Introduction

Ocean manganese nodules, also known as polymetallic nodules, are widely distributed on the seafloor of the global ocean at depths of about 4000–6000 m [1]. They are rich in vital metals, such as Mn, Cu, Co, and Ni, which are widely used in manufacturing industries, of which Cu, Co, and Ni are in great demand in high-technology industries [2, 3]. The abundances of Mn, Cu, Co, and Ni in ocean manganese nodules have been reported to be 168.06, 47.00, 113.60, and 64.00 times that of the earth's crust, respectively, and it is considered the most economically valuable marine mineral resource [4]. Since the mining cost of deep-sea mineral resources is much higher than that of terrestrial resources, and the value of Co and Ni is much higher than that of Mn, the simultaneous recovery of multiple valuable metals from ocean manganese nodules must be considered in order to improve the economics of their processing.

Generally speaking, ocean manganese nodules are aggregates consisting of iron-manganese oxides shells surrounding a core of clay minerals, which are essentially spherical, although their individual shapes can vary from nearly plate-like to flattened disk-like [5, 6]. In manganese nodules, manganese minerals are usually found in todorokite, birnessite, vernadite, and buserite, while iron mainly forms trivalent iron oxides and hydroxides [7, 8]. According to relevant studies, the crystallinity of manganese nodules is very poor and their main components, manganese oxides and iron oxides are present as disseminated associations [9, 10]. The valuable metals Cu, Co, and Ni exist in the aggregate of iron-manganese oxides in the form of isomorphism or adsorption [11]. Therefore, the key to extracting valuable metals from ocean manganese nodules is to destroy their original mineral structure.

In order to effectively extract valuable metals from the ocean manganese nodules, researchers in many countries have carried out a lot of research work and reported many extraction methods, which are mainly classified into two categories: wet reduction leaching and fire reduction leaching. Wet reduction leaching mainly includes cuprous ions reduction-ammonia leaching [12], ferrous ions reduction-acid leaching [13, 14], sulfuric acid leaching at high temperature and pressure [15], sulfur dioxide or sulfite reduction leaching [16, 17], reduction leaching with organic reducing agents [18, 19], and bioleaching [20, 21]. Recently, there are many related talking about bioleaching as a greener process for the extraction of metals from various sources [22,23,24]. However, the obvious disadvantages of wet reduction leaching are low efficiency and metal recovery [25]. Fire reduction leaching can be further divided into sulfation roasting-ammonia leaching [26], chlorination roasting-acid leaching [27], smelting reduction [28, 29], and reduction roasting-acid leaching [30,31,32]. The sulfation and chlorination roasting processes are not only very corrosive to the equipment, but they also emit highly polluting gases SO2 and Cl2. Smelting reduction is a typical pyrometallurgical process for processing ocean manganese nodules, but the high energy consumption is required to obtain a high recovery of valuable metals [33]. Besides that, due to the use of coal as a reducing agent, this process releases large amounts of pollutants such as sulfur dioxide, nitrogen oxides, and carbon dioxide into the atmosphere, causing serious environmental pollution problems [34]. In recent years, studies on the extraction of manganese from ocean manganese nodules or terrestrial manganese oxide ores by the reduction roasting-acid leaching process with biomass as a reducing agent have been widely reported [30, 31, 35,36,37]. It is believed that biomass pyrolysis reduction can effectively reduce acid gas NOx and SOx emissions compared with the traditional coal reducing agent, and at the same time, while also achieving the reduction of high-valent manganese oxide ores at lower temperatures, thus significantly reducing energy consumption [38]. Nevertheless, this process also generates a large amount of greenhouse gas CO2, and in addition, the tar produced by the pyrolysis of biomass is introduced into the leaching solution, causing unavoidable problems for the recovery of metals and the wastewater treatment.

Ammonium sulfate roasting-water leaching technology is considered a promising process due to its low roasting temperature, environmental friendliness, low cost, and high extraction efficiency [39,40,41]. This process enables the conversion of metal oxides to metal sulfates or ammonium salts at lower temperatures so that effective extraction of valuable metals can be achieved by leaching the roasted slag with pure water. Since 2017, the extraction of valuable metals from various ores or solid wastes by ammonium sulfate roasting technology has been widely reported such as laterite [42], blast furnace slag[43, 44], zinc oxidized ore [45], vanadium slag [46, 47], spent Nd-Fe-B magnets [48], bauxite residue [49], and spent lithium nickel cobalt manganese oxides [50, 51]. However, no report was found on the application of ammonium sulfate roasting-water leaching technology for extracting valuable metals from ocean manganese nodules.

