Schlenk Line
Schlenk Line
Schlenk Line
CoVer Essay
Alkyl and Aryl Derivatives of the Alkali Metals: Strong Bases and
Reactive Nucleophiles. 2. Wilhelm Schlenk’s Organoalkali-Metal
Chemistry. The Metal Displacement and the Transmetalation
Reactions. Metalation of Weakly Acidic Hydrocarbons. Superbases
Dietmar Seyferth
Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
I. Introduction its two components could not be separated. The new field of
organometallic chemistry was a very active one in the 1850s,
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
1
InPart 1 of this essay, the preparation of organoalkali-metal after Frankland had reported the preparation of dimethylzinc
compounds by reaction of organic halides with alkali metals in 1849. Among the chemists of the time who had begun
and with organoalkali-metal compounds was discussed. In Part research in organometallic chemistry was George Bowdler
2, the preparation of organoalkali-metal compounds by the metal Buckton in England.2 Buckton had reported his preparations of
Downloaded via 181.55.20.70 on November 18, 2020 at 20:22:35 (UTC).
displacement reaction (eq A) and by transmetalation (eq B) is ethyl derivatives of mercury, tin, and lead by the action of
considered. diethylzinc on halides of these metals in 1859.3 Having prepared
diethylmercury, which he wrote as HgC4H5 (using Berzelius’
RnM′ + nM f nRM + M′ (A) flawed atomic weight scale in which C ) 6 and O ) 8), Buckton
RnM′ + nR′M f nRM + R′nM′ (B) was led to wonder:3b,d
“The electro-negative character of the group Cn2Hn2+1 in
M ) alkali metal; M′ ) another metal, usually a heavy the class of organo-metals to which zincethyl belongs, may
metal (e.g. Hg, Sn, Pb) now perhaps be considered as established. Some interest,
nevertheless, attaches to the question whether sodium is
This account includes a discussion of the seminal work in this capable of displacing ethyl from mercuric ethyl. An answer
area of Wilhelm Schlenk. In view of the importance of his to this question would give us some means of judging the
pioneering contributions, I felt that a reasonably thorough position of ethyl, as regards its electro-negative function
discussion of all of his research in the broad area of organoalkali- towards the true metals.”
metal chemistry was warranted. Toward the end of Schlenk’s
research activities, Karl Ziegler came on the scene and a not Therefore, he carried out the experiment and described its
overly friendly interaction developed between the two when outcome (but without experimental details):3b,d
Ziegler entered the organoalkali-metal field. This required a “At ordinary temperatures, sodium has only a slow action
discussion of Ziegler’s work, and so the essay grew longer than on mercuric ethyl, but after the lapse of a few hours a
I had expected when I started my writing. Add to that the last voluminous grey sponge is formed, whilst the liquid
section on organolithium-derived superbases, the exciting more entirely disappears. This sponge-like body has the property
recent development in organoalkali-metal chemistry, and Part of spontaneous combustibility in a marked degree, and is
2 ended up quite long, with no room left for a discussion of liable to explosion from apparently very slight causes. By
structures and bonding, which I had originally intended to the application of a gentle heat, a strong rush of gaseous
include. matter is evolved which eudiometric experiments proved
to be a mixture of ethylene and hydride of ethyl, obviously
II. The Preparation of Organoalkali-Metal proceeding from the disintegration of a double molecule
Compounds by the Metal Displacement Reaction. of ethyl.”
George Bowdler Buckton, 1859 Buckton correctly interpreted his observations:
In Section I of Part 1 of this essay,1 the attempts by J. A. “...probably, perhaps, sodiumethyl is first formed in the
Wanklyn to prepare sodium alkyls were described. These reaction, and then decomposed by heat
involved the use of the metal displacement reaction for the
first time (eq 1), but the sodium alkyl, a strong, highly polar HgC4H5 + Na ) NaC4H5 + Hg
(C2H5)2Zn + 2Na f 2C2H5Na + Zn 2(NaC4H5) + 2Hg ) 2 NaHg + C4H4+ C4H5H
(1)
excess
The mercury is supposed here to be inert, and in no way
nucleophile and base, reacted with the excess of diethylzinc, a to determine the decomposition.”
Lewis acid, to give an adduct, the zincate complex NaZn(C2H5)3.
This compound was much more reactive than diethylzinc, about (2) For a picture and biographical sketch of Buckton, see an earlier
as reactive as one might have expected ethylsodium to be, but historical essay: ; Seyferth, D. Organometallics 2003, 22, 2346.
(3) (a) Buckton, G. B. Proc. R. Soc. London 1859, 9, 309. (b) Buckton,
G. B. Proc. R. Soc. London 1859, 9, 685. (c) Buckton, G. B. Ann. 1859,
(1) Seyferth, D. Organometallics 2006, 25, 2. 109, 122, 222. (d) Buckton, G. B. Ann. 1859, 112, 220.
(4) This chemistry has been discussed in Part 1 of this essay.1 (6) Schorigin, P. Ber. Dtsch. Chem. Ges. 1908, 41, 2717.
(5) Acree, S. F. Am. Chem. J 1903, 29, 588. (7) Schorigin, P. Ber. Dtsch. Chem. Ges. 1908, 41, 2723.
