Leaching of Oxyanionic Trace Contaminants from Metakaolin Geopolymers under Batch or Continuous Water Extraction

Abstract

This study investigates the largely omitted but surprisingly extensive leaching of oxyanionic trace contaminants (vanadium, arsenic, and chromium) from metakaolin geopolymers. Understanding the leaching of these toxic metal­(loid)­s is increasingly important as geopolymers are considered for applications in which they are in contact with water, such as in wastewater treatment. The leaching was quantified from 1 to 4 mm sized crushed geopolymers with batch and continuous water extraction at a pH of 11–13 (i.e., natural pH of geopolymers). The results revealed that the batch extraction (at a liquid/solid ratio of 10) resulted in a substantial release of vanadium (190–4300 μg/L or 3.8–86 μg/g geopolymer), arsenic (39–460 μg/L or 0.78–9.2 μg/g), and chromium (1.3–130 μg/L or 0.026–2.6 μg/g). These values exceed many regulatory guidelines related to environmental safety. However, the continuous extraction with a flow-through setup indicated that the high initial leaching diminished to safe concentrations (<100 μg/L vanadium, < 10 μg/L arsenic, and <50 μg/L chromium) after reaching a cumulative liquid/solid ratio of approximately 11.2. In both batch and continuous leaching, the total released amount of the metal­(loid)­s followed the trend As (4.4–100%) > V (0.8–62.9%) > Cr (0.02–33.2%). As a conclusion, metakaolin geopolymers may require washing before their use when the leaching of oxyanionic metal­(loid) contamination is relevant. Additionally, the batch leaching results (at liquid/solid = 10) of metakaolin geopolymers should be considered with caution, as they represent the initially high release of oxyanions, and thus may create an overly adverse impression of the leaching behavior.

1. Introduction

Geopolymers are synthetic low-Ca alkali-metal aluminosilicate products with an amorphous or nanocrystalline zeolite-like nanostructure and ceramic-like properties. They can be used as a sustainable concrete binder; however, in recent years, many potential high-value and more advanced applications have received attention such as water and wastewater treatment, coatings for controlled-release fertilizers, plant substrate material, or biomedical materials (e.g., implants or drug delivery). This interest was sparked due to the potentially more sustainable and low-cost production route of geopolymers in comparison to competing materials (e.g., conventional ceramics, zeolites, or activated carbons). ,

Metakaolin, formed by the dehydroxylation of kaolinite at approximately 750 °C, is the most well-known and commonly used aluminosilicate precursor for geopolymers. The use of metakaolin is convenient, as it is widely available (proven global reserves are 32 billion tons), and it has a more consistent composition in comparison to many industrial aluminosilicate side streams (e.g., fly ashes, metallurgical slags, or mine tailings). However, kaolinite deposits typically contain trace elements such as Sr, Zn, Co, Cu, As, V, and Cr , due to their adsorption on the mineral surface from the environment, elemental substitution in the mineral structure, or presence of minor quantities of other minerals such as illite, smectite, or goethite. When metakaolin is used as a geopolymer precursor, it is exposed to a pH of up to 13–14, leading to the dissolution of its structure and eventual reagglomeration to form the geopolymer. The pH in the pore solution of the metakaolin geopolymer remains high, at up to 13. The release of trace elements from crushed metakaolin geopolymer particles when exposed to water has been studied at the pH range of 1–14 using a batch extraction at a liquid/solid ratio of 10: it was shown that cationic trace metals (e.g., Sr, Zn, Co, and Cu) had minimal leaching at pH >6 but oxyanions (e.g., As and V) exhibited maximum solubility at pH of 9–12. The cationic metals are effectively stabilized via precipitation as hydroxides at high pH, adsorption on aluminosilicates, and isomorphic substitution of calcium. However, for oxyanions, no effective stabilization mechanisms exist in low-Ca geopolymers, and thus relatively high leaching has been observed from metakaolin geopolymer mortars and granules prepared from the metakaolin geopolymer with municipal solid waste incineration fly ash. In those studies, it was specifically identified that the source of oxyanions was metakaolin (e.g., the leaching amount of V increased upon increasing the metakaolin content and decreasing the fly ash content in a geopolymer formulation).

