Energy and Surface Performance of Light-Coloured Surface Treatments

Abstract

This study presents the evaluation of the photometric performance and energy-saving potential of light-coloured pavement mixtures (LCPMs) in road lighting applications, along with their effects on surface friction, macrotexture, and specularity. The application of LCPMs in tunnels can enhance road surface illumination, thereby improving driver visibility, increasing road safety and comfort, and reducing energy consumption per kilometre. While such surface treatments enable more efficient and cost-effective lighting, maintaining an optimal balance in surface performance poses many challenges due to the impact on concurrent targets in terms of friction, macrotexture, noise contribution, and specularity. Indeed, issues related to friction performance, macrotexture characteristics, and the concurring energy-saving potential of LCPMs remain insufficiently explored. To this end, investigations were conducted to assess the energy-saving potential of light-coloured surface treatments and to evaluate the photometric, frictional, and macrotexture properties of different densely graded LCPMs. A new method was set up and implemented to compare different surface treatments. The results indicate that light-coloured surface treatments increased the average luminance coefficient (up to 0.2406), with glass-containing mixtures offering greater potential for improved surface texture, friction, and energy-efficient road lighting.

1. Introduction

The road surface in the tunnel is usually constructed with traditional black asphalt pavement. However, lighting traditional black roads can cause problems such as higher energy consumption, urban heat island formation, and reduced stiffness due to the heat absorbed by the pavement [1]. However, tunnels should be equipped with adequate interior road lighting to provide drivers with sufficient visibility and visual comfort for safe and efficient travel [2]. Therefore, the application of light-coloured pavement mixtures (LCPMs) in tunnels stands out as a promising application as it can improve visibility for vehicle users by increasing road surface illumination, thus increasing road safety and comfort, and at the same time reducing energy consumption per kilometre [3,4].

Highway tunnel coatings can be designed using LCPMs that incorporate photoluminescent materials. LCPMs are specialised surface pavements produced either by mixing conventional black asphalt with various light-coloured aggregates, pigments, additives, and other materials at controlled temperatures, or by applying reflective materials directly onto the asphalt surface [5,6]. When photoluminescent additives are incorporated, these materials absorb photons from light sources and re-emit the stored energy as visible light. This self-luminous behaviour enhances pavement visibility within tunnels and contributes to reducing the need for external lighting. Moreover, the use of diffuse rather than specular reflection through matte (i.e., lusterlessness) finishes minimises the risk of glare, preserving visibility and driver safety [7]. The combination of diffuse reflection and increased brightness results in improved photometric efficiency, making LCPMs ideal for environments requiring enhanced visibility, such as underground tunnels and enclosed parking structures [8].

One of the primary benefits of LCPMs is their improved photometric performance, which translates to increased solar reflectance and luminance. These changes positively affect both surface temperature regulation and energy consumption for night time lighting. Typical new asphalt exhibits very low albedo values ranging from 0.05 to 0.10, which may increase slightly to 0.10–0.15 as the surface weathers [9]. In contrast, LCPMs can reach initial albedo levels of 0.4 to 0.8, significantly reducing solar radiation absorption and resulting in cooler surface temperatures [9]. Ferrari, et al. [10] stated that when the albedo level of asphalt surfaces increases to 0.4, the surface temperature decreases by 4–5 °C by reflecting more sunlight. Moreover, in a reflective pavement pilot area in Doha, coated sections showed up to 7 °C lower midday surface temperatures compared to adjacent uncoated asphalt [3]. These coatings also enhance the reflectivity of visible light. Studies have shown that using reflective coatings in urban environments can reduce lighting power needs by 20–30% while still maintaining visibility standards [2,3]. This is particularly advantageous in tunnels and parking lots, where artificial lighting represents a significant portion of energy consumption. The increased luminance of these surfaces enables the use of fewer or lower-wattage lighting fixtures. For example, Diao, Peng and Ling [3] reported that incorporating light-coloured aggregates like marble chips into asphalt significantly increased pavement brightness, reducing the need for tunnel lighting. Additionally, Moretti, et al. [11] indicated that LCPMs can lower the energy consumption of artificial lighting systems used for roadway visibility by up to 30%, thereby contributing to the reduction in greenhouse gas (GHG) emissions such as CO₂.

However, despite their energy and environmental benefits, the types of additives used in the production of LCPMs can alter the surface macrotexture of the pavement, which in turn affects wet skid resistance—a critical factor in ensuring traffic safety and driving comfort [12]. Therefore, macrotexture plays a vital role in ensuring friction and diffused light reflection. Fundamentally, coatings must preserve or enhance this texture to maintain adequate skid resistance and minimise glare. Historically, some light-coloured surface treatments slightly reduced skid resistance by smoothing over aggregate texture [4]. However, recent solutions have embedded anti-skid aggregates such as bauxite or angular sands into the coatings, which restored and often enhanced friction properties [13]. For instance, Kim, Choi, Lee, Rodrigazo and Yeon [7] achieved a British Pendulum Number (BPN) of 90, a value exceeding the performance of many untreated asphalts. At this point, an important issue is that the macrotexture remains “open and clean”. Indeed, soiling and rubber deposits can diminish both friction and reflectance. Moreover, texture also affects the nature of light reflection. A coarse surface helps disperse light uniformly, reducing glare and enhancing visibility. Hence, reflective pavements that maintain good macrotexture contribute to improved visual comfort, particularly in tunnels where lighting uniformity is critical [3].

Furthermore, durability remains a key concern when applying reflective coatings. While early light-coloured treatments sometimes underperformed in high-stress environments [4], recent advances in material engineering have significantly improved wear resistance. Polymer-modified coatings and epoxy-based binders now allow for reflective surfaces that retain durability under traffic and environmental exposure. Chen, et al. [14] observed 21–43% lower material loss under abrasion tests in reflective coatings compared to conventional dark asphalt sealants. Yi, et al. [15] tested an epoxy-based coating that maintained performance without visible cracking after 120 days of accelerated ageing, estimating a useful life of up to one year under high traffic. For low-traffic scenarios such as parking lots, the effective lifespan could extend to 3 years or more [9]. Surface enhancement with durable, light-coloured aggregates, such as quartz sand or ceramic granules, helps increase skid resistance and mechanical robustness. However, excessive aggregate content can lead to cohesion loss. An optimal balance was reported by [9,13,16], who found that a 30% sand mix achieved ideal wear resistance and skid performance. An exemplary durable coating was demonstrated by Kim, Choi, Lee, Rodrigazo and Yeon [7], who applied a methyl methacrylate coating with aluminium silicate. This coating demonstrated only 0.1% wear after 500,000 load cycles and maintained a BPN of 90, indicating superior skid resistance.

To sum up, from a sustainability perspective, LCPMs could offer multiple benefits: longer service life, lower energy consumption, and fewer GHG emissions.

Table 1. Comparative performance metrics of untreated and light-coloured treatments from a sustainability perspective.

Performance Metric	Conventional Asphalt (Untreated)	Light-Coloured Asphalt	Reference
Solar Reflectance (Albedo)	0.05–0.10 new; 0.10–0.15 aged	0.4–0.8 initial; 0.3–0.5 after weathering; up to 0.20 after ageing	[1,25]
Solar Reflectance Index	4–6%	45–65%	[26,27,28,29]
Average luminance coefficient (Q0)	0.076 cd·m−2·lx−1	0.076–0.157 cd·m−2·lx−1 for white-coloured asphalt	[2,30]
Illuminance	6.15 lx in tunnel (2 W LED power)	7.8 lx for light-coloured asphalt in tunnel (2 W LED power)
Lighting Efficiency	Baseline (100% lighting power required)	Up to 20–30% lower lighting power needed	[2,3]
Peak Surface Temperature	65–70 °C in hot climates	Typically 5–10 °C cooler (e.g., 58 °C)
Abrasion Loss	Higher material loss under wear	21–70% lower abrasion loss in tests	[1,9,14,31]
Skid Resistance (BPN)	50–70 (adequate when new)	Up to 90 with added aggregates (very high)	[7,17]
Service Lifespan	Not applicable (asphalt lasts 10–15 years before resurfacing)	0.5–3 years on low-traffic roads; 1 year on high-traffic roads; up to 7 years for special applications	[9,15]

