Matrix isolation, carbonyls prepared

Spectroscopists also saw the potential of reacting ligands with transition metal under matrix isolation conditions. Photolysis of metal carbonyls in organic (383 or inert gas matrices (39) had already been done, but atoms offered the possibility of step-wise addition of ligands. DeKock (40), Turner (41 ), and Moskovits and Ozin (42) made early contributions, but the work of the last two became dominant (32). By 1972, there were the two distinct branches in transition metal atom chemistry, the preparative and matrix spectroscopic studies. 

The most recent fairly comprehensive review Of the vibrational spectra of transition metal carbonyls is contained in the book by Braterman1. This provides a literature coverage up to the end of 1971 and so the subject of the present article is the literature from 1972 through to the end of 1975. Inevitably, some considerable selectivity has been necessary. For instance, a considerable number of largely preparative papers are not included in the present article. Tables A-E provide a general view of the work reported in the period. Table A covers spectral reports and papers for which topics related purely to vibrational analysis are not the main objective. Papers with the latter more in view are covered in Table C. Evidently, the division between the two is somewhat arbitrary. Other tables are devoted to papers primarily concerned with the spectra of crystalline samples — Table B — to reports of infrared and Raman band intensities — Table D and sundry experimental techniques or observations - Table E. Papers on matrix isolated species, which are covered elsewhere in this volume, are excluded. 

In the last three decades of the twentieth century, following Walter Hieber s retirement, four aspects of the research on mono- and polynuclear metal carbonyl complexes found particular attention. These were the preparation of highly reduced carbonyl metallate anions, the generation of stable metal carbonyl cations, the matrix isolation of uncharged metal carbonyls obeying or not the 18-electron rule and, last but not least, the giant metal carbonyl clusters. 

E. P. Kundig, M. Moskovits, and G. A. Ozin, Intermediate Binary Carbonyls of Palladium Pd(CO)n where n = 1-3 Preparation, Identification, and Diffusion Kinetics by Matrix Isolation Infrared Spectroscopy, Canad. J. Chem. 50, 3587-3593 (1972). 

Unlike Ni , Pd does not form a stable binary carbonyl. When PdCl2 is carbonylated in acetic anhydride, reduction takes place to give the polymeric [Pd(CO)Cl] . However, Pd(CO)4 can only be prepared and studied using matrix-isolation techniques. 

The hypothetical titanium heptacarbonyl Ti(CO)7 is as yet unknown. However, Ti(CO)6 has been prepared and characterized by matrix isolation techniques. Ti(CO)6, thus prepared, is found to be unstable above —200 °C Titanium carbonyls Ti(CO)x (x = 1 6) are formed during deposition of laser-ablated titanium atoms with CO during condensation in excess Ne, or on annealing and photolysis of the matrix. Titanium carbonyl complexes see Carbonyl Complexes of the Transition Metals) have also been observed spectroscopically (by IR) as intermediates in the low-temperature reaction of Ti(CH2Ph)4 and its dicyclohexylamine adduct with CO. ... 

One of the earliest studies of zero valent lanthanide metal vapor chemistry involved the matrix isolation reaction of the metals with CO (73, 74). Based on infrared data, a variety of zero valent metal carbonyl complexes, Ln(CO) , n = 1-6, were postulated. These were not preparative scale experiments, however, and the reaction products were not stable except at very low temperature. Hence, unambiguous confirmation of the formula and structure of these complexes could not be obtained. 

Condensation of metal vapors at low temperatures with CO allows the preparation or spectroscopic identification (matrix isolation technique) of several carbonylic species , e.g., Fe(CO)5, Ta(CO)s, etc., including the unstable carbonyls of the inner transition metals . Yields are low. 

Zero-valent palladium carbonyls without phosphine ligands have been prepared by two groups of workers by the technique of matrix isolation at low temperatures. Carbonyls of the type Pd(CO) (m = 1 to 4) have been characterized by IR spectra. Diffusion studies indicate that the lower carbonyls react readily with CO to give Pd(CO)4 (46, 167). 

