The Vespiary

The Hive => Novel Discourse => Topic started by: camxyl on April 06, 2004, 10:01:00 AM

Title: Benzodioxin MDA analogue?
Post by: camxyl on April 06, 2004, 10:01:00 AM
I understand how easy it is for any chemical novice to sit on chemdraw and draw up *new* exciting possible drugs but just thought I would throw this chemical into discussion. 

2,3-Dihydro-1,4-benzodioxin-6-aminopropane (?think thats correct)


The benzaldehyde appears to be available...

CAS = 29668-44-8

Pubmed showed that the aminoethanol analogue might have beta-blocking ability but thats all I could find...

2-Benzodioxinylaminoethanols: a new class of beta-adrenergic blocking and antihypertensive agents.

Lalloz L, Loppinet V, Coudert G, Guillaumet G, Loubinoux B, Labrid C, Beaughard M, Dureng G, Lamar JC.

Various 2-benzodioxinylaminoethanol derivatives were synthetized and investigated for beta-adrenergic blocking activity. Most compounds demonstrated a beta-blocking activity of a competitive type when evaluated in guinea pig atrial and tracheal preparations. Three compounds were more potent than practolol and propranolol. All compounds demonstrated antihypertensive properties in spontaneously hypertensive rats. The most active compound was 1-(1,4-benzodioxin-2-yl)-2-[N4-(2-methoxyphenyl)piperazino]ethanol (11), which at 2.5 mg/kg iv lowered blood pressure by 41%.

PMID: 6120237 [PubMed - indexed for MEDLINE]

Anyway, just wondered if anyone had any thoughts?

- OL
Title: Possible, but not very likely
Post by: Rhodium on April 06, 2004, 10:29:00 AM

Pihkal #110 (


This substance is rather similar, but has been shown to be relatively ineffective. I do not think that the MDA analog you drew up would show any MDMA-like action either, as anything more bulky added to the methylenedioxy ring has been shown to be rather ineffective. This is what Shulgin has to say on a similar topic: (

Title: Looks like those compound releases formaldeyde
Post by: Lilienthal on April 06, 2004, 12:01:00 PM
Looks like those compound releases formaldeyde on contact with acidic medium... Doesn't sound very healthy  :)
Title: A more interesting possibility?
Post by: Kinetic on April 07, 2004, 01:37:00 PM
The MDA/IAP 'hybrid' 5-(2-aminopropyl)-2,3-dihydrobenzofuran pictured below has been made by the Nichols research group but has recieved little attention here (save for

Post 476448 (

(Rhodium: "5-(2-aminopropyl)-benzofuran", Novel Discourse)
and a couple of other posts), although it has been touted as a 'non-neurotoxic MDMA analogue' (according to psychokitty in

Post 108607 (missing)

(dormouse: "Better than honey?  -Bella*****", Novel Discourse)
). Here it is:

Molecule: (

Non-neurotoxic MDMA analogue ("NC(C)Cc2ccc1OCCc1c2")

The benzaldehyde is not commercially available and the 2,3-dihydrobenzofuran precursor (which can then be formylated to the benzaldehyde) is expensive. But consider the following scheme:

Molecule: (

Dibromination of phenol ("c1ccccc1O>>c1cc(Br)cc(Br)c1O")

This appears to give the majority of the desired isomer best upon treatment of phenol with molecular bromine. The dibrominated phenol can then be alkylated with 1,2-dibromoethane and a base:

Molecule: (

Etherification ("c1cc(Br)cc(Br)c1O.BrCCBr>>c1cc(Br)cc(Br)c1OCCBr")

The next step is my favourite: A one-pot intramolecular (Parham) cyclisation followed by formylation upon treatment with DMF. The compound is first treated with 2-3 equivalents of ethylmagnesiumbromide. Because the aromatic ring gives a more stable Grignard than the ethylmagnesiumbromide, the Mg cation migrates, and the resulting aryl Grignard displaces the bromide on the end of its connected ethyl chain. The Mg cation from the other equivalent of Grignard then inserts between the C-Br bond para to the oxygen, which after treatment with DMF and hydrolysis gives the benzaldehyde. I've drawn it as two steps so it can be seen properly:

Molecule: (

Benzaldehyde formation part I ("c1cc(Br)cc(Br)c1OCCBr.[Mg++]Br.>>Br[Mg++]c2ccc1OCCc1c2")

Molecule: (

Benzaldehyde formation part II ("Br[Mg++]c2ccc1OCCc1c2.C(=O)N(C)(C)>>C(=O)c2ccc1OCCc1c2")

References for the dibromination:

Benzyltrimethylammonium tribromide in DCM/methanol, 87% yield: Bull.Chem.Soc.Jpn.; 60; 11; 1987; 4187-4189.

Bromine in CCl4: J.Prakt.Chem.; 332; 6; 1990; 1093-1098.

Bromine in GAA: J.Chem.Soc.; 85; 1904; 1228.

For the etherification:

Dibromophenol, dibromoethane, NaOH (aq.), 59% after 22h heating: J.Org.Chem.; 46; 7; 1981; 1384-1388.

For the benzaldehyde formation:

Tetrahedron Letters 41 (2000) 2269–2273

J.Org.Chem. 1981, 46, 1384-1388

Once you get the benzaldehyde it should be plain sailing with the usual Knoevenagel and reduction to the amine. It's likely that the yield of the etherification can be increased by using a more modern PTC method (and a large excess of dibromoethane: see

Post 196477 (missing)

(hest: "Re: New Amph.  more potent than LSD", Serious Chemistry)

When I get my hands on the remaining reagents required I'll give the above proposal a go. I've already tried the cyclisation/formylation but my vacuum isn't working and the impure benzaldehyde has been waiting for weeks already. Plus I'm pretty certain there will be some 4-ethoxybenzadehyde which will be difficult to separate. I'll probably distill to get an idea of the yield and then discard whatever I get back. I don't fancy ending up with a lot of 4-ethoxyamphetamine by accident.


Looks like those compound releases formaldeyde on contact with acidic medium... Doesn't sound very healthy  :)

Won't the same apply to MDMA? The 'acetal' between the oxygens is the same in both cases. I was wondering about this as a possible explanation for the far greater instability of the dioxole ring to Lewis/protic acids in comparison to the plain aryl ethers. Anisole, for example, is stable to standard Friedel-Crafts conditions whereas benzodioxole is entirely demethylated (AlCl3/acid chloride/DCM at 0oC).

Title: interesting!
Post by: Rhodium on April 07, 2004, 04:53:00 PM
Kinetic: Do you have those refs handy, so that you could post a suggested experimental?

Title: Unsure why it's active
Post by: Fastandbulbous on April 07, 2004, 09:20:00 PM
Hi Kinetic,

I’m a little puzzled as to the activity of the dihydrobenzofuran deriv that you mentioned in your post above. The reason being that in a previous paper (1), Nichols, investigated the activity of MMAI (5-methoxy-6-methyl-2-aminoindane) and MMA (3-methoxy-4-methylamphetamine) and found that they both substituted for MDMA in rats trained to discriminate MDMA activity. MMA has the methyl group in the para position and the methoxy group in the meta position with respect to the isopropylamine side chain (as MMAI has the isopropyl side chain formed into a symmetrical 5 membered ring, you cant really refer to para and meta orientation of the ring substituents), which makes sense, as it is effectively DOM with one of the methoxy groups removed. What puzzles me is that with the dihydrobenzofuran deriv you mention, the oxygen atom is in the para position (equates to para methoxy) to the sidechain, and the direct attachment of the carbon atom is in the meta position (equates to meta methyl), which has the orientation of the ring substituents the other way around; after checking, I cannot find any reports of 4-methoxy-3-methylamphetamine being active at all, nevermind substituting for MDMA in trained rats. As such, I would have thought that the benzofuran with the structure given below would be the compound with activity similar to MMA (and hence MDMA)


ref (1): Synthesis and pharmacological examination of 1-(3-methoxy-4-methylphenyl)-2-aminopropane and 5-methoxy-6-methyl-2-aminoindan: similarities to 3,4-(methylenedioxy)methamphetamine (MDMA)
Johnson MP, Frescas SP, Oberlender R, Nichols DE
J Med Chem, 1991; 34(5):1662-8

