Towards Safe Diatomite Sludge Management: Lead Immobilisation via Geopolymerisation

: Diatomite, a natural adsorbent rich in active silica, serves as a valuable precursor for geopolymer synthesis. The safe disposal of diatomite as a failed lead (Pb(II)) adsorbent is critical to prevent secondary contamination. This study investigated the immobilisation efficiency of geopolymerisation for Pb(II)-rich diatomite sludge. Low-grade diatomite with high ignition loss was utilised in the synthesis of alkali-activated geopolymers. It was demonstrated that the geopolymers achieved a compressive strength of 28.3 MPa with a 50% replacement rate of metakaolin by diatomite sludge, which was not a compromise in strength compared to that of the geopolymer with no Pb(II) (26.2 MPa). The leaching behaviour of Pb(II) was evaluated using water and acetic acid, yielding concentrations below 3 mg/L and immobilisation efficiencies of 95% in both scenarios. Analytical techniques including Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD), and scanning electron microscopy (SEM) elucidated the mineral composition and chemical environment of the geopolymers. These analyses revealed that Pb(II) migrated from diatomite pores, potentially forming soluble hydroxides under sufficient hydroxide, which then participated in condensation with silicon and aluminium monomers, effectively immobilising Pb(II) within amorphous aluminosilicate gels. Furthermore, the formation of the amorphous gels within diatomite pores hindered Pb(II) leaching, encapsulating Pb(II) effectively. This study presents a novel approach to immobilising heavy metals within building materials, enhancing mineral resource utilisation efficiency while addressing environmental contamination concerns.


Introduction
The contamination of surface and groundwater remains a critical environmental issue globally.Among various pollutants, heavy metal ions have attracted considerable attention due to their high toxicity, wide range of influence, and challenging biological degradation process [1,2].Diatomite, a naturally occurring material characterised by its high porosity and abundant surface hydroxyl groups, has demonstrated effective adsorption properties in water treatment applications [3][4][5][6].
However, the disposal of spent diatomite adsorbents laden with heavy metals poses a new environmental challenge.Improper management can lead to secondary contamination, undermining the initial benefits of diatomite's use in water treatment.Thus, ensuring the safe disposal of spent diatomite adsorbents is crucial for environmental remediation and for mitigating the risks associated with heavy metal pollution.Immobilisation technology presents a promising approach to managing solid wastes rich in heavy metals by reducing their toxicity and mobility, thereby enhancing environmental stability [7].
While cement has traditionally dominated immobilisation technology [8][9][10][11], its high carbon footprint spurred interest in more sustainable alternatives.Geopolymers, a class of low-carbon cementitious materials, offer a promising solution.Geopolymers possess superior properties such as refined pore structures, low permeability, high alkalinity, and exceptional chemical stability compared to conventional cementitious materials [12].Previous studies have highlighted their effectiveness in immobilisation applications [13,14].For example, Tan, J. et al. [15] investigated the feasibility of utilising construction and demolition waste-based geopolymers to stabilise solids in municipal solid waste incineration fly ash.The researchers found that geopolymers were effective in the immobilisation of various types of heavy metals in fly ash, with the highest efficiency observed for lead.Ji, Z. et al. [16] also observed that the stability of metals in geopolymers was increased, and the release risk was decreased.
The synthesis of geopolymers typically involves natural minerals like kaolinite and halloysite as raw materials [17][18][19].These minerals must undergo specific treatments to activate them into reactive amorphous aluminosilicates [20].Diatomite stands out as a natural mineral that exhibits the inherent reactivity of geopolymerisation without requiring additional processing [21][22][23].Diatomite is a fossil assemblage of diatom frustules and is mainly composed of amorphous hydrated silica (SiO 2 •nH 2 O) [24].It is an ideal source of active silicon, which is essential for geopolymerisation.This unique attribute presents a promising opportunity for immobilising diatomite sludge using geopolymers.Despite these advantages, research on the immobilisation of diatomite sludge through geopolymerisation remains limited, particularly regarding the mechanism of heavy metal transformation within this system.
This study explored the impact of varying amounts of diatomite on the compressive strength of geopolymers to determine the optimal volume of diatomite sludge for solidifying geopolymer bodies.The stability of the solidified bodies was evaluated through leaching behaviour and compressive strength tests.The mineralogical properties and microstructure were analysed to investigate the mechanisms underlying the immobilisation of lead-rich diatomite sludge via geopolymerisation.