As metakaolin geopolymers are studied, for example, as adsorbents in wastewater treatment, disinfecting filtration media in drinking water treatment, and many other applications in which the geopolymer would be in contact with water, understanding their leaching behavior becomes increasingly important to ensure environmental safety. In this study, the leaching of three oxyanionic trace contaminants (As, V, and Cr) is monitored with both batch and continuous water extraction setups from crushed metakaolin geopolymers prepared from five different commercial metakaolins. As, V, and Cr were selected for this study due to their toxicity. The World Health Organization (WHO) guideline values for them in drinking water are V 0.1 mg/L, total Cr 0.05 mg/L, and As 0.01 mg/L, while the US Environmental Protection Agency (USEPA) has the maximum levels of Cr 0.1 mg/L and As 0.01 mg/L but no guideline value for V. In addition, arsenic and chromium have also environmental threshold values in groundwater, maximum 0.005 and 0.01 mg/L, respectively, based on the European directive 2006/118/EC. Their leaching is also controlled from materials intended for earth construction, for example, by the Finnish Government decree 843/2017. The aim of this study is to provide currently lacking information about the leaching of trace oxyanionic contaminants over time from metakaolin geopolymers and evaluate the relevance of their leaching, which is essential when designing new potential applications for such materials.

2. Materials and Methods

2.1. Geopolymer Preparation

Three metakaolins (A, B, and C in Table ) and two kaolinites (D and E in Table ) were used in this study. Kaolinite samples were calcined at 750 °C for 5 h to convert them into metakaolin. The materials represent widely used commercial metakaolins and kaolinites in geopolymer studies (the suppliers and material trade names are shown in Table ). Composition of the metakaolins (by X-ray fluorescence) is shown in Table . In addition, As, V, and Cr contents of the metakaolins were determined by inductively coupled plasma mass spectroscopy (ICP-MS) analysis (XSeries II, Thermo Fisher Scientific) with aqua regia-hydrofluoric acid digestion. Laser diffraction (Beckman Coulter LS 13 320) was used to determine the particle size distributions of the metakaolins (Figure ). The mineral phases present in the metakaolins are shown in Figure . Each of the detected phases (quartz, muscovite, hematite, and Illite) is a common impurity in kaolinite deposits and their presence in the final product (metakaolin) depends on how the kaolinite has been processed.

1. Main Elements of Metakaolins Expressed as Weight % of Oxides and the IDs Used Later in the Study .

trade name and supplier	ID	Na2O (%)	MgO (%)	Al2O3 (%)	SiO2 (%)	K2O (%)	CaO (%)	TiO2 (%)	Fe2O3 (%)	LOI at 525 °C (%)	LOI at 925 °C (%)
Argical 1200, Imerys	A	0.06	0.13	39.28	55.75	0.91	0.10	1.95	0.01	1.0	1.9
PalMeta, Aquaminearls Finland	B	0.14	1.48	27.16	59.06	5.19	0.12	1.03	3.54	0.4	1.4
N/A, Thermo Fisher Scientific	C	0.06	0.34	38.73	55.62	1.93	0.07	0.09	1.17	0.2	0.3
MetaMax, BASF	D	0.19		42.50	51.11	0.10	0.07	1.49	0.42	0.3	0.6
N/A, Sigma-Aldrich	E	0.22	0.33	40.84	54.13	1.75	0.05	0.03	0.86	0.3	0.7

^aLOI = loss on ignition, N/A = the supplier did not use any trade name for their product.

1.

Particle size distribution of metakaolins. The particle size distributions according to the manufacturers (as d₅₀) are A 1.5 μm, D 1.3 μm, and B, C, and E unspecified.

2.

X-ray diffractograms of metakaolins.

Sodium silicate solution (Merck, SiO₂/Na₂O molar ratio of approximately 3.5 and 65 wt % water content) was mixed with sodium hydroxide pellets (98.7%, VWR) for 1 day to obtain a SiO₂/Na₂O molar ratio of 1.4. This solution was used as an alkali activator.

Alkali-activator solution and metakaolin were mixed in weight ratios of 0.96 to 1.51 (to obtain constant molar ratios for all geopolymers: Na₂O/Al₂O₃ = 1.0, H₂O/Na₂O = 11.8, and SiO₂/Al₂O₃ = 3.7) using a high-shear mixer (RW20, IKA Labor Technik) at 5000 rpm speed for 10 min. The fresh-state pastes were then poured into steel molds, cured at 60 °C for 24 h in plastic bags, and demolded. The samples were stored for 7 days in plastic bags and crushed and sieved to a particle size of 1–4 mm. Geopolymers prepared from metakaolins are identified as A_GP, B_GP, C_GP, D_GP, and E_GP according to the metakaolin used.

