The barbituric acids, their chemical struture, synthesis and nomenclature
Author: Leo Levi
Pages: 30 to 40
Creation Date: 1957/01/01
"No small art is it to sleep:
It is necessary for that purpose
to keep awake all day."
from Thus Spake Zarathrustra
Barbituric acids are sedative drugs comprising a vast class of synthetic substances with closely related chemical structures and similar pharmacological activities. They are odourless, white crystalline solids which are only slightly water-soluble and hence administered mainly by the oral route. Their sodium salts, on the other hand, dissolve quite readily in aqueous media, which, when sterile, are suitable for subcutaneous, intramuscular or intravenous injection. The therapeutic value of these drugs is unquestioned, and in spite of certain shortcomings-the ideal sedative is still to be discovered or synthesized-they have by far supplanted all other hypnotic agents and are manufactured on a truly magnificent scale. Their production in the United States has increased by more than 400% since 1933 and now appears greatly to exceed the amount needed for therapeutic purposes  . In 1936 total U.S.A. production was 231,167 lb. of which 174,188 lb. were sold, and by 1948 this figure had reached 672,000 lb. (1 lb. = 7,000 grains). In 1953 production dropped slightly (634,000 lb.), but a record high level was reached in 1954, when 798,000 lb. were manufactured. This is an amount equivalent to 3,619,728,000 capsules or tablets of 0.1 g. each, or approximately 22 therapeutic doses per person in that country  .
VON BAEYER'S "SYNTHESIS" OF BARBITURIC ACID
The parent compound of this broad class of substances, barbituric acid, was obtained by von Baeyer almost 100 years ago  . This brilliant organic chemist had actually not set out to synthesize this compound, but happened to isolate it during the course of his masterly investigations of uric acid and its derivatives (Fig. 1). Yet he considered the discovery of this substance an extremely important one. "Man wird sehen, wie sich diese Materialien in einfachster Weise um die Substanz N 2C 4O 3H 4, die ich Barbitursäure nennen will, gruppieren lassen und wie also die Frage nach der Konstitution der Harnsäure und ihrer Derivate auf die Untersuchung dieser Substanz zurückgekehrt ist." No reference can be found in his papers which would suggest a reason for the choice of the name "Barbitursäure "l and the origin of the name has remained somewhat obscure. Fieser assumed that von Baeyer thought of the German word "Schlüsselbart" (the bit or "beard" of a key, from L. barba, beard) and uric acid so as to stress his conception of this substance as the key compound in the large series of cyclic ureides  . A second explanation, however, has it that von Baeyer, while working on this compound and other uric acid derivatives, met a charming young lady called Barbara and chose the name "Barbitursäure" for scientific plus sentimental reasons . This interpretation was later accepted by Fieser in the second edition of his textbook, which no longer mentions the original ingenious assumption.
Von Baeyer did not show the structural formula of the various compounds he isolated, for Kakulu’s theory of valence -announced in 1859-had not yet been applied to cyclic compounds. Only ten years after its discovery was the six-membered heterocyclic ring system of the compound recognized by Mulder and represented as a cyclic diimide consisting of two nitrogen and four carbon atoms, three of which are adjacent  . Its physico-chemical properties were more rigorously described four years later by Conrad and Guthzeit .
1 The German word "Barbitursaäre" carries the accent on the third syllable and retaining such accentuation in "barbituric acid" and "barbiturates" appears to be both etymologically justified and euphoniously most preferable. (See "Der Grosse Brockhaus", Handbuch des Wissens in Zwanzig Bänden, 15. Auflage, F. A. Brockhaus, Leipzig 1929, p. 300.)
