Novel Alkali Extraction–Acid Precipitation–Chlorination Distillation Method for Co-extraction of Humic Substances and Germanium (Ge) from Ge-Rich Lignite

Abstract

The traditional combustion–chlorination distillation process used for extracting germanium (Ge) from lignite not only causes significant environmental pollution but also wastes the high-value humic substances (HSs), especially fulvic acids (FAs) in lignite. This study attempts to comprehensively recover Ge, humic acids (HAs), and FAs from lignite using an alkali extraction–acid precipitation–chlorination distillation (AEAPCD) method. The optimized process parameters are obtained as alkali extraction at 100 °C for 2 h using 0.3 mol/L NaOH as the extractant and a liquid–solid ratio of 9:1, acid precipitation of extracted liquid using HCl to adjust the pH value to 1, and the chlorination distillation of Ge-rich FA solution at 95 °C for 30 min. In this case, the recovery yields of HAs and FAs in their respective products are 88.83 and 91.42%, respectively, and the Ge recovery yields of GeCl₄, HAs, FAs, and alkali extraction residues are 64.95, 5.18, 9.91, and 19.96%, respectively. The mechanism study shows that germanium in lignite is mainly bound to HSs, followed by other organic compounds (most likely lignin-like substances). During alkaline extraction of lignite, HSs (especially FAs) first undergo neutralization reaction with NaOH to form soluble sodium salts, and the bound germanium is also extracted. As the concentration of NaOH solution increases, lignin-like substances undergo hydrolysis, generating a large amount of small-molecule soluble substances, such as p-coumaryl, coniferyl, and sinapyl alcohols, or their derivatives, dimers, trimers, etc., and germanium bound to them is also extracted. In addition, purification of crude FAs and HAs was also attempted in this work.

1. Introduction

Germanium (Ge), a highly valuable dispersed element, plays an important role in many high-tech fields, such as infrared optics, fiber optic communication, polymer catalysts, semiconductor materials, and solar panels. The global demand for germanium is growing at a rate of nearly 20% per year. , Many countries including the U.S.A, China, European Union have included Ge in their critical mineral list in light of their role in national security or economic development. ,

At present, more than half of germanium comes from lignite mines, especially those located in Lincang (Yunnan province), Wulantuga (Inner Mongolia) in China and Spetzugli (Primorye) in Russia. The average Ge content in these three coal-hosted Ge deposits is 850, 240, and 1025 g/t, extraordinarily higher than the world’s average Ge concentration of lignite (2 g/t). , Currently, the industrial mainstream method for extracting Ge from lignite is based on the combustion–chlorination distillation (CCD) method. It involves the vaporization of Ge into coal fly ash to obtain a rude Ge concentrate with a Ge grade of 0.3–1.2% by coal combustion, and then, the fly ash is leached by concentrated HCl and distilled to obtain a GeCl₄ concentrate. − Such techniques generate a large amount of waste residue after acid leaching and acidic waste liquid after distillation. Moreover, due to the high content of moisture, heavy metals (such as Pb, Cd, Cr) and low calorific value of lignite, the power generation efficiency is lower and the released toxic substance during combustion causes severe environmental pollution. , Meanwhile, the highly valuable humic substances (HSs) in lignite are wasted in the CCD process, with a price of up to US$ 1000 per metric ton.

Based on the solubility, HSs are commonly divided into alkaline soluble but acid insoluble humic acids (HAs) and fulvic acids (FAs), which can be dissolved at all pH values. They are both macromolecular substances composed of various complex aromatic components and aliphatic carbon chains and contain a large number of active oxygen-containing functional groups, such as carboxyl, carbonyl, phenolic hydroxyl, quinone, etc. , These structural features grant them weak acidity, redox properties, ion-exchange capacity, a strong complexation ability with metal ions, and colloidal properties, − resulting in various applications in many fields, such as agriculture (enhancing soil health, stimulating plant growth and enhancing their disease resistance), environmental remediation (restoring polluted environments by binding toxins, enhancing microbial diversity), livestock and poultry food (improving animal health and growth by enriching feed quality), fertilizer industry, wastewater treatment, biofuel production, energy storage, pharmaceuticals and cosmetics, etc. ,− Currently, commercial HAs and FAs are mainly extracted from leonardite and lignite by the method of alkaline extraction followed by acid precipitation, although it inevitably brings a certain degree of chemical pollution. , In addition, there are some emerging methods, such as hydrothermal extraction, enzymatic extraction, acid pretreatment, oxidation pretreatment, supercritical fluid extraction, microwave-assisted extraction, ultrasound-assisted extraction, etc. ,,

Germanium is mainly bound with organic matter of lignite, and very little germanium exists in the mineral form. ,, It preferentially associates with HSs, which has been confirmed by direct or indirect analysis. − Germanium can form organic complexes with phenolic hydroxyl, carboxyl, and hydroxyl groups in HAs or FAs, which are generally believed to be the main occurrence modes of germanium in lignite. Studies on the modes of Ge occurrence in lignite have shown that HSs and Ge can be extracted simultaneously from lignite using alkaline treatment. For example, Yakushevich et al. used a 0.5 M NaOH solution to extract Ge from lignite (with a Ge content of 592 g/t) from the Spetsugli area of the Pavlovskoe coalfield. 86% of germanium was enriched in the sodium humate solution. Zhang et al. used 2% NaOH solution to leach lignite from the Bangmai area of Lincang, and the results showed that about 75–96% of germanium entered the sodium humate solution. Wu et al. found that about 91% of germanium entered the sodium humate solution using a mixed solution of 1% NaOH + 0.1 M Na₄P₂O₇ to leach lignite from Lincang. Overall, the CCD technique routes currently used for Ge recovery from lignite have severe problems of low efficiency and environmental pollutions. The alkaline leaching method shows great advantages for the co-extraction of high-valued Ge and HSs from lignite. However, the optimum extraction conditions, leaching, and separation mechanisms of Ge and HSs from lignite remain unclear.

