3-phenylpropan-1-ol experiment

[Sydenhams chorea]

3-Phenylpropan-1-ol (hydrocinnamyl alcohol) is commonly available as a perfumery ingredient and should be easily prepared from the more available cinnamyl alcohol or cinnamaldehyde (cinnamon bark oil, cassia oil) by various means of reduction. For the following experiment the patent DE819400 was mainly used as a reference, and the following route is envisioned:


In the patent procedure 95 grams of 3-phenylpropan-1-ol is heated with 250 grams of "paraffin oxydation fatty acids" (basically any from a range of saturated fatty acids can be adapted) at 250°C for 3 hours, thus forming an ester. The heat is then increased to 340°C, upon which a mixture of (mainly) allylbenzene and propenylbenzene distills over. Increasing this temperature increases the propenylbenzene to allylbenzene ratio.

25 grams of freshly distilled 3-phenylpropan-1-ol was placed in a beaker and 66 grams of stearic acid (cosmetics grade) was added. This mixture was heated gently until all stearic acid had molten and 10 drops of 31% HCl were added as an esterification catalyst. The temperature was increased to 150°C with stirring, upon which strong bubbling was observed (water evolving). The patent asks for the mixture to be heated at 250°C until no more water distills, but considering that there was considerable water vapour evolving for more than an hour at 150°C, and the fact that the boiling point of 3-phenylpropan-1-ol is 235°C, it was concluded that the patent conditions are perhaps overkill, and it was decided to keep the mixture at 150°C for 3 hours. After 90 minutes the bubbling subsided and no more water vapour seemed to be evolved.

Note that this must be performed in a fume hood or in a well-ventilated area, as the smell of fumes, although not particularly toxic or corrosive, can be quite overpowering. The mixture was allowed to cool to 90°C and 15 ml toluene was added (careful!), and the temperature was again brought to 150°C until all toluene boiled off. (This was adapted from US4510093, where xylene is used as a "withdrawing agent" to speed up water removal as the lower-boiling azeotrope and as such the formation of the ester. Xylene was not at hand, so toluene was used.

After heating for 3 hours at 150°C the odour of the mixture changed completely, from the initial pleasant flowery scent of the hydrocinnamyl alcohol to a sharper, rather unpleasant smell resembling motor oil at the end.

A distilling flask was created from a scouting-type aluminium drinking bottle, a B14 three-way adaptor was fitted in the bottleneck with the help of Teflon-tape, it was wrapped in aluminium foil, and a water-cooled Liebig-condensor was attached (the black ester oil was added to the distilling flask first):



The flask was gently heated with a gas flame. Cracking noises were heard in the flask (formation of water) and ~25ml of a cloudy destillate was collected (from which a couple of milliliters separated):


(pic taken during the distillation process)

The mixture has not been characterized yet but it is highly suspected that it is a mixture of mainly allylbenzene and some propenylbenzene (and water). The smell of the distillate is again completely different, best described as a combination of an aromatic scent à la benzene or toluene and the stingy, pungent odor of allyl alcohol.

To characterize the end-product it will be completely converted to propenylbenzene by refluxing with KOH.

[Goldmember]

Nice! Do you have a preferred reduction in the first instance?

[Quantum Dude]

Great post. Very interesting.  Any specific reasons for using 31 % HCl instead of another mineral acid like H3PO4 for esterification ? Has the mixture of allylbenzenes been treated with polymerization inhibitors at all ?

[Sydenhams chorea]

Nice! Do you have a preferred reduction in the first instance?

I distilled mine from a commercially available source, but in the future I would definitely opt for electroreduction of cinnamaldehyde at a lead cathode, especially considering the wide availability and cheapness of cinnamon bark oil (90% cinnamaldehyde), and the ease of such an electroreduction (I have good experience with this technique, see elsewhere on this forum).