The ring system, assigned to the barbituric acid molecule, is capable of existing in four different configurations, all of which are in dynamic equilibrium with each other (Fig. 2) and subject to reversible interconversion. In the keto form, carbon atoms No. 2, 4 and 6 are part of carbonyl linkages, while in the enol form these carbon atoms are associated with hydroxyl groups on account of the migration of active hydrogens from neighbouring linkages. The keto form predominates in acid solutions, whereas the enolic form is mainly present in alkaline media[ 8] . Depending on the solvent and the hydrogen ion concentration of the system, one, two or all three of the carbonyl linkages may be affected by the enolization process - . However, because of the marked chemical reactivity of the methylene group flanked by two carbonyl groups, either of the hydrogen atoms associated with this linkage will migrate before any of the imide bonds are broken. Hence the mono-enol form of the molecule is generally believed to have the configuration shown in Fig. 2, and in aqueous systems with a pH range from about 5 to 8 it is considered as affording the greatest contribution to the over-all structure of the compound  - . Although none of the tautomeric isomers has ever been isolated, considerable chemical support for existence of the mono-enol as presented in Fig. 2 was secured by Wood and Anderson and strong physico-chemical evidence for its structure, mainly by means of ultraviolet analysis, has since also been obtained . It would appear, therefore, that the mono-enol form as shown in New and Nonofficial Remedies -which differs with respect to the site of enolization-is not supported by present chemical evidence and applicable only to illustrate the tautomerization of 5,5-disubstituted barbituric acids where the enolization process involves either of the alternate-CONHCO-groupings.
Each of the enolic forms may be considered to represent an unsaturated cyclic alcohol, the unsaturation being conjugated in the dienol and both conjugated and symmetrical in the trienol form. The hydrogen shifts leading to the formation of these configurations are accompanied by the establishment of imino-carboxylic acid or isoamide groups, -N = C-OH, which are stabilized by both resonance and partial association with the unsaturated heterocyclic ring system. Although influenced by the reaction medium, all these structural features of the molecule-e.g., an active methylene group, a conjugated and symmetrical ring system, plus a comparatively stable imino carboxylic acid group-serve to confer acidic properties on the compound and make it behave like an acid that is stronger even than acetic or benzoic acid (K 25° in water = 1.05 x 10-4).
Similar tautomeric equilibria (from the Greek tauto = the same) involving shifts of the imino hydrogens associated with the heterocyclic nucleus and concomitant establishment of one or two double bonds within the ring system are exhibited by 5,5-disubstituted barbituric acids. Recently Levi & Hubley, in a study comprising the characterization of clinically important barbiturates by means of complex formation and infra-red spectroscopy, presented a mechanism of reaction based on enolization of these compounds in aqueous pyridine  . Further contributions to elucidation of the keto-enol equilibrium processes are being made by Chatten & Levi  .
The tautomeric structures have the same molecular composition but differ in the position of one or more hydrogen atoms.
Because of the lack of an active methylene group flanked by two carbonyl linkages and the impossibility of forming a symmetrically conjugated ring system, all 5,5-disubstituted barbituric acids are less acidic than the parent compound. Conversion of the keto form to the enol form is catalysed by alkali. When dissolved in water and treated with aqueous sodium hydroxide it is the enol form of the compound that reacts, and as its concentration in the equilibrium mixture is depleted, more of the keto form changes over to the enol form, which in turn is neutralized again by the base. Eventually all acid is converted to the sodium salt, as shown in Fig. 3. Other metallic hydroxides-e.g., those of potassium, calcium, magnesium, etc. -may be similarly used to prepare the corresponding metallic salts.2
In addition to the labile equilibrium present between the various tautomeric or desmotropic forms, each of these in turn exists in equilibrium with other structures which differ from one another only by shifts of valence electrons representing unsaturation. As a result of these migrations and oscillations of atoms and electrons, which occur continuously within the molecule, the structure of barbituric acid does not correspond to any one of the different formulas shown, but represents the sum total of the structural contributions of its individual configurations. We cannot describe this molecule accurately by a single formula, for it is both a structure of more than one tautomeric form and a resonance hybrid of more than one electronic state. The keto form manifests the same type of resonance as does urea, while the trienol form would exhibit the type of resonance shown by the benzene ring (Fig. 4). This phenomenon also contributes to making the compound behave as a relatively strong acid, for the two nitrogens are prone to electron attraction and therefore the acidity function of the HO-C = group between them is substantially increased. Even in the 5,5-disubstituted barbituric acids, where the presence of alkyl or aryl groups prevents the molecule from becoming benzenoid in type, the electron displacement effects are still sufficiently strong to make these compounds react with alkali and allow the formation of metallic salts.
