Use of X-Ray Fluorescence for the Evaluation of Chloride Ion Penetration in Concrete

Abstract

Chloride ions are the primary aggressive agents responsible for the corrosion of reinforced concrete structures and, consequently, the reduction of their service life, which entails high repair costs. These ions penetrate the concrete through its porous network and depassivate the steel reinforcement, even in small quantities. The adoption of specific binders modifies the cementitious matrix, potentially hindering the progression of these ions. As a chloride detection technique, X-ray fluorescence (XRF) can be employed, which, in addition to identifying the elements present in a sample, allows the creation of intensity distribution maps. Thus, this study aimed to evaluate how the XRF technique can contribute to understanding chloride penetration and to identify which cement types most effectively hinder their advancement. To accelerate chloride ion migration, two methods were employed: ASTM C1202 and NT Build 492. Eight different types of cement were analyzed, and the results indicated high resistance to chloride penetration in cements containing slag and pozzolan, while cements with filler exhibited lower resistance.

Keywords:
Cement; concrete; XRF; M4 Tornado

1. Introduction

The annual global cost of corrosion exceeds 3.5% of the world's Gross Domestic Product (GDP), amounting to approximately USD 3.38 trillion¹. The corrosion of reinforced concrete structures constitutes a significant portion of this cost and remains one of the primary challenges in civil construction, particularly in aggressive environments such as coastal regions and industrial areas. Among the various factors contributing to structural degradation, the presence of chloride ions stands out due to their ability to penetrate concrete and reach the steel surface, breaking the passive layer and leaving it unprotected²,³.

Chloride ions can enter concrete not only from its constituents, such as calcium chloride (CaCl₂) based set accelerators, contaminated aggregates, or mixing water, but also from external sources, including industrial environments, seawater exposure, and marine atmospheres⁴. Additionally, higher temperatures promote increased chloride diffusion and, consequently, deeper penetration⁵,⁶. Even small amounts of chloride ions can be detrimental to reinforced concrete, as Cl^- anions participate in corrosion reactions but are not consumed in the process⁷. Once inside the concrete, chlorides can exist in three primary forms: chemically bound as Friedel's salt (Fs), physically adsorbed on the surface of calcium silicate hydrates (CSH), or as free ions the latter being the most aggressive form for structures⁸. The chemical entrapment in the form of Fs occurs because the AFm phase, one of the main hydration products of C₃A and C₄AF, exhibits a double-layered lamellar structure. This structure consists of two electrically positive layers surrounding a negatively charged layer, which captures Cl⁻ anions⁹.

Among the measures that can be adopted to enhance the durability of structures, the use of low-porosity concrete is considered the most effective way to mitigate chloride attacks¹⁰. Increased concrete cover thickness, particularly in highly aggressive environments, also plays a crucial role in protecting the reinforcement¹¹. Moreover, research indicates that various cementitious additions not only reduce permeability but also enhance free chloride retention, hindering their transport to the reinforcement. The incorporation of pozzolanic materials or slag into the concrete mix, for example, improves durability by promoting additional Fs and CSH formation, which contribute to a greater entrapment of free chlorides¹²-¹⁵. As the chloride concentration in concrete increases, the binding with CSH may become more significant, while the influence of Fs may decrease¹⁵.

Several factors complicate scientific analysis and treatment of concrete, including its complex structure, the predominantly on-site manufacturing process, and the ongoing evolution of its components over time¹⁶. In this context, studies implementing various techniques are essential for a better understanding of the phenomenon and its influencing variables. Previous research has demonstrated the effectiveness of techniques such as X-ray fluorescence (XRF) for identifying chlorides in concrete samples¹⁷-¹⁹. XRF-generated images can reveal the relative intensity of chlorine, allowing for the assessment of preferred ion migration paths, the element's distribution, and consequently, the most affected zones²⁰,²¹. Additionally, cracks can be detected, as well as their respective impact on the penetration of aggressive agents²². However, due to the characteristic energy of chlorine, certain spectrometer targets may alter the detected signal based on their composition. Rhodium (Rh) targets, for instance, can increase the background noise, requiring the use of filters to achieve more accurate results²³. Other chemical analysis techniques, such as Laser-Induced Breakdown Spectroscopy (LIBS), may also be employed for chloride detection in concrete²⁴,²⁵, but are not addressed in this study.

Thus, this research aimed to investigate the use of XRF for the characterization and analysis of chloride ion penetration in concrete samples produced with different types of cement, seeking to optimize the application of this technique and understand the influence of cement composition on the observed results.

2. Experimental Procedure

2.1. Materials

Eight types of Portland cement used in Brazil were individually used in this study: CP II E, CP II E-RS, CP II F, CP II F-RS, CP II Z, CP III-RS, CP IV, and CP IV-RS, all classified for a compressive strength of 32 MPa at 28 days. CP II cements with the suffixes E, F, and Z represent, respectively, composite cements with the addition of blast furnace slag, carbonate material, and pozzolanic material. The CP III and CP IV cements also contain blast furnace slag and pozzolanic material, respectively, but in higher proportions. Meanwhile, the RS suffix represents sulfate-resistant cements.

