发表论文-穆斯堡尔谱研究组.pdf

N0. 2006 – 2 ISSN 1880 – 327X 155 Gd, 166Er and 237Np Mössbauer Spectroscopic Studies on the Structure and Chemical Bonding in Lanthanide and Actinide Compounds Junhu Wang March 1, 2007 Abstract Making clear the structure and chemical bonding feature in f-block element compounds is not only important for the basic areas but also for the applied areas of lanthanide (Ln) and actinide (An) science. Mössbauer spectroscopy (MBS) – the nuclear gamma resonant spectroscopy – is a powerful tool for investigating the structure and chemical bonding feature in various materials. The electronic state of target atom is available to be directly reflected in its Mössbauer (MB) spectrum. The MB effect has been confirmed to 14 elements to Ln and 6 elements to An, however, only 151Eu MBS is comparatively applied up to now. Though valuable information is also available to be obtained from MBS of the other f-block elements, their MB spectrum measurements are more difficult because of the preparation of MB source by oneself, keeping the radiation source and absorber (sample) at near helium temperature. On one hand, we selected three such MB elements and performed a systematic investigation on the structure and chemical bonding in various materials containing Gd or Er element by using 155 Gd and 166Er MBS in connection with powder and/or single-crystal X-ray diffraction method. On the other hand, in relation to the nuclear waste management, 237Np MB spectroscopic studies on the structure and chemical bonding in various materials containing radioactive 237Np element were also conducted. Based on these investigations, differences on the structure and chemical bonding feature in f-block element compounds were discussed and much precious results were deduced. In this technical report, the research process and some results are reviewed. Keywords: f-block element; 155Gd, 166Er and 237Np Mössbauer spectroscopy; lanthanide compound; actinide compound; crystal structure; chemical bonding; 1 155 Gd, 166Erおよび237Npメスバウアー分光法によるガドリニウム、 エルビウムおよびネプツニウム化合物の配位構造と化学結合の研究 王 軍虎 中京大学生命システム工学部 （２００７年４月から中国科学院大連化学物理研究所） ２００７年３月１日 要旨 ｆ－ブロック元素化合物の化学結合および配位構造を解明することはランタノイドおよ びアクチノイド科学の基礎分野だけでなく応用面からも重要である。メスバウアー分光法は 原子核のgamma線に対する無反跳共鳴吸収現象を利用しており、メスバウアー核種の電 子状態についての知見を得ることができ、化学結合、配位構造および各種触媒機能材料 のIn-situ状態等に関する多くの情報を得ることができる。 ｆ ―ブロック元素のメスバウアー 効果についてランタノイドは 14 元素、アクチノイドは 6 元素で観測されるが、151Euのみ多く の研究が行われている。しかしながら測定には線源を自作しなければならなかったり、液体 ヘリウム温度近くに冷却しなければならなかったりと困難は伴うものの、有用な化学的な情 報が期待できる元素が多くある。本研究ではそのようなメスバウアー元素に注目し、155Gd, 166 Erおよび237Npメスバウアー分光法を用いて種々のガドリニウム、エルビウムおよびネプ ツニウム化合物を研究し、f－ブロック元素化合物の化学結合および構造に関してたくさん 貴重な知見を得た。ここで本研究の過程およびいくつかの研究結果をまとめて紹介する。 キーワード：ｆ－ブロック元素； 155Gd, 166Erおよび237Npメスバウアー分光法； ランタノイド 化合物； アクチノイド化合物； 配位構造； 化学結合 2 155 166 237 从 Gd, Er及 Np穆谱中获得的镧系和锕系元素化合物的结构和成键信息 王军虎 日本中京大学生命系统工学部 (从 2007 年 4 月开始: 中国科学院大连化学物理研究所) 2007 年 3 月 1 日 摘要 不论从基础科学还是从实际应用的角度来看,研究和解明镧系和锕系元素化合物材 料的结构和化学成键的机理都有很重要的意义。穆谱学是通过观测原子核对γ线的共鸣 吸收现象而研究核外电子举动的科学，在许多化学领域, 如物质结构，化学成键，催化 反应及催化剂功能材料的原位状态分析等方面已有广泛地应用。自 1958 年穆斯堡尔效 应发现以来, 镧系元素有 14 种，锕系有 6 种它们的穆谱观测法已被确立及报导，可目 151 前仅有 Eu穆谱学在物质材料研究中有较广泛地应用。造成这种状况的原因主要被认为 151 57 119 是镧系和锕系的其它的穆斯堡尔元素的谱图观测与 Eu及最常见的 Fe和 Sn相比有一 定的难度，如放射源必须自己制作，必须在接近液氦的低温下方可测量等。可是, 从这 样的元素的穆谱上同样可得到许多用其它谱学法难以得到的独特信息。作者着眼于这类 155 166 穆斯堡尔元素，制作了优质的穆谱单线放射源,在世界上首次系统开展了 Gd和 Er穆谱 学在各种钆和铒配位络合物材料的结构和化学成键中的应用研究。同时, 与日本原子力 237 研究所合作, 从放射性废物处理及无机结构化学的观点出发, 用 Np穆谱学结合差热分 析, 磁性测量及X衍射表征等对Np(VI)配位络合物也进行了系统研究。根据以上研究结 果, 作者系统对比考察了镧系和锕系元素化合物材料的结构异同及 4f和 5f轨道在化学 成键中所起的不同作用。本技术报告将对上述的研究过程及其代表性的一些研究结果做 一概述。 关键词: 155 166 237 Gd, Er和 Np穆谱学；镧系元素化合物；锕系元素化合物；配为结构；化学 成键机理；4f和 5f轨道 3 155 Gd, 166Er and 237Np Mössbauer Spectroscopic Studies on the Structure and Chemical Bonding in Lanthanide and Actinide Compounds Junhu Wang School of Life System Science and Technology, Chukyo University, 101 Tokodachi, Kaizu, Toyota, 470-0393 Japan (From April, 2007: Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Dalian, 116023 China) 1. Introduction Materials containing lanthanide (Ln) and/or actinide (An) elements have been widely applied in advanced functional material, nuclear fuel and nuclear waste management, due to their unique physico-chemical properties. However, the roles of 4f and 5f orbits in chemical bonding are still unclear. Though electronic structure calculation methods are used to make clear the roles of 4f and 5f orbits in chemical bonding and some valuable results have been reported so far, the evidences directly observed by experiment are still not many. Therefore, making clear the coordination structure and chemical bonding feature in the f-block element compounds is not only important in the basic areas but also in the applied areas of Ln and An science. On one hand, the 4f orbit of Ln is not traditionally considered to be a participant in chemical bonding. Recent years, mainly based on the electronic structure calculations, participation of 4f and 5d orbits in covalent bonding are evolved [1,2]. On this viewpoint, more detail investigation is needed, especially sufficient evidence directly obtained from an experimental basis. On the other hand, the 5f, 6s, 6p and 6d orbits of An are considered to participant in chemical bonding. This leads to (a) a bigger range of oxidation states than with Ln; (b) a greater tendency to covalent formation (but maybe involving 6d rather than 5f) in ions like AnO2+ and AnO22+ (most notably the uranyl ion, UO22+ and the neptunyl ion, NpO2+) [1]. The role of each orbit involved in chemical bonding of An needs to be investigated furthermore. A comparative study of Ln and An is a good way for understanding their coordination structure, chemical bonding feature, and fundamental physico-chemical property, and even for developing novel functional materials as luminous material, nuclear material. Mössbauer (hereafter called MB) effect refers to the resonant and recoil-free emission & absorption of Tgamma rays by atomic nuclei bound in a solid form [3]. Nearly fifty years ago, whilst working on his doctoral thesis under Professor Maier-Leibnitz in Heidelberg, Rudolf L. Mössbauer made this important discovery. Since the states of electrons around a MB atomic nuclei are available to be directly reflected in the hyperfine structure of its MB spectrum, Mössbauer spectroscopy (here after called MBS) or the nuclear gamma resonant spectroscopy has become a powerful tool to investigate the coordination structure and chemical bonding feature in various functional materials. The MB effect has been confirmed to have about 100 nuclear transitions in some 80 nuclides in nearly fifty elements. There are 14 elements to Ln and 6 elements to An that the MB effect is available to be observed as shown in Figure 1, however, only 151Eu MBS is comparatively applied. Though valuable information can be also obtained from MBS of the other f-block elements, their MB spectrum 4 measurements are more difficult because of MB source preparation, keeping the source and absorber (sample) at near helium low temperature. 1 H 18 He 2 13 14 15 16 17 Li Be B C N O F Ne Na Mg Al Si P S Cl Ar K 3 4 5 6 7 8 9 10 11 12 Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe Cs Ba * Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn Fr Ra ** Rf Db Sg Bh Hs Mt U un U uu U ub * Ln La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu ** An Ac Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr Figure 1. Mössbauer elements. (The elements observed Mössbauer effect so far are shown in pink) We selected such three MB elements and performed a systematic investigation on various functional materials containing Gd and Er by using 155Gd and 166Er MBS in connection with powder and/or single-crystal X-ray diffraction (XRD) method. At the same time, in relation with nuclear waste management, 237Np MB spectroscopic studies on various materials containing Np were also conducted. Based on a large quantity of experimental results, the differences on the coordination structure and chemical bonding feature in a large number of Ln and An compounds were studied and much precious information were deduced [4]. In this technical report, the research process and some results are reviewed. 2. 