Evaluating the Feasibility of Foamed Glass Aggregate in Lightweight Concrete Mix Designs

Abstract

Lightweight aggregate concrete is known for its potential to decrease overall building load and cost. Aero Aggregates’ Aerolite is a foamed glass aggregate (FGA) available in seven different sizes which has the potential to replace normal weight aggregates to create lightweight concrete. This research analyzes the feasibility of using FGAs in optimized concrete mix designs and employing those designs in a full-scale building. Nine different mix designs were created using optimization methods, including the Tarantula Curve and 0.45 power chart, to determine the ideal aggregate proportions. All mixes were cast in 0.1 m diameter, 0.2 m tall cylinders and tested after 7 and 28 days to determine unit weight (density), compressive strength, and modulus of elasticity. After testing, the optimal design was identified as 65% coarse and 15% fine aggregates to be replaced with FGAs because it gave the best unit weight and compressive strength for structural lightweight concrete. The optimal concrete mix design was used to create an example building model in RAM Structural Systems to prove that FGA concrete can reduce cost, materials required, and carbon emissions on a larger scale.

1. Introduction

Lightweight aggregate concrete is becoming increasingly popular in structural design due to the benefits of a lighter structure. The concrete is made lighter by replacing the typical gravel coarse aggregates and sand fine aggregates with a lighter weight material. Typically, lightweight aggregates are more porous, increasing water absorption and improving self-curing. This gives the added benefit of increased fire resistance, due to the low thermal conductivity and high heat resistance. It is also known to reduce acoustic transmission and improve the thermal resistance of the concrete. Lightweight concrete can be classified as structural, masonry, or insulating. The American Concrete Institute’s Guide for Structural Lightweight-Aggregate Concrete (ACI 213-14) shares the current data and findings on lightweight concrete, including recommended methods for proportioning, mixing, and placing [1]. Structural lightweight concrete is classified as concrete having a maximum dry unit weight (density) of 18.1 kN/m³ and a minimum 28-day compressive strength of 17.2 MPa.

Aero Aggregate’s Aerolite foamed glass aggregates (FGAs) are made from 99% recycled glass and are 85% lighter than typical aggregates. Currently used mainly for lightweight backfills in construction applications, foamed glass aggregate has the potential to improve the sustainability aspect and decrease the price of projects. FGA is sold by Aero Aggregates in seven different sizes, examples pictured in Figure 1, ranging from 0.03 to 12.5 mm. To create FGA from recycled glass pieces, the glass is heated and mixed with a foaming agent, which creates bubbles and voids within the grain. Using FGA in concrete mixtures has the potential to achieve the benefits of lightweight concrete while improving sustainability by reducing glass waste. All or a percentage of the stone aggregates in concrete can be replaced with FGAs, with the goal of achieving an adequate unit weight and compressive strength for use in structural applications. The strength of the aggregates themselves proves to be important in concrete mixtures as they act as the strength ceiling for the entire concrete member. Aero Aggregates has completed introductory research on the impacts of Aerolite FGA on concrete mixes, but further work can be performed to refine the mixture and gain more knowledge. The focus of this study is to determine the feasibility and benefits of using FGA to create lightweight concrete when compared to normal natural aggregate concrete.

Prior literature and research have analyzed methods for mix proportioning. According to Wang et al. (2018), the first step in the mix proportioning process is selecting the proper aggregate system [2]. Optimizing the aggregate gradations is critical to maintain proper workability in the mixture. The Tarantula Curve is one method developed to determine the optimal amount of aggregate retained on each sieve [2]. The curve utilizes an envelope to maintain the percentages of each aggregate and provide the most economical design. A combined aggregate system that fits within the envelope of the Tarantula Curve provides the best workability while using less cementitious materials. The 0.45 power chart provides a maximum density line to be matched with a combination of coarse and fine aggregates. The study concludes that an optimal aggregate gradation maintains within the envelope of the Tarantula Curve and remains as close to the maximum density line of the 0.45 power chart.

ACI 211.2-98, titled “Standard Practice for Selecting Proportions for Structural Lightweight Concrete”, details methods for selecting proportions for lightweight concrete mixture designs [3]. The strength of structural lightweight concrete mainly depends on the strength of the aggregates used. Lightweight concrete is usually proportioned based on trial mixes followed by adjustments. ACI 211.2-98 details two different methods to create a starting mixture design that can be adjusted and improved after laboratory testing. The first method involves calculations to determine the required slump, nominal maximum aggregate size, mixing water, and water–cement ratio. Method 2 is the preferred option, and it requires testing and adjusting trial mixes until desired properties are reached.

The strength of lightweight concrete is mainly dependent on the aggregate strength, the water–cement ratio, and the strength of the interfacial transition zone, which is the area of the cement paste immediately surrounding the aggregates. Research by Lo et al. (2006) concluded that the strength of the concrete tested is directly proportional to the aggregate crushing strength and aggregate density [4]. The 15 mm aggregate was found to have the highest density of those tested, and therefore highest strength. It was also determined that the number of pores in the cement paste and interfacial zone increases with the water–cement ratio, but this decreases the concrete strength.