summarises the comparative performance metrics of untreated and light-coloured treatments from a sustainability perspective. In Table 1, the performance criteria associated with lower energy consumption and reduced GHG emissions are grouped as solar reflectance, average luminance coefficient, lighting efficiency, and maximum surface temperature. At the same time, those related to coating durability include abrasion loss, skip resistance, and service lifespan. As shown in Table 1, the values for solar reflectance (albedo), solar reflectance index, average luminance coefficient and illuminance are significantly higher in reflective pavement such as LCPMs compared to conventional asphalt. For instance, reflective pavement applications in cities such as Los Angeles have demonstrated measurable reductions in ambient air temperatures—by 1–2 °C—leading to improved pedestrian comfort and reduced cooling energy demands in nearby buildings [2]. On a broader scale, Xu, et al. [17] modelled the climate effects of reflective surfaces and concluded that they can significantly reduce radiative forcing and CO₂ emissions. Another key benefit of reflective pavements is the reduction in lighting energy demand, particularly in energy-intensive environments like tunnels. For instance, Diao, Peng and Ling [3] reported that reflective coatings can lower electricity use for lighting by up to 30%, offsetting initial installation costs through long-term savings. Additionally, by reducing peak surface temperatures, these coatings help slow down pavement degradation processes such as oxidation and rutting. Some coatings incorporate recycled content, such as waste marble or ceramic by-products, enhancing their environmental profile [18]. Cost-wise, coatings involve a higher upfront investment and periodic reapplication [9]. However, life-cycle assessments often show that these costs are offset by reductions in maintenance frequency, energy bills, and environmental externalities [19]. Additionally, machine learning-based predictive models and 3D deep learning algorithms have just been incorporated into the studies to evaluate potential wear and texture characterisation and quantify service life extension in the design of more sustainable light-coloured roads, including concrete roads [20,21,22,23,24].

Based on this background, the long-term energy and environmental benefits of LCPMs appear highly promising. Their increasing implementation, particularly in tunnels, demonstrates their potential as sustainable infrastructure solutions within the broader context of global climate and energy challenges. However, concerns remain regarding their impact on surface friction, which may compromise traffic safety. In particular, issues related to friction performance, macrotexture, and actual energy savings require further investigation.

Furthermore, building on the ongoing expertise and technical insights gained through the IASNAF initiative and the currently active LIFE SILENT project (“Sustainable Innovations for Long-life Environmental Noise Technologies”, LIFE22-ENV-IT-LIFE-SILENT/101114310), this study marks the initial stage of a new research effort aimed at expanding and enhancing previous outcomes. LIFE SILENT continues to investigate dense-graded asphalt mixtures for durable and low-noise pavement applications. In alignment with this objective, the present work explores the potential synergy between surface friction longevity and energy savings, particularly in road tunnel environments. It focuses on evaluating the effects of various light-coloured dense-graded mixtures on energy savings and on the photometric, frictional, and macrotexture properties of pavement surfaces. By integrating outputs learned from ongoing projects, this study seeks to further advance multifunctional pavement design, enhancing both traffic safety and sustainability through innovative material applications. Figure 1 shows the specific, measurable, achievable, relevant, time-bound (S.M.A.R.T.) objectives of this study. To achieve these objectives, the studies detailed below were carried out:

Task 1. A comprehensive literature review was conducted to explore the relationship between photometric properties and energy savings (see Section 1).

Task 2. Experimental design and implementation (see Section 2):

Task 2.1 Preliminary testing on reference surfaces

Task 2.2. Design and production of identical dense-graded friction course samples, categorised as follows:

ST1: Reference surface (black, untreated).

ST2: Surface treated with white polymer.

ST3: Surface treated with a mixture of white polymer, white powder (cement + quartz sand), and water.

ST4, ST5: Similar to ST3, but with added glass fibres.

ST6, ST7: Surface treated with white polymer, white powder, water, glass fibres and glass aggregates.

Task 2.3. Testing procedures included the following steps for each sample:

• Initial non-destructive measurements of geometric, volumetric, mass, photometric, texture, and friction properties on the black surface.
• Preparation of surface treatment mixtures (see

  Table 2. Asphalt binder properties.

  Properties	Value	Standard
  Penetration (0.1 mm) at 25 °C	52–61	EN 1426 [32]
  Softening point (°C)	49–51	EN 1427 [33]
  Ductility at 25 °C (cm)	>100	EN 13589 [34]
  Elastic recovery (%)	9–11	EN 13398 [35]
  Dynamic viscosity at 100 °C (mPa.s)	5000–6000	EN 13702-2 [36]
  Dynamic viscosity at 135 °C (mPa.s)	300–800	EN 13702-2 [36]

  ,

  Table 3. Mechanical properties of aggregates (of HMA).

  Properties	Value	Standard
  Los Angeles abrasion (%)	18	EN 1097-2 [37]
  Flakiness index (%)	12	EN 933-3 [38]
  Polished stone value (PSV)	44	EN 1097-8 [39]
  Sand equivalent (SE)(%)	95.3	EN 933-8 [40]

  ,

  Table 4. The volumetric characteristics of HMAs.

  HMA Name	Density (Gmb) t/m3	Air Void (Av)%	Bitumen Content (B)%	Sand (S)%	Filler (F)%
  H1–H7	2.3–2.4	1–6	5.3–6.4	21–28	6.2–7.6

  ,

  Table 5. Supplementary material parameters.

  Material	Parameter	Specification	Basis for Parameter Determination
  White Polymer	Type	Acrylic and polymer-modified material.	Prior coating studies. Ensures adequate adhesion without excessive viscosity.
  White Powder	Composition	40% Portland cement + 60% quartz sand	Ratio selected from prior IASNAF/SILENT studies for binding and reflectance.
  Glass Fibres	Length	5 mm (short-cut fibres)	Literature suggests 5 mm is optimal for dispersion and surface reinforcement.
  	Diameter	~12 µm	Industry standard for chopped E-glass fibres.
  Glass Aggregate	Particle size distribution	≤4.75 mm, with 70% between 2.36 and 4.75 mm	This range balances stability, friction, and texture.
  Water	Ratio	5–7% of dry weight	To aid polymer activation, cement hydration, and uniform mixing.

  and

  Table 6. Material composition of HMA, H, and of the mixtures for surface treatment, ST.

  S. Treatment ID	ST1	ST2	ST3	ST4	ST5	ST6	ST7
  HMA Past Name	H1 = 92-UCEBFF-3SSFF-MDI	H2 = 78-UCB-5.30	H3 = 109-UCBFF-3SSFF-MDI;	H4 = 87-UCB-5.30	H5 = 97-UCB-5.30	H6 = 81-UCBFF-35SSFF- MDI	H7 = 1LOP
  HMA Project	IASNAF	IASNAF	IASNAF	IASNAF	IASNAF	IASNAF	C.03-2023
  HMA mass, g	1152.53	1133.07	1155.9	1150.83	1146.64	1150.99	1141.09
  White polymer, g	-	52	9.45	9.45	9.45	9.45	9.45
  	0%	100%	18.17%	18.17%	18.17%	18.17%	18.17%
  White powder, g (40% cement + 60% quartz)	-	-	39.11	37.92	36.72	30.75	31.94
  	0%	0%	75.21%	72.92%	70.62%	59.13%	61.42%
  Water, g	-	-	3.44	3.33	3.23	2.70	2.81
  	0%	0%	6.62%	6.40%	6.21%	5.19%	5.40%
  Glass fibre, g	-	-	-	1.3	2.6	1.3	2.6
  	0%	0%	0%	2.5%	5.00%	2.50%	5.00%
  Glass aggregate, g	- 0%	-	-	-	-	7.8	5.2
  		0%	0%	0%	0%	15.00%	10.00%
  White material total, g	0	52	52	52	52	52	52

  ), allowing the mixture to rest for 1 min before applying it to the HMA sample.
• Application of the mixture on the sample surface using tape to prevent lateral spillage.
• Two-day curing period.
• Post-treatment non-destructive measurements of the same properties.

Task 3. Evaluation of the effects of surface treatments on macrotexture and friction properties (see Section 3.1).

Task 4. Preliminary photometric testing and assessment of the impact of surface treatments on photometric indicators (see Section 3.2).

Task 5. Set up and implementation of a new method for the assessment of the optimum surface treatment design that maximises both energy efficiency and traffic safety, based on experimental findings and literature review.

2. Materials and Methods

This section addresses:

(1)

The materials produced. Basically, hot-mix asphalts (HMA) were prepared, and, on each sample, a given light-coloured surface treatment was applied.

(2)

The methods used for testing them.

2.1. Materials and Mix Design

HMA Samples Production

The asphalt binder supplied by the Peretti Petroli S.P.A. bitumen processing plant in Puglia, Italy, was used in dense-graded hot mix asphalts (HMAs). Asphalt binder properties are presented in Table 2.

