Mechanism and kinetics of reactive yellow 145 decolorization by catalytic nano zero-valent iron particles

Abstract

In this study, the decolorization of Reactive Yellow 145 (RY-145) dye, which is commonly used in the textile industry, was investigated using nano zero-valent iron (nZVI) particles. The synthesized nZVI particles were found to agglomerate, forming a porous structure that allows dye molecules to access the surface. It was determined that the main mechanism of the decolorization process involves the initial adsorption of RY-145 onto the nZVI surface, followed by a simultaneous reaction with hydrogen molecules formed on the surface. Experiments revealed that the highest decolorization efficiency (96.4%) was achieved under the most acidic conditions (pH 3.0), highlighting the importance of hydrogen molecule formation. Furthermore, increasing the amount of nZVI and decreasing the initial dye concentration enhanced both the decolorization rate and efficiency. The adsorption mechanism showed good compliance with the Langmuir isotherm model and was defined as a “favorable” process. However, due to the continuous dissolution of the nZVI surface during the reaction, it was concluded that the term "pseudo-adsorption" is more appropriate for this process. Kinetic analyses showed that the decolorization rate is best described by the pseudo-second-order kinetic model. The findings of this study highlight the potential of nZVI in wastewater treatment applications.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-47465-y.

Introduction

Nanotechnology is defined as a field of research and innovation related to creating ‘objects’ (usually materials and devices) at the scale of atoms and molecules. Recently, in many fields, nanotechnology, industrial¹ or domestic such as drug development, medical treatment^2–4, wastewater treatment, material synthesis, development of biosensors and bio-nanomaterials, etc. has been started to be applied^5,6. In chemical reactions, nano size particles have some advantage like large surface area, charge stabilization, relatively lesser heterogeneity, and the more effectively mass transfer. The last decades, zero-valent iron (ZVI) has attracted much attention from researchers in the removal of contaminants for its advantages in cost effective and environmentally friendly^7,8. The potential applications of ZVI in waste water⁹ and soil system have been widely explored, including removal of chlorinated organic compounds¹⁰, nitrobenzene¹¹, organic dyes¹², phenols¹³, inorganic anions¹⁴, metalloids¹⁵ and drugs⁹. On the other hand, as a result of the surface losing its reactivity due to the formation of iron hydroxide, studies have been intensified by decreasing to nano dimensions. Nano zero valence iron (nZVI) in the reaction medium has two properties: as both reducing agent^16–20 and a catalyst¹⁷ have reported that the reducing potential of Fe⁰ at neutral pH is −0.44 V. Mukherjee et al., 2016¹⁷ found that nano iron particles are good catalyst and reductant against several chlorinated organic compounds and metal ions.

On the other hand, nZVI not only provides the decolorization of refractory organic dyes but also facilitates their biodegradations²¹. It is well known that dyes with chromophoric systems such as azo, anthraquinone, and phthalocyanine give color to water even at low concentrations. These systems can absorb solar irradiation, and the intensity of this irradiation is vitally important for photosynthesis. Unfortunately, in deep water, solar irradiation may not reach the bottom or its intensity remains at a very low level²². Therefore, the decolorization of dyes is of vital importance for the aquatic environment. To this end, many researchers have tried to eliminate the risk of various types of dyes to the environment by using different decolorization and degradation methods. They can be listed as Fenton, UV/H₂O₂, UV/Fenton²³, Wet Air Oxidation^24,25, UV/TiO₂²⁶, Electro-Fenton^24,25, Ozonation²⁷, Sonolysis²⁸. In particular, the heterogeneous Fenton process has recently emerged as a highly promising alternative due to its superior stability, ease of catalyst recovery, and efficiency across a broader pH range compared to traditional homogeneous Fenton systems^29,30. Despite these advancements, the search for cost-effective and multi-functional materials continues. While nZVI has been widely studied, its dynamic role as a sacrificial reagent—where its metallic core is consumed to generate reactive species—is often overlooked in favor of viewing it as a static adsorbent. The novelty of this work resides in the quantitative deciphering of this sacrificial nature of green-synthesized nZVI, correlating the molar consumption of the metallic core with the volumetric production of reactive hydrogen. Furthermore, the specific porous structure of the green-stabilized nZVI (as confirmed by BET analysis) offers a unique perspective on the synergy between adsorption and reduction without requiring external noble metal catalysts.

The primary objective of this study was to achieve the decolorization of RY-145 dye using nZVI (nano zero-valent iron). To support this goal, our research focused on three key areas: a) We investigated whether nZVI functions as both a reducing agent and a catalyst during the decolorization process. b) We determined the impact of nZVI’s behavior in an acidic environment on hydrogen molecule production and changes in its surface area. c) We established the kinetics of the decolorization reaction and identified the most suitable kinetic model.

Material and method

Materials

Reactive Yellow 145 (RY145) is a textile dye from Kıvanç Tekstil (Adana, Turkey) with a molecular weight of 1026.2 g/mol, a pH of 5.9 (1% solution), a solubility of 200 g/dm³ at 60 °C, and a purity of 96.7%. Its Topological Polar Surface Area (TPSA) is 457 Ų³¹. Other chemicals, such as FeCl₃​⋅6H₂​O (98.0%) and NaBH₄​ (98.0%), were purchased from Merck (Germany). The chemical structure of RY-145 is shown in Fig. 1. The change in RY-145 concentration during a decolorization experiment was monitored at a wavelength of 422 nm using a double beam UV–VIS spectrometer (Shimadzu UV-2101PC UV–VIS scanning spectrophotometer). The decolorization efficiency (DE %) of the dye was calculated according to the following equation:

Fig. 1.

RY-145 chemical formula.

where C₀ (mg/L) is the initial concentration of RY-145 and C_t (mg/L) is the concentration at a given reaction time (t).

