The Late Eocene to Early Oligocene age Aksitero formation is the oldest sedimentary formation in the West Central Luzon basin of the Philippines, which serves as the sedimentary capping to the Zambales ophiolite (Yumul & Dimalanta, Reference Yumul and Dimalanta1997), a complete sequence of oceanic crust and upper-mantle rocks. It is situated close to Tarlac, Zambales and Pangasinan areas and has two recognizable members: the Bigbiga Limestone member and the Burgos member. The Bigbiga Limestone member is in the lower section of the Aksitero formation and consists of micritic limestones intercalated with tuffaceous turbidites. The Burgos member, on the other hand, occurs in the upper section of the Aksitero formation, consisting of interbedded limestone, calcareous–tuffaceous turbidity sandstones, siltstones and mudstones. Petrographic analyses of the limestones and turbidites in the Aksitero formation were conducted to determine the origin and diagenetic processes undergone by the sedimentary units. The hemipelagic limestones predominantly contain fragments of planktonic foraminifera and nanofossils, with evidence of replacements of the original micritic tests by cristobalite, zeolite (clinoptilolite and mordenite) and smectite, which suggest volcanic origins. The tuffaceous turbidites, on the other hand, have graded textures and beds showing complete Bouma sequences, and thus signify transport and re-deposition from turbidity currents (Garrison et al., Reference Garrison, Espiritu, Horan and Mack1979).

Exposed portions of the Burgos member of the Aksitero formation were also investigated in the SAILE mine located in Mangatarem, Pangasinan. SAILE Industries, Inc., is the only commercial producer of natural zeolite and bentonite mineral products in the Philippines with a Mineral Production Sharing Agreement issued by the Mines and Geosciences Bureau under the Department of Environment and Natural Resources in the Philippines. The samples collected from the mine site show beds of zeolite (clinoptilolite), bentonite and silica occurring in the turbiditic layers. Zeolitized layers, consisting particularly of clinoptilolite and mordenite, are common, occurring as cement or as secondary mineral precipitates in voids. Zeolitization was caused by burial diagenesis at depths of ~3.5–4.0 km and geothermal gradients of 15–20°C km^–1 (Garrison et al., Reference Garrison, Espiritu, Horan and Mack1979). The possible depositional settings included submarine eruption and gravity flows aided by turbidity currents that eventually transported sediments away from the vent towards a deeper basinal area, where calcareous hemipelagic ooze dominated (Garrison et al., Reference Garrison, Espiritu, Horan and Mack1979).

An alternative mechanism related the zeolite genesis in the Aksitero formation with the hyper-alkaline Manleluag hot springs (pH 10.0–11.5) in Mangatarem, Pangasinan (Vargas et al., Reference Vargas, Pascua, Arcilla, Honrado, Alexander, Namiki, Fujii, Yamakawa, Sato and McKinley2009; Arcilla et al., Reference Arcilla, Pascua and Alexander2011). The high pH of the hot spring is due to the hydrolysis of Mg-rich rocks (gabbros) and Fe hydrolysis and oxidation of the pyroxenes present in the Zambales ophiolite. The hyper-alkaline waters from this reaction may have dissolved silica, which re-precipitated at lower temperatures, leading to zeolite formation (Vargas et al., Reference Vargas, Pascua, Arcilla, Honrado, Alexander, Namiki, Fujii, Yamakawa, Sato and McKinley2009; Arcilla et al., Reference Arcilla, Pascua and Alexander2011).

