Introduction

In the process of modern social development, a large amount of ordinary Portland cement (OPC) is consumed every day for the construction of various infrastructures. According to statistics, OPC production is growing at a rate of 9% per year worldwide [1]. The OPC production process, not only consumes mineral resources and fossil energy but also produces a large amount of greenhouse gases [2]. The study found that the CO2 content emitted for the production of OPC accounts for 8% of global anthropogenic CO2 emissions, and this proportion is still growing [3, 4]. Such a huge amount of carbon emissions will aggravate environmental damage and is not conducive to the sustainable development currently advocated. In this context, alkali-activated materials (AAMs) with low energy consumption and carbon emissions have attracted people’s attention.

AAMs are new cementitious materials formed by the reaction of alkali activators and precursors rich in silicon and aluminum [5]. The most common use of alkali activators is NaOH, Na2SiO3 and Na2CO3 or a combination of the three [6, 7]. The precursors for the preparation of AAMs are mostly composed of industrial solid waste (e.g., granulated blast furnace slag (GBFS) [8], fly ash (FA) [9], steel slag [10]) and mine solid waste (e.g., coal gangue [11], bauxite [12]). Compared with OPC, AAMs can not only significantly reduce carbon emissions and energy consumption [6], but also help to absorb various types of solid waste. In addition, AAMs also have many excellent properties such as good compressive strength [13], excellent durability [14], heavy metal adsorption capacity [15], fire resistance [16], and are considered to be one of the ideal substitutes for OPC.

Phosphorus slag (PS) is a by-product produced during the extraction of yellow phosphorus. Globally, large amounts of PS are emitted every year, with China alone emitting more than 8 million tons a year [17, 18]. Due to the lack of effective treatment methods, these PS are often directly accumulated outdoors, occupying land, and causing toxic ions to penetrate the soil under rainwater erosion, causing environmental pollution [19]. Thus, PS recycling is of great importance. PS consists primarily of calcium oxide (CaO) and silica (SiO2), with a small amount of alumina (Al2O3) and phosphorus pentoxide (P2O5) present, and the content of the amorphous phase is greater than 90% [20], indicating that it has potential reactivity and can be used to prepare AAMs. However, due to the presence of phosphorus and fluorine in PS, the early hydration of the slurry is hindered, and the lower Al2O3 content also makes the activity of PS lower than that of GBFS [21]. To this end, some scholars have tried to use different methods to improve the activity of alkali-activated phosphorus slag (AAPS). Xie et al. [19] found that high alkali concentration conditions can accelerate the dissolution rate of AAPS and increase the number of hydration products. Jia et al. [22] found that increasing the alkali concentration can increase the early strength of AAPS, but it affects the compactness of the matrix in the later stage, which has a negative effect on the strength. Maghsoodloorad et al. [23] found that the compressive strength of AAPS can be increased by more than twice when the curing temperature of AAPS is increased from 25 °C to 90 °C under the same curing time. It is not difficult to see from the above results that the performance of AAPS depends on high temperature and high alkaline concentration to get play. However, this undoubtedly hinders the application of PS as AAMs precursor.

FA and GBFS are industrial by-products emitted from power plants and ironworks, respectively [24, 25]. Their chemical composition contains active silicon-oxygen-aluminum components, which are the two most commonly used precursor materials for AAMs. AAMs prepared with GBFS/FA as a precursor can exhibit good workability performance, mechanical properties and durability at room temperature [26, 27]. Moreover, when GBFS/FA is combined with some inert precursor materials to prepare AAMs, it can synergistically hydrate and improve its performance. Sun et al. [28] prepared AAMs by using solid waste white mud (mainly composed of calcium carbonate) produced by the papermaking industry as a filler and GBFS/FA together. The results show that the 28 days strength is better than that of OPC slurry even when the content of white mud in the slurry reaches 50%, and it also has better early strength. Huseien et al. [29] replaced GBFS/FA with waste ceramic powder in a ratio of 50–70% to prepare AAMs with a 28-day compressive strength greater than 70 MPa. Song et al. [30] prepared AAMs by replacing FA with steel slag and GBFS in a certain proportion, and found that the amorphous gel of AAMs containing steel slag increased and the porosity decreased.