In the current work, a novel alkali extraction–acid precipitation–chlorination distillation (AEAPCD) method was proposed for co-extraction of Ge, HAs, and FAs from Ge-rich coal. First, an alkaline solution was used to coextract Ge and HSs from lignite. Then, the separation of FAs and HAs was carried out using the acid precipitation method. The distribution of germanium between FAs and HAs was investigated. Finally, germanium was separated from FAs and HAs by using distillation. The effects of alkaline extraction, acid precipitation, and distillation conditions on the recovery yields of Ge, FAs, and HAs were intensively examined as a function of the concentration of the extractant, extraction temperature, solid-to-liquid ratios, etc. In addition, the purification of crude HAs and FAs was conducted in this work, and the mechanism of enhancing germanium extraction by increasing the concentration of alkaline solution was also explored using sequential chemical extraction experiments (SCEEs), LC–MS, and FTIR. Compared to the CCD process, the AEAPCD method can comprehensively recover the high-value components of Ge-rich lignite and avoid a series of environmental pollution problems caused by combustion and HCl leaching.

2. Experimental Section

2.1. Raw Materials and Chemicals

2.1.1. Lignite Sample

Bulk coal samples of approximately 20 kg were acquired from a local coal-hosted Ge mine in Wulantuga, Inner Mongolia. The coal sample was sequentially crushed by a jaw crusher and a rod mill, followed by sieving with a square-hole screen (1 mm). After drying at room temperature, the sieved coal particles were divided and packaged with 20 g in each sealed bag. Finally, all coal samples were stored in a dry container for subsequent use. The results of ultimate analysis, proximate analysis (including moisture (M), ash (A), volatile matter (V), and fixed carbon (FC), all expressed on an air-dried basis (AD)), Ge content, and HA and FA contents of the coal sample are shown in Table . Ultimate analysis was carried out using an EL cube CHNSO elemental analyzer (Elementar Corp., Hanau, Hessen, Germany). The determination of HA and FA contents is referred to Chinese standards GB/T 34766–2017 and GB/T 34765–2017, respectively. The general steps are as follows: using a mixed alkali solution of Na₄P₂O₇ and NaOH to extract HAs and FAs from lignite, the extraction solution was precipitated with H₂SO₄, filtered, and the filtrate was oxidized with K₂Cr₂O₇ under strong acid conditions. Excess K₂Cr₂O₇ was titrated with a standard (NH₄)₂Fe­(SO₄)₂ solution, and the mass fraction of carbon in FAs (relative to the mass of the lignite sample) was calculated based on the consumption of K₂Cr₂O₇. Then, this value was multiplied by 0.59 to obtain the FA content. After dissolution by dilute NaOH solution, the filter residue underwent the same steps as described above to obtain the HA content of lignite. The determination of the Ge content was carried out by using an inductively coupled plasma optical emission spectrometer (PerkinElmer Optima 8300). Prior to ICP-OES determination, the coal sample was subjected to ashing and digested with mixed acid (HNO₃+H₃PO₄+HF) according to the Chinese standard (GB/T 8207–2007).

1. Proximate Analysis, Ultimate Analysis, Ge Content, and HA and FA Contents of the Lignite Sample .

proximate analysis (wt%)							ultimate analysis (wt%)
Mad	Aad	Vad	FCad	HAs (wt%)	FAs (wt%)	Ge (g/t)	Cdaf	Hdaf	Odaf	Ndaf	Sdaf
13.26	10.40	33.19	43.15	7.00	6.19	249.90	68.28	8.69	19.11	1.67	2.25

^aNote: ad: air-dried basis; daf: dry ash free basis.

^b-by difference.

As shown in Table , the coal sample is characterized by high volatile matter, markedly high oxygen and hydrogen contents, and low carbon contents, classifying it as a typical lignite with moderate sulfur contents (2.25%). Particularly, it contains an exceptionally high germanium content, thus making it suitable as a feedstock for germanium extraction. In addition, this lignite sample also contains 6.19% FAs and lower contents of HAs (7%), which may be related to the coalification degree and coal-forming environment. The HA contents of Xilinhot and Huolinhe lignite were 14.9% and 11.25–14.6%, respectively, determined by Du et.al. and Cheng et al.; another lignite from Zalainuer contains 14.55% HAs. Similar to the three lignites, Wulantuga lignite in this work is produced in the eastern region of Inner Mongolia, and perhaps it has similar coal-forming environments and coalification degrees. According to the Chinese coal classification standards (GB/T 5751–2009), they all belong to old lignite with a light transmittance of 30%, while those with a light transmittance of >30–50% belong to young lignite. These two types roughly correspond to sub-bituminous coal and lignite in the United States. The lignites from Zhaotong and Xundian in Yunnan Province, China, have HA contents of 44.34 and 53.28%, respectively, both of which belong to young lignite. , Another factor that affects the lignite HA content is the determination method. There are currently multiple methods for determining HA contents, such as ISO5073:2021, GB/T11957–2001, GB/T 34766–2017, ISO19822, and the method provided by IHSS. ,, These methods all follow the principle process of alkali extraction and acid precipitation, but there are differences in specific operating parameters such as liquid–solid ratio, reagent concentration, dehydration method (filtration or centrifugation), alkali extraction, acid precipitation time, etc. Especially, the first three methods use K₂Cr₂O₇ oxidation + (NH₄)₂Fe­(SO₄)₂ titration to indirectly determine HA contents, while the latter two methods use HCl and HF to remove the ash in HAs to directly determine HA contents. In fact, the HA content of lignite from different origins varies between 1 and 85%, most of which are several tens of percent; however, there are also a few lignite with lower HA contents; for instance, the lignite from Bowman, North Dakota, USA, only contains 3.9% HAs. For Wulantuga lignite, although its HA content is only 7%, the coal sample contains a considerable amount of FAs (6.19%), the price of which is several times or even dozens of times that of HAs. This indicates that the joint extraction of HSs and Ge from Ge-rich lignite has a promising economic significance.