Many studies have shown the viability of using glass aggregate in lightweight concrete (Nemes and Jozsa, 2006, Bubenik et al., 2023, and Asztalos and Kocserha, 2020) [5,6,7]. The variables that impact concrete material properties, especially strength, when replacing traditional aggregates with foamed glass aggregate are numerous. Asztalos and Kocserha (2020) used 100% recycled glass foam granules obtained from waste glass as a lightweight concrete aggregate [7]. A series of concrete mixes were proportioned with water-to-cement ratios of 0.4 with 0 to 30 weight percentage of foam glass granules. The results showed that foam glass increased many material properties, such as thermal insulation and density, and decreased its water absorption. It was determined that the addition of glass foam granules significantly reduced the concrete strength. Strength data exhibited large standard deviations, while durability tests, including alkali–silica reactivity (ASR) and thermogravimetric analysis (TGA), confirmed the non-reactive behavior of foam glass aggregates, ensuring long-term structural performance [7].

The potential reactive nature of glass aggregates is of concern. Rajabipour et al. (2015) provide a thorough history and the current state of understanding with respect to reaction mechanisms and the effects of aggregate properties, pore solution composition, and exposure conditions on the severity of ASR [8]. Figueira et al. (2019) present an overview of relevant achievements concerning ASR processes, including conditioning factors, diagnostic and preventive measures, and test methods of ASR [9]. The focus of these studies was not on recycled materials or glass foam; however, many researchers have sought to reduce environmental issues and optimize current production processes by using recycled glass materials as an alternative to common concrete components. Recent investigations included waste glass and demolished brick [10], glass shards from municipal solid waste, [11], recycled waste medical glass [12], and glass foams produced from soda–lime glass waste and rice husk ash [13].

Gencel et al. (2024) noted that foam concrete has generally superior thermal insulation properties compared to conventional concrete, but the high content of Portland cement and high porosity makes foam concrete prone to physico-durability concerns [14]. The research used expanded perlite and fine-sized waste glass sand as the main aggregates in concrete mixes. It was concluded that the addition of glass sand improves physico-mechanical and durability properties of foam concrete and the study showed that it is possible to produce sustainable insulating foam concrete with waste glass sand.

Balachandran et al. (2024) investigated the relationship between the major mechanisms, their relative prevalence during the ASR reaction, and their influence on the pore solution chemistry and nature of ASR products by examining mortar with and without fly ash [15]. It was demonstrated that the addition of fly ash to the mortar resulted in generation of fewer ASR products than those produced in a control system.

Sharma et al. (2021) investigated the effects of foamed glass aggregates on the properties of concrete [16]. It was shown that the aggregate size plays an important role in material strength. Three different sizes of foamed aggregate were investigated, and the maximum strength concrete strength was obtained using the minimum size aggregate. It was concluded that the size of aggregate has a significant impact on the ultimate strength of the concrete. Additionally, it was concluded that the smaller the size of the aggregate, the higher the strength will be, and the larger the size of the aggregate, the lower the strength. The maximum concrete strength achieved was 16 MPa [16].

Research by Osfouri et al. (2025) on sustainable structural lightweight concrete containing foam glass aggregates showed that concrete made with foam glass granules exhibited higher compressive strength compared to concrete incorporating crushed foam glass [17]. Concrete with 20% foam glass replacement could meet lightweight concrete criteria, with a bulk density below 2000 kg/m³ and a compressive strength above 17 MPa. It was noted that compressive strength of granules made from waste glass was greater than concrete with granules made from virgin glass. The study concluded that foam glass is a sustainable alternative to natural aggregates in concrete [17].

Zheng et al. (2025) designed and developed a high-performance lightweight two-stage concrete composite (LTSCC), based on the concept of multiscale applications of various lightweight glass-based materials [18]. A slurry infiltration setup was used to achieve a uniform distribution of the lightweight glass-based materials. The foam glass coarse aggregates were preplaced in the mold and lightweight high-strength slurries that also incorporated foam glass filled the voids under vacuum pressure and gravity. Compressive strengths over 40 MPa were achieved for all mixes investigated. It was concluded that combining high-strength foam glass coarse aggregates with lightweight high-strength slurries can be used to develop high-strength foam glass aggregate concrete [18].

2. Materials and Methods

Each concrete mixture comprises different amounts of the same main components. Coarse aggregate stone, fine aggregate sand, and FGA were all used at varying proportions in different mixes. All mixes are made with Type I Portland cement and use a water–cement ratio ranging from 0.375 to 0.40. Two admixtures, Sika ViscoCrete Water Reducer and Lithium Stabilized Colloidal Silica (LiSol), were used in every mix. Water reducing admixture is added to increase the strength and workability of the mix. LiSol is added to the mixture to minimize the risk of alkali–silica reactions within the concrete in the future (Feng et al., 2005) [19]. Osfouri et al. (2025) provides a comprehensive review of ASR in concrete with foam glass aggregates [17].