155Gd, 166Er and 237Np MBS [5-7] Generally, three principal hyperfine interactions are available to be observed from MBS. They are monopolar electric interaction, quadrupolar electric interaction and dipolar magnetic interaction. The first one derives from the coupling between the charge distribution of the protons in the nucleus and that of the electrons penetrating the nuclear volume. It leads to the isomer shift, δ. The electronic charge density inside the nuclear volume results from the direct contribution of the s-electrons; the p, d or f electrons interact indirectly via a screening effect. The isomer shift provides information on the charge state, the bonding nature, the number and distance of the nearest neighbors, the number of sites containing the resonant atom and so on. The quadrupole electric interaction is produced by the coupling between the quadrupole moment Q of the nucleus and the non-spherical distribution of the electrical charges (ionic charge on the lattice and contribution of the valence electrons) which induces an electrical field gradient (EFG) at the site of the resonant nucleus. The quadrupole splitting, e2qQ, provides information on the symmetry of the coordination polyhedron and the valence 5 electron contribution. The dipolar magnetic interaction has its origin in the coupling of a magnetic dipole moment, µ, of a spin I nuclear level, with the effective magnetic field, Heff, at the nuclear site. Relaxation phenomena produced by dynamic interactions (spin fluctuations) lead to more or less well resolved spectra, according to the relaxation time, τ. When the relaxation time is very large, if compared with the Larmor precession period, a static hyperfine magnetic field is observed. In a more detailed discussion one would have to distinguish relaxation processes in a paramagnet or in an ordered magnet. Extensive experimental work has been reported previously and an excellent review has given by Czjzek [5] about 155Gd MBS, which gives information on δ and e2qQ. No relaxation phenomena have been discussed up to now on 155Gd MBS. 237Np MBS, which has been described in numerous articles and reviews, gives information on δ, e2qQ, Heff and τ. Among of them, research work is not so many to be reported up to now on 166Er MBS, which gives information on Heff, e2qQ and τ and the e2qQ value is available to be obtained only in the case of the existence of the effective magnetic field, Heff. MB study on coordination compound is mainly on Np. Before our research is reported, no any work has been done on Gd and Er. About experimental techniques, points in common, are that 155Gd, 166Er and 237Np MBS are restricted to low temperature because of the high energy of the gamma rays. That is to say, the radiation source and absorber must be in a cryostat. Furthermore, fine radiation sources for 155Gd and 166Er MBS are only available to be made for oneself. The additional description is separately given as below on the three MBS in short. A dilute solution of 155Eu diffused into Pd or the compound 154SmPd3 after neutron irradiation is a fine single-line source for 155Gd MBS. Three gamma transitions (60.0, 86.5, 105.3 keV) can be employed for 155Gd MBS. At present, only the 86.5 keV transition is employed extensively since its resulting MB spectrum has the best resolution. Since the change of the mean-square nuclear radius ∆ is negative for the 86.5 keV transition of 155 Gd, the small δ the large s-electron density at the resonant nucleus position is. The large range of δ values has been observed in the metallic Gd compounds from about -0.2 to nearly 0.9 mm s-1. Since the Q of the 86.5 keV state (I = 5/2) is very smaller than that of the ground state (I = 3/2), only the ground state splitting is resolved, resulting that the spectra have the appearance of doublets. When 155Gd nuclei are in a site of cubic point symmetry and experience a magnetic field, the MB spectrum is split into 12 lines due to the magnetic hyperfine interaction. There are three groups of 4 lines each since the two g factors are nearly equal for the ground and 86.5 keV states. The compound Ho0.4Y0.6H2 after neutron irradiation is a useful single-line source for 166 Er MBS and gamma transition of 80.6 keV is employed. Since the change of the mean-square nuclear radius ∆ is very small for the 80.6 keV transition of 166Er, the δ is not available to be observed. Because the 80.6 keV transition in 166Er is E2 transition, the ground state (I = 0) is not split, the 80.6 keV state is split into five equally spaced levels when the magnetic hyperfine interaction is existed, and the five equally spaced levels are further shifted when the quadrupole hyperfine interaction is existed. Metallic 241Am radioactive isotope is used as the source for 237Np MBS and gamma transition of 59.6 keV is employed. The special feature of 237Np MBS is that the oxidation 6 number of Np can be sensitively reflected from the δ value. For the 59.6 keV transition of 237 Np, the sign of the mean-square nuclear radius ∆ is negative and it is the same with that of the 86.5 keV transition of 155Gd, but the change is very larger. In the case of pure quadrupole hyperfine interaction, the spectrum shows five lines and is symmetric with respect to the central line when the electric field gradient is axial symmetry, it shows three lines when the asymmetry parameter, η is equal to 1. The spectrum, which is symmetrical for pure magnetic splitting, has 16 lines. In the case of non-collinear magnetic and quadrupole hyperfine interactions, the spectrum becomes more complex. 3. Preparation of MB Sources 155EuPd3 and 166Ho0.4Y0.6H2 3.1. 155EuPd3 Source for 155Gd MBS [8] In this study, firstly a fine single-line 155EuPd3 source (about 231 MBq) for 155Gd MBS was prepared by chemical synthesis and neutron irradiation of the compound 154SmPd3. 154 SmPd3 was only prepared by the conventional solid state reaction in a H2 atmosphere at high temperature. According to the previous report, the product was mixed hydrides after the mixture of 154Sm(HCOO)3 and PdHx was calcined at high temperature in a H2 atmosphere. However, the product obtained in this study was confirmed as 154SmPd3, not as 154SmPd3Hx or 154 SmHx and PdHx, by the channeling method through the nuclear reaction 1H(11B, α)αα. In addition, Pd fine particles used to synthesize starting material PdHx were 100 prepared by a chemical solution process (a) in order to increase the relative reaction 99 Secondly, the prepared 154SmPd3 (about 312.4 mg) pellet was wrapped with Al sheet with 99.99% purity and irradiated by JRR-3M-HR-1 reactor (the flux of neutrons: 6.0 x 1013 cm-2 s-1) for 67 h in Japan Atomic Energy Research Institute (JAERI, presently the Japan Atomic Energy Agency). After irradiation, the sample was left in JAERI for one month in order to wait for the Pd activity to die. Scheme of the nuclear reaction was shown as the follows: T (%) area. 98 100 99 (b) 98 97 -10 -5 0 -1 v / mm s 5 10 Figure 2. 155Gd Mössbauer spectra of GdPd3 (a) and cubic Gd2O3 (b) at 12 K obtained in this study. 154 Sm(n, γ)155Sm(β-, 22 min)155Eu(β-, 4.96 y)155Gd Finally, the prepared 155EuPd3 source was mounted to an Al holder and again wrapped by Al sheet with 99.99% purity. Araldite adhesive was used in order to mount the source on the holder tightly. Its high performance was proved by observing the 155Gd MB spectra of 7 known compounds, GdPd3 and cubic Gd2O3 as shown in Figure 2. The GdPd3 is iso-structural with the source compound, 154SmPd3 and its 155Gd MB spectra were known to show single-line pattern. The cubic Gd2O3 have two crystallographically inequivalent Gd3+ sites and the population is 3:1, which can be identified clearly by 155Gd MBS. The obtained results in this study indicate that the newly prepared source is single-line and fine enough to investigate the structural characteristic of materials containing Gd. 3.2. 166Ho0.4Y0.6H2 Source for 166Er MBS [9] The preparation of the 166Ho0.4Y0.6H2 MB source was according to the following procedure: the alloy of Ho0.4Y0.6 (about 200 mg) was wrapped in a Ti sheet which acted as oxygen getter and moreover wrapped in a Ta sheet, and then was put into a quartz tube. The quartz tube was connected to a vacuum system containing a manometer and put it into an electronic furnace. After evacuating, the quartz tube was heated to 1273 K for 2 h and then down to 1123 K. Dried hydrogen gas was through the quartz tube by a pressure of 200 Torr. The Ho0.4Y0.6 alloy was almost immediately reacted with hydrogen at 1123 K. In order to avoid the formation of trihydrate Ho0.4Y0.6H3, the quartz tube was cooled rapidly from 1123 K to room temperature by liquid nitrogen. The prepared Ho0.4Y0.6H2 dihydrate compound was checked by XRD that revealed it had a cubic CaF2-type structure. line-width of 2Γ = 7.7 mm s-1 at 4.2 K). As shown in Figure 3, the spectrum of ErH2 obtained in this study is single-line pattern and has a line-width of 2Γ = 8.0 mm s-1 at 12 K. It indicates the prepared 166 Ho0.4Y0.6H2 source is fine enough. Since the half life of 166Ho is short as 26.9 h, a 166 Ho0.4Y0.6H2 source is only available to be used about 7~10 days for measuring 3~4 samples. T (%) The prepared Ho0.4Y0.6H2 (about 55 mg) was pelletized into a disc (φ10 mm). The disc was wrapped by Al sheets and then irradiated at JAERI with JRR-3M PN-1 (neutron flux of 6.0 x 1013 cm-2s-1) for 9 min. The obtained 166 100 Ho0.4Y0.6H2 (about 1.5 GBq) MB source was evaluated by a standard absorber, ErH2. ErH2 was iso-structural with the 98 source compound, Ho0.4Y0.6H2 and its 166 Er MB spectrum was known to show a 96 narrower single-line pattern (the 94 92 -100 -50 0 v / mms 50 100 -1 Figure 3. 166Er Mössbauer spectrum of ErH2 at 12 K obtained in this study. 4. 