Heats of Immersion of Manganese Nodule, Manganese Dioxide, and Manganese-Iron Oxide YOSHIO KAWAI, MASAHIRO NITTA,* AND KAZUO AOMURA Department of Chemistry, Faculty of Engineering, Hokkaido University, Kita 13, Nishi 8, Kita-ku, Sapporo 060, Japan Received April 20, 1981; accepted August 10, 1981 The heats of immersion in water and thermal desorption spectra of manganese nodule, manganese dioxide, and manganese-iron mixed oxide were measured in order to clarify the surface properties of the nodule. The heat of immersion of manganese dioxide increased with increasing pretreatment temperature from 393 to 623 K and then decreased up to 723 K, accompanied by a structure change to sesquioxide; the heat of immersion of the nodule had a maximum value at 423 K. The results of TDS showed that chemisorbed water on the nodule was evolved at lower temperature and the surface was easily stabilized compared with manganese dioxide. For the mixed oxides, the heat of immersion had a minimum value at 26% of the iron content and the temperature at which water was evolved became higher with the iron content. On the basis of these results the surface of the nodule will be discussed.
heat of immersion of the two oxides in water (5). Many investigations have reported the structure, surface properties, and catalytic behaviors of manganese oxide (6). There are, however, few data on the heat of immersion of manganese oxide and manganese-iron mixed oxide. The purpose of this study is to clarify the surface properties of M n - N by measuring the heat of immersion in water, and comparing it with those of manganese oxide and manganese-iron oxide.
Manganese nodule (Mn-N) is expected to be utilized as a new metal resource in the near future because of its enormous deposits on the ocean floor (1, 2). The large specific surface area of M n - N and its composition containing various transition metal oxides have prompted us to use it as a catalyst and an adsorbent. We have previously shown that M n - N exhibits considerable catalytic and adsorptive activities toward several petrochemical reactions (3) and for the removal of heavy metal ions (4). However, due to the structural complexity of Mn-N, there have been few studies of its surface properties. One method of clarifying the surface properties of such a complex mixed oxide as M n - N is to obtain information about the main component. M n - N is composed mainly of manganese and iron oxides (1). Many workers have studied quantitatively and thermodynamically the surface properties of iron oxide and hydroxide by measuring the
Materials The Mn-N used in this study was collected from the Pacific Ocean floor near Hawaii. The crushed sample, excluding clay minerals which are impurities, was washed with distilled water and sieved to 20-42 mesh, then stored over saturated NH4C1 solution for at least 2 weeks to ensure a constant moisture content. The chemical composition of the Mn-N is shown in Table 1. The M n - N was evacuated at a given temperature under a pressure of 1.3 Pa for 3 hr before heat mea-
* TO whom queries concerning this paper should be sent. 47
Journal of Colloidand InterfaceScience,Vol. 88, No. 1, July 1982
0021-9797/82/070047-08502.00/0 Copyright © 1982 by AcademicPress, Inc. All rights of reproductionin any form reserved.
KAWAI, NITTA, AND AOMURA TABLE I Metal Content in M n - N Acid soluble part (wt%)
Acid insoluble part (wt%)
Mn Fe Ni Cu Oxygen, water and other metals
28.5 6.48 1.47 1.63
surement. MnO2 was obtained from a commercial reagent. It was crushed to the same size as that of M n - N sample. The manganese-iron mixed oxides were prepared by precipitation from solution of the metal chlorides and sodium hydroxide under conditions similar to that of sea water as described elsewhere (7). The precipitate was washed with distilled water and then dried at 383 K for 24 hr. The prepared powdered sample was pelletized by press and then crushed to the same size as that of M n - N sample. The mixed oxide was analyzed for manganese and iron by using an atomic absorption spectrometer. The MnO2 and mixed oxide were evacuated under the same conditions as those for M n - N before heat measurement except a pressure of 6.7 × 1 0 -3 Pa. Not the absolute value of heat of immersion at each evacuation temperature but the feature of the change of the heat with the temperature was compared between M n - N and MnO2 and mixed oxide, since the change of the heat of immersion with evacuation pressure was far less than that with evacuation temperature.