Aero Aggregates manufactures aerolite granulated foamed glass in seven different sizes. The aggregates meet the standards for ASTM C330 [20] and C331 [21] for structural concrete and concrete masonry units. The FGA sizes titled G1, G2, and G3 are classified as fine aggregates, while G5, G6, and G10 are coarse aggregates. G4 is an intermediate aggregate size that contains both fine and coarse particles. To determine accurate gradations for each size, a sieve analysis was conducted following ASTM C136 [22]. Measurements were taken to calculate the weight of aggregates retained on each size of the sieve. Each aggregate size was put through the stack of 25 mm, 19 mm, 12.5 mm, 9.5 mm, #8, #16, #30, #50, and #100 sieves to determine the weight retained on each. From these weights, percentages retained and passing are calculated to determine the full gradations of each FGA size, and the percentage of aggregate passing through each sieve size is shown in Figure 2.

A spreadsheet was created using the different aggregate gradations to optimize the combination of sizes and determine the best possible mix. The mix designs are portioned by volume percentage, then the volume is converted to weight using the specific gravity of each component of the mix. Three different graphs were created to optimize the design: the Tarantula Curve (Wang et al., 2018), the 0.45 power chart, and the coarseness factor chart (MT 122-17) [2]. Each mix design is created to balance the three graphs to give the most ideal proportions. The Tarantula Curve uses the determined aggregate gradations to create optimized proportions that minimize voids to increase strength. The predetermined aggregate gradations are used to find the percentage of the total retained on each sieve size. The individual percentage retained is graphed against the respective sieve size. Based on prior research in the field (Wang et al., 2018) [2], the Tarantula Curve gives a set of ideal proportions which will create a balanced, workable mix, as shown in Figure 3.

The optimal mix design will lie between the curves of the maximum and minimum percentages. The 0.45 power curve assesses overall gradation by looking at the percentage passing. There are four curves plotted on this graph to find the optimal aggregate proportions. The sieve opening sizes raised to 0.45 power is the independent variable plotted. The baseline curve, the power chart line, determines the percentage passing with the ratio of the sieve size being analyzed over the nominal maximum sieve size, all raised to 0.45 power. The values both 7 percent higher and lower than this baseline curve are also calculated and plotted to give an optimal range of gradations for a mixture design.

Once the optimal design is determined, each concrete batch is mixed in a 0.06 cubic meter capacity drum mixer and cast into eight 0.2 m tall, 0.1 m diameter cylinder molds. It should be noted that the lightweight aggregates were not pre-soaked or otherwise treated before being added to the mixer. After 7 days and 28 days have passed, respectively, cylinders are removed from the molds and weighed to determine the unit weight of that mixture. Compression tests were conducted to find the compressive strength of each cylinder and the average strength of each mix. The strength gain is evaluated over time by testing at both 7 and 28 days. A 26,700 kN capacity Forney testing machine is used to apply load to each concrete cylinder. Based on ASTM C39, the load is applied at approximately 0.24 MPa per second [23].

3. Results

3.1. Concrete Mixture Designs

Each concrete mixture is designed to optimize each of the three graphs to create the strongest and most workable mix. A range of mixes with varying amounts of both normal weight aggregates and lightweight FGAs are designed. All mixes have a volume of around 0.023 cubic meters. The Tarantula Curves used for mix design optimization are shown in Figure 4. The cement content was determined based on standard mixture design procedures such as ACI C211.1-22 [24], as well as previous mixtures conducted by Aero Aggregates and as part of this research program. The water content was adjusted for each mixture during mixing to achieve the desired workability of the paste. Every effort was made to keep the water content as low as possible while achieving the necessary workability. The nine concrete mix designs include a normal weight control mix, five combination mixes, and three all-lightweight mixes, all shown in

Table 1. Mix design proportions.

	MIX 1		MIX 2		MIX 3		MIX 4		MIX 5		MIX 6		MIX 7		MIX 8		MIX 9
	Vol %	W (kg)	Vol %	W (kg)	Vol %	W (kg)	Vol %	W (kg)	Vol %	W (kg)	Vol %	W (kg)	Vol %	W (kg)	Vol %	W (kg)	Vol %	W (kg)
Cement	19.07	13.44	17.59	12.54	19.33	13.94	14.84	10.58	17.38	12.43	17.38	12.43	16.08	11.76	16.37	12.09	16.80	12.09
Water	18.96	4.24	18.87	4.27	22.27	5.10	17.09	3.87	20.01	4.54	21.89	4.97	18.51	4.30	18.85	4.42	21.17	4.84
FA	0.00	0.00	0.00	0.00	0.00	0.00	26.48	15.59	18.09	10.65	14.55	8.56	0.00	0.00	30.00	17.66	24.67	14.52
CA	0.00	0.00	0.00	0.00	0.00	0.00	39.72	24.28	22.91	14.01	17.58	10.74	35.00	21.39	0.00	0.00	11.72	7.16
G1	3.60	1.04	2.45	0.71	7.77	2.23	0.00	0.00	0.60	0.17	0.61	0.17	7.00	2.01	0.00	0.00	1.23	0.35
G2	18.02	4.36	14.11	3.42	9.43	2.28	0.00	0.00	2.41	0.58	4.55	1.10	18.00	4.36	0.00	0.00	1.85	0.45
G3	10.21	1.87	5.52	1.01	5.55	1.02	0.00	0.00	0.90	0.17	1.82	0.33	5.00	0.92	0.00	0.00	0.93	0.17
G4	12.61	1.63	7.97	1.03	4.44	0.57	0.00	0.00	3.02	0.39	4.24	0.55	0.00	0.00	0.00	0.00	1.23	0.16
G5	10.81	1.17	2.45	0.27	3.88	0.42	0.00	0.00	2.41	0.26	3.03	0.33	0.00	0.00	0.04	0.43	3.08	0.34
G6	3.60	0.33	12.88	1.20	10.54	0.98	0.00	0.00	4.22	0.39	6.06	0.56	0.00	0.00	14.00	1.30	7.40	0.69
G10	0.00	0.00	15.95	1.48	14.98	1.39	0.00	0.00	6.03	0.56	8.49	0.79	0.00	0.00	17.00	1.58	10.48	0.97