Here are some references on different routes for the preparation of 3-phenyl-1-propanol I've copied from Orgsyn:

Ethyl cinnamate has been reduced to 3-phenyl-1-propanol with sodium and ethanol,
Bouveault and Blanc, Compt. rend., 137, 328 (1903); Bull. soc. chim. France, [3] 31, 1209 (1904).
Shorygin, Bogacheva, and Shepeleva, Khim. Referat. Zhur., 1940, 114 [C. A., 36, 3793 (1942)].
Clutterbuck and Cohen, J. Chem. Soc., 123, 2509 (1923).
Krushevskii and Gildburg, Russ. pat. 31,008 [C. A., 28, 3425 (1934)].

hydrogen and copper chromite,
Adkins and Folkers, J. Am. Chem. Soc., 53, 1095 (1931); Wojcik and Adkins, J. Am. Chem. Soc., 55, 4939 (1933).

and sodium and ammonia.
Chablay, Compt. rend., 156, 1022 (1913).

The alcohol has also been prepared by reduction of the glyceride of cinnamic acid with sodium and amyl alcohol;
Darzens, Compt. rend., 205, 682 (1937).

by reduction of cinnamic acid with lithium aluminum hydride;
Nystrom and Brown, J. Am. Chem. Soc., 69, 2548 (1947).

by reduction of cinnamoyl chloride with sodium borohydride;
Chaikin and Brown, J. Am. Chem. Soc., 71, 122 (1949).

and by reduction of ethyl dihydrocinnamate with sodium and ethanol.
Bouveault and Blanc, Compt. rend., 137, 328 (1903); Bull. soc. chim. France, [3] 31, 1209 (1904).
Burrows and Turner, J. Chem. Soc., 119, 428 (1921).

Cinnamaldehyde has been reduced to 3-phenyl-1-propanol with hydrogen and palladium.
Straus and Grindel, Ann., 439, 307 (1924).
Endoh, Rec. trav. chim., 44, 869 (1925).

platinum,
Skita, Ber., 48, 1692 (1915).
Tuley and Adams, J. Am. Chem. Soc., 47, 3063 (1925).
Vavon, Compt. rend., 154, 361 (1912); Ann. chim. et phys., [9] 1, 166 (1914).

or nickel,
Adkins and Billica, J. Am. Chem. Soc., 70, 695 (1948).
Braun and Kochendorfer, Ber., 56, 2175 (1923).
Delepine and Hanegraaff, Bull. soc. chim. France, [5] 4, 2087 (1937).
Palfray, Sabetay, and Gauthier, Compt. rend., 218, 553 (1944).

nickel in alkaline solution (no hydrogen),
Papa, Schwenk, and Whitman, J. Org. Chem., 7, 587 (1942).

lithium aluminum hydride,
Nystrom and Brown, J. Am. Chem. Soc., 70, 3738 (1948).

electrolysis at a mercury
Shima, Mem. Coll. Sci. Kyoto Imp. Univ., A12, 70 (1929).

or lead electrode,
Law, J. Chem. Soc., 101, 1030 (1912).

and with an unmentioned catalyst.
Walter, Ger. pat. 296,507 [C. A., 13, 368 (1919)].

Reduction of cinnamyl alcohol to 3-phenyl-1-propanol has been effected by use of sodium and ethanol,
Gray, J. Chem. Soc., 127, 1156 (1925).

sodium amalgam and water,
Hatton and Hodgkinson, J. Chem. Soc., 39, 319 (1881).
Rugheimer, Ann., 172, 123 (1874).

hydrogen and nickel
Tanaka, Chem. Ztg., 48, 25 (1924).

or palladium,
Straus and Berkow, Ann., 401, 151 (1913).

sodium and ammonia,
Chablay, Compt. rend., 143, 829 (1906); Ann. chim. et phys., [9] 8, 191 (1914).

and lithium aluminum hydride.
Hochstein and Brown, J. Am. Chem. Soc., 70, 3484 (1948).

Other syntheses have been brought about by reduction of ethyl ?,?-epoxy-?-phenyldihydrocinnamate with sodium and amyl alcohol;
Verley, Bull. soc. chim. France, [4] 35, 488 (1924).

by reduction of ethyl benzoylacetate with hydrogen and copper chromite;
Adkins and Billica, J. Am. Chem. Soc., 70, 3121 (1948).

by reduction of acetonephenyllactic acid with hydrogen and copper chromite;
Oeda, Bull. Chem. Soc. Japan, 10, 531 (1935).

by reaction of ethyl alcohol with sodium benzylate;
Guerbet, Compt. rend., 146, 300 (1908); Bull. soc. chim. France, [4] 3, 503 (1908).

