Author Topic: From Safrole to anywhere  (Read 1377 times)

0 Members and 1 Guest are viewing this topic.


  • Guest
From Safrole to anywhere
« on: September 16, 2003, 03:09:00 AM »
As promised to Rhodium here is the typed version (IR and NMR spectra and MS data were left out) of

J. Chem. Res. (M), 1982, 1142-1165


Some obvious errors were corrected (e.g. missing brackets or g instead of mg) but please notice that this article is a miniprint. So reading the text was quite difficult and there can bee erros especially concerning numbers. The scanned pages can bee requested via PM (each of the four has 4 MB!).

As far as Lego understands this article it seems to bee possible to obtain MD-P2P from safrole via MD-P2Pol.

An improved synthesis of indole derivatives related to Indomethacin from natural Safrole

Eliezer J. Barretro*, Paulo R.R. Costa**, Perola Regina V.R. Barros and Waldemir M. Queiroz.

* Departamento de Quimica, Universidade Federal de Sao Carlos, 13.560, Sao Carlos, S.P., Brazil.
** Nucleo de Pesquisas de Produtos Naturais, Universidade Federal de Rio de Janeiro, Bl. N, 21.941, R.J., Brazil.

The indole moiety is present in a great number of natural and synthetic biologically active compounds1. Particulary in the group of non-steroidal antiinflammatory agents (NSAIA)2, among other aryl acetic acids, indomethcain 13 occupies an outstanding position. In view of the continous efforts in the preparation and pharmacological evaluation of new NSAIA4, we became interested in preparing analogues of 1, using abundant and readily accessible natural products as inexpensive starting materials. With this objective in mind, the indolyl-acetic esters 2 and 3 were chosen as initial target compounds. Using natural safrole 45 as starting material it was possible to develop under mild conditions, a synthetic sequence of high yield in which the key in the preparation of 2 and 3 is a reductive cyclization of the appropriate nitro-[beta]-keto ester 5 and 6 (Scheme 1).

In order to obtain 5 (Scheme II) the double bond of 4 was oxidized affording epoxysafrole 7 in good yields. As expected6 this compound produces the alcohol 8 (90%) by a regiospecific oxirane ring hydrogenolysis (method B). Nitration of 8 in very mild conditions 7 (HNO3)/CHCl3) gave 9 in quantitative yield. Alternatively, the alcohol 8 (93%) was synthesized by an oxymercuration-demercuration sequence8 (method A) from 4. The next step in the choosen synthetic sequence was the preparation of the methylketones 10 and 11 that was attempted by oxidation of the alcohols 8 and 9 with pyridinium chlorochromate (PCC)9 (method A) or Jones' reagent10 (method B). The nitro methylketone 12 could be C-alkylated with ethyl bromoacetate, using the intensely violet colored benzylic enolate, prepared by treatment of 12 with sodium hydride in dimethylformamide11, to furnisch the desired ester (78%). On the other hand, the lithium benzylic enolate, generated by treatment of 10 with lithium diisopropylamide (LDA) in tetrahydrofuran 12, was C-alkylated to provide the ketoesters 11 (85%) which by nitration gave 5 (98%).

Scheme I

Scheme II

Scheme III

In order to obtain the nitro-[beta]-keto esters 6, we developed the synthetic route described in Scheme III. Thus, reaction of 4 with N-bromosuccinmide (NBS) in acetone/water produces a mixture of isomeric bromohydrins 13a-c13. This mixture was treated with sodium cyanide in methanol under reflux, and the crude reaction product 14 submitted to a careful alkaline hydrolysis14 (H2O2, EtOH, 6N NaOH) gave the desired amid 15 as the sole product to crystallize as a clean colourless material from the reaction mixture. This useful sequence can be run on a 30 g scale without isolation of intermediates, in 50% overall yield. Alternatively, the bromohydrin 13a can prepared in quantitative yield by direct bromination of 7. The formation of this dibromo compound 13a can be understood by initial bromination of the 5-position of the aromatic ring followed by regioselective opening of the oxirane15 ring by the hydrobromic acid liberated in the reaction medium. A pure sample of 14 could be obtained (65%) by reaction of 13a with sodium cyanide in methanol. The conversion of 15 into hydroxy acid 16 (90%) was attempted by akaline hydrolysis. A one pot procedure for the transformation of 16 into 17 was accomplished in 75% yield by hydrogenolysis of C-Br bond over Pd/C using methanol as solvent that permits the acid catalytic esterification by action of the hydrobromic acid formed in the reaction vessel. Finally, nitration of 17 gave 18 (90%) which, when submitted to oxidation with PCC9 or Jones' reagent10 provided the nitro-[beta]-keto ester 6 in 80% yield.