4 Organometallics, Vol. 28, No. 1, 2009
formed. After several days beautiful, deep blue crystals formed Acree investigated the benzophenone/sodium reaction a few
which on drying became lighter in color. This product, in years later.5 Following the views of Nef on reactions of sodium
solution or as the solid, was extremely air- and moisture- with organic carbonyl compounds,25 he described the initial
sensitive, which made its isolation as a pure material difficult. product as the disodium species 4 and said it reacted with a
The blue solutions were decolorized almost instantaneously on second molecule of benzophenone to give sodium-benzopinacol
exposure to air, and bulk samples of the blue solid oxidized so (5). This, he said, is marginally stable at room temperature,
rapidly in air that the whole solid mass began to glow. Careful regenerating 4 and benzophenone, writing eqs 8 and 9 (in the
drying of the hydrogen and diethyl ether was imperative, and Nef style25). In his experiment, Acree obtained the blue ether
the apparatus shown in Figure 5 served in the preparation of solution and blue solid but did not isolate and analyze the latter.
the alkali-metal/diaryl ketone products. The reaction was carried Instead he poured it into water. Beckmann and Paul’s formula-
out in the sealed separatory funnel. Upon its completion, the tion of the (C6H5)2CO/Na product as 3 Acree considered “very
unreacted sodium remained behind while the ether suspension improbable”.
of the product which had formed was drawn off through the
wide-bore stopcock into the cylindrical vessel F, in which an
atmosphere of dry hydrogen was maintained. The product was
retained on filter paper on the porcelain filter plate C and dried
under a stream of hydrogen. Flask G contained ether, which
was forced under hydrogen pressure into vessel F to wash the
product collected on the filter. According to a sodium analysis,
the product contained 1 g atom of sodium and 1 formula weight
of benzophenone. The hydrolysis of an ether solution of the
blue product gave, depending on reaction conditions, either a
mixture of benzopinacol, (C6H5)2(HO)C-C(OH)(C6H5)2, ben-
zophenone, and benzhydrol, (C6H5)2CHOH, or a mixture of only Schlenk had become interested in Beckmann and Paul’s
benzophenone and benzhydrol. Beckmann concluded that the paper, which reported such extraordinary results, because it led
blue product was 3, writing its formation as shown in eq 7 on
(25) Nef, J. U. Justus Liebigs Ann. Chem. 1899, 308, 283.
8 Organometallics, Vol. 28, No. 1, 2009
Scheme 3 On reaction of the blue product with iodine, the ketone and
sodium iodide were the products. Hydrolysis of the sodium
bis(biphenylyl) ketone product gave a mixture of the ketone
and (C6H5-C6H4)2CHOH but no pinacol product. Of special
interest was the reaction of benzopinacol with freshly prepared
1.5% sodium amalgam (but not with sodium alone) in anhydrous
ether at room temperature in the absence of air, which proceeded
with rapid evolution of H2 to give a deep blue solution of what
proved to be the sodium-benzophenone product of Beckmann
and Paul. At first sight, Schlenk said, this result might imply
that the blue solid is, as Acree had claimed, the disodium
derivative of benzopinacol 5. The reactions of the blue product
with oxygen and iodine then could be written as shown in eqs10
and 11. However, the intense colors of the alkali-metal-diaryl
ketone compounds, their extremely facile oxidation and their
him to suspect that a new class of compounds of trivalent carbon
was involved. His first publication on his work in this area was
a preliminary communication without experimental details,26
written, as he said, “in order to ensure us a leisurely continued
investigation of this subject.” To make this possible, in his last
sentence he requested his chemical colleagues for the time being
to leave this research area to him. As was customary in those
days, nobody intruded. Times have changed!
Substituting 4-biphenylyl groups for phenyl had served
Schlenk well in his study of the triarylmethyl radicals, and he hydrolysis, which usually proceeded with C-C bond rupture,
used the same approach in his investigation of alkali- spoke against such a “normal” formulation. Schlenk also
metal-diaryl ketone products, carrying out his experiments with considered the formulation of Beckmann and Paul improbable
bis(4-biphenylyl) ketone, (C6H5-C6H4)2CdO, and 4-biphenylyl in view of the benzopinacol/sodium-benzophenone product.
phenyl ketone, (C6H5-C6H4)(C6H5)CdO. This turned out to He concluded that all the observed chemistry was best explained
bring important experimental advantages. The products obtained in terms of a trivalent carbon derivative, one with reactivity
in the reactions of the sodium-benzophenone product with similar to that of the triarylmethyl radicals, and he favored
water and carbon dioxide were difficult to crystallize and formula 6 for the alkali-metal-diaryl ketone products.
separate. Substituting one or both of the phenyl groups of In the full paper which followed in 1913,27 the important
benzophenone by biphenylyl groups alleviated these difficulties. question concerning the molecular weight of the alkali-
In addition to sodium, potassium also was used in the study. A metal-diaryl ketone products was answered. A normal ebul-
reaction of potassium with (C6H5C6H4)(C6H5)CdO gave a green lioscopic molecular weight determination was not possible
product with a 1:1 ketone/potassium ratio (Anal. Found: K, because of the poor solubility of such products, but the solubility
13.14, 13.49. Calcd: K, 13.13). A molecular weight determi- of the potassium-biphenylyl phenyl ketone product in diethyl
nation of the blue sodium-derived product was of crucial ether immediately following its preparation was sufficient to
importance in order to distinguish between a trivalent, mono- allow the following indirect procedure. A solution of the ketone
meric species, 6, that Schlenk had envisioned and a dimeric in well-dried diethyl ether in a boiling point elevation apparatus
under a purified nitrogen atmosphere was heated to the boiling
point. When this had remained constant, some freshly cut (under
ether) shiny small pieces of potassium were added. The green
potassium-ketone product formed in solution, without precipi-
species of type 3. However, the deep blue solutions of the alkali- tation. During this process the boiling point remained constant,
metal-diaryl ketone products were not true solutions but which showed that the number of molecules in the solution had
contained the blue solids in colloidal or in polymeric (“hoch- not changed. It follows that each molecule of ketone had formed
komplex,” Schlenk said) forms. Thus, cryoscopic molecular one molecule of the potassium-ketone product and the latter
weight determinations were not possible. For this reason, was monomeric. Thus, Schlenk’s original formulation,
Schlenk in this early work resorted to chemical studies. A (C6H5-C6H4)(C6H5)COK in this case, was correct. In a control
notable similarity between the triarylmethyl radicals and the experiment, the boiling point of a sample of diethyl ether that
alkali-metal-diaryl ketone products was their extremely ready contained some clean pieces of potassium was determined.