Crushed stone aggregates supported with limestone mineral powder supplied by Costruzione Due Mari S.r.l. (Maida, Catanzaro, Italy) were used in the produced dense HMAs, and their mechanical and volumetric characteristics are given in Table 3. In addition, two different aggregate gradations were used for five different mix designs, and the curve is shown in Figure 2.

Moreover, dense-graded HMAs with identical aggregate gradation and bitumen content were prepared in accordance with the SUPERPAVE design criteria, following the EN 12697-3 [41] standard. The volumetric characteristics of HMAs produced are given in Table 4. After HMAs samples were produced, the surface treatments explained in Section 2.2 were applied on the HMA specimens.

2.2. Surface Treatment

In this study, four types of materials were used for light-coloured surface treatments on the asphalt pavement surface: white polymer, white powder (cement + quartz sand), glass fibre and glass aggregate. Water was used to mix the materials homogeneously and activate their contents. Figure 3 illustrates the materials used for these treatments.

Table 5 shows the supplementary material parameters. According to Table 5, white polymer is a white-coloured surface material consisting of a liquid mixture of emulsified acrylic resin, to ensure sufficient adhesion without excessive viscosity. White powder is composed of 40% white Portland cement and 60% quartz sand. Additionally, short-cut glass fibres with a length of 5 mm and a diameter of ~12 µm were incorporated, as literature highlights this size for optimal dispersion and reinforcement. Glass aggregate with a nominal maximum aggregate size lower than 4.75 mm, where 70% lies between 2.36 and 4.75 mm, was selected to balance stability, friction, and surface texture. Finally, water was added at a ratio of 5–7% of the dry weight to activate the polymer, ensure uniform mixing, and allow for cement hydration. Based on these materials, seven asphalt mixtures were designed in total, including six with light-coloured surface treatments and one untreated reference asphalt mixture. The material composition details of these mixtures, together with the untreated reference asphalt mixture (conventional black coloured), are given in Table 6.

Initially, the dimensions and weights of the HMAs were recorded. To prepare the ST mixtures, the procedure illustrated in Figure 4 was followed, using the material compositions listed in Table 6. Using the mixing ratios given in Table 6, the required amount of ingredients was manually mixed in a bowl at room temperature for one minute. The prepared mixture was then left to rest for one minute, during which a tape was wrapped around the sample edges to prevent lateral shedding and to allow proper placement of the material on the surface. After a fixed curing period of two days, the dimensions and weights of treated samples were measured again. The curing process is particularly important for ST2 applications, which contain a high proportion of white polymer, as the white colour tends to shift to a grey tone during curing (see Figure 5). However, in other ST applications, colour change was not detected.

The labelled HMAs subjected to surface treatments that reach their final properties upon curing, as seen in Figure 6, were categorised as follows:

Each sample consists of a bottom part (HMA, referred to as Hj) and an upper part (surface treatment, referred to as STj). ST1 represents the black-coloured reference asphalt mixture (i.e., in this case, no surface treatment was added). ST2 refers to the asphalt mixture treated only with white polymer using the microsurfacing method. ST3 was obtained by applying a composite material consisting of White polymer, white powder, and water onto the asphalt surface. ST4 and ST5 include the same composite material as ST3, with the addition of glass fibres. ST6 and ST7 also include glass aggregates. To produce LCPMs containing glass aggregate, the original aggregates were partly replaced with glass aggregates of equivalent mass and similar grain size distribution, thereby incorporating the glass aggregates into the asphalt concrete surface. To this end, it is noted that some recent papers have shown that it is possible to use crushed glass as a partial substitute for “fine” aggregates in road bituminous layers, where different percentages are indicated as optimal, and different glass granule sizes are specified. The literature points out that a glass content ranging from 8% to 15% is optimal for the mechanical properties of the asphalt mixture in terms of Marshall Stability, rutting resistance, and skid resistance. These studies also indicated the best performance for glass particle sizes lower than 4.75 mm. One example is the work of Alhassan, et al. [42], who identified a percentage of 8% glass (≤4.75 mm) as best for stability and air voids, while Afkhamy Meybodi [43], Shafabakhsh and Sajed [44] were in favour of 15% (2.36–4.75 mm) to achieve better skid resistance and rutting performance. Previous studies have shown that higher replacement levels of 20–50% could be used; however, they might lead to a reduction in moisture resistance and, thus, further design considerations should be incorporated [45,46]. The overall review of literature suggests that a 10% glass content (with particles ≤ 4.75 mm) is the fraction that is the most commonly recommended and is thus the most generally sustainable, cost-effective, and mechanically properties-improving fraction (cf. also ([47,48,49,50]).

2.3. Macrotexture and Friction Properties

For surface texture and friction properties of the produced samples, before and after surface treatment, the relevant standards and test setups are presented in

Table 7. Methods and their standard used to determine macrotexture and friction properties.

Test	Unit	Detail	Standards/References	Set Up/Literature
PTV*	-	Wet PTV typically 35–60 (higher = better). UK Highways minimum 36 (wet) for safety and design targets often ≥45 (low risk) or ≥60 (high risk) [54]. Note: values depend on aggregate polish and contamination.	BS 7976-2:2002/A1:2013 (pendulum test) [55]; EN 13036-4 (pendulum) [51]; ASTM E303 [56]; ISO 13473-2 (friction) [57]; PIARC Road Tunnels Manual [58].	British Pendulum Skid Tester (e.g., Munro or James Heal “Sliptest”); GripTester (Kassel); Deighton pendulum; BPN/PSV testers (though PSV is material only); Dynamic Friction Tester (DFT).
MTD*	mm	0.5–1.0 mm (new hot-mix asphalt, depends on mix); FHWA suggests 0.7 mm on high-speed roads (Lower texture 0.3–0.5 mm for old surfaces; >1.0 mm for open-graded friction courses.) [59].	ISO 13473-1:2019 (macrotexture by sand patch) [60]; ASTM E965 [61] (AASHTO T279 [62], volumetric sand patch); EN 13036-1 (sand patch method) [52]; PIARC Road Tunnels Manual [58]; FHWA/MaST [63].	Sand patch kit (volumetric method, e.g., ASTM sand cone); Circular Track Metre (CTM, e.g., Kj Law CTM-III or TRL CTM); laser profilers (e.g., Pavemetrics LCMS series; Dynatest/ViDiG LTS-1, Dynatest 365); high-speed texture profilers (Keyence LJ-X, LSL instruments).

. To assess skid resistance, the British Pendulum Skid Tester was used in accordance with EN 13036-4 [51], while surface macrotexture was evaluated using the Sand Patch method in line with EN 13036-1 [52]. For detailed procedures involving macrotexture depth (Mean texture Depth, MTD*) and skid resistance (Pendulum Test Value, PTV*), an improved testing protocol developed by Praticò and Astolfi [53] was followed. Specifically, the testing sequence included: mass/volumetric measurements, photometric analysis, macrotexture assessment, surface cleaning, skid resistance testing, and gentle surface drying. Variations in the MTD* and PTV* values were analysed before and after surface treatment. A decrease in either indicator reflects a reduction in surface texture or friction, which may compromise overall traffic safety.

2.4. Photometric Measurements

The reflective properties of asphalt pavement surfaces treated with white-coloured materials were evaluated using a series of photometric indicators, including luminance, illuminance, average luminance coefficient, specularity factor, and reflection factor. It is observed that the following list of standards dealing with pavement/tunnel lighting can be considered:

• CIE 144:2001 [64]: Road surface and road marking reflection characteristics (The report also offers recommendations for measuring road surface and marking reflectivity on site).
• CIE 88:2004 [65]: Lighting of tunnels and underpasses (where recommendations concerning the daytime and night time lighting are given).
• CIE 194:2011 [66]: On Site Measurement of the Photometric Properties of Road and Tunnel Lighting (where essential photometric quantities and types and characteristics of instruments, the conditions and methodologies of measurements, the evaluation of uncertainty in measurement results and data elaboration are discussed).
• CEN 14380:2004 [67]: Tunnel lighting (where the design is considered, based on photometric considerations).
• EN 13201-1:2015 [68]: Road lighting—Part 1: Guidelines on selection of lighting classes (where geometry of the road, markings, visual environment, difficulty of the navigation task, lack of visibility, risks of glares due to existing elements, local weather, specific users such as high rate of elderly or visually impaired people are considered to choose the appropriate lighting class).
• EN 13201-2:2015 [69]: Road lighting—Part 2: Performance requirements (where lighting classes for road lighting, aiming at the visual needs of road users, are considered).
• EN 13201-3:2015 [70]: Road lighting—Part 3: Calculation of performance (where the photometric performance of road lighting installations is considered).
• EN 13201-4:2015 [71]: Road lighting—Part 4: Methods of measuring lighting performance (where procedures for measuring various photometric parameters like luminance and illuminance are considered).
• EN 13201-5:2015 [72]: Road lighting—Part 5: Energy performance indicators (where energy efficiency in different road lighting installations is considered).
• UNI 10439:2001 [73]: Illuminotecnica—Requisiti illuminotecnici delle strade con traffico motorizzato (where important details about measurements and specifications are provided).