In addition to PS, other solid waste materials also have problems of poor workability or mechanical properties when they are used as precursors to prepare AAMs alone [31, 32]. It is considered to be the best way to solve this problem by using the synergistic coupling between different materials as precursors to improve each other’s performance [33]. Therefore, to realize the effective utilization of PS, alkali-activated PS-GBFS-FA (AAPGF) was prepared by using PS and GBFS/FA as precursors. The workability performance, mechanical properties and hydration products of AAPGF were investigated by replacing PS with different contents of GBFS/FA. The synergistic effect between GBFS/FA and PS was used to overcome the problem that AAPS was difficult to be effectively applied at room temperature and normal alkaline concentration. This paper makes up for the blank of PS and GBFS/FA as precursors to prepare AAMs, and also provides some reference value for the utilization of PS in AAMs.

Materials and methodology

Materials

The test PS was taken from Kunming Haifu Trading Co., Ltd (Yunnan Province, China), GBFS from Jiyuan Steel Mill (Henan Province, China), FA was taken from Yatai Co., Ltd (Liaoning Province, China). According to Table 1, PS and GBFS mainly contain calcium oxide (CaO), silica (SiO2) and alumina (Al2O3), PS exhibits a lower Al2O3 content and contains a small amount of phosphorus pentoxide (P2O5). FA is primarily composed of SiO2 and Al2O3, followed by CaO and iron oxide (Fe2O3). The particle size distribution was shown in Fig. 1. The median particle sizes (D50) of PS, GBFS and FA were 20.52 μm, 10.60 μm and 8.92 μm, respectively. The scanning electron microscopy (SEM) results of PS, GBFS and FA particles are shown in Fig. 2. PS and GBFS are irregularly angular, while FA is mostly spherical and a few are irregular. Figure 3 is the X-ray diffraction (XRD) test results. The image shows that both PS and GBFS have a large hump at 20–40° 2θ, which suggests the raw materials contain amorphous aluminosilicate compounds [20, 34]. There are calcite, quartz and magnesium oxide in the crystal phase of PS. The crystal phase of GBFS is mainly consist of gehlenite, dicalcium silicate and calcite. The phase of FA is mainly composed of mullite, quartz and a small amount of hematite, magnetite and other crystal phases, and the content of amorphous phase is low. The more crystal phase of FA is the reason for its weak hydration activity [24]. However, this characteristic proves beneficial in mitigating the swift condensation of GBFS, establishing a favorable coupling effect [27]. The standard sand for the test was taken from Xiamen City, Fujian Province, the particle size distribution and XRD analysis are shown in Ref [35]. The NaOH from Kemao Co., Ltd (Tianjin, China), with a purity of more than 96%. Na2SiO3 was obtained from Gongyi Borun Refractories Co., Ltd (Henan Province, China). The purity of Na2SiO3 is more than 98.5% and the modulus is 2 M. The two solutions were mixed as an activator.

Table 1 Chemical composition and content of materials (wt%)
Full size table
Fig. 1
figure 1

Particle size distribution of raw materials

Full size image
Fig. 2
figure 2

SEM images of raw materials: (a) PS, (b) GBFS, and (c) FA

Full size image
Fig. 3
figure 3

XRD patterns of raw materials

Full size image

Sample preparation

Alkali-activated PS-GBFS-FA (AAPGF) was prepared by replacing PS with GBFS/FA in the ratio of 30%, 40% and 50% denoted as PFG-0–PGF-3 based on varying GBFS/FA masses. In the precursor materials of AAPGF, since both PS and GBFS are high calcium materials, excessive calcium content has an adverse effect on the workability of AAMs [34]. To optimize the element content in the system, the GBFS content in GBFS/FA is kept unchanged, and the FA content is increased (accounting for 10–30% of the total mass of the cementitious material).