2 In accordance with accepted rules of chemical nomenclature such salts only should be called "barbiturates" and the term" barbituric acids" be exclusively applied to the corresponding non-metallic derivatives. However, the name "barbiturate" has come to be used quite generally as a generic name, denoting a particular group of drugs, and as such it also includes the acid form.
CONVERSION OF VERONAL ACID TO VERONAL SODIUM
Like the parent compound, all substituted barbituric acids are resonating structures, and since resonance manifests itself in a shortening and therefore strengthening of the interatomic linkages, these compounds are also relatively stable molecules. In the test tube (see Fig. 1) as well as in the body the heterocyclic ring system survives drastic chemical treatment. Barbital, for example, undergoes but little change in vivo, more than 80% usually being excreted unaltered in the urine . Phenobarbital also remains to a large measure undestroyed by the body and most metabolites of other barbituric acids contain the six-membered heterocyclic ring system still intact  . Direct oxidative attack on the side chains appears to be the most important phase of the chemical alterations these compounds suffer within the body. The belief that some barbituric acids are completely broken down to carbon dioxide, ammonia and urea as a result of hydrolytic ring cleavage can no longer be considered valid. Experiments utilizing barbituric acids labelled with the heavy isotope of nitrogen, N 15, and with C 14 have shown that only small amounts of the tracers appear in the exhaled carbon dioxide and the ammonia-urea fractions of the urine. This relatively recent technique for following up the fate of barbituric acids in the body will surely serve to revise much of our present still fragmentary knowledge concerning the metabolism of these compounds, for most of the analytical methods that have been in general usage for half a century are unreliable because of their lack of both sensitivity and specificity  , .
RESONANCE FORMS OF BARBITURIC ACID
The resonating structures have the same molecular composition and are identical in the position of all the atoms. They differ only in the location of valency electrons.
ALKALINE HYDROLYSIS OF A BARBITURIC ACID
Where R&rsquo' and R" represent alkyl or aryl groups.
It is to be emphasized, however, that the stability of the heterocyclic system manifests itself in acidic media only. The bonds between tile-NH- and -CO- groups are easily broken by alkaline hydrolysis, and ammonia, carbon dioxide and an alkali malonate are produced when a barbituric acid is treated with hot alkali (Fig. 5). Under suitably controlled conditions the reaction has also been reported to lead to the formation of substituted malonic acid amides  or dialkylacetylureas  which compounds may be obtained as microcrystalline precipitates and serve to detect and characterize these drugs. It is because of this susceptibility to alkaline hydrolysis that barbituric acid derivatives are not considered to be aromatics, which compounds are usually quite stable in both acidic and basic media.
GRIMAUX’S SYNTHESIS OF BARBITURIC ACID
The condensation of urea with malonic acid is accompanied by the formation of 2 moles of water which are subsequently taken up by the phosphorus oxychloride. Instead of this reagent acetic anhydride may also be used .
