Substitution reactions of Fe and Ru complexes with liquid phosphite ligands in the absence of solvents

Article abstract of DOI:10.1016/j.inoche.2008.06.002

The substitution reactions of solid piano-stool Fe and Ru complexes have been carried out with liquid phosphite ligands that also act as 'solvents' for the reaction. It was observed that the solvent-free reaction occurs more slowly than the corresponding reaction in a solvent. Addition of a catalyst, [CpFe(CO)₂]₂, did result in a more rapid substitution reaction but again the reaction was slower in the absence of a solvent. The reagent Me₃NO · 2H₂O has been shown to be a good decarbonylating agent for the solvent-free reaction.

Full text of DOI:10.1016/j.inoche.2008.06.002

Inorganic Chemistry Communications 11 (2008) 1082–1084
Contents lists available at ScienceDirect
Inorganic Chemistry Communications
journal homepage: www.elsevier.com/locate/inoche
Substitution reactions of Fe and Ru complexes with liquid phosphite ligands in
the absence of solvents
1
*
Apollinaire Munyaneza, Muhammad D. Bala , Neil J. Coville
Molecular Sciences Institute, School of Chemistry, University of the Witwatersrand, 1 Jan Smuts Avenue, Johannesburg 2050, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
The substitution reactions of solid piano-stool Fe and Ru complexes have been carried out with liquid
phosphite ligands that also act as ‘solvents’ for the reaction. It was observed that the solvent-free reaction
occurs more slowly than the corresponding reaction in a solvent. Addition of a catalyst, [CpFe(CO)2]2, did
result in a more rapid substitution reaction but again the reaction was slower in the absence of a solvent.
The reagent Me3NO ꢀ 2H2O has been shown to be a good decarbonylating agent for the solvent-free
reaction.
Received 2 April 2008
Accepted 3 June 2008
Available online 7 June 2008
Keywords:
Substitution reactions
Iron
Ó 2008 Elsevier B.V. All rights reserved.
Liquid phosphites
Solventless chemistry
Extensive studies of many so-called ‘solid–solid’ (solventless)
reactions in organic [1] and organometallic [2] chemistry have
shown that the reactions occur in the melt phase. In the melt stud-
ies between solid reagents that have been reported to date the gen-
eral consensus is that solvent-free (solventless) reactions occur
more rapidly than reactions performed in solvents due to enhanced
concentration effects. This would suggest that reactions performed
between a solid and a liquid reactant would also occur more rap-
idly than when performed in a solvent provided again that the so-
lid reactant melted/dissolved under the reaction conditions used.
These reagents would also provide a single liquid phase for the
reaction.
To explore this possibility we have investigated the reaction be-
tween liquid phosphite reagents, P(OR0)3, and solid RCpM(CO)2I
reactants to give RCpM(Co)[P(OR0)3]I (M = Fe, Ru; R = H, Me and
R0 = CH3, C2H5, C6H5) (Scheme 1).
The reactions of metal complexes with phosphite ligands (many
of which exist as liquids at room temperature) have been widely
reported in a range of solvents [3–6]. For example, trialkylphosph-
ite ligands P(OR)3 (R = CH3, C2H5) have been shown to react with
R0CpM(CO)2X complexes in benzene and toluene as solvents to
produce the compounds R0CpM(CO)[P(OR)3]X (where R0 = H, Me,
COOMe, etc.; M = Fe, Ru and X = Cl, Br, I) in high yields, indicating
that the reaction is quite facile [7–10]. The solvent-free reactions
The progress of the solvent-free substitution reaction (Scheme
1), in the absence and presence of [CpFe(CO)2]2 as a catalyst, was
readily monitored by FTIR and 1H NMR spectroscopy. Results are
shown in Table 1. Although, these reactions were performed at var-
ious temperatures (50–80 °C), Table 1 only shows results at 80 °C.
The reaction is very slow to non-existent at the lower tempera-
tures. The products were isolated and fully characterised (see Sup-
plementary materials for details). As can be seen from Table 1, both
the catalyzed and uncatalyzed reactions led only to the formation
of neutral products CpFe(CO)[P(OR)3]I.