Alkali activator is an important part of AAMs. Caldas et al. [32] and Marvila et al. [36] found that the type of alkali activator and sodium content have an important influence on the performance of AAMs. Therefore, according to the previous research of the research group [37], NaOH and Na2SiO3 were used as composite alkali activators, and the modulus was fixed at 1.5, and the water-binder ratio was 0.35. The specific ratio is shown in Table 2. Notably, the negligible Na2O content in GBFS/FA (Table 1) renders the impact of its addition on the alkali concentration in the reaction inconsequential. The treated materials were poured into the mixer according to GB/T 17,671–2021 for preparation. Subsequently, the stirred slurry was injected into the 40 mm × 40 mm × 160 mm steel mold twice. Initially, approximately half of the slurry was injected and vibration commenced. After the compaction, the remaining slurry is filled into the mold to start the second compaction. After the compaction is completed, the surface is flattened with a scraper. Curing to the specified age at a temperature of 20 ± 1 °C and RH ≥ 95%.

Table 2 Mix design (g)
Full size table

Test methods

Workability

The fluidity test, conducted in accordance with GB/T 2419–2005, utilized NLD-3 type jump table (China, Zhejiang Dadi Instrument Co., Ltd.) for testing. The mixed slurry was poured into a mold placed on the flow table and tamp with a tamper. Subsequently, the mold was lifted, and the flow table was initiated. Following 25 vibrations of the flow table, the average diameter of the vertical two directions was tested as the final result. The test process is shown in Fig. 4.

Fig. 4
figure 4

Fluidity test process

Full size image

The setting time according to GB/T 1346–2011, using a standard Vicat instrument (China, Shaoxing Exploration Instrument Factory) test. The prepared slurry was poured into the mold. The initial setting needle is immersed into the slurry from a height close to the surface of the slurry. When the test needle is 4 ± 1 mm away from the bottom of the mold and no longer sinks, the slurry reaches the initial setting state. Subsequently, the mold was flipped, and the slurry was deemed to have reached the final setting state when the final setting needle sank to a distance less than 5 mm into the slurry.

Mechanical properties

According to GB/T 17,671–2021, using WAW-1000 pressure testing machine (China, Shanghai Hualong Testing Instrument Co., Ltd.), the prism sample with a size of 40 mm × 40 mm × 160 mm was broken from the middle part and the compressive strength test was carried out. The results are valid averages of at least six samples.

Hydration

The main phase composition was tested by XRD. The instrument used is the D8 Advance X-ray diffractometer (Bruker, Germany). The operating voltage and current are 45 Kv and 40 mA respectively. The scanning range and speed are 5–70° 2θ and 5°/min respectively. The material functional groups and element bonding of the samples were analyzed by Fourier transform infrared spectroscopy (FTIR). The instrument used is a Nicolet iS50 FTIR Spectrometer, using a conventional KBr tablet test, test wave number range of 400–4000 cm− 1. The microstructure of the samples was analyzed by SEM and energy dispersive spectroscopy (EDS). The instrument used was Zeiss Gemini 300. Before testing, samples underwent slicing, resulting in a diameter of ≤ 1 cm and a thickness < 1 cm. The smooth surface was selected as the test surface. Subsequently, the sample surface was sprayed with gold (30 mA, 150 s) and put into the instrument for vacuum treatment.

Results and discussion

Fluidity

Figure 5 shows the AAPGF fluidity test results. According to the requirements of GB 175–2007, the fluidity should not be less than 180 mm. All groups, except for the PGF-0 group meet this requirement, and the addition of GBFS/FA has a good improvement effect on the fluidity of AAPGF.

