**Contents**




## **About the Editor**

**Stefano Ubaldini** is a senior research scientist, Member of the International Scientific Advisory Board of the Institute of Geotechnics of the Slovak Academy of Science (SAS), Member of the Board of Directors and of the Scientific Council of the High Tech Recycling International Interuniversity Research Center (HTR), and head of the Bio-Hydrometallurgy, Cyanidation and Electrochemical Laboratories of the Institute of Environmental Geology and Geoengineering of the Italian National Research Council (IGAG-CNR). He is a Member of the Operational Groups, in the framework of the European Innovation Partnership on Raw Materials (EIP-RMs) of Horizon 2020, as a representative of academia, research institutes and think tanks https://ec.europa.eu/growth/tools-databases/eip-raw-materials/en/members. He has been appointed by the European Commission to take part in the Governance of the EIP-RMs as Sherpa member of the High Level Steering Group (HLSG)/Sherpa Group (years 2013–2020).

Relevant skills, scientific activity and experience

Dr. Ubaldini's main area of expertise is in hydrometallurgy and bio-hydrometallurgy. His activity has been carried out mainly in the framework of the primary and secondary raw materials (mineral and materials industry) for over 30 years, including the fundamental research, hydrometallurgy and biometallurgy of low-grade ores, metal extraction from solutions, precious metals recovery with novel leachants, treatment of waste materials, process development and design. His special interest is the interdisciplinary approach (mineralogy, leaching kinetics and solution chemistry) to the problems involved in the extraction and the recovery of metal values from primary and secondary raw materials. He has participated in 15 EC projects and managed R&D projects with funds from industrial and governmental resources; he was the scientist responsible for the following EC projects: H2020, Raw Materials and Recycling, Brite-Euram, VII FP, Life-Environment, INCO-Copernicus and INTAS. Dr. Ubaldini is the author of more than 250 publications on national and international journals and proceedings of international congresses. He has participated, as a component of the scientific committees, international honorific committee, chairman or invited speaker, to numerous national and international conferences. He is Member of International Advisory Board, Guest Editor and referee of international journals in the frame of waste treatment, hydrometallurgy and bio-hydrometallurgy. In the framework of the programme "Cooperation" of the 7th FP and Horizon 2020, he is employed by the European Commission as an expert evaluator. He is an expert evaluator of research projects appointed by the National Center of Science and Technology Evaluation, Ministry of Education and Science, Astana, Republic of Kazakhstan. He is an expert evaluator of the projects of the Italian Ministry of Education, University and Research (MIUR). Dr. Ubaldini is an international evaluator of PhD theses. His current research interests are primary and secondary raw materials, with particular reference versus the hydrometallurgy and biometallurgy of low-grade georesources, with the main aim to recover base and precious metals and to develop sustainable and innovative technologies for the processing of industrial wastes (spent catalysts, industrial tailings, batteries, WEEE etc.) and mining residues, coming also from abandoned sites, and the treatment of waste waters.

## *Editorial* **Leaching Kinetics of Valuable Metals**

## **Stefano Ubaldini**

Istituto di Geologia Ambientale e Geoingegneria, CNR, Area della Ricerca di Roma RM 1—Montelibretti—Via Salaria Km 29,300, Monterotondo, 00015 Roma, Italy; stefano.ubaldini@igag.cnr.it; Tel.: +39-06-90672748

Received: 18 January 2021; Accepted: 18 January 2021; Published: 19 January 2021

## **1. Introduction and Scope**

Leaching is a primary extractive operation in hydrometallurgical processing, by which a metal of interest is transferred from naturally-occurring minerals into an aqueous solution. In essence, it involves the selective dissolution of valuable minerals, where the ore, concentrate, or matte is brought into contact with an active chemical solution known as a leach solution.

There are numerous hydrometallurgical process technologies used for recovering metals, such as: agglomeration; leaching; solvent extraction/ion exchange; metal recovery; and remediation of tailings/waste. Currently, hydrometallurgical processes have a wide range of useful applications, not only in the mining sector—in particular, for the recovery of precious metals, such as gold and silver—but also in the environmental sector, for the recovery of toxic metals (such as copper, nickel, zinc, manganese, arsenic, cadmium, chromium, lead) from wastes of various types, and their reuse as valuable metals, after purification.

Therefore, there is an increasing need to develop novel solutions, to implement environmentally sustainable practices in the recovery of these valuable and precious metals, with particular reference to the critical metals, that are those included in materials that are indispensable to modern life and for which an exponential increase in consumption is already a reality or will be in a short-term perspective (antimony, indium, vanadium, rare hearts, etc.). Consequently, the economics of the processes, which are closely linked to the kinetics of leaching, are of great importance.

For publication in this Special Issue, consideration has been given to articles that contribute to the optimization of the kinetic conditions of innovative hydrometallurgical processes—economic and of low environmental impact—applied for the recovery of valuable and critical metals.

I would like to thank the authors who accepted this invitation, helping us to produce a high-impact, high-quality Special Issue on the "Leaching Kinetics of Valuable Metals".

## **2. Contributions**

Researchers around the globe investigating the leaching kinetics of valuable metals have been invited to submit research papers, so that readers can recognize the common points between them. Among the submitted manuscripts, eleven articles have been published in the issue.

The papers are all of high scientific value, while the experimental activities carried out fall into various disciplinary sectors, confirming the importance of the studies of the leaching kinetics of valuable metals in different scientific and technological fields. Here, I will briefly summarize the content of the published papers.

Geochemical characterization studies and batch leaching experiments were conducted to explore the effects of a CO2 + O2 leaching system on uranium (U) recovery from ores obtained from an eastern limb of the Zinda Pir Anticline ore deposit in Pakistan [1]. This study provides new insights into the feasibility *Metals* **2021**, *11*, 173

and validity of the site application of U neutral in situ leaching. According to Asghar F. et al., further studies are needed to reveal the influencing mechanism of the U(VI) initial concentration on U recovery in the solid phase.

While considerable experimental material has accumulated on the purification of uranium-containing water using nanoscale zero-valent iron (nZVI), there are no comparative studies of the sorption properties of iron-containing composites of different composition based on clay minerals. Taking into account the importance of environmental studies based on natural minerals, the removal of uranium by nZVI supported on kaolinite, montmorillonite and palygorskite was investigated by Kornilovych et al., including the removal efficiency of uranium from contaminated groundwater with low and high mineralization [2].

The kinetics of dissolution of refractory sulfide gold-containing concentrates of the Yenisei ridge (Yakutia, Russia) by a solution of HNO3 in the temperature range of 70–85 ◦C was investigated by Rogozhnikov et al., leading to the conclusion that the increase in sulfur content in concentrate can be used to ensure the more energy-efficient oxidation of sulfide minerals [3].

A physical–chemical activation desilication process was proposed to extract silica from high alumina fly ash (HAFA) [4]. The effects of fly ash size, hydrochloric acid concentration, acid activation time and reaction temperature on the desilication efficiency, were investigated comprehensively. The results achieved by Gong et al. indicate that physical and chemical activation suppresses the formation of zeolite, thereby improving the desilication efficiency and further improving the A/S of the fly ash; results which are very advantageous for the next step of alumina extraction.

A study of the oxidation of thiosulfate, with oxygen using copper (II) as a catalyst, at a pH between 4 and 5 has been conducted by González-Lara et al. [5]. The basic idea was to avoid the formation of tetrathionate and polythionate, transforming the thiosulfate into sulfate. The nature of the reaction and a kinetic study of thiosulfate transformation, by reaction with oxygen and Cu2+ at a ppm level, have been determined and reported. The thiosulfate concentration was reduced from 1 g·L−<sup>1</sup> to less than 20 ppm in less than three hours.

The paper by Batnasan et al. [6] deals with the recovery of gold from waste printed circuit boards (WPCBs) ash by high-pressure oxidative leaching (HPOL) pre-treatment and iodide leaching followed by reduction precipitation. Under the optimal conditions, the percentage of gold extraction from the gold chips and the residue of WPCBs was 99% and 95%, respectively.

The objective of the experimental work carried out by Ubaldini et al. is the application of innovative and sustainable technologies for the treatment and exploitation of mining tailings from Romania [7]. The results obtained by application of the thiosulfate process on a low gold content ore were considered encouraging. The optimization of process parameters and operating conditions should permit the best results in terms of process yields to be achieved.