The Philippine natural zeolites (PNZs), consisting primarily of clinoptilolite and mordenite, are currently mined and used in applications including heavy metal removal (Cagomoc & Vasquez Jr, Reference Cagomoc and Vasquez2016; Olegario-Sanchez & Pelicano, Reference Olegario-Sanchez and Pelicano2017), nanotechnology (Olegario et al., Reference Olegario, Pelicano, Dahonog and Nakajima2019a, Reference Olegario, Pelicano, Felizco, Mendoza and Nakajima2019b), oil removal (Osonio & Olegario-Sanchez, Reference Osonio and Olegario-Sanchez2017), wastewater treatment (Barola et al., Reference Barola, Dusaban, Olegario-Sanchez and Mendoza2019; Dignos et al., Reference Dignos, Gabejan, Olegario-Sanchez and Mendoza2019) and, more recently, biomedical applications (Olegario-Sanchez & Felizco, Reference Olegario-Sanchez and Felizco2017; Olegario-Sanchez et al., Reference Olegario-Sanchez, Tan, Mendoza and Balela2017; Osonio & Vasquez Jr, Reference Osonio and Vasquez2018; Taaca & Vasquez Jr, Reference Taaca and Vasquez2018). Locally sourced natural zeolites remain largely unexploited in such applications due to variation in the purity and composition of the various deposits. In addition, the availability and versatility of synthetic zeolites have inhibited the commercialization of natural zeolites. However, the PNZs have already been utilized as raw materials for animal feeds, agricultural stocks and other industrial applications in the local market. Recently, they have been used for inactivation of Escherichia coli (Olegario-Sanchez et al., Reference Olegario-Sanchez, Tan, Mendoza and Balela2017) and as noble-metal catalytic converters (Guerrero et al., Reference Guerrero, Mendoza, Ong, Olegario-Sanchez and Ferrer2019) after modification with Cu. Nevertheless, both synthetic and natural zeolites have attracted growing interest regarding their use in various applications that involve ion-exchange processes.

To maximize the ion-exchange capacity of PNZ samples, preconditioning treatments should be performed. Preconditioning of zeolites has been carried out in the past to enhance their ion-exchange capacity (Ndayambaje, Reference Ndayambaje2011), although the methods applied may not be optimal for locally sourced zeolites. These modifications include acid treatment, de-alumination and de-cationization, ion exchange and thermal treatment. The processes remove cations from the zeolite structure and transform zeolites in a near-homoionic form with increased content of the target cation (Semmens et al., Reference Semmens and Martin1988; Ndayambaje, Reference Ndayambaje2011). Various conditioning techniques, however, may affect the ion-exchange behaviour of natural zeolites (Cerjan-Stefanović et al., Reference Cerjan-Stefanović, Ćurković and Filipan1996). The effects of NaCl, KCl and HCl were studied to evaluate the Cu²⁺ and Co²⁺ removal efficiency of South African clinoptilolite samples in column experiments (Mamba et al., Reference Mamba, Nyembe and Mulaba-Bafubiandi2010). The HCl-conditioned clinoptilolite was the most efficient in terms of metal removal (Mamba et al., Reference Mamba, Nyembe and Mulaba-Bafubiandi2009). In a similar study, Na-treated zeolites displayed excellent selectivity for Zn²⁺ followed by Mn²⁺, compared to unconditioned natural zeolites (Cerjan-Stefanović et al., Reference Cerjan-Stefanović, Ćurković and Filipan1996).

Among the conditioning agents, HCl has been reported to be an effective agent for improving the ion-exchange capacity of natural zeolites. In addition, HCl leaches Al³⁺ out of the clinoptilolite structure (Mamba et al., Reference Mamba, Nyembe and Mulaba-Bafubiandi2010). Moreover, Na treatment may replace certain cations on the zeolite framework, rendering it homoionic. The Na⁺ ions are easily exchanged in the zeolite structure, which greatly affects the exchange properties of zeolite (Semmens et al., Reference Semmens and Martin1988). Thus, the present work aims to observe the influence of a two-step preconditioning treatment of PNZ on the morphology, texture and structural properties of the PNZ samples. The PNZ samples were first acid conditioned using HCl, followed by Na⁺ exchange. For each step, the best-conditioned samples were used. The cation-exchange capacity and selectivity of the PNZ samples were assessed by ion exchange of PNZ with Cu²⁺ ions.

Experimental

Sample preparation

A total of 200 g of PNZ tuffaceous rock samples was obtained from the SAILE Mine in Mangatarem, Pangasinan. The samples were crushed and sieved to particle sizes of 1 and 3 mm. Granules were washed four times with deionized (DI) water and sun dried for 48 h. Further drying was performed prior to acid treatment by oven drying of the natural washed zeolites (NW-Z) at 150°C for 6 h.