Fig. 5
figure 5

AAPGF fluidity test results with different GBFS/FA contents

Full size image

As seen from the SEM image of PS, the PS particles have irregular shapes, more angular particles and larger particle size (D50 = 20.52 μm), which increases the friction between particles, thus the PGF-0 group has the lowest fluidity. Compared with PS, GBFS (D50 = 10.6 μm) and FA (D50 = 8.92 μm) have smaller particle sizes. They are distributed between PS particles, thereby improving the particle size distribution of AAPGF and enhancing its fluidity [38].

In addition, although the high amorphous phase content of GBFS can increase the early reaction rate and accelerate the consumption of mixed water, thereby reducing fluidity [39]. However, FA exhibits minimal involvement in the reaction during the early stages. The angular-free spherical particles can exert a morphological effect [40] and play a leading role in enhancing fluidity [41].

Setting time

Figure 6 shows the AAPGF setting time test outcomes. The incorporation of GBFS/FA greatly shortens the setting time of AAPGF leading to flash setting.

Fig. 6
figure 6

AAPGF setting time test results with different GBFS/FA contents

Full size image

This phenomenon is mainly related to the rapid dissolution of GBFS in an alkaline solution. GBFS swiftly dissolve the necessary Ca2+ ions for the reaction in the alkaline solution, providing more nucleation sites for the remaining dissolved substances [24]. This process promotes the formation of C-(A)-S-H gel, accelerate the hydration reaction, and shorten the setting time. It is found that the incorporation of FA into GBFS can improve the flash setting phenomenon caused by GBFS [42]. However, even with a 30% FA content in AAPGF, the flash setting phenomenon has not been alleviated. This is attributed to both PS and GBFS in AAPGF have higher CaO content, so the hydration reaction rate at room temperature is higher than that of FA [43, 44]. Consequently, a certain amount of FA has limited impact on the setting time.

Compressive strength

Figure 7 shows the results of the compressive strength tests for AAPGF. The compressive strength of the PGF-0 group stimulated by PS alone was lower. The compressive strength from 1 day–28 days was 4.2 MPa, 5.7 MPa, 8.51 MPa and 30.09 MPa, respectively. Significantly lagging behind other groups. Moreover, due to the presence of P2O5 in PS, the early strength development in the PGF-0 group was slower than that in the later period.

The presence of P2O5 results in the generation soluble phosphate ions (PO43−) in the solution. When PO43− reacts with Ca2+ ions, it forms insoluble substances, reduce the concentration of Ca2+ ions, and adhering to the surface of the particles [45]. This process adversely affects the normal progress of the hydration reaction. Moreover, the lack of Al element in PS is also one of the factors causing the low strength of PGF-0 group. Following the incorporation of GBFS/FA, the compressive strength of the PGF-1–PGF3 group was significantly improved. This enhancement is attributed to the ability of GBFS to rapidly dissolve Ca2+ and Al3+ ions in an alkaline solution, forming C-(A)-S-H gel. This can not only reduce the adverse effects of P2O5, but also supplement Al element for AAPGF, thus ensuring the development of strength.

Fig. 7
figure 7

AAPGF compressive strength test results with different GBFS/FA contents

Full size image

The PGF-1 group, with a 30% GBFS/FA content, exhibited the highest strength, reaching 72.65 MPa at 28 days. However, when the content of GBFS/FA is more than 30%, the compressive strength of each age decreases. This is related to the increase in the proportion of FA substituted PS in AAPGF. The activity of FA is low, and it basically exists in the form of filler in the system. Over time, although some active substances can precipitate Si4+ and Al3+ ions to form N-A-S-H gel [46], the number of its formation is extremely limited at room temperature, and most of them still exist in the form of inert filler. On the contrary, despite the relatively slow early reaction of PS, it gradually dissolves and form C-S-H gel in the later stage of hydration [21], ensuring the growth of strength. This also proves that in AAPGF, the hydration activity of PS is greater than that of FA.

XRD analysis

Figure 8 shows the XRD results of AAPGF. The primary phases identified encompass C-(A)-S-H gel, Portlandite, hydrotalcite and phase composition in raw materials. The N-A-S-H gel exhibits amorphous properties [47], and consequently, no relevant characteristic peaks were found in the XRD spectrum.