2.1.2. Chemicals

Chemicals of the analytical reagent grade as well as deionized water were used in the experiments. K₂Cr₂O₇, Na₄P₂O₇, NaOH, H₃PO₄, HF, and (NH₄)₂Fe­(SO₄)₂ were purchased from Tianjin Yongda Chemical Reagent Co., Ltd., HNO₃, H₂SO₄, HOCl₄, NH₄Cl, and H₂O₂ were purchased from Chengdu Kelong Chemical Reagent Co., Ltd., and CH₃COOH and CH₃COONa were purchased from Suzhou Dongbai Chemical Co., Ltd. The sodium type 732 resin and D401 resin used for purifying FAs were purchased from Basle Chemical Technology (Tianjin) Co., Ltd. Sodium type 732 resin can be transformed into hydrogen type 732 resin by soaking in 3% HCl for 4 h and rinsing with deionized water until neutral.

2.2. Process Flow of Co-extraction and Separation of Ge, HAs, and FAs from Lignite

Figure shows the process used for the corecovery of Ge, HAs, and FAs from lignite in this study. As mentioned above, alkaline extraction was first used to simultaneously extract HSs and Ge from lignite. Acid precipitation is a classic method for separating HAs and FAs according to their solubility differences in different pH solutions. HCl was selected as the reagent, because it can react with Ge-containing substances in HSs to form GeCl₄ that is easily separated from crude HAs and FAs using the chlorination distillation method , The remaining HAs and FAs after distillation were purified using water washing and ion-exchange methods, respectively.

1.

Process flow of co-extraction and separation of Ge, HAs, and FAs from lignite.

2.3. Alkali Extraction of HSs and Ge from Lignite

A series of single factor experiments on the alkali extraction process were conducted to optimize the operating parameters for Ge and HS separation from lignite. These parameters include the concentration of the extractant (0.1–0.5 mol/L, with a gradient of 0.1 mol/L), extraction temperature (80–120 °C, with an interval of 10 °C), liquid–solid ratio (6:1–10:1, with an interval of 1:1), and extraction time (0.5–2.5 h, with an interval of 0.5 h). Each experiment was conducted using NaOH as the extractant, a constant temperature water bath for heating (when the temperature exceeded 100 °C, a high-pressure reactor was used), and 2 g of lignite as the raw material. After extraction, the slurry was filtered and divided into the filter residue and filtrate (i.e., humate solution). The residue was washed multiple times with deionized water until the filtrate droplets were neutral. The recovery yields of Ge and HSs in the filtrate were obtained using eqs and , respectively:

$${\epsilon }_{\mathrm{Ge}‐\mathrm{Fi}}=(1-\frac{{\gamma }_{\mathrm{Re}}\times {\theta }_{\mathrm{Ge}‐\mathrm{Re}}}{249.9})\times 100\%$$ 1

$${\epsilon }_{\mathrm{HS}‐\mathrm{Fi}}=(1-\frac{{\gamma }_{\mathrm{Re}}\times {\theta }_{\mathrm{HS}‐\mathrm{Re}}}{13.19})\times 100\%$$ 2

where ε_Ge‑Fi and ε_HS‑Fi represent the recovery yield of Ge and HSs in the filtrate, respectively, %; γ_Re represents the yield of the filter residue, expressed as a decimal; θ_Ge-_Re is the Ge content of the filter residue, g/t; and θ_HS-_Re represents the HS content of the filter residue, %, the determination of which is based on ISO 5073–2021. The numbers 249.9 and 13.19 in eqs and refer to the Ge content and HS content of the lignite sample, respectively.

2.4. Separation of HAs and FAs by the Method of Acid Precipitation

The filtrate obtained after alkaline extraction and filtration is a mixed solution of HA and FA sodium salt. The pH value of the solution can be adjusted by adding acid to coagulate and precipitate HAs, which are insoluble in acidic solutions. After filtration, the separation of HAs and FAs can be achieved. HCl was used as a pH adjuster to investigate the recovery yields of HAs and FAs at four pH values (1, 2, 3, and 4). The recovery yields of HAs, FAs, and Ge can be obtained using eqs –:

$${\epsilon }_{\mathrm{HA}}=\frac{{\gamma }_{\mathrm{HA}}\times {\beta }_{\mathrm{HA}}}{7}\times 100\%$$ 3

$${\epsilon }_{\mathrm{FA}}=\frac{{\gamma }_{\mathrm{FA}}\times {\beta }_{\mathrm{FA}}}{6.19}\times 100\%$$ 4