Table 8. Methods and their standard used to determine photometric properties.

Test	Unit	Detail	Standards/References	Device
Luminance	cd·m−2	Threshold zone: 60–330 cd·m−2 (low to high speed/volume) Interior zone: 1–3 cd·m−2 in extra-urban low-speed tunnels; Exit zone often 5× interior.	CIE 088:2004 Guide for road tunnels [65]; CIE 061:1984 (tunnel entrance) [74]; ANSI/IES RP-8-22 [75]; PIARC Road Tunnels Manual [58]; EN 13201 (road lighting) [68];	Luminance metres (e.g., Hagner S5 S/N 119; Konica Minolta LS-110/LS-100; Topcon BM-7A; Tektronix/Keysight J-17L; Dr. Lange LLG-450 “ProfiLux”); high-range lux metres; certified colorimeters.
Illuminance	Lux	Entrance/threshold: on the order of 500–1000 lx (daylight conditions); Interior zone: tens of lx (e.g., 30–50 lx) depending on design (lower at night); Exit zone: several times interior (often 150–250 lx) to counter daylight.	CIE 088:2004 Guide for road tunnels [65], CIE 061:1984 (tunnel entrance) [74]; ANSI/IES RP-8-22 [75]; PIARC Road Tunnels Manual [58]; EN 13201 (road lighting) [68];	Illuminance metres/light metres (e.g., Hagner S5 S/N 119; Sekonic L-308, Extech HD450, Testo 540); many luminance metres (Konica Minolta, Topcon) can switch to lux mode; spectroradiometers.

shows methods, standard and their devices used to determine photometric properties. In this study, the procedure for photometric measurements was carried out in accordance with the EN 13201 [68,69,70,71,72] and UNI 10439 [73] standards, using the Hagner S5 S/N 119 device, as shown in Figure 7 and Figure 8 and described below:

Reflectance properties of asphalt pavement surfaces measurements were performed on laboratory-cast STj samples using a goniophotometric rig. STj represents the ID number of the surface-treated mixture (see Table 6). In Figure 7 and Figure 8, the sample is mounted horizontally and illuminated by a point-like light source positioned at a fixed height H above the sample centre. The source can be moved sideways by a distance r′ from the centre to set the incidence angle. A precision photometer (Hagner S5) is aligned to view the sample at an elevation angle α = 1° above the surface plane. This 1° observation angle mimics the low-angle view of a driver’s eye and is standard in CIE road lighting measurements. The angular coordinates are defined as follows: β is the azimuthal angle between the vertical plane containing the light source and the vertical plane containing the photometer, and γ is the incidence angle from the perpendicular, given by tan (γ) = r′/H. In practice, the sample (or source) is rotated in azimuth to achieve the desired β values. At the same time, the horizontal offset r′ is adjusted to set specific γ angles (e.g., including γ = 0 when r′ = 0). For each measurement geometry (β,γ), the surface illuminance E (in lux) and the surface luminance L (in cd·m⁻² ) in accordance with the UNI 10439 standard are recorded. Illuminance is measured on the pavement surface using a calibrated luxmeter, and the photometer measures luminance in the specified 1° direction. The key photometric quantity is the luminance coefficient q, defined as the ratio of luminance to illuminance:

$$q (\alpha ,\beta ,\mathsf{\gamma }) = {\displaystyle \frac{L}{E}}$$

(1)

With $\alpha$ = 1° fixed. At each $(\beta ,\mathsf{\gamma })$, the value of $q$ is computed from the measured L and E. The measurements are repeated over a grid of orientations: typically, $\beta$ is stepped through the full 180° range (e.g., in 45° increments), and for each $\beta$, the incidence angle γ is varied by changing r′ (including the cases tan$\mathsf{\gamma }$ = 0 and tan$\mathsf{\gamma }$ = 2, as required for specularity analysis). All data are compiled in tabular form, with columns for $\beta ,\mathsf{\gamma }$ (or tan$\mathsf{\gamma }$), illuminance E, luminance L, and the computed luminance coefficient q. From the tabulated $q (\beta ,\mathsf{\gamma })$ data, the average luminance coefficient (Q0) and the specularity factor (S1) are derived according to CIE road–surface conventions. The Q0 is defined as the solid-angle–weighted mean of $q$ over the relevant hemisphere of incidence. In practice, Q0 is computed by summing the measured q values multiplied by their solid-angle weights and normalising by the total solid angle (cf. Figure 9).

This metric represents the overall lightness of the surface. The specularity factor $S1$ quantifies how much of the reflection is concentrated in a specular component. By definition, $S1$ is the ratio of two reduced-luminance coefficients:

$$S1 = {\displaystyle \frac{r\left(\mathrm{0,2}\right)}{r \left(\mathrm{0,0}\right)}}$$

(2)

Here, $r1=r \left(\mathrm{0,0}\right)$ corresponds to the normal-incidence case, tan$\mathsf{\gamma }$ = 0), and $r2=r\left(\mathrm{0,2}\right)$ corresponds to the forward-skewed case (β = 0°, tan$\mathsf{\gamma }$ = 2). The factor cos³γ accounts for the inverse-square drop-off and obliquity of illumination. In the measurements, $r1=q (0,0)$ (since cos³γ = 1) and $r2=q (0,2){cos}^3\gamma$ with tan$\mathsf{\gamma }$ = 2; these are obtained directly from the table of $q$ by computing $q{cos}^3\gamma$ for the two cases. Finally, $S1$ is calculated as the ratio above. In addition, the values of Q0 assessed above were compared to the relationships after Corell and Sørensen [76]:

Q0 estimated = (0.957 × r1 + 0.746 × r2 + 104.5)/10,000

(3)

Qd estimated = (0.981 × r1 + 0.323 × r2+ 86.1)/10,000

(4)

Moreover, the above procedure follows the CIE “R-table” approach for characterising road pavements. In fact, CIE road–surface tables define representative $q (\beta ,\mathsf{\gamma })$ values for each pavement class (R1–R4), and classify surfaces by their Q0 and $S1$ parameters. Thus, the computed Q0 and $S1$ for the “white” material can be directly compared to these standard classifications.

Furthermore, it is also stated that reflection factors depend on material, surface texture, surface conditions, age of pavement, and angle of incidence (of the light). For a perfectly diffuse surface, the reflection factor is given by the product πL/E. The measurement of reflection factors essentially involves comparing two luminance values. In luminance measurement mode, the photometer is directed at the surface whose reflection factor is to be determined, and the luminance value is read from the display and recorded. Subsequently, a reference surface is placed over the same area, and the new luminance value is again read and recorded from the photometer. Furthermore, the reference surface—provided by Hagner—had a matte finish and a reported reflection factor of 0.956. Accordingly, the reflection factor of the tested surface can be calculated as follows:

$${\rho }_x={\displaystyle \frac{{\rho }_R{L}_x}{L_R}}$$

(5)

${\rho }_x$ is the reflection factor of the examined surface, ${\rho }_R$ is the reflection factor of the reference surface (in this case ${\rho }_R$ = 0.956), $L_x$ is the luminance of the examined surface (cd·m⁻²), and $L_R$ is the luminance of the reference surface (cd·m⁻² ).