The crystal structure and morphology-particle size of Nano-Zero-Valent-Iron (nZVI) were analyzed using X-ray diffraction (XRD) and scanning electron microscopy (SEM), respectively. For the XRD analysis, a Rigaku Miniflex diffractometer combined with a D/teX Ultra 250 detector was used, employing Cu-Kα radiation (λ = 0.154056 nm) in the 2θ range of 20–80°. SEM analysis was performed with a JEOL JSM-5500 instrument. The Brunauer–Emmett–Teller (BET) surface area was determined to be 44.7 ± 0.4 m²g⁻¹ using a Tristar 3000 surface analyzer (Micromeritics, USA).

Method

nZVI synthesis

First, a 1.000 dm³ solution of 0.045 M FeCl₃​⋅6H₂​O salt containing Fe³⁺ ions with an ionic diameter of 0.129 nm, was prepared. This solution was then transferred to a 2.000 dm³ reactor. After hydration, the ionic diameter of the Fe³⁺ expanded to 0.900 nm due to water molecules. To ensure a strictly anaerobic environment and to physically expel dissolved oxygen, the Fe³⁺ precursor solution was first heated to 90 °C. Subsequently, it was cooled to room temperature while high-purity nitrogen gas (N₂, 99.999%) was bubbled through the solution for 30 min. Simultaneously, without heating, a separate 1.000 dm³ NaBH₄ solution (0.250 M) was also purged with nitrogen gas for 30 min to remove any remaining dissolved oxygen. The 0.250 M NaBH₄ solution was added dropwise to the 0.045 M Fe³⁺ solution while the reaction mixture was stirred vigorously with a magnetic stirrer under a continuous nitrogen blanket to prevent the oxidation of nZVI.

After the entire 1.000 dm³ of NaBH₄​ solution was added, the reaction mixture was immediately filtered under reduced pressure using a nitrocellulose membrane filter (0.1 µm pore size, 47 mm diameter). The resulting very small, black-colored particulate product was washed three times with 50 cm³ of water to remove impurities. It was then washed another three times with 50 cm³ of methanol to remove water molecules. Finally, the product was dried under reduced pressure and stored under a nitrogen atmosphere. The effectiveness of this deoxygenation protocol was verified by XRD analysis of the fresh samples, confirming the metallic state of the particles. This procedure is detailed in Fig. 2.

Fig. 2.

Illustration of the synthesis steps of NZVI, including the diameters of atoms/ions/molecules and a demonstration of their relative volumes.

Decolorization experiments

The dye solutions used in the decolorization experiments were buffered to maintain a stable pH. The necessary pH adjustments were made by titrating a 0.100 M acetic acid solution with a 1.000 M sodium hydroxide solution, with the pH being monitored by a pH meter.

An RY-145 solution (0.100 mM) was prepared in different acetate buffers at pH 3, 7, and 11. Five hundred cm³ of the dye solution was placed into a 1.000 dm³ reactor (8 cm diameter, 20 cm height). While the reactor contents were mixed with a magnetic stirrer, a weighed amount of nZVI was added. One cm³ samples were taken every 10 minutes for 120 minutes before centrifugation. To determine the optimum quantity of nZVI, three different amounts of nZVI (0.200, 0.500, and 1.000 g) were studied at an initial dye concentration of 0.100 mM.

Sample preparing for analysis and determination of system parameters

For XRD and SEM analysis, the synthesized and filtered nZVI samples were dried using a vacuum freeze dryer (Dura-Dry VP190D) and then stored under a nitrogen atmosphere (Fig. 3a and Fig. 4). To observe the behavior of nZVI in the reaction mixture without RY-145, a series of preliminary experiments were conducted in 500 cm³ of a pH 3 buffer solution. In one of these experiments, after the addition of nZVI, the reactor was stirred for 5 minutes. The entire reactor content was then immediately filtered under reduced pressure, and the nZVI was dried using a vacuum freeze dryer. Finally, the dried nZVI was subjected to IR analysis using an FTIR instrument. This same procedure was replicated for a reaction time of 60 minutes, with the results shown in Fig. 5.

Fig. 3.

XRD patterns of nZVI particles before (a) and after (b) the decolorization process.

Fig. 4.

SEM images of nZVI different magnification.

Fig. 5.

(a): The FTIR spectrum of nZVI after 5 min in absence of RY-145. (b): FTIR spectrum of RY-145 dye. (c): FTIR spectrum of the RY-145 and nZVI reaction mixture after 10 min. (d): FTIR spectrum of the RY-145 and nZVI reaction mixture after 60 min..

Another experiment under the same conditions was designed to monitor dissolved Fe²⁺ ions. For this purpose, 2.0 cm³ samples were periodically collected from a reaction environment containing only a pH 3 buffer solution and nZVI at 0, 10, 20, 30, 40, 50, 60, 80, 100, and 120 minutes, and subsequently centrifuged. The supernatant was analyzed using an Atomic Absorption Spectrometer (AAS) (Agilent, 280FSAA).

To determine the potential hydrogen gas reservoir of the nZVI, 0.5 g of nZVI was treated with 10 cm³ of a 9.000 M Hydrochloric acid solution for 5 minutes. The volume of hydrogen gas formed was determined by its displacement of water in a 500 cm³ measuring cylinder. Before this measurement, the method was calibrated using iron powder (99%) purchased from Merck. For the calibration, iron powder was weighed in separate amounts of 0.05, 0.10, 0.20, 0.40, and 0.60 g, and the corresponding hydrogen gas volumes were measured at the same condition.

To determine if nZVI has a catalytic effect, hydrogen gas (purity, 99.999%) was passed through a 0.10 mM RY-145 buffer solution at a flow rate of 10 cm³min⁻¹ for 20 minutes, in the absence of nZVI. Five cm³ samples were collected separately at reaction times of 0 and 20 minutes and analyzed using a UV/Vis spectrometer (Fig. 6).

Fig. 6.

