Comparative Study of ASTM C1202 and IBRACON/NT Build 492 Testing Methods for Assessing Chloride Ion Penetration in Concretes Using Different Types of Cement

Abstract

Durability is crucial for reinforced concrete, directly influencing the service life of structures. The presence of aggressive agents, especially chloride ions, significantly impacts durability. This study investigates the differences between ASTM C1202 and IBRACON/NT Build 492 standards in concrete containing various types of cement designed for a characteristic compressive strength of 40 MPa. Forty-eight cylindrical samples were prepared using eight types of Portland cement, including those with blast furnace slag, filler, and pozzolanic materials. Chloride migration tests were performed according to the ASTM C1202/2022 and IBRACON/NT Build 492/1999 methodologies. At a 95% confidence level, the results indicated that concrete made with filler-containing cement (PCII F-SR and PC II F) showed the poorest chloride resistance, with charge passing values exceeding 4000 coulombs (ASTM C1202) and diffusion coefficients above 10 × 10⁻¹² m²/s (IBRACON/NT Build 492). In contrast, concrete containing high slag cement (PC III-SR) and pozzolan cement (PC IV) demonstrated superior resistance to chloride penetration, with charge passing values below 1500 coulombs and diffusion coefficients under 5 × 10⁻¹² m²/s. Notably, discrepancies in classification were observed, as PC II Z (fly-ash based) and PC II E-SR (slag-based) received different ratings under the two methods. ASTM C1202 was found to be more stringent than NT Build 492, highlighting significant variations in the classification criteria between these standards. Based on the findings, new interval values are proposed for classifying concrete regarding the risk of chloride ion penetration, particularly for the ASTM C1202 standard, in order to better align with performance-based durability criteria and improve classification accuracy.

1. Introduction

The penetration of chloride ions is one of the most aggressive pathological manifestations in reinforced concrete structures, triggering a corrosive process in the reinforcements and reducing the service life of the elements. This issue is responsible for a significant socioeconomic impact, since a considerable amount of financial resources are allocated to the maintenance and repair of structures that have suffered severe deterioration caused by chloride corrosion (Kim and Bumadian [1]; Angst and Elsener [2]; Dey, Kumar and Phani Manoj [3]). These ions can be present in the environment and penetrate through the concrete pores. Therefore, producing concretes with lower porosity is an effective way to combat attacks from aggressive agents.

Concrete durability is affected by the penetration of aggressive agents into its pore network. Permeability directly influences durability because permeability is associated with the formation of pathways through the pores, and the greater the number of pores, the less resistant it tends to be. It is also worth noting that the water/cement ratio, addition of materials, and hydration of the products are factors that interfere with this permeability (Ali and Qureshi [4]).

Pathological manifestations in civil construction are sometimes due to failures in project formation and neglect in execution. The observation and analysis of the environment are important steps to be taken before concrete production because it may contain elements harmful to the integrity of the mechanical and chemical resistance properties of the structures to be built, compromising their durability (Ribeiro et al. [5]). Among these harmful elements, chlorides stand out as ionic compounds present in the atmosphere in different forms, which can reach the concrete from setting accelerators containing CaCl₂, aggregates, mixing water, seawater and a marine atmosphere, among others. They can manifest in up to three main forms in concrete: chemically bonded, free, and adsorbed on the surface (Isaia [6]).

Chlorides are widely recognized as a significant contributor to the pathological manifestations of reinforced concrete structures, particularly when exposed to aggressive environments such as coastal or industrial areas. They infiltrate the concrete matrix, reaching the reinforcement and initiating corrosion even at relatively low concentrations. By disrupting the passive oxide layer naturally formed on steel surfaces, chlorides accelerate corrosion rates and significantly weaken the tensile strength of concrete, which largely depends on the reinforcement bars. The extent of their impact is influenced by factors such as the concrete permeability, cement composition, pH, and chloride concentration. Additionally, their widespread presence, stemming from de-icing salts, marine environments, and industrial emissions, amplifies their detrimental effects on concrete durability (Angst et al. [7]; Sun et al. [8]; Balestra et al. [9]; Rissardi et al. [10]). This aggressive nature underscores the critical importance of implementing targeted durability criteria in standards for chloride-exposed environments. Therefore, evaluating chloride penetration in concrete structures is vital for improving the durability and lifespan of reinforced concrete elements. To achieve this goal, several test methods have emerged for assessing concrete’s resistance to chloride penetration.

Accelerated methods for evaluating chloride ion penetration in concrete, such as ASTM C1202 [11] and NT Build 492 [12], are commonly used to assess the durability of concrete in aggressive environments. These methods both involve the application of an electric field to drive chloride ions through the concrete, but they differ in the mechanisms and metrics used to measure resistance to chloride penetration.

In the ASTM C1202 method [11], a concrete sample is subjected to a direct current, with chloride ions migrating through the specimen between two electrolyte solutions: a sodium chloride solution on one side and a calcium hydroxide solution on the other. The amount of electric charge passing through the concrete is measured, with a lower charge indicating better resistance to chloride ion penetration. This method primarily evaluates the overall permeability of the concrete and provides a rapid assessment of its resistance to chloride ingress.

On the other hand, NT Build 492 [12] employs a slightly different approach by applying an electric field between a chloride solution on one side and a sodium hydroxide solution on the other, similar to ASTM C1202. However, instead of measuring the total charge passed through the concrete, this method directly measures the depth of chloride penetration after a specified duration. The chloride penetration depth is determined using a colorimetric technique, which helps correlate the results with the concrete’s durability. NT Build 492 is more focused on assessing the actual extent of chloride migration within the material and is often considered to be more reflective of real-world conditions, particularly for long-term chloride exposure. While the ASTM C1202 method suggests applying a constant potential difference of 60 V, NT Build 492 recommends adjusting the potential difference based on the initial measurement of the electrical current passing through the system. Although both methods aim to indirectly quantify concrete’s resistance to chloride presence, procedural variations may lead to divergent results across different concrete samples.

The service life and performance of concrete structures are influenced by the selection of the materials composing them (Chandramouli et al. [13]). In this context, the choice of cement in composite manufacturing holds significant importance. Consequently, there are numerous studies on the influence of the type of hydraulic binder used in concrete production that aim to observe resistance to chloride ion penetration, since these directly interfere with properties in both the fresh and hardened states. Different compositions imply different types of cement. Therefore, the material’s behavior under chloride attack will not be the same after hydration.