.

3.1.1. Mix 1

Mix 1 is an all-lightweight concrete mix using six different sizes of foamed glass aggregates. A water–cement ratio of 0.43 is used, with an additional 80 milliliters (mL) of water reducing admixture. The maximum aggregate size is 12.5 mm, and the proportions of the mix are determined using prior research by Aero Aggregates. The very light weight of the FGA, combined with the porous structure of the aggregate absorbing some of the mixture water, led to difficulty creating a cohesive mix. When tested after curing for 7 and 28 days, the average unit weight is around 11.0 kN/m³. The cylinder contained many voids both on the surface and within which contributes to both the low unit weight and compressive strength. The strength saw no increase from 7- to 28-day testing, with an average compressive strength of 6.76 MPa.

3.1.2. Mix 2

The goal of Mix 2 is to improve the strength and workability of the first mix, which is attempted by utilizing the previously mentioned optimization methods. The maximum aggregate size is increased to include 25 mm diameter grains to strengthen the mix, and the cement content was reduced slightly to be more consistent with typical values for a 25 mm maximum aggregate size mixture. Similarly, to Mix 1, Mix 2 weighs around 11.0 kN/m³, much lower than the typical lightweight concrete which is around 18.1 kN/m³. Three cylinders are tested at 7 days, with an average compressive strength of 6.21 MPa. Three more cylinders are tested at 28 days to determine if there is any increase in strength. A slight increase in compressive strength can be noted, with an average of 6.90 MPa. The FGA particles fractured under this load, as seen by the failure surface shown in Figure 5. The porosity of the FGA and the low workability of the paste create many voids within the mix and weakens the cylinder, which also contributes to the low compressive strength.

3.1.3. Mix 3

Mix 3 uses the same aggregate proportions as the previous mix with the addition of 40% more cement. This placed the cement content well above a typical range, but increased uniformity and cohesion when mixing. This increased the unit weight of the cylinders to an average of about 11.8 kN/m³. This unit weight is closer to a typical lightweight concrete by ACI standards, but the addition of cement makes a less sustainable and more expensive mix. The load sustained by the cylinders at 7 days was significantly higher, with an average compressive strength of over 13.8 MPa. The addition of the extra cement gives about a 6.9 MPa increase in strength. The results varied for each cylinder and the strength did not increase over time, indicating that the strength is governed by the FGA strength.

3.1.4. Mix 4

The fourth mix is a normal weight mix used as a control group for data comparison. A 2-to-3 ratio of fine to coarse aggregate is used, with a water–cement ratio of 0.40 and cement content as recommended by the mixture design process described in ACI 211.1-22 [16]. As expected, this mix had the highest unit weight, with an average of 24.3 kN/m³. This is approximately double the unit weight of all lightweight mixes. The 7-day compressive strength was nearly 41.4 MPa at failure, with the three cylinders showing consistency. Three more cylinders were loaded at 28 days, and the average compressive strength increased to around 45.5 MPa. The normal weight aggregates are seen to increase in strength over time more than the FGAs, showing that improvement in the strength of the cement paste results in increased compressive strength for the concrete.

3.1.5. Mix 5

Mix 5 is a combination of both normal weight aggregates and FGAs. Lightweight aggregates replace 40% of the coarse aggregates and 20% of the fine aggregates. The addition of the FGA gives a wider range of aggregate grain sizes to further optimize the design, as seen in the Tarantula Curve of Figure 6. The goal is to create a lightweight design that can maintain strength due to the addition of normal weight aggregates. The combination mix composed of FGAs and normal weight aggregates had an average unit weight of 20.4 kN/m³. This lies in the range of a semi-lightweight concrete, somewhere between lightweight and normal weight. The three cylinders tested at 7 days had an average compressive strength of around 25.5 MPa. A strength increase of 5.5 MPa was noted when testing at 28 days, with an average of 31.0 MPa. While this concrete mix design reached an acceptable strength for use in structural applications, the weight is too high to be considered lightweight concrete.