UV/VIS spectrum of 0.10 mM RY-145 dye solution before and after hydrogen gas passing for 20 min.

Results and discussion

Characterization of nZVI and system parameters

Figure 1 presents a comprehensive, step-by-step model for the synthesis of nZVI particles. The process relies on the reduction of hydrated Fe³⁺ ions using NaBH₄​. Each stage is clearly depicted with its approximate relative dimensions and molecular structures. Initially, a Fe³⁺ ion with a 0.129 nm radius is dissolved in water. It becomes enveloped by surrounding H₂O molecules, each having a 0.282 nm radius, to form a hydrated Fe³⁺ structure with a total radius of 0.900 nm. This stable structure represents the initial state of the ion in an aqueous solution. Subsequently, upon the addition of the powerful reducing agent NaBH₄​ to the solution, H₂ gas formation is initiated. A key factor in this process is the exceptionally small 0.05 nm radius of a hydrogen atom. This minuscule size allows it to effortlessly permeate the gaps in the hydration shell to reach the central Fe³⁺ ions. They then directly reduce the Fe³⁺ ions to elemental iron atoms. As a direct result of this reduction, the newly formed elemental iron atoms rapidly agglomerate, giving rise to nZVI particles. This agglomerated structure is observed to have diameters ranging from 11.73 nm to 25.32 nm. These nZVI particles are approximately 60 to 190 times larger than the initial Fe³⁺ ions. To confirm the effectiveness of the synthesis procedure and the deoxygenation protocol, the freshly synthesized particles were characterized by XRD analysis (Fig. 3a). The XRD pattern exhibits a sharp and dominant diffraction peak at = 44.7°, which corresponds to the (110) plane of body-centered cubic (bcc) metallic iron (Fe⁰). Additionally, the secondary peak observed at = 65.1° is attributed to the (200) plane of Fe⁰, further confirming the crystalline metallic structure. The absence of significant characteristic peaks for iron oxides—such as magnetite at 35.4° or hematite at 33.1°—clearly demonstrates that the nZVI was successfully protected from oxidation during the synthesis stage. Furthermore, the crystalline nature of the synthesized particles allows for the determination of the average crystallite size () through the broadening of these diffraction peaks using the Scherrer equation:

where is the shape factor (0.9), is the X-ray wavelength (0.15406 nm for ), is the full width at half maximum (FWHM) of the (110) peak in radians, and is the Bragg angle. Based on the FWHM of the donant peak at = 44.68°, the average crystallite size was determined to be approximately 14.3 nm. The SEM analysis (Fig. 4B) provides strong visual confirmation of these XRD results. The direct measurements of individual particles primarily cluster around 14 nm (e.g., 13.49 nm and 14.55 nm), which shows a remarkable correlation with the calculated average crystallite size. The slightly larger diameters observed (up to 25.32 nm) are attributed to the clustering of these primary nanocrystals, consistent with the fact that the Scherrer equation identifies the size of coherent scattering domains rather than larger physical agglomerates. To further elucidate the porous nature of these agglomerates, N₂ adsorption–desorption (BET) analysis was performed. The synthesized nZVI exhibited a BET specific surface area of 44.76 ± 0.4 m²g⁻¹ and a total pore volume of 0.081 cm³g⁻¹. According to the Barrett-Joyner-Halenda (BJH)³² model, the average pore diameter was determined to be 7.25 nm, confirming a predominantly mesoporous structure. The large and small pores (100–400 nm) formed by this agglomeration are more clearly visible in Fig. 4B. These pores, supported by the calculated mesoporous diameter (7.25 nm), appear to be large enough for the RY-145 dye (with a Topological Polar Surface Area of 457 Ų) to penetrate. Additionally, the surface area of 44.7 ± 0.4 m²g⁻¹ is sufficiently large for the adsorption of the RY-145 dye on nZVI. Moreover, Fig. 5b and 5c provide evidence that RY-145 was adsorbed onto the nZVI. Figure 5b and 5c show the IR spectra of RY-145 and the nZVI after being added to the RY-145 solution for 5 min, respectively. The transmittance of the RY-145 chromophore groups is visible in the 500–1700 cm⁻¹ wavenumber range. In the IR spectrum of nZVI interacted with RY-145, slight shifts in the wavenumbers of these groups can be seen. This is an expected observation, as interactions on the nZVI surface can affect the wavenumber values. This situation indicates that RY-145 was adsorbed onto the nZVI surface; in other words, the nZVI surface was dyed with RY-145. The adsorption of RY-145 onto the nZVI surface raised the suspicion that the decolorization process could be taking place on the nZVI surface. This also hints that nZVI can simultaneously act as both a reducing agent and a catalyst. To test this situation, hydrogen gas was passed through a 0.10 mM RY-145 buffer solution (in the absence of nZVI) for 20 min, but no change was observed in the UV/Vis absorption spectrum of RY-145 (Fig. 6). This result serves as additional proof of nZVI’s catalytic effect in adding hydrogen to double bonds^16,33.

Fig. 7 shows the plot of the hydrogen gas volume generated versus the amount (in mmol) of standard Fe powder. Two lines appear in this figure: a theoretical plot, which represents the hydrogen gas volumes calculated stoichiometrically and using the ideal gas equation, and a second line, which represents the volumes measured via water displacement. The hydrogen gas volumes were corrected and plotted using parameters such as the ambient temperature (27 °C), the water vapor pressure at this temperature (26.7 mmHg³⁴), and the altitude of the working environment (23 m, atmospheric pressure 758.01 mmHg). Based on this plot, a correction factor of 1.256 was calculated, meaning the hydrogen gas recovery is about 80%. The experimental method is not in significant disagreement with the theoretical value and can therefore be used for calibration. Since the same losses will be accounted for when determining the potential hydrogen gas reservoir of 0.50 g of nZVI, the amount of iron (in mmol) can be calculated accurately from the produced hydrogen gas volume. Finally, the potential hydrogen gas reservoir for 0.50 g of nZVI was measured to be 161 cm³ (corrected value 155 cm³) and was calculated as 6.28 mmol. At the beginning of the decolorization reaction, the amount of RY-145 in the 500 cm³ reaction medium was 0.0125 mmol for a concentration of 0.025 mM. The RY-145 molecule contains 22 double bonds, such as C=C, C=N, and N=N, in its conjugate system, and the amount of these double bonds was calculated to be 0.275 mmol. These double bonds are potential functional groups that can undergo addition reactions with the hydrogen molecules formed in the reaction medium¹⁸. Over the course of the reaction, the amount of hydrogen molecules produced was about to 28 times (mol/mol) the amount of dye at the lowest concentration and 1.90 times the amount of dye at the highest concentration.