Title: An interesting possiblilty
Post by: Kinetic on April 08, 2004, 08:54:00 AM
Hi Fastandbulbous: :)

The Nichols research group made both isomers of the dihydrobenzofuran analogue. The one I depicted is compound 5 in

Synthesis and Pharmacological Examination of Benzofuran, Indan, and Tetralin Analogues of 3,4-( Methyenedioxy)amphetamine (

(, and your analogue is compound 4. I chose the former as it's much easier to make. Here is the quote from the paper on the pharmacology:

The results of the drug-discrimination studies are shown in Table I and II. Benzofurans 4 and 5 fully substituted in (S)-lc-trained and 3-trained rata with ED508 not significantly higher than those of the training drugs and potencies comparable to one another. Both 4 and 5 were potent in producing disruption in (S)-amphetamine-trained and LSD-trained rata, and neither compound substituted in (S)-amphetamine-trained or LSD-trained rata which were not disrupted (data not shown). These results indicate that, behaviorally, 4 and 5 resemble (S)-1c and 3, and have neither amphetamine-like nor LSD-like properties. Furthermore, these data indicate that the position of the oxygen atom in relation to the alkylamine side chain is not particularly significant in producing lc- or 3-like behavioral responses. Thus, both 4 and 5 might be expected to exhibit human psychopharmacology similar to one another, to 1c and 3, and, possibly, to the related entactogens la and lb [Kinetic's voice: 1a and 1b are MDA and MDMA, respectively]. This prediction is based on our previous observations that both la and lb fully substitute for the training drug lc.8,27 Since 4 and 5 also substitute for 1c, it seems highly likely that these compounds must produce a behavioral cue similar to that produced by la and lb.


I made the 2-(2',4'-dibromophenoxyethyl)bromide by dibromination of 2-phenoxyethylbromide with 2 eq. bromine in GAA. I didn't use a Lewis acid catalyst (as Nichols does in the similar procedure:

Post 364601 (missing)

(Barium: "Another way of inserting the aminopropyl", Serious Chemistry)) so I'm worried that there may have been some ring-monobrominated contaminant. The yield after recrystalisation from 1:1 methanol:acetonitrile (an excellent combination) was around 70%. I want to try the ring bromination on the more activated phenol before I take the procedure any further, to ensure the maximum amount of dibrominated ring product.

The ring-bromination of a phenol with molecular bromine is a pretty standard procedure but I can go to the library and type up the experimental from the Bull Chem Soc Japan article as well as the old J Chem Soc article (though I may have to order this one).

Here is the smaller of the two cyclisation articles (the other file hasn't taken kindly to the new upload feature). BuLi isn't necessary; Nichols manages a similar cyclisation with the Grignard:

A practical approach to highly functionalized benzodihydrofurans
Michael Plotkin, Sanyou Chen and P. Grant Spoors
Tetrahedron Letters 41 (2000) 2269–2273



A number of aromatic dibromides have been treated with 2–3 equivalents of n-butyllithium in order to initiate two sequential chemical events, a Parham cyclization and an intermolecular reaction with DMF.

And the abstract and selected experimental from the second article:

Oxygen Heterocycles by the Parham Cyclialkylation
Charles K. Bradsher and David C. Reames
J.Org.Chem. 1981, 46, 1384-1388


The addition of butyllithium at -100oC to w-bromoalkyl ethers of o-bromophenol (and its congeners) led to preferential exchange of the aryl bromine at position 2. The resulting organolithium reagents, under suitable conditions, cyclized to afford 2,3-dihydrobenzofurans (6), 3,4-dihydro-2H-l-benzopyrans (13), or 2,3,4,5-tetra-hydro-1-benzoxepins (16) in good yields, but less satisfactory reaults were obtained with the intermediate expected to produce 8-methyl-3,4,5,6-tetrahydro-2H-benzoxocin (19). w-Bromoethy1 and w-bromopropyl ethers of suitable dibromophenols were treated successively with 2 equiv. of butyllithium and an electrophile to yield derivatives of 6 and 13.


Phenoxyethyl Bromides (4).

These were prepared in yields of 47-61% by refluxing and stirring for 6-22h a mixture containing 0.120 mol of the phenol, 30.1 g (0.160 mol) of the ethylene bromide, 5.20 g of sodum hydroxide, and 90mL of water, esentially as described by Marvel and Tannenbaum6

(b) 5-Methyl-2,3-dihydro-2-benzofurancarboxyaldehyde (8).

Two successive equivalents of butyllithium were added to 4f at -100oC as described in the preceding experiment. Thirty minutes after the addition of the second equivalent of butyllithium, a solution of 1.61 g (22mmol) of dimethylformamide was added over 3 min. The solution was held at -100oC for 30 min and was then allowed to warm slowly to room temperature. Stirring of the suspension at room temperature was continued for 1 h. The reaction mixture was poured into 5% hydrochloric acid, the phases were separated, and the aqueous phase was extracted with ether (3x150mL). The organic materials were dried, concentrated, and distilled under reduced pressure. The yield of aldehyde 8 was 1.80 g (81%): bp 72-75.5oC (0.08 torr)

(c) 2,3-Dihydro-5-benzofurancarboxylic Acid (10).

(2,4-Dibromophenoxy)ethyl bromide (4e; 8.97g, 25mmol) was dissolved in dry THF (165 mL) and hexane (40 mL). The solution was placed in a 500-mL flask equipped for low-temperature lithiation and was cooled to -100oC. Butyllithium (27mmol) was added at such a rate that the temperature did not rise above -95oC. The solution was stirred at -100oC for 30min, and then 27 mmol of butyllithium was added at -100oC. A sample was taken after 30 min and processed, and examination by 1H NMR showed exchange to be complete. One hour (at -100oC) after the second addition of butyllithium, the mixture was poured into a slurry of solid CO2 in ether (150 mL). After the mixture had come to room temperature, the layers were separated. The organic phase was extracted with saturated sodium bicarbonate solution (3 X 150 mL). The combined bicarbonate solutions were washed once with ether and then acidified with hydrochloric acid. The resulting precipitate was recrystallized from ethanol, giving 2.50g (61%) of 7 as shiny colorless plates: mp 184-188oC; mp (pure).

6: Marvel, C. S.; Tannenbaum, A. L. Organic Syntheses; 1941; Vol. I, p 435.

Title: Cmpd 4 is a much better bet than cmpd 5
Post by: Fastandbulbous on April 12, 2004, 06:30:00 PM
Hi Kinetic

On your last post on this subject, you quoted a part of Nichols paper that said that both benzofuran compounds (4 and 5) substituted for compounds 1c and 3 in drug discrimination trials using trained rats. By extension, the paper went on to say that as compounds 1a and 1b also substituted for 1c and 3. From this, it went on to say that both 4 and 5 might be expected to exhibit human pharmacology similar to 1c and 3, and possibly, to the related entactogens 1a and 1b.
Indeed, 1a, 1b, 4 and 5 do substitute for 1c and 3, and this is due to the fact that all the aforementioned compounds are potent inhibitors of 5-HT uptake, but later on in the paper, it is stated that 1a and 1b are also inhibitors of dopamine and noradrenaline (norepinephrine) uptake

It is our hypothesis that 6, which possesses a significant catecholaminergic component, behaviorally lies closer to the la end of a spectrum However, 6 caused disruption in a large percentage of 3-trained rats, indicating that the behavioral effects of 1c and 3 may actually be slightly different. The explanation for this discrepancy may lie in the fact that 3 is highly selective for affecting serotonin uptake/release, and its behavioral cue probably reflects a more "purely serotoneric" response. In moving from 3 to Ic to la, an increased ability to affect catecholamine uptake/release is observed, and behavioral responses may reflect this change in the ratio of catecholamine to 5-HT effect.of behavioral effects than it does to the pure serotonin releasing, or 3, end. Thus, the compounds in this series can be ranked in order their ratio of catecholamine/5-HT effect such that la > 6 > lc > 3. These ratios of effects may significantly alter the behavioral pharmacology, since we have recently shown that 5-HT-releasing agents can markedly potentiate the effects of an indirectly acting dopaminergic drug.