In the article submitted by Cháidez et al., a copper leaching process from chalcopyrite concentrates using a low-pressure reactor is presented. The experimental results showed that it is possible to extract 98% of copper in only 3 h. This result indicates a fast process compared with others reported in the literature [8].

The experimental results obtained during the preparation of Al-Ni and Al-Ni-Mg alloys using the aluminothermic reduction of NiO by submerged powder injection, assisted with mechanical agitation, are presented and discussed by Silva Beltran et al. [9].

Zhovty Vody city, located in south-central Ukraine, has long been an important center for the Ukrainian uranium and iron industries. Uranium and iron mining and processing activities during the Cold War resulted in poorly managed sources of radionuclides and heavy metals. Widespread groundwater and surface water contamination has occurred, which creates a significant risk to drinking water supplies [10]. The results of the study conducted by Kornilovych et al. demonstrate the effectiveness of the use of the permeable reactive barrier (PRB) for ground water protection near uranium mine tailings storage facility (TSF). The greatest decrease was obtained using zero-valent iron (ZVI)-based reactive media and the combined media of ZVI/phosphate/organic carbon combinations.

Eudialyte is a promising mineral for rare earth elements (REE) extraction due to its good solubility in acid, low radioactivity, and relatively high content of REE. Ma et al. present a study assessing the two stage hydrometallurgical treatment of eudialyte concentrate: dry digestion with hydrochloric acid and leaching with water. The research reported in this paper [11], also as a novelty, explored the feasibility and efficiency of the REE extraction process at room temperature on a scale-up demonstration platform that precedes future industrial applications. Information on upscaling operations for the treatment of eudialyte is also missing in the overall literature.

#### **3. Conclusions and Outlook**

A variety of topics have composed this Special Issue, presenting recent developments within the field of leaching kinetics of valuable metals.

As Guest Editor, I am very happy for the success of this Special Issue. I am also proud of the final result, in addition to the high quality and originality of the contributions. I hope that all the scientific results in this Special Issue contribute to the advancement and future development of research in this field.

I would like to warmly thank all the authors for their contributions, and all the reviewers for their efforts in ensuring a high-quality publication. Sincere thanks also to the Editors of *Metals* for their continuous help, and to the *Metals* Editorial Assistants for the valuable and inexhaustible engagement and support during the preparation of this volume. In particular, my sincere thanks to Mr. Toliver Guo for his help and support.

**Conflicts of Interest:** The author declares no conflict of interest.

## **References**


*Metals* **2021**, *11*, 173


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2021 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article*

## **Geochemical Characteristics and Uranium Neutral Leaching through a CO2** + **O2 System—An Example from Uranium Ore of the ELZPA Ore Deposit in Pakistan**

**Fiaz Asghar 1,2, Zhanxue Sun 1,2,\*, Gongxin Chen 1,2, Yipeng Zhou 1,2, Guangrong Li 2,3, Haiyan Liu 1,2 and Kai Zhao 2,3**


Received: 11 October 2020; Accepted: 27 November 2020; Published: 1 December 2020

**Abstract:** Geochemical characterization studies and batch leaching experiments were conducted to explore the effects of a CO2 + O2 leaching system on uranium (U) recovery from ores obtained from an eastern limb of Zinda Pir Anticline ore deposit in Pakistan. The mineralogy of the ore was identified by Electron Probe Micro-analyzer (EPMA) and Scanning Electron Microscope-Energy Dispersive Spectrometer (SEM-EDS), showing that pitchblende is the main ore mineral. XRD was also used along with EPMA and SEM characterization data. Experimental results indicate that U mobility was readily facilitated in the CO2 + O2 system with Eh 284 mV and pH 6.24, and an 86% recovery rate of U3O8 was obtained. U speciation analysis implied the formation of UO2 (CO3)2 <sup>2</sup><sup>−</sup> in the pregnant solution. The plausible mechanism may be attributed to the dissolved CO2 gas that forms carbonate/bicarbonate ion releasing oxidized U from the ore mineral. However, U recovery in the liquid phase was shown to decrease by higher U(VI) initial concentration, which may be due to the saturation of Fe adsorption capacity, as suggested by an increase in Fe concentration with increasing initial U(VI) concentration in the solid phase. However, further studies are needed to reveal the influencing mechanism of U(VI) initial concentration on U recovery in the solid phase. This study provides new insights on the feasibility and validity of the site application of U neutral in situ leaching.

**Keywords:** geochemical characteristics; pitchblende; U neutral leaching; ELZPA ore deposit in Pakistan

## **1. Introduction**

Uranium (U) ore deposits are important to society as a primary material for the generation of nuclear power [1–3]. Interest in in situ leach (ISL) mining has grown considerably over the last 30 years [4] because of its significant advantages over conventional methods for mining, especially low-grade sandstone-type ore deposits. Research on U mobilization has focused on its migration and mineralization under natural conditions, as well as its impact on the environment [5–17]. The incorporation or release of U to or from mineral structures is determined primarily by its redox state and mineral solubility [18]. The change in oxidation state from U(IV) to U(VI) usually occurs in nature under oxidizing conditions at the geosphere/atmosphere interface or within the lithosphere from contact

with oxygenated groundwater [19]. Generally, sulfuric acid/iron sulfate and carbonate/bicarbonate are common leachants in acidic and alkaline leaching systems, respectively [20,21].

A number of leaching tests have been reported to study the U leaching mechanism from porous media under different leaching conditions [22–27]. The impact of parameters including pH, Eh, and alkalinity on U recovery parameter was investigated. Combing a modeling approach, U chemical leaching and adsorption tests were performed by Briganti et.al. [28], showing that pH and alkalinity were main factors regulating geochemical behavior of U during ignimbrite-water interaction. However, their study lacks the consideration of Eh and suggests that the greatest leaching of U occurs at 50 ◦C, which seems irrational in field trials [29,30]. Furthermore, U leaching efficiency was shown to be complicated by factors including chemicals' initial concentrations and Fe and other metals' adsorption, of which the influential mechanisms are not sufficiently studied.

The influential ISL parameters of U recovery can be classified into two different categories. One concerns technological factors and the other concerns natural factors where U ores are leached mostly under oxidizing conditions in order to convert U contained into minerals from a relatively insoluble tetravalent form (U(IV)) into water-soluble hexavalent U (U(VI)) [31]. Under the CO2 + O2 leaching condition, their partial pressure was kept constant for U dissolution. The constant pressure parameter may produce in situ conditions during laboratory experiments. Indeed, the dissolution of U minerals is promoted by increasing the partial pressure of O2 or CO2, which is covered in a later discussion. Hence, it is essential to describe redox conditions and solution chemistry in the natural prevalent environments in order to better understand the geochemical behavior of the U species.

The presence of the U species in a water-rock interaction system is attributed to the evolution of redox conditions. Although acidic and alkaline leaching of U from limestone and sandstone media has been practiced extensively in the past, the shortcomings presented the opportunity to investigate the use of neutral leaching of U to understand its leaching mechanism, considering that limited research has been published on U neutral leaching. The novelty of this work is to unravel the mechanism of U mobilization during the neutral leaching process from U ore of an eastern limb of Zinda Pir Anticline (ELZPA) ore deposit. Therefore, the objectives of this study were the following: (i) identify the deposit minerology and geochemical behavior of the U species and (ii) determine the effects of leaching parameters on the release of U from U-bearing sandstone.

## **2. Materials and Methods**

## *2.1. Geological and Hydrological Setting of the Study Deposit*

The study deposit is located within the eastern ward extension of the Sulaiman Range, Dera Ghazi Khan, Pakistan, known as the eastern limb of Zinda Pir Anticline (ELZPA) ore deposit. It is a typical sandstone-type U deposit. The exposed rocks in the Sulaiman Fold Belt range were deposited from the Triassic to Tertiary periods with an estimated thickness of more than 7000 m [32–36].

The ore body is mainly found at a burial depth of 70–120 m in the Neogene molasses sequence, commonly known as Siwalik, that was uplifted and deformed to form the Zinda Pir Anticline range. The strata dip steeply to the east with angles of between 75◦ and 80◦. The layers above and below the mineral-bearing strata are impermeable and retain confined pore water at varying water table depths. The water has a pH between 7 and 8, and consists mainly of SO4 <sup>2</sup><sup>−</sup>, HCO3 <sup>−</sup>, Na<sup>+</sup>, Ca2<sup>+</sup>, and Mg2<sup>+</sup> ions, as shown in Table 1.