Fig. 7.

The plot of the volume of hydrogen gas generated versus the amount of standard Fe powder.

On the other hand, the amounts of Fe²⁺ ions dissolved in the solution were monitored during the 120-minute reaction time, in the absence of RY-145, and the results are shown in Fig. 8. According to the figure, 3.24 mmol of dissolved Fe²⁺ was generated after the 120-minute reaction. This amount must correspond to a decrease in the amount of nZVI. Fig. 5d clearly shows that an Fe-O bond formed on the nZVI surface before it was completely oxidized and dissolved to yield Fe²⁺ ions, for example, after a 60-minute reaction time. The 574.08 cm⁻¹ wavenumber is specific to the Fe-O bond³⁵. Additionally, the wavenumbers corresponding to the chromophore groups almost completely disappeared.

Fig. 8.

Changes in nZVI and Fe²⁺ during the decolorization reaction in the absence of RY-145.

To verify the “oxidative dissolution” hypothesis and understand the structural evolution of nZVI during the decolorization of RY145, XRD patterns were recorded before and after the reaction (Fig. 3). The spent nZVI sample was obtained after 30 min of contact with the dye solution. Analysis of Pristine nZVI (Before Reaction). The XRD pattern of the fresh nZVI (Fig. 3a) is characterized by a dominant and sharp diffraction peak at 2 = 44.68°, which is indexed to the (110) plane of body-centered cubic (bcc) metallic iron ( Fe⁰ The high intensity and narrow width of this peak confirm the high crystallinity and metallic purity of the synthesized nanoparticles. A minor peak observed at 35.58° suggests the presence of a negligible pre-existing oxide layer (magnetite/maghemite) formed during the synthesis or handling stage. Post-reaction analysis of the spent nZVI (Fig. 3b) revealed significant structural transformations, offering a detailed insight into the chemical evolution and oxidation state of the nanoparticles following RY145 treatment.:

Depletion of the Fe⁰ core: The intensity of the characteristic (110) peak at 44.68° markedly decreased compared to the pristine sample. This reduction confirms the oxidative dissolution of the metallic iron core as it serves as a primary electron donor for the reductive cleavage of the dye’s azo bonds.

Identification of oxidation products: The emergence of new distinct peaks at 30.20°, 35.58° and 62.77° corresponds to the (220), (311), and (440) crystal planes of spinel iron oxides, specifically Magnetite (Fe₃O₄) or Maghemite ( Fe₂O₃) (JCPDS 19–0629³⁶).

Amorphous oxyhydroxides: The broad hump appearing at low angles (2 = 9–15°) indicates the formation of poorly crystalline or amorphous Iron Oxyhydroxides (FeOOH). These hydrated phases are typical corrosion products in aqueous media and play a vital role in the removal process through surface adsorption and complexation with RY145 molecules. Finally, the comparison between the two patterns illustrates a clear transition from a purely metallic state to a multi-phase core–shell structure. The shift from a single Fe⁰ peak to multiple oxide and hydroxide phases demonstrates that the removal of RY145 is not merely a surface phenomenon but a synergistic process. It involves the consumption of the iron core for reductive degradation and the subsequent utilization of the newly formed corrosion products for the adsorptive entrapment of the dye.

The effects of pH and the amount of nZVI on decolorization of RY-145 dye

Decolorization experiments were performed using a 0.10 mM solution of RY-145 dye and 0.500 g of nZVI in a acetate buffer. The reactions were conducted at pH 3.0, 7.0, and 11.0 for 120 minutes. The results showed that a decomposition rate of 96.4% was achieved at pH 3.0³⁷. Under neutral and alkaline conditions, the decolorization rates were 84.4% and 75.6%, respectively (Fig. 9). The amount of hydrogen molecules produced is inversely proportional to the pH of the medium, as hydrogen production increases with increasing acidity However, as shown in Fig. 6, hydrogen molecules alone are not sufficient for the addition reaction that leads to the saturation of the compound’s double bond. The reaction of a double bond with a hydrogen molecule is a well-known addition reaction that occurs on a catalyst surface. The proposed reduction mechanism involves the initial adsorption of RY-145 onto the nZVI surface. Subsequently, hydronium ions react with the nZVI surface to produce hydrogen molecules. Thus, the production of hydrogen molecules occurs simultaneously with the hydrogen addition reaction.

Fig. 9.

Decolorization of RY-145 at various pH values ([RY-145]₀ = 0.100 mM).

On the other hand, at pH 3, the decolorization of RY-145 increased with the amount of nZVI (Fig. 9). Decolorization percentages of 60.1%, 96.4%, and 99.4% were achieved for 0.100 mM RY-145 at 0.2, 0.5, and 1.0 g nZVI, respectively. Increasing the amount of nZVI results in an increase in two important parameters: the generation of hydrogen molecules and the adsorption surface area. This situation causes the decolorization rate of RY-145 to increase. The BET surface area of nZVI was determined to be 44.7 m²/g. Based on this value, the surface areas for 0.200 g and 0.500 g of nZVI can be easily calculated as 8.94 m² and 22.35 m², respectively. Similarly, since the molecular hydrogen reservoir of 0.50 g of nZVI was determined to be 6.28 mmol from Fig. 7, the hydrogen reservoirs for 0.20 g and 1.00 g of nZVI can also be calculated as 2.51 mmol and 12.56 mmol, respectively. As a result, the increase in the decolorization rate of RY-145 with the increase of these two parameters is compatible with scientific principles.