The other constituents used for producing the concrete specimens were:

• − Fine Aggregate: fine washed sand, with a fineness modulus of 1.71;

• − Coarse Aggregate: type 0 crushed stone;

• − Water: potable water supplied by Empresa Baiana de Águas e Saneamento (EMBASA);

• − Superplasticizer Admixture: MC-PowerFlow 1180 from Bauchemie.

The properties of the aggregates are presented in Table 1 and were determined according to NBR 16916: Fine Aggregate – Determination of Density and Water Absorption²⁶ and NBR 16917: Coarse Aggregate – Determination of Density and Water Absorption²⁷.

Table 1
Physical properties of the aggregates.

Property	Fine Aggregate	Coarse Aggregate
Real Specific Gravity (kg/m3)	2844.71	2719.36
Loose Bulk Density (kg/m3)	1639.29	1366.61
Compacted Bulk Density (kg/m3)	1768.74	1553.35
Fineness Modulus (-)	1.71	2.46
Maximum Diameter (mm)	1.18	9.50
Moisture Content (%)	2.94	0.32

2.2. Concrete dosage

The cylindrical concrete specimens, measuring 100 mm in diameter and 200 mm in length, were prepared according to Table 2, following the dosing method established by the Brazilian Portland Cement Association (ABCP), designed for a compressive strength (f_ck) of 40 MPa. A superplasticizer admixture was used to improve concrete workability and achieve the slump values specified for the research. The molding process adhered to the guidelines of NBR 5738: Concrete – Procedure for Molding and Curing of Test Specimens²⁸.

Table 2
Dosage per m³ of concrete.

Type of Cement	Specific Gravity (kg/m3)	Cement (kg/m3)		Sand (kg/m3)	Gravel (kg/m3)	Water (l/m3)	Additive (l/m3)	Slump (cm)
CP II E	3000		420.30	330.44	704.20	137.12	0.77	20.00
CP II E - RS	2990		420.44	329.21	704.43	137.21	0.96	20.00
CP II F	2980		420.58	327.98	704.66	137.29	1.16	20.00
CP II F - RS	3000		420.30	330.44	704.20	137.12	0.58	20.00
CP II Z	3030		419.89	334.06	703.51	136.87	0.39	20.00
CP III – RS	2950		421.00	324.22	705.37	137.55	0.48	20.00
CP IV	3100		418.97	342.21	701.98	136.30	0.77	20.00
CP IV - RS	2960		420.86	325.48	705.13	137.47	0.58	20.00

After 24 hours, the specimens were demolded and subjected to wet curing in a saturated calcium hydroxide solution for 28 days, followed by 168 days of air curing. Subsequently, chloride ion migration acceleration tests and X-ray fluorescence (XRF) detection analyses were performed. The specific mass (kg/m³) of each binder used is presented in Table 2.

2.3. Test methods

2.3.1. Chloride ion migration acceleration test

For the chloride ion migration acceleration tests, the cylindrical concrete samples were cut into four smaller sections (100x50 mm), from which only the central parts were used. The tests were then performed in accordance with two distinct standards: ASTM C1202²⁹ and NT BUILD 492³⁰, the latter recommended by the Brazilian Concrete Institute (IBRACON).

In both standardized procedures, the concrete slices are placed between two solutions in a specific container, where a cathodic sodium chloride solution is placed at the bottom and an anodic sodium hydroxide solution at the top. Subsequently, using a power supply, the negative and positive terminals are connected to the lower and upper solutions, respectively, in order to force the migration of chloride ions from the lower solution to the upper solution through the concrete matrix. Table 3 summarizes the key testing characteristics of each standard.

Table 3
Comparison between ASTM C1202 and NT BUILD 492.

CHARACTERISTICS	ASTM C1202	NT BUILD 492
Sample	Cylindrical with 100 mm diameter and 50 mm thickness	Cylindrical with 100 mm diameter and 50 mm thickness
Duration	6 hours	From 6 to 96 hours (with voltage adjustment after initial current at 30 V)
Voltage	60 V	30 V (initial)
Concept	Evaluate the amount of electric current passing through the concrete sample.	Evaluate chloride penetration through sample rupture and AgNO3 spraying
Chloride Diffusion State	Steady-state	Non-steady-state
Cathodic Solution	3% NaCl	10% NaCl
Anodic Solution	0,3 mol/L NaOH	0,3 mol/L NaOH

The apparatus used for both tests is shown in Figure 1a. A sealant was applied to prevent the upward migration of chloride ions through the lateral concrete/apparatus interface, along with a slight inclination during the procedure, as specified by the standards.

Figure 1
(a) Chloride ion migration testing apparatus; (b) Sample rupture using a press; (c) Broken samples.