The integration of mineral admixtures, such as granulated blast furnace slag, pozzolanic materials, and limestone filler, profoundly influences the durability of concrete. The inclusion of these supplementary cementitious materials in concrete production not only lowers the cost of the final product, but also frequently enhances its handling, strength, and/or durability (Srivastava et al. [14]). The successful utilization of these materials in varying proportions has resulted in the development of different types of Portland cement, such as those with slag additions (PCII E, PCII-SR, and PC III-SR), filler additions (PC II F and PC II F-SR), and pozzolanic material additions (PC II Z, PC IV, and PC IV-SR).

According to Silveira [15], pozzolans and slags, due to their high silica content, interact with the calcium hydroxide released during clinker hydration, resulting in the formation of compounds with bonding properties, enhancing some physical and chemical characteristics of the concrete microstructure, such as reducing voids, increasing the mechanical strength at advanced ages, and enhancing resistance to chemical attacks, such as chloride attack. However, the influence of pozzolanic additions on concrete durability can be diverse due to variations in the individual chemical and mineralogical compositions of each material, resulting in different reaction mechanisms (Ribeiro [16]). On the other hand, limestone filler additions improve the compactness and workability of concrete and mortar, establishing a connection between hydration products (Alyousef et al. [17]).

In this context, this study aims to investigate the differences in chloride penetration assessment results according to ASTM C1202 [11] and IBRACON/NT Build 492 [5,12], using eight types of Portland cement (PC) with a characteristic strength of 32 MPa at 28 days: PC II E, PC II E-SR, PC II F, PC II F-SR, PC II Z, PC III-SR, PC IV, and PC IV-SR; the specifications of these cements are mentioned in NBR 16697 [18].

2. Materials and Methods

The research was conducted using a quantitative–qualitative, exploratory, and experimental methodology. The concrete mixtures were proportioned using the method of the Brazilian Portland Cement Association (ABCP), using fine sand, coarse aggregate, and a water-to-cement ratio of 0.33. This relationship was obtained using the Abrams’ Curve for an expected compressive strength of 40 MPa. The accelerated tests for chloride ion penetration were performed at 28 days according to the recommendations of ASTM C1202 [11] and IBRACON/NT Build 492 [5,12].

2.1. Materials Characterization

The materials used for producing concrete test specimens were Portland cement with a characteristic strength of 32 MPa (PC II E, PC II E-SR, PC II F, PC II F-SR, PC II Z, PC III-SR, PC IV and PC IV-SR), fine aggregate (quartz sand) and coarse aggregate (granite), and superplasticizer additive MC-PowerFlow 1180 from Bauchemie^®. The superplasticizer, composed of ether-based polycarboxylate polymers, was utilized to enhance the workability of the fresh concrete while reducing the water content. This was achieved by reducing the attractive forces between cement particles and aggregates, resulting in a more fluid mixture without the need for additional water (Ma et al. [19]). The technical specifications of the superplasticizer can be found in (MC-Bauchemie [20]). The technical specifications of the cements are detailed in the Brazilian standard NBR 16697 [18]. Information regarding the characteristics of the cements used is summarized in

Table 1. Composition details of cements based on Brazilian standard NBR 16697 [18].

Type of Cement	Information
PC II E	Addition of blast furnace slag at a content of 6–34%.
PC II E-SR	Addition of blast furnace slag at a content of 6–34% and low content of tricalcium aluminate (C3A).
PC II F	Addition of carbonate material (filler) at a content of 11–25%.
PC II F-SR	Addition of carbonate material (filler) at a content of 11–25% and low content of tricalcium aluminate (C3A).
PC II Z	Addition of pozzolanic material at a level of 6–14%.
PC III-SR	Addition of blast furnace slag at a content of 35–75%, also considered Sulfate Resistant (SR) as it has a low content of tricalcium aluminate (C3A) or because it has a slag percentage between 60 and 75%.
PC IV	Addition of pozzolanic material at a level of 15–50%.
PC IV-SR	Addition of pozzolanic material at a content of 15–50%, being Sulfate Resistant (SR) due to its low content of tricalcium aluminate (C3A) or even due to its percentage of pozzolanic material between 40 and 50%.

.

Table 2. Physical properties of cements, aggregates and superplasticizer.

Cement Type	Specific Mass (kg/m3)	Loose Unit Mass (kg/m3)	Compacted Unit Mass (kg/m3)	Fineness Modulus	Maximum Diameter (mm)	Moisture
PC II E	3000 a
PC II E-SR	2990 b
PC II F	2980 b
PC II F-SR	3000 c
PC II Z	3030 d
PC III-SR	2950 e
PC IV	3100 f
PC IV-SR	2960 a
Fine aggregate	2845	1639	1769	1.71	1.18	2.94
Coarse aggregate	2719	1367	1553	2.46	12.5	0.32
Superplasticizer	1090 *

Information about the specific mass of cements was provided by the suppliers: ^a Liz cement brand; ^b Nacional cement brand; ^c Mizu cement brand; ^d Montes Claros cement brand; ^e CSN cement brand; ^f Faciment cement brand. * Information about the specific mass of the superplasticizer was obtained from the technical sheet provided by the supplier.

presents the physical properties of the cements and aggregates.

2.2. Methods

2.2.1. Mix Design of Concretes

The dosing method employed was the ABCP method, derived from the dosing guidelines of the American Concrete Institute (ACI) [21] but tailored to suit Brazilian material conditions. Utilizing the Abrams’ Law curve, the ABCP method calculates the water-to-cement ratio, taking into consideration the required 28-day concrete strength and the typical strength of the cement utilized. The selection of concrete strength adhered to the guidelines outlined in NBR 6118 [22] for highly aggressive environments.

Reference concretes were proportioned with an estimated compressive strength of 40 MPa and a water/cement ratio below 0.45, as recommended by standard NBR 6118 [22], and cast according to NBR 5738 [23]. A superplasticizer was utilized to adjust the slump of the concretes, targeting 20 mm. The concrete was designated based on the type of cement used in its production.