155Gd, 166Er and 237Np MB Measurements [8-10] The 155Gd and 166Er MB spectra were measured by using the prepared 155EuPd3 and 166 Ho0.4Y0.6H2 sources on a WissEl MB measuring system consisting of MDU-1200, DFG-1200 and MVT-1000, respectively. The 237Np MB spectra were measured on the same 8 model MB system at JAERI by using an assembled metallic 241Am source (about 600 MBq) purchased from Russian. The following points were in common for the three MB measurements. Both of the radiation source and samples were kept at low temperature in a cryostat equipped with a closed-cycle refrigerator. The MB gamma rays (86.5 keV for 155Gd, 80.6 keV for 166Er and 59.6 keV for 237Np) were separately counted with a pure Ge detector. The Doppler velocity was measured with a laser MB velocity calibrator WissEl MVC-450. The absorber thickness was 115 mg Gd cm-2, 200 mg Er cm-2 and 120 mg Np cm-2 for the 155 Gd, 166Er and 237Np MB measurement, respectively. The ordinary MB spectra (no relaxation phenomenon) were computer-fitted by using a sum of the Lorentz approximation. The paramagnetic relaxation 166Er and 237Np MB spectra were analyzed by using the relaxation-fitting procedure based on the Nowik and Wickman model. The value of δ is referred to the radiation source 155EuPd3 at 12 K for 155Gd and NpAl2 at 4.2 K for 237Np. 5. Crystal Structure of Ln(III)-β-Diketonato Complex (Ln = Gd or Er) [9, 11] Mainly based on the previous report of 151Eu MB studies, the structure and chemical bonding in Ln(III)-β-diketonato complexes (Ln = Gd or Er) were selected as one of main subject of our investigation. In order to give a more reasonable explanation to their MB results, single-crystal X-ray structural determinations were done to several prepared Ln(III)-β-diketonato complexes. Ten β-diketone ligands and their abbreviations used in this study are shown in Figure 4. H3C CH 3 OH O Hacac (CH3)3C CH3 F3C CF3 O OH Hdpm O OH Hdbm O OH Hpta O OH Htaa C(CH3)3 CF3 (CH3)3C CH 3 O OH Hbfa CF3 O O OH Hfta F3CF2CF2C OH O Hbza S CF3 O OH Htta C(CH3)3 O OH Hfod Figure 4. Ten β-diketone ligands and their abbreviations used in this study. 5.1. Material Preparation Polycrystalline samples of the Ln(III)-β-diketonato complexes (Ln = Gd or Er) were prepared by modifying the previous method except Gd(acac)3•3H2O. Gd(acac)3•3H2O was purchased from Aldrich Chemical Co. Inc. A typical procedure was particularly described by 9 preparing the Er(III) dpm complexes. In order to obtain single crystals with reasonable size and good quality for the three-dimensional X-ray structure analysis, numerous attempts were made. As a result, single-crystals of Gd(pta)3•2H2O 1, Gd(bfa)3•2H2O 2 and Er(pta)3•H2O 3 were obtained by recrystallizing their crude product from n-hexane, respectively. Single-crystals of Er2(pta)6 4 and Er(dpm)3 5 were obtained from subliming the crude products under reduced pressure at 423-473 K, respectively. Single-crystals of Er(dpm)3•H2O 6 were obtained through keeping the solution dissolving the 5 into n-hexane at 277 K after two days. Compositions of all the polycrystalline samples were decided by the chemical analysis. All the polycrystalline samples were dried in a vacuum desiccate over three days before the chemical analysis and the MB measurement were conducted. The C, H and N chemical analysis were carried out with a Perkin-Elmer Model 2400. Gd and Er contents were determined by the chelatometric titration on the asked samples with H4edta (ethylenediaminetetraacetic acid) standard solution. 5.2. Crystal Structure Determination Crystal structures of the six prepared β-diketonato complexes (separately noted 1, 2, 3, 4, 5, 6 as above) were determined by three-dimensional X-ray methods. The reflection data were collected on a Rigaku AFC5S diffractometer with the graphite monochromated Mo-Kα radiation (λ = 71.069 pm) at room temperature. The structure for the 5 was solved by a direct method using a SAPI92 program and those for the 1, 2, 3, 4 and 6 were solved by a heavy atom method using a DIRDIFF92 program, and the structures were expanded by using Fourier techniques. All the calculations were performed by using a teXsan crystallographic software package from Molecular Structure Corporation. The X-ray crystallographic files in CIF for the 3, 4, 5 and 6 were deposited as Document No. 75020 at the Office of the Editor of Bull. Chem. Soc. Jpn and also deposited at the CCDC, 12 Union Road, Cambridge CB21EZ, UK. And their copies can be obtained on request, free of charge, by quoting the deposition numbers 177648-177651. The X-ray crystallographic files in CIF for the 1 and 2 were also deposited at the CCDC, 12 Union Road, Cambridge CB21EZ, UK and their copies can also be obtained on request, free of charge, by quoting the deposition numbers 181237-181238. We are planning to publish their details, too. 5.3. Gd(III) and Er(III) pta Complexes For Gd and Er pta complex, crystal structures of Gd(pta)3•2H2O 1, Er(pta)3•H2O 3 and Er2(pta)6 4 were successfully determined. The Crystal structure of Ln pta complex has not yet been found to be reported. This report should be the first time. The results obtained from infrared spectroscopic studies indicate that anhydrous and monohydrate species are possible for the Ln pta complex. The crystal structures for 1, 3 and 4 determined in this study indicate that the Ln pta complex crystallizes not only in anhydrous and monohydrate structures but also in dihydrate structure. Figures 5a, 5b, 5c show the structures for 1, 3 and 4, respectively. The Gd(III) ion in 1 is eight-coordinated with three bidentate pta ligands and two water molecules. However, the Er(III) ion in 3 is seven-coordinated with three bidentate pta ligands and one water molecule. 10 Moreover, it is interesting that 4 has a dimeric structure as being observed in the Ln dpm complexes, Ln2(dpm)6 (Ln = La-Dy); each Er(III) ion is seven-coordinate and one of the chelating oxygen atoms of the pta ligand bridges Er(pta)3 fragments, serving as the seventh ligand atom. The coordination polyhedron around Gd(III) in 1 is a distorted square antiprism. This configuration has also been found in some other reported crystal structures as Gd(acac)3•3H2O and Eu(tta)3•2H2O. The coordination polyhedron around Er(III) in 3 is a slightly distorted monocapped trigonal prism as being commonly found in many monohydrate Ln β-diketonato complexes. The coordination polyhedron around each Er(III) in 4 is also a slightly distorted monocapped trigonal prism. Figure 5a, 5b. The molecule structures of Gd(pta)3•2H2O 1 (left) and Er(pta)3•H2O 3 (right). The mean bond length Gd-O (244 pm) for 1 is obviously longer than the mean bond length Er-O (227 pm) for 3 and (229 pm) for 4. However, the mean bond angle O-Gd-O (71.4˚) for 1 is clearly smaller than the mean bond angles O-Er-O (75.2˚) for 3 and (74.7˚) for 4 in the same pta chelate rings. The shortest Gd(III)-Gd(III) distance in 1 is 610.4 pm. The shortest Er(III)-Er(III) distance in 3 is 587 pm, being shorter than that of 1. The shortest Er(III)-Er(III) distance in 4 is 382 pm, being the intermolecular Er1-Er2 distance and clearly shorter than that of 1 and 3. Figure 5c. The molecule structure of Er2(pta)6 4. 11 5.4. Gd(III) bfa Complex For Gd and Er bfa complexes, only the crystal structure of Gd(bfa)3•2H2O 2 was successfully determined in this study. This is also the first report on the crystal structure of the Ln pta complex. The structure of 2 as shown in Figure 6 is similar to that of 1. That is to say, the Gd(III) ion in 2 is eight-coordinated with three bidentate bfa ligands and two water molecules and the coordination polyhedron around Gd(III) is a distorted square antiprism. Mean bond length Gd-O Figure 6. The molecule structure of Gd(bfa)3•2H2O 2. (245 pm) for 2 is almost similar to that of 1 (244 pm) and is clearly longer than that of Gd(acac)3•3H2O (237 pm) being reported as the Gd(III) ions coordinating with three bidentate acac ligands and two water molecules. Mean bond angle O-Gd-O (71.3˚) for 2 is almost the same with that of 1 (71.4˚) in the same chelate rings and is smaller than that of Gd(acac)3•3H2O (72.0˚). The shortest Gd(III)-Gd(III) distance in 2 is 609.2 pm, being almost the same with that of 1 (610.4 pm). 