Raw Materials
Diatomite (Dt) was sourced from Tongliao, Inner Mongolia, China, while metakaolin (MK) was derived from calcining kaolin, which was sourced from Maomin in Guangdong, China, for 2 h at 750 • C in a muffle furnace to obtain high reactivity [19].

Preparation of Pb(II)-Rich Dt Sludge
The lead nitrate (Pb(NO 3 ) 2 ), of analytical reagent grade, was dissolved in deionised water to form a homogeneous solution with a concentration of 250 mg/L.Dt was then mixed with this solution in the container to achieve a liquid-solid ratio of 10.The mixture was agitated using a thermostatic oscillator for 24 h at 25 • C to ensure the effective adsorption of Pb 2+ by Dt.Afterwards, the solids were separated by centrifugation at 4000 rpm for 20 min and dried in an oven at 60 • C for 5 d.The resulting dried powder was identified as Pb(II)-rich Dt sludge, and the amount of Pb(II) in the sludge was calculated by measuring the concentration of Pb 2+ in the supernatant.

Preparation of Geopolymers
In this study, the alkali-activator was a 10 mol/L sodium hydroxide solution (10 M NaOH).Preparation involved dissolving NaOH pellets in deionised water and allowing the solution to cool for at least 24 h before use.Geopolymer pastes were prepared using a cement mixer.Dt and the alkali-activator should be pre-blended by mechanical stirring for Minerals 2024, 14, 763 3 of 15 24 h to ensure the dissolution of the diatom frustule.MK was then added to the mixture in the mixer bowl.Mixing commenced at low speed for 60 s, followed by a 90 s interval during which the pastes adhering to the wall were removed and placed in the centre of the bowl and then mixing at a high speed for an additional 4 min.The resulting paste was then moulded into cubes (20 × 20 × 20 mm 3 ) and vibrated for 1 min to eliminate air bubbles that formed during mixing.Specimens were subsequently placed in an environmental chamber at 40 ± 1 • C and 98 ± 2% relative humidity (RH), covered with plastic wrap to prevent water loss.After one day, specimens were demoulded and left to cure at room temperature until testing.The specific details of the preparation are outlined in Table 1.Geopolymers resulting from this process were denoted as GDt 10 , GDt 20 , GDt 30 , GDt 40 , GDt 50 , and GDt 60 , respectively.

Leaching Tests
In this part, two different leaching methods were conducted to comprehensively assess the leaching behaviour of Pb(II) from the geopolymers.
One method utilised deionised water as the leachate to evaluate the potential release of Pb(II) under conditions simulating surface or groundwater exposure, following ASTM D8155-17 guidelines [25].Geopolymers were crushed to a particle size less than 5.0 mm.These particles were added to a container with deionised water at a weight ratio of 1:10 and agitated continuously for 18 h at 25 • C using end-over-end agitation equipment rotating about a central axis at 30 r/min.
The other method employed acidic leachate to evaluate the stability of geopolymers under extreme situations such as exposure to landfill leachate or acid rain, following the United States Environmental Protection Agency's (US EPA's) Toxicity Characteristic Leaching Procedure (TCLP), which is one of the most commonly used protocols [26,27].For alkaline materials, an acetic acid (HOAc) solution with a pH of 2.88 ± 0.05 was used.Geopolymers were crushed to a particle size less than 9.5 mm and mixed with HOAc solution at a ratio of 1:20.Leaching proceeded for 20 h at 25 • C using a thermostatic oscillator.
Leachates from both methods were separated by centrifugation at 4000 rpm for 20 min and filtered through a 0.45 µm filter.The concentrations of Pb 2+ in the leachates were determined using inductively coupled plasma optical emission spectroscopy (ICP-OES).All chemical reagents used were of analytical purity.