Table 3. Proportion of materials per m³ of concrete. SP means superplasticizer.

Type of Concrete	Cement (kg/m3)	Fine Aggregate (kg/m3)	Coarse Aggregate (kg/m3)	Water (L/m3)	w/c Ratio	SP (L/m3)
PC II E	420.30	330.44	704.20	137.12	0.33	0.77
PC II E-SR	420.44	329.21	704.43	137.21	0.33	0.96
PC II F	420.58	327.98	704.66	137.29	0.33	1.16
PC II F-SR	420.30	330.44	704.20	137.12	0.33	0.58
PC II Z	419.89	334.06	703.51	136.87	0.33	0.39
PC III-SR	421.00	324.22	705.37	137.55	0.33	0.48
PC IV	418.97	342.21	701.98	136.30	0.33	0.77
PC IV-SR	420.86	325.48	705.13	137.47	0.33	0.58

outlines the mix designs established for the fabrication of the test specimens, with a consistent anticipated compressive strength across all mixes. It is noteworthy that the utilization of cement with varying specific masses resulted in different mixture proportions of materials.

2.2.2. Chloride Ion Migration Test from IBRACON/NT Build 492

The Nordtest NT Build 492 [12] is a widely adopted method for assessing the non-steady-state ionic migration of chlorides in concrete. This method involves measuring the penetration of chloride ions into concrete by determining the concentration profiles within the sample. To prepare the samples, cylindrical specimens measuring 10 cm in diameter and 20 cm in height were cut into four slices, each 5 cm in height. The end slices were discarded, and the central slices were utilized for testing, following the recommendations outlined in the IBRACON practice [5]. Tests were carried out in triplicate for the eight types of concrete, totaling 24 samples. Figure 1a depicts the cutting scheme and selection of slices from the specimen.

According to the Nordic standard, samples must undergo a conditioning process in a desiccator, where they are kept under a vacuum of 1 to 5 kPa for 3 h. Subsequently, a saturated solution of Ca(OH)₂ and deionized water is added to the samples, which are then maintained under vacuum for an additional 1 h. Following this, the samples should be immersed under atmospheric pressure in the same solution for 18 ± 2 h (NT Build 492 [12]. However, due to the unavailability of desiccator equipment, an alternative immersion procedure was implemented. The samples were submerged in a Ca(OH)₂ solution for at least three days to ensure full saturation. The saturation process was monitored by measuring the sample mass every 24 h. The samples were considered saturated when their mass remained constant, which was achieved by the second measurement. This procedure is also suggested in standard NT Build 443 [24] and was adopted by Guignone [25] and Silva [26] for chloride penetration tests in hardened concrete. Concrete pre-conditioning in lime water is carried out to prevent calcium leaching and maintain a high pH environment (Dehghan et al. [27]). This conditioning also ensures that chlorides are not absorbed through capillarity. The complete saturation of the concrete ensures that ionic migration becomes the dominant mechanism for chloride transport, closely simulating the conditions found in marine environments where structures are exposed. In contrast, if the concrete is dry or unsaturated, capillary absorption would prevail as the primary transport mechanism (Helene [28]; Guimarães [29]; Gjørv [30]).

Figure 1. Schematic illustrations of (a) sample cutting and selection and (b) apparatus for measuring chloride ion migration based on NT Build 492 method [5,12]. Adapted from Silva [26].

Figure 1. Schematic illustrations of (a) sample cutting and selection and (b) apparatus for measuring chloride ion migration based on NT Build 492 method [5,12]. Adapted from Silva [26].

A testing apparatus was assembled for conducting the experiment, where the samples were placed inside 100 mm PVC tubes and sealed with silicone to prevent leaks. Figure 1b illustrates the assembly scheme of the test. A solution of sodium hydroxide and distilled water (0.3 M NaOH) was added to the top of the tubes containing the samples, while the reservoir in contact with the bottom of the sample was filled with a 10% sodium chloride (NaCl) solution. Stainless steel rods were used as the anode and cathode, with the anode connected to the positive pole of the power supply and positioned in the anodic solution, while the cathode was connected to the negative pole of the power supply and positioned in the cathodic solution.

At first, the power supply was set to deliver a voltage of 30V. Subsequently, the current passing through the samples was measured using a multimeter, and the voltage was adjusted based on

Table 4. Voltage and duration of chloride migration test. Adapted from NT Build 492 and IBRACON [5,12].

Initial Current, I0 (With 30 Volts) (mA)	Applied Voltage, U (After Adjustment) (V)	Possible New Current (mA)	Duration of the Test, t (h)
${\mathrm{I}}_0< 5$	60	${\mathrm{I}}_0 < 5$	96
$5 \le { \mathrm{I}}_0 < 10$	60	$10 \le { \mathrm{I}}_0 <20$	48
$10 \le { \mathrm{I}}_{0 }<15$	60	$20 \le { \mathrm{I}}_0 <30$	24
$15 \le { \mathrm{I}}_0 <20$	50	$25 \le { \mathrm{I}}_0 <35$	24
$20 \le { \mathrm{I}}_0 <30$	40	$25 \le {\mathrm{I}}_0 <40$	24
$30 \le { \mathrm{I}}_0 <40$	35	$35 \le {\mathrm{I}}_0 <50$	24
$40 \le { \mathrm{I}}_0 <60$	30	$40 \le { \mathrm{I}}_0 <60$	24
$60 \le { \mathrm{I}}_0 <90$	25	$60 \le { \mathrm{I}}_0 <75$	24
$90 \le { \mathrm{I}}_0 <120$	20	$60 \le { \mathrm{I}}_0<80$	24
$120 \le {\mathrm{I}}_0 <180$	15	$60 \le {\mathrm{I}}_0<90$	24
$180 \le {\mathrm{I}}_0 <360$	10	$60 \le { \mathrm{I}}_0 <120$	24
${\mathrm{I}}_0 \ge 360$	10	${\mathrm{I}}_0 \ge 120$	6

to determine the most appropriate test duration (in hours) according to the initial current. Figure 2 illustrates the ongoing test following the application of the initial voltage U.