Method The calorimeter and the technique for the heat of immersion measurement have been described previously (8). All measurements were carried out at 313 ___0.03 K. The specific surface area of the samples was measured by nitrogen adsorption at 77 K using the BET method. Journal of Colloid and Interface Science, VoL 88, No. 1, July 1982
The X-ray analysis was performed with an X-ray diffractometer using Fe-K~ radiation, 30 kV and 10 mA. A S T M cards were used to determine the crystal form of the sample. Thermal desorption spectra ( T D S ) were measured by using a mass spectrometer. A sample was degassed at room temperature for 1 hr under a pressure of 6.7 × 10 -3 Pa and then heated in an electric furnace at a heating rate of 5 K / m i n by a programming controller. A mass spectrum of the gas evolved from the sample was obtained for each 10 K rise in temperature. Amounts of water and oxygen evolved were estimated directly from the intensities at mass number of 18 and 32, respectively, since amounts of other gases were negligible. RESULTS AND DISCUSSION
Heat of Immersion of Mn02 The heat of immersion in water and specific surface area of MnO2 as a function of pretreatment temperature are shown in Fig. 1. The heat of immersion increases with temperature up to 623 K then decreases above 673 K. In the region up to 623 K, the active sites for adsorption are created by dehydration o f preadsorbed water and surface hy-
580 IZO 560 'E
:~ 480 70 460
Pretreatment Temperature /K
FIG. h Heat of immersion in water (©) and specific surface area ( t ) of Mn02.
HEATS OF I M M E R S I O N OF M n - F e OXIDE TABLE II Change of X-Ray Diffraction Pattern of MnO2 Pretreatment
1.54 W 2.03 W
droxyl groups. These sites are readily rehydrated with water in the immersion process. The large heat due to the rehydration may be most of the overall heat effect observed. When the pretreatment temperature increases, the dehydration of MnO2 is accelerated and the heat of rehydration in the immersion process increases accordingly. However, in the region above 673 K, the heat of immersion and specific surface area decrease, probably due to the fact that the structure changes from MnOz to Mn203. The change of structure of MnO2 and the behavior of water were confirmed by the resuits of X-ray analysis and TDS measurement. The change of X-ray diffraction pattern of MnO2 pretreated at each temperature is shown in Table II. In the region of pretreatment temperature from 393 to 623 K, /3form of MnO2 does not change. A change of the structure occurs at 673 and at 773 K, it is clearly observed that MnO: changes to more crystallized 7-Mn203. The TDS of MnO2 is shown in Fig. 2. The peaks of water at 363 and 513 K may be attributed to phys-
ically and chemically adsorbed water, respectively. It is suggested that the desorption of physically adsorbed water at 363 K gives only a little influence on the heat of immersion, whereas that of chemically adsorbed water at 513 K considerably activates the surface of MnO2. The activated surface
FIG. 2. Thermal desorption spectra of MnO2 and manganese nodule: O, H20, and 0, 02 from MnO2; A, H20, and A, 02 from manganese nodule. Journal of Colloid and Interface Science, Vol. 88, No. 1, July 1982
KAWAI, NITTA, AND AOMURA
would be easily rehydrated with water molecules in the immersion process and give a large amount of heat of rehydration. Consequently, a large heat of immersion is observed. The evolution of oxygen above 593 K is due to the transformation of MnO2 to Mn203 as confirmed by the X-ray analysis. Many researchers have reported that the heat of immersion of various metal oxides such as Fe203, SiO2, A1203, TiO2, and ZnO in water has a maximum value at a certain pretreatment temperature (5(d), 9). They maintain that the increase of the heat of immersion is due to the increase of heat of rehydration of the surface from which the hydroxyl groups had been removed and that the decrease of it is due to the formation of a more stable surface by the pretreatment at higher temperatures. The stabilized surface of these metal oxides has been formed without any change of the structure. However, an interesting phenomenon was observed in the present case, that is, the decrease in heat of immersion of MnO2 in water is accompanied by the change of the structure.