It is observed that the rate of the solventless reactions between
CpFe(CO)2I and phosphite ligands are slower when compared with
the reactions carried out in the presence of benzene as a solvent
[11]. For example the reaction between CpFe(CO)2I and P(OCH3)3
in refluxing benzene (ꢁ80 °C) yielded 60% CpFe(CO)[P(OCH3)3]I
after 5 min while under solventless conditions only 10% conversion
into the same product was obtained after 15 min at this tempera-
ture (80 °C).
Further, the reaction of a 1:1.1 ratio of CpFe(CO)2I:P(OMe)3 in
which the amount of solvent, toluene, was varied (0 ml, 1 ml,
5 ml, 10 ml) revealed that the reaction rate increased with increas-
ing volume of solvent. Thus, after 15 min of reaction at 80 °C, only
5% conversion was obtained when no solvent was used (0 ml), 12%
conversion for 1 ml of toluene, 18% conversion for 5 ml of toluene
and 20% conversion for 10 ml toluene.
In contrast, the solid–solid solvent-free reactions between
CpFe(CO)2I and phosphine ligands showed improved rates and con-
versions when compared to the products obtained from the same
reaction in benzene as a solvent [12].
have also been performed in the presence of
[CpFe(CO)2]2.
a catalyst,
* Corresponding author. Tel.: +27 11 717 6738; fax: +27 11 717 6749.
E-mail address: neil.coville@chem.wits.ac.za (N.J. Coville).
Present address: School of Chemistry, University of KwaZulu-Natal, Private Bag
X54001, Westville Campus, Durban 4000, South Africa.
1
This was further confirmed by reacting CpFe(CO)2I with a mix-
ture of both PPh3 and P(OMe)3 in a 1:1:1 and a 1:1:10 molar ratio
1387-7003/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.inoche.2008.06.002

A. Munyaneza et al. / Inorganic Chemistry Communications 11 (2008) 1082–1084
1083
P(OC2H5)3 but with P(OC6H5)3, MeCpFe(CO)[P(OC6H5)]I, was
obtained in 20% yield as detected by 1H NMR spectroscopy. In
contrast, in refluxing benzene, the catalyzed reaction between
MeCpFe(CO)2I with P(OMe)3 easily afforded the product
MeCpFe(CO)[P(OMe)3]I [17]. Similar solvent-free (melt) studies
with solid phosphines were observed to be accelerated when an
electron donating substituent is attached to the Cp ligand,
although the product obtained was the salt product, [MeCpFe(CO)2
[P(OR)3]I [12].
R
R
+
P(OR')3
M
I
P(OR')3
M
CO
CO
CO
I
M = Fe, Ru; R = H, Me and R' = CH3, C2H5, C6H5
Scheme 1. Reactions between RCpM(CO)2I and phosphite ligands.
This decrease in reaction rates with solid–liquid reactants was
further confirmed when the central metal was changed from Fe
to Ru. As anticipated, the solventless reactions between CpRu(CO)2I
and P(OR)3 ligands (R = CH3, C2H5, C6H5) did not occur at 80 °C in
the presence or absence of [CpFe(CO)2]2 as a catalyst after 1 h. Even
at a higher temperature of 100 °C and in the presence of the iron
dimer as catalyst, the conversion into the products, CpRu(CO)[-
P(OR)3]I, was less than 20% after 1 h of reaction. As a comparison,
the same reaction in refluxing toluene as solvent was complete
in less than 45 min [18].
Table 1
Solventless reactions between C5H5Fe(CO)2I and P(OR)3 to yield C5H5Fe(CO)P(OR)3I at
80 °C
Ligand
Time
(min)
Conversion to producta (%)
No
catalyst
[CpFe(CO)2]2
catalysed
Me3NO ꢀ 2H2O
catalysed
P(OCH)3
P(OC2H5)
P(OC6H5)
a
15
75
135
0
0
4
10
14
15
100
100
100
Thus, all the solventless reactions between RCpM(CO)2I (R = H,
Me; M = Fe, Ru) and P(OR)3 ligands studied are slower than when
performed in solvents, even in the presence of [CpFe(CO)2]2 as a
catalyst.