The ELZPA deposit is divided into four zones (locally), i.e., Lal-Ashab uraniferrous horizon (Z-I), Fowl creek uraniferrous horizon (Z-II), Zamdani uraniferrous horizon (Z-III), and Ghazzi uraniferrous horizon (Z-IV). All the zones except Z-I are bounded by lower and upper shales, and a volcanic ash layer deposited parallel to zone IV. The ore body (Z-I) is tabular upward and exhibits a slanting shape along the down dip. The host sandstone is overlain by 7 to 10 m thick brownish grey shale, running parallel to the strike of the host sandstone. The shale below the ore-bearing strata is however not continuously developed. The three predominant formations that form the major subdivision

of the Siwalik series are the Vihowa, Litra, and Chaudhwan Formations. The deposit is located in the middle part of the Litra Formation (Middle Siwalik). The hosted sandstone is grey to dirty grey, fine- to medium-grained and loosely to moderately cemented. Sometimes it is cross-bedded and contains limonitic and hematitic alternation bands. The sandstones in the lower part are moderately cemented, thus forming high peaks, whereas those in the upper part are less cemented and fragile, thereby forming a flatter topographic scenario. The average permeability coefficient and porosity of the rock aquifer are 0.61 m/d and 22%, respectively. The climate in the area is semi-arid to arid. The maximum and the minimum annual temperatures are 50 and 5 ◦C (average temperature is 30 ◦C). A frost period rarely occurs except for some seasonal snowfall on high peaks of the Sulaiman Range. The precipitation is about 200 mm, which falls mostly during the period of July to September and winter [37,38].


**Table 1.** Chemical composition of ground water (ppm).

## *2.2. Sample Collection*

The material was sourced from the sandstone-type U deposit that existed in the ELZPA, Sulaiman Range, Dera Ghazi Khan, Pakistan. At each sampling location, three replicate samples were collected and the unrepresentative materials were removed. The depth of the investigated boreholes spanned from 90 to 165 m. Samples were coated with wax during the core recovery process to avoid contact with atmospheric oxygen. All samples were transported to the laboratory for further treatment (discussed below).

## *2.3. Sample Preparation*

Representative (in composite) ore samples were prepared before being analyzed and for laboratory leaching experiments. Briefly, the natural grain size of the crushed ore sample was used for the leaching batch test. For this study, natural grain size means that the ore sample was first gently loosened without reducing the sand grain size. The corresponding particle size distribution ranges were <0.5 mm, 0.5 mm to 1.00 mm, and >1.00 mm. A total of 36 samples were dried, homogenized, and sieved. All the ore samples were grouped as mixed, high, medium, and low grade on the basis of % U3O8 concentration, as shown in Table 2. The coning and quartering method was used to prepare representative samples. A portion of the homogenized ore material was then milled to 200 mesh size for the determination of % U3O8 in each sample group.

## *2.4. Reagents*

All chemicals used in this work were of analytical reagent grade and used without further purification. The solutions and dilutions used for calibration were prepared using deionized water. Lixiviant solutions used for batch leaching experiments were prepared with laboratory tap water

(LTW), of which the composition is shown in Table 3. The representative ore sample was first grounded into a final 200 mesh size and pulverized before being analyzed (discussed below).


**Table 2.** Sample grouping.



## *2.5. Batch Leaching Process*

Batch extraction tests typically involve mixing a sample with a specific amount of leaching solution without renewal of the leaching solution [39]. Leaching experiments were carried out in batches with a liquid-to-solid (L/S) ratio of 2 in the autoclave reactor [40]. A Karl Kolb Scientific Technical Supplies autoclave (25 bar, BTR temperature of 300 ◦C, volume of 0.5 L, Dreieich, Germany) was used. An amount of 150 g of ore sample (TCS-04) was mixed in the prepared 300 mL of lixiviant (liquid) of varying conditions (variable bicarbonate solution concentration), while the ore samples (TCS-03 and TCS-02) were processed under only the best result of HCO3 − concentration. Experiments were performed to investigate the long-term release of the recalcitrant uranium from the ore under varying lixiviant solution conditions (variable bicarbonate solution concentration), and replicate tests were conducted under the same conditions. The leached solution was sampled at regular time intervals, i.e., from 0.5, 2.5, 7.5, 17.5, 41.5 h to days, and so on, until the equilibrium condition was acquired (i.e., no further change in U3O8 recovery in the leached solution). Leaching tests performed by the researchers indicated that the L/S ratio and grain size were important factors for determining the metal release; however, 24 h was not sufficient for a thorough assessment of leaching [41]. The leaching process was kept free from mechanical agitation at room temperature. The batch sampling process consisted of removing 10 mL of leached solution and filtering it through a 0.45 μm filter (Millipore, Merck, Germany). The pH and Eh were monitored throughout the batch process. The filtered sample was analyzed for U3O8 and HCO3 − concentration. Immediately after sampling, the sample was kept in a refrigerator until analysis. The HCO3 <sup>−</sup> concentration was controlled by adding ammonium bicarbonate (NH4HCO3) in the lixiviant solution. The pH of the leaching regulation system was set in the range of 6 to 6.6. The industrial oxygen gas (O2) was injected as oxidant to oxidize U(IV) into U(VI). The ratio of CO2 and O2 was fixed at 1:10 with a constant injection pressure of 2 bar throughout the batch process. At the end of each test, solid and liquid phase separation was accomplished by centrifugation. For centrifugation, 15 mL of the batch were transferred to 50 mL polycarbonate centrifuge tubes and centrifuged at 4000 rpm for 15 min. The volume of leachate recovered was measured and recorded. The pH and Eh were monitored with a Hanna Instruments HI 98121 device (Hanna, Hungary). The Hanna meter was calibrated for pH function with 4.0 and 7.0 buffers. The Eh calibration was performed using a Zobell solution.

## *2.6. Characterization Techniques*

Alkalinity of the leachate was measured by titration with HCl. The endpoint of the titration was determined by a color change. Briefly, methyl orange and phenolphthalein were used to detect the bicarbonates and carbonates, respectively. The liquid phase was analyzed for U3O8 through fluorimetry and spectrophotometrically using dibenzoyl methan as coloring agent, depending on the concentration of U3O8 in the leached solution. The cations Ca2+, Mg2+, Fe2+, Fe3+, Na+, and K<sup>+</sup> were measured by acid-base titration and flame photometry, respectively. The anions (SO4 <sup>2</sup>−, Cl−) were analyzed spectrophotometrically. All liquid samples were filtered using a syringe. The elemental composition and SiO2 were analyzed through atomic absorption spectrometer and spectrophotometrically, respectively, while the whole rock was analyzed through XRF. For solid material, U assays were carried out by taking 5 g of pulverized sample of the finely milled to 200 mesh size. The 5 g sample was moistened with (1.00 N) HNO3 for the removal of carbonates with effervescence, then 2 mL HClO4 and 10 mL HF were added; the sample was then heated to dry in a microwave oven-digester. Again, 10 mL HNO3 and 10 mL HF were added and the sample was heated to dry again. Followed by addition of 50 mL of (1.00 N) HNO3, and the mixture was leached for one hour, and finally it was diluted to a volume of 100 mL. Inorganic carbon was analyzed through acidification of the finely grounded sample using concentrated HCl followed by back titration with NaOH. Organic carbon (C org.) was analyzed by oxidizing the sample with a K2Cr2O7 and H2SO4 mixture followed by back titration using freshly prepared ferrous ammonium sulphates. Total carbons were analyzed in this manner and cross-checked using a sulfur analyzer (CS-800 Eltra, Germany).

XRD patterns of the U ore sample were recorded using a diffractometer (Bruker D8 ADVANCE, Karlsruhe, Germany) with CuKα radiation (40 kV, 40 mA) in a continuous scanning mode, and the 2θ scanning ranged from 3◦ to 80◦. EVA software (V4.2.1, developed by Bruker Corporation) was used to analyze the mineral composition with XRD data using the default crystallography open database COD 2013. An EPMA (JXA-8230) was used to identify the U ore mineral. A small sample (<0.5 g) of material was drawn from the fraction that was expected to be enriched with U mineral. The feed ore and leached residue sediment samples were characterized using a scanning electron microscope (SEM) (NovaNanao 450, Fei Czech, Co., Ltd.), while the distribution of elements was detected by energy-dispersive X-ray spectrometry (EDS) (Oxford X-MaxN).