by reaction of benzylmagnesium chloride with a mixture of ethylene chlorohydrin and ethylmagnesium chloride;
Conant and Kirner, J. Am. Chem. Soc., 46, 241 (1924).

by reaction of trimethylene oxide with phenylmagnesium bromide;
Searles, J. Am. Chem. Soc., 73, 124 (1951).

by condensation of benzylmagnesium chloride with ethylene oxide;
Ohara, Jap. pat. 4364 (1950) [C. A., 47, 3347 (1953)].

and by hydrogenolysis of 1-phenyl-1,3-propanediol over nickel-on-kieselguhr.
Arnold (to E. I. du Pont de Nemours & Co.), U. S. pat. 2,555,912 [C. A., 46, 3074 (1952)].

The reduction of 4-phenyl-m-dioxane to give 3-phenyl-1-propanol, as described here, is based on the procedure of Beets,
Beets, Rec. trav. chim., 70, 20 (1951).

who used sodium and diisobutylcarbinol. Other substituted m-dioxanes may also be converted to substituted 3-aryl-1-propanols by the same procedure.
Beets and Van Essen, Rec. trav. chim., 70, 25 (1951); Drukker and Beets, Rec. trav. chim., 70, 29 (1951).

3-Phenyl-1-propanol also has been obtained in 85% yield by the reduction of 4-phenyl-m-dioxane over copper chromite catalyst.
Emerson, Heider, Longley, and Shafer, J. Am. Chem. Soc., 72, 5314 (1950).

EDIT: added some of the above papers

[Goldmember]

Wow! Thanks,much  food for thought all in one convenient outlet.

Stupid question time, Would something similar result if phenylalanine was esterfied with stearyl or other fatty alcohol?

[Sydenhams chorea]

Great post. Very interesting.  Any specific reasons for using 31 % HCl instead of another mineral acid like H3PO4 for esterification ?

Because it's volatile. According to the patent the ester forms fine without any catalyst, the weakly acidic properties of stearic acid (pKa 10.15) apparently being sufficient. I figured the addition of HCl(aq) would speed up the formation of the ester, while after a short while of heating at 150°C all HCl would have been evaporated.

The HCl addition and the toluene as a "withdrawal agent" can probably be omitted and exchanged with a more professional approach such as a Dean-Stark trap.

Has the mixture of allylbenzenes been treated with polymerization inhibitors at all ?

Good point, I guess I should add some BHT or hydroquinone. But I wasn't planning on storing it too long, but begin the isomeration with KOH.

[Sydenhams chorea]

Propenylbenzene has a sweetish smell, it reminds somewhat of isosafrole. Much better than allylbenzene, I could smell this all day.

[coincoin]

Very nice... any expectations on yield ? characterization going on ?

[Sydenhams chorea]

Currently running another batch, will distill all propenylbenzene in one go.

Yields look quite good so far, similar volume of allylbenzene is obtained from 3-phenylpropan-1-ol, and the isomerization procedure, adapted from Osmium's method for isosafrole which refluxes isosafrole over KOH in a light vacuum around 130°C (but here the vacuum is omitted as the bp of allylbenzene is low enough, 156-157°C vs 232°C for safrole), yield around the same on first sight using TLC (looks like only propenylbenzene + crud after 16 hours, 85% is reported for safrole, still have to distill).

(edit: note that Rf values are about the same using chloroform:ethyl acetate:formic acid (85:15:1) but the smell is markedly different, I assume if allylbenzene still be present it would be a large spot/streak. Granted this isn't the best method to speculate upon, there is wishful thinking from my part...)

Man, I will be able to vaporize propenylbenzene in my living room for over a year, the thought of breathing that delicious sweetish odour every single day makes me so happy with life...

[zgoat65]

Awesome post!  Pleez keep us up to speed on yer final results.   joe says he has 32oz of cassia oil just layin around doing nothing.  Propenylbenzene sounds so much cooler

[Sydenhams chorea]

Awesome post!  Pleez keep us up to speed on yer final results.   joe says he has 32oz of cassia oil just layin around doing nothing.  Propenylbenzene sounds so much cooler

It is much cooler! Smells much better than any type of cinnamon, take my word for it.