2.2. Leaching Experiments

Leaching was evaluated with a batch water extraction according to the EN-12457-2 standard and a continuous water extraction using a flow-through column method. Fifty grams of 1–4 mm sized geopolymers were used in both experiments. In the column method, ultrapure water was distributed through geopolymer pieces using a constant empty bed contact time (EBCT) of 8.8 min for all geopolymers (bed volumes were 56.6–73.2 cm³ and flow rates 0.11–0.15 L/h), and water samples after the column were taken every 30 min for 5 h. The cumulative liquid-to-solid ratio (L/S, L/kg) during the flow-through leaching was calculated with eq :

$$\mathrm{cumulative}\left(\frac{\mathrm{L}}{\mathrm{S}}\right)=\frac{{\int }_0^{t}Q\mathrm{d}t}{m}$$ 1

where Q is the flow rate (L/h), t is the time (h), and m is the mass of the bed (kg).

In the batch leaching, ultrapure water was mixed with 1–4 mm sized geopolymer pieces at L/S = 10, and the samples were shaken for 24 h. After both leaching experiments, suspensions were filtered through a 0.45 μm membrane filter and water was analyzed for As, V, and Cr with an ICP-MS according to the standard SFS-EN ISO 17294-2:2023. In addition, the pH of water samples was measured with a WTW InoLab 7710 meter. In addition to the geopolymer samples, control batch leaching experiments were conducted with the studied metakaolins (i.e., not converted into geopolymers) in ultrapure water without pH adjustment (pH of 4.8–6.5 after 24 h shaking) and ultrapure water for which the pH was adjusted to 12.5 using NaOH (pH of 12.1–12.3 after 24 h shaking).

The total leached amounts (as %) of arsenic, chromium, and vanadium from continuous and batch experiments were calculated using eqs and , respectively.

$$\mathrm{total}\mathrm{leached}\mathrm{amount},\mathrm{continuous}(\%)=\frac{\sum _{t=0\mathrm{h}}^{t=5\mathrm{h}}(C_t\times V_t)}{m_{\mathrm{metakaolin}}\times C_{\mathrm{metal}(\mathrm{loid})}}$$ 2

where C _t (μg/L) is the As, V, or Cr concentration in the water sample collected at time t, V _t (L) is the volume of the water sample collected at time t, m _metakaolin (g) is the mass of metakaolin in the geopolymer sample used for the leaching experiment, and C _metal(loid) (μg/g) is the concentration of As, V, or Cr in metakaolin as analyzed by ICP-MS.

$$\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{h}\mathrm{e}\mathrm{d}\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t},\mathrm{b}\mathrm{a}\mathrm{t}\mathrm{c}\mathrm{h}(\%)=\frac{C_{\mathrm{b}\mathrm{a}\mathrm{t}\mathrm{c}\mathrm{h}}\times V_{\mathrm{b}\mathrm{a}\mathrm{t}\mathrm{c}\mathrm{h}}}{m_{\mathrm{m}\mathrm{e}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{a}\mathrm{o}\mathrm{l}\mathrm{i}\mathrm{n}}\times C_{\mathrm{m}\mathrm{e}\mathrm{t}\mathrm{a}\mathrm{l}(\mathrm{l}\mathrm{o}\mathrm{i}\mathrm{d})}}$$ 3

where C _batch (μg/L) is the As, V, or Cr concentration in the water sample collected after 24 h batch leaching, V _batch (L) is the volume of the water sample, and m_metakaolin (g) and C _metal(loid) (μg/g) are the same as given in eq experiment, and C _metal(loid) (μg/g) is the concentration of As, V, or Cr in metakaolin as analyzed by ICP-MS.

3. Results and Discussion

Figure presents the leaching results from the flow-through column experiment. The total contents of V, As, and Cr (Table ) appear to have poor correlation with the high initial leaching in Figure (e.g., D_GP has much higher V leaching but not higher total V content). Thus, the release of these elements is likely also controlled by factors such as the mineral phase to which the metal­(loid) is originally attached (i.e., bonding environment and speciation) and the reaction degree of the geopolymers. The results demonstrate that the leaching diminishes rapidly upon flushing the geopolymers and, for example, the WHO drinking water guideline values (i.e., V < 100 μg/L, As < 10 μg/L, and Cr < 50 μg/L) were reached after a cumulative L/S ratio of approximately 10, 11.2, and 4.5, respectively. The pH values during the flow-through column test decreased as residues of unreacted alkali activator were gradually removed.

3.