## **3. Results and Discussion**

## *3.1. Mineral Association and Geochemical Characteristics*

Mineral association in the U ore shown by sample TCS-04 is listed in Table 4 and the XRD pattern is shown in Figure 1. It shows that quartz and mica are the predominant phases with percentages accounting for 30.40% and 20.30%, respectively. The percentages of albite and labradorite are 14.40% and 10.80%, respectively. The quartz is angular with a 0.05 to 0.4 cm size suggesting a nearby source. The oligoclase and pyroxene have percentages of 8.0% and 7.7%, respectively, along with orthoclase accounting for 3.40%. Mostly weathered oligoclase shows polycrystalline orthoclase aggregates. The distribution of calcite is very uneven and it can be localized as calcareous cemented masses or calcareous cemented thin layers. The content of calcium carbonate in the non-calcareous uncemented core samples is about 1.50%.


**Table 4.** Mineral analysis (%) of uranium ore by XRD.

The minor minerals including amphibole, kyanite, staurolite, epidote, zircon, magnetite, and hematite possess percentages below unity. The content of magnetite and hematite is 0.4% and 0.10%, respectively.

**Figure 1.** XRD pattern of uranium ore.

The geochemical composition for the four groups' ore samples is shown in Table 5. SiO2 is the major composition with a percentage range accounting for 57.36–68.87%, followed by Al2O3, CaO, and CO2 with percentage ranges of 10.09–14.89%, 4.75–11.85%, and 4.10–10.35%, respectively. The percentages of oxides of Na, K, Mg and Fe are in the range of 1.00–2.00%, and TiO2 is less than unity.

**Table 5.** Geochemical composition of uranium ore samples (wt.%).


The U mineral in the investigated sample (TCS-04) of ore is mainly silica-bearing pitchblende and its composition analysis through EPMA is listed in Table 6. An electronic photograph is shown in Figure 2.


**Table 6.** Composition of pitchblende (%) analyzed by EPMA.

**Figure 2.** The backscattered electronic photograph of the uranium ore sample. Qz: quartz; Pit: pitchblende.

The U oxide content results of the examined mixed ore sample (TCS-04) and of the ore samples (TCS-03, TCs-02, and TCs-01) with three different grades (high, medium, and low) are shown in Table 7. The geochemical information about sample TCS-04 reflects an overview of the ore deposit with respect to presence of special element characteristics. This sample has an average of 363 ppm of U3O8 with a U(IV)/U(VI) ratio of 21 (Table 7) and shows the mobility of U in terms of the U oxidation state. This sample analysis shows that 4.55% of U3O8 in the ore deposit may exist in the oxidized form (UO2 <sup>2</sup>+). The oxidized form may be leachable without injecting an oxidant during the leaching process. However, 95.19% of U3O8 in the ore deposit exists in the reduced form (U(IV)), and might be made leachable by injecting an oxidant during the leaching process. A similar composition was observed in the other three groups (TCS-03, TCS-02, and TCS-01), where the oxidized form of U accounted for 12.23%, 9.50%, and 7.08%, respectively. The presence of the U species in ore samples was attributed to the evolution of redox conditions. This is indicated by the Fe species (Fe2<sup>+</sup> and/or Fe3<sup>+</sup>) being positively correlated to U(VI), as shown in Figure 3. The general reactivity of Fe significantly influences U geochemistry, and Fe (Fe2<sup>+</sup> and/or Fe3<sup>+</sup>) forms mostly oxide, oxyhydroxide, and S-mineral compounds with its substitution in U-minerals commonly [42]. Moreover, there is no well-defined correlation between the U species and organic carbon. It probably suggests that the oxidized U associated with inorganically dominated carbon as compared to organics. Recently, a similar study was documented from the eastern Suliman range of Dear Ghazi Khan, Pakistan [37,38] showing that U ore was mostly associated with organic matter (probably petroleum) as well as other phases of the ore (e.g., biotite). Consequently, the geochemistry of U-containing minerals indicates a relatively pre-complexed environment of the investigated ore deposit, where the occurrence of U(IV) was favored as compared to U(VI). It in turn indicates that the injection of oxidant was needed for the mobilization of the reduced form of U during in situ leaching. A similar case has been reviewed and reported by Shamim Akhtar [43], showing that U deposits in Pakistan were formed in a reduced environment. Various studies have reported that U solubility is affected by its speciation and aqueous chemistry including organic matter (OM) concentrations [44–46]. Low OM concentration was shown to promote U solubility, whereas high OM concentration may immobilize U [47]. Of all the inorganic ligands, carbonates are more active ones to complex U, as compared to others such as SiO2 <sup>4</sup><sup>+</sup> and PO4 <sup>3</sup><sup>−</sup> [48,49].

## *3.2. Neutral Leaching*

Different leaching methods including neutral, acidic, and alkaline have been reported in U mining sites [50–55]. The present study investigated the processing of neutral leaching with the pH value being monitored in the experimental scope. These operational conditions were suggested to favor the application of the neutral leaching since it avoids excessive consumption of reagents (discussed below).


**Table 7.** Special characteristics of uranium ore.

**Figure 3.** Trend of U(VI) and Fe(III) with organic carbon (C org.) in uranium ore.

## 3.2.1. Effect of Bicarbonate Concentration and Chemical Dynamics during Leaching

Concentrations of HCO3 − are shown as a function of time in Figure 4. Overall, two trends were observed during the course of experiments. In the first test, where 0.60 g/L of HCO3 − concentration was used for the preparation of the lixiviant solution, the U3O8 was determined to be 157 mg/L after equilibrium was obtained, while its initial concentration was of 12 mg/L before starting the batch processing (Figure S1). For the two tests (#2 and #3), 0.80 g/L and 1.00 g/L of HCO3 − concentrations were used, respectively, and the U3O8 was determined to be 25 mg/L and 64 mg/L, with an initial concentration of 12 and 5 mg/L before starting the batch processing, respectively. Similarly, in the fourth test (#4), 1.20 g/L of HCO3 <sup>−</sup> concentration was used, and the final concentration of U3O8 was determined to be 124 mg/L, with its initial concentration of 3 mg/L. The highest U recovery (86%) was found using a 0.60 g/L of HCO3 − concentration. The U recovery with a concentration of 1.00 g/L of HCO3 − was 45%, followed by a 0.8 g/L batch where recovery was 13%. The U recovery with 1.20 g/L of HCO3 − was 68%. The changes in uranium recovery for the tests are shown in Figure 5.

Consequently, the dissolution of the U was favored more easily with 0.6 g/L of HCO3 −, and only 8.70 days was needed to achieve high recovery leaching time. The lower limit of HCO3 − concentrations in the aforementioned tests shows a slightly gentle process fluctuation as compared to those on the higher side, and a general improvement in the recovery was observed as the HCO3 − concentrations increased from 0.8 to 1.2 g/L during the process. The effect of varying concentrations of the leaching agent (HCO3 −) on U recovery was studied by Elizângela and Ana [56]. Results obtained from the authors showed that 1.00 mol/L of HCO3 − was the best concentration, under which >95% recovery was obtained in 24 h via alkaline leaching, while a general improvement in the recovery was observed as the HCO3 − concentrations increased. However, the highest recovery of 86%, as a special result, was obtained from test #1 with 0.6 g/L of HCO3 <sup>−</sup>. The lowest recovery of the 0.8 g/L HCO3 − test may be due to lower bicarbonates and the constantly stable concentration of HCO3 − as U recovery may highly correlated with it [57]. The geochemical analysis results of ore and residuals are shown in Table 8. It can be seen that the CaO concentration in all test residues was decreased significantly. This may be due to the injection of CO2 gas causing the dissolution of calcite (Figure 6), as shown by Equation (1):

$$\rm CaCO\_3 + CO\_2 + H\_2O \to Ca^{2+} + 2HCO\_3^- \tag{1}$$

**Figure 4.** Change in HCO3 − concentration in the leaching solution.

**Figure 5.** Change in uranium recovery in the leaching solution.

The maximum dissolution of calcium with a concentration of 0.80 g/L of HCO3 − was 2.33%, followed by the 1.0 g/L batch where the dissolution was 1.56%, while U recoveries were 13% and 45%, respectively. In contrast, the calcium dissolution with 0.60 g/L of HCO3 − was 1.14%, and in the 1.20 g/L batch, 0.56% calcite was dissolved, with U recoveries being 86% and 68%, respectively. The relationship between U recovery and calcite dissolution is not totally positively correlated. However, 0.60 g/L concentrations of HCO3 <sup>−</sup> test have comparatively higher U3O8 recovery than that of 1.20 g/L HCO3 − test with a calcite dissolution range accounting for 1.14% and 0.56%, respectively.