15
75
135
5
8
17
15
18
20
15
75
135
9
16
17
31
38
42
Trimethylamine N-oxide has been demonstrated to be a useful
decarbonylating agent in CO substitution reactions of RCpM(CO)2I
complexes. These reactions (and indeed all reactions involving
Me3NO) have previously only been performed in solvents [19,20].
Result of the influence of the addition of Me3NO ꢀ 2H2O to the
solvent-free solid/liquid reactions is presented in Table 1. It can
be seen that at 80 °C in the presence of Me3NO ꢀ 2H2O, complete
reaction was obtained for all the ligands within 15 min of reaction,
as confirmed by IR spectroscopy. After isolation, the obtained prod-
ucts have been confirmed by spectroscopic studies to be
RCpM(CO)[P(OR0)3]I. The spectroscopic data (see Supplementary
materials) obtained via the solvent-free conditions are in agree-
ment with the characterization of the same products obtained ear-
lier by solvent procedures [9,11,17,18]. The solvent-free reactions
in the presence of Me3NO ꢀ 2H2O are much faster than reactions
carried out in the presence of solvent, e.g. reactions between
CpRu(CO)2X (X = Cl, Br, I) and phosphites were complete after 2 h
in xylene [21].
Based on 1H NMR data analysis.
in the absence of solvent (at 80 °C). After 15 min reaction, 1H NMR
spectroscopy data obtained by dissolving the product in CDCl3
indicated the formation of CpFe(CO)[P(OMe)3]I (<5%) and CpFe(-
CO)(PPh3)I (45%) at the 1:1:1 ratio. In the presence of excess phos-
phite (10 times), the CpFe(CO)(PPh3)I conversion was only 9%
while the CpFe(CO)[P(OMe)3]I content was still <5%. The liquid
phosphite ligand clearly inhibits the conversion reaction.
Similar results were obtained when the reaction was performed
in the presence of a catalyst, [CpFe(CO)2]2. The catalyst [CpFe(-
CO)2]2 was soluble in all phosphites studied.
Thus a solvent-free reaction does not always benefit from in-
creased localised concentration (and contact) effects between reac-
tants. Therefore, other factors have to be considered when
interpreting these reactions.
The steric [13] and electronic effects [14,15] associated with the
phosphite and phosphine ligands are not expected to play a major
role in the reaction as earlier studies have shown that the reaction
occurs by a SN1 mechanism [16]. Thus, the properties of these li-
gands acting as solvents need to be considered.
The reaction between Me3NO and metal carbonyl complexes is
known to involve reaction of the bound CO with the Me3NO
[19,20]. Thus, the ability to generate the CO substitution products
under solventless conditions suggests that the phosphite ligand
properties are not important here and the solubility of CO in the
phosphite plays no role in this reaction.
In conclusion, the solventless reactions between RCpM(CO)2I
(R = H, Me; M = Fe, Ru) and three liquid phosphite ligands, L
[L = P(OCH3)3, P(OC2H5)3, P(OC6H5)3] have been attempted. The sol-
ventless reactions proceeded more slowly when compared to the
reactions carried out in solvents such as toluene. Even the addition
of a catalyst such as [CpFe(CO)2]2 did not show any significant
improvement in the rate of the reaction. However, using solvent-
free conditions, the reagent trimethylamine N-oxide has success-
fully been used to rapidly afford 100% conversion of reactants into
products. These sets of experiments therefore show that the in-
creased rates observed in solvent-free reactions are not only due
to the increased concentration of reactants in direct contact with
each other. Also important are the solvent properties, i.e. the
chemical nature of the reactants. Thus, even though the phosphites
used in this study are liquids and hence should have a better ability
to ‘mix’ with solid reactants as compared to corresponding solid
phosphines, generally the solvent-free substitution reaction of
the phosphines are much faster than those of the corresponding
phosphites.