3. Results and Analyses

3.1. Effect of Surface Treatments on Friction and Macrotexture

In this study, the Sand Patch Method, based on ASTM E965 [61] and EN 13036-1 [52] standards and developed by Praticò and Astolfi [53], was used to determine the MTD* values of mixtures with and without surface treatment. Additionally, the CoreLok apparatus was used to evaluate the relevant volumetric properties in accordance with AASHTO T-331 [77] and ASTM D6752-09 [78] standards. Figure 10 presents the specimens tested using the Sand Patch Method along with their corresponding MTD* values, as well as the skid resistance and the corresponding PTV* values. The ST1 mixture, representing the untreated reference, meets the minimum MTD* requirement (0.4 mm) specified in many national and international contract specifications. However, the application of white polymer alone (i.e., ST2) resulted in reduced surface roughness and, consequently, lower macrotexture performance. The drop in MTD* from 0.4 mm (untreated ST1) to 0.3 mm for ST2 (only polymer) can be explained by the polymer filling surface voids between aggregates, thereby smoothing the macrotexture. When a viscous white polymer is applied to the rough asphalt surface, it flows into the gaps and hollows of the surface texture. This partially fills the macrotexture voids, effectively reducing the average depth of surface asperities. In other words, the polymer acts like a thin binder film that levels off some of the surface roughness. MTD*, as measured by the sand patch method, is proportional to the volume of voids at the surface. By occupying a portion of those voids, the polymer reduces that volume, hence lowering the MTD* (a reduction in ~0.1 mm, or 25%, in this case). This reduction in macrotexture due to polymer filling has important implications, since an overly smooth surface can negatively impact skid resistance (particularly in wet conditions). A simplified quantitative model based on sphere packing and liquids can be proposed to illustrate this effect. Let V0 be the initial void volume per unit area A, corresponding to the untreated surface’s MTD (where V0/A is approximately 0.4 mm). After applying a polymer layer of volume Vp per unit area, the effective void volume is reduced to (V0 − φ·Vp), where φ represents the fraction of the polymer that infiltrates into the surface voids. The new texture depth can be estimated as MTDnew ≈ MTD0 · (1 − φ·Vp/V0). In practical terms, if the polymer seeps into and fills (for example) ~25% of the original surface void volume, the MTD would drop by roughly 25%, consistent with the observed change from 0.4 mm to 0.3 mm. The viscosity of the polymer plays a critical role in determining φ (the degree of void filling). A lower-viscosity polymer will penetrate more easily and deeply into small voids and crevices between aggregate particles, yielding a higher φ and thus a greater reduction in macrotexture depth (cf. [79,80]). Conversely, a higher-viscosity (thicker) polymer may remain on top of the aggregate peaks without fully entering narrow gaps, resulting in a lower macrotexture change [81]. This concept is analogous to the behaviour of asphalt binders in surface seals: an excess of low-viscosity binder can “bleed” into surface voids and significantly diminish macrotexture and skid resistance [79]. In summary, the application of the white polymer introduced a thin film that partially filled the aggregate gap structure of ST1, smoothing out the surface profile and reducing MTD*, with the extent of smoothing linked to the polymer’s flow characteristics (viscosity) and the amount applied.

In contrast, the combined use of white polymer and white powder (i.e., ST3) significantly enhanced the MTD* of the mixtures. Furthermore, the incorporation of additives such as glass fibre and glass aggregate influenced surface texture, with higher contents generally corresponding to increased MTD* values. Among these, the effect of glass aggregate on MTD* was observed to be more pronounced and consistent. These findings confirm that surface treatments—particularly those involving white powder, glass fibre, and glass aggregate—are effective in improving surface roughness compared to untreated mixtures. However, it should be noted that the higher MTD* values could lead to reduced driving comfort, increased tyre wear and fuel consumption, as well as higher noise levels.

In addition to macrotexture, evaluating the effect of surface treatments on skid resistance is essential to balance acoustic and safety requirements, as well as to comprehensively assess the effectiveness of the white surface treatment materials used. According to Figure 10, the ST1 mixture exhibited a high level of skid resistance, with a PTV* value falling within the limits conventionally specified by the contract specifications (PTV* ≥ 55). Supporting the MTD* results, the application of white polymer alone (ST2) significantly reduced skid resistance and is considered unsafe. Interestingly, although the ST3 mixture met the MTD* requirements, the inclusion of white powder slightly improved the surface performance compared to the white polymer in ST2. Yet, it remained quite unsatisfactory, insufficient in terms of skid resistance. On the other hand, the incorporation of glass fibre and glass aggregate in ST4–ST7 substantially increased the PTV* values.

3.2. Results of Photometric Measurements

3.2.1. Pilot Tests

In order to define reference photometric conditions, initial tests were carried out on different surfaces under varied lighting conditions. Illuminance and luminance levels were tested under controlled laboratory and dark room environments, as well as under general environmental conditions such as indoor lighting and outdoor sunlight, using different light sources. The comprehensive results, as shown in

Table 9. Photometric parameters under various experimental conditions.

Sample Condition	Illuminance (lux)	Luminance (cd·m−2)	Expected Luminance (cd·m−2)
Dark Room Conditions
White colour	7.25	0.21	5
White colour	90.75	0.86	5
Black colour	90.14	0.60	5
ST1	90.07	0.69	5
Laboratory Conditions
ST1	92	3.28	5
ST1	180	2.91	10
ST1	270	3.85	15
ST1	360	7.87	20
White colour	7.5	1.84	5
White colour	15	3.98	10
White colour	22.5	5.22	15
White colour	30	6.65	20
White colour	450	100	-
Reference Checks			Reflectance (%)
Calibration reflectance—daylight	502.2	117.8	73.69
Calibration reflectance—daylight	521.5	111.3	67.05
Calibration reflectance—dark lab	63.79	6.82	33.59
General Conditions
Desk with light	780	185.2	-
Desk without light	307	65.7	-
Outdoors (sunlight)	6680	-	-
ST1—outdoors (sunlight)	-	151.2	-
ST1—with light	-	21.5	-
ST1—without light	-	7.56	-
Desk with light (repeated measure)	819.6	176.2	-
ST1—with light (repeated)	853.3	20.85	-
Different Lens (f = 500)
ST1—with light	865.3	21.36	-
White colour—with light	860.4	200.7	-
Desk with light	821.5	156.7	-
Calibration reflectance—with light	843.5	190.5	70.95

, describe the measured and expected luminance values of reference surfaces and sample ST1, providing a foundation dataset for further analysis of surface treatment effectiveness. The aim of this section was to create reference lighting conditions necessary for the comparison of various surface treatments. Under dark room conditions and quite wide beams, all surfaces exhibited relatively low luminance values due to the limited illuminance. However, among the samples tested, the ST1-untreated surface demonstrated lower values than the pure white colour surface. This indicates that the surface treatment can contribute to light reflectance even under minimal lighting. Moreover, in laboratory conditions, a controlled increase in illuminance led to corresponding increases in luminance for both ST1-untreated and pure white colour surfaces. While white surfaces exhibited a steep rise in luminance with increasing light levels, the ST1-untreated samples showed a more gradual and stable luminance response. Reflectance measurements under various conditions confirmed that environmental lighting has a significant impact on optical readings.

Under daylight conditions, calibration panels exhibited reflectance values in the range of 67–74%, whereas these values dropped to nearly half under dark room conditions. This highlights the importance of performing photometric measurements under standardised lighting environments. Based on the experience gained during the measurements, it was determined that conducting measurements in a laboratory environment under office lighting and during daylight hours would be appropriate. Accordingly, the measurements were carried out under these conditions to evaluate the effects of surface treatment methods on photometric properties.

3.2.2. R-Classification Characteristics

The S1 and Q0 values of LCPMs with different surface treatments were calculated using Equations (2) and (3).

The results are presented in

Table 10. Q0 and S1 of asphalt mixtures subjected to surface treatments.

ID Sample	Sample Number	$\mathbf{Q}0 (\overline{\mathit{x}}$)	Q0 (σ)	95% CIs=	$\mathbf{S}1(\overline{\mathit{x}}$)	S1 (σ)	95% CIs=	Grade Standard (CIE)
ST1	2	0.0388	0.0065	0.0298–0.0478	0.7218	0.0280	0.6830–0.7606	R2
ST2	2	0.1884	0.0855	0.0699–0.3069	0.1916	0.1287	0.0132–0.3700	R1
ST3	2	0.2253	0.0651	0.1351–0.3155	0.1162	0.0192	0.0896–0.1428	R1
ST4	2	0.2265	0.0579	0.1463–0.3067	0.1548	0.0011	0.1533–0.1563	R1
ST5	2	0.2406	0.0778	0.1328–0.3484	0.1052	0.0433	0.0452–0.1652	R1
ST6	2	0.2182	0.0291	0.1779–0.2585	0.1407	0.0199	0.1131–0.1683	R1
ST7	2	0.2292	0.0144	0.2092–0.2492	0.1622	0.0442	0.1009–0.2235	R1

and Figure 11 in terms of the average for the given surface treatment, ST.

In Table 10, for each surface treatment, ST, Q0 and S1 values are given, including the leading statistical indicators (average, standard deviation, and confidence interval, 95% CIs=).

In Figure 11, the x-axis reports Q0, while the y-axis reports S1. A comparison of various Q0–S1 pairs reported in different studies and standards with the results of the seven types of ST mixtures tested in this study is presented.

