications are categorized into eight main groups (Reindl 2003): (1) building materials; (2) concrete production; (3) construction aggregates; (4) industrial mineral uses; (5) building insulation; (6) asphalt paving; (7) remelt; and (8) others.  While the recycling of glass cullet is beneficial in most applications, the performance of the product should not be compromised as compared to products prepared with virgin materials.  Numerous advantages can result from the recycling of glass cullet: (1) reduced emissions and energy consumption during processing and manufacturing of virgin materials; (2) reduced consumption of virgin materials; (3) diminished consternation of public concerning emissions; (4) improved economic competitiveness of construction and manufacturing; and (5) reduced glass cullet disposed in landfills. The most common roofing product is asphalt shingles, which account for approximately 85% of roofing products used in the residential sector in the US (Leavell 2006).  A fiberglass roofing shingle is comprised of four major components: a substrate, asphaltic coating with mineral fillers, surface granules, and backdust.  Conventional materials used to produce shingles include fiberglass matting as the substrate, air-blown asphalt as the coating material, and aggregates.  Aggregates are classified by particle size as top surface granules, filler material, or backdust particles.  Typical particle sizes are listed in Table 1.  Materials most widely used as top surface granules include crushed slate, basalt, and trap rock (Pagen et al. 1986).  Finer silt-sized granules are used as backdust, and ideal materials are non-cementitious minerals, such as mica flakes, talc, or sand (Bondoc et al. 1988).  Examples of filler material include fly ash, recycled rubber, recycled shingles, fine grained carbonate rock, dolomite, trap rock, sand, stone dust, and limestone.  Limestone has dominated the market due to its naturally occurring abundance, satisfactory performance, and positive reactions with asphalt as it does not make it brittle or loose granules (Leavell 2006). Table 1.  Typical Particles Size Usage on Roofing Shingles Shingle Component Typical Particle Sizes Mineral Filler Material  45µm - 150µm  Backdust Particles 75µm - 595µm Top Surface Granules 595µm - 2,360µm 3 EXPERIMENTAL PROGRAM The objective of the experimental program was to evaluate the suitability of incorporating recycled glass cullet into the manufacturing process of asphalt roofing shingles and to test the hypothesis that the use of recycled glass cullet will increase solar reflective properties without affecting performance of asphalt roof shingles.  Table 2 provides an overview of the experimental conditions, variables, properties testing, and ASTM specifications.  The experimental program was divided in two phases.  The first phase of the experimental program was to characterize the engineering properties of glass cullet in comparison to conventional materials used in producing fiberglass shingles.  The engineering properties of interest included particle size distribution, specific gravity, absorption, void content, and soundness.  The second phase consisted of preparing laboratory shingles in order to measure and compare the solar reflective properties and strength performance of conventional and recycled glass roof shingles.  Conventional and recycled-glass modified fiberglass reinforced asphalt roofing shingles were prepared in the laboratory by varying the amount of glass cullet and conventional materials used in asphalt roof shingles, and acceptability was based upon the standard specifications described in ASTM D 3462.   3.1 Materials Description and Processing Three sources of recycled glass cullet were collected from C&D processing plants.  Each source was presorted by color and included one source of green glass and two sources of clear glass.  Approximately 005-2 85% of the glass fragments received were larger than the maximum size aggregate (2.36mm) used in producing asphalt roof shingles and needed to be processed into a smaller size.  With the use of a high performance mixer, the fragments were reduced, sieved, and fractionated into sizes used in the production of conventional shingle particle sizes, see Table 1.  The ground glass cullet was then utilized in lieu of conventional mineral aggregates as surface granules, filler material, and backdust.  Table 2.  