To further highlight the significance of the present study, the performance and structural properties of the synthesized nZVI were compared with relevant literature. While several studies have focused on nZVI-based remediation, the crystallite size often remains a limiting factor. For instance, Tarekegn et al.¹² reported nZVI particles with a significantly larger size for Methylene Blue removal, whereas our synthesis yielded much smaller crystallites (8.10 nm), providing a higher density of active surface sites. Although green synthesis routes, such as the one demonstrated by Abdelfatah et al.³⁸, offer eco-friendly alternatives, they often result in lower crystallinity compared to the well-defined bcc metallic iron ( Fe⁰) structure achieved in this work. Furthermore, our simplified synthesis route achieved a 96.4% removal efficiency for Reactive Yellow 145, which is highly competitive with complex modified systems like the MOF-derived N-doped nZVI@C rods reported by Ye et al.⁹. These comparisons underscore that the proposed nZVI provides a robust and efficient platform for the treatment of reactive dyes in aqueous solutions.

The effects of initial RY-145 concentration on decolorization

It was observed that the more dilute the initial solution, the faster the decolorization of RY-145 in catalytic reactions (Fig. 10). As shown in Fig. 10, only 58.9% of RY-145 was decolorized after 120 minutes with an initial concentration of 0.300 mM, whereas 100% decolorization was achieved in the same time frame when the initial concentration was 0.025 mM. An initial concentration of 0.100 mM resulted in a 92.4% degradation of the dye. While the total amount of decolorized RY-145 increased with a higher initial concentration, an inverse relationship was observed between the initial concentration and the decolorization rate³⁸. This is because at lower dye concentrations, the reactive sites on the nZVI surface are more readily available, leading to a higher decolorization level. Over the 120-minute decolorization reaction, 3.24 mmol of Fe²⁺ was formed, with the majority of the reaction completed within the first 20 minutes. The amount of Fe²⁺ formed in the first 20 minutes was measured to be 2.50 mmol. Stoichiometrically, the molar amount of hydrogen molecules produced is 1.0 times the amount of oxidized iron. Thus, 3.24 mmol of hydrogen was produced over 120 minutes, with 2.50 mmol produced in just the first 20 minutes.

Fig. 10.

Decolorization of RY-145 at various dosage of nZVI ([RY-145]₀ = 0.100 mM).

On the other hand, at the 20-min mark, the decolorization ratios were 92.12%, 87.60%, 80.70%, 57.56%, and 42.12% for initial dye concentrations of 0.025, 0.050, 0.100, 0.200, and 0.300 mM, respectively (Fig. 11). The corresponding amounts of reduced double bonds for these rates were 0.253, 0.482, 0.887, 1.266, and 1.390 mmol. These results indicate that the amount of hydrogen molecules produced in the reaction medium during the first 20 min is significantly greater than the amount of reduced double bonds. The fact that fewer double bonds are reduced despite the excess of hydrogen molecules can be explained by a decrease in the nZVI’s surface area. Initially, the reaction medium contained 8.93 mmol of nZVI. After 20 min, 6.43 mmol remained, meaning that approximately 30% of the nZVI had dissolved. Although there isn’t a direct one-to-one relationship between the amount of nZVI and its surface area, the reduction in nZVI amount undeniably led to a decrease in the total surface area. Consequently, while the rate and efficiency of RY-145 decolorization have an inverse relationship with increasing dye concentration, the amount of double bonds saturated by hydrogen molecules increased as the RY-145 concentration rose.

Fig. 11.

Decolorization of RY-145 at various concentrationof RY-145 (pH = 3, nZVI = 0.50 g).

The adsorption of RY-145 on nZVI

The Langmuir equation is a well-known and widely applied mathematical approach that describes the nature of adsorption. It has been successfully used, especially in heterogeneous systems, and is still in use today. The approach is based on the relationship between the rates of desorption and adsorption. The theory of the Langmuir equation can be briefly described by the following linear form:

1

Here: : is the equilibrium concentration of RY-145 after 120 min.

: is the adsorbed RY-145 concentration on nZVI surface.

and : are isotherm constants for particular solute–solvent combination.

The maximum adsorption capacity can be obtained from the reciprocal of the slope of the linearized Langmuir plot. In this section, the adsorption of RY-145 on nZVI was investigated, with the results presented in Fig. 12. Figure 12 shows the decrease in RY-145 concentration at five different initial concentrations after interaction with nZVI. At a pH of 3, the adsorption was successfully fitted to the Langmuir adsorption isotherm model. The regression coefficient was found to be 0.982 (Table 1). To further evaluate the surface characteristics, the Freundlich isotherm model was also applied (Fig. 12B). The model assumes that the adsorption energy decreases exponentially with the completion of the adsorption sites. The linear form of the model is expressed as

2

Fig. 12.

The Langmuir (A) and Froundlich (B) adsorption isotherms for various equilibrium concentration of RY-145. Initial concentrations are 0.025—0.300 mM.

Table.1.

Some constants of RY-145 in Langmuir isotherm and kinetic models.

