Life Cycle Assessment of Waste Glass Powder Incorporation on Concrete: A Bridge Retrofit Study Case

Abstract

The construction sector is responsible for some of the highest energy and natural resources consumption. In this context, new materials and solutions are created aimed at developing sustainable alternatives. While the literature presents papers that evaluate the mechanical and durability properties of concrete with glass waste powder and account for its environmental impact, no papers have executed the evaluation considering the retrofit of bridges. Furthermore, no papers evaluating the materials, construction, and maintenance could be found. Hence, this study proposes a technical and sustainable solution for the retrofit of the Third Bridge of Vitoria, an important intercity urban connector. This study evaluates both the technical and the environmental performance of structural concrete elements, considering the partial substitution of cement with glass waste powder and a baseline scenario with conventional concrete. The environmental impacts were evaluated through the life cycle assessment tool. The results indicate that incorporating waste glass powder in the prestressed hollow-core slabs as a partial cement replacement can improve the durability-related properties and mitigate environmental impact. It also shows that the manufacturing phase is the most impactful and that glass powder can significantly reduce the impact of maintenance.

1. Introduction

The construction industry is one of the sectors with the highest consumption of natural resources and energy [1]. According to the CO₂ emissions growth curve, maintaining current emission standards can lead to air quality pollution levels, creating an uninhabitable environment [2]. Furthermore, cement production represents 7% of all fossil fuel emissions [3], and while cement-based products are considered the second-highest material consumed on the planet [4], an increase in their use was observed in recent years [5]. Hence, reassessing the current production process and developing new sustainable alternatives is vital for the construction industry.

To contribute to the reduction in cement consumption to produce concrete and mitigate environmental impact, research has demonstrated promising results involving the incorporation of supplementary cementitious materials (SCMs) in concrete structures. These studies indicate that some SCMs can improve the performance of concrete regarding their mechanical aspects and durability. SCMs present the potential to compensate for the partial removal of cement or even enhance the properties of concrete by filling the pores of the material. If the proper dimensions and chemical properties are used, these materials can become pozzolanic materials, by combining themselves to calcium hydroxide and with the different types of cement [6,7,8,9].

According to Samad and Shah [10], the incorporation of SCM into concrete as a partial substitute for cement can not only reduce the use of cement in concrete structures, but also obtain varying effects such as strength gain, abrasion resistance, freeze-thaw and deicer-scaling resistance, drying shrinkage and creep, permeability, alkali-silica reactivity, chemical resistance and carbonation. Regarding some examples of SCMs that have the potential to enhance the mechanical properties and durability of concretes, the literature points to fly ash, silica fume, slag, and metakaolin. Regarding the use of slag, gains in properties related to durability and a slow gain in compressive strength were observed; however, the long-term compressive strength is higher compared to concrete without slag. The beneficial effects of slag concrete depend on the dosage ratios and curing conditions. A maximum replacement level of 50% is recommended for concretes with slag and a curing temperature of at least 20 °C.

There are also gains in mechanical properties and durability with the use of fly ash; however, in early age strength of concretes with fly ash it is slower than the concretes without fly ash because the pozzolanic reactions are slower than the hydration reactions and they start after about five days [10]. Metakaolin can obtain relevant gains to resist compression and durability such as resistance to chloride ions and alkali silica reactions by partially substituting the cement up to 20% [6,7,8].

In addition, there are studies that indicate that incorporating glass waste powder into concrete as cement substitutes can also be an important SCM and improve the mechanical properties and durability of the hardened concrete. After 56 days, there were gains in resistance to compression, considering a 20% cement substitution with 20 μm glass waste powder [11,12]. Tariq et al. [13] identified greater resistance to chloride penetration in concrete with the incorporation of waste glass than in concrete with fly ash and justified that this is due to the continuous evolution and refinement of the pore structure, which results in better long-term transport performance of the binder system and considerable extension in terms of service-life for concrete exposed to marine environment. Jurczak and Szmatuła [14] compared the use of fly ash incorporation with waste glass in concrete and identified that there were no significant differences in terms of compressive strength for the concrete class adopted. Ahmad et al. [15] identified a gain in compressive strength of 27% in concrete with the replacement of 20% cement with glass waste. The gains from the 20% replacement were greater than the 10% and 30% replacement. Lee et al. [16] showed an increase in chloride resistance with the partial replacement of 20% cement by glass waste in concretes with 56 days and an increase in compressive strength after 90 days with the same replacement content. It was also concluded that the pozzolanic re-action significantly contributed to the improvement of the degree of compaction of concrete. A study by Cassar and Camilleri [17] also showed that concrete with a glass waste substitution of 10% and 20% cement presented a higher resistance to penetration of chloride ions than concrete that had no waste residue incorporation. Matos and Sousa-Coutinho [18] identified resistance to chloride penetration, sulphate resistance, and a reaction to alkali silica without compromising the compression resistance with the incorporation of glass waste in concrete. Regarding the most suitable percentages for the partial replacement of cement by waste glass, research suggests benefits in the percentages between 10% and 20%; this will depend on the characteristics of the physical and chemical properties of the material [11,12,17,18].

In addition to the technological benefits presented with the use of waste glass as a partial substitute for cement, research indicates environmental benefits. According to Frederico and Chidiac [19], its incorporation presents the potential for (a) diversion of non-recycled waste from landfills to useful applications; (b) reduction of the negative effects of cement production, relevant to the consumption of non-renewable natural resources; (c) reduction in the use of energy for cement production and corresponding emissions of greenhouse gases.

However, to assess whether a particular material has the potential to reduce environmental impacts, analyses were carried out using the life cycle assessment methodology (LCA) (i.e., ISO 14040 [20] and ISO 14044 [21]), which stands out as an essential tool to quantify, evaluate, compare, and improve solutions for the environment. LCA studies comparing the incorporation of materials into concrete are present in the literature [22,23,24].

Regarding LCA adoption of waste instead of cement, Panesar et al. [22] and Celik et al. [23] incorporated fly ash into concrete and quantified the environmental benefits of this use, proving the environmental viability. Robson et al. [24] surveyed 127 traces of cementitious materials that used ornamental rock residues and carried out an LCA to verify the feasibility of using this residue without and with thermal treatment. Regarding the incorporation of waste glass into concrete, Mohan et al. [25] evaluated through an LCA the use of glass powder as a partial substitute for cement in concrete and mortar. Tucker et al. [26] performed an LCA comparing recycling and the use of glass as a pozzolan. The results showed similarities between the environmental and economic gains between both options. Deschamps et al. [27] evaluated, through an LCA, the use of glass waste in concrete, and the results pointed out the environmental benefits in the environmental indicators used such as incorporated energy and global warming potential.