After completing the test period, the currents were measured once again. Subsequently, the samples were removed from the apparatus, superficially dried, and subjected to diametral compression testing using an EMIC manual hydraulic press PCM 100C to fracture the specimens and expose an internal surface. Immediately following fracture, a solution of silver nitrate (0.1 M AgNO₃) was applied to the internal surface of the samples. Using a precise caliper with 0.1 mm accuracy, the depth of silver chloride (AgCl) precipitation was measured at 10 mm intervals, corresponding to the accelerated penetration of chloride ions (Xdi), approximately 15 min after solution application. To mitigate blockages caused by large aggregates and edge effects, measurements were taken only in areas located at least 10 mm from the edge of the specimen. Between five and seven measurements were recorded. Figure 3 provides a visual representation of the chloride penetration depth measurements.

With the X_di penetration data, the average X_d for each sample was calculated. Subsequently, the non-steady-state diffusion coefficient D_ns (×10⁻¹² m²/s) was determined using Equation (1). This equation, derived from Fick’s Second Law, describes the diffusion process of chloride ions through the concrete under the influence of an electric field. It establishes a relationship between the chloride concentration gradient and the rate of chloride migration, allowing for the quantification of ion movement in the material. The equation assumes a steady-state condition, where the chloride ion flux is proportional to the concentration gradient across the concrete surface (Tang, Nilsson and Basheer [31]).

$${\mathrm{D}}_{\mathrm{ns}}={\displaystyle \frac{0.0239\times (273+\mathrm{T})\times \mathrm{L}}{\left(\mathrm{U}-2\right)\times \mathrm{t}}}\times \left({\mathrm{X}}_{\mathrm{d}}-0.0238\sqrt{{\displaystyle \frac{\left(273+\mathrm{T}\right)\times \mathrm{L}\times {\mathrm{X}}_{\mathrm{d}}}{\mathrm{U}-2}}}\right)$$

(1)

where T denotes the mean of the initial and final absolute temperatures of the analyte (°C); U stands for the electrical potential difference (V); t represents the exposure time (h); X_d signifies the average value of the penetration depth (mm); and L indicates the thickness of the specimen (mm).

The NT Build 492 standard does not provide a classification for chloride ion penetration. Thus, Nilsson, Ngo, and Gjørv [32] proposed reference values and a classification based on the chloride migration coefficient data for concrete at 28 days (Dns₂₈). These values are displayed in

Table 5. Classification of chloride penetration through the non-steady-state migration coefficient proposed by [32], based on NT Build 492 [12].

Dns28 × 10−12 m2/s	Permeability of Chloride Ions
>15	High
10–15	Moderate
5–10	Low
2.5–5	Very low
<2.5	Negligible

.

2.2.3. Chloride Ion Migration Test from ASTM C1202

In this method, the preparation of samples, the alternative conditioning process, the insertion of samples into PVC tubes sealed with silicone, the preparation of the anodic solution, the positioning of the solutions in the reservoir and apparatus, and the connection of the electrodes to the power source followed the same procedure as outlined earlier for the NT Build 492 method. However, in the ASTM C1202 method, there is a change in the concentration of the cathodic solution, utilizing 3% sodium chloride (NaCl). The power supply held a steady voltage of 60.0 ± 0.1 V for the entire 6 h duration of the test. The current flowing through the specimens was monitored every 30 min using a multimeter. After the test concluded, the total charge passing (Q, in Coulombs) through was computed using Equation (2), with I₀ representing the current measured immediately after the voltage was applied, and I_t representing the current measured at times t (in minutes).

$$\mathrm{Q}=900 \times \left({\mathrm{I}}_0+2·{\mathrm{I}}_{30}+2·{\mathrm{I}}_{60}+...+2·{\mathrm{I}}_{330}+{\mathrm{I}}_{360}\right)$$

(2)

With the total charge passing (Q), the concrete was classified according to the risk of chloride ion penetration, following

Table 6. Classification of chloride penetration through the non-steady-state migration coefficient based on ASTM C1202 [11].

Total Charge Passing	Classification of Risk for Chloride Penetration
>4000	High
2000–4000	Moderate
1000–2000	Low
100–1000	Very low
<100	Negligible

.

2.3. Statistical Analysis

The data were analyzed using a completely randomized experimental design. An analysis of variance was employed to assess significant differences between treatments. Subsequently, upon detecting such differences, the Scott–Knott multiple mean comparison test was utilized. All assumptions underlying the analysis of variance, including normality and the homoscedasticity of residuals, were thoroughly checked. The significance level of the research was set at 5%. For data analysis, the R software, version 4.3.0 (R Core Team [33]) in conjunction with the ScottKnott package, version 1.3-0 (Jelihovschi et al. [34]), was employed.

3. Results

3.1. Resistance to Chloride Penetration—IBRACON/NT Build 492

Figure 4 presents measurements of the chloride penetration depth of the concretes. In Figure 4a, a comparison is made between cements containing slag. It is evident that chloride penetration is significantly lower in concrete PC II E compared to PC II E-SR, primarily due to the higher presence of C₃A, which aids in reducing chloride penetration. Chloride ions chemically bound to tricalcium aluminates (C₃As) form calcium chloroaluminate (Friedel’s salt), reducing the overall amount of free chlorides (Ribeiro et al. [5]). This indicates that the higher the C₃A concentration, the greater the capacity to bind free chlorides. Similarly, the lower chloride penetration in concrete PC III-SR compared to PC II E-SR suggests that the higher amount of slag also contributes to reducing chloride penetration.

This trend is also observed in cements containing pozzolanic materials (Figure 4b). However, the difference between concrete PC IV and PC IV-SR is less pronounced. Notably, when comparing PC IV and PC II Z concretes, the difference is much more significant, as the PC IV samples, which contain a higher amount of pozzolana in their composition, showed better performance. This indicates that the impact of the amount of pozzolanic material is greater than the difference in C₃A concentration between these concretes. This phenomenon occurs because the increased alumina content in fly ash enables the chemical binding of chloride ions, leading to a reduction in free chlorides within the concrete. Additionally, the pozzolanic reaction with calcium hydroxide further enhances the concrete’s ability to impede the transportation of chlorides (Liu et al. [35]; Moffatt, Thomas and Fahim [36]; Saldanha, Reddy and Consoli [37]; Hu et al. [38]; Xie et al. [39]; Medeiros-Junior et al. [40]).