Hence, considering the clean characteristics of ammonium sulfate roasting technology and its potential application in metals extraction, a sustainable low-temperature roasting and water leaching process for simultaneously extracting Mn, Cu, Co, and Ni from ocean manganese nodules was proposed in this study. The feasibility of the ammonium sulfate roasting process for the reduction of ocean manganese nodules was analyzed from a thermodynamic point of view. The effects of roasting temperature, ammonium sulfate dosage, and roasting time on mineral phase transformation and leaching efficiency of valuable metals were systematically investigated. And the effects of leaching temperature and leaching time on the leaching efficiency of valuable metals were explored. In addition, the mechanism for extracting valuable metals from ocean manganese nodules via ammonium sulfate roasting-water leaching was analyzed and determined.

Materials and Methods

Materials

The ocean manganese nodules used in this study were mined on the seabed of the Western Pacific Ocean and were provided by the China Ocean Mineral Resources R&D Association (COMRA). The ocean manganese nodules were crushed and ground to obtain samples with particle sizes less than 0.074 mm. Then, the samples were dried at 105 ℃ for 120 min before being used in roasting experiments. The 0.1 g of the ocean manganese nodules was completely dissolved in mixed acid (20 ml HCl, 5 ml HNO3, and 10 ml HF) and diluted to 100 ml. The chemical composition in this solution was analyzed by an inductively coupled plasma optical emission spectrometer (ICP-OES: Optima 7300 V, Perkin-Elmer, America) at radio frequency (RF) power of 1.35 kW and argon auxiliary flow rate of 0.45 L/min. The chemical composition in the ocean manganese nodules obtained by calculation is shown in Table 1. It can be seen that the chemical composition of the ocean manganese nodules was mainly Mn, Si, and Fe, and there are also minor amounts of Cu, Co, and Ni.

Table 1 Chemical composition of the oceanic manganese nodules (wt.%)

The X-ray diffractometer (XRD: 7000S/L, SHIMADZU, Japan) was used to determine the phase composition of the ocean manganese nodules, and the results are shown in Fig. 1. Disordered diffraction peaks were observed from the XRD pattern (Fig. 1), indicating that the crystallinity of the ocean manganese nodules was poor, and the existing phase compositions were primarily quartz, anorthite, birnessite, and todorokite. The ammonium sulfate used in the roasting experiment was of analytical grade and was purchased from Shanghai McLean Biochemical Technology Co., Ltd.

Fig. 1
figure 1

XRD pattern of the ocean manganese nodules

Experiments Procedures

The roasting and leaching experiments are depicted schematically in Fig. 2. The following were the specific experimental steps: The 10 g of ocean manganese nodules and ammonium sulfate were mixed evenly at a certain mass ratio in each experiment. The porcelain boat containing the mixed samples was placed in the tube furnace, and the power switch was turned on with the heating rate set to 20 ℃/min. Three parallel experiments were performed for each condition. During the roasting process, compressed air was introduced at a rate of 100 mL/min, allowing the volatilized ammonia gas to enter the gas washing bottle and be absorbed by the dilute sulfuric acid solution. After the roasting reaction, the roasting slag in the porcelain boat was taken out.

Fig. 2
figure 2

The schematic diagram of the roasting device and leaching device

The roasting slag was subsequently leached with pure water. The following were the leaching conditions: the liquid–solid ratio was 5:1, the leaching temperature was 60 ℃, and the leaching time was 60 min. After the leaching reaction, the slurry was filtered using a vacuum filter, and the residues were rinsed several times with pure water. Then, the leaching residues were dried at 105 ℃ for 120 min and weighed. The target metal contents of the roasted slag and leached residue were determined by ICP-OES. The metal leaching efficiency was calculated using the following Eq. (1) based on the principle of mass conservation.

$$ \eta \left( \% \right) = \left( {1 - \frac{{m_{1} c_{1} }}{{m_{0} c_{0} }}} \right) \times 100\% , $$
(1)

where \(\eta \) is expressed as leaching efficiency of metal (%); \({c}_{0}\) and \({c}_{1}\) are the concentrations of metal in roasting slag and leaching residue (g/g), respectively; \({m}_{0}\) and \({m}_{1}\) are the mass of roasting slag and leaching residue for each leaching experiment (g), respectively.

Results and Discussion

Thermodynamic Analysis of the Roasting Process

The thermal decomposition process of ammonium sulfate is a gradual transition process from the solid phase to the liquid phase to the gas phase. To understand the decomposition process of ammonium sulfate, a thermogravimetric analysis of ammonium sulfate was performed. Figure 3a shows TG and DTA curves of ammonium sulfate in the air atmosphere. The DTA curve for (NH4)2SO4 (Fig. 3a) shows two endothermic peaks, indicating that decomposition occurs in at least two steps. The weight loss ratio of (NH4)2SO4 between 25 ℃ and 350 ℃ was 21.36%, demonstrating that thermal decomposition of (NH4)2SO4 occurred, and the main reactions are expressed as Eqs. (26) [52,53,54,55]. According to the research of Kiyoura and Urano [53], (NH4)2SO4 thermally decomposed at temperatures above 100 ℃ to produce NH4HSO4, while NH4HSO4 and (NH4)2SO4 can react between 100 and 170 ℃ to form (NH4)3H(SO4)2. When the temperature was higher than 200 ℃, NH4HSO4 lost water to form NH2SO3H, which can be further decomposed into many gases (Eq. (5)). In addition, they also reported the formation of (NH4)2S2O7 via the reaction of NH4HSO4 and NH2SO3H at about 200 ℃. After 350 ℃, the TG curve dropped sharply, which was caused by the gasification reaction of (NH4)2S2O7 and (NH4)2SO4 (Eqs. (7) and (8)).