Characterisation
X-ray fluorescence (XRF) tests were conducted using a wavelength-dispersive X-ray fluorescence spectrometer (XRF-1800, Shimadzu, Japan).The sample mixed with borate flux was melted into glass beads for analysis.In order to determine the loss on ignition (LOI), 1.000 g samples were heated to 1010 • C for 10 min.
Fourier transform infrared (FTIR) spectroscopy was performed on a Bruker VERTEX 70 infrared spectrometer (Karlsruhe, Germany).The sample mixed with KBr was pressed into a transparent round slice.The spectrum of the slice was collected in a wavenumber range from 4000 cm −1 to 400 cm −1 under transmittance mode.A spectral resolution of 4 cm −1 was selected, and each spectrum was derived from the average of 60 scans.
FESEM was carried out with SU8010 (Hitachi, Japan) at an accelerating voltage of 1.5 kV.The equipped AMETEK energy-dispersive X-ray spectrometer (EDS; Hitachi, Japan) was used for the EDS spot analysis of the micro-component content with a voltage of 15 kV and a current of 20 µA.
ICP-OES tests performed with Agilent ICPOES730 (Santa Clara, CA, USA) at wavelength 220.353 nm were used to determine the concentration of Pb in leachate.
Compressive strength tests were performed using a YAW-300D Computer Control Electronic Compression testing machine (Lixian, China) with a load capacity of 300 kN.The applied load rate on the specimen was 500 N/s.A one-way analysis of variance (ANOVA) followed by Tukey's post hoc multiple comparison tests were employed to ascertain the discrepancies in compressive strengths attributable to the addition of Dt [28].

Properties of Raw Materials
The chemical compositions of Dt and MK were determined using XRF.The results presented in Table 2 indicated that Dt was composed primarily of silica.The loss on ignition (LOI) reached 20.74%, which exhibited a higher level than that of normal Dt (5%~12%) and was considered to be rich in organic matter [29,30].MK primarily contained aluminosilicate, with a silicon (Si)-to-aluminium (Al) molar ratio of 1.The mineral compositions were determined via XRD analysis, with Figure 1a showing the diffraction patterns of Dt and MK.Dt exhibited a broad hump peak in the range of 15 • to 30 • (2θ), which represented the amorphous silica phase with the same structure as opal-A [24,31].Furthermore, minor quantities of chlorite and quartz mineral impurities were detected in Dt.Similarly, MK displayed a comparable hump peak, contributable to the disruption of the kaolinite structure by removing the hydroxyl groups during the hightemperature treatment.This transformation changed ordered 1:1 layered aluminosilicate into a short-ordered amorphous phase [32,33].The remaining diffraction patterns in MK indicated the presence of muscovite, quartz, and kaolinite without dehydroxylation.
Figure 1b,c depicts the morphology of Dt and MK, respectively, as observed via FESEM.Dt revealed complete diatom frustules characterised by a highly porous disc-like shape, with diameters ranging from 10 to 30 µm.Numerous finely crushed frustules were also observed concurrently.MK retained a layered structure despite the loss of hydroxyl groups, showcasing a pseudo-hexagonal plate-like morphology stacked along the c-axis direction.also observed concurrently.MK retained a layered structure despite the loss of hydroxyl groups, showcasing a pseudo-hexagonal plate-like morphology stacked along the c-axis direction.

Compressive Strengths of Dt-Based Geopolymers
As a crucial indicator for assessing immobilisation effectiveness, the compressive strength of geopolymers showed a positive correlation with reduced heavy metal leaching [7].This part aimed to determine the optimal Dt content in geopolymers to achieve the Minerals 2024, 14, 763 6 of 15 desired compressive strength levels.Figure 2 illustrates the average compressive strengths of Dt-based geopolymers at 7 d and 28 d.Except for GDt 10 , all specimens showed improved compressive strength from 7 to 28 days.GDt 20 , GDt 40 , GDt 50 , and GDt 60 exhibited growth rates exceeding 20%, corresponding to an increase of approximately 5 MPa.This suggests ongoing geopolymerisation in most systems over the subsequent 21 days, enhancing compressive strength through the continued formation of amorphous aluminosilicates [34,35].Notably, GDt 30 showed the highest compressive strength at 7 d, reaching levels comparable to GDt 20 , GDt 40 , GDt 50 , and GDt 60 at 28 d.However, no further strength improvement was observed in GDt 30 at 28 d.