Mohan et al. [25] investigated, via adopting the LCA, the emissions of greenhouse gases and embodied energy, from the cradle to the gate, in the manufacture of 1 m³ of concrete with glass powder compared to 1 m³ of conventional concrete, each having a resistance to 30 MPa compression in 27 days.

Studies that aim to assess the durability of special works, such as bridges, were also identified. Vieira et al. [28] developed a model proposal to estimate the service life (SL) using real data on total chloride concentrations in pillars of the Terceira Ponte de Vitória, Espírito Santo, Brazil. Regarding studies involving LCAs of bridges, Zhang et al. [29] compared the environmental impact of two alternatives for replacing a bridge deck. Initial demolition, construction, and future maintenance were evaluated. Martin et al. [30] performed an LCA for a Spanish steel arch and concrete deck rail bridge. All stages were considered: the material stage (raw material extraction, production, and distribution); construction, use, and maintenance (repair and replacement); and end of life. The results showed that the main contributor to environmental impact was the manufacturing phase of the material, accounting for 64% of the total results. Dequidt [31] evaluated the impact of climate change on the life cycle of a Norwegian road bridge. Gervásio and Silva [32] carried out a comparative analysis of the life cycle of bridges composed of steel and concrete. Du and Karoumi [33] performed an LCA model on bridges as a guideline to quantify the environmental impact for railway bridge structures. The initial stage of manufacturing the material is responsible for the greatest environmental demand.

However, there are gaps in the research on the adoption of LCAs for bridges, also associated with aspects of durability, service life, and incorporation of residues. Regarding the incorporation of glass powder into concrete and the use of LCAs, few papers are available [25,26,27]. However, research that integrated an LCA of the incorporation of glass residue in concrete in bridge adequacy proposals was not found.

There is a relevant social demand in the municipality of Vitória, Espírito Santo, Brazil, related to the frequent suicide attempts on the Third Bridge of Vitória (Third Bridge VIXBR). In 2000, 4.8 deaths were registered per 100,000 inhabitants and, in 2016, it grew to 6.5 [34], an increase of 35%. In this context, this study sought to reduce these statistics, presenting a solution for the reduction in these occurrences on the Third Bridge VIXBR, which does not currently present an effective lateral barrier system against these attempts. Moreover, the destination of glass waste is also an issue in this particular municipality due to the industries located in Vitoria. The considered glass waste is the sludge obtained by the treatment, which cannot be reused in the manufacturing process and is typically discarded.

In this context, this research proposes a solution to the retrofit of the Third Bridge located in VIXBR. The solution has low technical and economic complexity and has the potential to mitigate environmental impact. The project has lateral closing and pedestrian passages that create a hybrid structure made of metallic and concrete components. To enhance the sustainability of the structure, the incorporation of glass waste as a substitute for cement was proposed. Laboratory examinations were conducted to ensure the incorporation of glass waste did not decrease important structural concrete properties such as resistance and durability. After the laboratory experiments, an LCA was conducted to evaluate the environmental impact from cradle to use, considering the maintenance of the structure.

Through this research, it was found that the use of glass waste with subsequent grounding provided lower environmental impact than the option of concrete without the waste, which can reduce the environmental impact related to repairs and maintenance by 50%. It was also possible to maintain the compressive strength and increase the resistance to chlorides, which is important mainly in sea bridges, where the demand for high-durability concrete is necessary.

2. Materials and Methods

2.1. Third Bridge VIXBR Technical Characteristics

The Third Bridge VIXBR was inaugurated in 1989 and connected the capital of Espírito Santo (Vitória) to the city of Vila Velha. Additionally, it represents a cultural landmark for Espírito Santo (Figure 1 and Figure 2) with an extension of 3330 m and a constant 18.30 m width. The bridge was constructed in a marine environment with a double traffic lane and no pedestrian or cyclist access/lanes.

2.2. Premises for the Technical Solution

The most important requests from users of the Third Bridge VIXBR were the creation of new traffic lanes due to the presence of frequent traffic jams during rush hours, the incorporation of pedestrian and bike lanes, and solutions to frequent suicide attempts. Additionally, pedestrian lanes can also be used by ambulances and police vehicles during rush hours.

The potential to incorporate glass powder into the prestressed hollow-core slabs as a technical solution to the Third Bridge VIXBR represents a possibility regarding technical and environmental issues. In this context, laboratory experiments were conducted to investigate the performance of concrete structures and different environmental impact assessment methodologies.

The LCA methodology from cradle to use for the environmental impact assessment, also considering the maintenance stage, was performed. Therefore, the environmental impact concerning material extraction and production, slab fabrication, construction, and a maintenance period of 100 years was evaluated. Additionally, as defined by Helene [35], the lifespan is the period in which chloride can affect the concrete reinforcement in critical conditions for despassivation. This paper assessed the lifespan using laboratory experiments for the different concrete mixes.

For the technical solution, elevated durability and potential environmental impact mitigations were considered for the material selection. This paper evaluated prestressed concrete hollow-core slabs in a 5 cm thick concrete layer. The support of the concrete elements was designed in metallic structures, using metallic I profiles to create low-weight structures that can overcome large spans.

Another critical point of the pedestrian lane is to allow the same visibility range as vehicle users, which also simplifies the structure. It is possible to visualize the structural system adopted and the strategies to meet the abovementioned requirements (Figure 3 and Figure 4).

2.3. Quantity Evaluation for the Proposal Execution

The evaluation of the needed materials in the pre-molded slab was performed through a structural estimate, which resulted in the details shown in Figure 5.

Additionally, the material consumption for concrete manufacturing was obtained according to the Helene e Terzian [36] method. The dosage planning for the concrete to be used in the bridge retrofit was based on a slump of 100 ± 10 mm after the incorporation of glass residues in the fresh concrete. The project for low core slabs considered twenty-six steel strands that were adopted in the inferior part of the slab with a 12.7 mm nominal diameter (26 PC-190 RB 12.7). Additionally, 8 steel strands were adopted in the superior part, with a 6.5 mm nominal diameter (6.5 PC-190 RB 6.5).

Table 1. Material quantity required to manufacture 1 m³ of a prestressed hollow-core slab.

Material Quantity	Control	GP1-10	GP1-20	GP2-10	GP2-20
Steel strand, 26 PC-190 RB 12.7 (kg)	51.48	51.48	51.48	51.48	51.48
Steel strand, 8 PC-190 RB 6.5 (kg)	3.44	3.44	3.44	3.44	3.44
Concrete (m³)	1	1	1	1	1
Cement (kg)	302.85	272.57	242.28	272.57	242.28
Glass powder (GP) (kg)	0	24.77	49.57	24.77	49.57
Gravel (kg)	1133	1133	1133	1133	1133
Sand (kg)	875.24	875.24	875.24	875.24	875.24
Water (kg)	181.71	181.71	181.71	181.71	181.71

shows the number of materials considering the functional unit of 1 m³ of the built slab.