In accordance with ASTM C1202, chloride diffusion can be evaluated through the total charge passed, measured in Coulombs, while NT BUILD 492 determines the migration coefficient, expressed in m²/s. The results from both tests can be correlated with the chloride penetration risk classifications, which can be obtained from ASTM C1202 test or another study³¹, which refer to the D_ns obtained by the NT Build 492 test.

After the completion of the test, the samples were fractured using a diametral compression press and subsequently analyzed via X-ray fluorescence (XRF).

2.3.2. XRF analyses

For each type of cement, oxide quantification measurements were performed using the Epsilon 1 spectrometer. To generate elemental maps illustrating the zones of chlorine accumulation and penetration depth, the M4 Tornado spectrometer by Bruker was employed (Figure 2). This equipment features an X-ray tube with a rhodium (Rh) anode, equipped with a 25 µm polycapillary optic and an SDD (Silicon Drift Detector) with an energy resolution below 142 eV.

Figure 2
X-ray spectrometry equipment.

During the analysis, a voltage of 30 kV and a current of 600 µA were applied, with a pixel size of 150 µm and an exposure time of 20 ms per pixel. All measurements were conducted under vacuum conditions at a pressure of 20 mbar. To minimize the influence of the Rh anode, a 12.5 µm aluminum filter was applied.

Although the samples had an average size of 100 mm in width and 50 mm in height, measurements were limited to the central area, excluding 10 mm from each side due to potential chlorine migration through the edges during the accelerated test procedure. Therefore, the mapped area measured 80x50 mm, corresponding to 533 pixels horizontally and 333 pixels vertically.

The images present a color scale, with red hues indicating higher concentration and black indicating lower chlorine concentration (Figure 3). Additionally, the images feature a measurement scale ranging from 0 to 50 mm, which assists in determining the chlorine penetration depth in the different types of concrete.

Figure 3
Color scale based on concentration percentage.

To standardize the determination of the penetration depth in the samples, two criteria were defined:

1. The presence of a region with higher concentration that demonstrates interconnected chlorine ion flow, excluding isolated, disconnected points;

2. Intensity greater than 40%, according to the intensity scale, which corresponds to the transition from blue to green hues.

The X-ray fluorescence (XRF) spectra analysis was performed using PyMca 5.9.2 software, provided by the European Synchrotron Radiation Facility (ESRF).

3. Results and Discussion

3.1. Oxide content

Through Table 4, it is possible to evaluate the oxide contents present in each of the eight types of cement used. The alumina (Al₂O₃) contents were found to range from 3.28% to 7.63%, with the highest values observed in the CP III - RS, CP IV, and CP II E cements. The elevated alumina content promotes the formation of calcium chloroaluminate, also known as Friedel's salt, which consequently reduces the percentage of free ions in the matrix and the chloride ion transport rate⁹,³². Regarding the calcium oxide (CaO) contents, it is noted that the cements with the highest concentrations were CP II F - RS, CP II F, and CP II E - RS, while the lowest contents were found in CP II Z, CP IV, and CP IV - RS.

Table 4
Oxide content of anhydrous cements.

Type of Cement	Al2O3 (%)	CaO (%)	Fe2O3 (%)	SiO2 (%)	Mn2O3 (%)	SO3 (%)	K2O (%)	TiO2 (%)	SrO (%)
CP II E	5.97	61.46	7.88	13.69	1.54	2.30	0.98	1.04	0.82
CP II E - RS	3.47	64.38	9.34	9.37	0.82	3.32	1.11	0.85	0.69
CP II F	4.08	64.36	9.22	9.76	0.16	3.06	1.65	0.99	0.15
CP II F - RS	3.28	65.74	13.52	7.12	0.59	2.52	0.54	0.88	0.09
CP II Z	3.91	52.16	21.55	11.53	0.27	2.56	0.90	0.81	0.31
CP III – RS	7.63	58.17	4.91	19.76	1.72	3.44	0.51	1.28	0.38
CP IV	7.36	57.58	6.16	19.90	2.40	2.26	1.37	1.12	0.15
CP IV - RS	5.88	61.79	7.19	13.79	1.36	2.28	0.95	1.11	0.83

3.2. Chloride ion penetration analysis

The use of the M4 Tornado allowed for the acquisition of average spectra and elemental maps, which enable the interpretation of regions with the highest intensity of the studied element, as well as its distribution and interaction with the cementitious matrix. The criteria presented in section 2.3.2 were applied to all sample types, and the results for each adopted cement are discussed below.

