cut to the chase...

On the Action of Organic Acids and their Anhydrides on the Natural Alkaloids.

C.R.A. Wright, D.Sc., Lecturer on Chemistry, St. Mary's Hospital Medical School, London. (Journal of The Chemical Society, Vol. 27 (1874), pp. 1031-1043.)

*Editor's Notes:

This is not the entire paper. Deleted sections are indicated by a "[...]". In most cases, deleted sections consist of quantitative analyses. These are not only uninteresting, but unimportant, since the molecular formulae for the alkaloids given in the paper are wrong. In the footnotes I attempt to describe how and why these early chemists went wrong in the above regard, and why we might care. (The editor is neither a chemist nor a historian -- please bear with him).

All temperatures are in Farenheit scale, as in the original.

1. Action of Acetic Acid on Codeine

In a former paper (Proc. Roy. Soc., 1872, p. 278), it was stated that attempts to form codeine derivatives by the action of glacial acetic acid at 100 met with only a small degree of success; when, however, codeine dried at 140 is heated to boiling with twice its weight of glacial acetic acid for eight hours (an inverted condenser being attached to retain the acid volatilised), a large quantity of a new base, diacetyl-codeine (1), is formed in virtue of the reaction,

C36H48N2O6 + 2C2H3O.OH = 2H2O + C36H40(C2H4O)2N2O6

On evaporating the resulting liquid to dryness on the water-bath, dissolving the residue in water, and adding sodium carbonate, an immediate white precipitate is thrown down, flocculent at first but soon becoming crystalline on standing; by rapid filtration this is separated from unaltered codeine, which is not immediately precipitated in this manner, save in very concentrated solutions. By dissolving the precipitate in dilute hydrochloric acid, and repeating the precipitation by sodium carbonate three of four times, a product is obtained quite free from codeine; the last precipitate is dissolved in ether or hot dilute alcohol, from either of which solvents the new base crystallises in bold, well-defined anhydrous crystals; it can now be recrystallised from benzene, chloroform, and boiling water without change, being readily soluble in all these solvents except water, in which it is only sparingly soluble even when boiling. Diacetyl-codeine forms a well-crystallised hydrochloride somewhat more soluble in water than codeine hydrochloride; when air-dry, the crystals contain C40H46N2O8.2HCl.4H2O. [...]

Attempts to prepare mono-acetylcodeine (1), C36H41(C2H3O)2N2O6, or to isolate it from the products of the action of acetic acid on codeine did not meet with success; it was, however, noticed that a larger portion of the first precipitate thrown down by carbonate of soda was soluble in excess of the precipitant than was the case with subsequent precipitations, whence it is possible that the monoacetylated base is formed together with diacetyl-codeine; hitherto, however, no means of separating it from unaltered codeine have been devised [...]

2. Action of Acetic Anhydride on Codeine

In order to see if higher acetyl derivatives are obtainable by the action of acetic anhydride, codeine was treated with a large excess of that substance in various ways; in every case, however, nothing but diacetyl-codeine resulted [...]

3. Action of Acetic Acid on Morphine

When morphine is boiled for several hours with twice its weight of glacial acetic acid, and inverted condenser being attached, a large amount is converted into a substance related to morphine in the same manner as diacetyl-codeine is to codeine; by dissolving the product in water, adding ammonia, and shaking up with ether, an ethereal solution is obtained which yields a copious crystalline hydrochloride on shaking with hydrochloric acid; this hydrochloride is but sparingly soluble in cold water, but can be recrystallised from that menstruum when hot without change. The crystals contain C34H36(C2H3O)2N2O6.2HCl.6H2O. [...]

When pure, neither the free base nor its salts give any coloration with ferric chloride; simultaneously with this base, however, there is formed a small quantity of a substance that does colour ferric chloride blue like morphine, but differs therefrom in being soluble in ether and in being precipitated by carbonate of soda in white amorphous flakes, soluble in excess of the precipitant; this substance does not seem to give a crystalline hydrochloride, and is apparently identical with the -diacetyl-morphine (2) described in   5, i.e, is isomeric with the above base, yielding a sparingly soluble crystalline hydrochloride, which is accordingly designated [alpha]-diacetyl-morphine (2).