**Table 8.** Geochemical analysis of ore and residues.

**Figure 6.** Concentration of oxides before and after leaching.

## 3.2.2. Effect of pH

The pH decreases quickly and then it remains stable (Figure 7); in addition, its relationship to Eh shows regular patterns (Figure 8). Different patterns of leaching as a function of pH were also observed (Figure 9). The leaching recovery of U increased with changing pH in the early stages of the process and then there was no further change in U recovery. For example, in the test using 0.6 g/L of HCO3 −, high U recovery was found, while the pH value reduced from 8 to 6.57. Mean and final values recorded for pH for each of the leaching protocols are shown in Table 9. Probably, the plausible mechanism may be due to the injection of CO2 gas. CO2 dissolved in water results in reducing pH of the system. The carbonate/bicarbonate ion in turn promotes the release of adsorbed U(VI) from the U-bearing mineral. The released U(VI) dissolved into the liquid phase depends upon contact time and mineral charge density. In addition, increasing H<sup>+</sup> results in the dissolution of gangue minerals like calcite (Figure 6). It can be seen (Figure 9) that the lowest pH value and highest Eh value (most oxidizing conditions) were observed for the best addition of HCO3 − (0.6 g/L). It should also be noted that pH showed almost a negative correlation with recovery (Figure 9a), i.e., more recovery resulted towards the lower pH of the neutral range, resulting in the formation of more H<sup>+</sup> in the solution. Briganti et al. [28] conducted a study using chemical leaching and adsorption tests with a simple modeling using PHREEQC. They found that the geochemical behavior of U during ignimbrite-water interaction was controlled mainly by temperature, pH, and solution chemistry (especially alkalinity). The main results of their work indicated that U was more easily mobilized by a slightly basic solution (pH 7.5). However, the results obtained from the research revealed that U was more easily mobilized by a slightly acidic solution (pH range 6 to 6.6) during nearly neutral leaching conditions. In many respects, leaching behavior as reflected by pH leaching tests and related characterization provide a better means of assessing environmental impact than an analysis of total composition, such as how the solubility changes if in situ pH changes occur. The pH is one of the key parameters that determines heavy metal mobility depending upon soil and sediment properties.

**Figure 7.** Change in pH measurements in the leaching solution.

The aqueous U(VI) carbonate system has been thoroughly studied by many researchers [58–64]. It is well accepted that the three monomeric complexes of general formula UO2(CO3), UO2(CO3)2 2−, and UO2(CO3)3 <sup>4</sup><sup>−</sup> present under the appropriate conditions [65].

In neutral to carbonate media, U is converted into a series of carbonate complexes (UO2CO3 (maximum fraction at pH 5), UO2(CO3)2 <sup>2</sup><sup>−</sup> (maximum fraction at pH 6.5), and UO2(CO3)3 4− (maximum fraction at pH 10 to 11)) when the pH of the solution is increased.

Nevertheless, in all the tests performed, the concentration of CO3 <sup>2</sup><sup>−</sup> was not detected in the leached solution during analysis because of the pH range condition was 6 to 6.6 (which indicates that UO2(CO3)2 <sup>2</sup><sup>−</sup> is the main carbonate complex in the condition studied in this work). In such cases, the recovery of U from a nearly neutral medium is possibly due to the formation of stable U carbonate complex-like UO2(CO3)2 <sup>2</sup><sup>−</sup> [66] via Equation (2):

$$\text{UO}\_2 + 1/2\text{O}\_2 + 2\text{HCO}\_3^{2-} \rightarrow \text{UO}\_2(\text{CO}\_3)\_2^{2-} + \text{H}\_2\text{O} \tag{2}$$

**Figure 8.** Trend of Eh and pH in leaching solution with different HCO3 − concentrations: (**a**) 0.60 g/L; (**b**) 0.80 g/L; (**c**) 1.00 g/L; (**d**) 1.20 g/L.

**Figure 9.** Trend of U3O8 recovery and pH in leaching solution with different HCO3 − concentrations: (**a**) 0.60 g/L; (**b**) 0.80 g/L; (**c**) 1.00 g/L; (**d**) 1.20 g/L).


**Table 9.** pH, Eh, and recovery changes against each of the leaching protocols investigated.

This is a stable U compound in the determined pH range. Many researchers have reported that when pH > 6, U(VI) is dominant in the presence of more complexes such as uranyl hydroxyl complexes and polynuclear uranyl hydroxyl complexes, and the reactivity of Fe inhibits U(VI) [67,68]. It may be the advantage of relatively less Fe (Table 4) in the investigated ores in terms of the experimental scope (pH 6 to 6.6). Under near-neutral conditions, U forms soluble complexes with carbonate and phosphates. Ma et al. [31] conducted an adsorption study of the U species and reported that these species are also influenced by the charge of the mineral surface, depending on the pH of the sorbate solution. Recently, Zhou et al. [46] reported from ISL research from mining sites that, as the pH value decreases, the adsorbed UO2 <sup>2</sup><sup>+</sup> can be replaced by protons and U concentration increases significantly in the leached solution.

## 3.2.3. Effect of U(VI) Initial Concentration

The effect of different U(VI) initial concentrations (in the solid phase) on U3O8 recovery was investigated. The results are listed in Table 10 and shown in Figure 10, showing that recovery was 86% at the lower initial U(VI) concentration (sample TCS-04). A higher U(VI) initial concentration in the solid phase leads to a lower recovery of U3O8 (Figure 10). This may be due to the saturation of Fe2<sup>+</sup> + Fe3<sup>+</sup> adsorption capacity as its concentration increases with increasing initial U(VI) concentration in the solid phase (Table 10). As the concentration of U(VI) increases in the liquid phase, UO2 <sup>2</sup><sup>+</sup> ions compete for adsorption sites or available functional groups [69]. It is reported that adsorption capacity continues to increase with the rise of initial U(VI) concentration, which may be due to the presence of more U(VI) ions around Fe [70]. Laboratory experiments focusing on U chemical species have been previously done as well for the purpose of researching in situ leaching of U ore, and have reported that the recovery of leaching U is decided jointly by both U content and its activity, according to Ma et al. [31]. Probably, it seems that another mechanism may be due to the presence of different U(VI) initial concentrations in the solid phase, as it is negatively correlated with U3O8 recovery (Figure 10).


**Table 10.** Initial U6<sup>+</sup> concentration in rock samples and uranium recovery in nearly neutral media.

A similar nature of correlation between U(VI) initial concentration (in the solid phase) and leaching recovery was examined under alkaline leaching. A total of five samples with varying U(VI) initial concentrations were tested. These were the five ore samples (FR1T-860C-28, FR1T-859C-24, FR1T-860C-26, FR1T-860C-20, and FR1T-860C-24, Table 11) out of a total of 36 collected ore samples, which were used in the neutral experimental study after sample grouping. The objective was to confirm the effect of this special influential parameter on leaching recovery comparison as a trial through other

leaching media. The results are listed in Table 11 and shown in Figure 10, showing that recovery was 58% at the lower initial U(VI) concentration, while the recovery decreased by increasing U(VI) initial concentration in the solid phase. The parallel leaching experimental study showed a similar negative correlation of U(VI) initial concentration with U3O8 recovery in the leached solution.

**Figure 10.** Trend of uranium recovery (in leaching solution) with initial U6<sup>+</sup> concentration (conc.) in rock samples.

.