3.2.1. Cements with addition of filler (CP II F and CP II F-RS)

The CP II F and CP II F-RS cements contain filler addition, which is a material with a finer particle size than clinker and contributes to better packing of the cement particles, thereby enhancing properties such as workability of mortars and concretes³³. For the samples subjected to the NT BUILD 492 test, it was observed that the chlorine peak exhibited higher intensity in the CP II F sample (Figure 4a), reaching a count of 1x10⁵, while for the CP II F-RS sample (Figure 4c), a value of approximately 4x10⁴ counts per second (cps) was obtained. This indicates that the first sample contains a higher amount of chlorine than the second, but it does not imply a greater penetration. Regarding the samples subjected to ASTM C1202, peaks of lower intensity, near the scales of 2x10⁴ and 4x10⁴ cps, were observed for the CP II F (Figure 4b) and CP II F-RS (Figure 4d) samples, respectively. Therefore, the behavior is opposite to the previous one, with CP II F-RS exhibiting a higher peak intensity.

Figure 4
Average spectrum of the samples, with those on the left subjected to NT BUILD 492 and those on the right subjected to ASTM C1202. (a) and (b) CP II F; (c) and (d) CP II F-RS.

Figure 5a and Figure 5c illustrate, respectively, the concrete samples made with CP II F and CP II F-RS subjected to the NT BUILD 492 migration acceleration test. According to the established criteria, a penetration of approximately 29 mm was observed for the former. In contrast, for the same type of cement, but sulfate-resistant (CP II F-RS), although there was no distinct penetration zone as in the previous case, the chloride ions were distributed throughout the sample with relative intensity, reaching up to 36 mm of depth in the left portion of the image.

Figure 5
Images of chloride ions in the samples, with those on the left subjected to NT BUILD 492 and those on the right subjected to ASTM C1202. (a) and (b) CP II F; (c) and (d) CP II F-RS.

In relation to Figure 5b and Figure 5d, with the same types of cement subjected to the ASTM C1202 acceleration test, a different behavior was observed. The concrete made with CP II F exhibited a penetration of approximately 14 mm, about half the penetration seen in Figure 5a. On the other hand, the chloride ions in CP II F-RS penetrated to about 33 mm. Nevertheless, in both tests, the sulfate-resistant cement (RS) demonstrated lower resistance to penetration, as the element was detectable throughout nearly the entire sample, indicating a low capacity to retain chlorides. In general, a similar behavior was observed, with high permeability to chloride ion penetration for cements containing filler addition.

An important observation is the highly irregular distribution of chlorine across the samples, lacking uniformity. What is noted are possible preferential paths followed by the element and specific zones of accumulation. It is also evident that the coarse aggregate appears as dark spots, devoid of chlorides. The ions primarily migrate through interconnected pores within the cementitious matrix and may encounter, at the paste-aggregate transition zone, a path of higher permeability and ease of transport.

In general, the peak intensities align with the images obtained. However, it is observed that the most pronounced characteristic peak identified in CP II F – NT BUILD 492 does not correlate with greater penetration depth, as evidenced in CP II F-RS – NT BUILD 492.

3.2.2. Cements with addition of Slag (CP II E, CP II E-RS and CP III-RS)

Portland cements of type II E are those that contain between 6% and 34% blast furnace slag, whereas CP III-RS cements have higher mass proportions, ranging from 35% to 75%³⁴. Analyzing the average spectra obtained for these types of cements, it is observed that, in general, they exhibited low counts, except for CP II E subjected to NT BUILD 492, as shown in Figure 6a. While this sample displayed a peak of 4x10⁴ cps, the others showed peaks of much lower magnitude, on the order of 0.6x10⁴ cps. Among the six samples, CP III-RS subjected to ASTM C1202 (Figure 6f) exhibited the characteristic peak with the lowest intensity.

Figure 6
Average spectrum of the samples, with those on the left subjected to NT BUILD 492 and those on the right subjected to ASTM C1202. (a) and (b) CP II E; (c) and (d) CP II E-RS; (e) and (f) CP III-RS.

Regarding the elemental maps, CP II E subjected to NT BUILD 492, shown in Figure 7a, demonstrated a penetration depth of 16 mm. On the right side of the image, an advance greater than this value is visible, but it is believed that an edge effect occurred due to the image composition. Chlorides likely had greater ease of transport at the interface between the side of the specimen and the apparatus used in the ion migration acceleration test.

Figure 7
Images of chloride ions in the samples, with those on the left subjected to NT BUILD 492 and those on the right subjected to ASTM C1202. (a) and (b) CP II E; (c) and (d) CP II E-RS; (e) and (f) CP III-RS.

For the sulfate-resistant cements, CP II E-RS and CP III-RS, represented by Figures 7c and 7e, respectively, a slightly lower penetration was observed, with chlorine being retained within the first 12 mm. In the case of CP II E-RS in Figure 7d, it is observed that chlorine is predominantly located within the first 18 mm of depth. The images show the presence of chlorine at low intensity, and the lower the total amount of the element in the sample, the smaller the concentration difference between regions, thus increasing the relevance of points with low concentration scattered throughout the image.