The CO substitution reaction in solvents has been proposed to
occur via a SN1 mechanism suggesting that CO dissociation is the
rate determining step. This would indicate that the CO dissociation
step is modified in the liquid phosphite. The solubility of CO in the
phosphites is suggested to be higher than in the solvents (or phos-
phines) used (i.e. the CO is not easily substituted because it does
not escape rapidly from the ‘solution’). This implies that the reac-
tion would be inhibited by the phosphite ‘solvent’ as the step to
regenerate starting material would be more favoured. Currently
no data exists to substantiate this proposal, but reactions with
Me3NO are consistent with this suggestion (see below).
The effect of having an electron donating group on the Cp ligand
was also studied. Thus, a series of solvent-free studies were con-
ducted with MeCpFe(CO)2I and the three liquid phosphite ligands
at 80 °C in the absence or presence of [CpFe(CO)2]2 as a catalyst.
In the absence of catalyst, no reaction occurred after 1 h for all
the ligands studied. Even after 1 h in the presence of a catalyst,
no reaction was observed by IR spectroscopy for P(OCH3)3 and

1084
A. Munyaneza et al. / Inorganic Chemistry Communications 11 (2008) 1082–1084
[4] Y. Sun, H.-S. Chan, P.H. Dixneuf, Z. Xie, J. Organomet. Chem. 691 (2006) 3071.
The solid–liquid study has further demonstrated the impact of
solventless reactions in organometallic chemistry.
[5] M. Diéguez, O. Pámies, C. Claver, Tetrahedron: Asymmetry 15 (2004) 2113.
[6] M.T. Reetz, P. Maiwald, C.R. Chim. 5 (2002) 341.
[7] R.J. Haines, A.L. Du Preez, L.L. Marais, J. Organomet. Chem. 24 (1970) C26.
[8] R.J. Haines, A.L. Du Preez, I.L. Marais, J. Organomet. Chem. 28 (1971) 405.
[9] N.J. Coville, M.S. Loonat, D. White, L. Carlton, Organometallics 11 (1992) 1082.
[10] R.J. Haines, I.L. Marais, C.R. Nolte, Chem. Commun. (1970) 547.
[11] N.J. Coville, E.A. Darling, A.W. Hearn, P. Johnston, J. Organomet. Chem. 328
(1987) 375.
[12] A. Munyaneza, M.D. Bala, N.J. Coville, Z. Naturforsch. 62b (2007) 453.
[13] C.A. Tolman, Chem. Rev. 77 (1977) 313.
[14] P.B. Dias, M.E. Minas de Piedade, J.A. Martinho Simoꢀes, Coord. Chem. Rev. 135/
136 (1994) 737.
Acknowledgements
We thank the NRF and the University of the Witwatersrand for
financial support.
Appendix A. Supplementary material
[15] J.C. Toledo, B.S. Lima Neto, D.W. Franco, Coord. Chem. Rev. 249 (2005) 419.
[16] J.D. Atwood, Inorganic and Organometallic Reaction Mechanism, second ed.,
Wiley-VCH, New York, 1997.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.inoche.2008.06.002.
[17] P. Johnston, M.S. Loonat, W.L. Ingham, L. Carlton, N.J. Coville, Organometallics
6 (1987) 2121.
[18] N.J. Coville, E.A. Darling, J. Organomet. Chem. 277 (1984) 105.
[19] M.O. Albers, N.J. Coville, Coord. Chem. Rev. 53 (1984) 227.
[20] T.Y. Luh, Coord. Chem. Rev. 60 (1984) 255.
References
[1] G. Rothenberg, A.P. Downie, C.L. Raston, J.L. Scott, J. Am. Chem. Soc. 123 (2001)
8701.
[21] D.A. Brown, H.J. Lyons, R.T. Sane, Inorg. Chim. Acta 4 (1970) 621.
[2] M.D. Bala, N.J. Coville, J. Organomet. Chem. 692 (2007) 709.
[3] I.S. Gonçalves, C.C.L. Pereira, C.A. Gamelas, A. Hazell, C.C. Romão, J. Organomet.
Chem. 692 (2007) 1593.