The total concrete needed for the construction of 1 hollow-core slab was 5 m³, and the analysis of the entire impact was obtained by multiplying the result by 534 hollow-core slabs (2670 m³ of concrete). For the total bridge length (i.e., 3330 m), 267 slabs were needed on each side, totaling 534 slabs, which amounted to 2670 m³ of concrete. Regarding the highest marine proximity, the corresponding extension of 1700 m totaled 1360 m³ of concrete. Figure 6 presents the installation bridge slab projection, referring to the complete length (3330 m) adopted for the LCA.

Considering a 302.85 kg/m³ cement consumption, approximately 808,609.5 kg of cement would be necessary. Additionally, the proposal of partial cement substitution for glass powder can increase up to 20%. Finally, the cement consumption reduction in the present contribution varied from 80,861 to 161,722 kg, with a 10% and 20% substitution, respectively.

The annual production of waste glass powder by the glass manufacturing company considered in the evaluation was 84,000 kg. Hence, the demand for 20% cement replacement would be attended in 2 years, and the 10% demand would be reached in one year. In the present contribution, the construction schedule considered was adequate to the monthly waste availability, fixated at 7000 kg. Hence, considering each slab with a partial 20% cement substitution requires 247.85 kg of waste to produce 28 slabs monthly. With only the bridge route with a higher need for durable concrete (route overseas: 1700 m), the waste availability would implicate a 10-month schedule until conclusion.

2.4. Materials and Experimental Methods

2.4.1. Waste Glass Powder

The waste glass originated from the final product of a water reuse system that treats the water of soda–lime flat glass polishing, and it was collected after the filter press, partially air-dried. For the laboratory analysis, the residue was kept in the kiln at 100 °C until a constant mass was reached. Subsequently, the waste was crushed using a ceramic mortar and pestle to break the lumps, producing the first glass powder (GP1) with an average particle size (µm) 42.17, and the average particle size (µm) of GP2 was 35.02. The Blaine specific surface area (cm²/g) for GP1 was 6121 and 8015 for GP2. Afterward, the glass powder was milled in a ball milling machine to increase the material’s reactiveness, producing ground waste (GP2).

For the grounding process, a ball mill was used that was composed of steel balls 40 mm in diameter and 0.278 kg. The internal diameter of the mill was 380 mm with a volume of 47 L and a capacity of 14 kg per batch. The method adopted evaluated from time to time the potential of dimensional reductions in the glass particles. The measurement was conducted in material that was contained in sieve #325 (45 µm), which resulted in a duration of 2 h. After two hours, little advancement was made, which fixed the milling time to 2 h. The process from glass polishing, transportation to the water reuse station, final residue destination, and milling to the elevation of the reactivity of the material is presented in Figure 7 and Figure 8 and demonstrates the reduction in the particle sizes and the advancement with milling time.

To execute the milling (GP2), the time required to achieve the minimum pozzolanic effect requirements according to ABNT NBR 12653 [37] (

Table 2. Limits for the classification of pozzolanic materials according to NBR 12653 [37]; results were obtained in the laboratory by the authors and detailed in Guignone et al. [38].

Properties	Limits ABNT NBR 12653 [37] Class E	Limits ABNT NBR 12653 [37] Class N	Glass Powder (GP1) (Class E)	Ground Glass Powder (GP2) (Class E)	Metakaolin (ME) (Class N)
Fineness by sieve #325 (45 µm) (% retained)	<20	<20	19.65	7.64	8.14
Pozzolanic activity with lime (MPa) [39]	≥6	≥6	4.14	6.28	11.53
Pozzolanic activity with cement (%) [40]	≥90	≥90	83	90	137
SiO2 + Al2O3 + Fe2O3 (%)	≥50	≥70	68.02	68.02	93.6
SO3 (%)	≤5	≤4	0.21	0.21	0.2
Loss on ignition (%)	≤6	≤10	0.55	0.55	2.5
Available alkalis in Na2Oeq * (%)	≤1.5	≤1.5	19.43	19.43	1.38

* Na₂Oeq = Na₂O + 0.658 K₂O.

) was tested.

The study concluded that 2 h were necessary to reduce the waste retained to 45 µm (#325) from 19.65% to 7.64% (maximum reduction obtained). The average particle size (µm) of GP2 was 35.02. The Blaine specific surface area (cm²/g) for GP1 was 6121 and 8015 for GP2. Afterward, the performance of the concrete with the two types of waste glass powder was evaluated for a 10–20% cement replacement. For the control mixture, cement CPV-ARI was used, because it has no additions in its composition, leading to a better understanding of waste incorporation benefits. The detailed results of the characterization of the materials obtained in the laboratory were developed by the authors and can be evaluated in Guignone et al. [38].

2.4.2. Laboratory Methods for Material–Technical Requirements

Cylindrical specimens (100 × 200 mm), cured in a humid chamber until reaching the test date, were tested for compressive strength. Additionally, the rapid chloride permeability test, recommended by ASTM C1202 [41], was applied, and chloride diffusion coefficient (Dns) tests were performed to estimate the service life (Figure 9). To determine the Dns, three methods were adopted (i.e., UNE 83987 [42]; NT BUILD 443 [43]; NT BUILD 492 [44]), and the average of the results was considered in the service life prediction and subsequently used in the LCA.

Table 3. Dimensions, number of samples, and experimental age of the samples.

Properties	Standard	Dimensions of Specimen	Number of Samples	Age of Testing	Category
Compressive strength	ABNT NBR 5739-07	Cylinder 100 mm in diameter and 200 mm in length	3	28, 56, and 91 days	Concrete
Rapid chloride permeability test (RCPT)	ASTM C 1202-12 [41]	Cylinder 100 mm in diameter and 50 mm in length	3	28 and 91 days	Concrete
Multi-regime method	UNE 83987-14 [42]	Cylinder 100 mm in diameter and 20 mm in length	3	28 days	Concrete
Bulk diffusion test (BDT)	NT BUILD 443-95 [43]	Cylinder 100 mm in diameter and 50 mm in length	3	28 days curing and 60, 120, and 180 days of exposure	Concrete
Rapid migration test (RMT)	NT BUILD 492 [44]	Cylinder 100 mm in diameter and 50 mm in length	3	91 days	Concrete
Slump	ASTM C143	-	3	At mixing	Fresh concrete
Density	ASTM C143	-	3	At mixing	Fresh concrete
Pozzolanic activity with lime (MPa)	ABNT NBR 5751-2015	Cylinder 50 mm in diameter and 100 mm in length	3	7 days	Mortar
Pozzolanic activity cement (%)	ABNT NBR 5752-14	Cylinder 50 mm in diameter and 100 mm in length	3	28 days	Mortar

details the dimensions, numbers, and age of execution of the laboratorial experiments to the samples studied.