For the samples subjected to the test prescribed by the American standard ASTM C1202, a different behavior was noted for the CP II E and CP III-RS samples. These did not exhibit a recognizable penetration pattern, as higher concentrations of chlorine were observed in isolated, distinct, and unconnected areas. This could be attributed to various factors, such as the low concentration of the element, which highlights the influence of the background and the noise generated in the test. As described in Section 2.3.2, an aluminum filter was used to attenuate the influence of the spectrometer target material (rhodium). Therefore, the noise mentioned is more likely related to scattered radiation, which contributes to the background signal. The inherent irregularity of the concrete samples may have influenced this aspect, and for images with low chlorine content, the background signal is more noticeable.

It is also evident that the cements with slag addition exhibited higher resistance to chloride ion penetration than the cements with filler, as seen both in the maps and spectra.

3.2.3. Cements with addition of Pozzolan (CP II Z, CP IV and CP IV-RS)

CP II Z cements contain an addition of pozzolanic material in their production, ranging from 6% to 14% by mass. In contrast, CP IV contain between 15% and 50% by mass. To be classified as sulfate-resistant, this percentage must be between 25% and 40%³⁵. Generally, cements with pozzolana additions form concretes with good durability³³.

Among the average spectra presented in Figure 8, it is observed that the results are very similar, showing no significant distinctions. The characteristic peaks approach the background, indicating low intensity of the element distributed across the area. Nevertheless, it is evident that CP IV subjected to NT BUILD 492 exhibits the most representative chlorine peak, while CP IV-RS shows the lowest count, indicating lower chloride ion intensity at the analyzed surface.

Figure 8
Average spectrum of the samples, with those on the left subjected to NT BUILD 492 and those on the right subjected to ASTM C1202. (a) and (b) CP II Z; (c) and (d) CP IV; (e) and (f) CP IV-RS.

Regarding chloride ion penetration, only items (a), (b), and (c) in Figure 9 exhibited behavior consistent with the criteria established for identifying penetration depth. For the remaining images, it was not possible to delineate a penetration profile, as no uniform variation, transitioning from higher to lower quantities in the ascending direction, was observed. Instead, there is a highly irregular distribution of chlorine in the analyzed area, with isolated points indicating higher intensity, but without a migration pattern or well-defined regions. This is likely due to the low concentration of the element in these samples, suggesting that they are cements resistant to chloride ion penetration.

Figure 9
Chloride ion images of the samples, with those on the left subjected to NT BUILD 492 and those on the right subjected to ASTM C1202. (a) and (b) CP II Z; (c) and (d) CP IV; (e) and (f) CP IV-RS.

For CP II Z and CP IV subjected to the NT BUILD 492 standard, ions penetrated to a depth of 15 mm, whereas for CP II Z subjected to ASTM C1202, a penetration depth of 17 mm was observed. This could be due to higher permeability resulting from interconnected pores. Furthermore, it is emphasized that peaks of higher intensity in the spectrum do not necessarily indicate greater penetration depths, which is the most relevant factor for reinforcing steel corrosion, as these are protected by concrete covers.

A clear distinction is observed between Figure 9c, which shows a higher quality image and allows for a notable observation of chlorine intensity zones, and the other items in the same figure. This is due to a higher concentration of chlorine, which consequently facilitates a more discernible intensity difference between regions.

3.2.4. Classification of cements according to resistance to penetration

Through the patterns identified in the figures, the penetration depths for each cement can be summarized in Table 5. Some samples did not exhibit identifiable and well-defined behavior, and thus were labeled as "non-detectable."

Table 5
Penetration depth by cement type.

Ranking	NT BUILD 492		ASTM C1202
	Sample	Depth penetration	Sample	Depth penetration
1º	CP II F-RS	36 mm	CP II F-RS	33 mm
2º	CP II F	29 mm	CP II E-RS	18 mm
3º	CP II E	16 mm	CP II Z	17 mm
4º	CP IV	15 mm	CP II F	14 mm
5º	CP II Z	15 mm	CP II E	Not detectable
6º	CP II E-RS	12 mm	CP IV-RS	Not detectable
7º	CP III-RS	12 mm	CP IV	Not detectable
8º	CP IV-RS	Not detectable	CP III-RS	Not detectable

From the table, it is evident that cements with fillers in their composition exhibited high penetration. CP II F-RS performed slightly worse than CP II F, likely due to its nature as a sulfate-resistant cement, which contains a lower quantity of aluminates in its composition, and consequently, a reduced capacity to chemically trap chlorides.

On the other hand, the cements that exhibited the lowest penetration were CP II E-RS, CP III-RS, and CP IV-RS, to the point of displaying images in which it was difficult to assess the zones of higher and lower chloride intensity. This is due to these cements containing, respectively, slag and pozzolana, which provide beneficial durability factors such as reduced porosity and the formation of additional CSH. Cements CP II E and CP II Z also contain such additions and benefit from these factors, albeit to a lesser extent, due to the lower proportion of these additions.