$${\epsilon }_{\mathrm{Ge}‐\mathrm{HA}}=\frac{{\gamma }_{\mathrm{HA}}\times {\beta }_{\mathrm{Ge}‐\mathrm{HA}}}{249.9}\times 100\%$$ 5

$${\epsilon }_{\mathrm{Ge}‐\mathrm{FA}}={\epsilon }_{\mathrm{Ge}‐\mathrm{Fi}}-{\epsilon }_{\mathrm{Ge}‐\mathrm{HA}}$$ 6

where ε_HA and ε_FA represent the recovery yield of HAs and FAs, respectively, %; γ_HA and γ_FA represent the yields of crude HAs and crude FAs, respectively, expressed as a decimal; β_HA refers to the total HA content of the crude HA product, %; β_FA refers to the FA content of the crude FA product, %; ε_Ge‑HA and ε_Ge‑FA represent the recovery yields of Ge in crude HA and FA products, respectively, %; and β_Ge‑HA refers to the Ge content of the crude HA product, g/t. The numbers 7, 6.19, and 249.9 in eqs – refer to the contents of HAs, FAs and germanium of the lignite sample, respectively.

2.5. Separation of Germanium from Crude FAs

This work only studied the germanium separation from crude FA solution, while that from crude HA solution was not considered because of its low Ge recovery yield (mentioned in section ). Crude FA solution was adjusted to pH 1 with HCl and then distilled in a water bath at 95 °C to evaporate Ge into the receiver in the form of GeCl₄. The experimental setup is shown in Figure . The Ge volatilization efficiency and Ge recovery yield under different distillation times are calculated using eqs and .

$${\mathrm{E}}_{\mathrm{Ge}‐\mathrm{Di}}=(1-\frac{100\times {\theta }_{\mathrm{Ge}‐\mathrm{Di}‐\mathrm{Re}}\times {\gamma }_{\mathrm{FA}}}{249.9\times {\epsilon }_{\mathrm{Ge}‐\mathrm{FA}}})\times 100\%$$ 7

$${\epsilon }_{\mathrm{Ge}‐\mathrm{Di}}={\epsilon }_{\mathrm{Ge}‐\mathrm{FA}}-\frac{100\times {\theta }_{\mathrm{Ge}‐\mathrm{Di}‐\mathrm{Re}}\times {\gamma }_{\mathrm{FA}}}{249.9}$$ 8

where E _Ge‑Di and ε_Ge‑Di represent the Ge volatilization efficiency and Ge recovery yield of the GeCl₄ product, respectively, %; θ_Ge‑Di‑Re represents the Ge content of crude FAs (dry basis) after chlorinated distillation. The value of 249.9 in eqs and refers to the Ge content of the lignite sample.

2.

Distillation equipment for germanium volatilization from crude FAs.

2.6. Purification of HAs and FAs

Water washing is commonly used to purify HAs, as HAs were mainly mingled with some soluble salts, especially NaCl, but HAs are insoluble in acidic and neutral environments. This mainly involved dispersing the crude HAs with an appropriate amount of deionized water, stirring thoroughly, centrifuging (4000 rpm for 15 min), pouring out the supernatant, and repeating this process until the color of the supernatant changed from yellow to colorless. The purity of the HA product (marked as P _HA) was determined by burning at 750 °C, which can be calculated according to eq. .

$$p_{\mathrm{HA}}=\frac{m_0-m_i}{m_0}\times 100\%$$ 9

where m ₀ represents the quality of the HA product before burning, %; m _i represents the quality of the residue after burning, %. Additionally, elemental analysis of HAs before and after purification was conducted.

The purification of the crude FA product adopted the method used by Zhang et.al. shown in Figure . The process is briefly introduced as follows: The crude FA solution was adjusted to pH 5, left at 4 °C for 12 h, centrifuged (4000 rpm, 10 min), and the precipitate was dried at 50 °C. The pretreated FAs were dissolved with deionized water to a concentration of 0.01 g/mL, and the pH of the solution was adjusted to 6–7. A conical flask containing Resin 732 (sodium type) cation exchange resin and FA solution with a solid–liquid ratio of 100 g/L was shaken at room temperature for 2 h. Then, filtration was carried out to obtain the primary purified FAs (filtrate (a)). Similar to the above operation except for the difference in the pH value of the solution, the D401 Na⁺ chelating resin and a hydrogen type cation exchange resin were also successively used for ion exchange to finally obtain the filtrate (c). Afterward, the filtrate was dried in a vacuum drying oven to obtain purified FAs. The purity of FAs (P _FA) before and after purification was determined using the roasting method (at 750 °C).

3.

Purification process flow of the crude FA product.

2.7. Elemental Analysis of HA and FA Before and After Purification

As common elements in coal or HSs, Ca, Mg, Fe, K, Al, and Si were selected for elemental analysis. In addition, Na and Cl were also selected, as large amounts of NaOH and HCl were added in the AEAPCD process. The determination of Cl content referred to a Chinese standard (HG/T 5938–2021). The analysis of other elements, such as germanium, was also carried out using ICP-OES. It should be noted that HAs were dissolved in a NaOH solution before being detected. Therefore, the Na content of HAs is the difference between the measured value and the amount added.

2.8. Sequential Chemical Extraction Experiment (SCEE)

The SCEE is the most commonly used quantitative analysis method for indirectly studying the occurrence modes of trace elements in coal; it can distinguish germanium with different modes of occurrence using different chemical regents. Because there was no unified standard for the SCEE of germanium from lignite, this study mainly refers to the methods proposed by several scholars. , The specific operating steps are listed in Table .