Preconditioning of the PNZ samples

Acid pretreatment of the PNZ samples

HCl preconditioning treatment

The NW-Z samples were soaked in HCl solutions (20% w/v) (Sigma Aldrich). The NW-Z powders were soaked in 1 M and 3 M HCl concentrated solutions for 6 and 12 h. Acid-treated zeolite (HC-Z) samples were washed 12 times with DI water until the pH of the supernatant was 6–7. Chlorine strips were also used as indicators to ensure that the zeolites were Cl^– free.

Experimental design

The HC-Z samples were prepared by varying the preparation variables, namely HCl concentration, soaking time and particle size, represented as A, B and C, respectively, using a 2³ factorial experimental design. The experimental levels and the range of the variables are listed in Table 1. The Si/Al ratio calculated from atomic absorption spectrometry (AAS) analyses was selected as the response variable. Each factor was examined at two levels: 1 M and 3 M for A, 6 and 12 h for B and 1 and 3 mm for C. The design of the study had eight different experimental conditions. Table 2 presents the series of trials of the 2³ factorial design. A total of 16 runs were performed in this study, with each experimental condition conducted in duplicate. The results of the experiment were analysed with MINITAB software. The analysis of variance (ANOVA) results for this design are presented in Table 3.

Table 1. Factors and levels for the 2³ factorial experimental design.

Table 2. Trials of the 2³ factorial experimental design with the Si/Al ratio as the response variable.

Table 3. ANOVA results for the 2³ factorial experimental design.

Chemical conditioning of the acid-pretreated samples

Na⁺ exchange

Two sets of optimization conditions were used for the Na⁺ ion exchange of 10% w/v HC-Z samples. The first optimization condition is focused on determining the best NaCl (99% Salinas) concentration (1 M, 2 M and 4 M) for the preconditioning of HC-Z samples at room temperature for 24 h. The second optimization condition used the best constant NaCl concentration to determine the best soaking time between 3, 6, 12, 18, 24, 48 and 72 h. The Na-Z samples were thoroughly washed with DI water until Cl^– free (AgNO₃ test). The filtered Na-Z samples were dried at 150°C for 6 h.

Characterization

Optimization was conducted to ensure utilization of the best pretreated samples (HCl and NaCl conditioning steps) before the Cu²⁺ ion exchange. Thus, this study assessed the as-prepared sample (NW-Z), the optimized acid-pretreated sample (HC-Z) and the Na⁺-exchanged zeolite sample (Na-Z). The Si/Al ratio for the optimization of HCl pretreatment as well as the saturation percentage of the Na⁺ loading were determined via AAS (Shimadzu AA-7000) at the Institute of Chemistry of the University of the Philippines Diliman (UP Diliman). Calibration for the AAS analysis was performed using a suitable calibration curve. The porosity and specific surface area of the untreated PNZ and HC-Z samples were determined using the Brunauer–Emmett–Teller (BET) method using Micromeritics Tristar II Plus equipment. The equipment utilized CO₂ and N₂ isotherms to calculate the pore-size diameter of the materials. CO₂ was used for porosity measurement and N₂ was used for surface area determination at 273 and 77 K, respectively. The samples were degassed at 90°C and 350°C for 2 and 8 h, respectively, with a ramp rate of 10°C min^–1. Data analysis was performed using MicroActive version 2.02 software. The Cu²⁺ uptake of zeolite was determined using microwave plasma-atomic emission spectroscopy (Agilent 4200). The chemical functional groups were detected by Fourier-transform infrared (FTIR; Nicolet MAGNA-IR 560) spectroscopy with a diffuse reflectance spectroscopy accessory. The zeolite phases of the samples were identified by X-ray diffraction (XRD; Shimadzu XRD-7000) operated at 40 kV and 30 mA with Cu-Kα radiation in the range 2–80°2θ at the National Institute of Geological Sciences, UP Diliman. The surface morphology of the samples was determined with scanning electron microscopy (SEM) using a Shimadzu SSX-550 SEM equipped with an energy-dispersive spectrometer (EDX; JEOL JSM 5300) at the Engineering Department of Hokkaido University (Japan). The composition of the samples was determined using the EDX.