Heat of Immersion of Manganese Nodule The heat of immersion in water and specific surface area of M n - N with pretreatment temperature are shown in Fig. 3. The I000 180 9OO
I 500 Pretreatment
FIG. 3. Heat of immersion in water (O) and specific surface area (O) of manganese nodule. Journal of Colloid and Interface Science, Vol. 88, No. 1, July 1982
specific surface area was practically constant in the range of the pretreatment temperature between 400 and 600 K and it increases between 600 and 700 K. The increase of the surface area might be due to the structural change, though there is no evidence in the X-ray diffraction result. The change in heat of immersion of M n - N can be explained in the same way as the case of metal oxides reported (9). The increase of the heat of immersion results from the activation of the surface, i.e., the formation of highly active sites and/or the increase in number of sites by the dehydration, and the decrease of the heat from the formation of a more stable surface as described above. Figure 3 shows that there is no relationshi p between the heat of immersion and the specific surface area. The heat of immersion of M n - N begins to decrease at a lower temperature (423 K) than those (570-770 K) of other metal oxides reported previously (9). This implies that the surface of M n - N is stabilized at a lower temperature. This is supported by the fact that the adsorbed water and surface hydroxyl groups on M n - N are removed at lower temperatures, as is evident from the result of TDS of M n - N shown in Fig. 2. Only a peak for the evolution of water is observed in the TDS of M n - N heat-treated in the range of temperature (_<600 K) at which two peaks are observed in the case of MnO2. The water evolution peaks of MnO2 appear at 373 and 523 K. The former is due to the physisorbed water and the latter due to the chemisorbed one and/or the surface hydroxyl groups. Below 600 K a broad peak of M n - N suggests that the sorbed water on M n - N is strongly physisorbed and/or weakly chemisorbed, that is, that there is not appreciable difference in bond strength between the two types of sorbed water. On the other hand, above 600 K both MnN and MnO2 have another water-evolution peak, respectively, due to the desorption of surface hydroxyl groups which are the most strongly bonded to the surface (isolated hydroxyls). In the studies of infrared spectros-
HEATS OF IMMERSION OF Mn-Fe OXIDE copy of the desorbability of sorbed water on metal oxides (14, 15), it has been confirmed that there are two kinds of surface hydroxyls on the oxide surface; one is isolated hydroxyls which survive after evacuation of the oxide at 723 K and the other is hydrogenbonded hydroxyls which are removed by evacuation at 573 K. It has been also revealed that rehydration shows a strong interaction of the evacuated surface with water molecules and reforms hydrogen-bonded hydroxyls, but only a small amount of isolated hydroxyls weakly interact with adsorbate molecules such as water and ammonia. In the present study, therefore, it could be said that water-evolution peak at >600 K has not so significance to the heat effect. A small amount of oxygen is evolved from M n - N above 533 K as shown in Fig. 2. In the case of MnO2, the evolution of oxygen occurs when MnO2 changes to Mn203, and this structure change accompanies with the decrease in heat of immersion. However, the heat of immersion of M n - N is not changed at temperature where oxygen is evolved. The X-ray diffraction pattern of M n - N showed only weak peaks but no changes of pattern in the whole region of experimental pretreatment temperature (<723 K).
to 1073 K and found that the heat maximizes at about 870 K (5(d)). From their results of iron oxide and the present result for MnO2 we cannot explain the variation of the heat of immersion of M n - N in water. It is reasonable to suggest that from the surface property of each single metal oxide it is difficult to deduce that of a mixed oxide such as Mn-N. It is, however, worthwhile to compare the surface property of the mixed oxide of manganese and iron which are the main components of M n - N with that of Mn-N. The specific surface area of the mixed oxide pretreated at 573 K with various atomic ratios of Fe/ (Fe + Mn) is shown in Fig. 4. As a matter of convenience, the sample of which iron content, F e / ( F e + Mn), are 0, 0.26, 0.45, 0.63, 0.81, and 1.0 are hereafter referred to as Fe0, Fe20, Fe40, Fe60, Fe80, and Fe 100, respectively. The specific surface area attains a maximum value at Fe40. The reason why the mixed oxide shows larger specific surface area than each of single metal oxides is considered as follows. The particles of iron and manganese hydroxides in aqueous solution are positively (10) and negatively (6(c)) 660