Compressive Strengths of Dt-Based Geopolymers
As a crucial indicator for assessing immobilisation effectiveness, the compressive strength of geopolymers showed a positive correlation with reduced heavy metal leaching [7].This part aimed to determine the optimal Dt content in geopolymers to achieve the desired compressive strength levels.Figure 2 illustrates the average compressive strengths of Dt-based geopolymers at 7 d and 28 d.Except for GDt10, all specimens showed improved compressive strength from 7 to 28 days.GDt20, GDt40, GDt50, and GDt60 exhibited growth rates exceeding 20%, corresponding to an increase of approximately 5 MPa.This suggests ongoing geopolymerisation in most systems over the subsequent 21 days, enhancing compressive strength through the continued formation of amorphous aluminosilicates [34,35].Notably, GDt30 showed the highest compressive strength at 7 d, reaching levels comparable to GDt20, GDt40, GDt50, and GDt60 at 28 d.However, no further strength improvement was observed in GDt30 at 28 d.The ANOVA was applied to compare the compressive strength among different Dt addition levels, followed by Tukey's post hoc multiple comparison.The results summarised in Table 3 indicate three distinct categories based on the 7 d compressive strength.The highest compressive strength (23.8 MPa) was achieved with 30 wt.% Dt, significantly different from all other specimens.GDt20, GDt40, GDt50, and GDt60 had strengths over 18.0 MPa, lower than GDt30 but not significantly different from each other.GDt10 exhibited the lowest compressive strength at 12.6 MPa.Consequently, the optimal addition amount of Dt in geopolymers for achieving a satisfactory 7 d compressive strength was found to be 30 wt.%.However, this impact of Dt changed with longer curing time.For the 28 d The ANOVA was applied to compare the compressive strength among different Dt addition levels, followed by Tukey's post hoc multiple comparison.The results summarised in Table 3 indicate three distinct categories based on the 7 d compressive strength.The highest compressive strength (23.8 MPa) was achieved with 30 wt.% Dt, significantly different from all other specimens.GDt 20 , GDt 40 , GDt 50 , and GDt 60 had strengths over 18.0 MPa, lower than GDt30 but not significantly different from each other.GDt 10 exhibited the lowest compressive strength at 12.6 MPa.Consequently, the optimal addition amount of Dt in geopolymers for achieving a satisfactory 7 d compressive strength was found to be 30 wt.%.However, this impact of Dt changed with longer curing time.For the 28 d compressive strength, no significant difference was observed among geopolymers with Dt additions ranging from 20 wt.% to 60 wt.%.This indicates that increasing Dt from 10 wt.% to 20 wt.% enhanced 28 d strength, but further increases did not affect compressive strength.Notably, GDt 10 showed decreased compressive strength from 7 to 28 d, attributed to inappropriate L/S and Si/Al ratios.The unsuitable L/S ratio caused drying shrinkage [36], while the low Si/Al ratio led to the formation of small oligomers, exacerbating cracking and reducing long-term strength compared to other specimens [37].These analyses demonstrated that incorporating 30% Dt significantly enhanced both 7 d and 28 d geopolymer strengths, which was conducive to the practical deployment of this material.Nevertheless, this study aimed to evaluate the potential of geopolymerisation for Pb(II)-rich Dt sludge immobilisation, suggesting higher Dt incorporation reduces the need for MK, thereby cutting CO 2 emissions from kaolinite calcination and reducing solidified volume.This improves immobilisation efficiency while lowering the carbon footprint.Consequently, 50% Dt addition is deemed optimal for subsequent research.