Description of the Experimental Program and ASTM Test Methods Variable Number of Levels Description Type of Aggregate  Location: Top Surface Granules 4 Green Glass Clear Glass #1 Clear Glass #2 Conventional Materials Usage of Pigment 2 Included with Surface Granules Not included with Surface Granules Phase I: Testing of Aggregate and Asphalt Binder N/A  Particle Size Distribution (ASTM C136 & D422) Specific Gravity (ASTM C128) Absorption (ASTM C128) Void Content (ASTM C1252) Soundness (ASTM C88) Rotational Viscosity (ASTM D4402) Phase II: Testing of Shingle Performance N/A Reflectance and Emittance (ASTM E903 & E408) Tear Strength (ASTM D1922) Conventional shingle aggregates were collected and included ceramic coated roofing granules for the top surface, high calcium limestone filler material, and crushed limestone (stone dust) as the backdust.  The #11 grade top surface igneous roofing granules were coated with a black ceramic pigment and were comprised of pulaskite, which is variation of syenite with little or no quartz.  The limestone filler material was a ground Calcium Carbonate product (96.7% CaCO3) with 98% passing the U.S. Standard Sieve Mesh #50 (0.300mm) and 70% passing the #200 mesh (0.075mm).  For the backdust, crushed limestone #10 was obtained, reduced, and then sieved to the required size and amount. Materials used to produce both conventional and glass modified shingles included air blown asphalt binder and fiberglass matting.  The binder consisted of oxidized air-blown asphalt that is suitable for use as shingle coating.  The fiberglass mat consisted of a non-woven web of glass fibers and served as the substrate for the shingle. A white pigment powder was used in the fabrication of the asphalt shingles.  To this end, the powder was mixed, saturated, and oven-dried with the glass surface granules prior to placement on the surface of the asphaltic coating.  The powder consisted of 98% pure titanium dioxide (TiO2) and was added at 8% by weight of the top surface granules.  The anatase-based ultra-fine powder had a specific gravity of 3.9.  The cool roof attributes of the modified shingle samples were evaluated with and without the addition of pigment powder. 3.2 Asphalt Shingle Preparation In order to prepare fiberglass reinforced asphalt shingles in the laboratory, the process was divided into three steps: preparation of the formwork, preparation of the asphalt blends, and aggregate preparation.  The preparation of the formwork included the thorough cleaning of an ample working surface for the 005-3 placement of a fiberglass mat substrate and then securing metal forms to ensure uniform dimension of the prepared samples.  The prototype sample dimensions were 76mm x 76mm x 3mm. Prior to pouring the asphalt binder in the prepared forms, all aggregates were weighted and fractionated to produce the samples described in Table 3.  For the specimens that did not include pigments, the following weights were used per sample: 25 grams of top surface granules, 5 grams of backdust, and 23.4 grams of filler materials.  The second embodiment consisted of the addition of the white (titanium dioxide) pigment powder to the surface granules.  To include pigment powder, two grams of powder were mixed with 23 grams of surface granules, saturated with water, and then oven-dried.   The filler material was then uniformly mixed with 12.6 grams of liquefied asphalt binder for each sample, which was heated to 204°C.  After mixing the filler material, 36 grams of asphalt coating mixture was reheated to become liquefied at 230°C.  The asphalt coating mixture was then poured in the formwork of each shingle and then placed into the oven at 204°C for impregnation of the fiberglass.  After one hour, the samples were removed from the oven and firm pressure was applied on the top surface granules in order to achieve 100% surface coverage.  Upon cooling of the shingles, the forms were removed and the shingles were separated from the working surface by applying heat using a forced air heat gun.  Finally, the backdust particles were applied to the back surface while keeping the underside heated.   Table 3.  