5.5. Er(III) dpm Complex For Gd and Er dpm complexes, crystal structures of Er(dpm)3 5 and Er(dpm)3•H2O 6 were determined in this study. Up to now, three kinds of complexes (nonaqua, monoaqua and dimmer) have been reported for the Ln dpm complex. The crystal structure of 5 has been reported previously and is confirmed by us in this study. To the Ln β-diketonato complex, this is the only one that the crystal structure of the monomer anhydrous complex has been solved up to now. Other reported structures of the Ln-β-diketonato complexes are hydrates, hydrated dimers and various adducts and the Ln(III) ion in those complexes are seven or eight coordinates. Crystal structures of the monoaqua for the Ln dpm complex have been reported for Eu(dpm)3•H2O and Dy(dpm)3•H2O. The dimeric structures for the Ln dpm complexes have been known for La-Dy. For Gd dpm complex, the crystal structures of the monoaqua and dimmer have been identified in this study. Figure 7 shows the observed XRD patterns for the prepared Gd dpm complex and the calculated XRD patterns for Pr2(dpm)6 based on the reported single-crystal X-ray structural data. It clearly indicates that the prepared Gd dpm complex is iso-structural to Pr2(dpm)6 and its chemical formula should be written as Gd2(dpm)6. Moreover, the monoaqua, Gd(dpm)3•H2O, can be obtained by recrystallizing the prepared Gd2(dpm)6 sample from 95% methanol. We attempted to determine the crystal structure of the monoaqua by three dimensional X-ray analysis, unfortunately, the obtained single-crystals were not good enough for the full measurement. As a result, its crystal system was determined as triclinic and its 12 Relative intensity (%) lattice parameters were determined as a = 1402.4(5), b = 1652.4(7), c = 1067.7(2) pm, α = 95.33(3), β = 104.25(2), γ = 115.07(3)˚. These results indicate that the obtained single-crystal is the monoaqua, Gd(dpm)3•H2O being iso-structural to Er(dpm)3•H2O and the reported 100 80 60 40 20 100 80 60 40 20 0 (a) (b) 8 12 16 20 2-theta (degree) 24 28 Dy(dpm)3•H2O. From the Figure 7. The observed XRD patterns for the prepared consideration of Ln contraction, Gd(III) dpm complex (a) and the calculated XRD it has been suggested that the patterns for Pr2(dpm)6 based on the reported monoaqua dpm complex could single-crystal X-ray structural data. not be formed for the Ln ions heavier than Dy. Our study, however, indicates that the monoaqua dpm complex is available to be formed for Er, even though the ion radius is smaller than that of Dy. As reported previously, Er(III) in 5 is coordinated with three bidentate dpm ligands. Interestingly, one of the dpm chelate rings lies in a mirror plane and the other two are symmetry related across this mirror plane. The crystal structure of 6 as shown in Figure 8 is similar to that of 3 and is iso-structural to the reported Dy(dpm)3•H2O. Er(III) in 6 is coordinated with three bidentate dpm ligands and one water molecule. The coordination polyhedron around Er(III) are a trigonal prism in 5 and a slightly distorted monocapped trigonal in 6 as same as being observed in 3 and the reported Dy(dpm)3•H2O. The mean bond length Er-O (223 pm) for 5 is clear shorter than that of seven-coordinated complexes, such as 6 (227.1 pm) and Dy(dpm)3•H2O (229.4 pm) and eight-coordinated complexes, such as 2 (244 pm). The difference 2.3 pm between the mean bond length Er-O of 6 and Dy(dpm)3•H2O compares favorably with the difference of the seven coordinated ion radius of Er (94.5 pm) and Dy (97.0 pm) and is well interpreted by the Ln contraction. The mean bond angle O-Er-O (74.3˚) in the same dpm chelate rings for 6 is almost the same with that of 5 (74.1˚), however, is slightly larger than that of O-Dy-O (73.6˚) for Dy(dpm)3•H2O. The shortest Er(III)-Er(III) Figure 8. The molecule structure of Er(dpm)3•H2O 6. distance is 998 pm in 5 and 558 pm in 6. 6. Information on the Coordination Structure and Chemical Bonding in Gd(III) Metal Complexes [11-20] 13 Eu and Np compounds, especially their metal complexes, have been studied by using 151 Eu and 237Np MBS, respectively. The δ of 151Eu can give valuable information on the bond T% characteristic as well as the oxidation state of Eu; the δ of 237Np can give information not only on the bond characteristic and the oxidation state of Np but also the coordination number (C.N.) and the mean Np-O bond distance. There is a potentially wide interest in MBS study on the structure and chemical bonding in Gd(III) metal complexes since Gd is located in the middle of the Ln series and can be considered as a representative of Ln in some case. In our study, a systematic investigation has been finished on the 100.0 structure and chemical bonding in some 99.8 (a) Gd(III) metal complexes by 155Gd MBS. 99.6 The subject of our investigation is various 100.0 kinds of Gd(III) complexes having 99.8 (b) different C.N. and different ratios of 99.6 coordinating oxygen to nitrogen atoms, 100.0 including β-diketonato complex, 99.5 (c) cyano-bridged complex, edta complex, 99.0 terpyridine (terpy) complex phthalocyanine (H2Pc) complex. and The 98.5 100.0 99.5 99.0 98.5 100.0 99.8 99.6 results indicate a tendency that the δ value (d) decreases with the increase in the C.N. and the number of the nitrogen atoms coordinating to Gd, showing that the Gd-O (e) and/or Gd-N bonds for the investigated -10 -5 0 5 10 Gd(III) metal complexes have a small -1 covalent contribution which is possible to v / mm s be deduced from the oxygen and/or Figure 9. 155Gd Mössbauer spectra for some nitrogen atoms of the ligands donating Gd(III) metal complexes at 12 K. (a) electrons to 6s, 5d and 4f orbits of Gd. Gd(pta)3 ･ 2H2O, (b) Gd(bfa)3 ･ 2H2O, (c) Figure 9 shows some 155Gd MB NaGd(edta) ･ 8H2O, (d) KGd[Fe(CN)6] ･ spectra for some Gd(III) metal complexes at 3H2O and (e) Gd(dpm)3. 12 K. The 155Gd MB parameters and configuration around Gd(III) are listed in Table 1. For a comparison, the data of Gd2O3 (cubic and monoclinic), GdF3 and pyrochlore-type Gd2Zr2O7 are also listed in Table 1. The eight-, nine- and ten-coordinated Gd complexes of N,N′-dimethylformamide (DMF), tetraethylene glycol (EO4), pentaethylene glycol (EO5) and 4,4′-bipyridine N,N′-dioxide (dpdo) are also listed in Table 1. All of the 155Gd MB results are obtained in our study. All of the 155Gd MB spectra are typical patterns of electric quadrupole interactions for 155Gd nucleus and the degree of quadrupole splitting are clearly different. The results for Gd2O3 and GdF3 reported here are generally in good agreement with those reported previously. As described above, δ is a measure of the s-electron density at the MB nucleus and can be influenced by the local structure around the MB nucleus. The definition of δ can be described as below: 14 Table 1. 155Gd Mössbauer parameters at 12 K and the coordination configuration around Gd(III) for some Gd(III) compounds. Code Complex δ1 e2qQ mm s-1 6.25 2Γ mm s-1 0.92 Coordination configuration GdF9 1 GdF3 mm s-1 0.67 2 Gd(η2-NO3)2(η1-NO3)(EO4) 0.68 3.53 1.19 GdO10 3 Gd(η2-NO3)2 (EO5) (NO3) 0.65 4.34 1.18 GdO10 4 Gd(EO5) (H2O) 3(ClO4)3 0.67 3.37 1.13 GdO9 5 Gd(NO3)6(µ-dpdo)3·2CH2Cl2 0.67 2.78 1.23 GdO9 6 NaGd(edta)･8H2O 0.62 4.72 1.10 GdN2O7 7 NH4Gd(edta)･6H2O 0.61 3.62 1.35 GdN2O6 8 Gd(bza)3･2H2O 0.64 4.43 1.11 GdO8 9 Gd(bfa)3･2H2O 0.63 3.09 1.28 GdO8 10 Gd(pta)3･2H2O 0.61 1.67 1.45 GdO8 11 Gd(tta)3･2H2O 0.60 7.26 1.23 GdO8 12 Gd(taa)3･3H2O 0.58 4.47 1.27 GdO8 13 Gd(acac)3･3H2O 0.57 5.64 1.38 GdO8 14 Gd(fta)3･3H2O 0.55 7.56 1.21 GdO8 15 Gd(fod)3･H2O 0.55 2.52 1.44 GdO8 16 Gd2Zr2O7 0.55 8.49 2.15 GdO8 17 Gd(dmf)4(H2O)3(µ-CN)3Fe(CN)5 0.66 3.40 1.02 GdNO7 18 GdCr(CN)6･4H2O 0.61 4.30 0.93 GdN6O2 19 GdFe(CN)6･4H2O 0.61 4.07 0.90 GdN6O2 20 GdCo(CN)6･4H2O 0.60 4.12 0.87 GdN6O2 21 KGdRu(CN)6･3H2O 0.60 4.81 1.01 GdN6O2 22 KGdFe(CN)6･3H2O 0.59 4.68 1.04 GdN6O2 23 Gd(dpm)3 0.65 6.49 1.15 GdO7 24 Gd(dbm)3･H2O 0.60 6.44 1.46 GdO7 25 Gd2O3 (monoclinic) 26 Gd2O3 (cubic) 27 28 GdPc2 Gd(terpy)3(ClO4)3 0.45 0.46 0.49 0. 51 0.50 0.41 0.40 5.36 2.78 0.49 5.53 10.85 3.65 1.55 0.92 0.70 0.64 1.55 GdO7 GdO7 GdO7 GdO6 GdO6 GdN8 GdN9 (1) Relative to the 155EuPd3 source; Error δ: 0.02 mm s-1; e2qQ and 2Γ: 0.05 mm s-1. δ = (4/5)πZe2(∆R/R)R2{|ψ(o)|2A - |ψ(o)|2S} Here, Z: atomic number; e: elementary electric charge; R: nuclear radius; |ψ(o)|2A and 15 3 0.65 4 5 6 0.60 0.55 8 9 10 7 11 12 13 14 15 16 17 23 18 19 20 24 21 22 26 25 26 25 25 0.45 GdO7 GdN6O2 GdNO7 GdO8 GdN2O6 GdN2O7 0.35 28 GdO6 GdN8 27 0.40 GdN9 0.50 GdO9 complexes with the DMF, EO4, EO5, dpdo, edta, CN, terpy and Pc complexes, a 2 GdO10 δ values of the β-diketonato 1 GdF9 The variation in δ for the investigated Gd(III) metal complexes (0.40 ~ 0.65 mm s-1) are smaller than that for the reported Gd intermetallic compounds (-0.2 ~ 0.9 mm s-1). However, comparing the 0.70 Isomer shift (δ) / mm s–1 |ψ(o)|2S: total electron density at the nuclear position of the MB atom in the absorber and radiation source, respectively. tendency shows that the δ Coordination configuration values decrease with an Figure 10. Plot of 155Gd Mössbauer isomer shift (δ) increase in the C.N. and the against coordination configuration around the Gd(III) ion nitrogen atoms coordinating in some Gd compounds. to Gd(III) can be found (see Figure 10), i.e., GdO10 ≈ GdO9 > GdN2O7 ≥ GdO8 ≈ GdN6O2 ≥ GdO7 > GdO6 > GdN8 ≈ GdN9. This indicates that the s-electron density at the MB nucleus is larger as the C.N. decreases and as the coordinating atoms change from oxygen to nitrogen. Furthermore, all of the δ values fall down between that of cubic Gd2O3 and GdF3 except for the terpy (GdN8) and Pc (GdN9) complexes. It means that there is a greater s-electron density at Gd nucleus in each of the investigated Gd(III) metal complexes than in GdF3 due to ∆R/R < 0 for 155Gd. In the terpy (GdN8) and Pc (GdN9) complexes, the s-electron density at Gd nucleus is even greater than in cubic Gd2O3. These results are consistent with those of the 151Eu MB spectroscopic studies on some Eu(III) metal complexes. In the case of 151 Eu, ∆R/R > 0, having a positive value, the δ values of the Eu(III)-β-diketonato complexes (+0.15 ~ +0.56 mm s-1), such as Eu(acac)3•H2O, Eu(tta)3•2H2O and Eu(dpm)3, also fall down between those of EuF3 (0 mm s-1) and the cubic Eu2O3 (+1.01 mm s-1). This has been concluded due to small covalent contribution in the Eu(III)-β-diketonato complexes since EuF3 can be considered as a purely ionic compound. The same should be applied on the