Metal carbonyl preparations can be divided into three main categories (1) dry methods, i.e., reactions occurring in the absence of any inert or reactive liquid, (2) wet methods, i.e., reactions occurring in the presence of an inert or reactive liquid, and (3) matrix-isolation techniques (see 14.6.1.5), by which a vaporized metal reacts with CO usually at temperatures around 20K. The latter low T method has been widely used recently for unstable metal carbonyls, their presence being usually established by IR spectroscopy. Examples are Cu(CO)3, AlfCOj, Ti(CO)g, and Ag(CO)2. We deal here only with the conventional methods (1) and (2), which are used almost exclusively for the preparation of thermally stable metal carbonyls. 

The pale yellow [Ni(PEt3)4] is also tetrahedral but with some distortion. In sharp contrast to nickel, palladium forms no simple carbonyl, Pt(CO)4 is prepared only by matrix isolation at very low temperatures and reports of K4[M(CN)4] (M = Pd, Pt) may well refer to hydrido complexes in any event they are very unstable. The chemistry of these two metals in the zero oxidation state is in fact essentially that of their phosphine and arsine complexes and was initiated by L. Malatesta and his school in the 1950s. Compounds of the type [M(PR3)4], of which [Pt(PPh3)4] has been most thoroughly studied, are in general yellow, air-stable solids or liquids obtained by reducing complexes in H2O or H20/EtOH solutions with hydrazine or sodium borohydride. They are tetrahedral molecules whose most important property is their readiness to dissociate in solution to form... 

Nearly all of the compounds were synthesized photochemically, which imposes a nuihber of quite strict limitations on larger-scale preparations of these compounds. The primary, or at least the initial, identification of these compounds has been via IR spectroscopy. (The sole exception has been Cp Re(CO)(C2H4)2 which was tentatively identified by NMR [18]). In most cases, characterization of the metal carbonyl moiety via v(C-O) IR bands is definitive, particularly if library spectra are available from liquid noble gas [23] or matrix isolation experiments [4]. This is because these bands are sharp and intense, and the C-O stretching vibrations are largely uncoupled from other vibrations of the molecule. Furthermore, the precise wavenumber of the bands are extremely sensitive to the oxidation state of the metal center as illustrated, for example, by Kazarian et al. in their study of hydrogen bonding to metal centers [25]. 

Sander and co-workers [94ACIE2212] have reported the first preparative-scale synthesis of a dioxirane via carbene oxidation. Thus, the relatively stable dimesityldioxirane 106 was prepared by the low-temperature O2 oxidation of carbene 104 in matrix isolation in CFCI3 an intermediate carbonyl oxide (105) was observed spectroscopically. 

Despite the extreme air and moisture sensitivity of organo-lanthanoid compounds, this is a rapidly expanding research area. An exciting aspect of organolanthanoid chemistry is the number of efficient catalysts for organic transformations that have been discovered (see Box 25.5). In contrast to the extensive carbonyl chemistry of the J-block metals (see Sections 24.4 and 24.9), lanthanoid metals do not form complexes with CO under normal conditions. Unstable carbonyls such as Nd(CO)g have been prepared by matrix isolation. Since organolanthanoids are usually air- and moisture-sensitive and may be pyrophoric, handling the compounds under inert atmospheres is essential. ... 

Hurlburt, Anderson and Strauss have reported the preparation of the first isolable silver carbonyl (as opposed to matrix isolated species). Additionally, there has been developments in gold halo-carbonyl chemistry, with a number of new complexes reported... 

Nickel is the only metal to react directly with carbon monoxide at room temperature at an appreciable rate, although iron does so on heating under pressure. Cobalt affords HCo(CO)4 with a mixture of hydrogen and carbon monoxide (p. 387). In general, therefore, direct reaction does not provide a route to metal carbonyls. The metal atom technique (p. 313) has been used to prepare carbonyls of other metals in the laboratory e.g. Cr(CO)g, but it offers no advantages over the reduction method discussed below. When metal vapours are cocondensed with carbon monoxide in frozen noble gas matrices at very low temperatures (4-20K) the formation of carbonyl complexes is observed. These include compounds of metals which do not form any stable isolable derivatives e.g. Ti(CO), Nb(CO) and Ta(CO)g as well as Pd(C0)4 and Pt(C0)4. Vibrational spectra of the matrix show that coordinatively unsaturated species such as Ni(CO) n = 1-3) or Cr(CO) (n = 3-5) are also formed under these conditions. 

---