Later in the same paper, it goes on to say that the compounds ability to inhibit the uptake of dopamine and noradrenaline plays a significant role in the psychopharmacology of said compounds, their being a spectrum of activity dependant upon the ratio of 5HT to catchecolamine uptake inhibition, compounds 1a and 1b also being potent inhibitors of catchecolamine uptake.

it is clear that all the compounds of this class have in common their ability to simultaneously release both 5-HT and the catecholamines DA and NE, albeit to varying degrees. Certain nuances of behavioral effect, however, seem to arise from shifts in the potency ratio in affecting catecholamine and 5-HT release. Thus, agents which have a lower potency in catecholamine systems, such as 5, behaviorally resemble agents like 3, while compounds with greater effects on catecholamine systems would be expected to elicit behavioral responses more similar to those produced by la.

From the data given in table III of that paper (see below), the relative values of 5HT to dopamine/noradrenaline inhibition can be calculated.


As can be seen from section c of the table (part a is present in paper, parts b and c are my calculations taken from data given in part a), the higher the relative inhibition of catchecolamines to 5HT, the closer the profile is to that of 1a (MDA) and the less like it is to the purely serotonergic 3 (MMAI). This places the order of how like 1a (and thereby unlike 3) the compounds are:-

                     1a > 4 > 6 > 5 > 3

As compound 6 is IAP, which has some similarity to MDA, but does not produce the extreme euphoria and feeling of closeness that MDA/MDMA does (from personal experience with IAP), it can be seen that whereas compound 4 is more like MDA than IAP, compound 5 is less like MDA than IAP, of the two, compound 4 is the most likely to have an overall activity like that MDA.

One other option is open to try to produce an activity profile like that of MDA/MDMA, and that is to use an indirectly acting catocholinergic compound (eg. amphetamine) with one of the above, in order to produce a 5HT to dopamine/noradrenaline uptake inhibition ratio more like that of MDA. As no data is given regarding the ability of any of the above to inhibit the enzyme monoamine oxidase (MAO), such an undertaking would have to proceed with great caution.

I have tried very small doses of amphetamine (aprox 2mg) taken in conjunction with 25mg of IAP, which did not result in any changes in pulse rate/ blood pressure that were different to 25mg of IAP taken on its own. I will eventually try a dose of aprox 20mg of amphetamine with 25mg of IAP, but as the neurotoxicity also seems to increase as the levels of dopamine increase, it may be some time before I can give a description of the action of said combination (it seems the neurotoxicity goes hand in hand with the ability to produce an MDA/MDMA type experience)

Structure of compounds 1a, 4, 5 and 6 (discussed above)


If Nichols (or anyone else) has taken that research any further, and someone has the appropriate references, I would be grateful for any more info.

Title: A success, a failure, and an alternative proposal
Post by: Kinetic on May 28, 2004, 09:05:00 AM
I decided to try the first two steps of the proposal I made above. Even if the compound turns out to be inactive, the chemstry is still very interesting, and some of it could be applied to the bis-benzodifuranyl analogue made by Nichols.

The first step was the dibromination of 2-phenoxyethyl bromide, which went very nicely:


200mmol 2-phenoxyethyl bromide
420mmol bromine
440mmol zinc chloride
Acetic acid

A solution of 40.2g 2-phenoxyethyl bromide and 60g zinc chloride in 100ml acetic acid was cooled to 10oC. Over 1.5 hours a solution of 21.6ml bromine in 25ml acetic acid was added, keeping the temperature at 10-15oC throughout. Stirring was continued without further cooling for 3 hours, then all was added to 500ml water. The precipitated solid was taken up in 100ml DCM followed by 2x50ml DCM. The combined extracts were washed with 250ml water, 2x100ml 1M potassium carbonate solution1, 100ml brine and dried over MgSO4. Removal of the solvent gave a brown solid weighing 70.7g (197mmol, crude yield 98%) which, after recrystallisation from 80ml 1:1 methanol:acetonitrile2, filtering and drying under vacuum, gave the title product a sparkling, light coloured solid.

Yield: 67.6g (188mmol, 94%)

1 The potassium carbonate washes should be omitted as they caused significant darkening of the organic layer.
2 100ml pure methanol should be an even better recrystallisation solvent.

The second step (one-pot cyclisation and formylation) gave a product which seems to be 5-ethyl-2,3-dihydrobenzofuran, judging by it's boiling point and apparent inability to react with semicarbazide. At higher temperatures it appears the aryl Grignard reacts more rapidly with the formed ethyl bromide than I had hoped, in a reaction reminiscent of Wurtz coupling.

To overcome the problem, a lower temperature could be used (the temperature reached 45oC during addition of the 2,4-dibromo-1-(2-bromo-ethoxy)-benzene to the excess of ethylmagnesiumbromide). Inverse addition - i.e. addition of 1 equivalent ethylmagnesiumbromide to 2,4-dibromo-1-(2-bromo-ethoxy)-benzene to form the ring, then cooling and addition of the rest of the ethylmagnesiumbromide, followed by DMF, is probably the best way. This is what I will try next time, though I doubt that will be soon.

But what seems more promising to me is the following proposal:

Starting from 4-hydroxybenzaldehyde is almost as innocuous as starting from phenol, with the advantage of having the formyl group already in place. Treatment of this with 1,2-dibromoethane, much like hest's work in

Post 196477 (missing)

(hest: "Re: New Amph.  more potent than LSD", Serious Chemistry)
, will give the ether:

Molecule: (

Etherification ("c1cc(C=O)ccc1O.BrCCBr>>c1cc(C=O)ccc1OCCBr")

Bromination in acetic acid should work just as well as the bromination of the phenol derivative above, as there is essentially only one position the bromine can take up. There are numerous examples in the literature for the similar bromination of 4-methoxybenzaldehyde:

Molecule: (

Bromination ("c1cc(C=O)ccc1OCCBr>>c1cc(C=O)cc(Br)c1OCCBr")

Now, the above compound is treated with 2 equivalents of ethylmagnesiumbromide. The first equivalent will add to the carbonyl, and the second will cause a Parham cyclisation as explained in the above post

Post 499568 (

(Kinetic: "A more interesting possibility?", Novel Discourse)
. This gives the phenyl-1-propanol derivative:

Molecule: (

Grignard addition followed by Parham cyclisation ("c1cc(C=O)cc(Br)c1OCCBr>>CCC(O)c2ccc1OCCc1c2")

The phenyl-1-propanol is of course readily dehydrated to the propenylbenzene, which can be though of as a safrole analogue:

Molecule: (

Dehydration ("CCC(O)c2ccc1OCCc1c2>>CC=Cc2ccc1OCCc1c2")

The propenylbenzene is then oxidised to the phenylacetone by the Hive's favourite Wacker oxidation, and from this the amphetamine or methamphetamine is made via any one of the usual methods.


For the bromination (the related 4-methoxybenzaldehyde -> 3-bromo-4-methoxybenzaldehyde transformation):

With bromine and iodine in CCl4: J. Amer. Chem. Soc., 39, 1917, 1711
With bromine in CCl4: Justus Liebigs Ann. Chem., 460?, 1928, 135
With bromine in acetic acid: J. Chem. Soc. Perkin Trans. 1, 1979, 829-837.

For the related Grignard addition of ethylmagnesiumbromide to 4-methoxybenzaldehyde:

Chem. Ber., 38, 1905, 1679
Tetrahedron Lett., 37 (48), 1996, 8767-8770.