In retrospect, the addition of aq. HCl as esterification catalyst is pretty stupid as it probably doesn't really do much as pertaining to dehydrating properties, key factor for a Fischer esterification. The reasonable molar excess of stearic acid, heating to at least 150°C and the use of toluene or preferably xylenes as withdrawal agent should be sufficient. A Dean-Stark trap would be neat and esthetically pleasing.

[zgoat65]

This is very appealing to me.  And all that lacks is the stearic acid. 

[Sydenhams chorea]

This is very appealing to me.  And all that lacks is the stearic acid. 

http://www.howtomakecandles.info/cm_article.asp?ID=ADDIT0003

Guess that means you won't be inviting me over for any candle-light dinners?

[zgoat65]

I was just about to pm you about my search results for  "cosmetic grade" stearic (and  candle makin was the majority of what i found).  Thanx for that clarification.  Gots to get my CC out and get me a candle kit on the way.  The wifey might like to actually make some candles as a hobby.  We could be all craftsy and stuff.

[coincoin]

stop sniffing propenylbenzene man don't get cancer before we know the outcome of this... send me the mass spectrometer data's in PM ^^

[Sydenhams chorea]

coincoin: don't have access to modern chromatographic or spectrophotometric equipment, sorry, TLC-thumping basement hobby chemist here... BTW Cancer is worth it, I'll start huffing that propenylbenzene shit. That sweet, tasty smell totally inebriates me.

zgoat: be aware that both stearic acid and stearin (the triglyceride) are used in candle making. Also, for cosmetics grade you ought to look into those new-age style shops that sell raw materials for making shampoos, perfumes etc. BTW palmitic acid will also work, as will any from a broad array of saturated fatty acids. Aim for a boiling point well above 300°C. Palmitic acid is the late Linda McCarthy's veggie alternative to cosmetics grade stearic acid.

*
Furthermore I want to add that the yield of my second batch was less spectacular than my first, yielding ~40 ml allylbenzene (after drying over MgSO₄) from 70 ml 3-phenylpropan-1-ol (+ 184 grams stearic acid). I did have the "luminous" idea of heating the mixture of alcohol and acid at 150°C for 5 hours in an open beaker  .

(As a side-note: Why do I always get such bright ideas when I'm seriously deprived of sleep and should be postponing my experiments rather than not knowing when to stop?  Obviously I lost a good deal by evaporation. Folks, don't be an idiot like me, use a Dean-Stark trap, or distill away the formed water as a low-boiling azeotrope with toluene/xylene.)

Also, when heating the ester with a moderate gas flame, you need some patience for the allylbenzene to form and distill over. Highest peak I had during both runs was 1 drop every 10 seconds. The greater part of the run it's more like 1 drop every 20-60 seconds. Cracking sounds (water condensate hitting hot oil) are always a good sign.

Another small note, at the end of the run, don't try to force the last drops out by liberally increasing the gas flame, when you see a thick smoke come over, it will solidify in the condensor because it's stearic acid (bp 383°C). This is the proverbial Porky Pig, i.e. "that's all folks".

[Sedit]

This is very appealing to me.  And all that lacks is the stearic acid. 

http://www.howtomakecandles.info/cm_article.asp?ID=ADDIT0003

Guess that means you won't be inviting me over for any candle-light dinners?

Or using shaving cream for that matter which it is a major constitute that drops out when shaving cream is diluted with water. Try it for yourself, a white precipitate of Stearic acid will drop out.


Sydenhams perhaps during the Ester pyrolysis you could figure out a means of trapping the water so that it does not form vapors that might be constantly fighting your heat source by attempting to lower the temperature to the boiling point of water.

[Sedit]

Do you believe that the reverse ester this would perform the same way?

If one where to form an ester with dihydrocaffeic would give the same resulting Allylbenzene?

Basically what I want to know is what becomes of the Stearic acid after the elimination of the water?

[lugh]

Obviously I lost a good deal by evaporation. Folks, don't be an idiot like me, use a Dean-Stark trap, or distill away the formed water as a low-boiling azeotrope with toluene/xylene.

This old post from the Hive might bee illuminating:

http://www.erowid.org/archive/rhodium/chemistry/equipment/simple2fancy.html

 

[coincoin]

Did you give another try to see the yields in a closed vessel ?