susceptibility to oxidation. The facility with which an ether Subsequently a weighed quantity of the ketone was added. In
solution of the blue sodium-benzophenone product was de- spite of the rather rapid formation of the potassium-ketone
colorized when exposed to air had been noted by Beckmann product, the molecular weight of the ketone could be determined
and Paul, but the reaction was not examined further. Schlenk from the boiling point elevation.
and Weickel found that air oxidation of the green sodium Also telling was the formation of the sodium-benzophenone
bis(biphenylyl)ketone product in ether resulted in immediate product by the reaction of a concentrated alcoholic solution of
decolorization and precipitation of a voluminous white solid sodium ethoxide with a suspension of benzopinacol in benzene,
which was found to be a mixture of (C6H5-C6H4)2CdO and a reaction that involved C-C cleavage to give the radical species
sodium peroxide. The latter could be washed out with water. (Scheme 4).
(26) Schlenk, W.; Weickel, T. Ber. Dtsch. Chem. Ges. 1911, 44, 1182. (27) Schlenk, W.; Thal, A. Ber. Dtsch. Chem. Ges. 1913, 46, 2840.
Organometallics, Vol. 28, No. 1, 2009 9
Scheme 4
compound, also a bright red solid, was even more reactive than
its triphenylmethyl analogue; it attacked pyridine and diethyl
ether and exhibited the usual extreme reactivity toward oxygen
and moisture. Its separation from the reaction mixture that also
contained suspended NaCl and excess [(CH3)4N]Cl was made
possible by the fact that it settled more slowly than the other In 1914 Schlenk published his first results of studies in a
two species did in petroleum ether (“Gasolin” Schlenk called new area of organoalkali-metal chemistry. The surprising
it) and so could be decanted from the heavier salts. reactions of sodium and potassium with diaryl ketones (vide
Pentavalent nitrogen compounds were a subject of interest supra) led him to consider the possibility of alkali-metal
and represented a challenge to later investigators. In 1926 Hager reactions with other organic compound classes that contained
and Marvel carried out reactions of methyl-, ethyl-, isoamyl, double bonds. An experimental investigation of the reactions
n-butyl-, and n-heptyllithium with a variety of tetraalkylam- of sodium with unsaturated compounds that contained CdC,
monium bromides and iodides in an attempt to prepare a CdN, and NdN double bonds followed.39
pentaalkylnitrogen compound.34 In all reactions only trialky- The reactions were carried out using sodium powder and well-
lamines were isolated from the reaction mixtures. None of the dried diethyl ether under a dry, oxygen-free nitrogen atmosphere
amines contained the alkyl groups of the alkyllithium used. Had in sealed tubes of the type shown in Figure 6. The sealed tubes
these workers studied these reactions in greater detail using a were shaken on a mechanical shaker at room temperature. The
higher-boiling solvent rather than petroleum ether, they would reaction times required were quite variable, from almost
have discovered the β-elimination-based olefin synthesis which instantaneous to several days. In general, the most reactive
Wittig reported over 30 years later.35 In one example, the unsaturated hydrocarbons were those in which the double bonds
reaction of n-heptyllithium with tetra-n-butylammonium iodide, were conjugated with one or more aryl groups. If the products
the hydrocarbon products should have been n-heptane from the were soluble in the ether medium, they could be isolated in
deprotonation step and butene-1 from the β-elimination reaction, pure form. If they were insoluble, they could be separated from
in addition to the tri-n-butylamine which was the only identified the excess sodium powder, but not in high purity, by virtue of
product.36 Twenty years later Georg Wittig also gave the the fact that usually the heavier sodium powder settled to the
preparation of a pentavalent nitrogen compound a try.37 In an bottom of the flask. The lighter product powder settled on top
attempt to prepare C6H5N(CH3)4, a solution of phenyllithium of the sodium and could be rinsed off with solvent.
was added under nitrogen to a suspension of [(CH3)4N]Cl in Sodium additions to hydrocarbons containing CdC bonds
ether and the reaction mixture was mechanically shaken for took several different courses. In the case of stilbene a slow
several days. The first experiments did not lead to an under- reaction resulted in formation of a dark brown-violet powder
standing of the reaction which had occurred. However, when that was insoluble in diethyl ether. Hydrolysis of the product
benzophenone was added to the reaction mixture and the product gave 1,2-diphenylethane. Schlenk concluded that it was the
was identified as [(C6H5)2C(OH)CH2N(CH3)3]Cl, it became clear disodium derivative which he wrote as 8 (eq 15) not as the
that phenyllithium had deprotonated tetramethylammonium dianionic compound, 9, as we would write it today.
chloride (Scheme 6). It was not clear how or if the LiCl was
bonded to the (CH3)3N-CH2 species which Wittig called
trimethylammonium methylide, an “ylide”. This question was
tackled by later workers. In any case, there was no evidence
for pentavalent nitrogen. Extension of the chemistry to phos- Tetraphenylethylene also gave a disodium derivative, 10, but
phonium salts by Wittig led to (C6H5)3PdCH2 and from that this was formed almost instantaneously, the result, as Schlenk
followed the Wittig olefin synthesis and the Nobel Prize in said, of the presence of four phenyl groups. Hydrolysis of the
Chemistry, awarded to Wittig in 1979.