Mineral Composition
The geopolymer solidified bodies were prepared according to the methods outlined in a preceding study, with Pb(II)-rich Dt sludge comprising 50% of the solid materials, while other preparation conditions remained constant.Figure 3 presents the XRD spectrum of the solidified body designated Pb-GDt 50 formed at 28 d, with GDt 50 serving as a control.
The XRD patterns revealed differences in the short-range ordering of amorphous phases among Dt, MK, and geopolymers [38,39].The geopolymers exhibited a typical amorphous aluminosilicate three-dimensional framework represented by a broad peak centred at 27 • (2θ), shifted to a higher angle compared to the Dt and MK patterns centred at 23 • (2θ).Both GDt 50 and Pb-GDt 50 displayed similar patterns without significant difference in pattern position and area, indicating that the presence of Pb(II) in Dt sludge did not influence the type of geopolymerisation products.While previous studies suggested that heavy metal ions might induce crystalline metal aluminosilicate phases like zeolite during geopolymerisation, no crystalline Pb(II)-containing phase was detected here [16].This absence could be attributed to the initial high Si/Al ratio and insufficient alkalinity of the reaction system, favouring the formation of a high-strength amorphous phase [40][41][42].However, geopolymerisation did convert a small portion of kaolinite to paragonite, a Na aluminosilicate with water.In natural settings, the release of Na + from the alteration is a common phenomenon that can result in the transformation of kaolinite into paragonite [43].The presence of structural defects in the heat-treated kaolinite in this study may have facilitated the occurrence of this transformation.Mineral impurities including albite, muscovite, and quartz did not participate in the reaction, filling the amorphous phase as fine aggregates.
Figure 4 depicts the FTIR spectra of the raw materials and the geopolymers GDt 50 and Pb-GDt 50 .In GDt 50 , vibration bands around 1018 cm −1 corresponded to the asymmetric stretching vibration of Si-O-(Si, Al) [44,45], shifting to a lower wavenumber compared to Dt and MK (1096 cm −1 ).This phenomenon was characteristic of the formation of the geopolymer amorphous phase [46].The aluminosilicates from the raw materials underwent a sequential depolymerisation and condensation process, which resulted in alterations to the chemical environment of the Si-O and Al-O bonds [47,48].Notably, the presence of Pb(II) did not further alter the position of Si-O-(Si, Al) vibration bands, which remained at 1018 cm −1 in Pb-GDt 50 .This did not match the findings of previous studies concerning the immobilisation of heavy metals in alkali-activated MK-based geopolymers.In these studies, heavy metals could alter the vibration bands by replacing Na + or participating in the condensation of Al(OH) 4 or Si(OH) 4 monomers [49,50].The unique porous structure of Dt and the preparation process of solidified bodies likely influenced this chemical behaviour of heavy metals.Figure 4 depicts the FTIR spectra of the raw materials and the geopolymers GDt50 and Pb-GDt50.In GDt50, vibration bands around 1018 cm −1 corresponded to the asymmetric stretching vibration of Si-O-(Si, Al) [44,45], shifting to a lower wavenumber compared to Dt and MK (1096 cm −1 ).This phenomenon was characteristic of the formation of the geopolymer amorphous phase [46].The aluminosilicates from the raw materials underwent a sequential depolymerisation and condensation process, which resulted in alterations to the chemical environment of the Si-O and Al-O bonds [47,48].Notably, the presence of Pb(II) did not further alter the position of Si-O-(Si, Al) vibration bands, which remained at 1018 cm −1 in Pb-GDt50.This did not match the findings of previous studies concerning the immobilisation of heavy metals in alkali-activated MK-based geopolymers.In these studies, heavy metals could alter the vibration bands by replacing Na + or participating in the condensation of Al(OH)4 or Si(OH)4 monomers [49,50].The unique porous structure of Dt and the preparation process of solidified bodies likely influenced this chemical behaviour of heavy metals.Moreover, newly formed weak bands around 1415 cm −1 in GDt 50 and Pb-GDt 50 were attributed to the O-C-O stretching vibration of carbonate salt, as evidenced by previous studies [51,52].This was a consequence of the carbonation of CO 2 in the atmosphere, which promoted the conversion of hydroxide to carbonate salt [53].The types of carbonates may include sodium carbonate (Na 2 CO 3 ) derived from a residual NaOH alkali-activator and lead carbonate converted from lead hydroxide (Pb(OH) 2 ) precipitate which was the product of the reaction between the alkali-activator and Pb 2+ adsorbed by Dt.Pores and cracks formed during water loss provided the main pathway for CO 2 ingress, accelerating hydroxide conversion.
Significantly, the organic-rich nature of Dt in this study was notable.Even after vigorous geopolymerisation, the vibration bands attributed to organic matter persisted in GDt 50 and Pb-GDt 50 spectra.Bands at 2926 cm −1 and 2853 cm −1 corresponded to antisymmetric and symmetric stretching vibrations of methylene (-CH 2 ), while the vibration band at 1385 cm −1 indicated the presence of methyl (-CH 3 ) [54].This suggested that organic matter Moreover, newly formed weak bands around 1415 cm −1 in GDt50 and Pb-GDt50 were attributed to the O-C-O stretching vibration of carbonate salt, as evidenced by previous studies [51,52].This was a consequence of the carbonation of CO2 in the atmosphere, which promoted the conversion of hydroxide to carbonate salt [53].The types of carbonates may include sodium carbonate (Na2CO3) derived from a residual NaOH alkaliactivator and lead carbonate converted from lead hydroxide (Pb(OH)2) precipitate which was the product of the reaction between the alkali-activator and Pb 2+ adsorbed by Dt.Pores and cracks formed during water loss provided the main pathway for CO2 ingress, accelerating hydroxide conversion.
Significantly, the organic-rich nature of Dt in this study was notable.Even after vigorous geopolymerisation, the vibration bands attributed to organic matter persisted in GDt50 and Pb-GDt50 spectra.Bands at 2926 cm −1 and 2853 cm −1 corresponded to antisymmetric and symmetric stretching vibrations of methylene (-CH2), while the vibration band at 1385 cm −1 indicated the presence of methyl (-CH3) [54].This suggested that organic matter in Dt can withstand strong alkaline environments.When attached to the surface of the diatom frustules, they could act as a barrier against the depolymerisation of Dt in alkaline solutions, thereby reducing frustule dissolution.