Fig. 8
figure 8

XRD patterns of AAPGF 28 at days with different GBFS/FA contents

Full size image

There is a broad hump in each group near 20–35° 2θ, indicating that there is an amorphous phase. Part of the amorphous phase is derived from raw materials, and the other part is newly generated by hydration reaction [48]. The addition of GBFS/FA does not significantly alter the hump area, so the strength difference could not be evaluated by the amorphous phase content. There are obvious C-(A)-S-H gel characteristic peaks in each group near 29° 2θ. The existence of C-(A)-S-H gel peak in PGF-0 group confirmed the hydration ability of PS in the later stage. Following the adding GBFS/FA, the C-(A)-S-H gel peak was enhanced, with the peak value reaching its maximum in the PGF-1 group. The Ca2+ and Al3+ ions decomposed from GBFS promoted the formation of C-(A)-S-H gel. However, with the increase of the proportion of low activity FA replacing PS, the C-(A)-S-H gel characteristic peaks in PGF-2 and PGF-3 groups gradually weakened, which was consistent with the strength test results.

In addition, unlike the PGF-0 group, a new crystal peak of hydrotalcite was found in PGF-1–PGF-3 groups. Hydrotalcite is the hydration product of Mg, Al and carbonate (CO32−) [47], forms due to the active Mg and Al elements present in GBFS/FA. Consequently, the incorporation of GBFS/FA improves the compressive strength by augmenting the number of gels such as C-(A)-S-H and hydrotalcite. In addition, with the increase of FA substitution ratio in AAPGF, the mullite and quartz phase gradually become obvious. This suggests that these two phases are relatively stable and not easy to react with other substances.

FTIR analysis

Figure 9 shows the FTIR results of AAPGF. There is a broad peak at 980–1001 cm− 1 corresponding to the antisymmetric stretching vibration of Si-O-T (T = Si, Al) bonds in the gel [49]. In comparison to the PGF-0 group, the wave number of the PGF-1–PGF-3 group decreased from 1001 cm− 1 to 980 cm− 1. The Al-O bond has a lower vibrational frequency than the Si-O bond [50], the decrease of wavenumber indicates a greater substitution of Si by Al in the reaction product, leading to the formation of more Al-containing gels.

Fig. 9
figure 9

FTIR spectra of AAPGF at 28 days with different GBFS/FA contents: (a) 400–4000 cm− 1 and (b) 600–1200 cm− 1

Full size image

It has been observed that when the wavenumber of the system is higher than that of C-A-S-H gel (950 cm− 1) and lower than that of N-A-S-H gel (1020 cm− 1), there were many types of gel products [28]. The wave number of the PGF-1–PGF-3group shifted from 980 cm− 1 to 996 cm− 1, and the wave number increased continuously and approached 1020 cm− 1. This trend suggests an augmented production of N-A-S-H gels. Usually, the SiO4 tetrahedra in the gel have Qn units (n = 0–4) due to the different numbers of bridging oxygen. The wave numbers of Q0–Q4 are located near 850 cm− 1,900 cm− 1,950 cm− 1,1100 cm− 1 and 1200 cm− 1, respectively [51]. The PGF-0–PGF-3 group produced an absorption peak near 950 cm− 1, signifying that the SiO4 tetrahedron in the product C-(N)-A-S-H gel mainly existed in the form of Q2 unit.

The band peaks near 1640 cm− 1 and 3450 cm− 1 are related to the bending vibration and tensile vibration of O-H in crystal water, respectively [52]. Compared with the PGF-0 group, the PGF-1 group exhibited the largest peak area and the highest strength, indicating that the most hydration products were generated. However, a notable reduction was observed in the PGF-2 and PGF-3 groups, attributed to the formation of N-A-S-H gel with low bound water content [53]. The band peaks near 1424 cm− 1 and 874 cm− 1 were related to the asymmetric tensile vibration of the O-C-O bond, indicating the occurrence of the carbonization reaction [51]. The band peak at 713 cm− 1 is the bending vibration of the Si-O-Al bond induced by the substitution of Si by tetracoordinate aluminum [Al (IV)] [54]. This observation further confirms the presence of Al-containing gel in the hydration product. Additionally, the band peak near 456 cm− 1 is caused by the bending vibration of the Si-O-Si bond and Si-O bond in the hydration product gel [55].