After this diversion into pentavalent nitrogen (which does
involve organoalkali-metal chemistry), we return briefly to
Schlenk’s triphenylmethylsodium chemistry. Schlenk and Ochs
investigated the reactions of triphenylmethylsodium with organic
carbonyl compounds.38 Carbonyl group additions of (C6H5)3CNa
(34) Hager, F. D.; Marvel, C. S. J. Am. Chem. Soc. 1926, 48, 2689. dark red ether solution of 10 gave tetraphenylethane. In the belief
(35) (a) Wittig, G.; Polster, R. A. Justus Liebigs Ann. Chem. 1956, 599, that these reactions involved sodium addition to the CdC bonds
13. (b) Wittig, G.; Polster, R. A. Justus Liebigs Ann. Chem. 1958, 612, as shown, Schlenk expressed surprise that such a reaction
102. occurred with tetraphenylethylene. After all, he noted, this olefin
(36) In the one experiment in which Hager and Marvel identified the reacted with chlorine to give a rather unstable tetraphenyldi-
hydrocarbon products, those from the reaction of [Et4N]Br with EtLi in
“heavy paraffin”, the observed formation of C2H6 and C2H4 in a roughly chloroethane and did not react with bromine at all. The idea
1:1 ratio was rationalized in terms of disproportionation of ethyl radicals that it was the sodium electrons that added to the LUMO of
from the decomposition of an assumed pentaethylnitrogen. This, of course, the CdC bond with concomitant formation of a solvated sodium
also is the result of the deprotonation/elimination sequence, which actually ion was one that would come only some years in the future.
had occurred.
(37) Wittig, G.; Wetterling, M.-H. Justus Liebigs Ann. Chem. 1946, 556,
193. (39) Schlenk, W.; Appenrodt, J.; Michael, A.; Thal, A. Ber. Dtsch. Chem.
(38) Schlenk, W.; Ochs, R. Ber. Dtsch. Chem. Ges. 1916, 49, 608. Ges. 1914, 47, 473.
12 Organometallics, Vol. 28, No. 1, 2009
Scheme 7
(40) Formulas 12 and 13 are reproduced as they appear in ref 39. Here (41) This may have been the first synthesis of polystyrene by anionic
also the benzene rings were not written as Kekulé structures with double polymerzation.
bonds. (42) Schlubach, H. Ber. Dtsch. Chem. Ges. 1915, 48, 12.
Organometallics, Vol. 28, No. 1, 2009 13
Scheme 8 Scheme 9
the reaction of metallic lithium with methyl chloride in diethyl doubt unwelcome entrance of Karl Ziegler51 into both of his
ether, as had been described by Ziegler et al.48 major areas of interest, trivalent carbon compounds (i.e.,
As noted above, ethyllithium was soluble in warm hydro- persistent organic free radicals) and organoalkali metal chem-
carbon solvents and was generally more tractable than meth- istry, when Ziegler began a vigorous pursuit of independent
yllithium. Its preparation thus was easier. Finely cut pieces of research in the early 1920s. Since Ziegler’s work impacted on
metallic lithium (in excess) were added, under nitrogen, to a Schlenk’s research and led to an active, not very friendly
solution of diethylmercury in benzene, and the resulting mixture controversy between them, an account of Ziegler’s research in
was heated at 65 °C with periodic shaking, for 3 days. The the area of organoalkali-metal chemistry must be included in
excess lithium, coated with lithium amalgam, sank to the bottom this essay.
of the flask. Filtration of the mixture at 70 °C (Figure 7) gave In his first full paper on organic radicals Ziegler reported an
a filtrate from which, as it cooled, large colorless crystals of interesting alkali-metal reaction (eq 21).52 The reaction was
ethyllithum crystallized. The crystallization was completed by believed to be one of the tetraphenylallyl radical, generated by
cooling to 0 °C. A wash with a small amount of cold petroleum some trace radical species, so that the postulated sequence was
ether gave pure C2H5Li (Anal. Calcd: Li, 19.44. Found: Li, that shown in eqs 22-24. This led Ziegler to a new synthesis
19.10) which melted at 95 °C under nitrogen. It could be distilled of organic radicals by interrupting the sequence of reactions at
at high temperatures, in part without decomposition. Like eq 23, in which a solution of the deeply colored organosodium
ethylsodium, it reacted with diethyl ether and was decomposed compound was added to a suspension of the organic perchlorate
in the process. In air ethyllithium inflamed, burning with a bright in ether. The RNa solution became lighter in color and the
red flame. perchlorate went into solution. However, the organic perchlo-
Phenyllithium was prepared using the same procedure. During rates were of limited applicability. The tetraphenylallyl per-
the 5-6 day reaction time at 65 °C a microcrystalline powder chlorate, for instance, was not very stable.
was formed. Decantation from the lithium/lithium amalgam
mixture followed by centrifugation of the phenyllithium suspen-
sion and washing of the isolated solid with benzene and
petroleum ether gave a pure product (Anal. Calcd: Li, 8.33.
Found: Li, 8.61). Even simpler was the preparation of phenyl-
lithium in high purity by the transmetalation reaction (eq 18).
Phenyllithium was isolated as a microcrystalline white powder
that inflamed in air, burning with a yellow flame.
An important contribution by Ziegler in this early work was
the discovery that the colored organoalkali-metal compounds
can be prepared by the cleavage of appropriately substituted
ethers by the alkali metals: e.g., eq 25.53 Further investigation
Although Schlenk described his preparations of the orga-
noalkali-metal alkyls and phenyls adequately in ref 20 when
he wrote the organoalkali-metal chapter in the Houben-Weyl
series,49 he was requested by the editor to do so in even greater
detail. This he did, with detailed instructions even for nitrogen
purification, including a final step of passing it through an ether of the ether cleavage reaction54 led to the preparation of
solution of triphenylmethylsodium (whose color served as an 2-phenylisopropylpotassium (19), a compound that was to be a
indicator for the absence of oxygen).