For the Parham cyclisation:

Articles dealt with given in the above post

Post 499729 (

(Kinetic: "An interesting possiblilty", Novel Discourse)

For the dehydration of phenyl-1-propanols to propenylbenzenes:

Propenylbenzenes from propiophenones (


Propenylbenzene from phenyl-1-propanol (

Title: Parham my mistake
Post by: Kinetic on May 31, 2004, 06:46:00 AM
Many thanks to azole for pointing out a mistake I had made, and for his suggestions on a workaround for the problem. His advice undoubtedly saved me a lot of time trying to work out why I had had so many failures.

The mistake, which I incorporated into my proposals throughout the thread, was the assumption that the ethylmagnesiumbromide used by Nichols in J. Med. Chem., 2001, 44 (6), 1003-1010 - article posted in

Post 364601 (missing)

(Barium: "Another way of inserting the aminopropyl", Serious Chemistry)
- was a catalyst; i.e., I assumed it metalated the aryl bromide, itself then forming ethyl bromide, which could then react with the Mg powder to reform ethylmagnesium bromide, which would again metalate the ring.

In fact, this is not the case: its role is simply as an activating agent (much like I2 is often used), to ensure the direct oxidative insertion of the Mg between the Ar-Br bond, which then displaces the chloride from the end of the connected side-chain, closing the ring. A reasonably thorough literature search indicates that metalation with Grignards only occurs in the presence of a catalyst (e.g. a Cu salt).

It's therefore unlikely I made 5-ethyl-2,3-dihydrobenzofuran as speculated in my above post; it also seems unlikely I could have made the benzaldehyde, or even closed the ring. There was however definitely a reaction; there was a temperature rise from 20oC to 45oC during addition of a THF solution of 2,4-dibromo-1-(2-bromo-ethoxy)-benzene to a THF solution of 2.5 equivalents of ethylmagnesiumbromide, as well as appreciable lightening of the dark 'metallic' Grignard colour.

At higher temperatures and with the heavier halides attached to the ethyl chain of the substituted 2-phenoxyethyl bromide, metalation of the side-chain, followed by elimination of ethene from the unstable product, is an increasingly prevalent side-reaction:

Molecule: (

Ethene elimination ("c1cccc(Br)c1OCCBr.CC[Mg++]Br.>>c1cccc(Br)c1O.C=C")

This could explain why the Nichols group chose to use 1,4-bis(2-chloroethoxy)benzene rather than 1,4-bis(2-bromoethoxy)benzene, and could also explain the reaction I observed.

Fortunately, the problem should be readily overcome. Starting from 4-hydroxybenzaldehyde, it would probably be best to use 1,2-dichloroethane in the etherification, to limit any unwanted metalation followed by elimination during the next step. The ring-bromination step will not be affected by the change of halogen.

Treatment of this brominated product with a slight excess of ethylmagnesiumbromide, which should react with the more electrophilic carbonyl (rather than the chloroethoxy group), will give an intermediate alcoholate:

Molecule: (

Intermediate alcoholate formation ("c1cc(C=O)cc(Br)c1OCCCl>>c1cc(C([O-])CC)cc(Br)c1OCCCl")

This intermediate is then added directly to a stirred suspension of Mg powder in THF; the excess ethylmagnesiumbromide acting as an initiator of the aryl Grignard formation and subsequent cyclisation in the same way Nichols uses it in the cyclisation of 1,4-bis(2-chloroethoxy)2,5-dibromobenzene. This leads us to the same product as before, 1-(2,3-dihydrobenzofuran-5-yl)-1-propanol:

Molecule: (

Ring closure ("c1cc(C([O-])CC)cc(Br)c1OCCCl>>CCC(O)c2ccc1OCCc1c2")

Note the two caveats: the ethylmagnesium must first react with the carbonyl, and not the chloroethoxy group (however, alkyl chlorides aren't particularly electrophilic); also, the intermediate alcoholate must be soluble in THF. Using sufficient solvent should overcome this problem - the horrible white goo which can precipitate during a Grignard reaction can often be overcome by using sufficient solvent. It may also be possible to use a better donor solvent, such as dimethoxyethane.

I've included the original article on the Parham cyclisation. There is nothing on the formation of dihydrobenzofuran analogues, but it's interesting nonetheless. The first thing that came to my mind was the applicability to aminoindanes and aminotetralins, as the 1-indanones and 1-tetralones (cyclised propiophenone/butyrophenone derivatives) can be made readily by this route:

Selective Lithiation of Bromoarylalkanoic Acids and Amides at Low Temperature. Preparation of Substituted Arylalkanoic Acids and Indanones
William E. Parham,* Lawrence D. Jones, and Yousry Sayed
J. Org. Chem., 40 (16), 1975


Edit: When synthesising 2,4-dibromo-1-(2-bromo-ethoxy)-benzene - as in the previous post, the flask was encased in aluminium foil throughout the reaction. Whether this is necessary I don't know, but the synthesis should read: 'A solution of 40.2g 2-phenoxyethyl bromide and 60g zinc chloride in 100ml acetic acid was cooled to 10oC. The flask was then encased in aluminium foil to exclude light. Over 1.5 hours...'
Title: Related PIHKAL entry
Post by: Kinetic on May 31, 2004, 03:54:00 PM
The first step from Shulgin's synthesis of

J (BDB) (

( - addition of propylmagnesiumbromide to piperonal - should be adaptable for the Grignard addition of ethylmagnesiumbromide to 3-bromo-4-(2-chloroethoxy)benzaldehyde. The crude yield is 96%:


The Grignard reagent of propyl bromide was made by the dropwise addition of 52g 1-bromopropane to a stirred suspension of 14g magnesium turnings in 50mL anhydrous Et2O. After the addition, stirring was continued for 10min, and then a solution of 50g piperonal in 200mL anhydrous Et2O was added over the course of 30min. The reaction mixture was heated at reflux for 8h, then cooled with an external ice bath. It was quenched with the addition of a solution of 75mL cold, saturated aqueous ammonium chloride. The formed solids were removed by filtration, and the two-phase filtrate separated. The organic phase was washed with 3x200mL dilute HCl, dried over anhydrous MgSO4, and the solvent removed under vacuum. The crude 62.2g of 1-(3,4-methylenedioxyphenyl)-2-butanol [Kinetic's voice: this should read 1-(3,4-methylenedioxyphenyl)-1-butanol], which contained a small amount of the olefin that formed by dehydration, was distilled at 98°C at 0.07mm/Hg to give an analytical sample, but the crude isolate served well in the next reaction. Anal. (C11H14O3) C,H.

Graphical abstract:

Molecule: (

Grignard addition to piperonal ("C(=O)c2ccc1OCOc1c2.CCC[Mg++]Br>>CCCC(O)c2ccc1OCOc1c2")

The next step in the synthesis is the dehydration of 1-(3,4-methylenedioxyphenyl)-1-butanol to 1-(3,4-methylenedioxyphenyl)-1-butene, which should be adaptable for the dehydration of 1-(2,3-dihydrobenzofuran-5-yl)-1-propanol to 1-(2,3-dihydrobenzofuran-5-yl)-1-propene. The yield is 93%, making the overall yield from piperonal 89%:


A mixture of 65g crude 1-(3,4-methylenedioxyphenyl)-2-butanol [Kinetic's voice: again, this should read 1-(3,4-methylenedioxyphenyl)-1-butanol] and 1g finely powdered potassium bisulfate was heated with a soft flame until the internal temperature reached 170°C and H2O was no longer evolved. The entire reaction mixture was then distilled at 100-110°C at 0.8mm/Hg to give 55g of 1-(3,4-methylenedioxyphenyl)-1-butene as a colorless oil. Anal. (C11H12O2) C,H.

Graphical abstract:

Molecule: (

Dehydration to 1-(3,4-methylenedioxyphenyl)-1-butene ("CCCC(O)c2ccc1OCOc1c2>>CCC=Cc2ccc1OCOc1c2")

Title: reactivity of RMgX towards RX
Post by: azole on June 01, 2004, 04:50:00 AM
To use more accessible Grignard reagents instead of RLi is a very attractive idea. However, the reactivity of RMgX may differ greatly from the reactivity of RLi. The mechanism of the reported* cyclization of 2,5-dibromohydroquinone bis(2-chloroethyl) ether with Mg/EtMgBr probably doesn't include metallation by EtMgBr. There are at least two articles to support this point of view.