Microstructure
Figure 5 presents SEM images and the EDS analysis results of GDt 50 and Pb-GDt 50 at 28 d.Geopolymerisation produced gels that encapsulated impurity minerals and unreacted raw material particles into a compacted unit.These gel products exhibited visible pores primarily formed by water loss during the curing [20], occasionally leading to crack formation due to gel shrinkage [55].In Figure 5a, undissolved diatom frustules and gel products filling the porous structure of the frustules are observed.This phenomenon indicates that the Al(OH) 4 monomers released from MK diffused into the pores, condensed, and then formed gel products that filled the pores, which may have enhanced the strength properties of the geopolymers.This is because, unlike mismatched structures between frustules and gels, a 'mortise-and-tenon' structure formed, improving particle bonding.

Compressive Strength
The compressive strength of Pb-GDt50 at 28 d measured 28.3 ± 0.7 MPa, indicating a slight increase compared to GDt50.This phenomenon was attributed to the consumption of OH − by Pb 2+ , which primarily affected the dissolution of diatom frustules without significantly altering MK's participation in geopolymerisation.Despite chemical composition alterations in gels, their quantity remained unchanged, supported by XRD analysis showing comparable areas of hump peaks representing the amorphous phase in both geopolymers.This is because MK is more prone to dissolution than diatom frustules due to T hydroxyl group removal, rendering aluminium in MK less stable; the alkalinity required for these Al element dissolutions is lower than that for the Si element in diatom frustules [63][64][65].Thus, the limited reduction in the alkalinity of the activator due to Pb 2+ was unlikely to have a significant effect on MK dissolution.Furthermore, the limited quantity of Pb(II) outside the frustules meant that even if they participated in the condensation, it did not reduce strength, as observed in other studies [66].This underscores the advantage of using MK-based alkali-activated geopolymers for effectively immobilising Pb(II)-rich Dt sludge, especially in achieving high compressive strength, a critical evaluation metric.The EDS results of both geopolymers indicated that the gel products were predominantly composed of Si, Al, and Na elements, given that Al participated in the condensation in a tetra-coordinated form, requiring Na + in aluminosilicate gel framework cavities for charge balance [56].Although Pb(II) in Dt sludge was a metal cation, it could not be immobilised similar to Na + since it already bound to OH − upon dissolution in the activator.Nevertheless, Pb(II) participated in geopolymerisation; in excess OH − , Pb(OH) 2 as the amphoteric hydroxide could further dissolve, forming soluble HPbO − 2 or Pb n (OH) (2n−m)+ m [57,58].Depending on the charge nature of the Pb-containing groups, they condensed with Al(OH) 4 or Si(OH) 4 monomers via dehydration or replaced Na + electrostatically [59].This was evidenced by reducing Si/Al and Na/Al ratios in Pb-GDt 50 compared to GDt 50 .The consumption of OH − by Pb 2+ weakened the dissolution of diatom frustules in the activator, decreasing available Si(OH) 4 monomers for condensation.The replacement of Na + with positively charged functional groups of Pb served to balance the overall charge of the framework, thereby reducing the necessity for Na + .However, EDS did not detect the Pb element, possibly because most Pb(II) was immobilised within the pores of diatom frustules, known for high-density adsorption sites binding Pb(II) in Dt sludge [60].On contact with the activator, Pb(II) bonded to these sites forming hydroxides, hindered by hydrogen