$$ \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) \, = {\text{ NH}}_{{4}} {\text{HSO}}_{{4}} \left( {\text{s}} \right) \, + {\text{ NH}}_{{3}} \left( {\text{g}} \right), $$
(2)
$$ \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) \, + {\text{ NH}}_{{4}} {\text{HSO}}_{{4}} \left( {\text{s}} \right) \, = \, \left( {{\text{NH}}_{{4}} } \right)_{{3}} {\text{H}}\left( {{\text{SO}}_{{4}} } \right)_{{2}} ({\text{s}}), $$
(3)
$$ {\text{NH}}_{{4}} {\text{HSO}}_{{4}} \left( {\text{s}} \right) = {\text{ NH}}_{{2}} {\text{SO}}_{{3}} {\text{H}}\left( {\text{s}} \right)_{{}} + {\text{ H}}_{{2}} {\text{O}}\left( {\text{g}} \right), $$
(4)
$$ {\text{3NH}}_{{2}} {\text{SO}}_{{3}} {\text{H}}\left( {\text{s}} \right)_{{}} = {\text{3SO}}_{{2}} \left( {\text{g}} \right) \, + {\text{ NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ 3H}}_{{2}} {\text{O}}\left( {\text{g}} \right) \, + {\text{ N}}_{{2}} \left( {\text{g}} \right), $$
(5)
$$ {\text{NH}}_{{4}} {\text{HSO}}_{{4}} \left( {\text{s}} \right) + {\text{NH}}_{{2}} {\text{SO}}_{{3}} {\text{H}}\left( {\text{s}} \right) \, = \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{S}}_{{2}} {\text{O}}_{{7}} \left( {\text{s}} \right), $$
(6)
$$ {3}\left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{S}}_{{2}} {\text{O}}_{{7}} \left( {\text{s}} \right) \, = {\text{ 2NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ 2N}}_{{2}} \left( {\text{g}} \right) \, + {\text{ 6SO}}_{{2}} \left( {\text{g}} \right) \, + {\text{ 9H}}_{{2}} {\text{O}}\left( {\text{g}} \right), $$
(7)
$$ {3}\left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) \, = {\text{4NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ N}}_{{2}} \left( {\text{g}} \right) \, + {\text{ 3SO}}_{{2}} \left( {\text{g}} \right) \, + {\text{ 6H}}_{{2}} {\text{O}}\left( {\text{g}} \right). $$
(8)
Fig. 3
figure 3

TG and DTA curves of the ammonium sulfate (a). Relationships between ΔG and T for main reactions that may occur during roasting (b)

Manganese and iron in the ocean manganese nodules mainly exist in tetravalent and trivalent forms, respectively, which are similar to MnO2 and Fe2O3, while copper, cobalt, and nickel exist in the aggregate of iron-manganese oxides in the form of isomorphism or adsorption. Depending on the thermal decomposition process of ammonium sulfate, the reactions of the following Eqs. (916) may occur in the roasting process. The relationship between ΔG and T for the main reactions between 0 and 600 ℃ was plotted using the professional software HSC Chemistry 6.0, and the results are shown in Fig. 3b. According to Fig. 3b, when the temperature exceeds 200 ℃, the reducing gas SO2 is generated by the thermal decomposition of ammonium sulfate. In the whole temperature range, ΔG of Eq. (9) is less than 0, indicating that SO2 can react with MnO2 in the ocean manganese nodules to form water-soluble MnSO4. The reaction process in Eq. (10) can react spontaneously at temperatures above 250 ℃, whereas the reaction process in Eq. (11) requires temperatures above 350 ℃. When the temperature is higher than 350 ℃, the ΔG of Eq. (12) starts to become negative, indicating that the conversion of Fe2O3 to Fe2(SO4)3 by ammonium sulfate roasting requires a roasting temperature higher than 350 ℃. However, if SO3 is present in the system, it can complete this conversion process at a lower temperature (Eq. (13)). The conversion of copper, cobalt, and nickel oxides into water-soluble sulfates all require roasting temperatures higher than 300 ℃.