In concretes containing only limestone filler (Figure 4c), the average chloride penetration is slightly lower in concrete PC II F-SR compared to PC II F, although the difference is not statistically significant given the error margin (see Figure 4d). Figure 4d provides a clearer comparison of the chloride penetration resistance across all concretes based on the average depth values. It shows that concretes PC II E, PC IV, and PC IV-SR demonstrate superior performance, whereas concretes with limestone filler exhibit poorer resistance.

Figure 5 shows statistical analyses along with the risk classification obtained from the non-steady-state chloride migration coefficients for each type of concrete. It was found that the concretes exhibiting the highest average resistance to chlorides, PC III-SR and PC-IV, are similar to each other, with D_ns equal to 1.37 × 10⁻¹² m²/s and 2.20 × 10⁻¹² m²/s, respectively, falling into the category of negligible chloride penetrability.

Other concretes demonstrating good resistance to chloride penetration and showing similarity were those produced with PC IV-SR and PC II E cements, with corresponding D_ns values of 3.30 × 10⁻¹² m²/s and 4.07 × 10⁻¹² m²/s, respectively, indicating very low chloride ion penetrability.

Concretes classified as having a low risk of chloride penetration were those produced with PC II E-SR and PC II Z cements, with Dns coefficients corresponding to 6.76 × 10⁻¹² m²/s and 8.91 × 10⁻¹² m²/s, respectively.

None of the studied concretes were classified as having a moderate risk of penetrability. However, those with PC II F and PC II F-SR demonstrated a high risk of chloride penetrability, indicating limited efficiency in resisting chloride penetration and therefore not being recommended for aggressive environments. These concretes exhibited similar behavior, with D_ns values of 15.11 × 10⁻¹² m²/s and 15.39 × 10⁻¹² m²/s, respectively.

3.2. Resistance to Chloride Penetration—ASTM C1202

Figure 6 shows a comparison of the results using statistical analysis, along with the risk classification, for the mean total charge passing of the various types of concrete. Concretes made with PC II F and PC II F-SR cements are similar to each other, both showing low efficiency in resisting chloride penetration. They exhibited total passing charges of 4455 C and 4716 C, respectively, placing them in the category of high chloride penetrability.

Concretes with PC IV and PC III-SR showed similar mean total passing charges, at 214 C and 376 C, respectively, classifying them as having a very low risk of chloride penetrability, similar to concrete with PC IV-SR, with a passing charge of 906 C.

Additionally, concretes with PC IV-SR and PC II E had similar mean total passing charges but differed in the classification of penetrability risk. Concretes with PC II E were classified as having low risk, with a passing charge equivalent to 1276 C.

Furthermore, concretes with PC II E-SR and PC II Z also exhibited similar mean total passing charges, categorized as moderately penetrable by chlorides, with charges of 2601 C and 2751 C, respectively.

Table 7. Summary of concrete classification according to ASTM C1202 and IBRACON/NT Build 492 methods.

Type of Concrete	Classification
	ASTM C1202 (Charge Passed)	IBRACON/NT Build 492 (Dns)
PC III-SR	Very low	Negligible
PC IV	Very low	Negligible
PC IV-SR	Very low	Very low
PC II E	Low	Very low
PC II E-SR	Moderate	Low
PC II Z	Moderate	Low
PC II F	High	High
PC II F-SR	High	High

summarizes the classification of concretes studied using both methods. As shown, there are divergences concerning the risk classification of chloride penetration using both methodologies.

The concretes produced with PC II F and PC II F-SR cements showed statistical similarities and the same risk classification for chloride ion penetrability across both test methods. This indicates that the lower C₃A content in PC II F-SR cement, although resulting in slightly lower values, was not significant enough to produce a statistical distinction. Fillers, being inert materials, can dilute the quantity of hydrated products, such as the C-S-H, which is critical for the strength and impermeability of concrete. This dilution may reduce the concrete’s resistance to chloride penetration.

For concretes incorporating pozzolanic materials, evaluated using the IBRACON/NT Build 492 method, significant statistical differences and distinct risk classifications for chloride penetrability were observed among the PC II Z, PC IV, and PC IV-SR concretes. Among these concretes, the PC II Z exhibited the poorest performance, attributed to its lower pozzolanic content, which reduces the availability of silica for portlandite conversion into C-S-H, thereby compromising the adsorption of free chlorides. Between the PC IV and PC IV-SR concretes, the superior performance of PC IV may be attributed to its higher C₃A content or differences in the amount of minerals added. These variations could stem from differences in cement manufacturing (see Table 1). In tests conducted according to ASTM C1202, PC IV and PC IV-SR concretes were classified as having the same chloride penetration risk. However, all concretes with pozzolanic additions showed statistical differences, reinforcing the trends observed with the IBRACON/NT Build 492 method.

For concretes containing slag additions (PC II E, PC II E-SR, and PC III-SR), the results from the IBRACON/NT Build 492 method revealed divergent risk classifications and significant statistical differences among the cements. The poorest performance was observed in concrete made with PC II E-SR cement, probably due to the lower C₃A content. The best performance was recorded for PC III-SR concrete, which contains a high slag addition, typically exceeding 60%; this is required for providing enhanced sulfate resistance (SR). While slag is less silica-rich compared to pozzolanic materials, its high addition percentage contributes significantly to the availability of the silica necessary for forming the C-S-H gel, improving the chloride penetration resistance.

4. Discussion

4.1. Discussion on the Relationship Between Type of Cement and Resistance to Chloride Penetration

The literature is comprehensive regarding the influence of different types of cement on chloride penetration resistance. Consistent with this study, Rodrigues [41] demonstrated that concretes with PC II F exhibit low resistance to chloride penetration, while those with PC IV-SR show high resistance. These results also align with the findings of Imai, Demeterko, and Aikin [42], where compared to PC II F concrete, PC IV concrete exhibited up to 73% higher chloride penetration resistance, while PC II Z concrete showed 49% higher chloride resistance. According to these authors, this difference in resistance is attributed to the reduction in the concrete porosity caused by the pozzolanic effect; thus, chloride ions tend to find penetration more difficult.