When comparing the two migration test methods, NT BUILD 492 and ASTM C1202, a similarity is observed for the cements CP II F-RS, CP II E-RS, CP II Z, and CP IV-RS with respect to penetration depth. CP III-RS also showed similar behavior when comparing, not the measurements, but rather the elemental maps. However, for the other cement types, such as CP II F, CP II E, and CP IV, significant differences were observed between the two migration tests, with higher values obtained for those subjected to NT BUILD 492. A likely reason for this is the voltage used in the tests. While the ASTM C1202 standard specifies a fixed voltage of 60 V for 6 hours of testing, the NT BUILD 492 test applies an initial voltage of 30 V, adjusting the voltage and testing duration depending on the initial current measured, which can vary from 24 to 96 hours.

All samples used in the NT BUILD 492 test were conducted for 24 hours, except for CP III-RS, which required 48 hours. Additionally, CP II E and CP IV showed adjusted voltages of 50 and 60 V, respectively, which may have enhanced ionic migration. Another factor that could influence significant differences between samples with the same cement type is the heterogeneity of the concrete and the interconnection of the pores.

It is possible to correlate the migration coefficient and passing charge from the respective NT BUILD 492 and ASTM C1202 tests with the penetration risk classifications, as mentioned in section 2.3.1. The correlation is presented in Table 6.

Table 6
Classification according to the penetrability of chloride ions.

Type of Concrete	NT BUILD 492		ASTM C1202
	Average (x10-12 m2/s)	Risk of penetration	Average (Coulomb)	Risk of penetration
CP II E	2.83	Very low	1092.72	Low
CP II E - RS	5.86	Low	2188.50	Moderate
CP II F	15.03	High	4233.96	High
CP II F – RS	15.22	High	4401.42	High
CP II Z	6.36	Low	2439.81	Moderate
CP III – RS	0.98	Negligible	221.91	Very low
CP IV	0.71	Negligible	152.55	Very low
CP IV – RS	2.05	Negligible	675.45	Very low

For a set of 3 samples per test, the average values of the non-steady-state migration coefficient in m²/s and the average passing charge in Coulombs are observed. A good correlation is noted between the tests, as the risk classification was the same for the vast majority of the concrete types. CP II F and CP II F-RS cements exhibited a high risk of penetration, while CP III-RS, CP IV, and CP IV-RS received a negligible classification regarding penetrability. These results are consistent with the analyses obtained by XRF. The most significant variations are observed for CP II E-RS and CP II Z, as they were classified as low risk for NT BUILD 492 and moderate risk for ASTM C1202.

4. Conclusions

The use of the XRF technique yielded significant results regarding the influence of different types of cement on the chloride ion penetration resistance in concrete.

1. Cements with filler additions exhibited the highest concentration and penetration depth of chloride ions, while cements containing blast furnace slag and pozzolana demonstrated a more resistant behavior. Among these, CP II E-RS, CP III-RS, and CP IV-RS stood out as the most resistant.

2. The samples prepared according to the NT Build 492 standard resulted in greater chloride ion penetration compared to ASTM C1202, which may influence the outcomes of other studies.

3. The XRF technique proved to be rapid, easy to interpret, and advantageous, particularly because it is non-destructive. Furthermore, the elemental maps provided valuable information regarding the position and transport of the studied element, zones of higher intensity, and behavior in relation to the concrete’s microstructure.

4. The observed and discussed average spectra reveal that more intense chloride peaks are not necessarily associated with greater penetration depths, but rather with higher amounts of the element, which may be located in the outer layers of the concrete.

5. Despite its advantages, the technique presented noise, especially in samples with low chloride ion intensity, making the perception of penetration more challenging.

5. Acknowledgments

This work was supported by the following entities: Coordination for the Improvement of Higher Education Personnel - Brazil (CAPES), the Graduate Program in Science, Innovation, and Modeling in Materials (PROCIMM) at the State University of Santa Cruz (UESC), the Mechanical Testing and Material Strength Laboratory (LEMER), and the Laboratory of Electronic Instrumentation and Analytical Techniques (LIETA) at the State University of Rio de Janeiro.

• Data Availability
  The data supporting the findings of this study are available at: https://data.scielo.org/dataset.xhtml?persistentId=doi:10.48331/SCIELODATA.AVQNMK&version=DRAFT