$$ {\text{MnO}}_{{2}} \left( {\text{s}} \right) + {\text{SO}}_{{2}} \left( {\text{g}} \right) = {\text{MnSO}}_{{4}} \left( {\text{s}} \right), $$
(9)
$$ {\text{MnO}}_{{2}} \left( {\text{s}} \right) \, + \, \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) \, + {\text{ SO}}_{{2}} \left( {\text{g}} \right) \, = {\text{ MnSO}}_{{4}} \left( {\text{s}} \right) \, + {\text{ 2NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ H}}_{{2}} {\text{O}}\left( {\text{g}} \right) \, + {\text{ SO}}_{{3}} \left( {\text{g}} \right), $$
(10)
$$ {\text{2MnO}}_{{2}} \left( {\text{s}} \right) + { 2}\left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) = {\text{ 2MnSO}}_{{4}} \left( {\text{s}} \right) + {\text{ 4NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ 2H}}_{{2}} {\text{O}}\left( {\text{g}} \right) \, + {\text{ O}}_{{2}} \left( {\text{g}} \right), $$
(11)
$$ {\text{Fe}}_{{2}} {\text{O}}_{{3}} \left( {\text{s}} \right) + { 3}\left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) = {\text{ Fe}}_{{2}} \left( {{\text{SO}}_{{4}} } \right)_{{3}} \left( {\text{s}} \right) + {\text{ 6NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ 3H}}_{{2}} {\text{O}}\left( {\text{g}} \right), $$
(12)
$$ {\text{Fe}}_{{2}} {\text{O}}_{{3}} \left( {\text{s}} \right) + {\text{ 3SO}}_{{3}} \left( {\text{g}} \right) \, = {\text{ Fe}}_{{2}} \left( {{\text{SO}}_{{4}} } \right)_{{3}} \left( {\text{s}} \right), $$
(13)
$$ {\text{NiO}}\left( {\text{s}} \right) + \, \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) = {\text{ NiSO}}_{{4}} \left( {\text{s}} \right) + {\text{ 2NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ H}}_{{2}} {\text{O}}\left( {\text{g}} \right), $$
(14)
$$ {\text{CoO}}\left( {\text{s}} \right) + \, \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) = {\text{ CoSO}}_{{4}} \left( {\text{s}} \right) + {\text{ 2NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ H}}_{{2}} {\text{O}}\left( {\text{g}} \right), $$
(15)
$$ {\text{CuO}}\left( {\text{s}} \right) + \, \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) = {\text{ CuSO}}_{{4}} \left( {\text{s}} \right) + {\text{ 2NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ H}}_{{2}} {\text{O}}\left( {\text{g}} \right). $$
(16)

Based on the above analysis, the metal oxides in the ocean manganese nodules can theoretically be converted into corresponding sulfates by ammonium sulfate roasting at temperatures ranging from 300 to 400 ℃. Subsequently, the valuable metals can be effectively extracted by water leaching.

Effect of Roasting Temperature

Roasting temperature is a crucial factor in the roasting process, which affects not only the thermal decomposition of (NH4)2SO4 but also the sulfation reaction between (NH4)2SO4 and ocean manganese nodules. The effect of roasting temperature on the leaching efficiency of the valuable metals in the ocean manganese nodules was investigated under the conditions of mass ratio for (NH4)2SO4 to ocean manganese nodules of 2:1 and roasting time of 120 min, and the experimental results are shown in Fig. 4.

Fig. 4
figure 4

Effect of roasting temperature on leaching efficiency of the valuable metals

As shown in Fig. 4, the leaching efficiencies of Mn, Cu, Co, and Ni were 28.87%, 31.02%, 14.69%, and 13.53%, while the leaching efficiency of Fe was only 4.96% when roasting was performed at 300 ℃. At this temperature, the (NH4)2SO4 most likely did not react strongly with the ocean manganese nodules. With the continuous increase of temperature, the leaching efficiency of the valuable metals gradually increased. When the roasting temperature was increased to 360 ℃, the leaching efficiencies of Mn, Cu, Co, and Ni increased to 94.20%, 78.42%, 69.32%, and 82.11%, and the leaching efficiency of Fe at this time was 32.48%. Because pure water was used as the leaching agent, some of the dissolved Fe was converted to Fe(OH)3 precipitation during the leaching process resulting in a low leaching efficiency for Fe. The leaching efficiency of the valuable metals was increased slightly by increasing the roasting temperature from 360 to 380 ℃. The leaching efficiencies of Mn, Cu, Co, Ni, and Fe reached 97.48%, 84.51%, 77.57%, 90.23%, and 38.37%, respectively, at the roasting temperature of 380 ℃. However, continuing to increase the roasting temperature, the leaching efficiency of the valuable metals remained with the results obtained at 380 ℃. Thus, the optimum roasting temperature was determined to be 380 ℃.