With the nitro-[beta]-keto esters 5 and 6 in hand we proceeded to the last key step of the synthetic route, viz, the reductive cyclization reaction16 to provide the indole derivations 2 and 3.

So, when a methanolic solution of 5 was submitted to catalytic reduction (H2, Pd/C 60 psi) for 15 minutes at room temperature, immediately followed by solvent elimination in vacuum and subsequent "flash" chromatography filtration over silica gel, the desired indole 2 could be obtained in 80% yield, in the form of a clear oil. On the other hand, catalytic reduction of 6, under identical conditions, afforded 3 as pale yellow crystals (85%).

In conclusion, the synthetic sequence here described is useful for the preparation of indoleacetic esters starting from the allylbenzene moiety present in various abundant natural compounds such as safrole 4. The potential biologically active indole derivatives 2 and 3 can be prepared in high overall yields (46% and 20% respectively) using easily acessible reagents and mild reaction conditions.

We are indebted to the Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (40.1424/80) for financial support and fellowships to PRVRB and VMQ, to Professor Walter B. Mors (NPPN, UFRJ), for his continous encouragement and to Professor Timothy J. Srcokson (UFSCan) for his valuable suggestions.


1. Pharm. Acta. Helv., 1973, 53, 65
2. Ann. Rept. Med. Chem., 1975, 10, 172, Ann Rept. Med. Chem., 1974, 9, 193
3. U.S Paten, 3, 242, 185 (Chem. Abstr., 1966, 64, 17555)
4. Inter-Alia: T.Y. Shen, J. Med. Chem., 1981, 24, 1; G.Y. Paris, D.L. Germaise, D.G. Cimon, L. Swett, G.W. Carter and P. Young, J. Med. Chem., 1980, 23, 9; D. Evans, D.W. Kunwell and T.A. Hicks, Australian (1970), 506, 085 (Chem. Abstr., 1980, 92, 169251t); M. Nagakura, T. Ota, N. Shimidzu, K. Kawamura, Y. Eto and Y. Wada, J. Med. Chem., 1979, 22, 48; J. Jeske and K. Stoskla, Acta Pol. Pharm., 1978, 35, 611.
5. W.B. Mors and C.T. Rizzini, "Botanica Economica", Epu, Sao Paulo, S.P., Brazil, 1976, p. 149.
6. R.L. Augustine, "Catalytic Hydrogenation", E. Arnold (Publishers), LTD, London and M. Dekker, Inc., New Yorik, 1965, p. 137
7. cf. D.R. Buckle; D.J. Outred, J.W. Ross., H. Smith, R.J. Smith, B.A. Spices and B.C. Gasson, J. Med. Chem., 1979, 22, 158.
8. H.C. Brown and P.G. Georghegan Jr., J. Org. Chem., 1970, 55, 1844.
9. E.J. Corey and W. Suggs, Tetrahedron Lett., 1975, 2647
10 The Jones' reagent was prepared as descripted in L.F. Fieser and M. Fieser, "Reagents for Organic Synthesis", John Wiley and Sons, Inc., New York, vol. I, p. 142
11. I.K. Stamos, Synthesis, 1980, 664
12 J.P. Albarella, J. Org. Chem., 1977, 42, 2009
13. The formation of 14c can be explained by the participation of a phenonium ion intermedite: Tetrahedron Lett., 1975, 4535.
14. C.R. Noller, "Organic Synthesis", Coll. II, John Wiley and Sons, Inc., New York, p. 586.
15. Treatment of epoxyde 7 with hydrobromic acid in acetone:water gives exclusively 4-(3-bromo-2-hydroxypropyl)-1,2-methylendioxybenzene.
16. We have published previously the cyclization of 12 to 5,6-methylendioxy-2-methylindole in 60% yield17. Using the modified conditions described here this transformation could be accomplished in 90% yield.
17. cf. E.J. Barreiro, P.R.R. Costa, R.T. de Wello and P.R.V.R. Barros, An. Acad. brasil. Cienc., 1981, 53, 65.