Langmuir adsorption constants			Langmuir–Hinshelwood rate constant		The pseudo-first-order Lagergren rate constant	The pseudo-second-order rate equation constants		The pseudo-second-order rate constants
KL (g−1)	(mM−1)	(mM.g−1)	(min−1)	(mM.min−1)	(min−1)	(mM.g−1.min−1)	(mM)	mM⁻1 g⁻1 min⁻1
71.4	201.6	0.354	0.0621 (r2 = 0.988)	7.12 × 10–4 (r2 = 0.920)	0.0020 (r2 = 0.532)	0.0341 (r2 = 0.999)	0.0506	1.0100 (r2 = 1.000) −0.1993 (- r2 = 0.994)
			0.0189 (r2 = 0.972)		0.0024 (r2 = 0.748)	0.0412 (r2 = 0.998)	0.0990
			0.0112 (r2 = 0.880)		0.0030 (r2 = 0.708)	0.0669 (r2 = 0.990)	0.1874
			0.0033 (r2 = 0.966)		0.0046 (r2 = 0.985)	0.0564 (r2 = 0.990)	0.2815
			0.0027 (r2 = 0.966)		0.0114 (r2 = 0.905)	0.0484 (r2 = 0.990)	0.3571

where, : is the amount of RY-145 adsorbed per unit mass of nZVI (mgg⁻¹), : is the equilibrium concentration of the dye (mgL⁻¹), : is the Freundlich constant related to the adsorption capacity, : is the heterogeneity factor (adsorption intensity). The Freundlich constant () was determined to be 3.077. Since > 1 (specifically 2 < < 10), the adsorption of RY-145 onto nZVI is considered highly favorable, reflecting a robust affinity. Although the Langmuir model showed a slightly better fit, the high correlation coefficient of the Freundlich model (r² = 0.9663) provides evidence for the multi-energetic nature of the surface sites during the sacrificial reaction process. The fact that 1/ (0.324) is well below unity indicates a strong bond between the RY-145 molecules and the nZVI surface, confirming that the surface remains highly active for dye localization even during the sacrificial dissolution of the iron core.

It has been discussed that RY-145 molecules must be adsorbed onto the nZVI surface before reduction can occur. In this decolorization reaction, the concentration of RY-145 is determined by subtracting the equilibrium concentration at time "t" from the initial concentration. However, it is possible that not all adsorbed molecules remain on the surface; some may be reduced and return to the solution as colorless products. Furthermore, the continuous dissolution of nZVI into the solution reduces the available surface area over time. Despite these constantly changing parameters, the fundamental constant is that the dye molecules must first interact with the surface to be reduced. Therefore, even if they later detach as a result of reduction, they decrease the concentration of RY-145 in the solution as if they were adsorbed. This conceptual framework allows the Langmuir equation to remain applicable to this dynamic system.

According to Fig. 12A, most of the RY-145 adsorption onto the nZVI catalyst occurred within 20 min. This rapid adsorption process can be described by a film diffusion model³⁹. The Langmuir isotherm constant for RY-145 was found to be 71.43 g⁻¹. The essential characteristic of the Langmuir equation can be expressed in terms of the dimensionless separation factor, R_L​, which was defined by Hall et al.,1966⁴⁰ as:

3

where a_L is the Langmuir isotherm constant and C₀​ is the highest initial RY-145 concentration (in mM). The value of R_L indicates the nature of the isotherms: unfavorable (R_L > 1), linear (R_L = 1), favorable (0 < R_L < 1), and irreversible (R_L = 0). The adsorption of RY-145 on nZVI was determined to be favorable, with an R_L value of 0.016.

However, a stable surface doesn’t exist because the nZVI surface continuously decreases due to dissolution. Despite this, the results were successfully fitted to the adsorption isotherm. In a typical adsorption process, the surface is stable and the amount of adsorbent does not decrease; only the adsorbent’s surface is covered by the adsorbate. In the interaction between nZVI and RY-145, both the surface and the adsorbate decrease. A total of 3.24 mmol of iron dissolved in the reaction solution, which corresponds to approximately 30% of the total iron. Consequently, the nZVI surface area has also decreased proportionally. Given these factors, a true adsorption process based on the Langmuir approach is not occurring throughout the entire process. Therefore, the term "pseudo-adsorption" is more appropriate due to the surface’s instability caused by continuous dissolution.

Adsorption/reduction mechanism

As illustrated in Fig. 13, the decolorization of RY-145 follows a pseudo-adsorption framework. To further validate the proposed reductive mechanism and rule out alternative pathways, additional diagnostic tests were performed:

• i)Investigation of the Oxidation Pathway: To evaluate the presence of reactive oxygen species (ROS), an iodide (I^-) oxidation test was conducted. In the presence of strong oxidants such as hydroxyl radicals (•OH), I^- is typically oxidized to I₂, resulting in a characteristic brown color. However, no color change was observed during the 30-min reaction, suggesting that an oxidative pathway was not the dominant mechanism under the studied experimental conditions.
• ii)Assessment of Coagulation/Precipitation: We investigated whether decolorization resulted from the formation of insoluble dye-metal complexes with Fe²⁺/Fe³⁺ ions. Throughout the process, the solution remained homogeneous with no detectable precipitate or floc formation. Furthermore, the FTIR spectrum (Fig. 5d) of the post-reaction mixture showed no characteristic peaks of the dye’s functional groups, which would have been present if even imperceptible precipitation had occurred.
• iii)Reductive Cleavage (The Dominant Pathway): Based on the systematic exclusion of the aforementioned pathways, it is concluded that RY-145 is primarily degraded via reductive cleavage. The Fe⁰ core of the nZVI acts as a potent electron donor, targeting the azo linkages (—N = N—) of the RY-145 molecule. This electron transfer breaks the chromophore, leading to the formation of colorless aromatic amines, which is consistent with the observed rapid decolorization.

Fig. 13.

Mechanistic illustration of the pseudo-adsorption and decolorization of RY-145 using nZVI at pH 3.0.

Initially, the dye anions are sequestered onto the positively charged oxide/hydroxide layer via electrostatic attractions and hydrogen bonding. Following this enrichment on the surface, the nZVI-mediated sacrificial reduction takes place, where the electron transfer from the Fe⁰ core leads to the cleavage of the C=C double bonds, effectively achieving decolorization. This reduction mechanism, involving electron transfer and nascent hydrogen production, is a well-established pathway for organic pollutant degradation via nZVI-based systems⁴¹.