4. Action of Acetic Anhydride in excess on Morphine(3)

When morphine is brought into contact with an excess of acetic anhydride, the following reaction takes place, tetracetyl-morphine being produced:--

C34H38N2O6 + 4(C2H3O)2O = C34H34(C2H3O)4N2O6 + 4C2H3O.OH.

The same result is brought about whether the materials are allowed to remain in contact at the ordinary temperature for several days or are heated to 100 or to 140 for some hours; on adding sodium carbonate to the product dissolved in water, a precipitate is obtained flocculent at first, but soon becoming crystalline; this dissolves readily in ether (morphine is virtually insoluble in that menstruum), and is obtained in fine anhydrous crystals by evaporation; it can also be recrystallised without change from hot alcohol or benzene, and from chloroform; long-continued boiling with alcohol partially decomposes it, whilst boiling with water quickly alters it. [...]

Tetracetyl-morphine gives no colour reaction with ferric chloride; like morphine, it is but sparingly soluble in ammonia and sodium carbonate, but in caustic potash it is readily soluble. When it is exactly  neutralised with dilute hydrochloric acid, a solution is obtained which deposits crystals on standing over sulfuric acid; these are extremely soluble in water, so that they only form just before the whole mass has evaporated to dryness; if any excess hydrochloric acid be present, partial decomposition takes place during the evaporation, acetic acid being evolved. [...]

The syrupy mother liquors from which tetracetyl-morphine has mostly crystallised, contain more or less [alpha]-diacetyl-morphine, which may be readily extracted by shaking up the ethereal liquor with hydrochloric acid, when the sparingly soluble hydrochloride of the [alpha] base crystallises out. [...]

5. Action of Acetic Anhydride not in excess on Morphine

When anhydrous morphine is heated with a quantity of acetic anhydride less than that requisite to form tetracetyl-morphine, various intermediate products appear to be formed according to circumstances; thus, when the requisite quantities are taken for the following equation:--

C34H38N2O6 + 2(C2H3O)2O = C34H34(C2H3O)2N2O6 + 2C2H3O.OH,

and the whole is heated to 100 for an hour, the chief product is that expressed by the above equation; but the diacetyl-morphine thus produced is not identical with the [alpha]-diacetyl-morphine above described; it yields a hydrochloride excessively soluble in water and incapable of crystallising, but drying up to a varnish, and it strikes a blue color with ferric chloride; the free base is readily soluble in ether and is precipitated from its salts in amorphous flakes by alkalis; it is readily dissolved in ammonia, sodium carbonate, and caustic potash; it is much less stable than [alpha]-diacetyl-morphine, ordinary morphine soon crystallising out from an ammoniacal solution of the base [...]

This base is for the present designated -diacetyl-morphine. [...]

8. On the Physiological Action of the above Morphine and Codeine Derivatives.

By F.M Pierce, M.D., L.R.C.P. (Lond.), Associate of the Owens College.

Doses of the anhydrous hydrochloride of [alpha]- and -diacetyl-morphines, tetracetyl-morphine, and diacetyl-codeine were subcutaneously injected into young dogs and rabbits, the quantities used in each instance being equivalent to 0.050 gram of anhydrous morphine hydrochloride, [...] with the following general results. No great amount of difference between the first three substances was noticeable, great prostration, fear, and sleepiness speedily following the administration, the eyes being sensitive, and pupils dilated, considerable salivation being produced in dogs, and slight tendency to vomiting in some cases, but no actual emesis. Respiration was at first quickened, but subsequently reduced, and the heart's action was diminished, and rendered irregular. Marked want of coordinating power over the muscular movements, and loss of power in the pelvis and hind limbs, together with a diminution of temperature in the rectum of about 4, were the most noticeable effects, the action on rabbits being less marked than that on dogs, and salivation and def&ealig;cation not being produced in their case.

In the case of -diacetyl-morphine, the moderate excitement produced after the first injection was, and the diminution of temperature in the rectum were somewhat more marked than with the others, tetracetyl-morphine prooducing the least excitement.