The 1H NMR spectra were run with Varian XL-100-12 (100 MHz) and Varian EM-360 (60 MHz) instruments, using the indicated solvent and Me4Si as internal reference. The i.r. spectra were recorded with a Perkin-Elmer 137-8 spectrophotometer, using KBr pellets. The u.v. spectra were recorded with a Beckman DGB spectrophotometer using EtOH (Uvasol) as solvent. The mass spectra were obtained with a Varian MAT-CH5-DF instrument coupled to a Varian MAT SS-100 computer system. The m.p. were determined in a Kofler apparatus. The hydrogenation reactions were performed in a Parr apparatus. Combustion analyses were carried out by CENPES - Petrobras (Rio de Janeiro, Brazil).

4-(2-Hydroxypropyl)-1,2-methylendioxy (8)
Method A - Oxymercuration-Demercuration of safrole (4)

Safrole (12.43 g, 76.7 mmol) was gradually added to a stirred solution of Hg(OAc)2 (25.0 g, 78.6 mmol) in THF (230 ml):H2O (75 ml). After 1 h at room temperature the yellow color was discharged and the reaction mixture was alkalinized (3M NaOH aq., 75 ml) followed by addition of a solution of NaBH4 (1.45 g, 38.15 mmol) in 3M aq. NaOH (75 ml). After 1 h the reaction mixture was saturated with NaCl, the organic layers separated and the aqeuous layer was extracted with EtOAc (4x250 ml). The combined organic layers were then washed successively with H2O (3x100 ml) and saturated brine, dried (anhy. Na2SO4) and evaporated to give 8 (12.7 g, 96%) as a clean viscous oil. An analytical sample was obtained by chromatography on silical gel using hexane-EtOAc (8:2) as eluent.

Method B - Hydrogenolysis of epoxysafrole (7)

A mixture of 7 (5.0 g, 27.7 mmoles) and 10% Pd/C (1.0 g) in MeOH (30 ml) was submitted to hydrogen pressure (65 psi). After 3 h the catalyst was filtered and the solvent evaporated to affort 8 (4.55 g, 90%) as a clear brown oil.

Procedure for nitrations


To a solution of 8 in CHCl3 (110 ml) was added HNO3 (d = 1.52, 3.62 ml). After 40 minutes at room temperature the reaction was neutralized with Na2CO3 aq., washed with H2O, dried (Na2SO4) and evaporated furnishing 9. (0.75 g, 100%) as yellow solid: m.p. 117-119°C from AcOH:H2O

5-Nitro-4-(2-oxo-1-carboxymethylpropyl)-1,2-methylendioxybenzene ethyl ester (5)

In a 1.5 mmol scale 11 was nitrated as described above to give as product a clear brown oil. This residue crystallized from AcOH-H2O gave 5 (80%): m.p. 96-98°C

5-Nitro-4-(3-methoxycarbonyl-2-hydroxypropyl)-1,2,-methylendioxybenzene (18)

In a 30 mmol scale 17 was nitrated as already described to give a brown solid. Recrystallization of the crude product from AcOH:H2O gave as yellow crystals: m.p.. 106-109°C

PCC-Oxidations (Method A)

Oxidation of 5-nitro-4-(3-methoxycarbonyl-2-hydroxypropyl)-1,2-methylendioxybenzene

A mixture of 18 (1.41 g, 5.0 mmol) and PCC (3.23, 15.0 mmol) in CH2Cl2 (75 ml) was allowed to react overnight at room temperature. The solvent was separated by filtration and residue washed with CH2Cl2 (2 x 50 ml). The combined organic fractions were evaporated and the residue filtered over silica gel (40 g) using CH2Cl2 as eluent, giving 6 (1.11 g, 80%) as a yellow solid: m.p. 86°C from AcOH-H2O

Oxidation of 4-(2-hydroxypropyl)-1,2-methylendioxybenzene (8

In a 3 mmol scale the oxidation of 8 as described above followed by filtration of the crude product over silica gel gave 10 (75%) as a colorless oil:
vmax 1720 cm-1, [delta]H (60 MHz, CCl4) 6.60 (3 H, narrow, m), 5.87 (2 H, s), 3.46 (2 H, s), 2.00 (3 H, s) ppm; m/z 178 (M+, 20%), 135 (100%).