Schlenk’s activity in the area of organoalkali-metal chemistry
continued into the late 1920s and early 1930s, dealing mostly
with those classes of compounds discussed above. In 1931 he
reported on yet another class, the alkali-metal enolates, prepared very useful reagent in Ziegler’s subsequent organoalkali-metal
as shown in eqs 19 and 20 and isolated as the pure solids.50 research. The cleavage of 2-phenylisopropyl methyl ether with
liquid sodium-potassium alloy proceeded rapidly in diethyl
ether. The resulting product, 22, a deep red ether-soluble
compound, was obtained in excellent yield.
In 1928 a preliminary communication by Ziegler and Bähr55
reported a reaction of 2-phenylisopropylpotassium that turned
(51) For a biographical sketch and picture of Ziegler see Part 1 of this
essay.1 Ziegler obtained his Ph.D. in 1920 and completed his Habilitation
in 1923; therefore, his earliest independent research was carried out in 1921/
During the early years, as he had requested, Schlenk did not 1922.
have competitors in his research, but this changed, with the no (52) Ziegler, K. Justus Liebigs Ann. Chem. 1923, 434, 34.
(53) Ziegler, K.; Thielmann, F. Ber. Dtsch. Chem. Ges. 1923, 56, 1740.
(54) Ziegler, K.; Schnell, B. Justus Liebigs Ann. Chem. 1924, 437, 227.
(48) Ziegler, K.; Nagel, K.; Patheiger, M. Z. Anorg. Allg. Chem. 1955, (55) (a) Ziegler, K.; Bähr, K. Ber. Dtsch. Chem. Ges. 1928, 61, 253.
289, 345. (b) Full paper (covering alkyllithium as well as C6H5(CH3) 2CK): Ziegler,
(49) Schlenk, W. In Houben-Weyl, “Die Methoden der Organischen K.; Crössmann, F.; Kleiner, H.; Schäfer, O. Justus Liebigs Ann. Chem. 1929,
Chemie”, 2nd ed.; Thieme: Stuttgart, Germany, 1924; Vol. IV, pp 945- 473, 1–35. (c) Ziegler, K.; Grimm, H.; Willer, R. Justus Liebigs Ann. Chem.
978. 1939, 542, 90. (d) In 1957, when the starting ether, C6H5(CH3) 2COCH3,
(50) (a) Schlenk, W.; Schmidt, R. Ber. Dtsch. Chem. Ges. 1925, 58, had become more readily available, Ziegler conducted further studies of
1189. (b) Schlenk, W.; Hillemann, H.; Rodloff, I. Justus Liebigs Ann. Chem. the reactions of C6H5(CH3) 2CK: Ziegler, K.; Dislich, H. Chem. Ber. 1957,
1931, 487, 135. 90, 1107.
Organometallics, Vol. 28, No. 1, 2009 17
Scheme 10 of 1910 he began his investigation of the interaction of sodium
with 1,3-dienes. In one experiment reported by Harries 9 g of
pure 1,3-butadiene and 0.5 g of sodium wire were sealed in a
glass tube and warmed at 35-40 °C on a water bath for 3 h.
During this time the liquid thickened around the sodium wire
into a brown, gelatinous mass. No buildup of pressure was noted
when the tube was opened. Addition of the contents of the tube
to dilute aqueous alcohol to destroy and remove unreacted
sodium converted the brown mass into a bright yellow, rubbery
material in nearly quantitative yield. Freshly prepared, the
product was soluble in ether, chloroform, and benzene. With
time, however, the solubility decreased and the material swelled
enormously when treated with solvent. The material could be
vulcanized, and the vulcanized product was superior to the
products derived from butadiene polymers obtained in other
ways in terms of tear strength and elasticity. The C,H analysis
out to be crucial in the further development of Ziegler’s scientific of the product from the sodium process agreed reasonably well
career: its addition to the CdC bond of appropriately activated with that calculated for (C4H6)n: Anal. Calcd: C, 88.89; H, 11.11.
olefins. This was the first carbometalation reaction (Scheme 10). Found: C, 87.92; H, 11.10. Its ozonolysis in chloroform
As Ziegler related, this was an accidental discovery. The reaction proceeded very slowly, giving a white, solid ozonide which,
that had been sought was the preparation of a dipotassium when dry, readily exploded.
derivative of stilbene as shown in eq 26, the potassium analogue As Harries noted, there were two remarkable features about
the sodium-induced polymerization of dienes. First, the sodium,
for the most part, was unconsumed; thus, it appeared to be a
catalyst. Second, although the product resembled rubber (Kaut-
schuk in German), it was different from natural rubber and
thermally produced synthetic rubber in terms of its behavior
on ozonolysis. Harries called these reactions “a surprising
of the violet-brown disodium stilbene that Schlenk had reported discovery” and called the products “Natrium-Kautschuk”. In
earlier. Instead of the expected violet-brown, ether-insoluble 1911 the structures of natural and synthetic rubber were not
solid, a soluble orange-yellow potassium compound resulted. known. Harries discussed them in terms of polymers of cyclic
Treatment of this solution with carbon dioxide followed by dimers, 22. Isoprene, in contrast to 1,3-butadiene, was not as
aqueous acid workup gave a carboxylic acid that was identified readily polymerized by sodium, requiring a reaction time of
as 21. This established that the potassium compound had been 50-100 h at 60 °C.
Scheme 11 Scheme 12
“Applications of Electricity in Chemistry.” As related by Even before 1933, but certainly after Hitler’s ascension to
his son Hermann he had obtained a new electrical apparatus power in Germany, Schlenk’s life became difficult. He made
but ignored warnings not to use this for the actual no secret of his beliefs, which were not in line with those of
demonstration and received a tremendous electrical dis- the National Socialists, and he did not abandon his Jewish
charge, which could be seen going in one finger and out friends and colleagues. He became persona non grata in Berlin,
the other hand. Schlenk was thrown to the floor, but the capital. In 1935 he was removed from his chair at the
revived and finished the lecture.” University of Berlin and transferred to the University of
Tübingen. There he became Professor of Chemistry and Director
Another move came in 1921 when Schlenk was chosen to of the Chemistry Institute. Tübingen was glad to have him, and
be the successor of Emil Fischer, who had died in 1919, as during the 71/2 years he spent there he was held in high regard.