Results and discussion

Preconditioning of PNZ

Acid conditioning of PNZ

Statistical analysis

The results of the statistical analysis are shown in Table 2. The Si/Al ratios of the synthesized HC-Z samples ranged from 5.9 to 8.4. Table 3 summarizes the main results of the statistical analysis using ANOVA. Τhe HCl concentration and particle size affect significantly the Si/Al ratio of the zeolite samples because they have p-values of <0.05. Moreover, these factors do not significantly influence each other, as evidenced by the high p-values (>0.1) for all interactions. In contrast, the soaking time (factor B) has a p-value of >0.05. Hence, the optimum acid pretreatment condition to be used in this study is to soak NW-Z samples with a particle size of 1 mm in a 3 M HCl solution for 12 h.

FTIR analysis of NW-Z and HC-Z samples

FTIR spectroscopy was employed to determine the functional groups present in the zeolite samples, as well as to ascertain that the Al–Si (both represented as T) cage of the samples remains intact after acid preconditioning. Characteristic zeolite bands were observed in the 400–1800 cm^–1 range (Fig. 1) (Gligor et al., Reference Gligor, Maicaneanu and Walcarius2010). The strong bands at 455 and 1042 cm^–1 represent the O–T–O angular deformation and T–O asymmetric external stretching, respectively, of the NW-Z sample. Characteristic bands at 798, 1439 and 1620 cm^–1 correspond to the T–O external symmetric stretching, T–O asymmetric internal stretching and H–O–H angular deformation of adsorbed water, respectively. The broad band at 3400 cm^–1 is attributed to the O–H bond stretching of adsorbed water molecules in the NW-Z sample (Taaca & Vasquez Jr, Reference Taaca and Vasquez2017).

Fig. 1. FTIR spectra of the NW-Z and HC-Z samples.

The bands at 400–1200 cm^–1 were enlarged, and these are shown in Fig. 2. There are slight shifts in the characteristic bands of NW-Z after the acid preconditioning. The positions of the shifted bands are summarized in Table 4. This effect of the acid pretreatment may be an indication of degradation to the zeolite framework (Ndayambaje, Reference Ndayambaje2011).

Fig. 2. FTIR spectra of the NW-Z and HC-Z samples in the 1200–400 and 2200–1200 cm^–1 range.

Table 4. Characteristic FTIR bands (cm^–1) of the NW-Z and HC-Z samples.

Chemical composition of the zeolite samples

Table 5 lists the chemical compositions of the NW-Z and HC-Z samples obtained by EDX analysis. During acid pretreatment, most cations were leached from the HC-Z. The Si/Al ratios are 4.90 for the NW-Z samples and 7.48 for the HC-Z samples. The change in the cation and Si/Al content of the PNZs is due to proton attack at the Al sites. Al–OH or cationic Al–OM bonds can be a result of proton attack at the Al sites (Cakicioglu-Ozkan, Reference Cakicioglu-Ozkan2010). The possible reaction mechanisms that caused the de-alumination and affected the Si/Al ratio of the PNZs are described in equations 1–3. Aluminium dissolution in the PNZ samples occurred upon soaking in the HCl solution. This was associated with adsorption of protons at the cationic sites close to the Al sites and removal of any monovalent (M⁺) or bivalent (M²⁺) cations at the zeolite edges (Doula & Ionnou, Reference Doula and Ioannou2003).

(1)$${\rm Al}-{\rm OH} + {\rm H}_{\lpar {{\rm aq}} \rpar }^ + \to {\rm Al}-{\rm OH} + {\rm H}_{2\lpar {{\rm aq}} \rpar }\to {\rm Al}_{\lpar {{\rm aq}} \rpar }^{3 + } + {\rm H}_2{\rm O}$$

(2)$${\rm Al}-{\rm O}{\rm M}^ +{+} {\rm H}_{\lpar {{\rm aq}} \rpar }^ + \to {\rm Al}-{\rm OH} + {\rm M}^ + &#x00A0;&#x00A0;&#x00A0;&#x00A0;&#x00A0;&#x00A0;&#x00A0;\;$$

(3)$${\rm Al}-{\rm O}{\rm M}^{2 + } + 2{\rm H}_{\lpar {{\rm aq}} \rpar }^ + \to {\rm Al}-{\rm O}{\rm H}_2 + {\rm M}_{\lpar {{\rm aq}} \rpar }^{2 + }&#x00A0;&#x00A0;&#x00A0;&#x00A0;&#x00A0; $$

Table 5. SEM-EDX analyses of the composition (wt.%) of the NW-Z and HC-Z samples.