2.5. Analytical Methods

Service Life Prediction

To analyze the service life (SL) of the structural elements in concrete, the methodology chosen focused on the estimation of the advancement of ion chloride into the concrete reinforcement and the observance of critical conditions in the armor. The choice was based on the location of the bridge: an environment with a high exposure to salt fog, and because chlorides are responsible for the most aggressive phenomenon related to the process of corrosion of the armors, producing a point corrosion, also known as pitting corrosion, as described by Helene [35] and Ribeiro [45].

Using the chloride diffusion coefficients (Dns) and adopting the Fick second law of diffusion in the non-steady state, it was possible to generate a graph of concrete service life versus chloride penetration depth (Helene [34] and Ribeiro [45]). Equation (1) was developed using the second law of Fick:

$$\begin{array}{l}{P}_{Cl}=2(z)\sqrt{Dt}\\ erf(z)=1-\frac{C_{cl}-C_0}{C_s-C_0}\end{array}$$

(1)

where D is the diffusion chloride coefficient (cm²/year), t is the service life in years, erf is the Gauss error function (z), P_Cl is the chloride penetration depth in which the chloride concentration reached a critical level for depassivation of reinforcement (cm), C₀ is the initial chloride concentration inside the specimens (mass, %), C_s is the chloride concentration on the surface of the specimens (%), and C_Cl is the concentration of chlorides in the depth P_Cl and time t (%).

To generate the graph, some parameters were fixated. The upper limit considered for the depassivation of the reinforcement was 0.4%, related to the cement mass. This value is the only limit of the reinforced concrete according to the International Federation of Structural Concrete (CEB [46]), and it is the average value between the Brazilian norm (ABNT NBR 6118 [47]) of 0.5% and the American norm (ACI-318 [48]) of 0.3%. The chloride concentration adopted in the surface of the concrete (Cs) was 0.6%. Nunes et al. [49] evaluated concrete structures with an age over 15 years and different distances from the sea and obtained values around Cs = 0.6% for up to 630 m. Additionally, Guimarães et al. [50] suggested 0.6% for a distance from the sea between 680 and 5000 m. Hence, the value adopted in the present study was consistent with the literature, admitting variations for the current application.

2.6. Life Cycle Assessment

According to ISO 14040 [20] and ISO 14044 [21], the LCA methodology is composed of four stages: (a) scope and goal definition; (b) life cycle inventory; (c) life cycle impact assessment; (d) interpretation. Additionally, this methodology aims to quantify the environmental impact of a determined product or system throughout its life cycle.

2.6.1. Goal and Scope Definition

This paper aimed to evaluate the environmental impact of the cradle (material fabrication), construction, maintenance, and use of prestressed concrete hollow-core slabs for a technical solution, also considering transport. The paper also aimed to compare different drying alternatives with the environmental impact related to the milling option of using glass powder as a substitute to cement.

Table 4. Functional units used in the current study to evaluate each different aspect of the environmental impact categories analyzed.

Analysis Type/Aim	Mass Functional Unit (1 kg)	Volume Functional Unit (1 m³)	Total Construction (534 Slabs)	Compressive Strength (1 MPa)	Compressive Strength and Service Life (years·MPa)
Comparison between drying methodologies (natural or artificial)	X
Comparison between milling scenarios (milling or not milling)	X
Comparative analysis of the product fabrication for different mixes		X
Comparative analysis of the complete cradle-to-grave bridge retrofit, considering different mixes and slab installations as well as the maintenance for a period of 100 years			X
Comparative analysis of how different compressive strengths can affect the total impact				X
Comparative analysis of how different compressive strengths and service life can affect the total impact					X

shows the functional units used in paper.

2.6.2. Life Cycle Boundaries

The system boundaries are identified in Figure 10.

Material extraction and fabrication, construction, use, and maintenance were considered in the present contribution. Additionally, a 100-year service life for the bridge was considered, based on Younis et al. [51] and Panesar et al. [22], for maintenance on the hollow-core slabs. Maintenance impact also considered the lifespans estimated through experimental analysis for each concrete mix evaluated. In summary, laboratory tests were necessary to evaluate the performance of the concrete (mechanical properties and related to durability) in order to compare the most suitable type of concrete to be adopted in the solution for the bridge. The chloride resistance tests produced the chloride diffusion coefficient measurements, necessary to estimate the service life of the concrete that could be adopted in the solution for the bridge. Using the life cycle assessment methodology and using service life estimates, it was possible to estimate the environmental impact in the bridge maintenance phases.

2.6.3. Data Quality Indicators

Data quality requirements for the present evaluation were (a) temporal correlation: 2020; (b) geographic correlation: Espírito Santo, Brazil, for the only company that beneficiates glass in the state and adopts a water reuse system; (c) technological correlation: technology available in the Brazilian market.

2.6.4. LCI, Materials’ Life Cycle Inventory, and Waste Glass Powder Life Cycle Inventory

The manufacturer’s information relating to the fabrication of rotary and spray dryers was adopted. Additionally, two technical visits were performed in the beneficiary glass enterprise to analyze the technical solution used in glass polishing. The cement producer responsible for the drying, milling, and waste reservation was also visited. Concerning the other main processes, this study’s most recent and representative data found in the literature were used [52,53,54,55,56]. In the absence of these, the comprehensive and current data available in Ecoinvent 3.3 were used [57,58].

Table 5. Materials and cement Life Cycle Inventory.