Flow-through column leaching results and pH change are shown as a function of a cumulative L/S ratio. The horizontal lines represent the WHO guideline value for drinking water, and vertical lines denote the cumulative L/S ratio when the values are reached.

2. Total As, V, and Cr Contents of Metakaolins Analyzed by Aqua Regia-Hydrofluoric Acid Digestion and ICP-MS and Batch Leaching of As, V, and Cr from Metakaolins at Two pH Conditions.

	total content			batch leaching (pH = 4.8–6.5)			batch leaching (pH = 12.1–12.3)
ID	As (μg/g)	Cr (μg/g)	V (μg/g)	As (μg/g)	Cr (μg/g)	V (μg/g)	As (μg/g)	Cr (μg/g)	V (μg/g)
A	8.9	3.1	100	0.036	0.027	0.28	5.6	0.31	21
B	2.4	5.8	153	<0.01	1.1	7.1	0.54	1.0	21
C	∼1.0	210	170	0.46	0.096	1.0	2.7	0.1	1.8
D	12	100	120	<0.01	<0.01	1.3	0.66	0.14	25
E	∼1.5	120	160	0.18	0.085	0.21	5.5	0.094	0.38

^aBatch leaching at pH 12.1–12.3 exceeded this value, which implies that As recovery in aqua regia-hydrofluoric acid digestion and ICP-MS was not 100%. Thus, this value can be considered an approximation.

As shown in Table , the batch leaching results (at L/S = 10 following the standard EN-12457-2) exceed the guideline values, so none of the materials could be used for earth construction or in contact with drinking or groundwater. Three of the geopolymers even exceeded the inert waste leaching limits. Thus, it is obvious that the EN-12457-2 standard (which is used as a basis for many legislative limits) provides an exceedingly adverse impression of the leaching of oxyanions due to the high initial leaching.

3. Arsenic, Chromium, and Vanadium Batch Leaching Data (L/S = 10, According to Standard EN-12457-2) and pH Values.

		A_GP	B_GP	C_GP	D_GP	E_GP
Arsenic	(μg/g)	7.6	1.0	0.78	2.4	9.2
	(μg/L)	460,	39,	120,	50,	380,
Chromium	(μg/g)	0.026	0.58,	2.6,	0.042	0.88,
	(μg/L)	44	130,	2.1	29	1.3
Vanadium	(μg/g)	6.2	86	48	3.8	58
	(μg/L)	2900	2400	190	4300	130
pH		12.5	12.22	12.78	12.61	12.48

^aExceeds at least one category of earth construction maximum leaching values in Finland (according to VNa 843/2017).

^bExceeds the inert waste landfilling limit (according to 2003/33/EC).

^cExceeds the WHO drinking water guideline values.

^dExceeds the groundwater threshold value (according to 2006/118/EC).

When comparing the batch leaching results of metakaolins (Table ) at a similar pH as those in Table (i.e., approximately 12.5), the leaching of arsenic is systematically higher from geopolymers. For chromium, the samples A_GP, B_GP, and D_GP exhibit lower leaching from geopolymers in comparison to metakaolins, while the samples C_GP and E_GP have higher leaching from metakaolins. This could be due to the speciation of Cr, since Cr­(III) is likely more efficiently stabilized in geopolymers in comparison to Cr­(VI). Samples C_GP and E_GP were those prepared from kaolinites freshly calcined before the preparation of geopolymers, which may contribute to a higher proportion of Cr­(VI). For vanadium, it appears that some samples have a higher leaching than that of geopolymers (i.e., B_GP, C_GP, and E_GP), while others have higher leaching than that of metakaolins (A_GP and D_GP). Finally, when comparing leaching from metakaolins at pH values of 4.8–6.5 and 12.1–12.3 (Table ), it can be seen that leaching at alkaline conditions is systematically higher.

The total leaching amounts of the metal­(loid)­s (as mass-%) evaluated from the mass balance (i.e., ratio of the cumulative leached amount in batch or continuous extraction to the total amount) are shown in Table . In both batch and continuous leaching modes, the total released amount of the metal­(loid)­s followed the trend As (4.4–100%) > V (0.8–62.9%) > Cr (0.02–33.2%).