In order to investigate the essential reason for the influence of roasting temperature on the leaching efficiency of the valuable metals, the roasted slag obtained at different roasting temperatures was analyzed by XRD, and the results are presented in Fig. 5. As shown in Fig. 5, the XRD pattern of the roasted slag at the roasting temperature of 300 ℃ was dominated by the diffraction peaks of (NH4)2SO4, in addition to a few diffraction peaks of NH4HSO4, NH2SO3H, and MnSO4, indicating that (NH4)2SO4 did not undergo strong thermal decomposition nor strong sulfation reaction with the ocean manganese nodules at this temperature. The diffraction peaks of a small amount of MnSO4 may be formed by the reaction of the reducing gas SO2 produced by thermal decomposition for ammonium sulfate with MnO2, as shown in Eq. (9) in “Thermodynamic Analysis of the Roasting Process”. When the temperature was increased to 320 ℃, the diffraction peaks of the roasting slag were basically the same as 300 ℃, but the number of diffraction peaks of (NH4)2SO4 was reduced. The diffraction peaks of (NH4)2SO4 were significantly reduced when the roasting temperature was 340 ℃, and many diffraction peaks of (NH4)2Mn2(SO4)3 with lower intensity appeared.

Fig. 5
figure 5

XRD patterns of roasting slag at different roasting temperatures

The diffraction peaks of roasted slag changed significantly when the roasting temperature was increased to 360 °C, and the diffraction peaks of (NH4)2SO4, NH4HSO4, and NH2SO3H disappeared. The main diffraction peaks changed to (NH4)2Mn2(SO4)3, and a large number of diffraction peaks of NH4Fe(SO4)2 appeared. This indicates that the (NH4)2SO4 reacted strongly with the Fe2O3 in the ocean manganese nodules only when the roasting temperature was raised to 360 ℃, so the leaching efficiency of Fe increased significantly as the roasting temperature was increased from 340 to 360 ℃. The types and intensities of diffraction peaks in roasted slag XRD patterns remained essentially unchanged after 360 ℃, with the only difference being that the diffraction peaks of MnSO4 increased, which was formed by the decomposition of (NH4)2Mn2(SO4)3. According to the above analysis, the sulfation reaction could be completed when the roasting temperature was 380 ℃, and the iron-manganese oxides in the ocean manganese nodules were converted into (NH4)2Mn2(SO4)3, NH4Fe(SO4)2, and MnSO4.

Effect of Ammonium Sulfate Dosage

The effect of (NH4)2SO4 dosage on the leaching efficiency of the valuable metals was explored at the roasting temperature of 380 ℃ and the roasting time of 120 min (Fig. 6). As can be seen from Fig. 6, the amount of (NH4)2SO4 has a significant effect on the leaching efficiency of the valuable metals. The leaching efficiencies of Mn, Cu, Co, Ni, and Fe were only 47.52%, 39.01%, 27.93%, 57.31%, and 22.17%, respectively, for the 0.5:1 of (NH4)2SO4 dosage (The (NH4)2SO4 dosage of 0.5:1 means that the mass ratio of (NH4)2SO4 to ocean manganese nodules is 0.5:1, and similar expressions are used in the rest of the paper), which may be since the amount of (NH4)2SO4 was not enough to convert all the iron-manganese oxides in the ocean manganese nodules into water-soluble sulfates, and the original mineral phases of the ocean manganese nodules were not completely destroyed. With the increase of (NH4)2SO4 dosage, the leaching efficiency of the valuable metals also increased significantly. When the dosage of (NH4)2SO4 was increased to 1.25:1, the leaching efficiencies of Mn, Cu, Co, Ni, and Fe reached 96.14%, 83.77%, 75.25%, 87.52%, and 36.28%, respectively. After 1.25:1, the leaching efficiencies of Mn, Cu, Co, Ni, and Fe increased only very slightly with the increase of (NH4)2SO4 dosage. The increase of (NH4)2SO4 dosage will significantly increase the production cost, and therefore, the optimum (NH4)2SO4 dosage was chosen as 1.25:1.

Fig. 6
figure 6

Effect of (NH4)2SO4 dosage on leaching efficiency of the valuable metals

The roasted slag at different (NH4)2SO4 dosages was characterized by XRD, and the results are shown in Fig. 7. The diffraction peaks of MnSO4, (NH4)2Mn2(SO4)3, Fe2(SO4)3, and NH4Fe(SO4)2 were observed from the XRD patterns of the roasted slag when the (NH4)2SO4 dosages were 0.5:1 and 0.75:1. The stronger MnSO4 and Fe2(SO4)3 diffraction peaks may be formed due to the decomposition of small amounts of (NH4)2Mn2(SO4)3 and NH4Fe(SO4)2 generated under these conditions. When the (NH4)2SO4 dosage was increased to 1:1, the diffraction peaks of (NH4)2Mn2(SO4)3 and NH4Fe(SO4)2 increased significantly, while the diffraction peaks of Fe2(SO4)3 disappeared. When the amount of (NH4)2SO4 was increased to 1.25:1, the diffraction peaks of (NH4)2Mn2(SO4)3 and NH4Fe(SO4)2 further increased, while some disordered diffraction peaks basically disappeared. The diffraction peaks in the roasted slag at (NH4)2SO4 dosage of 1.5:1 or 2:1 were consistent with the results at 1.25:1, indicating that the original mineral phases of the ocean manganese nodules were completely destroyed at (NH4)2SO4 dosages of 1.25:1, and all the iron-manganese oxides had been converted into sulfates. This explains exactly why the leaching efficiency of the valuable metals remained stable after the dosage of (NH4)2SO4 was 1.25:1. Excess (NH4)2SO4 will be decomposed into many gases during the roasting process, which undoubtedly lead to the waste of resources.