## 3.2.4. Effect of Redox Conditions

Trends of redox potential (Eh) values are shown as a function of time (Figure 11). Different patterns of oxidation as a function of Eh were observed. The test with 0.6 g/L of HCO3 − has a high leaching rate, and the Eh value was enhanced more than three times (from 90 to 284 mV), while the test with 1.2 g/L of HCO3 − enhanced the Eh value less than three times and a similar Eh increment was observed in the tests with 0.80 and 1.0 g/L of HCO3 − concentrations, as shown in Figure 12. For the most part, the leaching of U increased with increasing Eh, although there was then no further change in U recovery at slightly increased Eh. Mean and final values recorded for Eh for each of the leaching protocols are shown in Table 9. It should also be noted that Eh showed almost a positive correlation with recovery, i.e., a higher recovery rate in higher oxidizing conditions at the later stage of the process (Figure 12), in the test with 0.6 g/L of HCO3 −. The slope of the figure from different curves was 0.409, 0.050, 0.204, and 0.410 against HCO3 − concentrations of 0.6, 0.8, 1.0, and 1.2 g/L, respectively. The Eh was almost directly affecting U mobilization. It is well accepted that Eh of the solution is the key factor affecting the dissolution of tetravalent uranium during the leaching of U, which is related to the composition and content of the variable valence ions in the solution. Zhou et al. [46] reported that change in the uranium concentration in the leached solution is time-lagged and its peak occurred synchronously with that of high Eh.

**Figure 11.** Change in Eh measurements in the leaching solution with different HCO3 − concentrations.

**Figure 12.** Uranium recovery and Eh variations in leaching solution with different HCO3 − concentrations: (**a**) 0.60 g/L; (**b**) 0.80 g/L; (**c**) 1.00 g/L; (**d**) 1.20 g/L.

## **4. Conclusions**

This investigation included a geochemical characterization study and nearly neutral (pH: 6–6.6) leaching mechanisms of uranium (U) ore from the ELZPA deposit. The ore mineral was determined to be dominated by pitchblende, as revealed by EPMA and SEM-EDS analytical results. Leaching experiments with a CO2 + O2 system were performed to investigate U mobility and transformation. Parameters of pH, Eh, alkalinity, and major ion(s) were shown to influence U mobilization. U3O8 recovery was achieved to a high level (86%), mainly in the form of a stable UO2(CO3)2 <sup>2</sup><sup>−</sup> complex in the leached solution. The plausible U migration mechanism may be due to the oxidizing conditions, and the

dissolved CO2 gas via a formation of protons releasing oxidized U form the uranium-bearing minerals. However, the injection of CO2 may cause a dissolution of gangue minerals (i.e., calcite). Comparatively, the overall uranium recovery in different ore group tests was significantly negatively correlated with prevalent organic complexed oxidized U in the solid phase under nearly neutral and alkaline leaching environments. Therefore, further studies are needed to explore such influential parameters within the research scope.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2075-4701/10/12/1616/s1, Figure S1: Change in uranium concentration in the leaching solution.

**Author Contributions:** Conceptualization, Z.S. and Y.Z.; methodology, G.L.; software, F.A.; validation, Y.Z., G.L., G.C. and H.L.; formal analysis, F.A. and K.Z.; investigation, Z.S. and F.A.; resources, Z.S.; data curation, F.A. and K.Z.; writing—original draft preparation, F.A.; writing—review and editing, Y.Z. and G.C.; visualization, F.A.; supervision, Z.S.; project administration, Z.S.; funding acquisition, Z.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the program of the National Natural Science Foundation of China (NSFC) (Nos. 41772266, 42072285 and 41662015). The APC was funded by East China University of Technology.

**Acknowledgments:** We thank the reviewers for their comments, which have played a great role in improving the manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Uranium Removal from Groundwater and Wastewater Using Clay-Supported Nanoscale Zero-Valent Iron**

## **Borys Kornilovych 1,2,\*, Iryna Kovalchuk 2, Viktoriia Tobilko <sup>1</sup> and Stefano Ubaldini <sup>3</sup>**


Received: 17 September 2020; Accepted: 23 October 2020; Published: 26 October 2020

**Abstract:** The peculiarities of sorption removal of uranium (VI) compounds from the surface and mineralized groundwater using clay-supported nanoscale zero-valent iron (nZVI) composite materials are studied. Representatives of the main structural types of clay minerals are taken as clays: kaolinite (Kt), montmorillonite (MMT) and palygorskite (Pg). It was found that the obtained samples of composite sorbents have much better sorption properties for the removal of uranium from surface and mineralized waters compared to natural clays and nZVI.It is shown that in mineralized waters uranium (VI) is mainly in anionic form, namely in the form of carbonate complexes, which are practically not extracted by pure clays. According to the efficiency of removal of uranium compounds from surface and mineralized waters, composite sorbents form a sequence: montmorillonite-nZVI > palygorskite-nZVI > kaolinite-nZVI, which corresponds to a decrease in the specific surface area of the pristine clay minerals.

**Keywords:** uranium; clays; nanoscale zero-valent iron; groundwater; wastewater

## **1. Introduction**

Pollution of the water basin in sites where uranium ore is mined and processed is one of the most important environmental problems, requiring an effective and rational solution [1]. The sources of surface and groundwater pollution are, first of all, the areas of development of uranium deposits, storages of wastes from hydrometallurgical processing of uranium ores, sites where uranium where enriched [2–4]. Water pollution with uranium compounds is often accompanied by pollution of other toxic elements, for example, arsenic [5].

Contaminated groundwater in places of extraction and processing of uranium ores, in addition to the high content of compounds, is characterized by high mineralization. The latter is a few grams per litre, mainly due to sulfates of Ca and Mg, which are formed due to the use of sulfuric acid in the technological processes of leaching [6]. In such waters, it is mainly in the composition of negatively charged sulfate or carbonate complexes, which significantly complicates the processes of water purification [7].

Sorption methods using ion exchange resins and synthetic inorganic sorbents are most widely used for deep purification of sewage and surface waters from uranium complexes [8–10]. At the same time, if it is necessary to treat large amounts of treated water, the economic factor becomes principal.

Some of the new effective environmental protection technologies that havebeen developed and widely tested in recent years are technologies based on the use of highly reactive nanoscale zero-valent iron. The use of coarsely dispersed zero-valent iron is quite common in environmental practice in the construction of permeable reaction barriers that are installed in the path of contaminated groundwater and serve as a collector of various toxicants [11,12]. The advantage of using nZVI is the possibility of its use for groundwater purification without the application of permeable reaction barriers, the construction of which requires significant capital costs.

Suspensions based on nZVI can be pumped into the ground along the path of groundwater directly in front of contaminated areas. Further, they are carried with underground streams, promoting the decomposition of organic toxicants or sorption of inorganic pollutants on themselves.

The first successful results on the use of nZVI were obtained during the decomposition of organic toxicants (chlorinated solvents, pesticides, dyes) [13,14]. Subsequently, the effectiveness of this technology was shown for a variety of inorganic pollutants, including heavy metals [15] and natural radionuclides [16,17]. However, the potential risks for natural ecosystems when using nZVI is insufficiently studied so far [18].

The problem with the use of nZVI for in situ and ex situ technologies is the insufficient colloidal stability of its suspensions, easy aggregation, and difficulty in separating nZVI from the treated solution. To solve it, various polymers and surfactants are commonly used [19,20]. Another approach to increase the stability of suspensions is their anchoring onto a solid matrix, which allows not only to increase their stability and stability to oxidation, but also to expand the scope of their application in traditional sorption technologies [21]. Active carbons [22], amorphous silica [23,24], layered double hydroxides [25], carbonized fungi [26], graphene and composites based on it [27,28] and others were successfully investigated.

Some of the most widely used matrices as the supports of nZVIare clay minerals, which have a significant specific surface area and rather high sorption properties in relation to radionuclides and heavy metal ions [29,30]. Iron-containing composite sorbents based on clays have enhanced sorption characteristics in relation to heavy metals and radionuclides compared to the initial minerals [31]. Representatives of almost all major classes of clay minerals were studied as matrixes: kaolinite [32], montmorillonite [33,34], illite [35], palygorskite [36].

While considerable experimental material accumulated on the purification of uranium-containing water using nZVI, there was no comparative study of the sorption properties of iron-containing composites of different composition based on clay minerals. Taking into account the importance of environmental studies based on natural minerals, the removal of uranium by nZVI supported on kaolinite, montmorillonite and palygorskite was investigated, including the removal efficiency of uranium from contaminated groundwater with low and high mineralization.