3.1.6. Mix 6

Mix 6 is designed based off the results of Mix 5. This design was altered to decrease the unit weight by adding more FGA. G5, G6, and a percentage of G4 are used to replace 55% volume of the normal weight coarse aggregate. G1, G2, G3, and the remaining percentage of G4 are used to replace 35% of the normal weight fine aggregate. The previous optimization methods are used again, and the aggregate proportions are adjusted for both the Tarantula Curve and 0.45 power chart. The average unit weight of the cylinders was 18.1 kN/m³, a typical value for lightweight concrete. The addition of a higher percentage of FGA decreased the unit weight to a more acceptable value for lightweight concrete but caused a corresponding strength decrease. The 7-day compressive strength averaged 19.3 MPa, which increased to 22.1 MPa at 28 days. The cylinders appear to have less voids on the surface than an all-lightweight mix, but some voids are still visible in the split cylinder, as seen in Figure 7.

3.1.7. Mix 7

Mix 7 uses normal weight coarse aggregates while the fine aggregates are replaced with G1, G2, and G3 FGAs. The average unit weight of this concrete is 18.5 kN/m³, typical for a lightweight concrete. When mixing, it appeared that the glass fine aggregates did not combine well with the coarse aggregates, likely due to the weight difference. The mix appeared segregated and dry, even after the addition of 1000 g of extra water. The compressive strength of the cylinders tested at 7 days averages 8.96 MPa. The strength only slightly increased from 7 days to 28 days, and the cylinders appear brittle with many voids present. Overall, the fine FGAs do not create a cohesive mix with the coarse aggregates, with the results showing a dry mix lacking the paste needed for strength.

3.1.8. Mix 8

Similarly Mix 8 will replace an entire component with FGAs, the coarse aggregates in this instance. Normal weight fine aggregates are used with G5, G6, and G10 coarse FGAs. The water content in this mix is increased to account for the porosity of the coarse FGAs. As expected, the unit weight of this mix is lower than the previous mix which replaced all the fine aggregate, with a value of around 15.4 kN/m³. However, the coarse FGAs mixed well with the rest of the concrete components, and the mix appeared mostly cohesive and uniform. The 7-day compressive strength testing yielded an average value of 12.4 MPa and did not increase at 28 days. This may indicate there was not enough time for hydration to increase strength before the porous aggregates absorbed water from the cement paste. Looking at the results for Mix 7 and Mix 8, Mix 8 has a lower unit weight due to the use of the larger-size glass aggregates. Although the larger FGA grains have a lower strength than the coarse aggregates used in Mix 7, the Mix 8 cylinders were able to hold more load. This is likely due to the improvement of the cement paste and resultant decrease in the void space in Mix 8. Although both mixes appeared to have too much water absorption that limited the hydration reactions, this can be improved with a higher amount of normal weight aggregates, as tested in Mix 9.

3.1.9. Mix 9

Mix 9 is designed using the results from both Mix 7 and 8. The goal is to increase the use of coarse FGAs to create a lighter, more cohesive mix, but still include normal weight coarse aggregates to achieve the desired strength. The aggregates in the mix are 43% FGA, with lightweight aggregates replacing 65% of the coarse aggregates and 15% of the fine aggregates. After curing, the cylinders did not contain voids visible on the surface, as seen in Figure 8, contrary to some previous mixes. The unit weight falls into the ideal range for structural lightweight concrete, with an average of 18.2 kN/m³. Eight cylinders were tested for compressive strength, and the average 7-day strength was found to be 25.5 MPa. The FGA still acts as the strength ceiling in this mix, indicated by the lack of strength increase over time. The average 28-day compressive strength increased to about 26.9 MPa, around 1.4 MPa higher than the first test.

4. Discussion

4.1. Results Comparison

Overall, there are some patterns that reveal themselves as the cylinders are tested. The goal for this lightweight concrete was a unit weight of around 18.1 kN/m³. Using only glass aggregate, the unit weight of the concrete was about half of that desired value in Mixes 1, 2, and 3. On the other end, the control mix using normal weight aggregates creates concrete with a unit weight of around 24.3 kN/m³. A larger percentage of FGA correlates with a lower unit weight concrete due to the lighter aggregates. The goal unit weight is achieved in Mixes 6, 7, and 9 when a percentage of the normal weight aggregates are replaced with FGAs.

An increase in FGA percentage is directly correlated with a decrease in compressive strength in most cases. As previously shown, when FGAs are added into a mix, the cylinders tend to fail due to the aggregates fracturing. The FGA appears to absorb more water in the mix than a normal aggregate would, likely due to the porosity of the grains. This does not allow adequate hydration reactions, causing many of the lightweight concrete mixes to be dry and brittle. Figure 9 shows the correlation between percentage of FGA, unit weight, and compressive strength in each mix.

Disregarding Mixes 7 and 8, where an entire component of the mix design was replaced with FGA, a trend can be seen between the compressive strength, unit weight, and percentage of FGA in the mix. The control mix with only normal weight aggregates has both the highest compressive strength and unit weight, while the all-lightweight mixes have the lowest of both properties. Mixes 7 and 8 replace the fine aggregates and coarse aggregates, respectively, with FGAs, leading to more segregated, weaker mixes that do not follow the trend. Each mix containing FGAs is limited by the strength of the aggregates, and all fail due to shearing through the cylinder which splits the aggregates. Mix 9 is best able to balance the ideal unit weight of 18.2 kN/m³ with a compressive strength of almost 27.6 MPa. The relationship between the strength and unit weight of each mix can be seen in Figure 10. The control mix, Mix 4, has the highest strength-to-unit weight ratio, but Mixes 5 and 9 show the best relationships of the lightweight mixes.

