Formation of an alkyne during degradation of metal-alkylidyne complexes

Article abstract of DOI:10.1039/c3dt50921d

The compound [(Ot-Bu)₃WCt-Bu] (1) (t-Bu = C(CH₃) ₃) decomposes upon contact with water and several organic products are formed, including di-tert-butylacetylene, t-BuCCt-Bu. This process is reminiscent of the degradation of trinuclear metal-alkylidyne complexes in which free carbynes are ejected into solution, couple and form alkynes along with many other products. The reactivity pattern of the resulting t-BuC carbynes that includes extensive hydrogen abstraction, cleavage of alkynes and lack of reactivity towards alkenes is indicative of a quartet (S = 3/2) spin state. A similar spin state was assigned to other RC (R = alkyl) species that were produced by degrading M₃-alkylidyne (M = transition metal) complexes in water. t-BuCCt-Bu is also produced during thermal decomposition of solid 1. In 1977 Fischer and co-workers reported a very similar process in which solids of Br(CO)₄CrCR¹ and Br(CO)₄CrCR² were co-thermolyzed to produce R¹CCR², R ¹CCR¹, and R²CCR². Fischer had considered the involvement of free carbynes in the making of the alkynes but later resorted to other explanations. The current results suggest that his original proposal is indeed valid.

Full text of DOI:10.1039/c3dt50921d

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Formation of an alkyne during degradation of
metal–alkylidyne complexes†
Cite this: Dalton Trans., 2013, 42, 12248
Ophir Levy, Sanaa Musa and Avi Bino*
The compound [(Ot-Bu)3WuCt-Bu] (1) (t-Bu = C(CH3)3) decomposes upon contact with water and several
organic products are formed, including di-tert-butylacetylene, t-BuCuCt-Bu. This process is reminiscent
of the degradation of trinuclear metal–alkylidyne complexes in which free carbynes are ejected into
solution, couple and form alkynes along with many other products. The reactivity pattern of the resulting
t-BuC carbynes that includes extensive hydrogen abstraction, cleavage of alkynes and lack of reactivity
towards alkenes is indicative of a quartet (S = 3/2) spin state. A similar spin state was assigned to other
RC (R = alkyl) species that were produced by degrading M3–alkylidyne (M = transition metal) complexes
in water. t-BuCuCt-Bu is also produced during thermal decomposition of solid 1. In 1977 Fischer and
co-workers reported a very similar process in which solids of Br(CO)4CruCR1 and Br(CO)4CruCR2 were co-
thermolyzed to produce R1CuCR2, R1CuCR1, and R2CuCR2. Fischer had considered the involvement of
free carbynes in the making of the alkynes but later resorted to other explanations. The current results
suggest that his original proposal is indeed valid.
Received 7th April 2013,
Accepted 24th June 2013
DOI: 10.1039/c3dt50921d
www.rsc.org/dalton
community for the past 35 years. To date, this work has been
cited in the literature only ten times2 and in none of these
Introduction
In 1977, Fischer and coworkers published a groundbreaking publications a direct reference to the surprising findings has
paper entitled “Transition Metal Carbyne Complexes, XXXIII been made. Later, these results were linked without a clear
Thermal Cleavage of Carbyne Ligands from trans-Bromo-tetra- reason to a totally different process, namely alkyne metathesis,
carbonyl(organylcarbyne)chromium Complexes”.1 In this in which alkynes swap their substituents as in: 2R1CuCR2 ⇄
report they described experiments in which mono-nuclear R1CuCR1 + R2CuCR2.3a This reaction requires the breaking
chromium–carbyne complexes that contain a metal–carbon and forming of carbon–carbon triple bonds.3 It is widely
triple bond decompose, whether in the solid state or in solu- accepted that this action is a metal-assisted process that
tion, and produce alkynes in yields higher than 30%. They requires a metal center.4
also showed that a suspension of two compounds with
In 2005 we reported that a carbon–carbon triple bond is
different carbyne ligands produced a mixed alkyne under mild formed by decomposition of a bis(alkylidyne)tri-molybdenum
temperature (eqn (1)).