Heat of Immersion of Manganese-Iron Oxide M n - N mainly consists of manganese and iron oxides as shown in Table I. Many researchers have already reported the heat of immersion of iron oxide in water. Ishikawa et al. observed that the heat of immersion of FeO(OH) in water results in a minimum value between the pretreatment temperatures of 470 and 570 K (5(b)). They found also that the heat of immersion increases above 570 K because of the formation and hydration of an a-Fe203 surface. Zettlemoyer et aL reported that the heat of immersion of a-Fe203 in water increases up to 673 K (5(c)). Morimoto et al. measured the heat of immersion of oz-Fe203 pretreated up
V 6zo 600
60 ~ 560
:= 50 540
F e / ( F e + Hn)
FIG. 4. Heat of immersion in water (O) and specific surface area (O) of manganese-iron oxide. Journal o f Colloid and Interface Science. Vol. 88, No. 1, July 1982
KAWAI, NITTA, AND AOMURA
charged, respectively. When iron or manganese hydroxide particles exist alone, an electrostatic repulsion occurs between the particles. In this case, the particles do not aggregate with each other but grow individually. The surface area of the precipitate thus formed is small. On the contrary, when both positive particles of iron hydroxide and negative particles of manganese hydroxide exist in a solution, the electrostatic attrac1.0
T e m p e r a t u r e /K FIG. 5. T h e r m a l d e s o r p t i o n s p e c t r a of m a n g a n e s e iron oxide: O, H 2 0 ; a n d O, 02. Journal of Colloid and Interface Science, Vol. 88, No. l, July 1982
tion which occurs between their both particles aggregates them. The neutralization of charge by the aggregation accelerates the aggregation more and more. Therefore, the precipitate produced in such a solution has large spaces in it, so that the specific surface area becomes larger. The reason why the surface area remains constant for Fe60 and Fe80 is not clear. The heat of immersion of the mixed oxide in water is shown in Fig. 4. The heat of immersion of Fe0 is larger that that of Fe20. The TDS of Fe0 shown in Fig. 5 indicates that the surface hydroxyls of Fe0 remain even after heating 573 K, the temperature at which the samples are pretreated before measurement of their heat of immersion. In this case, hydrogen bonding between the hydroxyl groups on the Fe0 surface and water molecules greatly contribute to the observed heat of immersion of Fe0 in H20. For the mixed oxides in the range 0.2-0.8, the heat of immersion increases with increasing iron content as shown in Fig. 4. This implies that the surface of the mixed oxides becomes more activated toward rehydration as the iron content increases. Similarly, the TDS of the mixed oxides indicates that the temperature at which water is evolved increases as the iron content increases, as shown in Fig. 5. The atomic ratio of M n - N used here, Fe/ (Fe + Mn), is estimated to be 0.19 from the composition shown in Table I and this value is near that of Fe20 mixed oxide sample. As described above, chemisorbed water on M n N surface is easily desorbed and the surface is stabilized at a lower temperature than that of manganese dioxide. In the case of the mixed oxide, it is shown in Fig. 5 that the surface of Fe20 is more stable than those of the other mixed oxide samples, since water adsorbed is evolved at the lowest temperature among those of the others. These results show that the surface of M n - N is somewhat characterized by synthetic manganese-iron oxide which has the same atomic ratio of iron to manganese as that of Mn-N. It
HEATS OF IMMERSION OF Mn-Fe OXIDE