4.2. Structural Application

A typical structural application of lightweight concrete is as part of a composite system, with a concrete slab on top of a metal deck supported by steel beams, girders, and columns. The slab cover thickness is mainly governed by the required fire rating of the building. According to the International Building Code (IBC), to achieve a 2 h fire rating a normal weight concrete slab requires a minimum thickness of 4.6 inches while a lightweight slab requires 3.6 inches without any additional fireproofing [25]. Lightweight concrete is known to be more fire resistant because the aggregates have better thermal stability due to their processing and the decreased thermal conductivity of the concrete, which allows the slab to be thinner and use less material to achieve the fire rating. The differing slab thicknesses have a quantifiable impact on the dead load of the building, therefore changing the required size of all supporting members.

An example structural design of a generic building was analyzed to determine the material impact due to the use of lightweight or normal weight concrete. A 5-story building containing a slab on a Vulcraft 75 mm metal deck on each level was analyzed, using RAM Structural Systems to design the beams and columns for each model. The modeled building is 76.8 m by 28.7 m with seven bays, as shown in Figure 11. Steel I-beams of varying sizes will be used for all beams and columns.

4.2.1. RAM Model

The previously mentioned building is used to create two different models in RAM Structural Systems governed by fire resistant design. The first version models a 75 mm metal deck with an 82.5 mm lightweight slab cover on top. The lightweight concrete has a unit weight of 18.1 kN/m³ and a compressive strength of 20.6 MPa, similar to the previously mentioned Mix #9. The second model contains the same metal deck but with a 114 mm normal weight slab cover on top of the metal deck. This is 27.6 MPa concrete with a unit weight of 22.8 kN/m³. A thinner concrete slab is used for the lightweight concrete based on code requirements resulting from the increased fire resistance of concrete made with lightweight aggregates. Both models have an applied superimposed dead load of 0.72 kN/m² on all floors and 0.96 kN/m² on the roof in addition to the self-weight of the building. A reducible live load of 4.79 kN/m² is applied across the surface of each floor.

The models are then input into the RAM Steel Beam Designer to determine the required member sizes, which are designed according to AISC 360-16 [26]. Because of the heavier dead load due to the normal weight concrete slab, larger steel members are needed to support the load. The columns required to support the load of each floor are also designed using RAM Steel Column. Highlighting a specific bay of the building model, the differences in loads and deflections can also be analyzed. A 10.98 m bay on the first floor of the building model was looked at, as shown in Figure 12.

In Figure 12, the column loads indicated include both the self-weight of the structure (steel and concrete) as well as the superimposed dead and live loads. The deflections represent the maximum deflection occurring along the length of the beam. The beams and girders supporting the lightweight slab experience different loads than those supporting the normal weight slab due to the dead load. The lightweight slab provides at least a 44.4 kN lower load on each gravity member, therefore requiring lower capacity members. Based on the generated design, the most common steel member used in the lightweight concrete building model is W24x76, while the most common in the second model is W27x84. More deflection occurs in the girders spanning across the beams supporting the lightweight slab, likely due to the lighter beams used. Deflection in the beams is similar between the two models even though different sizes of exterior beams are used.

The columns on the first floor support the cumulative load of the upper floors of the building, so the required capacity is lower for the lightweight concrete building due to the smaller dead load. Smaller and lighter weight columns can be used on every floor as shown by the first model, further decreasing the required steel for design. For the lightweight design, it was determined that various W10 and W12 members will be needed, while the heavier model also required several W14 columns.

4.2.2. Material Use

Overall, the buildings require 502,000 kg and 560,000 kg of steel, respectively, with 58,000 kg more needed for the version with the thicker, normal weight slabs.

Table 2. Structural steel tonnage.

	LWC	NWC
Story	Beam Weight [kg]
1	72,845	88,880
2	95,555	104,590
3	95,555	104,590
4	95,555	104,590
5	95,555	104,590
Columns	45,990	53,095
Total	501,045	560,320

shows the steel takeoff and the difference in weight for each floor of the two building models. With the use of lightweight concrete in building design, this example shows an 11% decrease in total steel required.

FGAs and other lightweight aggregates are more fire resistant than typical aggregates; therefore, a thinner lightweight slab can be used to achieve an adequate fire rating. The difference in the 83 mm and 114 mm concrete slabs leads to a 349 cubic meter difference in the amount of concrete for the entire building. The lighter overall building weight will also reduce the size of the foundations, and the materials needed.

4.3. Cost Analysis

Utilizing less concrete in a building design can help lower project costs, but the price difference in the lightweight aggregates must be considered. Per Aero Aggregates, their foam glass aggregates cost USD 0.88 per kg, higher than usual aggregate costs. Coarse and fine aggregates cost around USD 0.02–0.04 per kg based on prior material purchases. A large percentage of the costs come from the cement at USD 0.40 per kg, so the higher amount of cement in the lightweight concrete mixes also impacts the cost. Mix 4, the control mix with normal weight aggregates, is the least expensive at USD 191 per cubic meter, due to the absence of FGAs. The all-lightweight mixes have a price increase of over USD 400 per cubic meter of concrete, with a total cost of USD 605 per cubic meter. The unit cost for Mix 9, which includes both lightweight and standard weight aggregates, is shown in

Table 3. Cost (in USD) of lightweight concrete Mix 9.