Main Process	Flux Type	Flux	Value	Unit
Cement (1 kg)	Materials	Clinker	0.9025	kg
		Plaster	0.0475	kg
		Limestone filler	0.05	kg
	Energy/Processing	Electricity	0.0555	kWh
		Packaging	1	kg
	Emissions	Particulate matter	0.09	g
Clinker (1 kg)	Materials	Limestone, crushed	1.3	kg
		Clay	0.2	kg
		Sand	0.1	kg
		Iron ore	0.03	kg
	Energy	Petroleum coke	101.08	g
		Charcoal	3.46	g
		(Hard) Mineral coal	3.45	g
		Firewood (bundle, energy wood)	4.6	g
		Diesel oil (burned)	59.1	kJ
		Natural gas	0.33	dm³
		(Heavy) Fuel oil	0.22	g
		Electricity	0.0555	kWh
	Emissions	Carbon dioxide	0.947	kg
		Carbon monoxide	0.47	g
		Nitrogen oxides	2.17	g
		Sulfur oxides	0.32	g
Limestone filler (1 kg)	Materials	Limestone, crushed	1	kg
	Energy	Electricity	0.06334	kWh
Limestone, crushed (1 kg)	Materials	Limestone	1	kg
	Energy	Electricity	0.000255	kWh
		Diesel oil	0.0034	MJ
	Emissions	Particulate matter < 2.5 μm	1.75 × 10−5	kg
		Particulate matter > 2.5 μm < 10 μm	7.51 × 10−5	kg
		Particulate matter > 10 μm	0.000148	kg
Limestone, raw (1 kg)	Materials	Calcite (mineral)	1	kg
		Water	2.14 × 10−2	dm³
	Soil	Occupation	1.34 × 10−3	m²a
	Energy/ Processing	Explosive (blasting)	1.82 × 10−5	kg
		Electricity	2.73 × 10−5	kWh
		Diesel oil (burned)	0.0333	MJ
	Emissions	Evaporated water	2.14 × 10−2	dm³
		Particulate matter < 2.5 μm	6.08 × 10−6	kg
		Particulate matter > 2.5 μm and <10 vm	5.89 × 10−5	kg
		Particulate matter > 10 μm	0.000159	kg
Clay (1 kg)	Material	Clay mineral	1	kg
	Soil	Occupation	1.67 × 10−4	m²a
	Energy	Diesel oil (burned)	0.0288	MJ
Gypsum (1 kg)	Miscellaneous	Miscellaneous (unmodified inventory)	-	-
Iron ore	Miscellaneous	Miscellaneous (unmodified inventory)	-	-
Sand	Materials	Quartz sand	1	kg
	Soil	Occupation	1.25 × 10−3	m²a
	Energy	Diesel oil (burned)	0.0609	MJ
Gravel (1 kg)	Materials	Granite	1.05	kg
		Groundwater	8.07 × 10−3	dm³
	Soil	Occupation	1.34 × 10−3	m²a
	Energy	Explosive (blasting)	1.45 × 10−4	kg
		Electricity	3.72 × 10−3	kWh
		Diesel oil (burned)	8.28 × 10−3	MJ
		Lubricant oil	6.00 × 10−3	g
		Internal transportation	1	km
	Emissions	Particulate matter < 2.5 μm	6.08 × 10−6	kg
		Particulate matter > 2.5 μm and <10 um	5.89 × 105	kg
		Particulate matter > 10 μm	1.59 × 10−4	kg
Potable water	Miscellaneous	Miscellaneous (unmodified inventory)	-	-

presents a summary of the LCI for the cement and materials.

Material removal and transportation to the cement production facility were considered in this study. Additionally, this paper considered that further waste processing (drying, unraveling, and milling) was performed in the cement fabrication site. The following options were evaluated: (a) natural drying, (b) artificial rotary dryer, (c) artificial spray drying, and (d) natural drying followed by a rotary dryer. After this preliminary analysis, the air-dried (naturally dried) option was evaluated, considering and not considering waste milling.

Natural drying consisted of the use of a tractor with a towable leveling planer and towable disc plow. The process began with the waste spread over the leveling plane, obtaining approximately 2.4 × 10^–3 h machine per ton of waste. Considering the plow working speed was 6 km/h, 2.6 × 10^–3 h machine per ton was obtained. After, the loaders transferred the waste to a dump truck. With 40 s for each volume of the blade, 3.5 × 10^–3 h machine per ton was obtained, totaling 8.5 × 10^–3 h machine/t, the same as 30 s machine/t. The final machine-hours value increased by 50% (46 s/t), adopting a conservative approach due to the assumptions used in the estimates, which increased the uncertainty of the evaluation. The removal process considered that two pieces of equipment were used to remove, crush, and homogenize soils, each one with the capacity to remove 6 m³/h and with a 2 HP (1.5 kW) electric motor [59]. Therefore, the specific consumption was 0.156 kWh/t glass waste. In the rotary dryer process, the residue was previously open air-dried to achieve 40 kg of water evaporated per ton of waste (4% humidity).

According to Mujumdar [60], the typical rotary dryer (Figure 11) energy consumption with a capacity ranging from 30 to 80 kg water/h.m³ varies from 4600 to 9200 kJ/kg evaporated water. Hence, with an average value of 6900 kJ/kg evaporated water, the total energy consumption was 0.276 MJ/kg for the dry waste glass. For a 20 m³ dryer, considering the average capacity of 60 kg water/h·m³, it would be able to dry 1200 kg water/h, or about 9600 kg in an 8 h day, equivalent to 240 t of waste glass powder processed per day. To use the spray dryer, the waste needs to be in a pumpable condition. Thus, this paper considered that the drying process followed a filter press. In this context, the water that needed to be evaporated was 150 kg per ton of dry GP (15% humidity). Additionally, the typical spray dryer energy consumption, with a capacity of 1–30 kg water/h·m³ was 4500–11,500 kJ/kg evaporated water [60]. With the average value of 8000 kJ/kg of evaporated water, the final consumption resulting was 1.2 MJ/kg of dry residue.

For waste milling, the values in Rebello et al. [61] and Alves et. al. [62] were used to estimate energy consumption. Rebello et al. [62] carried out a comparative assessment of the life cycle of waste processing from ornamental stones, sand, clay, and limestone and adopted a consumption of 0.01180 kWh clay milling, based on the national average for thermal energy consumption according to Alves et al. [62]. This value refers to mass milling (considering mills, vibrating screens, elevators, conveyor belts, and granulators). Due to the absence of specific information related to waste glass milling, reference material was adopted that included approximate physical properties (clay) as shown in

Table 6. Life Cycle Inventory: milling and drying of the waste glass powder.

Main Process	Flux Type	Flux	Value	Unit
GP, rotary dryer (1 kg)	Materials	GP (naturally dried)	1.04	kg
	Energy	Heat, natural gas	0.276	MJ
	Emissions	Evaporated water	0.04	dm³
GP, spray dryer after press filter (1 kg)	Materials	GP, cake (press filter)	1.15	kg
	Energy	Heat, natural gas	1.2	MJ
	Emissions	Evaporated water	0.15	dm³
GP, natural drying (1 kg)	Materials	GP, cake (press filter)	1.106	kg
	Energy	Machine	0.046	s/kg
		Electricity	0.000156	kWh
	Emissions	Evaporated water	0.106	dm³
GP, milling process (1 kg)	Materials	Glass waste	1.0	kg
	Energy	Electricity	0.01180	kWh

.