## **2. Materials and Methods**

The object of the study was a layer silicate kaolinite (Al4Si4O10(OH)8) with a structure of 1:1 type. This clay was taken from the Glukhovets deposit (Ukraine), and has the most perfect crystal structure among the kaolinites from other numerous kaolin deposits of Ukraine. Among the representatives of layered silicates with a 2:1 structure (smectites), montmorillonite from the Cherkassk deposit (Ukraine) with a structural formula (Ca0.12Na0.03K0.03)0.18(Al1.39Mg0.13Fe0.44)1.96(Si3.88Al0.12)4.0O10(OH)<sup>2</sup> × nH2O was chosen. Fibrous clay mineral palygorskite (also known as attapulgite) with a structural formula Mg5Si8O20(OH)<sup>2</sup> × 4H2O was also taken from the Cherkassk deposit (Ukraine).

The purification of natural minerals from impurities of quartz, feldspars, carbonates, aluminium and iron oxides was carried out.X-ray powder diffraction (XRD) patterns were recorded on X-ray diffractometer DRON-4-07 (Russia) in the range of 2–40◦ (2θ) using CuKα irradiation.

The parameters of the porous structure: specific surface area (SSA), total pore volume (V), average pore radius (r) of the natural and composite sorbents were determined by the BET method from nitrogen adsorption isotherms obtained on a Nova 2200e gas analyzer (USA), pore distribution(Rmax)—by BJH and DFT methods [37].

To obtain composite sorbents "clay mineral/nZVI" with a mass ratio nZVI: clay mineral~0.2: 1 modified method was used [38,39]. To do this, in a 0.2M solution of Fe(NO3)3. 9H2O, weighed clay was introduced.The resulting suspension (pH2) was quantitatively transferred to a three-necked flask and the ion Fe3<sup>+</sup> reduction process was performed with a solution of sodium borohydride (NaBH4) in a nitrogen atmosphere. The resulting composite sorbent was then separated from the liquid phase by centrifugation and washed three times with alcohol. The resulting precipitate was dried under vacuum at a temperature of 60 ◦C and crushed to obtain a fraction ≤0.1 mm.

Water purification from uranium ions was performed using natural clay minerals and composite sorbents. Solutions were prepared with distilled water and uranyl trihydrosulfate salt (UO2SO4·3H2O). 1M NaCl solution to create ionic strength (~0.01) was used. The pH of the solutions was adjusted with 0.1M solutions NaOH and HCl.

Mineralized waters were used solutions, whose composition corresponded to the composition of underground mineralized water near the liquid waste storage facility for hydrometallurgical processing uranium ores at the Centre of Ukrainian Uranium Industry (Zhovti Vody) by anions: HCO− <sup>3</sup> —450; Cl−—180; SO2<sup>−</sup> <sup>4</sup> —2830; NO<sup>−</sup> <sup>3</sup> —130 mg/l and by cations: Ca2<sup>+</sup>—576; Mg2+—209; (Na<sup>+</sup> <sup>+</sup> <sup>K</sup>+)—391; NH<sup>+</sup> <sup>4</sup> —0.92; Ni2<sup>+</sup>—<0.05; <sup>C</sup>*u*2+—<0.03; <sup>C</sup>*o*2+—<0.06; Mnsum—0.10; Zn2<sup>+</sup>—<0.01; Pb2<sup>+</sup>—<0.19; Cd2+—<0.01; Fesum—0.05 mg/L [40]. The output solutions were prepared on the basis of the corresponding sodium salts, the total salt content was—5280 mg/L, pH 7.2.

Sorption of uranium ions was performed under static conditions at room temperature with continuous shaking of the samples for 1 h (the volume of the aqueous phase—50 mL, the amount of sorbent—0.1 g). Inthe end, the liquid phase was separated by centrifugation (6000 rpm) and the equilibrium metal concentration was determined spectrophotometrically (UNICO 2100UV) using an Arsenazo III reagent at a wavelength of 665 nm.

## **3. Results**

X-ray analysis of samples shows the presence of only small impurities in pristine minerals (Figure 1).

**Figure 1.** X-ray powder diffraction (XRD) patterns of pristine and nanoscale zero-valent iron (nZVI)—modified clay minerals.

On diffractograms (XRD) of all modified minerals, there are weak reflections at 0.252 and 0.202 nm corresponding to crystalline phases of zero-valent iron (α-Fe), iron oxide (FeO), and also, in time, smaller quantities of goethite (FeOOH). Clays kept their structures in mixtures, since no structural changes between neat and supported clays were found.The amount of iron applied to the surface is, according to the chemical analysis of the solution after treatment of the modified samples with hydrochloric acid, 0.17 mg/g for all minerals.

By nature, the nitrogen sorption isotherm on the pristine kaolinite (Figure 2), according to the modified de Boer classification, belongs to type II (b) isotherms and are typical for nonporous sorbents with a small macroporous component [41].

**Figure 2.** Nitrogen adsorption–desorption isotherms of pristine and nZVI—modified clay minerals: (**a**)—kaolinite and montmorillonite,(**b**)—palygorskite.

The narrow hysteresis loop of type H3 on isotherms is the result of capillary condensation in the structural aggregates of kaolinite between weakly interconnected flat elementary packages of the mineral. The nature of the nitrogen adsorption curves on the samples of montmorillonite and paligorskite are similar to kaolinite and, thus, also belong to type II (b) isotherms with a hysteresis loop type H3. The calculated characteristics of the porous structure of the samples are given in Table 1. As can be seen from the nitrogen isotherms, the specific surface area of samples sharply decreases after the surface modification. Such reduction is stipulated by aggregation processes of small clay particles by nZVI and practically complete closing of micropores with nZVI films and the resultant blocking of the access of nitrogen molecules to these pores.


**Table 1.** Characteristics of the porous structure of the samples.

Sorption (q, μmol/g) of uranium ions on pristine and modified samples are shown in Figure 3. On the kaolinite surface, sorption can occur on active centres of two types: ditrigonal siloxane wells on the basal surfaces of kaolinite particles and hydroxyl groups on broken bonds in tetrahedral (Si-O-Si) and octahedral (Al-O-Al) sheets on edge surfaces of the particles [42].

**Figure 3.** Sorption isotherms of U(VI) ions on pristine kaolinite and kaolinite-nZVI.

However, for the pristine Glukhovets kaolinite, due to its perfect structure and the absence of heterovalent isomorphic substitutions in the mineral structure, the charge of the structural packages is close to zero and, therefore, sorption at low concentrations of uranium ions in solution occurs mainly on hydroxyl groups located on edge surfaces of the particles.

Such sorption centres on the edge surfaces, depending on the pH of the medium, can have the composition >Si-OH and >Si-O<sup>−</sup> on the broken tetrahedral sheet of the mineral and >Al-OH2 <sup>+</sup>; Al-OH and >Al-OH− on a broken octahedral sheet of the mineral and sorption of uranium ions on the surface occurs with the formation of strong inner-sphere surface complexes.

The results obtained for the modified samples indicate that the sorption values of uranium ions at pH 6 are several times higher than those for the pristine minerals (Figure 3). As it was established, synthesized nZVI has a core–shell structure and the thin film on the surface of the particles consists of iron oxides such as FeO, Fe2O3 and Fe3O4 and its hydroxides [14,17,43,44]. Fe(0) is in the centre of the particles. As a result of such composite structure, the removal of uranium (VI) from water by nZVI is also may be affected by the redox transformation of a highly soluble form of U(VI) to less soluble U(IV)with direct electron transfer at the Fe(0) surface [16,17,45]:

$$\mathrm{Fe}^{0} + 1.5\mathrm{U}\mathrm{O}\_{2}^{2+} + 6\mathrm{H}^{+} = \mathrm{Fe}^{3+} + 1.5\mathrm{U}^{4+} + 3\mathrm{H}\_{2}\mathrm{O} \tag{1}$$

In order to elucidate the synergetic effect for the Kt-nZVI composite sorbents, the removal of U(VI) by nZVI alone was also investigated. As shown in Figure 3, the removal efficiencies of U(VI) by nZVI alone wassignificantly lower than that of Kt-nZVI.

The dependence of the sorption values of uranium ions on the surface of kaolinite particles from pHis presented in Figure 4. The corresponding curves have a characteristic form with a marked minimum in the acidic and alkaline pH areas.