Oxidation of 5-nitro-4-(1-hydroxypropyl)-1,2-methylendioxybenzene (9)

In a 30 mmol scale the oxidation of 9 as already described gave 12 (86%) as yellow crystals: m.p. 139-141°C from AcOH-H2O

Jones Oxidations (Method B)

Oxidation of 5-Nitro-4-(2-hydroxypropyl)-1,2-methylendioxybenzene (9)

To an ice cold solution of 9 (1.1 g, 4.9 mmol) in CH3COCH3 redistilled from KMnO4 (70 ml) was added dropwise a solution of Jones' reagent10 until a slight excess was present (2.0 ml), followed by addition of a saturated solution of NaHSO3 (10 ml). The organic layer was separated and the residue extracted with EtOAc (3 x 100 ml). The combined organic extrats were washed with aq. NaHCO3 (50 ml), H2O (50 ml), saturated brine, dried (Na2SO4) and evaporated. The residue crystallized from AcOH-H2O gave 12 (0.89 g, 80%).

Oxidation of 4-(3-hydroxypropyl)-1,2-methylendioxybenzene (8)

In a 1.5 mmol scale 8 was oxidized as above giving 10 in 78% yield, after purification by preparative TLC over silica gel.

Oxidation of 5-Nitro-4-(3-methoxycarbonyl-2-hydroxypropyl)-1,2-methylendioxybenzene (18)

In a 10 mmol scale the Jones' oxidation of 18 was run as above affording 6 as a crystalline yellow solid crystallized from AcOH-H2O in 72% yield.


  • Guest
From Safrole to anywhere, Pt. 2
« Reply #1 on: September 16, 2003, 03:11:00 AM »
Alkylation of 5-Nitro-4-(2-oxopropyl)-1,2-methylendioxybenzene (12)

A solution of 12 (0.5 g, 2.2 mmol) in dry DMF (20 ml) was added slowly with stirring to a cooled suspension of NaH (0.10 g, ~3.0 mmol) in DMF (2 ml). The violet colored mixture was stirred for 1 h at room temperature, followed by cooling (0°C) and rapid addition of ethyl bromoacetate (0.35 ml, 3.17 mmol). The violet color disappeared and after 0.5 the reaction mixture was diluted with cold water (40 ml) and extracted with EtOAc (4 x 50 ml). The organic layer was washed with H2O, saturated brine, dried (Na2SO4), and evaporated giving a clear brown oil. Filtration of the crude product through silica gel (CH2Cl2 as eluent) gave 5 as pale yellow crystals (0.54 g, 78%).

Alkylation of 4-(2-oxopropyl)-1,2-methylendioxybenzene (10)

A solution of 10 (0.2 g, 0.56 mmol) in dry THF (6 ml) was slowly added at -78°C to a solution of lithium diisopropylamide in 2 ml of THF (prepared from 1.76 ml of 1.42 n-butyllithium in hexane 0.36 ml of diisopropylamine). The mixture was stirred for 15 min at -78°C and warmed to room temperature during 30 min. After cooling to -78°C ethyl bromoacetate (0.16 ml, 1.5 mmol) was added and the mixture was allowed to stir for 2.5 h at -78°C and then the reaction was quenched with 2.0 ml of saturated NH4Cl. Ether (35 ml) and H2O (5 ml) was added, and the layers seperated. The organic layer was washed successively with 10% HCl (3 x 25 ml), H2O (5 ml), saturated brine, dried (Na2SO4) and evaporated. The crude product obtained as a clear brown solid was filtered over silica gel giving an incolor oil (0.22, 11, 85%)

Epoxysafrole (7)

Peracetic acid was added to a cold solution of 4 (10 ml, 67.0 mmoles) in CHCl3 (500 ml), and the mixture was left at room temperature for 72 h. The CHCl3 solution was washed successively with H2 (150 ml), 5% NaHCO3 (3 x 50 ml), 10% aqueous NaI (50 ml), 10% NaHSO3 (50 ml) and saturated brine (2 x 50 ml), dried over Na2SO4 and evaporated. Chromatography of the residue on silical gel with n-hexane afforded 7 (8.56 g, 80%) as an incolor oil: [delta]H(100 MHz, CDCl3) 6.69 (3 H, narrow, m), 5.90 (2 H, s), 3.06 (1 H, m), 2.34-268 (3 H, m), 2.50 (1 H, dd, J=5.0 and J=3.0 Hz) ppm; m/z (M+ 178) (55%), 135 (100%).