Description of the Shingle Specimens Sample ID Top Surface Material Filler Material X1 Control 1: Ceramic Coated Granules Limestone X2 Control 2: Ceramic Coated Granules Clear Glass 1 A Green Glass Limestone B Clear Glass 1 Limestone C Green Glass Green Glass D Clear Glass 1 Clear Glass 1 C1 Green Glass and Pigments Green Glass D1 Clear Glass 1 and Pigments Clear Glass 1 G1 Clear Glass 2 and Pigments Clear Glass 2 3.3 Reflectance and Emittance Testing   The main purpose of using recycling of broken and waste glass cullet in the manufacturing of asphalt roofing shingle is to alleviate heating and cooling loads in buildings by reducing solar heat flux on the roof and heat island effects by increasing the reflectivity of the roof.  To assess this ability, samples were exposed to a controlled light source from a spectrophotometer in order to measure solar reflective properties in accordance with the Standard Test Method for Solar Absorptance, Reflectance, and Transmittance of Materials Using Integrating Spheres (ASTM E903).  Solar absorptance and reflectance were measured by using a spectroreflectometer with an absolute integrating sphere, 15°/h (Model: LPSR 200IR, AZ Technology).  Measurements of emittance were produced by a spectrafire with an absolute ellipsoidal cavity, 15°/h (Model: TESA 2000, AZ Technology).  Prepared samples for reflectance testing are presented in Table 3; sample dimensions were 76mm x 76mm x 3mm.  To consider variability in reflectance and emittance measurements, two to six readings of three replicates were conducted.   Calculations of Solar Reflectance Index (SRI) under standard solar and ambient conditions are based on Approach II of the Standard Practice for Calculating Solar Reflectance Index of Horizontal and Low-Sloped Opaque Surfaces (ASTM E 1980). Based upon measurements of solar absorptance  and thermal emissivity   of the shingle samples, the SRI was calculated for three convective coefficients   that correspond to low, medium, and high wind conditions at 5, 12, 30 W•m–2•K–1, respectively: 005-4  [1]   [2]   3.4 Tear Strength Testing To assess roof shingle tear strength, prepared asphalt shingles were tested using a pendulum tear strength tester according to the Standard Test Method for Propagation Tear Resistance of Plastic Film and Thin Sheeting by Pendulum Method (ASTM D 1922).  Conventional and glass-modified shingles were compared to industry minimum strength performance standards as specified in Standard Specification for Asphalt Shingles Made from Glass Felt and Surfaced with Mineral Granules (ASTM D 3462).  Tear strength was determined using a Thwing-Albert ProTear Electronic Elmendorf Tear Tester.  Initial tear strength testing included sample types X1, C1, and D1 shown in Table 3.  These samples were made using 36 grams of asphalt coating mixture as outlined in the aforementioned asphalt shingle preparation process.  However, results were inconclusive due to the thickness of the shingle, which increased strength values beyond the testing capabilities of the Elmendorf Tearing device.  To achieve measurable results, the procedure was modified to reduce the thickness of the shingle by using 20 grams of asphalt coating mixture.   4 RESULTS AND ANALYSIS 4.1 Particle Size Analysis Table 4 illustrates the particle size characteristics of the three sources of glass cullet.  Based on these characteristics, collected glass cullet sources appear to be well-graded.  However, glass cullet sources are too coarse to be directly used in glass-modified asphalt shingles without processing.  The glass cullet particle sizes were reduced with the use of a commercial grade blender with 101.6mm blades and were then fractionated into sizes suitable for use in roof shingles as top surface granules (2.36mm – 595µm), backdust particles (595µm – 75µm), and filler material (150µm – 45µm).  The cumulative particle size distribution curves of the ground glass cullet are shown in Figure 1 (a to c) and are compared to the particle sizes of conventional aggregate used as top surface granules, backdust particles, and filler material.  Results show that the laboratory grinding process effectively reduced the size particles of the recycled glass cullet and the nominal maximum aggregate size of the glass cullet, rendering it similar to conventional ceramic coated top surface granules at 2.36mm with 7.4 to 7.7% fines.  The fineness modulus was calculated for each aggregate type, and glass sources and ranged from 2.59 to 2.78 whereas the markedly smaller conventional aggregate fineness modulus was 1.93.   (a) 005-5  (b)  Figure 1: Cumulative Particle Size Distribution Curve of (a) Top Surface Granules (b) Backdust Particles, and (c) Filler Material Table 4.  (a) Particle Size Characteristics of Recycled Glass Cullet Sources and (b) Comparison of Specific Gravity, Absorption, and Soundness of Conventional and Recycled Glass Cullet (a) Characteristics Green Glass Clear Glass 1 Clear Glass 2 Coefficient of Uniformity (Cu) 4.01 4.52 4.21 Coefficient of Curvature  1.43 1.07 1.41 Fineness Modulus (FM) 5.63 5.34 5.63  4.2 Reflectance, Emittance, and Solar Reflective Index (SRI) The mean solar reflectance and thermal emittance results are summarized in Table 5.  In general, reflectance results were influenced only by the top surface material employed and were unaffected by the type of filler material used.  Figure 2 shows the ceramic-coated roofing granules on the top surface of the conventional shingles and compares it to the shingles prepared with top surface glass granules, with and without the addition of white pigment powder.  To measure the effectiveness of glass cullet as a top surface material, mean reflectance values were compared for samples with the same filler material.  Samples A, B, C, and D include glass cullet as a top surface material, which resulted in an increased reflectance as compared to Samples E and F.  The utilization of glass cullet as top surface granules 005-6 resulted in a mean reflectance range increase from 0.029 to 0.050.  T-test statistical analyses with 95% confidence level were performed and showed significance for Samples E vs. A (P-value = 0.004), Samples E vs. B (P-value = 0.015), and Samples F vs. D (P-value = 0.0005). Table 5.  Results of Reflectance, Emittance, and Solar Relfective Index (SRI) ID Material Composition Solar Reflectance Thermal Emittance at 300K Convection Coefficient Top Surface Filler Low Med High 5 W/m²K 12 W/m²K 30 W/m²K X1 Control 1: Ceramic Coated Granules Limestone 0.040 0.917 -0.1 0.1 0.2 X2 Control 2: Ceramic Coated Granules Clear Glass 1 0.036 0.917 -0.6 -0.4 -0.2 A Green Glass Limestone 0.069 0.928 4.4 4.3 4.2 B Clear Glass 1 Limestone 0.090 0.906 5.2 5.6 6.1 C  Green Glass Green Glass 0.069 0.918 3.6 3.7 3.9 D Clear Glass 1 Clear Glass 1 0.086 0.911 5.1 5.5 5.8 C1 Green Glass & Pigments Green Glass 0.263 0.917 28.0 28.1 28.3 D1 Clear Glass 1 & Pigments Clear Glass 1 0.254 0.921 27.1 27.2 27.2 G1 Clear Glass 2 & Pigments Clear Glass 2 0.275 0.933 30.6 30.4 30.2                                                  (a)                                   (b)                                       (c)                                                    (d)                                    (e)                                       (f) 005-7 Figure 2: Qualitative Comparisons of the Top Surface of Shingle Samples Prepared with and without Glass Cullet  Although the increase was statistically significant, glass granules on the top surface alone do not meet the standards to be classified as a cool roof material.  Samples C1, D1, and G1 represent the case in which a white pigment powder was added to enhance the reflectance performance of the glass-modified shingles.  The addition of white pigment powder to the top surface granules resulted in a mean reflectance increase from 0.168 to 0.194 for the samples with glass cullet top surface granules.  T-test statistical analyses with 95% confidence level were conducted and showed significance differences for Samples C vs. C1 (P-value < 0.001) and Samples D vs. D1 (P-value < 0.001).  The overall mean increase in reflectance from conventional materials to top surface glass granules with pigments was 0.218.  A t-test statistical analyses with 95% confidence level was performed and showed significant differences for Samples F vs. D1 (P-value < 0.001).   4.3 Tear Strength Initial testing involved Samples X1, C1, and D1 so that ten constant radius measurements could be tested.  However, the results of initial tear strength testing were inconclusive as no readings were obtained.  Despite using the heaviest pendulum (6400gf) available for the Elmendorf device, the pendulum was unable to swing freely to tear the specimen, as the mass of the pendulum was supported by the sample.  In order to decrease the strength of the shingle and to obtain useable readings, this test method was repeated by modifying the procedure to produce thin laboratory shingles.  The amount of coating was decreased from 36 to 20 grams, which was the thinnest specimen achievable using the procedures outlined in the experimental plan.  