Specific surface area and pore-size distribution

The specific surface area and pore-size distribution of the zeolite samples before and after acid pretreatment were determined by BET analysis. The N₂ adsorption isotherms of the HC-Z samples are of type 1 according to the International Union of Pure and Applied Chemistry (IUPAC) classification. Changes in the specific surface area and porosity (pore volume and pore size) of the PNZ samples were detected using the adsorption isotherm data evaluated using the t-plot, Langmuir and Dubinin–Astakhov methods (Table 6). The specific surface area and pore volume of the HC-Z samples increased by 62% and 32%, respectively, relative to the PNZ samples. The increases in the micropore volume and specific surface area are attributed to the formation of secondary micropores that are larger than the main pore system (7–20 Å) due to the opening of pockets on the zeolite (Cakicioglu-Ozkan, Reference Cakicioglu-Ozkan2010; Hernandez, Reference Hernandez2014). The increased specific surface area and porosity of the HC-Z samples may be due to the increased open spaces in the zeolite framework. The SEM-EDX results show that most cations were removed or redistributed within the zeolite framework (Table 5).

Table 6. Specific surface area and porosity of the NW-Z and HC-Z samples.

The nature of the acid used affects the structure of zeolite because ligands (L) such as Cl^– ions form complexes on the zeolite surface. The Al or Si atoms (represented as T) on the zeolite surfaces are centres for the formation of outer-sphere complexes on the protonated zeolite surface (T–OH²⁺) with the ligands (L^n–), as shown in equation 4. The use of HCl led to adsorption on the external surface to form outer-sphere complexes that follow the ion-exchange reactions shown in equations 4 and 5. Equation 5 shows the formation of inner-sphere complexes on the protonated sites of zeolite (Doula & Ionnou, Reference Doula and Ioannou2003).

(4)$$\eqalign{{\rm T}-{\rm OH} + {\rm L}^{{\rm n}-} + {\rm H}_{\lpar {{\rm aq}} \rpar }^ + & \to {\rm T}-{\rm OH}_2^ +{+}\, {\rm L}_{\lpar {{\rm aq}} \rpar }^{{\rm n}-} \cr & \to {\rm T}-{\rm OH}_2^ +{-}\,{\rm L}^{{\rm n}-}}$$

(5)$${\rm T}-{\rm OH} + {\rm L}^{{\rm n}-} + {\rm H}_{\lpar {{\rm aq}} \rpar }^ + \to {\rm T}-{\rm OH}_2^ +{+}\, {\rm L}_{\lpar {{\rm aq}} \rpar }^{{\rm n}-} \to {\rm T}-{\rm L}^{{\rm n}-}$$

As a result, the use of HCl can improve the properties of the external surface and facilitate the extraction of Al atoms. It is also very effective at increasing the size of the micropores (pore volume) of the PNZs without destroying their crystal structure (Cakicioglu-Ozkan & Ulku, Reference Cakicioglu-Ozkan and Ulku2005). This, in turn, indicates that the use of HCl for preconditioning treatment is effective for producing more porous adsorbents.

Na⁺ conditioning of the HC-Z samples

After acid pretreatment, Na⁺ ions should be introduced to the zeolite that has had free pore spaces formed inside it and on its surface by the acid pretreatment. The cation-exchange capacity (CEC) of zeolite is improved by rendering the zeolite homoionic via Na⁺ pretreatment (Ndayambaje, Reference Ndayambaje2011). The HC-Z samples were soaked under various conditions to determine the best concentration to use, as well as the best soaking time. Then, the dried Na-Z samples were soaked with a Cu solution to exchange Na⁺ ions with Cu²⁺ ions. The Cu²⁺ ion was chosen because further functionalization of the PNZs may include Cu modification for bacteria inactivation (Olegario-Sanchez et al., Reference Olegario-Sanchez, Tan, Mendoza and Balela2017). The amount of Cu²⁺ ions in the eluate was determined by AAS. A small Cu²⁺ content in the eluate would indicate that most of the Na⁺ ions were displaced by the Cu²⁺ ions.