Professor of Chemistry and Director of the Chemical Laboratory He was busy with teaching, administration, and research, as well
of the Fredrich Wilhelm University of Berlin. In Berlin he as with work on the second volume of his textbook of advanced
continued his organosodium research, assisted by Ernst Berg- organic chemistry. (The first had been completed in Berlin in
mann, who had been a Schlenk student and subsequently, an 1932.) With the start of World War II in 1939 things became
Assistent after his Habilitation. He did, however, venture, if difficult in the Chemistry Institute. A severe shortage of funds
only briefly, into other areas. In 1923, interested in the developed. Schlenk had finished the second volume of his
mechanism of the catalytic hydrogenation of unsaturated organic textbook in 1939 and had started on a third. The demands on
compounds, he looked into the possibility of one of the then Schlenk’s time were such that he did no further research. He
currently discussed mechanisms, a route involving labile metal had published his last paper in 1933 (the last one on organoal-
hydride intermediates. He chose nickel because of previous work kali-metal chemistry in 1931). No publications resulted from
with this metal by others. Using (not surprisingly) an organo- his years in Tübingen. One last blow came in 1942 when he
metallic approach, Schlenk and Weichselfelder treated 0.17 g was expelled from the German Chemical Society (GDCh.). He
of anhydrous NiCl2 with approximately 0.15 mol of C6H5MgBr died during the war in 1943.76
in diethyl ether with agitation at 14 °C.74 The reaction mixture The reaction of diorganomercurials with alkali metals is rarely
immediately became brown. During the course of the reaction used these days for the preparation of organoalkali-metal
purified hydrogen was admitted from a gas buret; H2 absorption compounds. It served Schlenk and Holtz well and was still used
was rapid, taking up 4H per Ni. The product was a black solid. in the 1930s, but then the better preparative routes of Ziegler,
Decomposition of a dry sample of the latter by addition of Wittig, Gilman and Morton (see Part 1) were developed and
alcohol and measurement of the evolved hydrogen established these became the methods of choice.
the formula NiH2. This solid was found to be the most active
hydrogenation catalyst known in terms of rate and activity at
IV. Preparation of Organoalkali-Metal Compounds by
room temperature. However, it had a fatal flaw: even traces of
the Transmetalation Reaction
air immediately “killed” its activity at room temperature. This
was not Schlenk’s only work with the Grignard reagent. Another procedure for the preparation of organoalkali-metal
Schlenk had married Mathilde von Hacke in 1905. The couple compounds is the transmetalation reaction, in which there is
had three sons, Wilhelm Jr. (1907-1975), Fritz (1909-1998), substituent exchange between an organoalkali-metal compound
and Hermann (1914-2003), all of whom became Ph.D. chem- and an organic derivative of another metal, usually a heavy metal
ists. With Wilhelm Jr. Schlenk published an important paper in (mercury, tin, lead, antimony, bismuth, selenium, tellurium), and
organomagnesium chemistry on the equilibrium shown in eq in a few cases silicon and even phosphorus and iodine (eq 29).
28,
Scheme 19
tive rates in the case of CH3Li, but when the less reactive (eq 59) which increases the yield of tetraphenyltin while
CH3MgBr was used, the two steps could be studied diminishing the (C6H5)3SnCH2Li yield.
separately.119b In THF solution at room temperature only the
transmetalation reaction occurred, giving (C6H5)2(CH3)PO and
C6H5MgBr. When this reaction mixture was heated at reflux,
metalation, giving C6H6 and (C6H5)2P(O)CH2MgBr, occurred. In the organolead series,126 reaction of [(C6H5)3Pb]2CH2 with
The (C6H5)3PS/CH3Li reaction did not take place in diethyl ether, C6H5Li in THF at -70 °C gave (C6H5)3PbCH2Li in nearly
but addition of THF resulted in formation of a deep red solution quantitative yield. This reagent was fairly stable in solution to
of (C6H5)2P(S)CH2Li.119c 20 °C. In a similar manner, [(C6H5)3Pb]3CH could be converted
As a further example of the utility of the transmetalation to [(C6H5)3Pb]2CHLi by reaction with phenyllithium. Reaction
reaction in organic synthesis is Schöllkopf and Gerhart’s of phenyllithium with [(C6H5)3Pb]2CHGe(C6H5)3 and with
preparation of diethylcarbamoyllithium (eq 56).120 Addition of [(C6H5)3Pb]2CHAsPh2 provided the mixed reagents
a ketone to the reaction mixture gave the expected (C6H5)3PbCH(Li)Ge(C6H5)3 and (C6H5)3PbCH(Li)As(C6H5)2,
(C2H5)2NC(O)C(OH)R2. respectively. These examples are given to show how broad and
varied the possibilities for synthesis of multimetal organometallic
compounds are. In group 15, (C6H5)2AsCH2Li,127 (C6H5)2-
SbCH2Li,128 and (C6H5)2BiCH2Li129 have been prepared by such
Li/Sn transmetalation.