Results showed that the average tear strength for both conventional and glass-modified shingles was substantially greater than the minimum shear strength of 1,700gf recommended by ASTM D3462, and results of a one-way ANOVA showed no significant difference between means at a level of significance of 0.0885.  It is recommended to develop assembly methods to decrease the thickness of the coating layer of the laboratory samples and to pursue the use of industrial manufacturing processes in order to further evaluate the effects of glass fillers on actual production samples 4.4 Economic Evaluation of Glass Modified Asphalt Shingles To perform a cost analysis of glass modified asphalt roofing shingles compared to conventional shingles, material costs for conventional aggregates and glass cullet as top surface granules and as filler material were estimated.  Ceramic coated black roofing granules were quoted at $145-185 per ton and approximately $100 per ton for the uncoated headlap granules by Specialty Granules Inc. (SGI) and 3M Industrial Mineral Division.  Because shipping and freight costs are so varied, the prices were given as free on board (FOB) at the granule manufacturing plant, where the buyer pays for all transportation costs.  High calcium limestone filler material was quoted between $17-40 per ton by Lhoist North America and Great Lakes Calcium.  The estimated price for recycled glass was provided by Strategic Materials Inc. and Dlubak Glass, and roofing granules were approximately $131-135 per ton and $150 per ton for filler material.  However, if the market demand increases, the price for glass roofing aggregates could eventually compete with container cullet pricing and 15.8mm minus plate glass recycling at $78-100 per ton.  The cost of obtaining white pigment powder at a rate of 8% by weight of the surface granules was $3.00 per pound.   According to the United States Department of Energy (USDOE), although cool asphalt shingles currently sell for up to $0.50 per ft2 more than conventional asphalt shingles (USDOE, 2013), the cost increase associated with implementing recycled glass granules with pigments is approximately $0.112 per ft2.  Conventional ceramic coated granules, which are utilized at a rate of 0.50 pounds per ft2, can be replaced by glass granules without the use of pigment powder for a cost reduction of $0.008 per ft2 in material cost alone.  Glass filler material at a rate of 0.46 pounds per ft2 can be utilized at a cost increase of $0.03 per ft2 in place of limestone filler material.  However, since filler material had no effect on reflectance or emittance, it was not included in the economic analysis.  According to the Department of Energy Cool Roof Calculator, the estimated savings associated with changing reflectance to 0.30 from a black roof for residential homes in Baton Rouge, LA is $0.061 per ft2 per year, resulting in $91.50 savings per year for a 005-8 typical 1500 ft2 residence (USDOE, 2014).  With the annual savings achieved from increasing reflectance, the payback period to offset the increased cost of glass granules with pigments is estimated at 1.8 years, which is a relatively short period of time as compared to the service life of a residential roof. 5 CONCLUSIONS Results show that glass cullet received from source recycling plants are generally coarse grained aggregate and therefore must be ground or crushed in order to be applied into the manufacturing process of asphalt roof shingles.  Typical asphalt shingles are characterized by low reflectance, and results show that reflectance values were influenced only by the type of top surface aggregate utilized, and the type of filler material had no effect on reflectance values.  The results show that a typical black ceramic coated asphalt shingle produced reflectance values from of 0.036-0.040.  Replacing the top surface granules with green glass produced reflectance values of 0.069 and clear glass from 0.086-0.090.  In order to achieve cool roof attributes, the addition of a white pigment mixed together with the top surface granules increased reflectance to 0.263 for green glass and 0.254-0.275 for clear glass.   Glass modified shingles were tested for durability by measuring resistance to tearing.  