The results of the Cu²⁺ exchange are listed in Table 7. After 24 h, the eluates of the Na-Z samples treated with a 4 M NaCl concentration contained the smallest number of Cu²⁺ ions, corresponding to a Cu uptake of 99.88%. Although the Cu uptake values of the Na-Z samples do not differ greatly, increasing the Na⁺ concentration may also increase the amount of Cu²⁺ ions adsorbed by the zeolite.

Table 7. Cu uptake (%) for the Na-Z samples modified with various concentrations of NaCl.

Figure 3 shows the optimized soaking time of Na-Z samples treated at a 4 M concentration, determined by the Cu²⁺ uptake (%) from aqueous solutions containing 100 ppm CuSO₄. The initial Cu²⁺ uptake observed on the Na-Z samples soaked for 3 h was 84.87%, which increased to 88.72% and 90.50% in the Na-Z samples soaked for 6 and 18 h, respectively. After 24 h, the Na-Z samples had adsorbed 99.69% of the Cu²⁺ ions in the Cu solution, which remained constant even after soaking for 48 and 72 h. Hence, the Na-Z samples displayed maximal Na⁺ adsorption after immersion in the NaCl solution for 24 h. Furthermore, PNZ samples have shown maximal Cu uptake at 24 h in previous studies (Olegario-Sanchez et al., Reference Olegario-Sanchez, Tan, Mendoza and Balela2017).

Fig. 3. Saturation vs. soaking time of Na-Z samples in 1 M CuSO₄ for 3, 6, 12, 18, 24, 48 and 72 h.

Cu²⁺ loading comparison between the NW-Z, HC-Z and Na-Z samples

Table 8 shows the Cu uptake percentages of the various PNZ samples in this work. The NW-Z samples displayed 99.40% Cu²⁺ uptake, indicating that Ca²⁺, K⁺, Na⁺ and Mg⁺ exchangeable cations detected by ED-XRF analysis (Table 5) are exchanged with Cu²⁺ ions. The HC-Z displayed a lower Cu uptake (96.62%), indicating a lower CEC for Cu²⁺ ions. Finally, the Na-Z samples yielded the greatest Cu²⁺ uptake (99.88%), which, although comparable to the Cu uptake of NW-Z, shows that the presence of exchangeable Na⁺ ions enhances the Cu adsorption capacity of the zeolite, as ion exchange is the mechanism of adsorption. In addition, the Na-Z samples showed improved textural properties compared to the NW-Z samples due to the acid conditioning step. The NW-Z material still contains impurities, similar to the starting PNZ material.

Table 8. Cu uptake (%) of the various PNZ samples.

Structural and morphological analyses of the zeolite samples

Phase structure identification of the zeolite samples

The XRD traces of the NW-Z, HC-Z, Na-Z and Cu-Z samples are shown in Fig. 4. The major phase of PNZs is clinoptilolite (C), with minor mordenite (M), quartz (Q) and feldspar (F) (Alberti, Reference Alberti1975; Koyama & Takeuchi, Reference Koyama and Takeuchi1977; Concepcion-Rosabal et al., Reference Concepcion-Rosabal, Rodriguez-Fuentes, Bogdanchikova, Bosch, Avalos and Lara2005; Taaca & Vasquez Jr, Reference Taaca and Vasquez2017). The intensities of the clinoptilolite peaks in the XRD traces of HC-Z, Na-Z and Cu-Z decreased compared to those of the NW-Z samples, in accordance with previous studies (Petrov, Reference Petrov, Ming and Mumpton1995; Taaca & Vasquez Jr, Reference Taaca and Vasquez2017). This is due to partial dissolution of the clinoptilolite structure after chemical modification.

Fig. 4. XRD traces of the NW-Z, HC-Z, Na-Z and Cu-Z samples.

Morphology of the zeolite samples

The SEM images of the NW-Z and Na-Z samples are shown in Fig. 5. The original clinoptilolite has a rough surface and consists of loose and stacked particles (Fig. 5a). The Na-Z sample has a rougher surface, which may be due to the formation of micropores after HCl treatment (Fig. 5b). Acid treatment may unclog the zeolite pores, rendering them available for adsorption (Sprynskyy et al., Reference Sprynskyy, Lebedynets, Zbytniewski, Namieśnik and Buszewski2005).

Fig. 5. SEM images of (a) NW-Z and (b) Na-Z. Scale bars: 10 μm.