According to the R-Classification by the CIE, mixtures ST1 fall into the R2 category based on their specular coefficient (between 0.42 and 0.85), while ST2, ST3, ST4, ST5, ST6, and ST7 are classified as R1 (<0.42). From a design standpoint, the distinction between the R1 and R2 classes is critical. As illustrated in Figure 11 and summarised in Table 10, the untreated surface (ST1) belongs to the R2 class, where higher specular reflection increases the risk of glare, particularly problematic in enclosed environments such as tunnels. Conversely, all treated surfaces (ST2–ST7) fall into the R1 class, characterised by diffuse reflection that reduces glare and provides a more uniform light distribution. Notably, several R1-classified mixtures (e.g., ST5) exhibited significantly higher Q0, indicating a strong potential for energy savings while simultaneously enhancing visual comfort. These findings suggest that, for tunnel applications, R1 surfaces with high Q0 values offer a preferable balance between energy efficiency and glare control, whereas R2 surfaces may require additional lighting adjustments to achieve the same performance.

On the other hand, all surface treatment materials contributed to an increase in the Q0 (ranging from 0.0388 to 0.2406). Among these, glass fibre and glass aggregate showed the most significant impact on improving Q0 compared to white powder and white polymer. Additionally, S1 values of mixtures containing glass aggregate (ST6 and ST7) increased. Given these results, it becomes essential to consider further how the particle size of glass aggregate influences both Q0 and S1.

In this context, it is necessary to highlight that the quartz component used is smaller than 4.75 mm (it is sand). This affects macrotexture and reflectance-related properties (Q0 and S1), yielding a finer surface texture and more numerous reflective facets, thereby increasing the total projected reflective area toward observers, together with its mineralogical composition [82].

Figure 11. Average Q0 and S1 compared to the literature and R-table. Note: Data adapted from Gidlund, et al. [83] and Li, et al. [84].

Figure 11. Average Q0 and S1 compared to the literature and R-table. Note: Data adapted from Gidlund, et al. [83] and Li, et al. [84].

In summarising, ST5 exhibited the highest Q0 and the lowest S1 values, while all surface treatments (except that for the zero case, ST1) were in the “diffuse area, where according to Figure 11 and Zhu, Li, Long, Zhou and Zhou [2], diffuse reflective pavement materials provide the most effective energy savings in terms of reflectance performance in road lighting.

3.2.3. Reflection Characteristics of Light-Coloured Asphalt Mixtures Under the Different Surface Treatments

For the effect of surface treatment methods on the reflectance factor and the reflection properties of light-coloured mixtures by projecting the reduced-luminance coefficient along the driving direction in spherical coordinates.

Table 11. Photometric parameters under various experimental conditions.

Sample Name	Luminance (cd·m−2)	Reflection Factor (RF)
Calibration reflectance—with light	190.5	0.956
Black colour—with light	9.76	0.05
ST1—with light	21.36	0.11
ST2—with light	50.02	0.25
ST3—with light	144.2	0.73
ST4—with light	192.0	0.97
ST5—with light	162.8	0.82
ST6—with light	184.5	0.93
ST7—with light	166.5	0.84

shows the calculated reflection factors of surfaces exposed to various surface treatments.

The data in Table 11 were measured under the following conditions: both the illumination intensity and the measurement geometry were kept constant during the measurement. Specifically, all measurements were conducted during the daytime under stable office lighting conditions. The distance of the photometer from the surface is fixed at 15 cm. An f = 500 lens (focal length = 500 mm) was used throughout the measurement. When the results in Table 11 are examined, it is noted that white-coloured surface treatment materials could increase the reflection factor of the untreated asphalt surface to the range of 0.25–0.97. Figure 12 presents a radar plot of the photometric properties of the mixtures along with macrotexture and surface friction variables, expressed as percentages. White powder and polymer materials provided a dramatic decrease in S1 value by 73% and 81% compared to black asphalt in samples ST2 and ST3, respectively. However, an increase in white polymer content alone led to a limited enhancement in the reflection factor. Additionally, it was observed that the white polymer caused a decrease in MTD* and PTV* values compared to black asphalt. Among the evaluated materials, white powder stood out as the most effective component in increasing the reflection factor and Q0. Regarding the glass-based materials, both glass fibre and glass aggregate significantly improved the reflection factor. However, when the glass fibre content exceeds 2.5%, a decrease in the reflection factor is observed, which may be attributed to the non-uniform distribution of fibres within the mixture.

For a more in-depth analysis, 2D projections of the R-table, representing standard Portland cement concrete and asphalt concrete surfaces, were derived onto the plane, as shown in Figure 13a,b. The objective was to demonstrate that the tests conducted on the pure black-coloured surface and the untreated asphalt sample (designated as ST1) comply with the UNI 10439 standard, and that light-coloured asphalt surfaces exhibit a higher reflection factor compared to the grey-toned Portland cement concrete surface.

Figure 14 shows 2D projection R-table values for samples with the surface treatments. Figure 14a–e enables a comparison of the reflection indicatrix visualising the directional light behaviour of asphalt surfaces for different surface treatment methods. It is noted that the vertical line (x = 0) corresponds to $\beta =90°$ (the relatively common situation where the light is on the left with respect to the driver direction), the extreme curve to the left usually corresponds to $\beta =180°$, the extreme curve to the right usually corresponds to $\beta =0°$(configuration not far from the condition of the driver in a tunnel), while the generic point of a curve is the same r value of the r-table mentioned above projected as indicated ($x=r·sin\gamma ·cos\beta ; z=r·cos\gamma$, where $r={10}^4·L·{\left(cos\gamma \right)}^3/E$). The maximum range on the y-axis (basically a distortion of L/E) ranges from about 350 (standard asphalt concrete, UNI 10439 [73], and ST1), to about 700 (standard Portland cement concrete, UNI 10439 [73]) to more than 750 (ST2–ST7). It should be noted that since ST1 represents traditional black asphalt, its scale is the same as asphalt concrete (see Figure 13b), while ST2–ST7 are visualised to be the same as Portland cement concrete (see Figure 13a) due to their light colour. However, the reflectance of light-coloured surface treatments, including sample ST2, is significantly higher than that of grey Portland cement concrete. Consequently, the measured values of the samples exceed the Portland cement concrete scale at certain points.

ST1 exhibits an asymmetric and irregular distribution. Stronger reflections are observed for x > 0 (β = 0–90°), whereas weaker reflections occur for x < 0 (β = 90–180°). This behaviour indicates that light is concentrated at specific angles, producing a zebra effect in which bright stripes appear only in certain directions due to the surface’s anisotropic reflection. The low Q0 value (0.04) further confirms the dark nature of the surface. A comparison of ST1 and ST2 projections shows that the maximum ordinate value of ST2 is nearly eight times higher, demonstrating the significant improvement in light reflectance achieved with the white polymer additive. The low S1 value of ST2 (0.1916) suggests a highly diffuse reflection, while its high Q0 (average 0.19) indicates enhanced brightness. Similarly, the addition of white powder in ST3 further increased the reflection capacity by raising the maximum ordinate value. The primary distinction between ST4 and ST5 is the additional 2.5% glass fibre content, which noticeably improves reflectivity. In terms of glass additions, the highest reflection capacities along the x-axis were observed in ST6 and ST7, both containing glass aggregate. Although a slight asymmetry is observed in the x > 0 region of ST7, it is noteworthy that glass aggregate can increase the S1 value while maintaining a relatively high Q0.

3.3. Evaluation and Ranking of Light-Coloured Pavement Mixtures

In this section, the following new method was set up and implemented to rank the LCPMs:

• Step 1. The indicators 1/S1, Q0, MTD, and PTV were selected as the most important in terms of energy and safety, with a “direction” towards benefits (the higher the indicator, the better). Indeed, a high specularity of pavement, especially when wet, can be dangerous due to reduced visibility and increased glare for drivers and pedestrians, potentially leading to accidents. This phenomenon is a significant safety concern, as it can create hazardous conditions on roads and pathways. Consequently, 1/S1 was selected as the indicator of concern.
• Step 2. For each indicator, Ii, a lower specification limit (LSLi) was selected, based on contract specifications and on the literature (2.38; 0.10; 0.4 mm; and 58, respectively).
• Step 3. For the given treatment STj, the differences Ii-LSLi were derived, assigning zero when Ii-LSLi < 0 (algorithm: if (Ii-LSLi > 0,Ii−LSLi,0))
• Step 4. For the given treatment STj, the feature scaling was applied to the differences Ii-LSLi (considered as a new feature, Ni). The normalisation of Ni = Ii-LSLi was carried out so that each feature contributes approximately proportionately to the final ranking. In more detail, rescaling was carried out using min-max normalisation to scale the range to [0, 1]. The algorithm used is: Ni’ = (Ni−MIN(Ni))/(MAX(Ni)−MIN(Ni))
• Step 5. For each treatment STj, the cumulative product, CPj, of the four new indicators above, Ni’ = (Ii−LSLi)’ was derived. The product of the sequence (CPj) enabled the clear separation of the solutions STj into two sets: unsatisfactory solutions (those with zero for one or more characteristics) and more or less satisfactory solutions (those with characteristics that comply with the specification limits set up in contract specifications or literature).
• Step 6. The ranking (for the second set of STj above) was derived based on the cumulative product CPj > 0, arranged from highest to lowest (cf.