SEM-EDS analysis

Figure 10 is the SEM image of AAPGF. In the PGF-0 group, unhydrated PS particles with angular shape are observable, and the number of hydration products is limited, with insufficient overlap. Simultaneously, a certain number of pores are evident. Upon the addition of GBFS/FA, it is difficult to observe the presence of unhydrated particles in the PGF-1–PGF-3 group. Moreover, more large-sized hydration products are formed around the particles. This observation indicates that GBFS/FA can promote the dissolution of PS and increase the number of hydration products. However, it is noted that some large pores are evident in the group mixed with GBFS/FA. This is related to the high activity of GBFS in the early stage of hydration, resulting in uneven distribution of hydration products and subsequently increasing the gap between particles [37].

Fig. 10
figure 10

SEM images of AAPGF at 28 days with different GBFS/FA contents: (a) PGF-0, (b) PGF-1, (c) PGF-2, and (d) PGF-3

Full size image

The EDS energy spectrum data of AAPGF at 28 days are shown in Table 3 representing the average values from six points in different regions. The chemical composition of the hydration products is mainly composed of Ca, Si, Al and Na. The presence of these elements confirms the co-existence of C-(N)-A-S-H gel and N-A-S-H gel in AAPGF [34], which is consistent with the FTIR results. The element distribution in the PGF-0 group is mainly composed of Ca and Si elements, with minimal Na and Al elements, indicating that the hydration products at this time are mainly C-S-H gel. Following the addition of GBFS/FA, both the Ca/Si ratio and Si/Al ratio gradually decreased, and the Na/Ca ratio gradually increased. The increase of the proportion of GBFS/FA replacing PS will gradually reduce the concentration of Ca2+ ions and an increased release of Al-O bonds. In response, Na+ ions are adsorbed into the Si-O-Al bond, forming C-(N)-A-S-H and N-A-S-H gels with a high degree of polymerization to balance the charge [56].

Table 3 EDS energy spectrum data of AAPGF at 28 days (at%)
Full size table

Conclusion

To make effective use of PS solid waste, AAPGF was prepared by using PS and GBFS/FA as precursors. The effects of different GBFS/FA contents on the workability performance, mechanical properties and hydration products of AAPGF were investigated. The conclusions are as follows:

  1. 1.

    GBFS/FA increases fluidity by optimizing the particle distribution of AAPGF and exerting morphological effects. Due to the high activity of GBFS, a large amount of C-(A)-S-H gel can be generated in the early stage of hydration, resulting in the flash setting of AAPGF, and the increase of FA ratio cannot alleviate the flash setting caused by GBFS.

  2. 2.

    GBFS/FA increases the strength of AAPGF. When the replacement ratio is 30%, the 28 days strength can reach the highest 72.65 MPa. PS can react to form C-S-H gel in the later stage of hydration, so the increase of the proportion of FA replacing PS leads to the decrease of compressive strength.

  3. 3.

    GBFS/FA increased the amount of C-(A)-S-H gel in AAPGF, accompanied by the formation of hydrotalcite gel. In addition, GBFS/FA also increases the content of Al and Na elements, transforms the hydration products from C-S-H gel to C-(N)-A-S-H and N-A-S-H gel, and improves the degree of polymerization. The SiO4 tetrahedron mainly exists in the form of Q2 unit.

  4. 4.

    GBFS/FA significantly reduced the number of unhydrated AAPGF particles. However, the high activity of GBFS in the early stage will lead to uneven distribution of hydration products and the formation of large pores.