	Weight (kg per m3 of Concrete)	Cost (USD/kg)	Cost (USD per m3 of Concrete)
Water	211.7	$0.00	$0
Cement	529.3	$0.40	$210
CA	635.6	$0.03	$19
FA	313.5	$0.04	$13
G1	15.5	$0.88	$14
G2	19.6	$0.88	$17
G3	7.4	$0.88	$7
G4	7.0	$0.88	$6
G5	14.7	$0.88	$13
G6	30.1	$0.88	$27
G10	42.6	$0.88	$38
Total			$362

.

Scaling these costs up to an entire project, the effect of the lightweight concrete can be seen. The cost of Mix 9 at USD 362 per cubic meter will be used because it provides an adequate unit weight and strength for structural lightweight concrete. As previously mentioned, the lightweight concrete allows for a thinner slab, decreasing the total amount of material used in the building. Even though there is a lower amount of concrete used, the cost is higher due to the more expensive lightweight concrete. On the other hand, the cost of steel used, estimated at USD 4.4 per kg, is lower because of the lighter and smaller members required (Steel Dynamics, 2024) [27]. The total project cost is lower for the lightweight building model by around USD 170,000, as shown in

Table 4. Total cost of building materials.

	Steel			Concrete			Total (USD)
	Cost (USD/kg)	Mass (kg)	Total (USD)	Cost (USD/m3)	Volume (m3)	Total (USD)
LWC	$4.40	501,045	$2,205,000	$362	908	$329,000	$2,534,000
NWC	$4.40	560,320	$2,465,000	$191	1258	$240,000	$2,705,000

. Overall, the lightweight concrete mixes with FGAs are more expensive to obtain, but the effect on the dead load of the building can lower total project costs, doing so in this example by over 6%.

The effect of using lightweight FGA concrete on the total project cost would be affected by the building characteristics and materials selected for the structural systems. Only one building configuration was examined in this study, and while it is representative of many common buildings the cost savings realized here would likely be different in other structures with different framing systems, loads, or materials. However, the results from this analysis show that using lightweight FGA concrete can be cost competitive with more traditional concrete in common steel framed construction.

4.4. Sustainability Impact

Another aspect that varies from normal weight to lightweight concrete is the impact on sustainability and carbon emissions. Instead of depleting natural resources by using mined aggregates, Aero Aggregates’ FGAs are created from 99% recycled container glass. Environmental product declarations (EPDs) are used as a standardized method for sharing quantifiable environmental information to compare products [28,29,30]. Aero Aggregates is in the process of completing the life cycle analysis needed to determine the environmental impact of FGAs, but companies with similar products have this information available. Foamit and Glaspor, developed by The Foamit Group, are also foamed glass aggregates produced from waste glass in a range of sizes [31]. Since these products are comparable to Aerolite, the carbon emission values presented in the EPDs will be used for this analysis.

Global warming potential (GWP) is measured in units of kilograms of carbon dioxide (CO₂) emissions per ton of product. This is quantified for all stages of a product’s life cycle, focusing on the early stages of material supply, transport, and manufacturing. Based on GWP values pulled from the product EPDs, Foamit and Glasopor (NEPD00280E) contain approximately 167 and 177 kg of CO₂ per ton, respectively [31,32]. These embodied carbon values can be altered based on the bulk density of each Aerolite FGA grain size to reflect the GWP. Average values from the literature and other EPDs are used to determine the embodied carbon of normal weight gravel and sand aggregates (EPD-154) [20]. Portland cement is known to be a large producer of greenhouse gases, with about 850 kg of embodied carbon per ton (EPD-20084) [29]. The carbon calculations are scaled up to determine the amount in each of the nine mix designs. Cement has such a high level of embodied carbon that the mixes with a high amount of cement content, including Mixes 1, 2, and 3, have a larger total GWP. Overall, this has the most impact on the CO₂ levels in each mix, as shown in

Table 5. GWP in each concrete mix.

MIX	FGA	Cement	Fine Agg	Coarse Agg	Total
	[kg CO2/Ton]
1	64.0	162.4	0.0	0.0	226.4
2	72.0	149.6	0.0	0.0	221.6
3	62.6	164.1	0.0	0.0	226.7
4	0.0	125.8	1.2	1.5	128.5
5	24.1	147.9	0.8	0.8	173.7
6	35.3	147.9	0.7	0.7	184.5
7	24.4	136.9	0.0	1.3	162.5
8	46.4	139.4	1.4	0.0	187.2
9	32.6	142.8	1.1	0.4	177.0

. The normal weight concrete mix, Mix 4, has the lowest amount of carbon, due to the increased GWP of FGA when compared to standard aggregate. The mixes with a combination of FGAs and normal weight aggregates have embodied carbon values between the control mix and all FGA mixes.