Transition-metal-containing organolithium reagents can be
A number of organolithium reagents containing an organo- prepared by a number of standard procedures, including
metallic or organometalloidal substituent have been prepared. transmetalation. The first reported preparation of ferrocenyl-
A useful, general procedure for the preparation of such lithium was effected by Li/Hg exchange (eq 60).130
compounds was developed by Dieter Seebach, who applied it
to the preparation of (R-phenylselenomethyl)-121,122 and (R-
phenyltelluromethyl)lithium reagents.123 This protocol involved
the preparation of a starting material containing two or more
of the organometallic or organometalloidal groups in question
and displacing one of them with an organolithium reagent (eq
57). Thus, reaction of 1 molar equiv of n-butyllithium with 1,1′-Dilithiodibenzenechromium, free of the monolithiated
(C6H5Se)3CCH3 in THF at -78 °C gave (C6H5Se)2(CH3)CLi, species usually obtained as a byproduct in the direct lithiation
while (C6H5Se)2CH2 gave C6H5SeCH2Li. (C6H5Te)2CH2 reacted of dibenzenechromium, could be prepared by the reaction shown
in a similar manner (eq 58). Such Li/Se transmetalation has in eq 61.131 ((Chloromercurio)cyclobutadiene)iron tricarbonyl
been used to prepare R-lithioalkylsilanes from underwent Li/Hg exchange with phenyllithium at -78 °C to
RSeCH(R′)Si(CH3)3 selenides.124 give the useful monolithio derivative (eq 62).132 Another
(119) (a) Seyferth, D.; Welch, D. E.; Heeren, J. K. J. Am. Chem. Soc. (126) (a) Kauffmann, T.; Kriegsmann, R.; Rensing, A.; König, R.;
1963, 85, 642. (b) Seyferth, D.; Welch, D. E.; Heeren, J. K. J. Am. Chem. Steinseifer, F. Chem. Ber. 1985, 118, 370. (b) Kauffmann, T.; Rensing, A.
Soc. 1964, 86, 1100. (c) Seyferth, D.; Welch, D. E. J. Organomet. Chem. Chem. Ber. 1985, 118, 380. (c) Rensing, A.; Echsler, K.-J.; Kauffmann, T.
1964, 2, 1. Tetrahedron Lett. 1980, 21, 2807.
(120) Schöllkopf, U.; Gerhart, F. Angew. Chem., Int. Ed. Engl. 1967, (127) Kauffmann, T.; Altepeter, B.; Klas, N.; Kriegsmann, R. Chem.
6, 805. Ber. 1985, 118, 2353.
(121) (a) Seebach, D.; Büttner, G. Angew. Chem., Int. Ed. Engl. 1969, (128) Kauffmann, T.; Joussen, R.; Klas, N.; Vahrenhoust, A. Chem. Ber.
8, 451. (b) Seebach, D.; Peletier, N. Chem. Ber. 1972, 105, 511. 1983, 116, 473.
(122) See also: (a) Dumont, W.; Bayet, P.; Krief, A. Angew. Chem., (129) (a) Steinseifer, F.; Kauffmann, T. Angew. Chem., Int. Ed. Engl.
Int. Ed. Engl. 1974, 13, 804. (b) Anciaux, A.; Eman, A.; Dumont, W.; Krief, 1980, 19, 723. (b) Kauffmann, T.; Steinseifer, F.; Klas, N. Chem. Ber. 1985,
A. Tetrahedron Lett. 1975, 1617. (c) Van Ende, D.; Dumont, W.; Krief, A. 118, 1039.
Angew. Chem., Int. Ed. Engl. 1975, 14, 700. (d) Reich, H. J.; Bowe, M. D. (130) (a) Seyferth, D.; Helling, J. F. Chem. Ind. (London) 1961, 1568.
J. Am. Chem. Soc. 1990, 112, 8994. (b) Seyferth, D.; Hoffmann, H. P.; Burton, R.; Helling, J. F. Inorg. Chem.
(123) Seebach, D.; Beck, A. K. Chem Ber. 1975, 108, 314. 1962, 1, 227.
(124) Dumont, W.; Krief, A. Angew. Chem., Int. Ed. Engl. 1976, 15, (131) Elschenbroich, C.; Schmidt, E.; Metz, B.; Massa, W.; Wolcadlo,
161. S.; Burghaus, O. Z. Anorg. Allg. Chem. 1999, 625, 875.
(125) Kauffmann, T.; Kriegsmann, R.; Altepeter, B.; Steinseifer, F. (132) Dieck, H. Ph.D. Dissertation, University of Texas at Austin, Austin,
Chem. Ber. 1982, 115, 1810. TX, 1968.
Organometallics, Vol. 28, No. 1, 2009 27
However, the two triphenyltin substituents could be displaced
in a stepwise manner133 (Scheme 20).
Scheme 20
less polar than HMPA, such as THF and diethyl ether, which
usually are used, they are at best transient intermediates. In
diethyl ether, on the basis of the stereochemical results noted
above, a four-center transition state could be involved. The
kinetics of some n-butyllithium/R-aminoalkyltin compound
transmetalations were studied by Klein and Gawley using the
rapid injection NMR procedure (119Sn NMR in this case).115
The effect of solvent on the rate was significant: too fast to
measure in THF, too slow in diethyl ether, and usually of a
convenient rate in 40/60 Et2O/THF.