4. Estimated Total Leached Amount (as %) of Arsenic, Chromium, and Vanadium from Continuous and Batch Experiments .

		A_GP	B_GP	C_GP	D_GP	E_GP
arsenic	batch	87.5%	24.1%	∼100%	7.3%	∼100%
	continuous	24.2%	14.5%	79.3%	4.4%	∼100%
chromium	batch	24.0%	33.2%	0.4%	0.5%	0.02%
	continuous	7.4%	17.4%	0.01%	0.3%	0.03%
vanadium	batch	49.2%	23.3%	1.9%	62.9%	1.4%
	continuous	15.5%	11.9%	0.8%	38.1%	0.8%

^aCalculated as a ratio of the leached amount to the total amount.

^bMass balance calculation exceeded 100%, and thus, the total release of the element is assumed.

The leaching of vanadium is due to the high solubility of VO₄ ^3– (which is the predominant V species at pH 11–12). Arsenic has several possible soluble species at pH 11–12, depending on the redox potential: AsO₄ ^3– or HAsO₄ ^2–, where As has oxidation state +5, or H₂AsO₃ ^– or H₃AsO₄, where As has oxidation state +3. The low total quantity of As in comparison to V or Cr (Table ) likely explains its high total leached amount (Table ). For chromium, the only soluble species at pH 11–12 is CrO₄ ^2– (in which Cr has oxidation state VI).

According to the results above, and if the WHO drinking water guideline values are considered as the target (they are the strictest of the considered limits), metakaolin geopolymers should be rinsed until a cumulative L/S ratio of approximately 11.2 is reached to ensure low enough leaching in their subsequent use. Approximate concentrations of As, Cr, and V in the washing water from each geopolymer when it is rinsed with water to a cumulative L/S ratio of approximately 11.2 are shown in Table . This means, as an example, that for each ton of metakaolin geopolymer, 11.2 tons of washing water would be generated. Thus, it appears that the metakaolin geopolymer can be in some cases a significant source of secondary pollution if there is no initial washing. On the other hand, the prewashing of geopolymers can lead to additional significant costs (the washing water probably requires treatment for V, As, or Cr, which can be achieved with, for example, coagulation/flocculation or precipitation).

5. Concentrations of As, Cr, and V in the Washing Water when Geopolymers Are Rinsed up to a Cumulative L/S Ratio of Approximately 11.2 (i.e., when the WHO Drinking Water Guidelines Are Achieved).

geopolymer	As (μg/L)	Cr (μg/L)	V (μg/L)
A_GP	113.6	12.0	816.3
B_GP	20.9	75.8	1092.8
C_GP	39.3	1.1	65.3
D_GP	27.6	13.3	2384.9
E_GP	133.7	1.7	63.7

4. Conclusions

This study investigates the leaching behavior of vanadium, arsenic, and chromium from geopolymers prepared from five different metakaolins. The leaching was assessed using either batch or continuous flow-through water extraction setups. The natural high pH of geopolymers (as indicated by the high pH of the leachates, approximately 12.3–12.7) promotes the leaching and mobility of vanadium, arsenic, and chromium, as they are present as VO₄ ^3– and CrO₄ ^2–, while arsenic can be present in several oxyanionic forms. The batch leaching experiment at L/S = 10 exhibited high leaching of the studied metal­(loid)­s, which clearly exceeds, for example, guideline values for drinking water, groundwater, or materials intended for earth construction. However, when the leaching was assessed using a continuous flow-through setup, the release of vanadium, arsenic, and chromium diminished below the guidelines when the cumulative L/S ratio of 11.2 was reached. Thus, the results highlight that metakaolin geopolymers could be washed before use, especially in their emerging applications in water treatment. The washing water may have high concentrations of the metal­(loid)­s (up to 134 μg/L As, 76 μg/L Cr, and 2400 μg/L V), and thus may require additional treatment. Second, the commonly used batch leaching test at L/S = 10 (e.g., EN-12457-2 standard) may result in misleadingly high release of oxyanionic components, considering that their release decreases very rapidly under continuous water extraction after exceeding L/S = 10. Alternative to washing, inclusion of Ca, Fe, and/or sulfate to the geopolymer structure could help to immobilize oxyanions via precipitation and sorption or structural uptake (e.g., into ettringite or layered double hydroxides). However, in certain emerging applications of geopolymers, such inclusion of Ca, Fe, and sulfate may not be possible, as it may deteriorate the wanted properties of geopolymers (e.g., adsorption of aqueous ammonium is an example of such a case).

The raw data used to prepare this article is presented either directly in the article or in the Supporting Information.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c10438.

• The raw data of the continuous leaching experiments (mass of geopolymer in each experiment was 50.0 g) (Table S1) (PDF)

The authors declare no competing financial interest.