Conventional and glass modified shingles exceeded the minimum standard for tear strength of 1700 grams and showed no significant differences between the means. Compared to a conventional black shingles, initial costs analysis shows only a modest increase in material costs of $0.112 per ft2 to produce glass modified asphalt shingles with pigments, but annual savings of $0.061 per ft2 per year in building energy consumption can be achieved through the increased reflectance. Based on the results of this study, it is concluded that glass cullet can be successfully blended with conventional materials to produce a sustainable asphalt shingle that has a solar reflectance of greater than 25% without compromising performance.   6 ACKNOWLEDGEMENTS The author would like to acknowledge Ghesquiere Plastic Testing Inc. for providing tear strength testing, Lhoist North America for providing limestone filler material, BlendTec for providing a mixer,  Strategic Materials Inc. for providing glass cullet, 3M Industrial Mineral Products for ceramic coated roofing granules, and the Louisiana Transportation Research Center (LTRC) for granting us access to their laboratories. 7 REFERENCES Bondoc, Alfredo A., Duane A. Davis, Stanley P. Frankoski, and Bruno E. Magnus. United States of America Patent 4717614. 1988. Glass Packaging Institute. October 23, 2013. http://www.gpi.org/recycling/glass-recycling-facts. Leavell, Daniel N. "Roofing Materials." In Industrial Rocks and Minerals, 7th Edition, by Metallurgy and Exploration Society for Mining, 1173-1178. 2006. Pagen, Charles A, George Stepien, Jr., and Paul A. Morris. United States of America Patent 4,588,634. 1986. Reindl, John. "Reuse/Recycling of Glass Cullet for Non-Container Uses." Dane County Department of Public Works, 2003. US Department of Energy, U.S. Department of Energy. DOE Cool Roof Calculator. February 18, 2014. http://web.ornl.gov/sci/roofs+walls/facts/CoolCalcEnergy.htm. U.S. Department of Energy, Office of Energy Efficiency & Renewable Energy. Cool roofs are ready to save energy, cool urban heat islands, and help slow global warming. December 2013. https://www1.eere.energy.gov/buildings/pdfs/cool_roof_fact_sheet.pdf. USEPA. "U.S. Environmental Protection Agency." State and Local Climate and Energy Program, Heat Island Effect. February 21, 2014. http://www.epa.gov/hiri/index.htm.    005-9  REDUCING HEAT ISLAND EFFECT BY USING RECYCLED GLASS CULLET IN ASPHALT SHINGLESMicah J. KileticoMarwa Hassan Louay MohammadIntroduction Continuous use of non-renewable energy sources leads to an increase in urban annual mean air temperature → Urban Heat Island Effects Urban areas are 1 to 3oC warmer than their surroundings. Evening 12o Difference. Exposed and highly absorptive materials used on roofs are a major contributing factor to increasing the heat island effectIntroductionAtlantaIntroduction EPA estimates that 11.5 to 12.8 million tons of glass waste are generated every year  Only 38% being recovered in various recycling activities: New glass and bottle containers/Secondary applications Every metric ton of recycled glass reduces production of carbon dioxide by as much as 315 kilogramsRecycling of Glass Cullet Secondary recycling applications Building materials Concrete production Construction aggregates Industrial mineral uses Building insulation Asphalt paving RemeltObjectives Evaluate the use of recycling glass cullet in the manufacturing of fiberglass roof shinglesReduce disposal problems for glass culletIncrease reflectivity of residential roof as compared to conventional asphalt shinglesBackground Typical particle size used in asphalt shingles:Shingle Component Particle Size Range U.S. Standard SieveTop Surface Granules 2.36mm - 595µm Mesh #8 - #30Backdust Particles 595µm - 75µm Mesh #30 - #200Mineral Filler Material 150µm - 45µm Mesh #100 - #325Experimental Program Three sources of recycled glass cullet were sampled from C&D processing plantsGround glass cullet was used in lieu of conventional mineral aggregatesAnatase-based TiO2 white pigment was added to increase reflectivityExperimental Program Testing of glass cullet and asphalt binder: Particle size distribution Specific gravity Absorption Void content Soundness Rotational viscosity Shingle performance: Reflectance and emittance Tear strengthPrepared Shingle SpecimensSample ID Top Surface Material Filler MaterialX1 Control 1: Ceramic Coated Granules LimestoneX2 Control 2: Ceramic Coated Granules Clear Glass 1A Green Glass LimestoneB Clear Glass 1 LimestoneC Green Glass Green GlassD Clear Glass 1 Clear Glass 1C1 Green Glass and Pigments Green GlassD1 Clear Glass 1 and Pigments Clear Glass 1G1 Clear Glass 2 and Pigments Clear Glass 2Asphalt Shingle