5-Bromo-4-(3-bromo-2-hydroxypropyl)-1,2-methylendioxybenzene (13a) and 5-bromo-4-(3-hydroxy-2-bromopropyl)-1,2-methylendioxybenzene (13b

To a solution of 4 (0.6 g, 3.7 mmol) in acetone (8.0 ml), H2O (2.0 ml) was added portionwise 1.58 g (8.0 mmol) of NBS freshly crystallized from H2O. After stirring 1.5 h at room temperature the solvent was evaporated resultng in a water insoluble oily residue which was soluble in EtOAc (100 ml). The organic layer was washed successively with 10% NaHCO3 (3 x 30 ml), water (3 x 30 ml) and saturated brine, dried (Na2SO4) and evaporated to give a crude prodcut as a dark brown oil (90%). Application of this product to a column of silica gel and elution with 50% hexane-CHCl3 afforded 13a* (0.45 g, 40%), which was recrystallized from hexane to give the analytical sample: m.p. 91-92°C
13b (0.13 g, 12%) which was also recrystallized from n-hexane to give an analytical sample: m.p. 90-91°C
a third compound (13c) was eluted from the chromatographic column as an oil
* This compound was the only product obtained in quantitative yield by direct bromination of the epoxy safrole (7) in acetone-water as solvent.

5-Bromo-4-(3-cyano-2-hydroxypropyl)-1,2-methylendioxybenzene (14)

To a solution of 13a (1.7 g, 5.0 mmol) in MeOH (10 ml) was added portionwise 0.98 g (5.0 mmol) of NaCN. After 4 h under reflux the solvent was evaporated and the residue was dissolved in 200 ml of EtOAc. The organic solution was successively with water (4 x 100 ml), saturated brine and dried (Na2SO4). After evaporation a brown oil was obtained, which was purified over silica gel using 5% MeOH in CHCl3 as eluent, to afford 14 (0.92 g, 65%)

5-Bromo-4-(3-carbamoyl-2-hydroxypropyl)-1,2-methylendioxybenzene (15)

To a solution of 14 (8.0 g, 28.1 mmol) in EtOH (60 ml) was added dropwise 1.2 ml aq 6N NaOH followed by 18 ml of 30% H2O2. The initial exothermic reaction (40-50°C, ~1 h) was maintained with stirring at 50°C for 3.5 h. After cooling to 0°C in ice-bath the mixture was filtered to furnish a white precipitate which was washed with cool water, dried in vacuum (78°C) and recrystallized from EtOH to give the amide 15 (7.65 g, 90%): m.p. 200-202°C
The amide (15 ) can be synthesied from safrole (4) in 30 g scale with 50% yield using the sequence already described here without bromohydrins (13a-c) separation, in a practical experimental procedure that consist in the treatment of the crude product (13a-c) with NaCN in EtOH followed by solvent evaporation and subsequent treatment with H2O2/NaOH/EtOH. The amide 15 was the sole product to crystallize from the reaction medium.

Hydrolysis of 5-bromo-4-(3-carbamoyl-2-hydroxypropyl)-1,2-methylendioxybenzene (15)

A mixture of 6.0 g (20 mmol) of 15 and 400 ml of 1N NaOH was refluxed with vigorous stirring until ammonia elimination was finished (~8h). The reaction mixture was cooled (0°C) and acidified with conc. HCl. The precipitate formed was filtered, washed with water and dried in vacuum (78°C), affording 16 as a solid: m.p. 156-158°C (MeOH)

4-(3-methoxycarbonyl-2-hydroxypropyl)-1,2-methylendioxybenzene (17)