Barbituric acid itself does not possess hypnotic properties and is considered to be a non-toxic chemical . Only when alkyl, aryl or alicyclic groups are introduced into the 5-position may the resultant compounds show marked physiological activity. Disubstitution appears to be essential, however, for conferring to the molecule hypnotic, sedative or anaesthetic properties. This phenomenon was observed by E. Fischer, one of Baeyer's famous students, and von Mering in 1903-i.e., forty years after the synthesis of barbituric acid had been reported . These workers used the diethyl analogue known since 1882  for clinical experiments and thus came to note its soporific effects. The product was introduced into medicine under the trade name of Veronal3 and it is still recognized as one of the best sedatives available. In 1911 phenylethyl barbituric acid was synthesized in the laboratories of I. G. Farben (D.R.P. 247, 952) and one year later its clinical evaluation reported by Geissler and Hauptmann  , the latter in particular noting the anticonvulsant properties of the compound. It came to be marketed under the trade name of Luminal. After the war of 1914-18 a rush to produce further modification of the original compound not covered by German patents became widely apparent, and since then the use of the malonyl ureas has been extensive among the medical profession. More than a thousand barbituric acids have already been synthesized-Chem. Abstr., Decennial Index 1937-46 lists over nine hundred-and their number is increasing steadily. Many of the newer compounds have been claimed to be superior to the older products with regard to potency, margin of safety and duration of action, but no more than about two dozen are commonly used in therapy. Veronal (barbital) and Luminal (phenobarbital), the two oldest representatives of the series, were both accepted by the United States and the British Pharmacopoeia, being first recognized by the U.S.P.X in 1926 and the B.P. in 1914 and 1932 respectively. Both preparations are also listed in the Pharmacopoea Internationalis. Of the newer preparations, only pentobarbital sodium, which is chemically identical with the proprietary barbiturate known as Nembutal, is to be considered an official drug. (Report of the Council on Pharmacy and Chemistry 1937, page 157.)
3 It has been reported that van Mering coined this name while approaching Verona on a trip through Italy  .
SYNTHESIS OF IPRAL
During the five decades these compounds have been in clinical usage the classical methods of preparation have practically remained unchanged. It is somewhat surprising that von Baeyer did not attempt to synthesize barbituric acid, as he knew that its two main structural units were derived from urea and malonic acid  . The synthesis was accomplished in 1879 by the French chemist Edouard Grimaux  , as shown in Fig. 6, and it soon became a general method for the preparation of substituted barbituric acids. Today-as fifty years ago-most of these compounds are still made by condensing urea with alkylated malonic acid esters in the presence of a slight excess of sodium or other metallic alkoxides dissolved in either ethyl or isopropyl alcohol (Fig. 7). The condensation is usually carried out by heating the mixture in an autoclave for several hours  . Neutralization of the reaction product (a barbiturate) with hydrochloric acid yields the crude free barbituric acid, which is then purified by recrystallization from suitable solvents. In most cases the urea-ester condensation proceeds quite smoothly and good product yields are obtained. Preparation of properly substituted malonic acid esters, on the other hand, is often a laborious process, The alkylated malonates are generally prepared by reacting the diethyl ester with aklyl halides, but in most cases alkylation of the ester leads to formation of a mixture of mono- and dialkyl derivatives with some of the starting material that has remained unchanged also being present. Hence considerable treatment of the reaction product is necessary to isolate the pure monoalkyl compound into which another radical is subsequently to be introduced.