Gd(III)-β-diketonato complexes since the nature of the chemical bonding in GdF3 can also be considered to be closest to purely ionic. Thus, the Gd-F bond in GdF3 is available to be assumed to be free of covalent contribution. The difference in δ values for cubic Gd2O3 and GdF3 should be considered as the difference in the degree of covalent contribution in their chemical bonding. The tendency of δ observed in this study should indicates that a small covalent contribution exist in the Gd-O and Gd-N bonds of the investigated Gd(III) metal complexes. The decrease in δ value means the increase of covalent contribution in the Gd-O and Gd-N bonds of the investigated Gd(III) metal complexes. In general, there are two possibilities being enable to cause the decrease of δ value for 16 155 Gd: (a) increase electron density of the 6s orbit; (b) decrease of electron density of the 5d and/or 4f orbit, resulting as the decrease of the shielding effect and then increasing s-electron density at the Gd nucleus. Since the covalent contribution in the Gd-O and Gd-N bonds is probably through oxygen and/or nitrogen atoms of ligand group donating electrons to 5d, 4f and 6s (6p) orbits of Gd(III), it is adaptable that the decrease of the δ value is due to the resultant result of the above two possibilities. As mentioned above, the electronic structure calculations have indicated that 5d and 4f orbits participate in the covalent bonding in Ln compounds. Here, 155Gd MB study indicates that the covalent bonding in Ln compounds is not only possible to be related to their 5d and 4f orbits but also their 6s (6p) orbits. The electric quadrupole coupling constant, e2qQ, give a direct measure of the magnitude of the electric field gradient (EFG) at the MB nucleus. The EFG is often produced by charges at greater distance (the lattice EFG) and by valence electrons (the valence EFG). In the case of 155Gd, the lattice EFG would be dominant in most situations since Gd(III) (4f7) has the high symmetric valence electron distribution. Thus, the e2qQ value is sensitive to the change of the symmetry of the local structure around Gd(III). The e2qQ values of the investigated Gd(III) metal complexes spread out from 1.67 to 7.56 mm s-1. It should be considered to reflect well the difference in the symmetry of the coordination polyhedron around Gd(III) in these Gd(III) metal complexes. On the other hand, there is a trend that e2qQ values of the seven-coordinated complexes, such as Gd2(dpm)6 and Gd(dbm)3•H2O, are larger than that of the eight-coordinated complexes, such as Gd(pta)3•2H2O 1 and Gd(bfa)3•2H2O 2. This can be also observed from the Er(III) metal complexes as described in the next section. It is reasonable since the local symmetry around Gd(III) is higher for a square antiprism than that for a monocapped trigonal prism. Gd(fod)3•H2O having the smaller e2qQ value could be considered to have an eight-coordinated dimeric hydrate structure as same as that of the reported Pr2(fod)6•2H2O. The line-width (2Γ) for the investigated Gd(III) metal complexes range from 1.11 to 1.46 mm s-1. This is broader than that of the ordinary Gd compounds, such as GdPd3 (0.89 mm s-1) obtained in this study. 7. Paramagnetic Relaxation MB Spectra of 166Er [9, 19] MB effect of the 80.6 keV transition in 166Er has been effectively applied to investigate the structural and magnetic properties of Er containing alloys and intermetallic compounds, but 166Er MB spectra of the ordinary paramagnetic erbium compounds are scarcely observed, especially the paramagnetic Er(III) metal complexes. An attempt applying the 166Er MBS for investigating the structural chemistry of various paramagnetic Er compounds, especially various functional paramagnetic Er(III) metal complexes, has been done in our study. As a preliminary experiment of 166Er MB spectroscopic study on the structural chemistry of Er(III) metal complexes, 166Er MB spectra for Er(III) formate, acetate, oxalate and cyano-bridged complexes were firstly observed as shown in Figure 11. A wide variety in shape from five line absorption to broad absorption spectrum was observed. It is clearly different from the shapes of the 166Er MB spectra for Er(HCOO)3•2H2O and Er(HCOO)3. This is considered to be mainly caused by the different paramagnetic τ. So far there were a few studies on the paramagnetic relaxation phenomenon observed in 166Er MBS, it is probably because the paramagnetic relaxation results a complex shape of the MB spectrum which is 17 1.00 100 0.95 99 T Er(HCOO)3·2H2O T 0.98 98 20 ns Er(HCOO)3 0.002 ns 0.80 1.00 1.00 0.98 T T 0.95 0.96 0.90 0.020 ns 2.0 ns 1.00 1.00 Er(CH3COO)3 0.98 T T 0.98 0.96 0.96 0.94 Er2(C2O4)3·3H2O 1.00 0.98 Er[Fe(CN)6]·4H2O T 0.98 96 0.090 ns 0.20 ns 0.94 1.00 98 0.96 0.96 0.10 ns 0.94 0.12 ns 0.94 -100 0.90 0.85 0.96 97 100 99 98 97 100 99 98 97 100 99 98 97 100 T T (%) 1.00 -50 0 -1 v / mm s 50 -100 -50 0 -1 v / mm s 50 100 -100 -50 0 -1 v / mm s 50 100 Figure 11. 166Er Mössbauer spectra for some Figure 12. The simulated 166Er Mössbauer paramagnetic Er(III) metal complexes at 12 K spectra for several relaxation times obtained in this study. according to the Nowik and Wickman model with gμHeff = 22 mm s-1 (610 T), 2Γ = 8 mm s-1, δ = 0 mm s-1 and e2qQ = 0 mm difficult to be fitted by Lorentzian lines. The reason of the paramagnetic s-1. Zeeman Splitting, Δ is assumed to be relaxation effect is considered to be as follows: zero. the Er(III) ion with electronic configuration 4f11 has three unpair electrons. All of energy levels of the f orbits are degenerated into 52 folds in higher symmetrical crystal field. If the Er(III) ion is put into lower symmetrical crystal field, the energy level degenerated into 52 folds will be split into several energy sub-states. The energy sub-state of J = ±15/2 Kramers doublet can become the ground level due to the spin-orbital interaction in some situation. In this modulation, the relaxation phenomenon can be taken place between the upper spin +15/2 and the down spin –15/2. Figure 12 shows some simulated 166Er MB spectra for several τ with gµHeff = 22 mm s-1 (610 T), 2Γ = 8 mm s-1, δ = 0 mm s-1 and e2qQ = 0 mm s-1 by a relaxation-fitting procedure based on the Nowik and Wickman model. The Zeeman splitting, Δ is assumed to be zero. The relaxed 166Er MB spectra change sensitively with the change in τ about τ = 0.002 ~ 20 ns. It indicates the relaxation-fitting procedure based on the Nowik and Wickman model can be used to evaluate the τ of the relaxed 166Er MB spectrum being comparable with the nuclear Larmor precession. The paramagnetic relaxation has two mechanisms, lattice-spin relaxation and spin-spin relaxation, and sometimes the spin-spin distance has been suggested to be important to determine τ. Therefore we carried out a systematic study by using various Er(III)-β-diketonato complexes to examine the relation between the spin-spin distance and τ by using 166Er MBS in connection with the results of three dimensional X-ray analysis. 18 100.0 99.5 99.0 100.0 100.0 (a) 99.5 (c) T% T% (b) 100.0 99.0 (d) 99.5 100.0 100.0 99.5 100.0 99.5 99.0 100.0 99.5 99.0 -100 (c) 99.5 99.0 100.0 99.8 99.5 -100 (b) 99.5 100.0 99.5 99.0 100.0 99.5 (a) (d) -50 0 50 -1 v / mm s 100 Figure 13a. The 166Er Mössbauer spectra for Er(III)-β-diketonato complexes at 12 K. (a) Er(pta)3•H2O 3, (b) Er(dpm)3•H2O 6, (c) Er2(pta)6 4, and (d) Er(dpm)3 5. (e) (f) (g) -50 0 50 -1 v / mm s 100 Figure 13b. The 166Er Mössbauer spectra for Er(III)-β-diketonato complexes at 12 K. (a) Er(fod)3•H2O, (b) Er(pta)3•H2O 3, (c) Er(dpm)3•H2O 6, (d) Er(acac)3•H2O, (e) Er(bfa)3•H2O, (f) Er(taa)3•H2O, (g) Er(dbm)3•H2O. Figures 13a and 13b show the 166Er MB spectra for the investigated that the investigated Er(III)-β-diketonato complexes at 12 K. It is clear 166 Er(III)-β-diketonato complexes is paramagnetic relaxed Er MB spectra. The MB parameters are summarized in Table 2. The relative transmission for each complex is less than 1%, being much smaller than that of ionic compounds and alloys having several percent. It indicates that the recoilless fractions of these complexes are very small even at 12 K. The spectra show a wide variety in absorption shape depending on τ; a broad single absorption with the short τ and a five-line absorption with the long τ. The estimated τ is spread out from 0.1 to 1 ns. The estimated τ for the 4 is 0.1 ns, whereas those for the 2 and the 3 are 0.4 and 0.5 ns, respectively. This is the same order to the increase in the shortest Er(III)-Er(III) distances in their crystal structure: 382 pm for the 4, 558 pm for the 2 and 587 pm for the 3. Thereby, τ correlates to the intermolecular Er(III)-Er(III) distance. In another word, the spin-spin relaxation is important in the Er(III)-β-diketonato complexes. This is confirmed by examining the MB spectra for the monohydrate complexes (see below). Interestingly, the τ for the 1 is, however, unexpectedly short as 0.4 ns though the shortest Er(III)-Er(III) distance (998 pm) is prominently longer than the other dpm and pta complexes. It will be interpreted as the τ is 19 dominated by not only the shortest Er(III)-Er(III) distance but also the