**Figure 4.** Sorption of U(VI) ions on pristine kaolinite and kaolinite-nZVI as a function of pH.

The last is due to the peculiarities of the chemistry of the clay surface and the complex chemistry of aqueous solutions of uranium salts [46]. In the acidic pH area, the sorption characteristics of clays in relation to uranyl cations UO2 <sup>2</sup><sup>+</sup> are determined by the neutral and positive surface charge resulting from protonation of surface groups and the formation of >>Si-OH, >Al-OH and >Al-OH2 <sup>+</sup> groups. In the alkaline area, dissociated >Si-O− and >Al-O−groups predominate on the clay surface, and uranium in solution exists mainly in the form of neutral UO2(OH)2, UO2CO3 and negatively charged species (UO2)2CO3(OH)3 − [7,47]. This causes insignificant values of sorption of uranium compounds in these pH areas, because positively charged uranium species which predominated in the acidic pH areas are not adsorbed onto the positively charged clay surface, and negatively charged uranium species in the alkaline pH areas are not adsorbed onto the negatively charged clay surface. In the neutral pH area, the charge of the kaolinite surface and the charge of uranium ions have opposite signs, which influences the proceedings of sorption processes and the appearance of maximum on the curves of the dependence of sorption values from pH.

Montmorillonite has a labile structure with the possibility of significant expansion of the inter-package distance during the penetration of polar water molecules between flat structural packages. This determines the availability for ion exchange processes not only of the outer but also of the inner surface of the particles of this clay mineral, which determines a significant increase in the sorption of uranium ions compared to kaolinite (see Figure 5). The nature of the dependence of sorption values from pH for montmorillonite corresponds to that for kaolinite, however, is less pronounced (see Figure 6). This is due to the smaller relative contribution of the edge surfaces of this mineral to its total specific surface area and the increased role of sorption centres on the basal surfaces of particles (ditrigonal wells) whose behavior is independent of pH [42].

**Figure 5.** Sorption isotherms of U(VI) ions on pristine montmorillonite and montmorillonite-nZVI.

**Figure 6.** Sorption of U(VI) ions on pristine montmorillonite and montmorillonite-nZVI as a function of pH.

Particles of chain silicate paligorskite have an elongated rod-like shape, the volume of which is pierced by zeolite-like channels [48]. Sorption of ions occurs on the outer surface of the mineral, the value of which is greater than that of kaolinite and montmorillonite established by N2 sorption. However, montmorillonite particles show an outstanding property to delaminate into individual silicate layers or thin packets of them in solutions. So, the values of the specific surface area of dispersions of montmorillonite samples in solutions are much higher than of air-dried samples (maximum crystallographic value of SSA are ~750–780 m2/g [41]). Therefore, the values of the sorption of uranium on both the pristine and modified minerals are the average values between those for kaolinite and montmorillonite (see Figures 7 and 8).

**Figure 7.** Sorption isotherms of U(VI) ions on pristine paligorskite and paligorskite-nZVI.

**Figure 8.** Sorption of U(VI) ions on pristine paligorskite and paligorskite-nZVI as a function of pH.

When considering the sorption processes in mineralized waters, it is important to analyze the forms in which uranium exists under these conditions. *Medusa* software, which is widely used in analytical practice, was used for calculations [49]. The main solid phases of uranium in aqueous systems are insoluble hydrates UO3 · H2O or UO2(OH)<sup>2</sup> (lg*Ksp*= −20.34 – −23.5) [47,50] and carbonate UO2CO3 (lg*Ksp*= −13.21 – −14.26) [51].At the same time, sulfate, carbonate, phosphate and nitrate complexes of uranium may be the main species in water.The composition of main anions of underground mineralized water near the liquid waste storage facility for hydrometallurgical processing uranium ores at the Centre of Ukrainian Uranium Industry (Zhovti Vody)are: HCO− <sup>3</sup> —450; Cl−—180; SO2<sup>−</sup> <sup>4</sup> —2830; NO<sup>−</sup> <sup>3</sup> —130 mg/L. The high affinity of uranyl ions to the nitrate is known, but the content of nitrate-ions in Zhovti Vody mineralized water is much smaller thanthe content of carbonate-ions and sulfate-ions. Therefore, we did not include nitrate-ions complexes in the speciation calculations. Zhovti Vody-mineralized waters, even with a sufficiently high content of uranium in them, are characterized by almost complete binding of uranium to sulfate complexes UO2SO4

and UO2(SO4) 2− <sup>2</sup> in the acidic area, as well as in carbonate complexes UO2CO3, UO2(CO3) 2− <sup>2</sup> and UO2(CO3) 4− <sup>3</sup> in neutral and alkaline areas (see Figure 9).

Sorption processes in mineralized waters were studied using montmorillonite and paligorskite. The pH dependence curves of uranium sorption values have a characteristic form for both types of clays with a maximum in the pH area of 5–7 (Figure 10).

**Figure 10.** Sorption of U(VI) ions on pristine and nZVI-modified montmorillonite and palygorskite from mineralized waters as a function of pH.

Sorption of uranium complexes occurs primarily due to the exchange of hydroxyl ions of the hydroxide film on the surface of nanoscale iron particles:

$$\text{M-(OH)}\_{\text{n}} + \text{UO}\_{2}(\text{SO}\_{4})\_{2}{2}^{2-} \rightarrow \text{M-(OH)}\_{\text{n-2}}\text{UO}\_{2}(\text{SO}\_{4})\_{2} + 2\text{OH}^{-}\tag{2}$$

$$\text{M-(OH)}\_{\text{n}} + \text{UO}\_{2}(\text{CO}\_{3})\_{2}^{2-} \rightarrow \text{M-(OH)}\_{\text{n-2}}\text{UO}\_{2}(\text{CO}\_{3})\_{2} + 2\text{OH}^{-}\tag{3}$$

When considering the mechanism of sorption of uranium compounds on the surface of clays, especially of rather high concentration of uranium, the possibility of precipitation of sparingly soluble hydroxides and others(hexavalent uranium compounds schoepite, carnotite, tyuyamunite, etc. [52]) cannot be ruled out. In mineralized waters, in contrast to diluted waters, the solubility of insoluble salts of uranium is significantly increased. Therefore, at concentrations of salts corresponding to those in mineralized waters from 3–5 to 10–20 g/L [52], the precipitation of uranium solid phases is not expected and the most probable mechanism of uranium binding is only the sorption mechanism.

Sorption isotherms of uranium from mineralized waters were obtained at pH 7.2, which corresponds to the pH value of real groundwater (Figure 11).

**Figure 11.** Sorption isotherms of U(VI) ions on pristine and nZVI-modified montmorillonite and palygorskite from mineralized waters.

Sorption isotherms were analyzed using Langmuir and Freundlich equations. The results of the calculations of the corresponding coefficients are shown in Table 2.


**Table 2.** Langmuir and Freundlich parameters for the adsorption of uranyl ions onto pristine and nZVI-modified minerals.

The obtained isotherms are well described by the equation of monomolecular Langmuir sorption (correlation coefficient R2 = 0.968–0.996) which assumes the energy homogeneity of the active centres accordingly. For the empirical Freundlich equation, which is suitable mainly for describing the starting areas of isotherms, the correlation coefficient is lower (R2 = 0.934–0.991).

## **4. Conclusions**

Thus, different in structure clays are effective cheap matrices for obtaining efficient sorption materials based on zero valence iron for purification of uranium-contaminated surface and mineralized groundwater, which is typical for areas of uranium ore mining and processing.In terms of sorption capacity, the composite samples form a sequence: montmorillonite-nZVI > paligorskite-nZVI > kaolinite-nZVI, which corresponds to a decrease in the specific surface area of the pristine clay minerals.

Removal of U(VI) from mineralized waters occurs primarily due to the exchange of hydroxyl ions of the hydroxide film on the surface of nanoscale iron particles by uranium sulfate and carbonate complexes.

Another possible mechanism of immobilization of uranium compounds is the reduction of uranium (VI) to uranium (IV) by electron transfer from the volume of nZVI particles through a hydroxide film to their surface with the formation and deposition of much less soluble compounds of the latter.However, it is necessary to specify the main reactions in the removal of uranium by immobilized ZVI for its further practical applications.

**Author Contributions:** Conceptualization, B.K.; methodology, I.K. and V.T.; validation, B.K. and I.K.; formal analysis, I.K.; investigation, I.K., V.T.; writing-original draft preparation, B.K., S.U.; writing—review & editing, B.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** Author is grateful to the staff of Institute for Sorption and Problems of Endoecology and Igor Sikorsky Kyiv Polytechnic Institute for the help and support.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article*