Similarly, Pereira [43] assessed the influence of cement type on the chloride migration coefficient and observed that PC IV cement exhibited lower values compared to PC II F. The PC IV cement also demonstrated better performance in studies conducted by Tavares [44], Crauss [45], and Marques and Ribeiro [46], which compared it with various types of hydraulic binders. According to Helene et al. [47], concrete produced with PC III-SR cement with a water-to-cement ratio ranging from 0.45 to 0.65 exhibited chloride penetration classified as very low according to the ASTM C1202 method, consistent with the results found in this study.

Ribeiro et al. [48] utilized six types of Portland cement in their studies (PC II E, PC II F, PC II Z, PC II Z-SR, PC IV, PC V ARI-SR), where they stated that the water/cement ratio, porosity, fineness of the binder, as well as the total alumina content are significant factors in chloride diffusion. Among the cements used, PC IV had the highest alumina percentage, approximately 9.1%, while the lowest total alumina content was 3.4% for PC II F. Thus, for a water/cement ratio of 0.50, the concrete with PC IV exhibited a chloride migration coefficient in the non-steady state equivalent to 3.23 × 10⁻¹² m²/s, which corresponds to 61% of the value of the coefficient for PC II F concrete. It is also noteworthy that the concretes with PC II Z and PC II E displayed migration coefficients classifying them as low penetrability. Despite Ribeiro’s studies [48] not employing the same water/cement ratio as this research, there is a noticeable trend in the performance of concrete depending on the type of cement used. Concretes produced with PC IV cement, characterized by higher alumina content, and the specific type of mineral additives directly influence improved particle packing, which reduces the porosity and permeability of the cementitious matrix composites.

Concerning the influence of limestone filler, numerous studies have shown that this supplementary material can undergo reactions with calcium aluminate hydrates, resulting in the formation of a stable AFm phase, specifically calcium monocarboaluminate. These reactions contribute to an increased total solid volume and reduced overall porosity (Matschei, Lothenbach and Glasser [49]; Celik et al. [50]; Pelletier-Chaignat [51]; Lothenbach [52]). However, the impact of limestone filler on the chloride ion diffusion coefficient remains a topic of debate in the literature. While some studies suggest that the addition of limestone powder can slightly reduce the chloride ion diffusion coefficient by influencing matrix tortuosity (Hornain et al. [53]), others have found contradictory results. For instance, Wang et al. [54] discovered that concrete containing limestone powder exhibited higher long-term chloride penetration compared to plain cement concrete. This discrepancy is attributed to the coarser pore structure and reduced production of C-S-H, resulting in a decreased chloride-binding capacity.

Sun et al. [55] demonstrated through the ASTM C1202 method that the resistance to chloride penetration is influenced by the content of limestone filler added. They found that the resistance to chloride ion penetration remained relatively similar to that of plain concretes at both 28 and 90 days with limestone filler additions of up to 8%. However, as the additions increased to 16% and 24%, there was a notable increase in the charge passed, indicating a decrease in resistance to chloride ion penetration.

While this study did not delve deeply into the primary impact of fillers and the mechanisms involved, focusing instead on the comparison between rapid chloride migration tests, the literature is abundant on this topic and can help explain the phenomena observed. It is well established that C₃A plays a significant role in chloride penetration resistance. The interaction between these minerals and chloride ions often results in the formation of calcium chloroaluminates, such as Friedel’s salt (Ca₄Al₂(OH)₁₂C_l2·4H₂O) and Kuzel’s salt (Ca₄Al₂(OH)₁₂Cl(SO₄)_0.5·5H₂O) (Thomas et al. [56]). This reaction primarily occurs through the interaction of chloride ions with AFm hydrates (Birnin-Yauri and Glasser [57]), with C₃A as the main precursor mineral. The formation of calcium chloroaluminates reduces the concentration of free Cl⁻ ions in the pore solution, thereby decreasing chloride penetrability (Birnin-Yauri and Glasser [57]; Chang et al. [58]). Consequently, sulfate-resistant cements, which typically have lower levels of C₃A, tend to exhibit greater chloride penetration compared to non-sulfate-resistant cements.

It is also known that filler-based cements tend to mitigate the formation of AFm phases by forming carboaluminate hydrates instead (Lothenbach et al. [52]; Péra, Husson and Guilhot [59]). This occurs because the formation of calcium carboaluminate hydrates is preferential due to their lower standard Gibbs free energy during formation (Lothenbach; Winnefeld [60]). With fewer AFm phases, the formation of Friedel’s salt, which binds chlorides, is compromised. Even though carboaluminates can be converted into Friedel’s salt in chloride-rich environments, as reported in (Shi et al. [61]), carboaluminates are relatively stable phases. The exchange reaction required to form Friedel’s salt involves more steps and is less thermodynamically favorable compared to the direct formation reaction using monosulfate or other aluminate phases. This stability means that, although the reaction can proceed, it is not as efficient as the direct formation of Friedel’s salt. Since carboaluminates are less effective at binding chloride ions, more chloride ions remain free in the pore solution, leading to higher chloride ion mobility and an increased risk of reinforcement corrosion. This might be related to the underperformance of filler-based cements.

It can also be evidenced that other supplementary materials, such as slag or pozzolans, play a more substantial role in enhancing concrete’s resistance to chloride penetration. This is because C-S-H can also bind with chloride ions and becomes more influential as the concentration of chloride ions increases.

According to Chang et al. [58], Thomas et al. [56] and Chang et al. [62], Friedel’s salt is the primary chloride-binding phase at low chloride concentrations. However, as the concentration increases, C-S-H takes over as the main chloride-binding phase. At low chloride concentrations, the formation of Friedel’s salt is favored due to its high affinity for chloride ions and the availability of aluminate phases. Conversely, at high chloride concentrations, once the aluminate phases are saturated, C-S-H takes over due to its larger capacity for chloride binding through surface adsorption and structural incorporation. This behavior explains the superior performance of pozzolanic and slag-based cements in chloride-rich environments, as well as the inferior performance of filler-based cements due to the dilution of cement.

In real-world scenarios, chloride-induced deterioration in concrete structures typically takes decades to manifest. This delayed effect is mainly due to the lower chloride concentrations and, of course, the absence of the electrical charge used in accelerated tests to stimulate ion penetration. Consequently, the performance of concrete with different types of cement may differ between accelerated tests and real-world conditions, especially under low chloride concentrations. In such cases, concrete made with pozzolanic cement could perform similarly to concrete made with a cement that has a high C₃A content but lower levels of supplementary materials.