Fig. 7
figure 7

XRD patterns of roasting slag at different (NH4)2SO4 dosages

Effect of Roasting Time

Based on the results in “Effect of Roasting Temperature” and “Effect of Ammonium Sulfate Dosage”, the effect of roasting time on the leaching efficiency of the valuable metals was studied at the roasting temperature of 380 ℃, and the mass ratio of (NH4)2SO4 to ocean manganese nodules of 1.25:1, and the results are presented in Fig. 8.

Fig. 8
figure 8

Effect of roasting time on leaching efficiency of the valuable metals

As shown in Fig. 8, the leaching efficiency of the valuable metals gradually increased with the prolongation of roasting time. The leaching efficiencies of Mn, Cu, Co, Ni, and Fe increased from 75.55%, 62.48%, 50.46%, 68.77%, and 3.97% to 88.58%, 81.26%, 74.78%, 87.19%, and 33.78%, respectively, when the roasting time increased from 10 to 70 min. After 70 min, the leaching efficiencies of Cu, Co, Ni, and Fe remained constant, while that of Mn was further increased with the increase of roasting time. When the roasting time was 90 min, the leaching efficiencies of Mn, Cu, Co, Ni, and Fe reached 96.04%, 83.61%, 75.03%, 87.24%, and 34.89%, respectively. After 90 min, the leaching efficiency of the valuable metals did not improve with increasing roasting time. Consequently, the optimal roasting time was determined to be 90 min.

Figure 9 displays the XRD patterns of the roasted slag at different roasting times. As shown in Fig. 9, although more diffraction peaks of (NH4)2Mn2(SO4)3 and NH4Fe(SO4)2 have appeared in the roasted slag at the roasting time of 10 min, some diffraction peaks of (NH4)2SO4, NH4HSO4, and (NH4)3H(SO4)2 were observed, indicating that (NH4)2SO4 did not completely react with the ocean manganese nodules. When the roasting time was extended to 30 min, the diffraction peaks of (NH4)2SO4 in the roasted slag were reduced. And when the roasting time was increased to 50 min, only a small amount of diffraction peaks of (NH4)2SO4 and (NH4)3H(SO4)2 were present in the roasted slag, whereas the diffraction peaks of (NH4)2Mn2(SO4)3 and NH4Fe(SO4)2 were significantly increased, and CaSO4·2H2O was converted to CaSO4. The diffraction peaks of (NH4)2SO4 and its intermediate products in the roasted slag completely disappeared at the roasting time of 70 min, indicating that the sulfation reaction was basically completed. Compared with 70 min, the diffraction peaks of MnSO4 in the roasted slag were stronger at the roasting time of 90 min, which might be the reason for the further improvement in the leaching efficiency of Mn. After 90 min, the XRD pattern of the roasting slag did not change. The changes in the phase composition of the roasted slag at different roasting times revealed the reason for the increase in the leaching efficiency of the valuable metals with increasing roasting time.

Fig. 9
figure 9

XRD patterns of roasting slag at different roasting times

Water Leaching

The roasted slag obtained at the roasting temperature of 380 ℃, the mass ratio of (NH4)2SO4 to ocean manganese nodules of 1.25:1, and the roasting time of 90 min was used as the research object for the water leaching process. The effects of leaching temperature and leaching time on the leaching efficiency of the valuable metals were investigated, and the results are presented in Fig. 10.

Fig. 10
figure 10

Effect of leaching temperature (a) and time (b) on leaching efficiency of the valuable metals

Figure 10a demonstrates the effect of roasting temperature on the leaching efficiency of the valuable metals at the roasting time of 60 min and the liquid-to-solid ratio of 5:1. As shown in Fig. 10a, the leaching efficiencies of Mn, Cu, Co, and Ni gradually increased with the increase of leaching temperature, while a decreasing trend in the leaching efficiency for Fe was observed. This is because the increase in leaching temperature favors the hydrolysis reaction of Fe in the solution. When the leaching temperature was increased from 20 to 40 ℃, the leaching efficiencies of Mn, Cu, Co, and Ni increased from 87.12%, 68.33%, 71.45%, and 83.09% to 96.12%, 83.57%, 75.08%, and 87.22%, while the leaching efficiency of Fe decreased from 43.09 to 35.03%. After 40 ℃, the leaching efficiencies of Mn, Cu, Co, and Ni remained stable, and the leaching efficiency of Fe decreased slightly. The main purpose of this study was to extract Mn, Cu, Co, and Ni from the ocean manganese nodules, so the best leaching temperature was chosen to be 40 ℃.