Factors influencing the course and mechanism of Grignard reactions. XVII. Interchange of radicals in the reaction of Grignard reagents and organic halides in the presence of metallic halides.
M. S. Kharasch and C. F. Fuchs
J. Org. Chem.
, 10, 292-297 (1945).


   Unlike RLi, organomagnesium compounds typically do not react with organic halides (R'Hal) to form R'MgX. However, there are exceptions, e.g. alpha-haloketones (

None (

( ) and haloacetylenes (see below). In the presence of CoCl2 the halogen-metal exchange does take place to some extent, and various coupling products are formed.

Factors determining the course and mechanism of Grignard reactions. XVIII. The effect of metallic halides on the reactions of Grignard reagents with 1-chloro-3-phenylpropane, cinnamyl chloride, and phenylethynyl bromide.
M. S. Kharasch, F. L. Lambert, and W. H. Urry
J. Org. Chem.
, 10, 298-306 (1945).


   In the absence of transition metal halides, 1-chloro-3-phenylpropane does not react with Grignard reagents; cinnamyl bromide enters "normal" Wurtz-type coupling reactions, and phenylethynyl bromide undergoes halogen-metal exchange. The situation is completely altered on addition of CoCl2 or other metal catalysts.

   The reactions of halogen-magnesium exchange were reviewed recently. Shame on me, I didn't know.

Highly Functionalized Organomagnesium Reagents Prepared through Halogen-Metal Exchange
P. Knochel et al.,
Angew.Chem. Int. Ed. Engl.
, 42(36), 4302-4320 (2003).

10.1002/anie.200300579 (

  It appears that arylbromides bearing electron-withdrawing substituents (or a chelating group ortho- to the halogen) do react with RMgX. Thus, 2,4-dibromoanisole is converted into 2-MeO-5-Br-PhMgCl by treatment with 2 eq. of i-PrMgCl in THF at 40° for 5 h. Subsequent reaction with CO2 affords the corresponding carboxylic acid in 90% yield (see the following ref.).

Metal-halogen exchange between polybromoanisoles and aliphatic Grignard reagents: a synthesis of cyclopenta[b]benzofurans
H. Nishiyama et. al.,
J. Org. Chem.
, 57, 407 (1992). (

No DOI found.


Post 185131 (missing)

(hest: "New Amph.  more potent than LSD", Serious Chemistry)
, (

Title: 2,3-Dihydrobenzofurans without BuLi
Post by: Kinetic on June 01, 2004, 07:03:00 AM
Great azole!

Here is another very interesting article: on the synthesis of 5-substituted 2,3-dihydrobenzofurans from 2-(2-bromophenoxy)ethyl chlorides, using only Mg in the cyclisation. The article also has a high-yielding PTC alkylation of bromophenols, referenced in

Post 196477 (missing)

(hest: "Re: New Amph.  more potent than LSD", Serious Chemistry)

The Synthesis of 5-Substituted 2,3-Dihydrobenzofurans
Ramon J. Alabaster, Ian F. Cottrell, Hugh Marley, Stanley H. B. Wright*
, 12, 1988, 950-952.


The preparation of 2,3-dihydrobenzofurans 6 from 2-(2-bromophenoxy)ethyl chlorides 3 by reaction with magnesium in a development of the Parham cyclialkylation reaction is described. A high yielding procedure using phase-transfer catalysis has also been developed for the preparation of the intermediate chloroethyl ethers 3 from bromophenols 2. The 5-hydroxy derivative 15 may be obtained from 2,3-dihydrobenzofuran (6a) by reaction with electrophilic agents followed by oxidation.


Edit: Commenting on your edit: So if 2,4-dibromoanisole reacts with 2eq. of i-propylmagnesiumchloride in THF at 40°C, it seems likely that 2,4-dibromo-1-(2-bromoethoxy)benzene would react with 2.5eq. ethylmagnesiumbromide at 20-45°C. So maybe I did cyclise the ring after all! But still, it's impossible to say without further analysis.
Title: Grignard formylation article
Post by: Kinetic on June 11, 2004, 12:53:00 PM
The following article - part of which has been posted in

Post 60192 (missing)

(yellium: "Another route to 2C-[BDE]", Chemistry Discourse)
- has a high-yielding and simple formylation of phenylmagnesiumbromide, which should be readily adaptable to suit the in-situ prepared 5-(2,3-dihydrobenzofuranyl)magnesiumbromide, as well as other interesting substituted aryl systems (such as 5-bromo-1,3-benzodioxole):

Synthetic Methods and Reactions; Part 109. Improved Preparation of Aldehydes and Ketones From N,N-Dimethylamides and Grignard Reagents
George A. Olah, G. K. Surya Prakash, Massoud Arvanaghi
, 3, 1984, 228-230


Title: 3-Br-4-(2-chloroethoxy)BA failed to cyclize
Post by: azole on July 28, 2004, 09:32:00 AM
SWIM has made some synthetic studies along the route proposed by Kinetic (

Post 510530 (

(Kinetic: "Parham my mistake", Novel Discourse)

1) 4-Hydroxybenzaldehyde was successfully alkylated. SWIM managed to get a 69% yield of a pure product using 1,2-dichloroethane in a non-optimized procedure. The published1 procedure deals with 1-bromo-2-chloroethane, a more reactive alkylator, and only a 30% yield is achieved.

2) Bromination of the above aldehyde gave 40% of the monobromo product. The optimum reaction conditions are yet to be found.

3) The reaction of 3-bromo-4-(2-chloroethoxy)benzaldehyde with 4 eq. of EtMgBr in THF gave no cyclization product after a 2.5 h reflux, as can be inferred from the NMR spectra. Apparently, only addition of EtMgBr has occurred.


   Of course, one can make a dimethyl acetal from 3-bromo-4-(2-chloroethoxy)benzaldehyde and cyclize it with Mg/THF as described above (

Post 510718 (

(Kinetic: "2,3-Dihydrobenzofurans without BuLi", Novel Discourse)

Experimental part


Bu4NBr FW 322.36
1,2-Dichloroethane FW 98.96, d 1.256, bp 83°
4-Hydroxybenzaldehyde FW 122.12
K2CO3 FW 138.21
Na2SO3 FW 126.04

   A 250 ml RBF equipped with a magnetic stirrer, a reflux condenser, and a gas bubbler connected to the top of the condenser was charged with 4-hydroxybenzaldehyde (18.06 g, 0.148 mol), anhydrous sodium sulfite (~0.2 g, 1.6 mmol), tetrabutylammonium bromide (2.39 g, 7.41 mmol, 5 mol. %), potassium carbonate (21.5 g, 0.156 mol), 1,2-dichloroethane (60 ml), and ethylene glycol (40 ml). The mixture was stirred at reflux for 24 h (at 17 h point gas evolution still continued, and TLC showed incomplete reaction). Then the mixture was cooled; water was added to dissolve the precipitated KCl,  followed by toluene (60 ml). The contents of the flask were transferred to a separatory funnel, the flask was rinsed with toluene (10 ml), and the rinsings also were added to the separatory funnel. After shaking, the organic (upper) layer was separated and washed with 10% aq. KOH (4×25 ml). The aqueous layer was extracted with toluene (15 ml), and the extract was also washed with KOH solution (4×10 ml).

   The combined organic phases were filtered* through a mixture of silica gel (height 2 cm) and anhyd. Na2SO4 (height 0.5 cm) placed on a glass filter (diam. 4 cm); the adsorbed product was eluted with a mixture of toluene (80 ml) and ethyl acetate (20 ml), and the combined solutions were evaporated under reduced pressure to give a crude product (27.7 g) as a yellow oil, which solidified on standing in a refrigerator.