Finally, it is important to emphasize the effect of the sulfate content on cement. The sulfate content significantly influences the formation of AFt and AFm phases, particularly ettringite and monosulfate, respectively. Ettringite forms when sulfate is sufficiently available, while monosulfate forms when the sulfate content is insufficient to fully react with the C₃A phase. The ettringite formed in these systems is chemically stable, as chloride ions do not react with it. In contrast, monosulfate is the primary precursor for the formation of Friedel’s salt, which effectively binds chlorides. The absence of monosulfate reduces the system’s ability to chemically bind chloride anions, with chloride likely being adsorbed into the C-S-H phase instead. Therefore, the binding of chlorides to the hydrated calcium aluminate or calcium silicate phases in cement also depends on the sulfate content (Paul et al. [63]; Bertola et al. [64]).

4.2. Discussion About Testing Methods

To compare chloride resistance results using the methodologies outlined in ASTM C1202 and NT Build 492, concretes were prepared with the same initial characteristic strength but using different types of cement for accelerated chloride ion penetration tests.

Throughout the testing process, it became apparent that despite both methodologies having similar sample preparation procedures, they each possess unique characteristics that can impact the final outcome. As noted by Cascudo [65], ASTM C1202 employs an excessively high voltage, regardless of the sample’s initial current, leading to a rise in temperature and potentially causing the solution to boil.

According to Menna Junior et al. [66], temperature fluctuations represent a variable that can obscure the true performance of concrete in the ASTM C1202 charge passing method. The authors noted a direct correlation between the maximum temperature and the charge passed in each migration cell, regardless of the cement type used. The highest charge passed occurred when the temperature peaked during the test. Furthermore, the increased movement kinetics of ions in concretes with a high water-to-binder (w/b) ratio contributed to temperature elevation. This phenomenon was primarily attributed to opposite polarities in the solutions, which enhance ion conductivity and result in a rise in temperature within the solution. This limitation underscores the importance of considering temperature effects in this test.

Additionally, it is important to emphasize that the stability of calcium aluminum monochloride hydrate, commonly known as Friedel’s salt, is also influenced by temperature (Renaudin et al. [67]; Balonis [68]; Suryavanshi, and Narayan Swamy [69]). Hino Junior et al. [70] observed that the chemical reaction leading to the formation of Friedel’s salt, as identified by NT Build 492, was not detected in the accelerated C1202 migration procedure. This discrepancy may be attributed to the shorter test duration and elevated temperatures in the accelerated procedure, which hinder the formation of Friedel’s salt.

In the IBRACON/NT Build 492 method, temperature checks are required both at the beginning and end of the test. Additionally, the method adjusts the applied voltage to account for the concrete’s quality. In contrast, the ASTM C1202 method maintains a constant initial load throughout the test. At the start of the IBRACON/NT Build 492 test, various samples exhibited significantly different initial currents (voltage Ui = 30 V), ranging from 6.8 mA to 84.6 mA. After voltage adjustments, seven sample groups were established with 24 h test durations, whereas concretes with PC III-SR cement necessitated a 48 h test period. Conversely, the ASTM C1202 test recorded currents immediately following the application of a 60 V voltage, ranging from 6.94 mA to 187.30 mA. Higher initial currents generally indicate a swifter migration process and lower resistance to chloride penetration.

Examining Equation (2) reveals that it solely relies on measured currents to determine the total passing charge, a parameter used to classify concrete with respect to the chloride penetrability risk. Consequently, it was noted that, to classify concrete as having negligible chloride penetrability, the average current must be below 4.63 mA, significantly complicating concrete classification within this category.

When comparing the results from both classification methods with those of other researchers, similarities emerged with studies conducted by Fedumenti [71]. The authors utilized PC II F cement to produce C40 class concrete, yielding a chloride migration coefficient of 30.90 × 10⁻¹² m²/s, classifying it as highly penetrable to chlorides via NT Build 492. This classification aligns with the ASTM C1202 method, which resulted in a total passing charge equivalent to 4930 coulombs.

Moreover, Hino Junior et al. [70] found that C45 class concretes (with a compressive strength of 45 MPa) containing PC II Z cement exhibited a total passing charge of 3658.5 C, categorized as moderately penetrable. Conversely, concrete with PC IV cement displayed a total passing charge of 1272 C, classified as low risk. This concrete also exhibited a non-steady-state migration coefficient equivalent to 5.8 × 10⁻¹² m²/s for PC II Z, indicating a low risk of chloride penetrability. Conversely, PC IV concrete showed a migration coefficient equivalent to 1.67 × 10⁻¹² m²/s, classifying it as having a negligible probability of chloride ion penetration. It is noteworthy that the same concrete may not consistently maintain the same classification across both test methodologies concerning the chloride penetration risk. Furthermore, it is important to highlight that the ASTM C1202 standard is more stringent compared to NT Build 492.

Once again, concrete produced with PC IV cement demonstrated satisfactory results in Wally’s [72] study. The probability of chloride penetration was classified as very low through the NT Build 492 method, with a migration coefficient equivalent to 2.84 × 10⁻¹² m²/s, showing a result very close to the classification of negligible chloride penetrability. Similarly, through the ASTM C1202 method, this same concrete achieved a passing charge of 859.44 C, thus being classified as having a very low risk of chloride ion penetrability. This indicates a convergence with the results of this research using the ASTM C1202 method.

Figure 7 shows the average Dns and charge passed for all concrete samples, along with the correlation between both techniques. As depicted in Figure 7a, the Dns values and the charge passed exhibit similar trends among the concretes, with a high correlation between the methods. However, discrepancies in the risk classifications are evident, as previously highlighted in Table 7.

In view of the discussions mentioned earlier, it is evident that the ASTM C1202 method considers fewer variables in classifying the risk of chloride penetrability compared to IBRACON/NT Build 492. Moreover, it applies a significantly high voltage value without factoring in the initial quality of the concrete under study (initial current). This, coupled with the stringent classification intervals, as previously highlighted, poses challenges.