cluster in aqueous solution at room temperature.5 Recently, we
have shown by using isotope labeling and other experiments
L5Cr uCR1 þ L5CruCR2 ! R1CuCR2 þ R1CuCR1 þ R2CuCR2 that the formation of alkynes by decomposition of metallo-
alkylidyne complexes in water proceeds via the formation of
ðL ¼ carbonyl or halide; R ¼ alkyl or arylÞ
free carbynes. These extremely reactive species can couple
with each other and produce alkynes (2RC → RCuCR) or react
ð1Þ
with solvent molecules and other substrates and form a large
number of organic molecules of various chain lengths.6 We
have also demonstrated that the coupling of two carbynes pro-
ceeds via a conical intersection7a and may occur without any
involvement of metal centers.7 One of the complexes that have
been examined in our laboratory is the mono-nuclear tung-
sten–alkylidyne complex tris(2-methyl-2-propoxy)(2,2-dimethyl-
propylidyne)tungsten(VI), [(OCMe3)3WuCCMe3]4b,8 (1) (Me =
methyl), also known as “Schrock Alkyne Metathesis Catalyst”
(see Fig. 1). Schrock and coworkers reported in 1983 that the
main decomposition product of this compound in aqueous
Although the possibility of forming free carbynes followed
by coupling had been mentioned at the beginning of their
report, Fischer and his coworkers preferred to abandon such
an approach and resorted to an unexplained metal-assisted
mechanism. This is probably the reason why such important
results have remained largely ignored by the scientific
Institute of Chemistry, The Hebrew University of Jerusalem, Edmond J. Safra
Campus, Givat Ram, 91904 Jerusalem, Israel. E-mail: bino@mail.huji.ac.il
†Electronic supplementary information (ESI) available: Analytical techniques.
See DOI: 10.1039/c3dt50921d
12248 | Dalton Trans., 2013, 42, 12248–12251
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
which two [t-Bu–C]0 moieties, from two different metal centers
of complex 1, interact with each other either as free carbynes
in solution or by some unknown metal-assisted process. To
the best of our knowledge, there are no straightforward metal-
assisted mechanisms that describe the formation of alkynes
from two mono-nuclear metal–alkylidyne compounds,
especially from insoluble substances that degrade upon
contact with water. Aggregation of metal centers via oxo/
hydroxo bridges as a means for RC coupling is unreasonable
because the minimal W⋯W distance should be about 3.5 Å,
far too long to allow an encounter between alkylidyne carbon
atoms. All other products in Table 1 are the result of hydrogen
abstraction followed by coupling (entries 3 and 4), hydrogen
abstraction (entry 5), water reduction (entry 6) and oxygen and
hydrogen abstraction (entry 7).
The formation of such a diverse list of products correlates
with the presence of a very reactive species such as carbyne in
aqueous solution.6
Similar products were detected upon decomposition of
compound 2 in H2O. This reaction also leads to the formation
of the alkyne t-BuCuCt-Bu along with several other organic
molecules.11
Interestingly, in addition to the products in Table 1, we also
detected two more unexpected products, namely 2-butyne
and 4,4-dimethyl-2-pentyne (neopentyne). It is reasonable to
assume that the source of these compounds is a small amount
of the ethylidyne derivative, [(Ot-Bu)3WuCMe], that co-crystal-
lizes with 1. Compound 1 is commercially produced by react-
ing [(Ot-Bu)3WuW(Ot-Bu)3] with neopentyne, t-BuCuCMe,
Fig. 1 Structure of [(OCMe3)3WuCCMe3] (1). Green W; red O; gray C; purple
(CMe3).
media is an oxo-tungsten(VI)–neopentyl complex9 but did not
mention the formation of organic products in the reaction
except for 2,2-dimethylpropane (neopentane).9a
We have recently reported that the decomposition of 1 in
water leads to the formation of the alkyne t-BuCuCt-Bu (t-Bu =
CMe3) along with several other organic molecules.6 Here we
report details of these experiments in H2O or D2O along
with the results for the decomposition of 1 in the solid state.