2. Experimental Procedures for the SCEE of Germanium from Lgnite.

procedure	chemical reagents and conditions	definition
(1)	20 mL of 0.1 mol/L CH3COONa solution was added to 0.5 g of lignite, shaken at room temperature for 16 h, and centrifuged at 4200r/min for 30 min; the supernatant was collected. The precipitate was then washed with 10 mL of deionized water, shaken for 15 min, and centrifuged. The mixture of the two supernatants was used to determine the germanium content.	the water-soluble and exchangeable state
(2)	The residue of (1) was extracted with a solution of 0.04 mol/L NH4OH·HCl+25% CH3COOH (v/v), following the same extraction procedure as before.	the carbonate and Fe–Mn oxide bound state
(3)	The residue from (2) was extracted using a mixture of 1% NaOH and 0.1 mol/L Na4P2O7, following the same extraction procedure as before. The extraction solution was evaporated and ashed. The ash was then digested with HNO3–HF-HClO4 mixed acid and analyzed for germanium content.	the humic substance bound state
(4)	The residue of (3) was extracted with 30% H2O2, following the same extraction procedure as before. The extraction solution was evaporated and ashed, and the subsequent operation was the same as (3).	the other organic matter bound state
(5)	The residue of (4) was dissolved with HNO3–HF-HClO4, and the digestion solution was used to determine the germanium content.	the residue state

2.9. Two-step Alkaline Extraction–Acid Precipitation (AEAP) Experiments

The alkaline concentration during the alkaline extraction process has a significant impact on Ge extraction (mentioned in section ). To further study the extraction product properties at different alkaline concentrations, a two-step AEAP experiment was designed. As shown in Figure , the two-step alkali extraction experiment had the same conditions except for different alkaline concentrations. The specific parameters were a liquid–solid ratio of 9:1, an extraction temperature of 100 °C, and an extraction time of 2 h. Two acid precipitation experiments were conducted under the same conditions.

4.

Two-step alkaline extraction–acid precipitation experiment process.

As shown in Figure , filtrate (a), filtrate (b), filter residue (a), and filter residue (b) were obtained from the two-step AEAP experiments, and these products were characterized by the methods of Fourier transform infrared (FTIR) spectrometry and liquid chromatography–mass spectrometry (LC–MS). The FTIR analysis was performed on an infrared spectrometer (Nicolet iS20, USA). All samples and KBr powder were mixed, ground uniformly, and pressed into tablets for testing at room temperature; scans were made in the spectral range of 400–4000 cm^–1 with a resolution of 4 cm^–1. It should be noted that filtrate (a) and (b) were vacuum-dried into powder before FTIR. The 4000 QTRAP LC–MS/MS system was used to determine the mass distribution of the four products (filtrate (a), filtrate (b), filter residue (a), and filter residue (b)).

3. Results and Discussion

3.1. Influence of Alkali Extraction Conditions on the Recovery Yields of Ge and HSs

3.1.1 Influence of Alkali Concentration

The effect of the NaOH concentration on Ge and HS extraction is shown in Figure . The experiments were conducted at 100 °C for 2 h with a liquid–solid ratio of 10:1. It can be seen that the HS recovery yield of the filtrate is not significantly affected by the concentration of NaOH solution. When the concentration of NaOH solution increases from 0.1 to 0.5 mol/L, the recovery yield of HSs only varies within a small range (88.7–91.4%). This indicates that 0.1 mol/L NaOH is sufficient to extract HSs from lignite, and another 10% or so of HSs in the filter residue is likely to be bound to alkaline earth metal ions (Ca²⁺, Mg²⁺, etc.) from lignite, which can be extracted using a mixed solution of NaOH and Na₄P₂O_7.

5.

Ge content of the filter residue and the recovery yields of Ge and HSs of the filtrate under different NaOH concentrations.

The alkali concentration has a significant impact on the Ge recovery yield of the filtrate. As the NaOH concentration increases from 0.1 to 0.5 mol/L, the Ge recovery yield significantly increases from 58.53 to 85.11%. A further increase of NaOH concentration has few influences on Ge extraction. Note that when the NaOH concentration is 0.3 mol/L, the Ge recovery yield reaches 80.09%, and only a 5.02% increase in the Ge recovery yield was obtained by increasing the NaOH concentration to 0.5 mol/L. Considering the reagent costs and benefits, 0.3 mol/L is suggested as the optimum alkali concentration for the extraction experiments.

3.1.2. Influence of Extraction Temperature

The effects of the temperature on the extraction of Ge and HSs are shown in Figure . As the extraction temperature increases, the recovery yields of HSs and Ge of the filtrate gradually increase and stabilize at 100 °C. At the same time, the Ge content in the filter residue gradually decreases. This indicates that the reaction temperature has a promoting effect on the extraction of Ge and HSs, but excessive temperature is futile, which can increase the extraction cost. Compared with alkali leaching of HSs, that of Ge is more affected by the extraction temperature because of its larger curve slope.

6.

Ge content of the filter residue and recovery yields of Ge and HSs of the filtrate under different extraction temperatures.

3.1.3. Influence of Extraction Time

As shown in Figure , with the extension of extraction time, the recovery yields of HSs and Ge of the filtrate gradually increase and tend to stabilize after 2 h, while the germanium content in the filter residue gradually decreases. Furthermore, from the slopes of recovery yield curves of Ge and HSs, it can be seen that the extraction time has a greater impact on the extraction of HSs than that of Ge.

7.

Ge content of the filter residue and recovery yields of Ge and HSs of the filtrate under different extraction time.