In response, Hino Junior et al. [70] proposed a new classification by adjusting the interval range for the non-stationary chloride diffusion coefficient based on NT Build 492. In contrast, considering the evaluation that the outcomes derived from NT Build 492 are more equitable, a novel approach has been recommended. This involves modifying the interval range of the passing charge for the ASTM C1202 standard, as illustrated in

Table 8. Proposed classification regarding the risk of the penetration of chloride ions based on the passing charge from ASTM C1202.

Original Charge Passing (C)	Adjusted Charge Passing (C)	Risk of Chloride Ion Penetration
>4000	>4500	High
2000–4000	3000–4500	Moderate
1000–2000	1500–3000	Low
100–1000	750–1500	Very low
<100	<750	Negligible

. The new values were based on the linear fitting curve in Figure 7b.

Table 9. Summary of concrete classification according to the ASTM C1202 and IBRACON/NT Build 492 methods with adjustments to the classification intervals of the ASTM C1202 method.

Type of Concrete	Risk Classification Regarding Chloride Penetration
	ASTM C1202 (Charge Passed)	IBRACON/NT Build 492 (Dns)
PC II E	Very low	Very low
PC II E-SR	Low	Low
PC II F	Moderate	High
PC II F-SR	High	High
PC II Z	Low	Low
PC III-SR	Negligible	Negligible
PC IV	Negligible	Negligible
PC IV-SR	Very low	Very low

delineates the risk classifications for chloride penetrability using the updated passing charge values.

Based on the classifications outlined in Table 9, it is evident that seven out of the eight types of concrete studied achieved consistent classifications across both methods following adjustments to the classification intervals recommended by the ASTM C1202 method. The only outlier was the PC II F concrete, which received a moderate penetrability classification by ASTM C1202 and a high penetrability classification by IBRACON/NT Build 492.

4.3. Discussion About Standards and Specifications

International standards play a crucial role in defining the performance of concrete in aggressive environments, including those exposed to chlorides. Examples include EN 206:2013 (Europe) [73] and its complementary standards, such as BS 8500-2023 (UK) [74] and DIN 1045-2 (Germany) [75]. Other standards include NBR 6118:2014 (Brazil) [22], ACI 318-19 (USA) [76], CSA A23.1-24 (Canada) [77], IS 456:2000 (India) [78], AS 3600:2018 (Australia) [79], SANS 10100-2 (South Africa) [80], and NF P18-710 (France) [81]. These standards are often tailored to suit local conditions, allowing for regional adjustments in response to specific environmental factors. Despite the variations in the exact values and requirements, many of these standards share fundamental performance criteria, such as a maximum w/c ratio, minimum compressive strength, and adequate cover depth to protect reinforcing steel. In addition to these primary criteria, recommendations include chloride penetration resistance testing and the use of certain cement types, particularly those incorporating supplementary cementitious materials to enhance durability in chloride-rich environments.

The variability in performance criteria across different standards, along with regional adaptations, provides manufacturers with flexibility in meeting durability requirements while considering the availability of local materials and environmental conditions. This balance between flexibility and standardized criteria facilitates the development of concrete solutions that are both contextually appropriate and durable in aggressive exposure environments. However, while the cement type may not always be explicitly specified, and while testing for chloride penetration (especially non-accelerated tests) can be time-consuming, the choice of cement plays a crucial role in ensuring long-term durability. It could therefore be beneficial for some standards to explicitly define cement types, particularly for environments prone to chloride-induced corrosion, to further enhance the performance of concrete and reduce uncertainties in achieving required durability outcomes [76]).Finally, it is important to highlight alternative diffusion tests, such as ASTM C1543 [82] and NT Build 443 [24], which has been widely used to assess chloride ion migration in concrete. These tests generally take longer and do not involve the introduction of an electrical potential difference. ASTM C1543 (the ponding test) applies a 3% NaCl solution to one face of prismatic concrete specimens, while the opposite face is exposed to 50% relative humidity. NT Build 443 recommends saturating the concrete in lime-saturated water before immersing the specimens in a 16.5% NaCl solution. These tests focus on chloride ion migration rather than conductivity or permeability.

Both tests provide a more direct measurement of chloride migration and are particularly valuable for future research aimed at assessing the influence of different cement types on concrete’s resistance to chloride-induced deterioration—an approach that, to the best of our knowledge, has not been thoroughly investigated.

5. Conclusions

This study revealed that the ASTM C1202 method considers fewer variables in classifying chloride penetrability compared to IBRACON/NT Build 492. Despite this, there is a strong correlation in the results, with a coefficient of determination (R²) of 0.93. However, despite the good correlation between the classification variables regarding the chloride penetrability risk, normative classification intervals may lead to discrepancies in the risk classification of chloride penetrability between the two methods, with only three convergences in the classification results. Notably, concretes classified as having negligible penetrability by the IBRACON/NT Build 492 method were classified as having very low penetrability in ASTM C1202.

Based on the rapid chloride permeability test results, concretes produced with cements predominantly containing filler as an additive consistently show high chloride ion penetrability in both classification methods. Conversely, concretes made with cements rich in slag and pozzolan demonstrate medium chloride ion resistance, typically classified as having negligible or very low chloride ion penetrability. However, it is essential to emphasize that verifying these findings through chloride diffusion tests is advisable, particularly when utilizing these results to forecast durability using derived coefficients (Szweda et al. [83]). The literature shows a contradiction in establishing a correlation between the results of chloride diffusion and chloride migration tests. While Szweda et al. did not find a relationship between the values of accelerated chloride penetration for different concretes, some studies indicate a reasonable correlation between the chloride diffusion results obtained in non-steady-state tests and effective chloride diffusion (Al-Sodani et al. [84]; Pontes et al. [85]; Argiz, Moragues and Menéndez [86]; Bogas and Gomes [87]). While the rapid chloride permeability test provides valuable initial data on chloride ion penetrability, confirming these results through chloride diffusion tests is recommended for robust durability assessments, as outlined in standard ASTM C1202. By doing so, leveraging correlations from accelerated methods can enhance the overall evaluation of concrete durability under various conditions.

Lastly, attention was drawn to the classification rigidity of the ASTM C1202 method. Thus, new interval values are proposed to classify concretes regarding the risk of chloride ion penetration for the ASTM C1202 standard.