C.N. Although we could not explain this completely at this stage, the difference in C.N. will lead to the difference in the energy level of the Er(III) ion through a crystal field effect, causing the difference in τ. Indeed, τ for the five Er(III) edta complexes investigated in this study also depends on the C.N. We are planning to publish their details, too. Table 2 166 Er Mössbauer parameters of the Er(III)-β-diketonato complexes at 12 K. e2qQ mm s-1 2Γ mm s-1 ns Heff T Er(fod)3･H2O 6.1 4.1 1.0 711 [Er(pta)3(H2O)] 7.0 5.6 0.5 654 [Er(dpm)3(H2O)] 4.3 a) 8.0 0.4 690 [Er(dpm)3] 0.1 8.0a) 0.4 525 [Er(dbm)3(H2O)] -0.1 a) 8.0 0.3 294 Er(fta)3 -2.4 8.3 0.2 640 Er(acac)3･H2O -2.7 7.5 0.2 598 Er(tta)3･2H2O 1.8 a) 8.0 0.2 (316) [Er2(pta)6] 3.6 8.4 0.1 621 Er(bfa)3･H2O 1.0 7.0 0.1 651 Er(taa)3･H2O 2.0 6.8 0.1 582 Er(bza)3･2H2O -4.0 5.7 0.1 (433) Complex τ a). The line-width (2Γ) was fixed at 8.0 mm s-1. Since τ depends on the C.N., we will focus our interest on the seven-coordinated Er(III)-β-diketonato complexes. As shown already, the C.N. for the Er(III) ions in the two monohydrate 2 and 3 are seven since the water molecule participates in the coordination. This would be safely extended to the other monohydrate Er(III)-β-diketonato complexes though the crystal structure are not known. The XRD pattern indicates that Er(dbm)3•H2O is iso-structural to Ho(dbm)3•H2O having a monocappted octahedral coordination with C3 symmetry. Thus, Er(dbm)3•H2O can not be included to the following discussion. Indeed, the MB spectrum is clearly different from other seven-coordinated complexes. However, this suggests that the electronic configuration may also play an important role in paramagnetic relaxation. The values of the τ are widely spread among the monohydrate complexes. The order of the τ is as follows: Er(fod)3•H2O > Er(pta)3•H2O 3 > Er(dpm)3•H2O 2 > Er(acac)3•H2O > Er(bfa)3•H2O > Er(taa)3•H2O. This indicates that the τ depends on the substitute. The substitutes (R1, R2) of the β-dikentone (R1COCH2COR2) ligands in the above order is (C3F7, t-Bu), (CF3, t-Bu), (t-Bu, t-Bu), (CH3, CH3), (Ph, CF3) and (CF3, CF3). This order is essentially that of the decrease in the bulkiness of the substitutes. Since the Er(III)-Er(III) distance will depend on the size of the substitutes, we can suppose that the τ depends on the intermolecular Er(III)-Er(III) distance as the results. Because we could not know the shortest Er(III)-Er(III) distance by single-crystal X-ray structure determination, we measured the 20 density of the crystal, expecting the density of the crystal would reflect the intermolecular distance. Such an attempt was, however, unsuccessful. No correlation between the τ and the density could be observed. This would be reasonable since the packing of the molecules in the crystal was not always the same and the substitution of H atom to F atom would result in the increase in density. Since the MB spectra with longer τ give resolved five-line absorption, we can obtain the value of e2qQ in good accuracy. Judging from the absorption shape, the values of e2qQ for Er(fod)3•H2O, the 3, the 2, and the 1 is considerably reliable. Interestingly, the values for the monohydrate complexes except for Er(dbm)3•H2O are obviously larger than that of the anhydrous complex of the 1. This suggests that there would be some differences in population of the valence orbits between the six- and seven-coordinated complexes. This is reasonable since the local symmetry around the Er(III) ion is higher for trigonal prism than that for monocapped trigonal prism being also confirmed in the seven- and eight-coordinated Gd(III)-β-diketonato complexes. The value of Er(dbm)3•H2O is clearly smaller than that of the other monohydrate complexes. This indicates the local symmetry around the Er(III) ion in Er(dbm)3•H2O is higher than that of the other monohydrate complexes being in good agreement with the coordination structure. Concerning with the values of Heff, the most of the values range between 500 and 720 T, being close to that of metallic erbium (742 T). The reliability of the smaller values for Er(tta)3•2H2O and Er(bza)3•2H2O, figured in parentheses, is rather low. This is resulted from the rapid relaxation and weak absorption. 8. 237Np MS Spectroscopic Studies on Neptunyl(VI) Compounds [21-24] Since the gamma ray 59.5 keV MB resonance in 237Np (5/2 → 5/2) has been found in 1964, various oxides, fluorides, oxyfluorides and polycarboxylic compounds of Np, from III to VII oxidation states, have been investigated by using 237Np MBS. The δ of 237Np is spread out from +40 to –80 mm s-1 relative to NpAl2 at 4.2 K and clearly correlated with the oxidation state of Np. Correlation between the δ and the C.N. of Np has also been found to the Np(VI) compounds. A correlation between the δ and the mean Np-O bond distance has also been reported for the oxygen-coordinated Np(VI) compounds: the δ of 237Np is increased with the increase of the mean Np-O bond distance. However, about the correlation between the δ and the mean Np-O bond distance, there are some experimental evidences for the six- and eight-coordinated Np(VI) compounds, no data can be found for the seven-coordinated Np(VI) compounds. Although a lot of data of 237Np MB spectroscopic studies on the Np compounds have been reported, investigation for paramagnetic relaxation phenomenon is still insufficient. This study focuses not only on the inorganic structural chemistry of actinide science, but also on the nuclear waste management. We carried out a systematic investigation on some novel neptunyl(VI) compounds by 237Np MBS, in combination with XRD, thrmogravimetric (TG) analysis and magnetic susceptibility measurement. 8.1. Neptunyl(VI) Trinitrato Complex, M[NpO2(NO3)3] (M = NH4+, K+) [21] 21 In this study, the results indicate that the magnetic hyperfine splitting of NH4, K and Rb salts observed in their 237Np MB spectra are due to slow paramagnetic relaxation. The δ and e2qQ values of NH4, K and Rb salts are the same within the experimental error. The Rietveld analyses indicate that the mean Np-O bond distances of NH4, K and Rb salts are very close to one another. The relationship between the mean Np-O bond distance and the δ of 237Np value has been re-confirmed to the neptunyl(VI) compound. These studies suggest that the environment around Np(VI) in NH4, K and Rb salts are similar to each other and the influence of the different M cation is relatively small for the neptunyl(VI) trinitrato complexes. NH4 and K salts were prepared as follows: firstly, the neptunyl(VI) nitrate salt, NpO2(NO3)2•xH2O, was prepared by adding concentrated nitric acid into a 0.1 M Np(V and VI) stock solution and then evaporating it. Then NH4 salt (brown precipitation) was obtained by the addition of a slight excess NH4NO3 to the neptunyl(VI) nitrate solution, and evaporating it at about 333 K. Subsequently, K salt and freshly prepared Rb Salt were also obtained by the same method by using KNO3 and RbNO3, respectively. Freshly prepared NH4, K and Rb salts were identified by XRD. Their crystal structures were refined by RIETAN-97β, which was based on the Rietveld method. A magnetization measurement of NH4 salt was performed for a polycrystalline sample by using a SQUID magnetometer (MPMS, QD). The magnetic susceptibility measurement was made from 2 K to room temperature. Revision of the diamagnetism was not performed. 3500 3000 2500 Intensity 2000 1500 1000 500 0 40 60 80 2θ / ° Figure 14. Result of the Rietveld refinement for NH4[NpO2(NO3)3] (Plus marks: observed, solid line: calculated; Rwp = 12.67 %, Rp = 9.78 %). The result of the Rietveld refinement of NH4 salt with space group R-3C is shown in Figure 14. The results suggest that different M cations can influence the lattice parameter and the unit-cell volume, but can hardly influence the bond distance and the bond angle between the Np atom and the coordinated oxygen atom. As shown in Figure 15, the structure of the [NpO2(NO3)3]- anion in NH4, K and Rb salts adopts a hexagonal-bipyramidal 22 Figure 15. Structure of the [NpO2(NO3)3]- anion in the neptunyl(VI) trinitrato complexes. 1000 100.0 800 4.5 K 99.8 99.7 100.00 χ-1 / mol•emu-1 99.9 600 μeff = 1.59 μB / Np 400 T (%) 99.95 99.90 μeff = 2.14 μB / Np 200 10 K 99.85 0 100.04 0 50 150 200 250 300 T/K 100.00 Figure 17. Plot of the reciprocal molar susceptibility against temperature for 99.96 30 K 99.92 -200 100 -100 0 NH4[NpO2(NO3)3] (Experimental: μeff = 100 1.59 μB / Np, theoretical: μeff = 2.14 μB / Np) -1 v / mms B Figure 16. 