2.6.5. Concrete Production

In this paper, all mixes considered were analyzed as they would be produced in the same batching plant, in the company responsible for the execution of pre-fabricated concrete. The energy consumption of this type of company was not available in a Brazilian inventory. In this context, the study of Bushi, Finlayson, and Meil [63] was used. The concrete produced in the United States was analyzed through the National Ready Mixed Concrete Association (NRMCA). Additionally, diesel quantity for internal transportation was 0.148 gallons/yd³ of concrete, according to the adaptation of the study of Marceau, Nisbet, and Vangeem [64].

2.6.6. Prestressing Slab Procedure

After the mixture, the concrete was transported to the prestress manufacturing site. In this stage, steel strands with high resistance were tensioned using hydraulic prestressing jacks. The pre-tensioning method was used, which means strands were tensioned before the mixed concrete was released into the forms. After curing, the final anchorage of the strand was released. The prestress carried out by the hydraulic jacks and the control system for cranes, winches, and steel cutting machines was employed and taken to the storage yards. The Life Cycle Inventory of the prestressed concrete slab included: (a) quantity of raw materials (active reinforcement steel); (b) electricity consumption (pre-tensioning machines, use of cranes, and steel cutting machines); (c) fuel consumption for transporting materials. For the amount of energy related to the pre-tensioning process, a prestressed concrete pole industry was adopted as a reference, detailed in the studies by Phrommarat and Arromdee [65].

2.6.7. Construction: Slab Installation

This item comprehends the life cycle stage from the exit of the prestressed slabs from the manufacturing site to the installation on the bridge. It also includes installation using hoisting. Additionally, this paper considered the delivery of 7 slabs per week, calculated according to waste availability. Each truck could carry 2 slabs; therefore, 3 trips were required per week, consisting of a 40 km round trip for each trip. Considering 2 years of service, that amounted to 11,520 km. The Ecoinvent database was used for lifting, considering a diesel engine operating 4 h a day, then 80 h a month, and 960 h a year. This study considered 2 years of lifting activities from the slabs, totaling 1920 h.

2.6.8. Use and Maintenance

• Maintenance

This maintenance analysis and repair phase was divided into two main processes: (a) 1 simple visual inspection every year; (b) 1 leading inspection every 5 years [63]. Accidental repairs were not considered due to the data availability. The following inspection procedures were adopted based on Manual Statens Vegvesen n. 136 [59] and Dequidt [31] for the 100 years: (a) Simple visual inspection, every year; no special equipment is necessary for this checkup. Therefore, the impact of this stage derives from the team transportation to the bridge location, which can be achieved by 1 vehicle, also considering the return from the inspection site (20 km). (b) Main inspection, every 5 years; this inspection type requires an elevator. It was estimated that the machine would be rented at a 50 km distance from the bridge site in the metropolitan region of the state of Espírito Santo. Hence, the transportation amounted to 1000 km, based on 100 years total service life. This paper selected a building elevator from the database, considering the machine would be working 8 h, totaling 160 h in 100 years.

• Repairs

The service life of the materials was determined in laboratory tests to assess the repair necessity over time. According to NBR 6118 [47], for aggressive marine environments, the adoption of 50 mm prestressed concrete cover in slabs is recommended. Therefore, based on the chloride migration and diffusion tests results, it was possible to obtain the non-stationary chloride diffusion coefficients and estimate the service life, obtaining the following results regarding the repair periods presented in Figure 12.

In the repair life cycle stage, this paper considered that part of the concrete cover would be replaced by the same material provided for the solution. In addition to the concrete execution, the operation of a 1 ± 1.0 kW breaker hammer was also estimated. Concrete demolition yield considered was 1 m³ period for 1 h with a breaker hammer

The repair paper was constituted of (considering the 100-year horizon): (a) people displacement using light vehicles; (b) demolition of damaged concrete; (c) repair with concrete (most representative elements). The repair activities adopted the premises described by Panesar [22]: The first repair activity occurs at the end of the estimated service life for each concrete mixture. In general, the higher service life of concrete leads to fewer repair needs and less material is consumed, considering the total service life of 100 years for the bridge. Furthermore, this study did not consider the environmental impact associated with other equipment related to concrete repairs. The following repair procedure was adopted based on the references above with adaptations. We assumed that 20% of the total concrete of the slab was affected (damaged) and 50% of the affected concrete (damaged) needed to be replaced (adaptation of Panesar [22]).

Firstly, maintenance actions consisted of removing the concrete cover and providing an adequate surface for coating adhesion. Then, a bonding layer was applied between the old and the new concrete (not considered in this study due to the low significance of the final impact and since it was the same material for all case scenarios). Finally, the concrete was placed to provide new protection against corrosion of the reinforcement. The amount of concrete needed for the repair is shown in

Table 7. Production, maintenance, and repair Life Cycle Inventory.

Main Process	Flux Type	Flux	Value	Unit
Concrete mixer (1 m³)	Energy	Electricity	5.05	kWh
		Diesel oil	27.73	MJ
Pre-tension, crane, and cut control system (1 m³)	Energy	Electricity	0.770	kWh
Steel production (1 m³)	Material	Hot rolling of steel	1.000	kg
	Material	Steel, low league	0.370	kg
	Material	Steel, no league	0.630	kg
	Transportation	Internal transportation	20	km
Construction–transport for installation on the bridge	Energy	Internal transportation	11,520	km
Construction–lifting	Energy	Electricity	1920	h
Simple visual inspection (once each year; 100 years)	Transportation	Internal transportation–low-weight vehicle	3200	km
Main inspection (5 times every year; 100 years)	Transportation of the elevator	Internal transportation	1000	km
	Transportation	Internal transportation–low-weight vehicle	400	km
	Equipment	Elevator operation	160	h
Repair–transportation to site (100 years)	Transportation	Internal transportation—low weight vehicle	400	km
Repair concrete–REF (100 years)	Material	Concrete (repair)	2136	m³
	Energy	Electricity (concrete demolition)	2136	kWh
Repair concrete—RV1-10 (100 years)	Material	Concrete (repair)	1869	m³
	Energy	Electricity (concrete demolition)	1869	kWh
Repair concrete—RV1-20 (100 years)	Material	Concrete (repair)	1468.5	m³
	Energy	Electricity (concrete demolition)	1468.5	kWh
Repair concrete—RV2-10 (100 years)	Material	Concrete (repair)	801	m³
	Energy	Electricity (concrete demolition)	801	kWh
Repair concrete—RV2-20 (100 years)	Material	Concrete (repair)	400.5	m³
	Energy	Electricity (concrete demolition)	400.5	kWh

.

2.6.9. Transportation Distances

The distances for the materials used in the clinker, cement, limestone, and lime are shown in

Table 8. Transportation distances of the materials.