The effect of leaching time on the leaching efficiency of the valuable metals was researched at the leaching temperature of 40 ℃ and the liquid-to-solid ratio of 5:1, and the results are displayed in Fig. 10b. It can be seen from Fig. 10b that the valuable metals in the roasting slag can be quickly leached, and the leaching efficiencies of Mn, Cu, Co, and Ni have reached 91.21%, 74.29%, 72.47%, and 78.94% at a leaching time of only 10 min. This is due to the conversion of the valuable metals in the ocean manganese nodules into the corresponding sulfates by ammonium sulfate roasting, which can be well dissolved in pure water. When the leaching time was increased to 30 min, the leaching efficiencies of Mn, Cu, Co, Ni, and Fe reached 95.87%, 83.53%, 75.02%, 87.04%, and 35.27%, respectively. After 30 min, the leaching efficiency of the valuable metals no longer increased with the prolongation of leaching time. Hence, the optimal leaching time was determined to be 30 min.

The leached residue obtained under optimal leaching conditions was analyzed by XRD, and the results are shown in Fig. 11. As shown in Fig. 11, the main diffraction peak of the leached residue was SiO2, which is due to the fact that SiO2 is water-insoluble and was retained in the slag during the leaching process. In addition, silicates such as anorthite in the ocean manganese nodules were converted to sulfates and SiO2 during the roasting process. Although the leaching efficiency of Fe was only 35.27%, only a small amount of NH4Fe(SO4)2 and (NH4)2Fe2(SO4)3 diffraction peaks could be observed in the XRD patterns. This is because a large amount of Fe formed the Fe(OH)3 precipitates without crystal structure during the leaching process. The presence of only a small amount of diffraction peaks of (NH4)2Mn(SO4)2 in the leached residue indicates that a large amount of Mn has been dissolved into the solution, which was consistent with the experimental results.

Fig. 11
figure 11

XRD patterns of leaching residue

Mechanism Analysis of Ammonium Sulfate Roasting-Water Leaching Process

According to the above study, at 300 ℃, a portion of (NH4)2SO4 was decomposed into NH2SO3H, which was then decomposed into SO2, NH3, and H2O. At this time, SO2 can react with MnO2 in the ocean manganese nodules to convert it into MnSO4 (Eq. (9)), while CuO, CoO, and NiO can be converted into the corresponding sulfates, and the reaction process is as follows Eqs. (1719). However, (NH4)2SO4 cannot be completely decomposed at lower roasting temperatures. As the roasting temperature increased above 360 ℃, a violent sulfation reaction occurred between (NH4)2SO4 and iron-manganese oxides in the ocean manganese nodules, as shown in the following Eqs. (2022). When the temperature reached 380 ℃, the sulfation reaction was completed. From the reported literature[39, 40, 56], it is known that CuO, CoO, and NiO were converted to (NH4)2Cu(SO4)2, (NH4)2Co(SO4)2, and (NH4)2Ni(SO4)2 at this temperature, respectively. The reaction process is as follows Eqs. (2325).