We also report the decomposition products of a related
compound, tris(2,2,-dimethylpropyl)(2,2-dimethylpropylidyne)-
tungsten(VI), [(Np)3WuCCMe3]8,10 (2), and of the Fischer-type
analog, [Cl(CO)2(TMEDA)WuCCMe3] (3) (TMEDA = N,N,N′,N′-
tetramethylethylenediamine), in H2O. We believe that the for-
mation of alkynes in these systems, both in solution and in
the gas phase, proceeds via coupling reactions between free
carbynes as originally mentioned by Fischer and coworkers.
thus forming
a mixture of 1 and a small amount of
[(Ot-Bu)3WuCMe].12 Co-degradation of the two derivatives
should also produce methyl carbynes that couple in solution
and form 2-butyne or react with t-Bu–C species and form neo-
pentyne. A similar reaction was observed when ethylidyne and
propylidyne tri-molybdenum clusters co-degraded in solution,
producing CH3C and CH3CH2C carbynes. Subsequently, these
species coupled and formed 2-pentyne along with 2-butyne
and 3-hexyne.6 Decomposition of compound 2, however, does
not yield 2-butyne or neopentyne, probably because of the fact
that commercial 2 is not prepared by using neopentyne.8,10
The detachment of t-Bu–C carbynes from complex 1
suggests that the formal oxidation state of the tungsten atom
Results and discussion
Compound 1 is an air- and moisture-sensitive off-white crystal- becomes +3 with three d electrons. Since only colorless W(VI)
line solid. Upon contact with water, the insoluble material complexes are formed upon decomposition of 1 in H2O, we
decomposes and produces an insoluble white solid of an assume that W(III) intermediates are oxidized by the solvent to
oxo-tungsten(VI)–neopentyl complex in high yields. The produce an oxo/hydroxo tungsten(VI) complex and H2. The
organic compounds that are produced in this reaction whether presence of gaseous dihydrogen in the headspace of the reac-
in H2O or in D2O were analyzed in our laboratory by using gas tion vessel was experimentally substantiated in our laboratory
chromatography/mass spectrometry (GC/MS) and are listed in in support of the above assumption (see the ESI, Fig. S1†).
Table 1.
Carbynes can be formed in either their doublet ground
tert-Butanol (entry 1) is formed upon protonation of the state (S = 1/2) or quartet excited state (S = 3/2). It is known that
tert-butoxide ligands followed by detachment of the resulting doublet carbynes are very reactive toward alkenes, insert into
alcohol molecules from 1. The most intriguing product in this C–H bonds and are not prone to H-abstraction reactions.13 By
list is the alkyne 2,2,5,5-tetramethyl-3-hexyne (di-tert-butyl- contrast, the quartet excited state was reported to be long-
acetylene) (entry 2). Such a compound that contains a carbon– lived14 and to prefer reactions with radicals and H abstrac-
carbon triple bond can be formed by a coupling reaction in tion.14a,15 We have recently shown that carbynes that are
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 12248–12251 | 12249

View Article Online
Paper
Dalton Transactions
Table 1 Organic products of the decomposition reactions of 1 in H2O or D2Oa
Entry
1
Product
1 in H2O
1 in D2O
2-Methyl-2-propanol
2
3
4
5
6
2,2-5,5-Tetramethyl-3-hexyne
2,2-5,5-Tetramethyl-3-hexene
2,2-5,5-Tetramethyl-3-hexane
2,2-Dimethylpropane
2,2-Dimethylpropanal
7
2,2-Dimethylpropanol
a Since oxo-tungsten(VI)–neopentyl is produced in 82–90% yields,9 the overall yield of the products in the table is estimated to be ∼10%.
t-Bu–C carbyne with respect to CH3C, although both species
react with water and form the corresponding aldehyde.
Solid 1 undergoes thermal decomposition under dry nitro-
gen at about 45 °C forming di-tert-butylacetylene along with
other products.17 The process of alkyne formation from 1 is
Scheme 1 Reaction of tert-butyl carbyne with propyne in aqueous solution.
analogous to several other systems in which thermolysis of
metal–carbyne complexes such as [Br(CO)4CrCR] or [(CO)9Co3CR]
prepared by degradation of metal–alkylidyne complexes do not generates the corresponding alkynes.1,18 In order to examine
react with alkenes and abstract hydrogen atoms readily.6 We whether a Fischer-type WuCCMe3 system would also produce
have concluded that their chemical behavior is indicative of a an alkyne upon decomposition, we prepared the compound
quartet (S = 3/2) state, and showed by computational methods [Cl(CO)2(TMEDA)WuCCMe3] (3).19 Unlike Fischer’s obser-
that the formation of a carbon–carbon triple bond from two vations that tungsten analogs of [Br(CO)4CruCR], namely
quartet carbynes is a feasible process.7a We have also found [Br(CO)4WuCR], had not produced alkynes upon thermolysis,
another marker reaction for quartet carbynes such as methyl thermal treatment of compound 3 did produce the alkyne di-
carbyne, namely the interaction with alkynes: MeC + R1CuCR2 → tert-butylacetylene. This alkyne was also generated upon
MeCuCR1 + MeCuCR2 + R1C + R2C. In water, R1C and R2C vigorous stirring of solid 3 in slightly basic aqueous solution.