The catalytic kinetic of decolorization of RY-145 by nZVI

Catalytic decolorization kinetic studies for RY-145 concentrations ranging from 0.025 to 0.300 mM were conducted under reaction conditions of pH 3 and with 0.500 g of nZVI. The Langmuir–Hinshelwood rate expression is successfully used for heterogeneous catalytic degradation to determine the relationship between the initial degradation rate and the initial concentration of the organic substrate (Eq. 4)^23,24,26.

4

Here, [RY145]₀ represents the initial concentration of RY-145 (mM), K_RY145 ​ is the Langmuir–Hinshelwood adsorption equilibrium constant (mM⁻¹), and k is the rate constant of the surface reaction (mMmin⁻¹). The apparent rate constant (k_app) values for each initial concentration were determined from the slopes of the straight lines obtained by plotting − ln [RY-145]/[RY-145]₀ versus reaction time (Fig. 14). The k_app values were found to be 0.0620, 0.0190, 0.010, 0.0030, and 0.0027 min⁻¹ for RY-145 concentrations of 0.025, 0.050, 0.100, 0.200, and 0.300 mM, respectively (see Table. 1). By plotting the initial concentrations against 1/k_app​, the reaction rate constant and the adsorption constant can be calculated. In this type of calculation, a second-order rate constant of 7.12 × 10⁻⁴ mMmin⁻¹ was found, but the regression coefficient was a value lower than 0.950 (Fig. 15). As can be seen in Table. 1, the regression coefficients for the determination of k_app​ all have values greater than 0.950, with the exception of one (0.888), while the regression coefficient for the determination of k′′ was observed to be 0.920. These results indicate that the Langmuir–Hinshelwood rate model is not suitable for explaining the decolorization rate.

Fig. 14.

Linear transform -ln [RY145]/[RY145]₀ vs. time for RY-145.

Fig. 15.

Determination of the adsorption equilibrium constants and degradation rate constants for RY-145.

On the other hand, although the pseudo-first-order Lagergren model⁴² (Eq. 5) describes the adsorption rate, in this study, the adsorption rate can be evaluated as the decolorization rate. The linear plots obtained from the pseudo-first-order model are illustrated in Fig. 16.

5

Fig. 16.

Pseudo first-order adsorption kinetics of RY-145.

In this equation:

: is the adsorbed dye concentration on the nZVI surface at any given time, t (mMg⁻¹).

:​ is the dye concentration adsorbed at equilibrium (mMg⁻¹).

k′ : is the pseudo-first-order rate constant (min⁻¹).

The data from this study unfortunately did not yield a stable regression coefficient when fitted to Eq. 5. However, a strong fit was observed using the pseudo-second-order rate model⁴³ (Eq. 6), which resulted in highly successful regression coefficients (Fig. 17).

6

Fig. 17.

Pseudo second-order adsorption kinetics of RY-145.

Here, is the initial adsorption rate constant for the pseudo-second-order kinetic model (mM g⁻¹ min⁻¹). According to Eq.6, when you plot ​ versus , the inverse of the slope gives you the ​ value, and the inverse of the y-intercept gives you the value. The results can be seen in Table 1. can be written as a function of as shown below.

7

Here, (mM⁻¹ g⁻¹ min⁻¹) is the pseudo-second-order rate constant. is the remaining concentration of RY-145 in the solution at equilibrium, and ​ is the amount of RY-145 adsorbed at time t. When the values are plotted against the ​ values, the slope should be ​ (Fig. 18). As can be seen in Fig. 18, contrary to expectations, two lines are formed instead of a single one, with one exhibiting a negative slope. In traditional kinetic models, a negative rate is typically considered physically inconsistent. However, in this sacrificial system, this observation reflects an ‘apparent’ negative rate, serving as a direct manifestation of the dynamic and sacrificial nature of the nZVI surface. This result suggests that the adsorption rate slows down significantly after the value of 0.0351 mM² at the intersection point. Regarding the kinetic behavior, these negative rate tendencies demonstrate that the pseudo-second-order model cannot be applied to nZVI, as it assumes a stable surface area which, in this case, is continuously decreasing. As evidenced by the intraparticle diffusion plots (Fig. 19), RY-145 molecules successfully penetrate and are trapped within the internal pore structure of the nanoparticles. As the reaction proceeds, the metallic iron core (Fe⁰) undergoes significant 'oxidative dissolution,' with experimental data (Fig. 8) indicating that approximately 36% of the iron content is dissolved⁴⁴. This structural erosion of the nZVI framework leads to the collapse of the pores where dye molecules had previously been sequestered (pseudo-adsorption). Consequently, these unreacted dye molecules are re-released into the bulk solution as their hosting matrix dissolves, surpassing the reduction rate at this specific stage. Furthermore, this dissolution reduces the reactive Fe⁰ mass and triggers surface passivation, leading to a decline in nascent hydrogen production. According to the theoretical justifications for negative-order dynamics in heterogeneous systems, this rise in concentration is a physical manifestation of matrix erosion, confirming the synergistic 'pseudo-adsorption–catalytic reduction’ mechanism. Although the pseudo-second-order equation has its shortcomings here, the intersection points and the subsequent negative rate provide a clear indication that a diffusion-dominant process has begun.

Fig. 18.

The calculation of dimensional rate constant of RY-145 depending on pseudo second-order adsorption kinetics constants.

Fig. 19.

Intraparticle diffusion kinetics for various RY-145 concentrations at pH 3.

Moreover, according to Fig. 4, nZVI has a numerous pore structure; therefore, the intra-particle diffusion theory as shown below was also applied to the data, and the results were given in Fig. 19.