bonds between silica hydroxyl groups and hydroxides diffusing from pores [61].The formation of gel products within the pores resulted in the retention of Pb(II) within the pores.Consequently, the quantity of the Pb element in the gel products situated outside the frustules was low, which prevented the concentration from reaching the detection limit of EDS.It is important to note that participation in framework construction was not the sole means of achieving immobilisation.Physical encapsulation also played a crucial role in this process [62].The compressive strength of Pb-GDt 50 at 28 d measured 28.3 ± 0.7 MPa, indicating a slight increase compared to GDt 50 .This phenomenon was attributed to the consumption of OH − by Pb 2+ , which primarily affected the dissolution of diatom frustules without significantly altering MK's participation in geopolymerisation.Despite chemical composition alterations in gels, their quantity remained unchanged, supported by XRD analysis showing comparable areas of hump peaks representing the amorphous phase in both geopolymers.This is because MK is more prone to dissolution than diatom frustules due to T hydroxyl group removal, rendering aluminium in MK less stable; the alkalinity required for these Al element dissolutions is lower than that for the Si element in diatom frustules [63][64][65].Thus, the limited reduction in the alkalinity of the activator due to Pb 2+ was unlikely to have a significant effect on MK dissolution.Furthermore, the limited quantity of Pb(II) outside the frustules meant that even if they participated in the condensation, it did not reduce strength, as observed in other studies [66].This underscores the advantage of using MK-based alkali-activated geopolymers for effectively immobilising Pb(II)-rich Dt sludge, especially in achieving high compressive strength, a critical evaluation metric.

Leaching Behaviour
In order to gain a more comprehensive understanding of the stability of the geopolymer solidified body composed of sludge, which was formed from natural Dt at saturated adsorption levels, a series of leaching experiments were conducted to assess the leaching of Pb(II) from Pb-GDt 50 in water and HOAc leachate.The immobilisation rate (I) of the solidified body was calculated according to the following equation: where C 0 is the leaching concentration of the Dt sludge (mg/L), and C is the leaching concentration of the geopolymer solidified body (mg/L).The results are presented in Table 4.The immobilisation rates of Pb-GDt 50 were approximately 95% under the influence of both leachates, with a minor quantity of Pb(II) still precipitated.This is closely related to the pore structure and chemical state of Pb(II) within the solidified body [7,12].While specific pore structure characterisation was not conducted in this study, the presence of detected carbonates suggests the existence of interconnected open pores, potentially serving as pathways for carbon dioxide and Pb(II) migration.
Regarding the chemical state, it was observed that the immobilisation efficiency slightly decreased when HOAc was employed as the leachate in this study.Despite this, geopolymers remain a viable option for immobilisation, as the decrease in efficiency was more pronounced with cement in acidic leachates [67].Furthermore, geopolymers utilising solid waste rich in calcium as a principal raw material have shown poor acid resistance, attributed to calcium's role in compromising the stability of the solidified body [68,69].In contrast, the absence of calcium in the raw material used in this study ensures superior chemical stability, reducing the risk of heavy metal migration in acidic environments like landfill leachates and acidic rain.