However, as Langer noted in his chapter in ref 141a for the an effective organosodium initiator for 1,3-diene polymerization
first of these,144a “Claims lost to 3,451,988 and 3,458,586” (both (see pp 16-17 of ref 1). Most of the polymerization studies
Langer patents) and for the second144b “Claims lost to 3,769,345” were done with the system obtained by mixing n-amylsodium,
(also a Langer patent). Presumably, this was the result of a lower R-olefin such as 1-propene, and isopropyl alcohol. The
litigation involving ESSO, Sun Oil, and the U.S. Patent Office. olefin and i-PrOH reacted with n-amylsodium to give a reagent
The alkyllithium-TMEDA combination was found to be a containing an allylsodium compound and sodium isopropoxide
superbase in its reactivity. Not only did n-C4H9Li-TMEDA (as well as NaCl from the n-amylsodium preparation). This was
lithiate toluene to benzyllithium and propene to allyllithium and a very potent polymerization initiator (“ALFIN catalyst”).147
sec-C4H9Li-TMEDA lithiate 1-butene to crotylithium, but also However, more pertinent to the present discussion, Morton also
n-C4H9Li-TMEDA lithiated benzene to give phenyllthium.145 found that addition of sodium isopropoxide or, better, potassium
All of these organolithium reagents were formed in high yield isopropoxide, to a suspension of n-amylsodium in alkane
as TMEDA complexes. This might, in some cases, be a medium increased the already high bascity of n-amylsodium:
disadvantage since, if acid hydrolytic workup was not possible, more rapid reactions gave higher yields of metalation product.148
the separation of TMEDA might be a problem. Also, in the The most spectacular observation made was that n-C5H11Na/
case of the lithiation of toluene, 1-5% of nuclear metalated iPrONa could metalate ethylene directly to give vinylsodium.148b
byproducts (meta and para isomers) were present. This is not a Morton’s discovery of the “alkoxide effect” was never cited by
problem with benzyllithum prepared by the transmetalation the later workers, who apparently rediscovered the alkoxide
reaction. Other n-C4H9Li-chelating tertiary diamine reagents effect some 20 years later. Perhaps they were too focused on
were prepared, but the TMEDA complex was the most reactive. organolithium chemistry to recognize that the concept had been
The main industrial interest in these C4H9Li-diamine complexes discovered earlier in organosodium chemistry by Morton.
derived from the fact that they are excellent initiators for the The story continues in the mid 1960s when three research
homo- and copolymerization of ethylene, styrene, and 1,3- groups independently rediscovered the “alkoxide effect”. In all
dienes,141b,146 but their outstanding utility as bases stronger than three cases alkyllithium reagents, usually n-butyl-, sec-butyl-,
n-butyllithium alone soon was recognized. and tert-butylllithum, were “activated” by the addition of 1
The most powerful basic alkali-metal-derived systems are molar equiv or more of a potassium or sodium alkoxide. The
obtained by adding 1 or 2 molar equiv of a sodium or potassium three groups reported such systems almost at the same time:
alkoxide, preferably one that is soluble, to the alkylithium
solution. Such mixed reagents are called “superbases”. (147) Morton, A. A.; Patterson, G. H.; Donovan, J. J.; Little, E. L. J. Am.
As reported in Part 1 of this essay,1 such superbases originally Chem. Soc. 1946, 68, 93.
(148) (a) Morton, A. A.; Holden, M. E. T. J. Am. Chem. Soc. 1947, 69,
were discovered accidentally in 1946 by Avery A. Morton at 1675. (b) Morton, A. A.; Marsh, F. D.; Coombs, R. D.; Lyons, A. L.; Penner,
the Massachusetts Institute of Technology during his search for S. E.; Ramsden, H. E.; Baker, V. B.; Little, E. L.; Letsinger, R. L. J. Am.
Chem. Soc. 1950, 72, 3785. (c) Morton, A. A.; Claff, C. E., Jr. J. Am.
Chem. Soc. 1954, 76, 4935. (d) Morton, A. A.; Claff, C. E., Jr.; Collins,
(145) Smith, W. N. In ref 141a, Chapter 2, pp 23-55. F. W. J. Org. Chem. 1955, 20, 428. (e) Morton, A. A.; Eisenmann, J. L. J.
(146) Kamienski, C. W. In ref 141a, Chapter 7, pp 163-176. Org. Chem. 1958, 23, 1469.
30 Organometallics, Vol. 28, No. 1, 2009
(160) Lochmann, L. Collect. Czech. Chem. Commun. 1987, 52, 2710. (166) Schade, C.; Bauer, W.; Schleyer, P. v. R. J. Organomet. Chem.
(161) Lochmann, L.; Fossatelli, M.; Brandsma, L. Recl. TraV. Chim. 1985, 295, C25.
Pays-Bas 1990, 109, 529. (167) Pi, R.; Bauer, W.; Brix, B.; Schade, C.; Schleyer, P. v. R. J.
(162) (a) Lochmann, L.; Trekoval, J. J. Organomet. Chem. 1987, 326, Organomet. Chem. 1986, 306, C1.
1. (b) Lochmann, L.; Petránek, J. Tetrahedron Lett. 1991, 32, 1483. (168) Brandsma, L.; Verkruijsse, H. D.; Schade, C.; Schleyer, P. v. R.
(163) Andringa, H.; Verkruijsse, H. D.; Brandsma, L.; Lochmann, L. J. J. Chem. Soc., Chem. Commun. 1986, 260. Parsons et al., who used Morton’s
Organomet. Chem. 1990, 393, 307. procedure with some modification, reported vinylsodium yields of 43-
(164) Lochmann, L.; Fréchet, J. M. J. Macromolecules 1996, 29, 1767. 45%: Parsons, T. D.; Silverman, M. B.; Ritter, D. M. J. Am. Chem. Soc.
(165) Lochmann, L.; Trekoval, J. Collect. Czech. Chem. Commun. 1986, 1957, 79, 5091.
51, 1439. For more detailed studies, see: (a) Lochmann, L. J. Organomet. (169) Brandsma, L., Verkruijsse, H. D. PreparatiVe Polar Organome-
Chem. 1989, 376, 1. (b) Lochmann, L. J. Organomet. Chem. 1989, 364, tallic Chemistry 1; Springer: Berlin, 1987,; pp 18-19. This book is an
281. (c) Brandsma, L.; Mal’kina, A. G.; Lochmann, L.; Schleyer, P. v. R. excellent guide to the preparation and use of organolithium, -sodium, and
Recl. TraV. Chim. Pays-Bas 1994, 113, 529. -potassium reagents.
32 Organometallics, Vol. 28, No. 1, 2009