Material Transported	Material Produced	Origin	Destination	Distance (km)
Clinker	Cement	Cachoeiro do Itapemirim, ES	Greater Vitória, ES	160
Limestone filler	Cement	Cachoeiro do Itapemirim, ES	Greater Vitória, ES	170
Gypsum	Cement	Araripe, CE	Greater Vitória, ES	1700
Sand	Clinker	Cachoeiro do Itapemirim, ES	Cachoeiro do Itapemirim, ES	5
Limestone raw material	Crushed limestone	Various sources	Clinker/Limestone filler	5
Clay	Clinker	Cachoeiro do Itapemirim, ES	Cachoeiro do Itapemirim, ES	5
Iron ore Cement	Clinker Concrete slab	Greater Vitória, ES Greater Vitória, ES	Cachoeiro do Itapemirim, ES Greater Vitória, ES	160 5
GP	Concrete slab	Greater Vitória, ES	Greater Vitória, ES	6.5
Sand and gravel	Concrete slab	Greater Vitória—ES	Greater Vitória—ES	5
Steel	Concrete slab	Greater Vitória—ES	Greater Vitória—ES	20
Concrete slab	Bridge	Greater Vitória—ES	Greater Vitória—ES	21.5

. All transportation distances were performed using road vehicles. In this context, EURO 3 with a 16–32 t capacity was chosen in the Ecoinvent database.

2.6.10. Life Cycle Impact Analysis (LCIA)

This study evaluated the global warming potential (GWP) and embodied energy (EE), widely used in LCA studies and with a strong correlation with other impact indicators [66,67]. Additionally, GWP has been the most commonly used impact category to study the effect of transport on environmental performance in a product [22,67], and it transforms all gases that contribute to the greenhouse effect into the unit of kg of CO₂ eq. Furthermore, the EE indicator analyzes consumed energy, considering direct and indirect use in all product life cycle processes. In addition, this paper considered that analyzing methods that represent several categories simultaneously was important. Thus, three multiple category methods were initially chosen, which resulted in a single total impact indicator (TI). In total, three impact indicators were adopted from 4 different LCIA methods: (a) global warming potential (GWP): kg CO₂ eq. (IPCC method [68]); (b) embodied energy (EE): MJ (cumulative energy demand (CED) method (Frischknecht et al. [69]); (c) total impact (TI): Pt (Impact2002+ (JOLLIET et al. [70]); (d) EDIP 2003 (HAUSCHILD et al. [71]). All simulations were performed using SimaPro 9 software.

3. Results and Discussion

3.1. Drying Options for GP

Environmental impact due to the natural drying process was almost imperceptible (Figure 13). Additionally, in the GWP category, the rotary dryer had environmental impact equivalent to the limestone filler alternative (Figure 13A). In the other impact categories, embodied energy and total impact, the limestone filler and the gravel had worse performances when were compared to the rotary dryer (Figure 13B,C).

The spray dryer presented the highest impact among the drying proposals, up to four times the impact of the rotary dryer and 73% of the environmental impact related to the milling for GWP (Figure 13A). Additionally, the spray dryer impacted 4.33 (Figure 13B) and 4.32 times (Figure 13C) more than the rotary dryer option, considering embodied energy and the total impact assessed with IMPACT 2002+, respectively.

In this context, the spray drying option was discarded for further analysis. Although the rotary dryer presented a similar environmental burden as other materials (such as gravel and limestone filler), based on a margin of ±25%, the option was rejected due to the already significant environmental impact from the additional processing of milling needed to produce the GP2. Hence, only the natural drying option (GP1) and natural drying and milling (GP2) were adopted in the comparison among concretes.

3.2. Concrete Environmental Performance Comparison

Considering the analysis of environmental impact for 1 m³ of a manufactured hollow-core slab, a reduction in the environmental impact with partial cement replacement was achieved for all LCIA methodologies evaluated (Figure 14). Additionally, the maintenance or the increase in the concrete compressive strength in the samples tested was noted.

Mohan et al. [25] evaluated glass waste in concrete as a partial replacement for cement and its environmental impact. They identified that, as the strength of the concrete increased, the environmental impact increased, contrary to what was identified in the present research. On the other hand, Tucker et al. [26] identified that the incorporation of glass waste was environmentally advantageous, including impacts similar to the glass recycling process. In research by Mohan et al. [25], for the analysis of conventional concrete (310 kg/m³ of cement) an impact of 290 kg CO₂ eq. was found. The authors also considered glass waste of 20% replacement (62 kg/m³), which resulted in 260 kg CO₂ eq.

In the present contribution, the conventional concrete uses 302.85 kg/m³ of cement, totaling 475 kg CO₂ eq (Figure 14A), and with the replacement of 20% of the total cement by glass powder (49.57 kg/m³), the total is 414 kg CO₂ eq. According to Mohan et al. [25], concrete with the addition of glass waste reduced 0.48kg CO₂ eq for each kg of waste incorporated into the material as pozzolanic material. This research, however, showed that the environmental gain in global warming potential can reach 1.23 kg CO₂ eq.

Using the same methodology for energy use, Mohan et al. [25] reported 3.23 MJ saved with each kilogram of glass waste used to make concrete, while this study found around 8 MJ (Figure 14B). Therefore, with the use of the glass powder, gains related to the reduction in the emission of potential gases for GWP and reduction in EE were noted.

Regarding the total impact (Figure 14C), it was found that the most expressive impact for the manufacture of 1 m³ of the slab was damage to human health. Additionally, replacing cement by the waste up to 20%, either considering the milling or without this additional step, produced equivalent reductions. The milled residue, however, had better performance regarding its mechanical properties.

It is also important to highlight that resistance to compression was maintained after 20% cement substitution with glass waste powder after 28 days. After 56 and 91 days, the mixture with 20% ground glass powder (GP2-20) had results slightly higher than those of the control mixture. These results can be explained by the physical effect of pore-filling, where the empty pores left by the cement paste are filled by the glass particles, reducing the porosity and contributing to the retention of the mixture water, enhancing the cement hydration process. Another process that may have contributed to the compressive strength gain was the possible pozzolanic effect of the glass waste. This possible chemical reaction between the glass powder and the cement (due to the formation of more stable compounds, such as the production of C–S–H originated from the reaction of calcium hydroxide and water) diminished the empty spaces in the transition zone between the paste and the aggregates [17,18].

Additionally, chloride penetration resistance was also evaluated (Figure 15). The ASTM C 1202 [41] test was adopted in which the results indicated that after 28 days, with a 20% cement replacement, moderate chloride penetration was found. However, in the control mixture, a high chloride penetration was observed in both 28 and 91 days. These results showed that in 28 days of curing, the concrete with glass powder acquired around three times more resistance. After 91 days, all concretes had low chloride penetration, except for the non-milled waste that was considered a 10% replacement (GP1-10).