237Np Mössbauer spectra for K[NpO2(NO3)3] at 4.5, 10 and 30 K. B Relative intensity geometry consisting of the Np(VI) ion coordinated with three bidentate nitrate ions and two oxide ions. The mean Np-O bond-distances of NH4 and K salts are 228.5 pm and 228.3 pm, respectively. They are very close to that of the reported Rb salt (228.4 pm) within the experimental error. The 237Np MB spectra of K salt at 4.5, 10, and 30 K are shown in Figure 16. Magnetic hyperfine splitting can be clearly observed from 4.5 to 30 K. Although 10 ns neptunyl(V and VI) complexes are thought to be paramagnetic in ordinary, a ferromagnetic neptunyl(V) formate complex, 1 ns NpO2(OOCH)•H2O, was recently discovered below 12 K. The magnetic hyperfine splitting in the MB spectra of NH4, K and Rb salts may 0.2 ns arise from two possibilities: one is slow paramagnetic relaxation, the other one is the 0.05 ns presence of a magnetically ordered state in the measured temperature range. It is not easy to distinguish the two possibilities by only using 0.003 ns the results of their 237Np MBS. The magnetization of NH4 salt measured -200 -100 0 100 -1 v / mm s Figure 18. Simulated 237Np Mössbauer spectra for several τ with Heff = 290 T, 2Γ = 5 mm s-1, δ = -38 mm s-1 and e2qQ = 230 mm s-1. (Heff // Vzz) by SQUID gives useful information about distinguishing the two possibilities. A plot of the reciprocal molar susceptibility of NH4 salt against the temperature is shown in Figure 17. The reciprocal molar susceptibility nearly 23 follows the Curie-Weiss law from 2 K to room temperature. The value of the effective magnetic moment (μeff) was estimated to be ~1.59 μB / Np, which was smaller than the B 6+ 1 theoretical value (2.14 μΒ / Np) of Np (5f ). These SQUID data and the 237Np MB spectra indicate that NH4 salt is paramagnetic down to 2 K. The magnetic hyperfine splitting observed in NH4, K and Rb salt’s MB spectra are due to slow paramagnetic relaxation. Figure 18 is shown some simulated 237Np MB spectra for several τ with Heff = 290 T, 2Γ = 5 mm s-1, δ = -38 mm s-1 and e2qQ = 230 mm s-1, being based on the paramagnetic relaxation model. According to the simulated spectra, the 237 Np MB spectra of NH4, K and Rb salts are similar to the modulation that the τ are longer than about 1 ns. The simulation also insists that the magnetic hyperfine splitting is due to slow paramagnetic relaxation. The δ and e2qQ values of NH4, K and Rb salts are the same within the experimental error, respectively. The δ values are typical for Np(VI) compounds. We know, there is a linear relationship between the δ value and the mean Np-O bond distance for the Np(VI) compounds. In this study, almost the same Np-O bond distances of NH4, K and Rb salts have almost the same δ values. This is consistent with the linear relationship between the δ values and the mean Np-O bond distances established for the neptunyl(VI) compounds. 8.2. 237Np MS Spectroscopic Studies on Neptunyl(VI) Oxalate and Hydroxide [22-24] In recent years, Krot and his co-workers have reported continuously some new neptunium compounds, especially for the seven- and oxygen-coordinated neptunyl(VI) complexes as NpO2C2O4•3H2O. This shows much room for the 237Np MB spectroscopic study. Here, 237Np MB spectra for NpO2C2O4•3H2O and an amorphous neptunyl(VI) hydroxide, NpO2(OH)2•xH2O, prepared by NaBrO3 oxidation are discussed. NpO2C2O4•3H2O was prepared as follows: into a small amount of Np(VI) stock solution, an excess amount of LiOH at 343 K was added until a dark brown precipitate of neptunyl(VI) hydroxide was obtained. The precipitate was separated, washed by distilled water and dried at room temperature. The dried precipitate was re-dissolved into concentrated HNO3. 7% H2C2O4 solution was dropped into the freshly prepared Np(VI) nitrate solution until a grayish green crystalline, NpO2C2O4•3H2O, was obtained. The grayish green crystalline was separated quickly, washed by 0.2 N HNO3, alcohol and ether. Amorphous neptunyl(VI) hydroxide, NpO2(OH)2•xH2O, was prepared by the following procedure: firstly, Np(VI) stock solution was reduced into Np(V) solution by adding H2O2. A neptunyl(V) hydroxide precipitate was obtained after an excess amount of NaOH was added to the Np(V) solution. The neptunyl(V) hydroxide was re-dissolved into concentrated HNO3. The obtained Np(V) nitrate solution was oxidized into freshly Np(VI) solution by adding NaBrO3. Amorphous neptunyl(VI) hydroxide, NpO2(OH)2•xH2O, was obtained by adding NH4OH to the freshly prepared Np(VI) solution. The oxalate and the hydroxide were examined by XRD. Thermal behavior of the hydroxide was investigated in a dry atmosphere on a Sartorius MP8 electric microbalance by recording the TG curve. The 237Np MB spectra for the oxalate were computer-fitted by the relaxation procedure. The Heff value was assumed as 300 T and the 2Γ was fixed to 8 mm s-1. The 237Np MB spectra for the hydroxide were computer-fitted by using the sum-of-Lorentzian 24 approximation. Values of the magnetic moments related to the 237 Np MB transition, used for the 100.00 99.00 fitting procedure are μg = 2.8 x 10-28 97.00 100.00 99.20 NpO2(OH)2•xH2O (amorphous) 98.40 T (%) m2 for the ground state and μe = 1.5 x 10-28 m2 for the excited state. The XRD patterns indicate that the oxalate sample is pure and iso-structural with U(VI)O2C2O4•3H2O. U(VI)O2C2O4•3H2O is a seven-coordinated complex and the oxalate groups are tetradentate, each bridging two uranyl(VI) ions. Only one water molecule is coordinated with the U atom. In order to obtain the pure oxalate sample, the final grayish green crystalline must be NpO2 98.00 97.60 100.00 99.90 NH4[NpO2(NO3)3] 99.80 99.70 100.02 100.00 99.98 [NpO2C2O4(H2O)]•2H2O 99.96 -200 -100 0 100 -1 v / mm s separated quickly. If not, some Figure 19. 237Np Mössbauer spectra for several Np Np(VI) ions will be reduced by compounds at 10 K. H2C2O4, some bright green precipitate will be observed in the solution. However, solid state of the neptunyl(VI) oxalate is stable within three weeks though it has been reported that even in the cold the neptunyl(VI) oxalate gradually decomposed as a result of intermolecular reduction of the Np(VI) ions. This was demonstrated by the XRD patterns of our oxalate sample before and after three weeks. The XRD patterns indicate that the hydroxide is non-crystalline phase and clearly different to the other four kinds of neptunyl(VI) hydroxides prepared in this study. Since the loss of the weight is more than one water molecular when an accurately weighted sample was heated at 100 ˚C for 4 h, the amorphous neptunyl(VI) hydroxide is written as NpO2(OH)2•xH2O. Table 3 237Np Mössbauer parameters of some Np compounds. mm s e 2qQ mm s-1 2Γ mm s-1 NpO2 NpO2(OH)2•xH2O (amorphous) [NpO2C2O4(H2O)]･2H2O* NH4[NpO2(NO3)3] -6.1(4) -43.7(5) -46.8(5) -36.5(1) 177(1) 174(3) 243(1) 2.8(5) 7.4(7) 8 4.8(8) NpO2(OH)2 -46.2(1) 193(1) 4.8 NpO2(OH)2•H2O (orthorhombic ) NpO2(OH)2•H2O (hexagonal ) -39.9(1) -43.4(1) 179(1) 149(1) 11 11 NpO2(OH)2•xH2O•yNH3 (x+y = 1 ) -44.6(1) 168(1) 4.6 Compound δ (NpAl2) -1 *The values of Heff and 2Γ were fixed to 300 T and 8 mm s-1, respectively. 25 Heff T T K 300 288(1) 10 10 10 10 Figure 19 shows the the δ/NpAl2 NpO 2 F2 CN oxalate and the amorphous -1 ( mm s ) NpO 2 CO 3 hydroxide at 10 K. For a NH 4 [NpO 2 (NO 3)3 ] comparison, 237Np MB spectra of - 3 0 a standard absorber of NpO2 and Na[NpO 2 (CH 3 COO) 3 ] prepared NH4[NpO2(NO3)3] at 10 NpO 2 (NO 3) 2 •xH 2 O -3 5 NpO 2 (OH) 2•H 2 O (hexagonal) K are also shown in Figure 19. Am orphous NpO 2 (OH) 2•xH 2O 8 Their 237Np MB parameters were -4 0 NpO 2(OH) 2•xH 2 O•yNH 3 listed in Table 3. The absorption shape of NpO2 show a K 2(NpO 2)2 V 2O 8 -4 5 (NH 4)2Np 2 O 7•H 2 O symmetrical single line since (NH 7 4)3NpO 2F 5 NpO2 has the cubic fluorite -5 0 [NpO 2C 2 O 4(H 2O )]•2H 2O structure, no electric quadrupole K 3 NpO 2F 5 splitting and magnetic hyperfine B a N p O 4 splitting can be observed. - 5 5 β -N a 2 N p O 4 6 NH4[NpO2(NO3)3] show a K 2 N p O 4 complex patterns due to electric - 6 0 L i4 N p O 5 B a 2 C o N p O 6 quadrupole interactions and slow N p F 6 paramagnetic relaxation. However, Figure 20. Plot of isomer shift (δ) against the oxalate at 10 K shows broad coordination number of Np for the Np(VI) paramagnetic relaxation pattern compounds. The data in bold are from this study. and the τ is clearly shorter than that of NH4[NpO2(NO3)3]. The hydroxide shows a pure electric quadrupole splitting patterns as well as that of the other four kinds of neptunyl(VI) hydroxides prepared in this study and no Np(V) species was observed. Since a Np(V) species was observed in the spectrum of a neptunyl(VI) hydroxide prepared by ozone oxidation, NaBrO3 was thought to be better than ozone as a reducing agent. The δ values of the oxalate and the hydroxide were estimated as –46.8(3) and –43.7(3) mm s , respectively. These are in the characteristic range for the Np(VI) species. Figure 20 -1 shows the correlation between the δ and the -35 C.N. for Np(VI) compounds. The δ value of the oxalate falls down in the reported δ range of the Np(VI) seven-coordinated -40 δ / mm s-1 Figure 21. Plot of isomer shift (δ) against the mean Np-O bond distance for the oxygen-coordinated Np(VI) compounds. The mean Np-O bond distances may be deduced from iso-structural U(VI) compounds by using the relation: ∆RUVI RNpVI = 0.01 nm, where R is the ionic radius of the six-coordinated ion. 1: Ba2CoNpO6, 2: Li4NpO5, 3: K2NpO4, 4: β-Na2NpO4, 5: BaNpO4, 6: NpO2(NO3)2･ xH2O, 7: NaNpO2(CH3COO)3, 8: Rb[NpO2(NO3)3], 9: NpO2CO3. 