6. References

• ¹ NACE International. (2016). International Measures of Prevention, Application, and Economics of Corrosion Technology Study (IMPACT) [cited 2025 Jan 7]. Available from: http://impact.nace.org/documents/Nace-International-Report.pdf
  » http://impact.nace.org/documents/Nace-International-Report.pdf
• ² Silva LK, Cabral AEB. Levantamento de manifestações patológicas em estruturas de concreto armado no estado do Ceará. In: Encontro Nacional de Tecnologia do Ambiente Construído; 2014; Maceió, Alagoas, Brasil. Anais. Porto Alegre: ANTAC; 2014. p. 1626-34. http://doi.org/10.17012/entac2014.21.
• ³ Mahima S, Moorthi PVP, Bahurudeen A, Gopinath A. Influence of chloride threshold value in service life prediction of reinforced concrete structures. Sadhana. 2018;43(7):1-19. http://doi.org/10.1007/s12046-018-0863-5
  » http://doi.org/10.1007/s12046-018-0863-5
• ⁴ Isaia GC, editor. Concreto: ensino, pesquisa e realizações. Vol. 2. São Paulo: IBRACON; 2005. 600 p.
• ⁵ Nguyen TS, Lorente S, Carcasses M. Effect of the environment temperature on the chloride diffusion through CEM-I and CEM-V mortars: an experimental study. Constr Build Mater. 2009;23(2):795-803. http://doi.org/10.1016/j.conbuildmat.2008.03.004
  » http://doi.org/10.1016/j.conbuildmat.2008.03.004
• ⁶ Dousti A, Rashetnia R, Ahmadi B, Shekarchi M. Influence of exposure temperature on chloride diffusion in concretes incorporating silica fume or natural zeolite. Constr Build Mater. 2013;49:393-9. http://doi.org/10.1016/j.conbuildmat.2013.08.086
  » http://doi.org/10.1016/j.conbuildmat.2013.08.086
• ⁷ Ribeiro DV, coordenador. Corrosão em estruturas de concreto armado: teoria, controle e métodos de análise. São Paulo: Campus; 2014.
• ⁸ IBRACON. Guia para determinação do perfil de concentração e profundidade de penetração de cloretos. São Paulo: IBRACON; 2021.
• ⁹ Baquerizo LG, Matschei T, Scrivener KL, Saeidpour M, Wadsö L. Hydration states of AFm cement phases. Cement Concr Res. 2015;73:143-57. http://doi.org/10.1016/j.cemconres.2015.02.011
  » http://doi.org/10.1016/j.cemconres.2015.02.011
• ¹⁰ Chandramouli, K, Srinivasa Rao P, Seshadri Sekhar T, Pannirselvam N, Sravana P. Rapid chloride permeability test for durability studies on glass fibre reinforced concrete. ARPN J Eng Appl Sci. 2010;5(3):67-71.
• ¹¹ ABNT: Associação Brasileira de Normas Técnicas. NBR 6118: projeto de estruturas de concreto. Rio de Janeiro: ABNT; 2023.
• ¹² De Oliveira AM, Cascudo O. Effect of mineral additions incorporated in concrete on thermodynamic and kinetic parameters of chloride-induced reinforcement corrosion. Constr Build Mater. 2018;192:467-77. http://doi.org/10.1016/j.conbuildmat.2018.10.100
  » http://doi.org/10.1016/j.conbuildmat.2018.10.100
• ¹³ Andrade C, Buják R. Effects of some mineral additions to Portland cement on reinforcement corrosion. Cement Concr Res. 2013;53:59-67. http://doi.org/10.1016/j.cemconres.2013.06.004
  » http://doi.org/10.1016/j.cemconres.2013.06.004
• ¹⁴ Figueiredo CP, Santos FB, Cascudo O, Carasek H, Cachim P, Velosa A, et al. The role of metakaolin in the protection of concrete against the deleterious action of chlorides. Rev IBRACON Estrut Mater. 2014;7:685-708. http://doi.org/10.1590/S1983-41952014000400008
  » http://doi.org/10.1590/S1983-41952014000400008
• ¹⁵ Chang H, Fan S, Guo Z, Ma Y, Li Z, Zuo Z, et al. Chloride binding behavior of calcium silicate hydrates and Friedel’s salt of pastes. J Build Eng. 2023;76:107318. http://doi.org/10.1016/j.jobe.2023.107318
  » http://doi.org/10.1016/j.jobe.2023.107318
• ¹⁶ Coimbra MA, Libardi W, Morelli MR. Estudo da influência de cimentos na fluência em concretos para a construção civil. Ceramica. 2006;52(321):98-104. http://doi.org/10.1590/S0366-69132006000100015
  » http://doi.org/10.1590/S0366-69132006000100015
• ¹⁷ Danner T, Geiker MR, De Weerdt K. µ -XRF characterisation of chloride ingress and self-healing in cracked concrete. In: XXIII Symposium on Nordic Concrete Research & Development; 2017; Aalborg. Proceedings. Aalborg: Norsk Betongforening; 2017. p. 3-6.
• ¹⁸ Bran-Anleu P, Wangler T, Nerella VN, Mechtcherine V, Trtik P, Flatt RJ. Using micro-XRF to characterize chloride ingress through cold joints in 3D printed concrete. Mater Struct. 2023;56(3):51. http://doi.org/10.1617/s11527-023-02132-w PMid:36909254.