Diacetyl-codeine produced results of precisely the same character, but somewhat less marked; the want of muscular coordination was less noticeable, whilst salivation was most profusely caused with dogs; moreover, whilst in the case of the morphine derivatives, the animals only fully recovered in about 24 hours, about 8 hours sufficed for recovery in the case of diacetyl-codeine. [...]

Editor's Footnotes:

1. All the morphine derivatives mentioned in the article ("diacetyl-codeine", "diacetyl-morphine", and "tetracetyl-morphine") are so named due to a miscalculation of the molecular formulae of morphine and codeine. (For instance, those of us who have not been living in caves all know that heroin is diacetyl-morphine and not "tetracetyl-morphine"). The confusion may be cleared simply by dividing by two: the molecular formula of morphine base, for instance, is C17H19NO3, while that given in the article is  C36H38N2O6. The latter is, in fact, two morphine molecules.

I would guess that early calculations were based on the assumption that the salts the alkaloids formed with divalent anions (e.g., morphine sulfate, codeine sulfate) had a 1:1 ratio of ionized-base to sulfate. Since, in fact, two codeine (or morphine) molecules are neutralized by one sulfuric acid molecule, the former assumption would lead to a molecular formula with each element present in twice its actual proportion.

This explains, for one thing, why it was impossible to form "monoacetyl-codeine" (or monoacetyl-morphine). Given the assumed molecular formulae, this would mean adding half a radical -- monoacetyl-codeine would be "halfacetyl-codeine". In reality, "diacetyl-codeine" and "diacetyl-morphine" are monoacetyl-codeine and monoacetyl-morphine, respectively.

2. Now that we know that "diacetyl-morphine" is really monoacetyl-morphine, what is the significance of these two "isomers" of monoacetyl-morphine? (It should be understood that the term "isomer", in those days, was used simply to refer to compounds with the same molecular formula but different physical properties. This is not how the term is used today). According to Small and Lutz in The Chemistry of the Opium Alkaloids, a report commissioned by the U.S. Public Health Service in 1932, "when morphine is heated with acetic acid two acetyl derivatives are formed; [alpha]-acetylmorphine and -acetylmorphine probably have respectively the phenolic and alcoholic hydroxyl groups acetylated." That is, [alpha]-acetylmorphine is 3-monoacetylmorphine (3-MAM) and -acetylmorphine is 6-monoacetylmorphine (6-MAM).

That "[alpha]-diacetyl-morphine" and "-diacetyl-morphine" refer to these two compounds respectively is made plausible by a number of observations. First, 6-MAM is frequently found in analyses of seized illicit heroin. It is believed to be present as a result of incomplete acetylation of morphine by acetic anhydride. As this is the "isomer" identified by Wright as the product of morphine treated by acetic anhydride "not in excess", it is very likely that his -diacetyl-morphine and our 6-MAM are one and the same. Also, the observation by Pierce that the -isomer had distinct pharmacological properties is suggestive, since 6-MAM is known to be significantly more potent than morphine. (Pierce's findings are puzzling in many respects, however, and are difficult to draw clear conclusions from).

"[alpha]-diacetyl-morphine" is more difficult to pin down. 3-MAM is found in forensic analyses of heroin and is considered indicative of its age, since heroin exposed to heat is known to degrade to form 3-MAM. It is likely that the "diacetyl-morphines" identified by Wright are, in fact, mixtures of 6- and 3-acetylmorphine and diacetylmorphine, with one or the other monoacetyl- derivative predominating.

3. This section describes a procedure that Wright could not have known would have such an impact on the world. Indeed, Wright's method bears remarkable resemblance to that used by "jungle chemists" today. Compare to this passage from the DEA publication Opium Poppy and Heroin Cultivation in Southeast Asia:

...morphine hydrochloride bricks are pulerized and the dried powder is is placed in an enamel or stainless steel rice cook pot. Acetic anhydride is then added...The pot lid is tied or clamped on, with a damp towel for a gasket, [Wright's "inverted condenser" apparatus] The pot is carefully heated for about two hours, below boiling, at a constant temperature of 185 Farenheit...Sodium carbonate, at 2.5 pounds per pound of morphine is dissolved in hot water and added slowly to tge liquid until effervescence stops. This precipitates the heroin base...