Preparation                               (a)                                 (b)                                  (c)                          (d)                                   (e)                                  (f)                                  (g)    (h)                                   (i) Solar Reflectance Measurements Solar Reflection, Absorption, Transmittance and Solar Reflectance Index (SRI) was measured using a spectrophotometer instrument  based on ASTM E903 and ASTM E 1980𝜒𝜒 = (𝛼𝛼 − 0.029𝜖𝜖)(8.797 + ℎ𝑐𝑐)(9.5205𝜖𝜖 + ℎ𝑐𝑐)𝑆𝑆𝑆𝑆𝑆𝑆 = 123.97− 141.35𝜒𝜒 + 9.655𝜒𝜒2Tear Strength Testing Prepared asphalt shingles were tested using a pendulum shear strength according to ASTM D 1922Results and Analysis 85% of glass cullet were too large to be used in producing asphalt roof shinglesGlass cullet was processed to smaller size Particle size distribution for Filler Material01020304050607080901000.0010.010.1110100Percent FinerParticle Diameter (mm)Green GlassClear Glass 1Clear Glass 2Results and Analysis Particle size distribution for top surface granules0204060801000.0010.010.1110100Percent FinerParicle Diameter (mm)CeramicCoatedGranulesGreen GlassClear Glass 1Clear Glass 2Results and Analysis Particle size distribution for Backdust Particles0204060801000.0010.010.1110100Percent FinerParicle Diameter (mm)CrushedLimestoneGreenGlassClear Glass1Clear Glass2Solar Reflectance Measurements The use of glass cullet as top surface granules resulted in a mean reflectance increase from 0.029 to 0.050. The addition of white pigment powder resulted in a significant increase in reflectance from 0.029 to 0.263 Differences were statistically significant (P-value = 0.004) Reflectance levels would meet cool roof thresholdSolar Reflectance AnalysisIDMaterial CompositionSolar ReflectanceThermal Emittance at 300KConvection CoefficientTop Surface FillerLow Med High5 W/m²K 12 W/m²K 30 W/m²KX1Control 1: Ceramic Coated GranulesLimestone 0.040 0.917 -0.1 0.1 0.2X2Control 2: Ceramic Coated GranulesClear Glass 1 0.036 0.917 -0.6 -0.4 -0.2A Green Glass Limestone 0.069 0.928 4.4 4.3 4.2B Clear Glass 1 Limestone 0.090 0.906 5.2 5.6 6.1C Green Glass Green Glass 0.069 0.918 3.6 3.7 3.9D Clear Glass 1 Clear Glass 1 0.086 0.911 5.1 5.5 5.8C1Green Glass & PigmentsGreen Glass 0.263 0.917 28.0 28.1 28.3D1Clear Glass 1 & PigmentsClear Glass 1 0.254 0.921 27.1 27.2 27.2G1Clear Glass 2 & PigmentsClear Glass 2 0.275 0.933 30.6 30.4 30.2Tear Strength Results Results for both conventional and glass-modified shingles were substantially greater than the minimum shear strength of 1,700 gf No significant difference between both sets (P=0.089)Cost Analysis With annual savings from increasing reflectance, payback period is 1.8 yearsConventional AggregatesRecycled Glass Recycled Glass & Pigmentsa) Cost of Top Surface Granules ($/ft2) $0.041 $0.033 $0.033b) Cost of Pigments ($/ft2) - - $0.120c) Cost of Limestone Fillers ($/ft2) $0.007 $0.007 $0.007d) Total Cost of Aggregates [d=a+b+c]($/ft2)$0.048 $0.040 $0.160e)Difference in Cost from Baseline Conventional Materials ($/ft2)Conventional Aggregates ($0.048)Glass Aggregates ($0.040-$0.048=-$0.008)Glass & Pigments ($0.160-$0.048=$0.112)Baseline -$0.008 $0.112f)Reflectance Savings based on Department of EnergyCool Roof Calculator($/ft² per year)Baseline $0.016 $0.061g)Estimated Payback Period [g=e/f](years) BaselineMaterials costs are cheaper in addition to savings in reflectance1.8 yearsConclusions Recycled glass is an effective alternative as a replacement to conventional aggregates in asphalt roof shingles When processed and crushed to size specifications, recycled glass can be used as top surface granules, backdust, or filler material. Conclusions Green glass cullet increased reflectance by 1.7-1.9 times the value of black ceramic coated granules Clear glass cullet reflectance was 2.25-2.50 times greater than conventional aggregate Use of glass cullet with white pigment powder greatly improved surface reflectance Up to a level that meets the cool roof thresholdConclusions Compared to a conventional black shingles, initial costs analysis shows only a modest increase in material costs of $0.112 per ft2 to produce glass modified asphalt shingles with pigments Annual savings of $0.061 per ft2 per year in building energy consumption can be achieved through the increased reflectanceThank You!

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52660.1-0076476/manifest

Comment

Related Items