   TLC (Merck F254 SiO2 plates, visualisation in UV light and with 0.5% 2,4-dinitrophenylhydrazine soln. in dil. H2SO4; eluent : CHCl3 - Me2CO 19:1 v/v) showed 2 major spots: Rf 0.65 (4-(2-chloroethoxy)benzaldehyde) and Rf 0.53 (presumably (OHCC6H4OCH2)2, since it was not detected in the product upon distillation), along with traces of 4-hydroxybenzaldehyde, Rf 0.24,  and another aldehyde byproduct, Rf 0.19.

   Distillation in a vacuum of an oil pump gave two fractions : bp 100-108 °C (1.83 g) and bp 108-115 °C (lit. bp 110 °C (0.1 mm Hg)1; 138-142 °C (2 mm Hg)2) . These were combined and redistilled. After a small forerun (0.96 g), the product was collected (18.83 g, 69%; lit.1 yield 30% from 1-bromo-2-chloroethane), which solidified on standing (mp 29-30 °C; lit.1 mp 31°C). The smell is similar to that of anisaldehyde, but not so intense. TLC showed a small admixture of 4-hydroxybenzaldehyde. The forerun solidified below 20 °C.

   1H NMR (200 MHz, CDCl3): ? (ppm) 9.85 (s, 1H, CHO), 7.80 (m (AA'BB'), 2H, H-2, H-6), 6.98 (m (AA'BB'), 2H, H-3, H-5), 4.27 (t, 2H, J = 5.7 Hz, -OCH2CH2Cl), 3.82 (t, 2H, J = 5.7 Hz, -OCH2CH2Cl).

   13C NMR (50 MHz, CDCl3): ? (ppm) 190.56 (CHO), 162.94, 131.80, 130.21, 114.67 (benzene ring), 67.97 (-CH2O-), 41.48 (-CH2Cl).

*This step (actually, a short-column chromatography) was originally designed to get rid of Bu4NBr and the byproduct having lower Rf value, presumably 4-(2-hydroxyethoxy)benzaldehyde. Instead, the solution can be simply dried with Na2SO4.


Br2 FW 159.82, d 3.119
4-(2-Chloroethoxy)benzaldehyde FW 184.62, d25 1.22462
ZnCl2 FW 136.28

   A solution of bromine (1.8 ml, 35 mmol) in glacial AcOH (5 ml) was added to a solution of 4-(2-chloroethoxy)benzaldehyde (~5 ml, 6.18 g, 33.5 mmol) and zinc chloride (0.91 g, 6.7 mmol) in glacial AcOH (15 ml) dropwise with magnetic stirring over the course of 8 min (slight exothermy). The reaction flask was protected from light with aluminum foil. The mixture was allowed to stand at rt for 2 h 15 min. The product has virtually the same Rf value as the starting aldehyde in a number of eluents tested (CHCl3 - Me2CO 19:1, petroleum ether - EtOAc 7:3, benzene - EtOAc 9:1 v/v), so TLC appeared to be useless.

   Another portion of Br2 (0.4 ml, 7.8 mmol) was added (obviously, this was a mistake, see below), and the mixture was allowed to stand for 1.5 h. Then 2% aq. Na2SO3 was added with stirring until the bromine color disappeared; the volume of the mixture was brought to ~100 ml with water, the organic layer separated, and the aqueous layer extracted with CHCl3 (3×10 ml).

   The combined organic phases were washed with water (60 ml), 5% aq. NaOH containing 0.5% Na2SO3 (2×60 ml), dried (Na2SO4), and evaporated to give a colorless oil (7.75 g), which solidified on standing in a refrigerator. This was dissolved in a mixture of petroleum ether (20 ml), CCl4 (10 ml) and PhMe (7 ml) at ~50 °C. On cooling to rt, crystals formed. The mixture was cooled to +4 °C; petroleum ether (10 ml) was gradually added to complete crystallization; the crystals were filtered off, washed with cold CCl4, then with a 1 : 3 mixture of CCl4 with petroleum ether, and dried. Yield 3.585 g (40%) of white crystals, mp 84-85 °C, with a "pesticide" smell. Rf 0.37 (petr. ether - EtOAc 7:3 v/v).

   The mother liquors and washings were evaporated to dryness. Low-temperature crystallization of the residue from methanol gave 1.95 g of white crystals with mp 50-53 °C, probably a dibrominated product, with Rf 0.57 (petr. ether - EtOAc 7:3 v/v). The melting point was raised to 53-54 °C after recrystallization from CCl4 - petroleum ether. The solubility of the byproduct in CCl4 is much higher than that of the main product. After standing at rt for several days, the crystals turned to a yellow liquid.

   1H NMR (200 MHz, CDCl3): ? (ppm) 9.84 (s, 1H, CHO), 8.06 (d, 1H, J = 1.9 Hz, H-2), 7.79 (dd, 1 H, J = 1.9 Hz, J = 8.5 Hz, H-6), 6.98 (d, 1H, J = 8.5 Hz, H-5), 4.36 (t, 2H, J = 5.9 Hz, -OCH2CH2Cl), 3.89 ((t, 2H, J = 5.9 Hz, -OCH2CH2Cl).

   13C NMR (50 MHz, CDCl3): ? (ppm) 189.43 (CHO), 159.21, 134.64, 131.08, 130.92, 113.01, 112.55 (benzene ring), 69.14 (-CH2O-), 41.11 (-CH2Cl).


3-Bromo-4-(2-chloroethoxy)benzaldehyde FW 263.52
EtBr FW 108.97, d 1.460
Mg FW 24.31

   To a solution of EtMgBr prepared from EtBr (4.0 ml,  54 mmol) and Mg (1.93 g, 79 mmol) in abs. THF (50 ml) under argon was added a solution of 3-bromo-4-(2-chloroethoxy)benzaldehyde (3.48 g, 13.2 mmol) in abs. THF (20 ml) dropwise with stirring and cooling in a water bath (30-35 °C, internal temperature). The addition took 13 min. No precipitate has formed. The mixture was refluxed for 2.5 h (after 0.5 h TLC showed a single spot of the product, Rf 0.46 in CHCl3 - Me2CO 19 : 1 v/v; no changes were noted on further heating) and cooled to 0 °C. A saturated aq. solution of NH4Cl (50 ml) was carefully added to the reaction mixture (Caution: ethane evolution!), followed by PhMe (50 ml). The organic layer was separated; the aqueous layer was extracted with PhMe (2×20 ml), and the combined extracts were washed with 10% aq. NaOH (3× 25 ml), dried (Na2SO4) and evaporated. The resulting product (viscous yellowish oil) gave a positive Beilstein test, and its NMR spectra were consistent with the title structure (cf. 3 and 4).

   1H NMR (200 MHz, CDCl3): ? (ppm) 7.50 (s, 1H, H-2), 7.17 (d, 1H, J = 8.6 Hz, H-6), 6.84 (d, 1H, J = 8.6 Hz, H-5), 4.46 (t, 1H, J = 6.4 Hz, CHOH), 4.25 (t, 2H, J = 6.0 Hz, -OCH2CH2Cl), 3.83 (t, 2H, J = 6.0 Hz,  -OCH2CH2Cl), 2.54 (br. s, 1H, OH), 1.69 (m, 2H, -CHOH-CH2CH3), 0.86 (t, 3H, J = 7.3 Hz, CH3).

   13C NMR (50 MHz, CDCl3): ? (ppm) 153.7, 139.2, 131.0, 127.2, 126.0, 114.4, 113.7, 112.3 (arom. ring; note the signals of admixtures), 74.6 (CHOH), 69.3 (-CH2O-), 41.5 (-CH2Cl), 31.7 (-CHOHCH2CH3), 9.9 (CH3).

1 J. Org. Chem., 18, 1380 (1953).


Patent US2568579 (


3   The 13C NMR spectrum of 1-phenyl-1-propanol, found in

NMRShiftDB - NMR web database (

(, is as follows: 140.50, 126.53, 128.33, 127.76 (arom. ring), 77.30 (CHOH), 29.24 (CH2), 9.84 (CH3).