  » http://doi.org/10.1617/s11527-023-02132-w
• ¹⁹ Chinchón-Payá S, Torres Martín JE, Rebolledo Ramos N, Sánchez Montero J. Use of a handheld x-ray fluorescence analyser to quantify chloride ions in situ: a case study of structural repair. Materials (Basel). 2021;14(3):571. http://doi.org/10.3390/ma14030571 PMid:33530387.
  » http://doi.org/10.3390/ma14030571
• ²⁰ Geiker M, Robuschi S, Lundgren K, Paraskevoulakos C, Gundlach C, Danner T, et al. Concluding destructive investigation of a nine-year-old marine-exposed cracked concrete panel. Cement Concr Res. 2023;165:107070. http://doi.org/10.1016/j.cemconres.2022.107070
  » http://doi.org/10.1016/j.cemconres.2022.107070
• ²¹ Silva SFM. Estudo da penetração de cloretos por migração iônica em concretos incorporados com fibras utilizando a técnica de XRF [dissertation]. Ilhéus: Universidade Estadual de Santa Cruz; 2023.
• ²² Moradllo MK, Sudbrink B, Hu Q, Aboustait M, Tabb B, Tyler Ley M, et al. Using micro X-ray fluorescence to image chloride profiles in concrete. Cement Concr Res. 2017;92:128-41. http://doi.org/10.1016/j.cemconres.2016.11.014
  » http://doi.org/10.1016/j.cemconres.2016.11.014
• ²³ Menzies A, Buegler M, Tagle R, Reinhardt F. Investigation of concrete by means of micro-XRF. Microsc Microanal. 2021;27(S1):3182-5. http://doi.org/10.1017/S1431927621011004
  » http://doi.org/10.1017/S1431927621011004
• ²⁴ Labutin TA, Popov AM, Raikov SN, Zaytsev SM, Labutina NA, Zorov NB. Determination of chlorine in concrete by laser-induced breakdown spectroscopy in air. J Appl Spectrosc. 2013;80(3):315-8. http://doi.org/10.1007/s10812-013-9766-8
  » http://doi.org/10.1007/s10812-013-9766-8
• ²⁵ Bryukhova AS, Kuznetsov AA, Seliverstova IV, Popov AM, Labutin TA, Zorov NB. Evaluation of aging of reinforced concrete structures by laser-induced breakdown spectroscopy of reinforcement corrosion products. J Appl Spectrosc. 2020;87(5):800-4. http://doi.org/10.1007/s10812-020-01073-4
  » http://doi.org/10.1007/s10812-020-01073-4
• ²⁶ ABNT: Associação Brasileira de Normas Técnicas. NBR 16916: agregado miúdo - determinação da densidade e da absorção de água. Rio de Janeiro: ABNT; 2015.
• ²⁷ ABNT: Associação Brasileira de Normas Técnicas. NBR 16917: agregado graúdo – determinação da densidade e da absorção de água. Rio de Janeiro: ABNT; 2015.
• ²⁸ ABNT: Associação Brasileira de Normas Técnicas. NBR 5738: concreto procedimento para moldagem e cura de corpos de prova. Rio de Janeiro: ABNT; 2016.
• ²⁹ ASTM: American Society for Testing and Materials. ASTM C1202: standard test method for electrical indication of concrete’s ability to resist chloride ion penetration. West Conshohocken: ASTM; 2010.
• ³⁰ NORDTEST. NT BUILD 492: concrete, mortar and cement-based repair materials: chloride migration coefficient from non-steady-state migration experiments. Finlândia: NORDTEST; 1999.
• ³¹ Nilsson L, Ngo MH, Gjørv OE. High performance repair materials for concrete structures in the Port of Gothenburg. In: Second International Conference on Concrete under Severe Conditions - Environment and Loading; 1998; London. Proceedings. Boca Raton: CRC Press; 1998. p. 1193-8.
• ³² Rasheeduzzafar, Al-Saadoun SS, Al-Gahtani AS, Dakhil FH. Effect of tricalcium aluminate content on corrosion of reinforcing steel in concrete. Cement Concr Res. 1990;20(5):723738. http://doi.org/10.1016/0008-8846(90)90006-J
  » http://doi.org/10.1016/0008-8846(90)90006-J
• ³³ Battagin AF, Rodrigues H. Recomendações sobre o uso dos distintos tipos de cimento Portland nas diferentes aplicações. Rev Concr Const. 2014;73:30-8.
• ³⁴ ABNT: Associação Brasileira de Normas Técnicas. NBR 16697: Cimento Portland - Requisitos. Rio de Janeiro: ABNT; 2018.
• ³⁵ Ribeiro DV, coordenador. Princípios da ciência dos materiais cimentícios: produção, reações, aplicações e avanços tecnológicos. Curitiba: Appris Publishing; 2021.

Edited by

• Associate Editor:
  Eliana Muccillo.

• Editor-in-Chief:
  Luiz Antonio Pessan.

Data availability

The data supporting the findings of this study are available at: https://data.scielo.org/dataset.xhtml?persistentId=doi:10.48331/SCIELODATA.AVQNMK&version=DRAFT

Publication Dates

• Publication in this collection
  01 Sept 2025
• Date of issue
  2025

History

• Received
  07 Jan 2025
• Reviewed
  24 May 2025
• Accepted
  29 June 2025