Furthermore, improvement in the resistance against chloride penetration with the adoption of glass waste was also observed by Lee et al. [16], Kamali and Ghahremaninezhad [12], and Matos and Coutinho [18]. Finally, regarding the present results, the concrete with 10% cement replacement without milling can be excluded from the analysis. The options GP1-20, GP2-10, and GP2-20 showed reductions in environmental impact associated with good results in resistance to chlorides.

Figure 16 shows the environmental impact in the endpoint category for 1 m³ of the hollow-core slab. It compares the environmental indicator results with the chloride diffusion coefficient for concrete, measured adopting three laboratory testing methodologies. Notably, the environmental impact tended to reduce with a higher percentage of waste glass powder incorporation, to replace cement, as occurred in GP1-20 and GP2-20. These types of concrete have higher efficiency and lower environmental impact. Additionally, a significant decrease in human health endpoint indicators was also found for the glass powder alternatives.

Figure 17 demonstrates the results regarding the relationship between the total impact of endpoint, the compressive strength and the service life of the concretes. For this purpose, a mixture of two or more properties was performed to build a specific functional unit. It was found that there is a correspondence between the environmental impact and the compressive strength, demonstrating that the use of waste glass as a partial substitute for cement can generate environmental benefits even considering the impact per 1 MPa of strength. If the service life of the concrete were also part of the evaluation, the benefit of using the residue would occur in a more accentuated way.

3.3. Total Environmental Impact of Each Life Cycle Stage

After the evaluation of the environmental impact of 1 kg of material, referring to the drying and milling of the waste and its comparison with other materials, and 1 m³ of constructed slab for comparison with the laboratory examinations, it was necessary to evaluate the total impact referring to the fabrication of all slabs of the bridge proposal. This analysis was performed to compare the results with the construction, use, and maintenance for a 100-year horizon for the technical proposal for the Third Bridge VIXBR. The results for the fabrication of the slab for the entire bridge and the installation environmental impact for 100 years of inspections and repairs are shown in Figure 18.

The highest environmental impacts refer to the material fabrication phase. These results agree with the finding of Martin et al. [30], who performed an LCA on a standard bridge, with the following stages: material stage (extraction of raw material, production and distribution), construction, use and maintenance, and end of life. The results showed that the main contributor to environmental impact was the material production phase, accounting for 64% of the total results. The results of this research also corroborate with Du and Karoumi [33] in which an LCA on railway bridges was performed. The authors concluded that the initial stage of manufacturing the material is responsible for the most significant environmental demand. The present study found that the control mixture presented a manufacturing-related impact of 55% (Figure 19).

When considering the total environmental impact for the fabrication of the solution and not only the analysis of 1 m³ of the constructed slab, the differences among the environmental impact methodologies became less perceptible, which indicates a similarity among the results. Additionally, the environmental impact related to the maintenance and repair for 100 years was around 80% of the fabrication cost for the concrete reference solution and up to 15% for partial substitution of 20% of the milled waste (GP2-20) according to Figure 14C and the methodology IMPACT 20002+, totaling a reduction of 65% of the environmental impact.

It is possible to reduce the environmental impact related to global warming (GWP) for the manufacture of materials using GP2-20 (Figure 14A) by 163,090 kg CO₂ eq. For 100 years, we would have 1630.9 kg CO₂ eq. per year. According to Feiz et al. [72], the production of 1000 kg clinker can generate 800 kg CO₂ eq. In 100 years, cement replacement by glass powder could offset the production of about 204 tonnes of clinker for cement manufacture. According to Feiz et al. [72], the CEM III/A 42.5 cement, equivalent to the Brazilian CPIII (ABNT NBR 16697 [73]) contains 47% clinker and 45% blast furnace slag, which could reduce clinker production in 400 tons of cement.

Regarding the impact related to the repair activities, the reference concrete showed that around 60% of the total are related to the manufacturing impacts. On the other hand, for GP2-20, 11% of the repair impacts are related to material fabrication. This difference represents a reduction of 49% and 614,140 kg CO₂ eq. Considering the same understanding related to manufacturing, we would have around 768 tons of clinker production avoided. Adding up the manufacturing and repair impacts, a total of 972 tons of clinker production would be avoided. Hence, the environmental impacts referring to material milling would not have adverse effects on the environmental impact. Quite the contrary, they would have the highest demand of concrete use with waste glass powder, the highest benefits to the environmental indicators analyzed. Finally, glass powder presents less repair cost due to the lower necessity of repair over time.

The usage of up to 20% of milled waste had around 5 times fewer repair necessities. Additionally, it is also important to emphasize that for the environmental impact related to the milling, in the evaluation of 1 kg of material and compared with other drying methodologies, the impact from the fabrication of 534 slabs becomes insignificant. The difference between concretes with or without milling is not perceived. Lastly, the use of concrete with glass powder presented an advantage regarding the mechanical properties, durability, and mitigation of environmental impact.

4. Conclusions

Environmental impact related to waste milling was significant when compared to waste drying, sand, gravel, and limestone indicators, considering the evaluation of 1 kg of material. However, when the evaluation focused on the material incorporated for the manufacture of one slab, the difference was not significant.

Additionally, with a 20% cement replacement, in the manufacture of all slabs, glass powder (GP2-20) could decrease the environmental impact related to repairs and maintenance by 50%, considering the maintenance horizon of 100 years. Furthermore, the fabrication phase presented the most significant environmental impact. Finally, considering total impact analyzed by the IMPACT 2002+ methodology, a significant difference between the relevance of the maintenance and the fabrication phases was observed.

The scenario considering a 20% cement substitution for waste glass powder in the concrete fabrication decreased 1.23 kg CO₂ eq. and 0.008 GJ energy use for each kilogram of waste incorporated in the material. Another important finding showed that glass waste could reduce 972 tons of clinker in concrete production, decreasing 2000 tons of cement manufacture of the type CEM III/A 42.5, equivalent to the Brazilian CPIII.

This research also showed that using GP2-20 can ensure that the compressive strength is maintained, chloride resistance is increased, and a reduced environmental impact is identified. Additionally, the lesser need for repairs can lead to economic benefits, when considering the long-term analysis of the project.

Finally, this paper identified that the gain in durability in concrete due to the glass powder incorporation provided a significant reduction in the environmental impact. The technical proposal to prevent suicide also presents an important social value to the community, offering a solution to frequent suicide attempts on the bridge, and access to ambulances and police vehicles, among others.