4   In the patent application (

some 13C spectra of dihydrobenzofurans are presented. Thus, in 5-bromo-2,3-dihydrobenzo[b]furan-7-carboxylic acid the chemical shift of C-2 is 71.78 ppm; that of C-3 is 27.88 ppm.
   Nichols didn't publish the 13C NMR spectra of the benzofurans obtained by his group. In 1H spectra the chemical shifts of the benzylic CH2 groups of dihydrobenzofurans were at ~3.2 ppm.
Title: Pesticides
Post by: Kinetic on July 28, 2004, 01:29:00 PM
Wow azole, what a post!

Thankyou so very much for trying this out! Although it's a shame the cyclisation doesn't work as hoped, your post was certainly an inspirational way in which to let us know. If only alkyllithiums were more accessible; ethyllithium would hopefully still do the trick.

It's interesting that you noted the smell of 3-Bromo-4-(2-chloroethoxy)benzaldehyde as pesticide-like; that was the first thing I thought when I smelled 2-phenoxyethyl bromide/chloride.

I made some 2-phenoxyethyl chloride using the procedure in Synthesis, 1988, 12, 950-952 (article posted above in

Post 510718 (

(Kinetic: "2,3-Dihydrobenzofurans without BuLi", Novel Discourse)
. I first modified the procedure, using less 1,2-dichloroethane and water, and half the amount of PTC. The 65% yield was lower than the expected 90%, so I tried it again, following the procedure almost to the letter (only using TBAB as PTC, and sodium metabisulfite instead of bisulfite). The yield was again 65%. After distillation the (pesticide-smelling, but rather pleasant to me) product was a solid in the fridge and slowly melted when removed, so had a melting point slightly below room temperature. The pure product should melt around 25oC.

The third time I tried the procedure I got only 39% yield, after using only two equivalents of NaOH. I inferred from this that the amount of 1,2-dichloroethane (80ml or 125ml for a 100mmol reaction gave the same yield) was not critical, nor was the amount or type of PTC (5mol% aliquat 336 or 10mol% TBAB). Only decreasing the amount of NaOH seems to lower the yield. This may be useful for anyone wanting to try hest's work in

Post 196477 (missing)

(hest: "Re: New Amph.  more potent than LSD", Serious Chemistry)

As the cyclisation doesn't work on the alkoxide, I hope that the original idea (treatment of 2-(2,4-dibromophenoxy)ethyl chloride with 2 equivalents of Mg, followed by treatment with DMF1) can still be made to work, as very similar work has been reported in the literature. The stabilising effect of the aryl ether oxygen in the ortho-position should ensure the cyclisation occurs before the para position reacts, so this should not interfere with the cyclisation as the alkoxide seems to. Hopefully we will soon see.

I feel inspired enough by your post to set off another synthesis of 2-phenoxyethyl chloride. I will then have enough to dibrominate it - using bromine and a catalytic amount of zinc chloride this time - and then attempt my yet-untested proposal of cyclisation with Mg (probably initiating the reaction with a small amount of ethylmagnesiumbromide1).

1 I know for certain that both the formlyation procedure and the initiation of aryl Grignards by ethylmagnesiumbromide occurs readily: see

Post 517404 (

(Kinetic: "Two formylation procedures", Novel Discourse)

TFSE is full of 5-year old threads ending with "I'll report back tomorrow!", but I feel the least I can do is try this again. I will report back in due course (but not tomorrow, though).
Title: Why must we insist on doing things the hard...
Post by: phenethyl_man on July 29, 2004, 12:07:00 AM
Why must we insist on doing things the hard way?  It's not as if benzofuran itself is difficult or expensive to obtain..  Nichols has already found that the tetrahydrobenzodifuran (5a) analog of 2,5-DMA is much weaker than the benzodifuran analog (6a, which just so happens to be as potent as the DOX-series of psychedelic amphetamines [1]).  Here are the compounds I am talking about:


Thus, I would put a good wager on the fact that the monobenzofuran, pictured above, would have considerably higher potency than the corresponding dihydro analogue.  Here's what Nichols had to say about the reasoning behind this:

"An additional trend that can be observed in Tables 1 and 2 is that the benzo[1,2-b;4,5-b']difuran-containing compounds (series 6) bind with higher affinity and exhibit increased potency relative to the corresponding tetrahydrobenzo[1,2-b;4,5-b']difurans (series 5), indicating that the compounds in series 6 posses more favorable interactions with the agonist binding site.  This may be due to the increased hydrophobicity of the extended tricyclic aromatic nucleus in 6a-c relative to the tetrahydro congeners 5a-c and a resulting greater tendency to partition into the hydrophobic receptor binding site.  It is also possible that the exteneded aromaticity of the benzo[1,2-b;4,5-b']difurans (series 6) may result in enhanced affinity by increasing the effective aromatic surface area on the ligand available for for favorable pi-stacking interactions with the agonist binding site, while still maintaining some (albeit weaker) hydrogen-bond acceptor properities of the furan oxygen atoms.  It is interesting to note that although potency is generally increased for the aromatic compounds 6a-c relative to the tetrahydro compounds 5a-c, the intrinsic activity of these compounds remaines largely unchanged. [1]"

If you are still set on dihydrobenzofuran, reduction of benzofuran still seems like a much more viable option taking into account both difficulty and economic factors.  However, that's just my opinion..

[1] J. Med. Chem. 2001, 44, 1003-1010

Title: Reactive in the wrong position
Post by: Rhodium on July 29, 2004, 05:18:00 AM
Why must we insist on doing things the hard way?  It's not as if benzofuran itself is difficult or expensive to obtain.

No, but it is difficult to manipulate - the 3-position on the furan ring is more reactive than any of the benzene ring positions (similar to indole), so therefore it is difficult to turn it into a benzofuranyl-aminopropane.

Title: d'oh.. I completely overlooked that.
Post by: phenethyl_man on July 29, 2004, 05:20:00 PM
d'oh.. I completely overlooked that.  Instead, let me propose these two steps from p-hydroxybenzaldehyde to a desired aldehyde:


The only possible problem I see here is perhaps obtaining chloroacetaldehyde, but it *is* available if you look hard enough.  250g of a 45% aqueous solution is available commercially for around $10..

p-formyl-phenoxyacetaldehyde when refluxed with GAA and a lewis acid *should* close the ring, forming the wanted aldehyde in good yield.

Title: Total Synthesis of Benzofuran
Post by: Rhodium on July 29, 2004, 07:09:00 PM

Total Synthesis of Benzofuran


Equip a 1-litre three-necked flask with an efficient double surface reflux condenser, a mechanical stirrer and a thermometer, the bulb of which is within 2cm of the bottom of the flask. Place a warm solution of 80g of sodium hydroxide in 80 ml of water in the flask, add a solution of 25g (0.266 mol) of phenol in 25 ml of water and stir. Adjust the temperature inside the flask to 60-65°C (by warming on a water bath or by cooling, as may be found necessary); do not allow the crystalline sodium phenoxide to separate out. Introduce 60g (40.5 ml, 0.5 mol) of chloroform in three portions at intervals of 15 minutes down the condenser. Maintain the temperature of the well-stirred mixture at 60-70°C during the addition by immersing the flask in hot or cold water as may be required. Finally heat on a boiling water bath for 1 hour to complete the reaction. Remove the excess of chloroform from the alkaline solution by steam distillation. Allow to cool, acidify the orange-colored liquid cautiously with dilute sulfuric acid and again steam distill the almost colorless liquid until no more oily drops are collected. Extract the distillate at once with ether, remove most of the ether from the extract by distillation on a water bath using a rotary evaporator. Transfer the residue, which contains phenol as well as salicylaldehyde, to a small glass-stoppered flask, add about twice the volume of saturated sodium metabisulfite solution, and shake vigorously (preferably mechanically) for at least half an hour, and allow to stand for 1 hour. Filter the paste of bisulfite compound at the pump, wash it with a little alcohol, and finally with a little ether (to remove the phenol). Decompose the bisulfite compound by warming in a round-bottomed flask on a water bath with dilute sulfuric acid, allow to cool, extract the salicylaldehyde with ether and dry the extract with anhydrous magnesium sulfate. Remove the ether by flash distillation and distill the residue collecting the salicylaldehyde (a colorless liquid) at 195-197°C. The yield is 12g (37%).