produce the corresponding carboxylic acids, R1COOH and This indicates that the nature of the precursor, whether
R2COOH.6 It is evident from the products listed in Table 1 that Schrock, Fischer or any other metal-carbyne type, such as ionic
t-Bu–C is extensively engaged in H-abstraction processes. In trinuclear alkylidyne complexes with intermediate oxidation
order to examine its reactivity towards alkynes we allowed com- states, does not play a role in the formation of the alkyne.
pound 1 to decompose in water in the presence of propyne, In all systems, degradation of the parent complex leads to free
HCuCCH3 (Scheme 1). The formation of t-BuCuCH and carbyne formation that, in turn, couple and form alkynes.
t-BuCuCMe, in addition to products from Table 1, clearly indi-
Fischer’s conclusion that free carbynes had not been
cates a reactivity pattern similar to that of the methyl carbyne.6 involved in the formation of alkynes was based on the fact
We have also found that t-Bu–C does not react with alkenes.16 that thermal treatment of Br(CO)4WuCR had not formed an
Hence, the overall reactivity of t-Bu–C that includes carbon– alkyne. In contrast, we have shown that compound 3, a similar
carbon triple bond formation through the coupling of two carb- Fischer-type tungsten–carbyne compound, does yield an
ynes, extensive hydrogen abstraction, cleavage of alkynes, and alkyne upon decomposition. Thus, Fischer’s original hypo-
lack of reactivity towards alkenes suggests a quartet spin state. thesis regarding the formation of free carbynes might be valid
However, the marker reaction of CH3C in water to form acetic after all. His experiment with mixed solids (eqn (1)) supports
acid does not occur with t-Bu–C. No 2,2-dimethyl-propanoic the assumption that in all the above reactions free carbynes
acid (pivalic acid) has been detected in the reaction mixture are ejected into either solution or the gas phase and sub-
of 1 in water. This is probably due to higher stability of the sequently generate alkynes by coupling. We intend to isolate
12250 | Dalton Trans., 2013, 42, 12248–12251
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
and investigate such species by the use of matrix-isolation
techniques.
Notes and references
1 E. O. Fischer, A. Ruhs and D. Plabst, Z. Naturforsch. Teil B,
1977, 32B, 802–804.
2 According to Thomson Reuters Web of Knowledge.
3 (a) X. Wu and M. Tamm, Beilstein J. Org. Chem., 2011, 7,
82–93; (b) F. Pennella, R. L. Banks and G. C. Bailey, Chem.
Commun., 1968, 1548–1549.
4 (a) T. J. Katz and J. McGinnis, J. Am. Chem. Soc., 1975, 97,
1592–1594; (b) J. H. Wengrovius, J. Sancho and R. R.
Schrock, J. Am. Chem. Soc., 1981, 103, 3932–3934.
5 A. Bino, M. Ardon and E. Shirman, Science, 2005, 308, 234–
235.
Experimental
Materials and sample preparation
Compounds 1 (98%) and 2 (98%) were purchased from Strem
Chemicals, Inc. and were shipped and kept refrigerated under
anaerobic conditions. Compound 3 was prepared according
to a literature procedure, using tert-butyllithium and oxalyl
chloride (∼98%).19
Decomposition of 1 or 2 in H2O or D2O
2 mL of H2O or D2O were thoroughly degassed in a stream of
argon. 2–7 mg of the solid sample (either 1 or 2) were
suspended in water in a sealed vial. After a few hours, a
headspace or a solution sample was collected and transferred
into the GC/MS. In the case of 2, the reactions were performed
only in H2O.
6 B. Bogoslavsky, O. Levy, A. Kotlyar, M. Salem, F. Gelman
and A. Bino, Angew. Chem., Int. Ed., 2012, 51, 90–94.