In general, alkylated malonic acid esters are more readily prepared than arylated ones and therefore alkylated barbituric acids heavily outnumber those containing aromatic substituents (Table I). Phenobarbital, for example, cannot be prepared by treating ethyl-diethylmalonate with a phenyl halide. The aryl group must be introduced at quite an early stage of the synthesis (Fig. 8), and the introduction of a second such substituent at carbon atom No. 5 is not feasible in this manner  , because the required ester (ethyl diphenylmalonate) is cleaved into diethyl carbonate and ethyl diphenylacetate by sodium ethoxide in alcoholic solution. The compound has been prepared relatively recently by an entirely different route-namely, the condensation of benzene with alloxane in the presence of sulfuric acid  . Aromatic radicals other than phenyl-e.g., furfuryl, thienyl, piperidyl, α-naphthyl, cyclopentenyl, etc. joined directly to carbon atom No. 5-have also been synthesized, and in some instances the introduction of two such groups at this position has been accomplished. In addition to di-substitution at carbon atom No. 5, substitution of either one or both of the imide hydrogens may be brought about-and N-substituted barbituric acids be obtained-by using appropriately configurated ureas as starting material. The condensation of thiourea instead of urea with
SYNTHESIS OF LUMINAL
(Text continued on page 40)
substituted malonic acid esters yields thiobarbituric acids, which compounds have a sulfur atom in place of an oxygen atom attached to carbon atom No. 2. Thiobarbituric acid was prepared by Michael in 1887 (35), and the first disubstituted derivative (5,5-diethyl-2-thiobarbituric acid) synthesized by E. Fischer in 1904 (32). All thiobarbituric acids are more acidic than the corresponding barbituric acids, a phenomenon which may be related to their rapid onset and short duration of action. Of the large number of homologues that have been prepared, only thiopental, the sulfur analogue of U.S.P. pentobarbital sodium (5-ethyl-5-(1-methylbutyl)-2-thiobarbituric acid), has been officially recognized. The compound was first accepted by the British Pharmacopoeia in 1932 (7th Add.) and the United States Pharmacopoeia in 1947 (U.S.P. XIII). It is also listed in the Pharmacopoea Internationalis (Vol. I., 1951).
The chemical nomenclature of the barbituric acids conforms to the Geneva System and is therefore without ambiguity. Considerable confusion, however, has been caused by the plurality of trade names that have been and still are used synonymously to designate a given compound. Luminal and Veronal, for example, are known under 18 and 11 synonyms respectively; and these numbers are a fair indication of the popularity these drugs enjoy. Of the newer products, only Evipal appears to have gained such wide acceptance (14 synonyms). The ending commonly used for the free acids is "al" and the name of the metal is added as a separate word to indicate the corresponding salt for-example, barbital and barbital sodium. The trade names assigned to these compounds by their various manufacturers-along with the corresponding chemical names, structural formulae and molecular composition of the free acids-are listed in Table I for those representatives of the series which have assumed clinical significance either in this country or in Europe.
The author wishes to thank Dr. L. I. Pugsley for permission to publish this review and for the valuable interest he has shown in this work. He would also like to express his appreciation to Messrs. J. C. Potter and Miss Munro of the National Research Council for drawing the formula charts.
ISBELL, H. et al., "Chronic Barbiturate Intoxication ". An Experimental Study; Arch. Neur. and Psychiatry, 64, 1 (1950).02
United States Tariff Commission: Synthetic Organic Chemicals; Production and Sales, Washington, D.C., United States Government Printing Office, 1948, 1951, 1953 and 1954.03
von BAEYER, A., "Untersuchungen über die Harnsäuregruppe ", Annalen, 127, 1, 199 (1863); 130,129 (1864).04
FIESER, L. F., Organic Chemistry, D.C. Heath and Company, Boston, 1944, p. 247.05
COHEN, W. A. T., "Chemisch-Historische Aanteekeningen" De nomenclatuur van enkele organische zuren, Chemisch Weekblad, 40, 176 (1943).06
MULDER, E., Berichte, 6, 1233-7 (1873).07
CONRAD, M. & GUTHZEIT, M., "über Barbitursäure", Berichte, 14, 1943 (1881).08
MACBETH, A. K., NUNAN, T. H. & TRAILL, D., "Halogen Compounds of Barbituric Acids ", Jour. Chem. Soc., 1248 (1926).09
STUCKEY, R. E., "The Ultra-Violet Absorption Spectra of Barbituric Acid and its 1-Methyl and 1,3-Dimethyl Derivatives", Quarterly Jour. Pharm. and Pharmac., 13, 312 (1940).10
FOX, J. & SHUGAR, D., "Absorption Spectra and Structure of Barbituric Acid Derivatives as a Function of pH ", Bull. Soc. Chim. Belge, 61, 44 (1952).11
LOOFBOUROW, J. R. & STIMSON, M. M., "Ultraviolet Absorption Spectra of Nitrogenous Heterocyclic Compounds ", Jour. Chem. Soc., 1275 (1940).12
WOOD, J. K., "The Acidity Constants of Some Ureides and Uric Acid Derivatives ", Jour. Chem. Soc., 89, 1838 (1906).13
WOOD, J. K. & ANDERSON, E. A., "The Constitution of the Salts of Barbituric Acid ", Jour. Chem. Soc., 95, 979 (1909).14
BILTZ, H., "Die Salzbildung der Barbitursäure und ihrer Derivate ", Ann., 404, 186 (1914).15
STUCKEY, R. E., Private communication.