3.1.4. Influence of the Liquid–Solid Ratio

As shown in Figure , the liquid/solid ratio has little effect on the HS recovery yield of the filtrate. When the liquid–solid ratio is 6:1, almost all HSs can be extracted by NaOH. However, as the liquid–solid ratio increases, the Ge recovery yield gradually increases and stabilizes at the liquid–solid ratio of 9:1. At the same time, the Ge content in the filter residue gradually decreases. The liquid–solid ratio is related to whether the coal particles can fully contact OH^– and Na⁺ ions in the alkaline solution; more importantly, it affects the total amount of OH^– and Na⁺ ions in the solution. Figures and show that the total amount of OH^– and Na⁺ ions in the alkaline solution is the key factor affecting Ge extraction. The higher the concentration of alkaline solution and the higher the liquid–solid ratio, the more OH^– and Na⁺ ions in the solution and the more germanium can be extracted.

8.

Ge content of the filter residue and the recovery yields of Ge and HSs in the filtrate under different liquid–solid ratios.

In summary, the optimal experimental parameters obtained from the single factor experiments of alkali extraction are a NaOH concentration of 0.3 mol/L, an extraction temperature of 100 °C, an extraction time of 2 h, and a liquid–solid ratio of 9:1. In this case, the recovery yields of Ge and HSs of the filtrate are 80.04 and 90.6%, respectively, and the Ge content in the filter residue is 59.2 g/t. The alkali leaching behavior of Ge is different from that of HSs. HSs appear to be more easily extracted with low NaOH concentration. To fully extract Ge from lignite, a sufficient amount of OH^– and Na⁺ ions seems necessary.

3.2. Influence of the Solution pH Value on the Recovery Yields of FAs/HAs and the Distribution of Germanium

Acid precipitation is a classic method for separating HAs and FAs, based on their differences in solubility in acidic solutions. As shown in Figure , when the pH value is 4, the HA recovery yield of the crude HA product is very low because the yield of crude HAs is low. As the pH value decreases, the HA recovery yield gradually increases. When the pH value is 1, the recovery yield reaches 88.83%. The trend of FA recovery is different from that of HAs during the acid precipitation process. The recovery yield of FAs changes in a narrow range (from 93.04 to 91.42%); the slight decrease may be caused by the entrainment of some FA molecules during HA sedimentation.

9.

Influence of the solution pH value on the recovery yields of FA/HA and germanium distribution.

As shown in Figure , Ge is mainly allocated to FA rather than HA. It is minimally affected by the pH value, although the HA recovery yield is very high (88.83%) when the pH value is 1 and the fact that the HA content is higher than that of FA in the lignite sample. This indicates that germanium in lignite mainly has affinity for FA than for HA. Compared with HA, the presence of more oxygen-containing acidic functional groups (such as carboxyl and phenolic hydroxyl groups) in FAs favors the enrichment of Ge during the coal formation process.

3.3. Effect of Distillation Time on Ge Recovery Yields

The volatilization efficiencies and recovery yields of germanium under different distillation times are listed in Figure . In the early stages of distillation, germanium quickly volatilizes in the first 30 min; but after 30 min, it is mainly H₂O rather than GeCl₄ that volatilizes from the solution. At 60 min, the volatilization efficiency and recovery yield of germanium reach 86.77 and 64.95%, respectively, slightly higher than those at 30 min. At this stage, the FA solution is heated to a dry state. However, there is 13.23% of germanium (measured by volatilization efficiency) still remaining in the residue. The possible reason is that as distillation proceeds, the Cl^– concentration in the solution gradually decreases. In the later stage of distillation, the Cl^– ions are insufficient to form GeCl₄ and cannot be evaporated. The improvement of the volatilization rate of germanium can be realized by supplementing the HCl.

10.

Ge volatilization efficiency and recovery yield of the distillate at different times.

3.4. Purification of HAs and FAs

As shown in Table , the purity of crude HAs is relatively high (95.5%), containing high sodium and varying amounts of elements, such as calcium, magnesium, iron, etc. After several rounds of water washing, the purity of HA increases to 98.89%, where 62.22% of sodium ions, 55.78% of potassium ions, and 52.29% of chloride ions are removed. Nevertheless, the removal rate of other ions is low; the reason may be that these ions form stable complexes with oxygen-containing functional groups in HAs. After purification, there is still 4908.4 g/t of sodium ions present in HAs, which may be due to electrostatic adsorption. When the pH value is >4.4, the surface of HAs shows a negative charge, causing cations to be adsorbed. Additionally, this may also be caused by the cation−π electron interaction.

3. Indicators of HAs and FAs Before and After Purification.

	elemental analysis (g/t)
samples	Ca	Mg	Fe	K	Na	Si	Al	Cl	purity (%)
crude HAs	547	103	1678	131	12993	1431	1037	10533	95.5
purified HAs	392	86	1242	58	4908	1098	1006	5025	98.87
crude FAs	3419	601	682	1928	227394	221	143	230933	38.33
purified FAs	1052	322	116	69	1066	205	127	209968	69.94

Clearly, the large amount of sodium and chloride ions in the crude FA is the main factor leading to its low purity (shown in Table ). After purification, the vast majority of sodium ions are removed, while a change in the chloride ion content is not significant, because the resins used in the experiment are cation exchange resins. Other metal elements have different removal rates, ranked from high to low, such as Fe > Ca > Mg > Al, possibly due to the varying degrees of adsorption of these metal ions by chelating resins. The silicon content does not change much, which may be related to the negative charge of its ions. Overall, the purification of FAs should include an anion exchange resin to remove chloride ions and further improve the purity of FAs.