9 : CN = 6 : CN =7 8 7 6 : CN = 8 -45 NpO2C2O4•3H2O -50 5 -55 -60 4 3 1 0.200 2 0.210 0.220 0.230 0.240 Np-O mean bond distance / nm 26 compounds. It is consistent with the result of the XRD analysis. The δ value of the hydroxide falls down in the reported overlap δ range of the Np(VI) seven- and eight-coordinated compounds. Comparing the results of NpO2(OH)2•H2O (orthorhombic), NpO2(OH)2•H2O (hexagonal) and NpO2(OH)2•xH2O•yNH3, the amorphous hydroxide is probably a eight-coordinated compound. Figure 21 shows the plot of the δ value against the mean Np-O bond distance for the oxygen-coordinated Np(VI) compounds including the data of the oxalate. The mean Np-O bond distance of the oxalate was deduced from U(VI)O2C2O4•3H2O by using the relation: Δ(RUVI – RNpVI) = 0.01 nm, where R was the ionic radius of the hexa-coordinated ion. The δ value of the oxalate is located between the six- and eight-coordinated Np(VI) compounds and the linear correlation between the δ value and the mean Np-O bond distance is confirmed to the six-, seven- and eight- coordinated Np(VI) compounds. Mean Np-O bond distance of the hydroxide is estimated as Np-O ≈ 0.225 nm from Figure 21. The e2qQ values of the oxalate and the amorphous hydroxide were estimated as 174(3) mm s-1 and 177(1) mm s-1, respectively. It is the same degree with that of the other four kinds of neptunyl(VI) hydroxides prepared in this study, but clearly smaller than that of NH4[NpO2(NO3)3]. The τ of the oxalate was estimated as about 0.02 ns. 5. Summary As reported above, abundant information on the coordination structure and chemical bonding in Ln and An compounds is available to be obtained from 155Gd, 166Er and 237Np MBS in connection with other analysis technologies. Based on the reported results of the electronic structure calculations, a large quantity of 155Gd MB experiment results have pointed out deeply that a small covalent contribution exists in the Ln-O (or N) bonds of various Ln(III) metal complexes and the covalent bonding is not only possible to be related to their 5d and 4f orbits but also to their 6s (6p) orbits; the covalent contribution in the Ln-O (or N) bonds should be through oxygen and/or nitrogen atoms of ligand group donating electrons to 5d, 4f and 6s (6p) orbits of Gd(III) ions. 166Er spectroscopic studies on Er(III) paramagnetic compounds should also suggest that there is a small covalent contribution in their Er-O (or N) bonds since the results indicate that the relaxation time is related to not only the coordination structure and spin-spin distance but also the electronic configuration. Certainly, comparing with the participation of the 5f orbit in chemical bonding, the contribution of the 4f orbit in chemical bonding is very few. This viewpoint is available to be confirmed from all of the 155Gd, 166Er and 237Np MB parameters obtained in this study. In the case of 237Np MBS, the δ and e2qQ valves are spread out in a large range and very sensitive to the variation of the oxidation state, coordination structure and the distance between the Np atom and the nearest coordinating atom. However, the 155Gd, 166Er and 151Eu MB parameters are only spread out in a narrow range though a tendency that the δ values of 155Gd decrease with the increase in the coordination number and the nitrogen atoms coordinating to Gd(III) has been confirmed in our study. The difference should not only originate from their different mean-square nuclear radius ∆ and their different magnitude of quodrupole moment, Q. Furthermore, the difference features on the coordination structure are also clearly observed from 155Gd, 166Er and 237Np MBS between the Ln and An compounds. Since the 27 participation of the 4f orbit in chemical bonding is very few, the coordination structure in Ln compounds can be approximately considered as a closest packed structure. This means that the ligand is available to be coordinated to Ln(III) ion in all directions if some spaces are still unoccupied around the Ln(III) ion. This structural feature should be only able to cause a small electric field gradient (EFG) in the Ln nucleus position in the Ln compound, as observed from 155 Gd, 166Er and 151Eu MBS. However, in the case of the An compounds, since the participation of the 5f orbit in chemical bonding is large, the covalent contribution in ions like AnO2+ and AnO22+ become greater. As a result, all ligands are only possible to be coordinated to An in the nearly perpendicular direction to linear (O-An-O)+ or (O-An-O)2+ ions. This structural feature should be the main reason why there is a large EFG existing in the Np nucleus position in the Np(V and VI) compound, as observed from 237Np MBS. This study was finished when the author was at Toho University as a doctoral student and at Japan Atomic Energy Research Institute (JAERI) as a joint researcher. References 1. S. A. Cotton, “Lanthanide and Actinide Chemistry”, John Wiley & Sons (2006). 2. H. Nakamatsu, T. Mukoyama, and H. Adachi, “Ionic and Covalent bonds in CeO2 Crystal”, Chem. Phys. Lett. 247 (1995) 168. 3. http://www.mossbauer.org 4. J. Wang, “Mössbauer Spectroscopic Studies on Gadolinium, Erbium and Neptunium Compounds”, Doctoral Dissertation, Toho University, 2002 (in Japanese). 5. G. Czjzek, “Mössbauer Spectroscopy of New Materials Containing Gadolinium” in Mössbauer Spectroscopy Applied to Magnetism and Materials Science, G. J. Long and F. Grandjean (Eds), Plenum Press, New York, 1993, Vol. 1, p.373. 6. J. Stöhr and J. D. Cashion, “160Dy and 166Er Mössbauer Studies of Concentrated and Diluted Rare-Earth Dihydrides: Single-Line Compounds and Crystal-Field Effects”, Phys. Rev. B 12 (1975) 4805. 7. J. Jove, L. He, J. Proust, M. Pages, and P. Pyykkö, “Mössbauer Spectroscopy as a Nuclear Probe for Solid State Transuranium Chemistry”, J. Alloys Comp. 177 (1991) 285. 8. J. Wang, M. Takeda, and T. Shishido, “Development and Evaluation of a New 155 Eu/154SmPd3 Source for Use with 155Gd Mössbauer Spectroscopy”, J. Nucl. Mater. 340 (2005) 52. 9. J. Wang, M. Takahashi, and M. Takeda, “Crystal Structures and 166Er Mössbauer Spectra of Some Er(III)-β-diketonato Complexes”, Bull. Chem. Soc. Jpn. 75 (2002) 735. 10. M. Nakada, M. Saeki, N. M. Masaki, and S. Tsutsui, “Preparation of source and sealed absorber holder for 237Np and238U Mössbauer measurements”, J. Radioanal. Nucl. Chem., 232 (1998) 201. 11. J. Wang, “Structure and Bonding in Some Gd(III) Metal Complexes Studied by 28 Three-Dimensional X-Ray Analysis and 155Gd Mössbauer Spectroscopy”, to be published. 12. J. Wang, M. Takahashi, T. Kitazawa, and M. Takeda, “155Gd Mössbauer Spectra of Some Gd(III)-β-Diketonato Complexes”, J. Radioanal. Nucl. Chem. 255 (2003) 195. 13. J. Wang, J. Abe, T. Kitazawa, M. Takahashi, and M. Takeda, “155Gd Mössbauer Spectroscopic Study of GdM(CN)6•4H2O (M = CrIII, FeIII and CoIII) and KGdM(CN)6•3H2O (M = FeII and RuII)”, Z. Naturforsch. 57a (2002) 581. 14. M. Takeda, J. Wang, T. Nishimura, K. Suzuki, T. Kitazawa, and M. Takahashi, “155Gd Mössbauer Isomer Shifts and Quadrupole Coupling Constants of Gadolinium Complexes”, Hyperfine Interactions, 156-157 (2004) 359. 15. J. Wang, M. Takeda, and A. Nakamura, “Influence on the Structural Characteristics of Defect solid Solution GdxZr1-xO2-x/2 with 0.18 ≤ x ≤ 0.62 under Longer Term Annealing Studied by 155Gd Mössbauer Spectroscopy”, J. Nucl. Mater. 340 (2005) 33. 16. J. Wang, A. Nakamura, and M. Takeda, “Structural Properties of the Fluorite- and Pyrochlore-Type Compounds in the Gd2O3-ZrO2 System xGdO1.5-(1-x)ZrO2 with 0.18 ≤ x ≤ 0.62”, Solid State Ionics 164 (2003) 185. 17. J. Wang, H. Otobe, A. Nakamura, and M. Takeda, “Correlation of Crystal Structures with Electric Field Gradients in the Fluorite- and Pyrochlore-Type Compounds in the Gd2O3-ZrO2 System”, J. Solid State Chem. 176 (2003) 105. 18. A. Nakamura, H. Otobe, J. Wang, and M. Tekeda, “155Gd Mössbauer Spectroscopic and X-ray Diffraction Study of the Zr1-xGdxO2-x/2 (0 < x ≤ 1.0) System”, J. Phys. Chem. Solids 66 (2005) 356. 19. M. Takeda, J. Wang, M. Takahashi, T. Shishido, A. Yoshikawa, and A. Nakamura, “166Er and 155Gd Mössbauer Spectroscopy of R–Rh–B (R = Er, Gd) System”, J. Alloys and Compounds 408-412 (2006) 371. 20. J. Wang, “Structural Characterization of the GdxZr1-xO2-x/2 System by 155Gd Mössbauer Spectroscopic Study” Chapter 8 in Nuclear Materials Research Developments, J. E. Keister (Ed), Nova Science Publishers (ISBN:1-60021-432-0), New York, 2006, in press. 21. J. Wang, T. Kitazawa, M. Nakada, T. Yamashita, and M. Takeda, “Structural Characterization of Neptunyl(VI) Trinitrato Complexes: M[NpO2(NO3)3] (M = NH4+ and K+)”, Bull. Chem. Soc. Jpn. 75 (2002) 253. 22. J. Wang, T. Kitazawa, M. Nakada, T. Yamashita, and M. Takeda, “237Np Mössbauer Spectra for Neptunyl(VI) Oxalate NpO2C2O4•3H2O and Amorphous Neptunyl(VI) Hydroxide NpO2(OH)2•xH2O”, J. Nucl. Sci. Technol. Suppl. 3 (2002) 429. 23. T. Saito, J. Wang, T. Kitazawa, M. Takahashi, M. Takeda, M. Nakada, T. Nakamoto, N. M. Masaki, T. Yamashita, and M. Saeki, “Synthesis of Neptunyl(VI) Hydroxides and Their 237Np Mössbauer Spectra”, J. Radioanal. Nucl. Chem. 239 (1999) 319. 24. M. Nakada, T. Kitazawa, T. Saito, J. Wang, M. Takeda, T. Yamashita, and M. Saeki, “A New Type of Neptunyl(VI) Hydroxide which is Topologically Similar to α–UO2(OH)2”, Bull. Chem. Soc. Jpn. 76 (2003) 1375. 29