New and Nonofficial Remedies, J. B. Lippincott Co., Philadelphia, 1951, p. 234.16
LEVI, L. & HUBLEY, C. E., "Detection and Identification of Clinically Important Barbiturates ", Analytical Chemistry, 28, 1591 (1956).17
CHATTEN, L. & LEVI, L., "The Reaction of Barbituric Acids with p-Nitrobenzylchloride". To be published.18
MAYNERT, E. W. & VAN DYKE, H. B., "Absence of Localization of Barbital in Divisions of the Central Nervous System ", Jour. Pharmacol. Exptl. Therap., 98, 184 (1950).19
MAYNERT, E. W. & VAN DYKE, H. B., "The Metabolism of Barbiturates ", Jour, Pharm. & Exptl. Therap., 96, 217 (1949).20
WILLIAMS, R. T., "The Metabolism of Drugs and Toxic Substances ", Annual Review of Biochemistry, 20, 441 (1951).21
BERGMANN, F., "Colorimetric Determination of Amides as Hydroxamic Acids ", Anal. Chem., 24, 1367 (1952).22
WOLFF, C. J., "Die Identifizierung der Barbitale ", Acta Pharm. Intern., 2, 107 (1951).23
New and Nonofficial Remedies, J. B. Lippincott Co., Philadelphia, 1951, p. 235.24
FISCHER, E. & VON MERING, J., "über eine neue Klasse von Schlafmitteln ", Therapie der Gegenwart, 5, 97 (1903); Therap. Monatshefte, 17, 208 (1903).25
CONRAD, M. & GUTHZEIT, M.," über Barbitursäure-derivate ", Berichte, 15, 2844 (1882).26
GEISSLER, W., "Luminal, a New Strongly Acting Hypnotic ", Münch. Med. Wochenschrift, 59, 922 (1912) (C.A. 6, 1809 (1912)).27
HAUPTMANN, A., "Luminal bei Epilepsie", Ibid., 59, 1907 (1912).28
ZUNTZ, N., Münch. med. Wochenschr., 55, 400-2 (1908).29
GRIMAUX, É., "Synthèse des dérivés uriques de la série de l'alloxane ", Bull. Soc. Chim. France, 31, (2) 146 (1879).30
BILTZ, H. & WITTEK, H., "über alkylierte und acylierte Barbitursäuren ", Berichte, 54, 1035 (1921).31
FISCHER, E. & DILTHEY, A., "über C-Dialkylbarbitursäuren und über die Ureide der Dialkylessigsäuren ", Annalen, 335, 334 (1904), Merck, D.R.P. 146, 496.32
DOX, A. W. & THOMAS, A., "5,5-Diarylbarbituric Acids ", Jour. Am. Chem. Soc., 45, 1811 (1923).33
McELVAIN, S. M., "5,5-Diphenylbarbituric Acid", Jour. Am. Chem. Soc., 57, 1303 (1935).34
MICHAEL, A., "über neue Reaktionen mit Natriumacetessig und Natriummalonsäureaether ", Jour. Prakt, Chem., 35, (2) 456 (1887).