  Table 12. Ranking of LCPMs.

  Surface Treatment	ST1	ST2	ST3	ST4	ST5	ST6	ST7
  Indicators, Ii								LSLi
  1/S1	1.39	5.22	8.60	6.46	9.50	7.11	6.16	2.38
  Q0	0.04	0.19	0.23	0.23	0.24	0.22	0.23	0.10
  MTD	0.39	0.34	1.02	0.93	1.08	1.02	0.90	0.40
  PTV	69.95	35.76	46.45	74.22	74.93	69.95	74.93	58
  New Indicators, Ni = Ii−LSLi > 0, 0
  Ref: 1/S1	-	2.84	6.22	4.08	7.12	4.72	3.78
  Ref: Q0	-	0.09	0.13	0.13	0.14	0.12	0.13
  Ref: MTD	-	-	0.62	0.53	0.68	0.62	0.50
  Ref: PTV	11.95	-	-	16.22	16.93	11.95	16.93
  Scaled Features (min–max normalisation), Ni’
  Ref: 1/S1	-	0.40	0.87	0.57	1.00	0.66	0.53
  Ref: Q0	-	0.63	0.89	0.90	1.00	0.84	0.92
  Ref: MTD	-	-	0.92	0.78	1.00	0.92	0.73
  Ref: PTV	0.71	-	-	0.96	1.00	0.71	1.00
  CPj	-	-	-	0.39	1.00	0.36	0.36
  Ranking	Excluded	Excluded	Excluded	2nd	1st	3rd	4th

  ).
• Step 7. A robustness check was carried out (cf.

  Table 13. Robustness check.

  Scenario	ST1	ST2	ST3	ST4	ST5	ST6	ST7
  All (see above)	Excluded	Excluded	Excluded	2nd	1st	3rd	4th
  S1 out	Excluded	Excluded	Excluded	2nd	1st	4th	3rd
  Q0 out	Excluded	Excluded	Excluded	2nd	1st	3rd	4th
  MTD out	Excluded	Excluded	Excluded	2nd	1st	4th	3rd
  PTV out	Excluded	Excluded	2nd	4th	1st	3rd	5th

  ), leaving one indicator out at a time. The robustness check was very satisfactory, resulting in ST5 always being the best, ST1 and ST2 always being excluded, and the remaining surface treatments having an intermediate ranking. This implies that conclusions remain consistent and there is a substantial stability of results, ensuring that findings are valid and reliable.
• Table 12 and Table 13 summarise the application of the method and the ranking obtained.

In more detail, according to the CPj results (cf. Table 12), the sequence ST5 > ST4 > ST6 > ST7 was obtained (where the 3rd and 4th ST are basically the same except for the third decimal). ST1, ST2, and ST3 were excluded from the ranking because they did not meet the expected minimum requirements. Among the suitable mixtures, ST5 stands out due to its highest Q0 value (on average 0.24), lowest S1 value (0.1052) and homogeneous distribution, indicating greater energy savings (particularly use in tunnels), and increased safety by reducing the glare risk, while also supporting traffic safety through favourable surface texture and friction.

On the one hand, the two best LCPMs, ST5 and ST4, are composed of the same materials (white polymer, white powder, and glass fibre), with the only difference being that ST5 contains twice the amount of glass fibre as ST4. As a result, ST5 exhibits a CPj value higher than ST4, highlighting the dominant effect of increasing the glass fibre content to 5% on the suitability of the mixtures in terms of photometric properties, macrotexture, and surface friction.

Apart from ST5, the distance between ST4, ST6, and ST7 is quite negligible, which calls for further investigation and study when comparing their characteristics. As shown in

Table 14. Various comparable aspects of ST applications.

Parameter/Aspect	Relevant International Standard	Application to ST4–ST7
Road surface reflection characteristics	CIE 144:2001 [64]—Road Surface and Road Marking Reflection Characteristics	ST4–ST7 can be benchmarked by measuring Q0 and S1, allowing classification relative to international reference surfaces.
Reference surface classes (dry conditions)	CIE 66:1984 [85]—Road Surfaces and Lighting (R1-R4 classes)	ST4–ST7 may correspond to higher-reflectance classes (closer to R1–R2) due to inclusion of reflective aggregates and fibres.
Reference surface classes (wet conditions)	CIE 47:1979 [86]—Road Lighting for Wet Conditions (W1–W4 classes)	Wet reflectance of ST4–ST7 can be evaluated against W-class standards to ensure no excessive specular glare in tunnels.
Roadway lighting performance requirements (luminance, uniformity, glare)	CIE 115:2010 [87] and EN 13201-2:2015 [69]	ST4–ST7 must enable compliance with required luminance and uniformity levels in road/tunnel lighting design.
Tunnel lighting design (zone-based luminance, uniformity, glare control)	CIE 88:2004 [65]—Guide for the Lighting of Road Tunnels and Underpasses	Light-coloured ST4–ST7 mixtures support energy-efficient tunnel lighting by increasing pavement luminance without excessive glare.
Measurement of diffuse luminance coefficient (Qd)	ASTM E2302-03a(2022) [88]—Test Method for Luminance Coefficient under Diffuse Illumination	Provides a standardised way to measure and verify the brightness of ST4–ST7 in line with international practice.

, these studies may include various aspects such as road surface reflection characteristics, reference surface classes (under dry and wet conditions), roadway lighting performance requirements, and tunnel lighting design.

Moreover, the process explained below was followed to calculate the potential energy savings of surface-treated designs with respect to ST1 (black asphalt), based on Q0.

(1)

The conditions considered in the study by Moretti Moretti, Cantisani, Di Mascio and Caro [11], which also incorporate Italian standards, were considered. In their studies, tunnel lighting is divided into two main components: permanent lighting and reinforcement lighting. Permanent lighting is applied along the entire tunnel length (0–750 m), whereas reinforcement lighting provides high-level illumination during daytime in the tunnel’s entrance section (0–577 m) and is organised into seven different reinforcement sections (zones). These zones correspond to functional areas, including the access zone, threshold zone, transition zone, interior zone, and exit zone. The zone distances and the design luminance values are presented in

Table 15. Design parameters defined by Moretti, Cantisani, Di Mascio and Caro [11].

Design Parameters Per Zone	Permanent	Zone 1	Zone 2	Zone 3	Zone 4	Zone 5	Zone 6	Zone 7
Reinforcement zone distance (m)	0–750	0–80	80–115	115–155	155–200	200–250	250–350	350–577
Design luminance (cd·m−2)	2	130	110	80	40	20	10	5

.

(2)

The basis of this study is the lumen method, which shows the relationship between the number of lamps needed in the tunnel and the Q0 values obtained with ST applications. In this context, the number of lamps required for each zone is calculated using Equations (6) and (7).

$${\mathrm{N}}_{lamp}={\displaystyle \frac{{\mathrm{E}}_{req}\times \mathrm{A}}{FL\times LLF\times Cu}}$$

(6)

$${\mathrm{E}}_{req}={\displaystyle \frac{Design luminance}{{Q0}_{ST\left(i\right)}}},\hspace{1em}\hspace{1em}(i=\mathrm{1,2},\mathrm{3,4},\mathrm{5,6},7)$$

(7)

where ${\mathrm{N}}_{lamp}$ is the number of lamps required for each zone, ${\mathrm{E}}_{req}$ is required illuminance level (lux, where one lux is one lumen per square metre), $\mathrm{A}$ is the area of the road zone (m²), FL is the luminous flux of the lamp (lumen), LLF is the maintenance factor (0.9), and Cu is the coefficient of utilisation (0.5). $\mathrm{A}$ is obtained by multiplying the zone distance by the 9 m road width. ${\mathrm{E}}_{req}$ is obtained by dividing the design luminance (cd/m²) value specified for each zone by Q0 (cd/m²/lux). These Q0 values are the average values given in Table 10 for each ST application.