While the addition of FGAs contributes to an increase in embodied carbon in concrete mixes overall, there are other environmental benefits from its use. FGAs are composed of recycled container glass, so no new resources are being depleted. The reuse of previously discarded glass aims to minimize the amount of waste in landfills. The use of lightweight concrete minimizes the total material required, as previously mentioned, also saving resources. When looked at in terms of an entire building, the total embodied carbon is being lowered when the reduction in weight of steel members due to the use of lightweight concrete is included. Hot rolled steel I-beams have a GWP of around 1000 kg CO₂ per ton, based on various manufacturer EPDs (EPD, Nucor) [30].

Table 6. Total building GWP.

Material	LWC	NWC
	GWP [kg CO2]
Concrete	2840	2850
Steel	552,000	618,000
Total	554,800	620,900

shows the total embodied carbon in the previously mentioned building example, totaling the amount in the steel and concrete aspects of the structure. Mix 9 is used for the lightweight concrete as it has an adequate unit weight and compressive strength.

Using concrete with recycled materials also gives a project a higher leadership in energy and environmental design (LEED) rating. FGA concrete could add up to 13 points in the materials and resources category of the LEED scorecard, potentially taking a project from a silver to gold or gold to platinum rating. Overall, the sustainability benefits of FGA and lightweight concrete are evident due to the glass waste reused instead of discarded in landfills and the total materials saved, reducing carbon emissions.

5. Conclusions

The goal of this research was to determine the feasibility and benefits of using FGA to create lightweight concrete. Nine different concrete mixes were designed by optimizing the aggregate gradations to create the strongest lightweight mixture. The mix designs ranged from a control mix of normal weight aggregates to three all-lightweight mixes, with combinations using only a percentage replacement of FGAs also tested. After 7 and 28 days, respectively, the cylinders were tested to determine unit weight and compressive strength. The use of lightweight concrete was also analyzed on a larger scale for a 5-story building composed of concrete slabs on top of metal decks supported by steel members. RAM Structural System was used to determine the difference in loads, deflections, and material use between a building with a lightweight concrete slab compared to a normal weight concrete slab. The cost and environmental impact of each mix design was also calculated to determine the overall benefit of each mix. A summary of all the experimental findings is provided in

Table 7. Unit weight, compressive strength, cost, and GWP of each concrete mix design.

MIX	Unit Weight	Strength	Cost	GWP
	[kN/m3]	[MPa]	[$/m3]	[kg CO2]
1	11	6.8	$635	13.8
2	10	6.9	$605	12.8
3	12	12.4	$604	14.0
4	24	45.8	$191	10.1
5	20	31.0	$439	12.1
6	18	21.9	$401	12.2
7	19	11.0	$491	11.7
8	15	10.6	$345	11.9
9	18	26.9	$361	11.8

.

By comparing the unit weight, strength, cost, and global warming potential, the benefits of each mix design are determined. The testing of each of the concrete mixes revealed the impact and behavior of FGAs within the mix. Since this study was focused on evaluating the feasibility of FGA concrete, the scope of the research is limited. The mixtures evaluated in this study all used the same aggregates and material components, and so additional research could be conducted looking at other materials or aggregate sources to improve the performance of the concrete. Given the compressive strength of the FGA concrete in this study, it would only be suitable for flexural applications. Additional work would need to be completed to evaluate the durability of the concrete before it is used in conditions where it would be exposed to moisture or other corrosive environments. Despite these limitations, FGA shows potential as a sustainable, cost-effective material for use composite floor systems.

Based on the research and experimental testing conducted, the following conclusions can be drawn:

• It was determined that a 65% replacement of coarse aggregate and a 15% replacement of fine aggregate gives the optimal results, as performed in Mix 9. This mixture gives the highest strength with the standard unit weight for lightweight concrete and has a moderate cost and environmental impact. Mix 9 cylinders appeared to have the least amount of voids and were cohesive during mixing. Therefore, Mix 9 is the only lightweight mix design recommended for structural use and future research testing.
• When analyzed on a full building scale, Mix 9 uses less overall material than normal weight concrete due to the reduced dead load and smaller steel members. As shown, this reduces the total cost and sustainability impact of the entire structure, even though the concrete mix is more costly on a unit volume basis. This demonstrates the benefits of lightweight concrete and how it can improve structural design.
• Mix design optimization methods, including Tarantula Curve and 0.45 power chart, can give ideal aggregate proportions that maximize the strength of the concrete. This is mainly seen in the control mix, Mix 4, which had high compressive strength even after curing for only 7 days.
• The all-lightweight mixes, Mixes 1, 2, and 3, have the lowest unit weight and the lowest strength. The weaker aggregates significantly decrease the strength ceiling of the concrete. FGAs are more expensive than typical aggregates, so these mixes also cost more. They also require more cement to bind the aggregates, which increases the embodied carbon in the mixes. Overall, the all-lightweight mixes are not practical and would be difficult to use in structural applications.
• A combination of stone aggregates and FGAs is the only way to achieve the desired unit weight of 18.1 kN/m³ while maintaining a viable strength for structural applications. Mixes 7 and 8 used FGAs for all the fine and coarse aggregates, respectively, but the mixes segregated in the mixer and contained many voids which decreased the strength. A combination of coarse and fine FGAs is better for mix uniformity.