4. Discussions on the Mechanism of Enhancing Ge Extraction by Increasing Alkaline Concentration

Section mentions that an alkali concentration of 0.1 mol/L is almost sufficient to extract all of the free HSs from lignite, and 58.53% of germanium bonded to them is also extracted accordingly. When the alkali concentration is further increased to 0.3 mol/L, the Ge recovery yield significantly increases to 80.09%. Different alkali leaching behavior of Ge and HSs as a function of alkaline concentration is critical to designing reasonable technique routes with low cost and high environment benefits. Among the factors, the modes of Ge occurrences play a significant role in the leaching behavior. It is necessary to determine the Ge associations in different organic or inorganic compounds in lignite.

4.1. Modes of Occurrences of Germanium in Lignite

Figure shows the proportion of various modes of occurrences of germanium in lignite; as shown in the figure, the proportion of germanium in the humic substance bound state is the highest, which is similar to the result mentioned in section that 58.3% of germanium was extracted when using a 0.1 mol/L alkaline concentration to extract lignite. In addition to HS-bound Ge, 32.67% of organically associated Ge was detected. This reveals that besides humic substances, there are some other types of organic compounds in lignite containing germanium that cannot dissolve when the alkaline concentration is low. As for the other three occurrence forms of germanium, namely, water-soluble/ion-exchange state, carbonate and Fe–Mn oxide bound state, and residual state, their total proportion is less than 5%, indicating that the affinity between Ge and inorganic minerals is relatively low.

11.

Various modes of occurrences of germanium in the lignite sample.

4.2. MS Characterization Results

A two-step AEAP experiment was conducted to obtain filtrate (a), filtrate (b), residu e­(a), and residue (b) (shown in Figure ), and their FTIR and LC–MS characterization results can provide some important information regarding the Ge-rich organic substances in lignite. As shown in Figure , the distribution of mass spectral peaks in the mass spectrum of residue (a) is continuous and concentrated, while that in the mass spectrum of residue (b) is clustered and dispersed, indicating that residue (a) contains more types of organic substances. In addition, residue (a) contains a small number of substances with a mass-to-charge (m/z) ratio greater than 800, while residue (b) contains more high-molecular-weight substances, some of which can even reach an m/z ratio of 1150.

12.

Mass spectra of residues (a) and (b).

Coincidentally, most of the MS peaks marked out in the mass spectrum of residue (a) and (b) match the chemical molecular weight of depolymerization products of lignin mentioned in several studies. − As listed in Table , most of these substances are monomer units (p-coumaryl, coniferyl, and sinapyl alcohols) that make up the lignin macromolecular structure or their derivatives, dimers, trimers, etc. These indicate that these substances are likely to come from lignin-like substances in lignite, which exist as precursors of HSs. Lignin is an important organizational structure of coal-bed plants, and the typical structure of lignin in coal-bed plants is still retained in lignite during coalification. Lignite contains more lignin-like substances because of its lower maturity during the coalification stage, with the structure being closer to the remains of primitive plants. Alkaline solution can promote the depolymerization of lignin. ,

4. Some Main Depolymerization Products of Lignin and Their Molecular Weights.

The main low-molecular-weight substances in residues (a) and (b) are very similar, with their MS peaks located at 150.88, 152.88, 208.90, 210.98, 266.80, 268.80, 270.81, 326.76, 365.21, and 393.27. Residue (a) is commonly termed primary humic acid, so residue (b) can be termed regenerated HAs due to their similarity in composition. Both types of HAs contain active oxygen-containing functional groups such as phenolic hydroxyl, hydroxyl, methoxy, etc. which are easily complexed with germanium. Therefore, increasing the alkaline concentration can extract more germanium.

As shown in Figure , filtrate (a) (referred to as primary FAs) is mainly composed of substances with molecular weights less than 300, and the corresponding MS peaks of these substances are continuous and concentrated. The composition of regenerated FAs (filtrate (b)) is relatively simple, and the corresponding MS peaks of these substances are clustered, which may be related to the depolymerization law of lignin. The depolymerization products are generally distributed in a regular pattern of monomer units (or their derivatives), dimer, trimer, tetramer, pentamer of monomer units, etc.

13.

Mass spectra of filtrate (a) and (b).

4.3. FTIR Characterization Results

As shown in Figure A, residue (b) has a stronger peak near 3410 cm^–1 compared to residue (a), which corresponds to the O–H stretching vibration of alcohols, phenols, or carboxyl groups, indicating a higher proportion of acidic functional groups in residue (b). The composition of residue (b) (see Figure ) is relatively simple and mostly consists of monomeric units (or their derivatives) of lignin and the polymers of these monomers, most of which contain acidic functional groups such as phenolic hydroxyl, carboxyl, and hydroxyl groups. The absorption peaks at 2927 and 2831 cm^–1, which exist in Figure A rather than Figure B, are attributed to the asymmetric and symmetric stretching vibrations of methylene, which often exist as bridge bonds, indicating that residues (a) and (b) have more high-molecular-weight substances than filtrates (a) and (b). This is consistent with the experimental phenomena and MS analysis results. Except the three absorption peaks mentioned above, residues (a) and (b) also have several almost similar absorption peaks located at 1710, 1630, 1440, 1380, and 779 cm^–1, indicating the presence of carbonyl, aromatic, methoxy, phenol, and aromatic ring disubstitutions, respectively. In addition, the broad peak near 1220 cm^–1 in residue (a) and the peak at 1050 cm^–1 in residue (b) are also characteristic absorption peaks of the phenolic hydroxyl. The absorption peak at 1150 cm^–1 indicates the presence of ether bonds. From the peak height ratio of the absorption peaks corresponding to 1630 and 1710 cm^–1, it can be seen that the peak height ratio of residue (b) is higher, indicating that residue (b) may contain a larger proportion of aromatic substances.