should be also considered that the surface properties of M n - N are not characterized sufficiently by the Mn-Fe mixed oxide. It is suggested that there is a significant contribution of minor elements of M n - N to the surface properties. We have reported that minor elements such as Cu, Co, and Ni of M n - N contributed to increase the surface area (13). It is very difficult to know clearly their effects on the heat of immersion and TDS of water. Because the effects become multiplicate due to the large number of minor elements. The X-ray analysis of the mixed oxides has revealed that the crystal forms of Fe0 and Fel00 are y-Mn203 and/or Mn304 and a-Fe203, respectively, although the mixed oxides are amorphous (7). Hase et al. investigated the structure of (Fe, Mn)203 and showed that the octahedra of metal ion of (Fe, Mn)203 becomes more distorted tetragonally when the iron content decreases (11). Though their preparation method of the sample is very different from that of the present study, the present result of the heat of immersion of the mixed oxides is well interpreted by their conclusion, that is, the surface of the mixed oxide becomes stable when the octahedra of the metal ion are distorted tetragonally. This deduction is also supported by the fact that the surface of MnO2 becomes stable when MnO2 changes into 3'-Mn203. Also, the structure of MnO2 is rutile type where the manganese ion coordinates octahedrally, while in case of 3'MnzO3 the manganese ion has a tetragonally distorted structure (12). Thus it could be concluded that a distorted structure of coordination of manganese ion contributes to the formation of more stable surfaces of manganese and manganese-iron oxides. The heat of immersion of Fel00 is smaller than those of other mixed samples. The Xray analysis revealed that Fel00 is crystallized as o~-Fe203, while the mixed oxides are amorphous. The TDS of Fel00 showed that Fel00 pretreated at 573 K does not possess surface hydroxyls which interact with water
molecules. Therefore the surface of Fel00 is stable resulting in a small heat of immersion. From the facts described above, we may conclude that the surface properties of MnN are in part characterized by manganeseiron mixed oxide, that is, the results of the heat of immersion and TDS of them showed that the surface heterogeneity of the nodule concerned to water adsorption took an intermediate behavior between manganese and iron oxides. It was also suggested that the role of minor metal components such as Co, Cu, and Ni of M n - N in its surface properties could not be neglected. ACKNOWLEDGMENT The authors' thanks are due Dr. K.-I. Tanaka of Hokkaido University, Japan for his help in measurement of thermal desorption spectra. REFERENCES 1. Glasby, G. P., "Marine Manganese Deposits," Elsevier, New York, 1977. 2. Hubred, G. L., Marine Mining 2, 475 (1980). 3. (a) Weisz, P. B., J. Catal. 10, 407 (1968). (b) Wu, S. C., and Chu, C., Atmos. Environ. 6, 309 (1972). (c) Matsuo, K., Nitta, M., and Aomura, K., J. Japan Petrol. lnst. 18, 697 (1975). (d) Nitta, M., Matsuo, K., and Aomura, K., Chem. Lett. 1977, 325 (1977). 4. Nitta, M., Kurokawa, Y., and Aomura, K., J. Japan Petrol. Inst. 24, 219 (1981). 5. (a) Healey, F. H., Chessick, J. J., and Fraioli, A. V., J. Phys. Chem. 60, 1001 (1956). (b) Ishikawa, T., and Inouye, K., Bull. Chem. Soc. Japan 46, 2665 (1973). (c) Zettlemoyer, A. C., and McCafferty, E., Z. Phys. Chem. 64, 41 (1969). (d) Morimoto, T., Katayama, N., Naono, H., and Nagao, M., Bull. Chem. Soc. Japan 42, 1490 (1969). 6. (a) Moore, T. E., Ellis, M., and Selwood, P. W., J. Amer. Chem. Soc. 72, 856 (1950). (b) Kobayashi, M., and Kobayashi, H., J. Catal. 27, 100 (1972). (c) Healy, T. W., Herring, A. P., and Fuerstenau, D. W., J. Colloid Interface Sci. 21, 435 (1972). 7. Matsuo, K., Nitta, M., Aomura, K., Shokubai 19, 287 (1977). 8. Nitta, M., Kawai, Y., and Aomura, K., J. Japan Petrol. lnst. 23, 172 (1980). Journal of Colloid and Interface Science, Vol. 88, No. 1, July' 1982
KAWAI, NITTA, AND AOMURA
9. (a) Young, G. J., and Bursh, T. P., J. Colloid Interface Sci. 15, 361 (1960). (b) Morimoto, T., Shiomi, K., and Tanaka, H., Bull Chem. Soc. Japan 37, 392 (1964). (c) Morimoto, T., Nagao, M., and Hirata, M., Kolloid-Z. Z. Poly. 225, 29 (1968). (d) Wade, W. H., and Hackerman, N., J. Phys. Chem. 65, 1681 (1961). 10. Parks, G. A., and de Bruyn, P. L., J. Phys. Chem. 66, 967 (1962).
Journalof Colloidand InterfaceScience, Vol.88, No. 1, July 1982
11. Hase, W., Kleinsttick, K., and Schulze, Z. Kristallogr. 124, 428 (1967). 12. Wells, A. F., "Structural Inorganic Chemistry," 4th ed., p. 458. Oxford Univ. Press, London/New York, 1975. 13. Matsuo, K., Nitta, M., and Aomura, K., J. Catal. 54, 446 (1978). 14. Morimoto, T., Yanai, H., and Nagao, M., J. Phys. Chem. 80, 471 (1976). 15. Lewis, K. E., and Parfitt, G. P., Trans. Faraday Soc. 62, 204 (1965).