7 (a) D. Danovich, A. Bino and S. Shaik, J. Phys. Chem. Lett.,
2013, 4, 58–64; (b) O. Levy and A. Bino, Chem.–Eur. J., 2012,
18, 15944–15947.
8 R. R. Schrock, D. N. Clark, J. Sancho, J. H. Wengrovius,
S. M. Rocklage and S. F. Pedersen, Organometallics, 1982, 1,
1645–1651.
Decomposition of 1 in water in the presence of propyne
2 mL of H2O were thoroughly degassed in a stream of argon.
The vial was then flushed with propyne. 3–4 mg of 1 were sus-
pended in the propyne-saturated water. After a few hours, a
headspace sample was collected and transferred into the GC/MS.
9 (a) I. Feinstein-Jaffe, S. F. Pedersen and R. R. Schrock,
J. Am. Chem. Soc., 1983, 105, 7176–7177; (b) I. Feinstein-
Jaffe, J. C. Dewan and R. R. Schrock, Organometallics, 1985,
4, 1189–1193.
10 D. N. Clark and R. R. Schrock, J. Am. Chem. Soc., 1978, 100,
6774–6776.
Solid-state thermolysis of 1
3–4 mg of 1 were transferred in a sealed vial to a pre-heated 11 Table 1, entries 2–5 were detected.
water bath. The water temperature was set to 47 °C. After 1.5 h, 12 (a) M. L. Listemann and R. R. Schrock, Organometallics,
a headspace sample was collected and transferred into the
GC/MS.
1985, 4, 74–83; (b) R. R. Schrock, Acc. Chem. Res., 1986, 19,
342–348; (c) A. Fürstner and P. W. Davies, Chem. Commun.,
2005, 2307–2320.
Decomposition of 3 in basic aqueous solution
13 (a) B. P. Ruzsicska, A. Jodhan, H. K. J. Choi, O. P. Strausz
and T. N. Bell, J. Am. Chem. Soc., 1983, 105, 2489–2490;
(b) J.-C. Loison and A. Bergeat, Phys. Chem. Chem. Phys.,
2009, 11, 655–664.
14 (a) Z. Hou and K. D. Bayes, J. Phys. Chem., 1993, 97, 1896–
1900; (b) Z. Hou and K. D. Bayes, J. Phys. Chem., 1994, 98,
6324–6326.
2 mL of 0.1 M NaOH were thoroughly degassed in a stream of
argon. 3–4 mg of the solid sample were suspended in the
aqueous solution in a sealed vial. After a few hours, a headspace
sample was collected and transferred into the GC/MS.
Solid-state thermolysis of 3
3–4 mg of 3 were transferred in a sealed vial to a pre-heated 15 (a) O. P. Strausz, G. J. A. Kennepohl, F. X. Garneux,
water bath. Water temperature was set to 80 °C. After 3 h,
a headspace sample was collected and transferred into the
GC/MS.
T. DoMinh, B. Kim, S. Valenty and P. S. Skell, J. Am. Chem.
Soc., 1974, 96, 5723–5732; (b) C. B. Kellogg, E. Valeev,
J. M. Galbraith, J. E. Fowler and H. F. Schaefer III., Mol.
Phys., 1996, 89, 1695–1705.
Characterization methods
16 No apparent reactions with alkenes such as cis-2-hexene or
cyclohexene were observed.
Samples were analyzed by GC/MS. For further technical infor-
mation see the ESI.†
17 Table 1, entries 1–3 and 5 were detected along with butene,
2-butyne and neopentyne.
18 J. U. Khand, G. R. Knox, P. L. Pauson and W. E. Watts,
J. Organomet. Chem., 1974, 73, 383–388.
Acknowledgements
The authors thank B. Bogoslavsky and the E. Y. Tshuva group 19 (a) A. Mayr, G. A. McDermott and A. M. Dorries,
for technical assistance and advice. The authors also thank
Profs. A. Mayr and S. E. Biali and Dr. C. E. Strasser for fruitful
discussions.
Organometallics, 1985, 4, 608–610; (b) G. A. McDermott,
A. M. Dorries and A. Mayr, Organometallics, 1987, 6, 925–
931.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 12248–12251 | 12251