$$ {\text{2CuO}}\left( {\text{s}} \right) \, + {\text{ 2SO}}_{{2}} \left( {\text{g}} \right) \, + {\text{ O}}_{{2}} \left( {\text{g}} \right) \, = {\text{ 2CuSO}}_{{4}} \left( {\text{s}} \right), $$
(17)
$$ {\text{2CoO}}\left( {\text{s}} \right) \, + {\text{ 2SO}}_{{2}} \left( {\text{g}} \right) \, + {\text{ O}}_{{2}} \left( {\text{g}} \right) \, = {\text{ 2CoSO}}_{{4}} \left( {\text{s}} \right), $$
(18)
$$ {\text{2NiO}}\left( {\text{s}} \right) \, + {\text{ 2SO}}_{{2}} \left( {\text{g}} \right) \, + {\text{ O}}_{{2}} \left( {\text{g}} \right) \, = {\text{ 2NiSO}}_{{4}} \left( {\text{s}} \right), $$
(19)
$$ {\text{2MnO}}_{{2}} \left( {\text{s}} \right) \, + \, \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) \, + {\text{ 2SO}}_{{2}} \left( {\text{g}} \right) \, = \, \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{Mn}}_{{2}} \left( {{\text{SO}}_{{4}} } \right)_{{3}} \left( {\text{s}} \right), $$
(20)
$$ {\text{2MnO}}_{{2}} \left( {\text{s}} \right) \, + { 3}\left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) \, = \, \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{Mn}}_{{2}} \left( {{\text{SO}}_{{4}} } \right)_{{3}} \left( {\text{s}} \right) \, + {\text{ 4NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ 2H}}_{{2}} {\text{O}}\left( {\text{g}} \right) \, + {\text{ O}}_{{2}} \left( {\text{g}} \right), $$
(21)
$$ {\text{Fe}}_{{2}} {\text{O}}_{{3}} \left( {\text{s}} \right) \, + { 4}\left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) \, = {\text{ 2NH}}_{{4}} {\text{Fe}}\left( {{\text{SO}}_{{4}} } \right)_{{2}} \left( {\text{s}} \right) \, + {\text{ 6NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ 3H}}_{{2}} {\text{O}}\left( {\text{g}} \right), $$
(22)
$$ {\text{CuO}}\left( {\text{s}} \right) \, + { 2}\left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) \, = \, \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{Cu}}\left( {{\text{SO}}_{{4}} } \right)_{{2}} \left( {\text{s}} \right) \, + {\text{ 2NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ H}}_{{2}} {\text{O}}\left( {\text{g}} \right), $$
(23)
$$ {\text{CoO}}\left( {\text{s}} \right) \, + { 2}\left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) \, = \, \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{Co}}\left( {{\text{SO}}_{{4}} } \right)_{{2}} \left( {\text{s}} \right) \, + {\text{ 2NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ H}}_{{2}} {\text{O}}\left( {\text{g}} \right), $$
(24)
$$ {\text{NiO}}\left( {\text{s}} \right) \, + { 2}\left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{SO}}_{{4}} \left( {\text{s}} \right) \, = \, \left( {{\text{NH}}_{{4}} } \right)_{{2}} {\text{Ni}}\left( {{\text{SO}}_{{4}} } \right)_{{2}} \left( {\text{s}} \right) \, + {\text{ 2NH}}_{{3}} \left( {\text{g}} \right) \, + {\text{ H}}_{{2}} {\text{O}}\left( {\text{g}} \right). $$
(25)

Figure 12 shows the mechanism diagram for extracting valuable metals from ocean manganese nodules by the ammonium sulfate roasting-water leaching process. Overall, under the roasting temperature of 380 ℃, the mass ratio of (NH4)2SO4 to ocean manganese nodules of 1.25:1, and the roasting time of 90 min, the iron-manganese oxides in the ocean manganese nodules were converted to (NH4)2Mn2(SO4)3, MnSO4, and NH4Fe(SO4)2, and the Cu, Co, and Ni encapsulated in the iron-manganese oxides were converted to (NH4)2Cu(SO4)2, (NH4)2Co(SO4)2, and (NH4)2Ni(SO4)2. As a result, water leaching can effectively extract valuable metals from the roasted slag. The gases released during roasting such as NH3, SO3, and H2O can be absorbed with dilute sulfuric acid to obtain ammonium sulfate solution, and then the ammonium sulfate products can be obtained by evaporation and crystallization. Ammonium sulfate products can be used in the roasting process, which realizes the recycling of ammonia while lowering production costs.

Fig. 12
figure 12

The mechanism diagram for extracting valuable metals from ocean manganese nodules by the ammonium sulfate roasting-water leaching process

Conclusions

A sustainable low-temperature roasting and water leaching process for simultaneously extracting Mn, Cu, Co, and Ni from ocean manganese nodules was proposed. Systematic research has confirmed that this method can effectively extract Mn, Cu, Co, and Ni from ocean manganese nodules. The experimental results showed that under the roasting temperature of 380 ℃, the mass ratio of (NH4)2SO4 to ocean manganese nodules of 1.25:1, and the roasting time of 90 min, the original mineral phases of the ocean manganese nodules were completely destroyed and the iron-manganese oxides in it were converted into (NH4)2Mn2(SO4)3, MnSO4, and NH4Fe(SO4)2, while the Cu, Co, and Ni wrapped in the iron-manganese oxides were converted into (NH4)2Cu(SO4)2, (NH4)2Co(SO4)2, and (NH4)2Ni(SO4)2.

Subsequently, the roasted slag was leached with pure water, and the leaching efficiencies of Mn, Cu, Co, and Ni reached 95.87%, 83.53%, 75.02%, and 87.04% at the leaching temperature of 40 ℃, the roasting time of 30 min, and the liquid–solid ratio of 5:1. The leaching efficiency of Fe was only 35.27%, which was due to the formation of a large amount of Fe(OH)3 precipitation during the leaching process. Due to the high value of Co, however, the leaching efficiency of the proposed process was mild, so the recovery of the remaining Co from the leaching residue should be investigated in the future. The waste gases (NH3 and SO3) released during roasting can be absorbed with dilute sulfuric acid to obtain ammonium sulfate crystals for recycling, which realizes the recycling of ammonia while lowering production costs. The roasting temperature of 380 ℃ is much lower than that of pyrometallurgical techniques proposed in other studies, resulting in significant energy savings for the process.