14.

Infrared spectra of four products. A-residue (a) and (b); B-filtrate (a) and (b).

The infrared spectra of filtrates (a) and (b) shown in Figure B indicate that they have similar functional groups, except for significant differences in the low wavenumber region. Both of them contain aromatic compounds (1640 cm^–1) clearly. The absorption peaks at 1440 and 1106 cm^–1 indicate the presence of methoxy, and the absorption peaks at 1203 and 3406 cm^–1 indicate the presence of the phenolic hydroxyl. The absorption peaks located at 881 and 815 cm^–1 indicate that there may be meta-substitutions on benzene rings. A more prominent shoulder peak near 1730 cm^–1 in the infrared spectrum of filtrate (a) and the presence of hydroxyl vibrations (3406 and 3229 cm^–1) suggest that filtrate (a) may contain more carboxylic acid; this is consistent with the fact that filtrate (a) (see Figure ) contains more butyric acid.

4.4. Leaching Mechanisms of Ge during the Alkali Leaching Process

Based on the SCEE results of the lignite sample and the LC–MS and FTIR analysis results of the four products obtained from the two-step AEAP experiment, we can draw a basic conclusion: Germanium in lignite is mainly bound to HSs, followed by lignin-like substances. In other words, HSs and lignin-like substances are carriers of germanium in coal because these substances contain a large number of active oxygen-containing functional groups (carboxyl, phenolic hydroxyl, hydroxyl, etc.), which can form coordination bonds with Ge. This has been confirmed by several experiments. ,, So, it is highly likely that Ge in lignite will exist in the form of HS–Ge or lignin–Ge complexes. As shown in Figure , referring to the host–guest two-phase model proposed by Haenel, the main components of coal are three-dimensional carbon structures containing a large number of polycyclic aromatic hydrocarbons connected by aliphatic chains and ether chains, while HSs, lignin-like substances, and their complexes with germanium are the mobile phase. Due to the presence of more acidic groups and better water solubility compared to HAs, FAs preferentially complex with germanium from the swamp water solutions during the peatification stage of coal-forming plants. Therefore, germanium in lignite is mainly composed of FA–Ge complexes. In addition, there are also small amounts of lignin-like substances in lignite, and the germanium adsorbed by these substances likely entered the cell wall during the growth period of coal-forming plants and was preserved because lignin contains fewer acidic functional groups and lacks an advantage in competing with FAs and HAs for germanium adsorption.

15.

Speculation on the mechanism of germanium release during alkaline leaching of lignite.

HSs including HAs and FAs belong to the group of polybasic weak acids. Their water solubility is closely related to their molecular weight and the number of hydrophilic functional groups (mainly carboxyl, phenolic hydroxyl, and hydroxyl) they contain, as these groups can ionize H⁺ in alkaline solutions, turning themselves into anions and becoming hydrophilic. As shown in Figure , due to the smaller molecular weight and higher proportion of acidic functional groups, FAs will preferentially undergo neutralization reaction with NaOH and enter the solution. At the same time, germanium complexed with them is also released into the solution. As the concentration of alkaline solution increases, weakly acidic HAs with a higher molecular weight and their complexed germanium also dissolve into the solution. When the concentration of alkaline solution continues to increase, lignin-like substances in coal undergo depolymerization under alkaline catalysis, generating substances similar to HSs by breaking the ether bonds between monomers. These substances carry germanium into the alkaline solution.

5. Conclusions

This study proposed an alkali extraction–acid precipitation–chlorination distillation method, which can comprehensively recover the high-valued organic and inorganic components of Ge-rich lignite in an environmentally friendly way. Alkaline extraction can simultaneously extract HSs and Ge in lignite. By performing extraction experiments with different operational conditions, it is found that the 0.1 M NaOH solution can extract 88.7% of HSs and 58.3% of germanium (which is closely bound to HSs). The occurrence forms of germanium in lignite are diverse. Increasing the concentration of alkaline solution can further extract other forms of germanium in lignite, which may be related to the fact that alkali can promote the depolymerization of lignin-like substances in lignite. The depolymerization products are mostly the basic structural units that make up lignin macromolecules or their derivatives, dimers, trimers, etc. These substances are soluble because they contain hydrophilic functional groups such as phenolic hydroxyl, carboxyl, hydroxyl, etc. These functional groups are also easily complexed with germanium. Moreover, germanium is mainly allocated to FAs rather than HAs. In comparison to the traditional chlorination distillation process requiring concentrated acid, a lower acidity is sufficient to separate germanium from the FA solution in this study. The solution needs to be maintained within a certain acidity range (pH1) throughout the process. Compared with the purity of crude HAs (95.5%), the lower purity of crude FAs (38.33%) is mainly caused by the introduction of a large amount of NaOH and HCl in the AEAP process. After purification with two cation exchange resins and one chelating resin, Na and K in FAs were almost removed, and Ca, Mg, Fe, and Al were also removed to varying degrees. However, due to the presence of a large amount of Cl^– ions, the purity of purified FAs remained relatively low (69.94%). In future work, anion exchange resins could be used to remove Cl^– ions from the FA.

The authors declare no competing financial interest.