8

Here, k_int is the intraparticle diffusion rate constant (mM.g⁻¹min^1/2), is the amounts of adsorbed RY-145 concentrations on the adsorbent (mMg⁻¹) at time t. k_int was found to be 4.8 × 10^–4, 10.7 × 10^–4, 23.5 × 10^–4, 37.4 × 10^–4, and 73.5 × 10⁻⁴mMg⁻¹min^1/2 for RY-145 concentrations of 0.025, 0.050, 0.100, 0.200, and 0.300 mM, respectively. With the exception of the 73.5 × 10^–4mMg⁻¹min^1/2 diffusion rate, the other obtained rates are not linearly compatible due to a regression coefficient lower than 0.950. As the initial concentration of RY-145 increased, k_int increased, depending on the chemical potential difference. At dilute RY-145 concentrations, the reduction rate of RY-145 by hydrogen molecules can be higher than k_int. Additionally, due to the continuous dissolution of nZVI in the reaction medium, the number of pores decreased, causing k_int to deviate from linearity. At higher RY-145 concentrations, the diffusion rate, which depends on the chemical potential, began to be more dominant than the hydrogen reduction rate. As a result, the regression coefficient started to approach a value of 1.

Assessment of adsorbent fate and potential for regeneration

The sustainability of the nZVI-based treatment process was evaluated by monitoring the iron dissolution behavior. Under optimized conditions, approximately 36% (3.24 mmol) of the initial 0.50 g iron dose (8.95 mmol Fe⁰) was found to dissolve into Fe²⁺ ions after 120 min. This substantial dissolution confirms that nZVI operates as a sacrificial electron donor, where the loss of metallic mass is directly proportional to the generation of the active hydrogen gas reservoir (155 cm³) required for the rapid reduction of the RY-145 dye. While this consumption limits traditional reusability of the pristine material, the system offers significant potential for a closed-loop cycle through chemical recovery. Therefore, the focus of this study remained on characterizing the sacrificial reaction kinetics and the pseudo-adsorption framework, rather than traditional reusability, due to the intentional consumption of the reactive metallic core. The final transformation of nZVI into stable iron oxides (magnetite/goethite) ensures that no toxic residues are released into the environment, and their magnetic nature allows for easy recovery, preventing any unintentional release of nanomaterials.

Recent studies by Adachi et al. present evolving strategies for catalyst stability and reuse. In their initial work, high degradation efficiency was achieved using iron-doped hydrochars, though the study focused primarily on process optimization³⁰. In a subsequent study, the team utilized thermal calcination at 400 °C to restore bimetallic catalyst activity²⁹. Most recently, it has been demonstrated that chemical re-reduction provides a sophisticated alternative; Adachi et al. (2025)⁴⁵ showed that NaBH₄ can effectively etch passivation (oxide) layers on nZVI surfaces and reduce dissolved Fe²⁺ ions back to the metallic Fe⁰ state, even in oxygenated environments.

Our study suggests that for nZVI systems, this chemical re-reduction using NaBH₄ provides a more targeted and less energy-intensive alternative to thermal methods. This dual regeneration approach—recovering both the magnetic solid core and the leached aqueous iron—ensures that the iron content is not wasted but rather “recharged” for subsequent high-performance cycles, significantly enhancing the overall economic viability and sustainability of the process.

Conclusion

In this study, the potential for removing Reactive Yellow 145 (RY-145) dye using nano-zero-valent iron (nZVI) particles was investigated. The synthesized nZVI particles exhibited a porous, agglomerated structure that facilitates dye access to the active sites. The process achieved a remarkable maximum decolorization efficiency of 96.4% within 120 min under acidic conditions (pH 3.0). This high efficiency is directly linked to the generation of an active hydrogen gas reservoir (calculated as 155 cm³), which drives the reductive degradation of the dye molecules.

The decolorization mechanism was found to be a synergistic process beginning with the adsorption of RY-145 onto the nZVI surface, followed by simultaneous reduction. Although the data showed a good fit with the Langmuir isotherm model, the continuous dissolution of the nZVI—where 36.2% (3.24 mmol) of the iron mass was consumed during the reaction—led to the conclusion that "pseudo-adsorption" is a more accurate term for this sacrificial process. Kinetic analyses revealed that while the pseudo-second-order model is traditionally used, it exhibits deficiencies in describing nZVI systems due to the constantly decreasing surface area. The formation of dual kinetic lines and the transition to negative rate values in Fig. 18 provide significant evidence that the process shifts from a surface-reaction-controlled stage to a diffusion-dominant stage.

In light of these findings, nZVI is proven to be an effective and promising material for treating azo dyes in low-pH industrial wastewater. For future research, several directions are suggested to improve the sustainability of this process:

1. Closed-Loop Regeneration: Future studies should focus on the chemical re-reduction of spent nZVI residues and dissolved Fe²⁺ ions using NaBH₄, as suggested by recent literature⁴⁵, to establish a cost-effective material cycle.

2. Surface Stabilization: Developing bimetallic systems or carbon-supported nZVI composites could minimize the rapid sacrificial dissolution, thereby extending the catalyst’s lifetime.

3. Scale-up Applications: Transitioning from batch studies to continuous flow-through packed-bed reactors is recommended to evaluate the long-term performance and pressure drop characteristics for industrial-scale applications.

4. While this study provides a comprehensive understanding of the reaction mechanism and the oxidative fate of nZVI during RY145 removal, further research exploring the influence of coexisting ions (such as , , and ) in complex textile wastewater matrices will be essential for the practical and large-scale application of this technology.

Author contributions

**Erdal Kuşvuran** : Writing—original draft, Visualization, Software, Resources, Investigation, Data curation. **Fatma Midik** : Writing—original draft, Investigation, Formal analysis, Data curation. **Sermin Gul** : Conceptualization and methodology.

Data availability

The data that support the findings of this study are available from the corresponding author, **Erdal Kusvuran**, upon reasonable request. E-mail: **ekusvuran@cu.edu.tr**.

Declaration

Competing interests

The authors declare no competing interests.