Conclusions
The safe disposal of waste adsorbents presents significant environmental challenges.This study investigated the immobilisation mechanisms of Pb(II)-rich diatomite sludge through geopolymerisation, demonstrating its efficacy and potential applications.Natural diatomite was utilised as a source of active silica, leveraging its unique macroporous structure and silica hydroxyl groups to form robust bonds with geopolymer gels.The interaction between gels and the pores of diatom frustules created a 'mortise-and-tenon' structure, achieving a notable compressive strength of 26.2 MPa with a 50 wt.%replacement rate.By using this feature, geopolymer solidified bodies with a Pb(II)-rich diatomite sludge holding rate of 50% were prepared.The solidified bodies demonstrated a 95% immobilisation efficiency of Pb(II) in the presence of both pure water and acetate leachate.Through comparative mineralogical and microstructural analyses, it was observed that the geopolymerisation immobilised Pb(II) through three primary pathways: • The Pb(II) present in the pores of the undissolved diatom frustules underwent in situ binding with the activator, resulting in its encapsulation by the gels that were subsequently generated in the pores.• The dissolution of frustules released Pb 2+ , which then participated in the condensation of silicon and aluminium monomers in the formation of soluble hydroxide under high alkalinity.• The continuous consumption of the activator by metakaolin dissolution prompted the conversion of soluble hydroxides of Pb(II) to precipitates, which were encapsulated by the gels in the subsequent reaction.
The current study underscores geopolymerisation as an effective method for the disposal of Pb(II)-rich diatomite sludge, offering a novel approach to mitigate secondary pollution from failed diatomite adsorbent.Furthermore, the adequate compressive strength of the solidified bodies supports the development of environmentally sustainable building materials, leveraging toxic solid wastes as low-carbon footprint cementitious materials.
In summary, geopolymerisation represents a promising avenue for the environmentally sound management of hazardous waste materials, merging environmental stewardship with innovative materials science for sustainable development.

Figure 1 .
Figure 1.(a) XRD patterns of Dt and MK, (b) SEM image of Dt, and (c) SEM image of MK.Figure 1.(a) XRD patterns of Dt and MK, (b) SEM image of Dt, and (c) SEM image of MK.

Figure 1 .
Figure 1.(a) XRD patterns of Dt and MK, (b) SEM image of Dt, and (c) SEM image of MK.Figure 1.(a) XRD patterns of Dt and MK, (b) SEM image of Dt, and (c) SEM image of MK.

Figure 2 .
Figure 2. The compressive strength of Dt-based geopolymers at 7 d and 28 d with the error bar representing the upper and lower 95% confidence interval.

Figure 2 .
Figure 2. The compressive strength of Dt-based geopolymers at 7 d and 28 d with the error bar representing the upper and lower 95% confidence interval.

Figure 5 .
Figure 5. SEM images and EDS analysis results (indicated by the position of the circle) of (a) GDt50 and (b) Pb-GDt50 at 28 d.The Si/Al and Na/Al ratios were calculated based on the EDS results from at least 10 points per sample.

Figure 5 .
Figure 5. SEM images and EDS analysis results (indicated by the position of the circle) of (a) GDt 50 and (b) Pb-GDt 50 at 28 d.The Si/Al and Na/Al ratios were calculated based on the EDS results from at least 10 points per sample.

Author Contributions:
Conceptualisation, H.G.; methodology, Z.H. and B.Z.; formal analysis, Z.H.; resources, P.Y.; data curation, Z.H.; writing-original draft preparation, H.G.; writing-review and editing, B.Z., T.Y., and T.P.; project administration, H.G.; funding acquisition, H.G. and P.Y.All authors have read and agreed to the published version of the manuscript.Funding: This research was funded by the Guangdong Basic and Applied Basic Research Foundation (Grant No. 2021A1515110941), GDAS' Project of Science and Technology Development (Grant No. 2022GDASZH-2022010104, 2023GDASZH-2023010104), and the National Natural Science Foundation of China (Grant No. 42272043).

Table 1 .
Mix design.The liquid-to-solid (L/S) ratio was defined as the ratio between the mass of the 10 M NaOH solution and the total mass of the solid phase.

Table 3 .
The ANOVA test followed by Tukey's post hoc test and compressive strength grouped by the mass percentage of Dt.

Table 4 .
The immobilisation performance of geopolymers under different leachates.