The total number of lamps required for each zone is calculated as ${\mathrm{N}}_{total}$ and $P_{installed power}$ is calculated according to the power of lamp type ($P_{lamp})$ using Equation (8):

$$P_{installed power}={\mathrm{N}}_{total}\times P_{lamp}$$

(8)

$$P_{total}=P_{installed power}\mathrm{\ast }t\mathrm{\ast }UPF$$

(9)

The total power ($P_{total})$ value is calculated using Equation (9). In this equation, the annual usage period ($t$) of the lamp is taken as 8760 h per year while the average used power factor ($UPF$) is considered as 0.535 depending on external conditions.

$$Energy saving \left(\%\right)={\displaystyle \frac{P_{total}\left(ST1\right)-P_{total}\left(STi\right)}{P_{total}\left(ST1\right)}}\times 100$$

(10)

(3)

Consequently, in this study, design luminance values for each zone, the reinforcement zone distances and road width, light loss factor, coefficient of utilisation, power of lamp type, annual usage period and average used power factor for Equations (6)–(9) to be used to calculate the potential energy savings of all ST applications are fixed. However, the required number of lamps for each zone varies according to the Q0 values determined according to ST applications, and the potential energy savings are calculated using Equation (10) with respect to ST1 (reference).

Table 16. Potential energy savings (%) of ST applications.

Surface Treatment	Potential Energy Savings (%) (Rounded to the Nearest Integer)
ST1	-
ST2	79%
ST3	82%
ST4	82%
ST5	83%
ST6	82%
ST7	83%

shows the calculated potential energy savings (%) of ST applications.

According to Table 16, it was determined that the ST5 design provided the highest energy savings with a value of 83%, compared to the reference design. In addition, it is also anticipated that potential energy savings of approximately 80% can be achieved by reducing the number of lamps needed.

Even if Q0 (as well as the remaining parameters) could undergo degradations over time (with consequent need for maintenance), when these results of photometric properties of LCPMs are examined in the light of studies conducted in the literature from the perspective of energy saving, the expectations for the environmental benefits align with the broader literature.

This consistently highlights the energy-saving potential of light-coloured pavements, particularly in tunnel environments where lighting constitutes the dominant share of energy consumption. The primary key to this is that increasing pavement reflectance elevates the road surface luminance, reducing the need for intense or frequent lighting to meet visibility standards. For instance, high-albedo pavements have been reported to reduce tunnel lighting power requirements by approximately 25–40% compared to conventional dark asphalt. According to the European Asphalt Pavement Association (EAPA), light-coloured or clear pavements offer superior luminance and contrast, resulting in up to 40% savings in lighting energy use. Field and modelling studies further support these findings. Zhu, Li, Long, Zhou and Zhou [2] demonstrated that using a reflective stone aggregate pavement reduced per-kilometre lighting power by 25–29.2%, depending on the design luminance criteria. Salata, et al. [89] reported a 41.5% reduction in annual tunnel lighting energy use when switching to a high-reflectance cool asphalt mix. Moretti Moretti, Cantisani, Di Mascio and Caro [11] reported that light-coloured pavements could reduce the installed power needed to satisfy minimum lighting requirements by up to 30%. Their laboratory tests confirmed that clear pavements provide an average luminance six times greater than standard asphalt, and even after surface darkening due to traffic, they maintain roughly three times higher reflectance. These findings are consistent with the results of the present study, in which the Q0 value of ST5 was found to be six times higher than that of conventional black asphalt (ST1). Moreover, Bocci and Bocci [90] measured the photometric properties of a clear asphalt mixture in two Italian tunnels. They found that its reflectance and luminance were several times higher than traditional black asphalt, resulting in approximately 30% lower lighting power while still meeting illumination standards. Accordingly, it can be considered that a ST design with an average Q0 value of 4–6 times higher than the reference black surface can provide 80% energy savings.

As a result, considering the optimum lighting conditions specified in the study and the lighting conditions that maximise energy savings, this study showed that approximately 79–84% energy savings can be achieved in ST applications at the very beginning, in line with the literature. However, this expectation should be validated through supporting simulations and field studies in future research, particularly considering national standards such as UNI 11095:2011 [91] and international guidelines like CIE 88:2004 [65], which define minimum luminance and uniformity requirements for each tunnel zone. When specularity complies with no-glare conditions (cf. [84]), it is well established that high-albedo surfaces not only enhance energy efficiency but also support compliance with these regulatory standards by providing a higher baseline illuminance. In practice, this may allow designers to reduce lamp wattage or increase luminaire spacing without compromising safety.

4. Conclusions and Recommendations

Building on research into durable, long-lasting, and low-noise road pavements (as conducted in the LIFE SILENT project mentioned above), a maintenance scenario was envisaged in which the mechanical properties of the friction course remain satisfactory while only the surface performance requires maintenance. In this context, the potential of LCPA, designed with innovative white materials, to simultaneously enhance traffic safety and sustainability—particularly in road tunnel environments—was highlighted. Accordingly, this study aimed to evaluate the photometric performance and energy-saving potential of LCPMs in road lighting applications, as well as their effects on surface friction and macrotexture. Based on laboratory experiments and optical performance results, the following conclusions can be drawn:

(1)

The average luminance coefficient (Q0) of the pavement was increased through surface treatment applications by improving its reflectance capability.

(2)

The specular reflection coefficients of LCPMs containing varying types and amounts of white materials were found to decrease significantly compared to traditional black asphalt.

(3)

Surface treatments containing only white polymer were observed to reduce surface macrotexture and to lower microtexture performance.

(4)

Surface treatments incorporating white powder, glass fibre, and glass aggregate were shown to be more effective than white polymer in enhancing surface macrotexture relative to the reference mixture. Notably, the addition of glass materials was found to improve surface texture, friction, and Q0 value.

(5)

The reflectance factor of untreated asphalt surfaces was increased from 0.11 to 0.97 through surface treatments, indicating a significant improvement in light reflection. Surface treatments containing glass materials were particularly effective in enhancing the reflection factor. Conversely, the use of glass fibres in amounts exceeding 2.5% was observed to reduce reflection performance.

(6)

Based on the new method established and implemented to compare different surface treatments considering multiple characteristics, ST5, which exhibited the highest Q0 value (0.2406) and the lowest S1 value (0.1052), was selected as the optimal surface treatment design. ST5 represented the most favourable combination of energy efficiency and safety, alongside enhanced visual comfort. Among the remaining six surface treatments, three (ST4, ST6, and ST7) were determined to achieve a satisfactory balance of surface properties, while the other three (ST1, ST2, and ST3) failed to meet the selected specification thresholds.

(7)

Treated surfaces improved brightness with Q0 up to 0.24 cd·m⁻²·lx⁻¹ (ST5) (vs. 0.04 for the untreated samples), and reduced glare with S1 as low as 0.10, enabling 83% tunnel lighting energy savings.

(8)

Glass-enhanced treatments (ST4–ST7) achieved PTV 70–75 (vs. safety threshold ≥55) and MTD 0.9–1.1 mm (vs. minimum 0.4 mm).

(9)

ST5 (Q0 = 0.24, S1 = 0.10, PTV = 75, MTD = 1.1 mm) was the best performer, combining the highest brightness, the lowest glare, and excellent skid resistance/texture.

(10)

ST1 (Q0 = 0.04, PTV = 70, MTD = 0.4) and ST2 (Q0 = 0.08, PTV = 36, MTD = 0.3) failed to meet one or more safety/lighting requirements and are not recommended.

(11)

Fibre/mineral-reinforced coatings (e.g., ST5) are expected to be more durable, with literature showing 21–43% lower abrasion loss, but require tunnel field validation.

The findings of this study are expected to provide insights that will support future research on the evaluation of friction, macrotexture, and energy efficiency of LCPMs over time and under real-world conditions, particularly within the framework of life-cycle assessment and life-cycle cost analysis for environmental and sustainable optimisation throughout their service life. In future studies, investigating the macrotexture, surface friction, abrasion loss, and photometric properties of LCPMs, particularly under heavy traffic conditions, and evaluating the findings through life-cycle assessment and life-cycle cost analysis will facilitate the development of more durable surface treatments. In addition, the results may be further supported by simulation and field studies in addition to laboratory experiments, and an in-depth analysis of the effects of material chemical composition on reflection characteristics for traffic safety may be conducted. Moreover, pilot implementation of high-performance treatments such as ST5 in tunnels, supported by field trials, durability monitoring, and integration into tunnel lighting design standards, is recommended to ensure long-term safety and energy efficiency.