A solution of 16 (13.3 g, 55.8 mmol) in MeOH (100 ml) and 10% Pd/C (1.0 g) was submitted to hydrogen pressure (40 psi) for 2 h at room temperature. The catalyst was removed by filtration and washed with MeOH. The combined solvents were treated with a mixture of EtOAc (500 ml) and aq. sodium carbonate (3 x 100 ml). The organic layer was successivly washed with water (2 x 50 ml), saturated brine and dried over anhydrous Na2SO4, filtered and evaporated to give 17 (9.1 g, 87%). An analytical sampe was obtained after filtration of the product over silical gel using 20% EtOAc-hexane to give a colourless oil:

Synthesis of 2-methyl-5,6-methylendioxyindole-3-acetic acid ethyl ester (2) and 5,6-methylendioxyindole-2-acetic acid methyl ester (3)

A mixture of 5 (100 mg, 0.32 mmol) or 6 (100 mg, 0.36 mmol), methanol (5 ml) and 10% Pd/C (20 mg) was submitted to hydrogen pressure (40 psi) during 15 min at room temperature. The catalyst was removed by filtration and methanol was carefully evaporated to give a clear oil, that was quickly filtred over silica gel (15 g) in a vacuum chromatography column using dry dichlormethane as eluent. The desired fractions after catious evaporation gave 2 (67 mg, 80%) as an oil


  • Guest
HUGE smiles!!!!
« Reply #2 on: September 16, 2003, 07:29:00 PM »
Holy shit!!! I know this is the veritable gold mine that many bees have been looking for for years! No more 8-hour long wackers to get to the beloved ketone!!!!
Lego, my hat's off to you for finding this.... WOOHOO!
Anyone wanna try a scaled-up run and post a writeup?
Also, THF is very hard to get... can any substitutions bee made here? thanks peace


  • Guest
One of the best articles posted lately
« Reply #3 on: September 17, 2003, 01:38:00 AM »
The article is really a gem, and a hard-to-find gem as well. I didn't believe it at first when I found the abstract for it, it simply seemed to have everything, despite its main topic being as boring as 'indolic non-steroidal antiinflammatory agents'... But after having displayed the citation in the "wanted references" thread for a while, suddenly three people emailed me the article almost simultaneously, but as it was in very small print, it was very hard to read without zooming. But now, thanks to Lego (good job!), we can finally all read it!

Not only are all the obvious transformations in this article nice, but there are also a few 'implicit' synthetic routes in it, which may be missed by many:

* Heating 6-Bromo-Safrole Bromohydrin (13a) or its isomer 13b with a solution of potassium carbonate will ring-close the bromohydrin to the corresponding epoxide, 6-Bromo-Safrole Epoxide (Like compound 7 with a bromine atom added on the ring's 6-position).

* Hydrogenation of the above 6-Bromo-Safrole Epoxide (H2, 40-60 psi, 10% Pd/C, MeOH) will hydrogenolyze the aromatic bromine atom (just like 16 is easily transformed to 17 under those conditions), and the 2,3-epoxy side chain will be reduced to the 2-alcohol (just like 7 is reduced to 8 in 90% yield under those conditions). Thus this one-pot removal of the bromine atom and reduction of the epoxide to the 2-alcohol will thus result in MDP2Pol.

* Now when we have a working method of epoxidizing Safrole to its terminal Epoxide, let's go take a look at the methods we have at our disposal for turning that one directly into MDP2P - one example is the procedure by C. Venturello, who has isomerized the terminal epoxide of safrole in almost quantitative yield using

anhydrous LiI in tetraglyme (130°C, 5h)

( . There are several other methods out there for isomerizing allylbenzene epoxides to phenylacetones, but this is one of the best.

Finally, I must discourage poeople from using the oxymercuration-demercuration of safrole using mercury(II)acetate. Mercury is a cumulative toxin, and is not only bad for your health, but also the environment. The method uses twice the weight of mercury(II)acetate compared to safrole!


  • Guest
The Reaction of Safrole With Bromine
« Reply #4 on: September 18, 2003, 12:42:00 PM »
Here is an article which relates to the bromination reactions discussed above:

Solvent Effects in the Reaction of Safrole With Bromine
Paulo R.R. Costa & Jaime A. Rabi

Tetrahedron Letters 51, 4535-4538 (1975)