o-Formylphenoxyacetic acid

To a mixture of 35 ml (40 g, 0.33 mol) of salicylaldehyde, 31.5 g (0.33 mol) of chloroacetic acid and 250 ml of water contained in a 500-ml, two-necked round-bottomed flask fitted with a stirrer unit, add slowly with stirring a solution of 26.7 g (0.66 mol) of sodium hydroxide in 700 ml of water. Heat the mixture to boiling with stirring and reflux for 3 hours. The solution acquires a red-brown color. Cool and acidify the solution with 60 ml of concentrated hydrochloric acid and steam distill to remove unreacted salicylaldehyde; 12ml (14g) are thus recovered. Cool the residual liquor which first deposits some dark red oil which then solidifies; on standing, almost colorless crystals appear in the supernatant solution. Decant the supernatant solution and crystals and filter off the crystals, and air dry; the yield of almost pure product, m.p. 132-133°C, is 21 g. The solidified red oil may be extracted with small quantities of hot water, the extracts treated with decolorizing charcoal and cooled, to yield a further 6g of product; total yield 27g.


Heat under reflux for 8 hours a mixture of 20 g (0.11 mol) of o-formylphenoxyacetic acid, 40g of anhydrous sodium acetate, 100 ml of acetic anhydride and 100ml glacial acetic acid. Pour the light brown solution into 600 ml of iced water, and allow to stand for a few hours with occasional stirring to aid the hydrolysis of acetic anhydride. Extract the solution with three 150 ml portions of ether and wash the combined ether extracts with 5 per cent aqueous sodium hydroxide until the aqueous layer is basic; the final basic washing phase acquires a yellow color. Wash the ether layer with water until the washings are neutral, dry the ethereal solution over anhydrous calcium chloride and remove the ether on a rotary evaporator. Distil the residue and collect the benzofuran as a fraction of b.p. 170-172°C. The yield of colourless product is 9.5g (91%).

Reference: A. I. Vogel, Textbook of Practical Organic Chemistry, 5th Ed., Longman (1989)

Title: Interesting route to vinyl-dihydrobenzofurans..
Post by: phenethyl_man on July 29, 2004, 09:36:00 PM
2,3-bis(2-hydroxyethyl)phenol is converted to 4-(2-Chloroethyl)-2,3-dihydrobenzofuran utilizing the vilsmeier reagent, triethylamine and acetonitrile as the solvent.  The vilsmeier reagent is easily prepared from DMF or NMF and POCl3.

The product is then converted to the styrene using a PTC, NaOH and KI.


4-(2-Chloroethyl)-2,3-dihydrobenzofuran (6).

To a 500-mL, three-necked, round-bottomed flask equipped with a mechanical stirrer, thermocouple, and N2 was added Vilsmeier reagent (19.3 g, 151 mmol, 2.5 equiv). The flask was cooled to an internal temperature -15 °C to -20 °C and then charged with 76 mL of CH3CN. To the cooled yellow slurry was added 2,3-bis(2-hydroxyethyl)phenol (1) (10.97 g, 60 mmol, 1 equiv) in portions over 20 min. The reaction was exothermic, and careful control of the temperature to -15 °C was required. The reaction was stirred for 30 min until it was complete as judged by the disappearance of the triol by HPLC. A dilute solution of Et3N (24.39 g, 241 mmol, 4.0 equiv) in acetonitrile (1:1 by volume) was added slowly while the temperature was maintained between -15 °C and -20 °C. The reaction was heated to 50 °C for 3 h until the mono-imidate was completely converted to the chloroethyldihydrobenzofuran. The reaction mixture was cooled to room temperature and quenched with 25 mL of water; all the salts dissolved. The mixture was concentrated to one half the original volume and charged with 80 mL of MTBE and 35 mL of water. The organic phase was separated and washed twice with 55 mL of 10% w/v H3PO4 solution in 10% w/v brine. The MTBE phase was then assayed by GC and quantitated against a 4-(2-chloroethyl)-2,3-dihydrobenzofuran (6) standard and found to contain 10.1 g of 6 (92%). Samples for NMR assays were prepared by removing MTBE under reduced pressure. 1H NMR (300 MHz, CDCl3) ä 2.95 (t, 2 H), 3.15 (t, 2 H), 3.70 (t, 2 H), 4.45 (t, 2 H), 6.60 (d, 1H), 6.70 (d, 1H), 7.05 (t, 1 H); 13C NMR (CDCl3) ä 30, 37.5, 46, 72.5, 110, 122.5, 128, 130, 137.5, 162. HRMS [M + H] calcd for C10H11ClO, 182.05; found, 183.05.

4-Vinyl-2,3-dihydrobenzofuran (2).

To a 500-mL, three necked flask containing the MTBE solution of 4-(2-chloroethyl)-2,3-dihydrobenzofuran (6), 28 mL of water, 31.7 mL (600 mmol, 10 equiv) of 50% w/w NaOH, 13.8 mL (21 mmol, 0.35 equiv) of 40% w/v tetrabutylammonium hydroxide, and 1.0 g (6.0 mmol, 0.1 equiv) of solid KI were added. The reaction mixture was heated to 50 °C for 3 h until the reaction was found to be complete by HPLC (<1 relative area % of 6 remaining). On completion of the reaction, the phases were separated at 50 °C to minimize the loss of the product into the rag layer. The rag layer was discarded. The organic phase was cooled and washed once with 45 mL of 0.5 M sodium thiosulfate in 10% w/v brine followed by a wash with 45 mL of 1 N sodium hydroxide in 10% w/v brine. The MTBE solution contained 7.3 g (83%) of 4-vinyl-2,3-dihydrobenzofuran (2). Samples for NMR assays were prepared by removing MTBE under reduced pressure or by distilling crude oil as mentioned in the text. 1H NMR (300 MHz, CDCl3) ä 3.15 (m, 2 H), 4.45 (t, 2 H), 5.25 (d, 1 H), 5.65 (d, 1H), 6.6 (d, 1H), 6.7(d, 1H), 6.95 (d, 1H), 7.05 (t, 1H); 13C NMR (CDCl3) ä 28.84, 70.78, 108.25, 115.33, 117.48, 124.65, 127.87, 134.15, 134.52, 160.09. HRMS [M + H] calcd for C10H10O, 146.07; found, 147.07.

Ref:  Organic Process Research & Development 2003, 7, 547-550

Title: 2,3-dihydrobenzofuran from 2-chlorophenyl ethanol
Post by: phenethyl_man on July 29, 2004, 11:45:00 PM
Here is a nice simple route to 2,3-dihydrobenzofuran from 2-chlorophenyl ethanol..  The only drawback is the requirement of a very strong base (NaH) to achieve good yields.  In the article they get no yield using potassium carbonate, and only a 18.2% yield using NaOCH3.  Toluene can be used alone without the ethyl acetate causing only a slight drop in yield (81.4%).


A typical procedure:

To a dry, 100 mL, three-necked, round-bottomed flask equipped with condenser, thermocouple, and stir bar was added 2-chlorophenyl ethanol (1.0 g, 5.23 mmol) with toluene (12 mL). To the flask was then added NaH (0.165 g, 6.54 mmol, 1.25 equiv.) with toluene (4 mL). The reaction mixture was heated at 40°C for 15 min before cooling to room temperature. Then CuCl (0.026 g, 0.26 mmol, 0.05 equiv.) was added with toluene (4 mL) and EtOAc (0.03 g, 0.26 mmol, 0.05 equiv.). The reaction mixture was again heated at reflux for 24 h. The reaction mixture was cooled to room temperature, quenched with water (20 mL), filtered over Celite and the cake was washed with MTBE. The aqueous layer was separated and extracted with 2x25 ml MTBE. The combined organic layers were then washed with 100 mL 1N HCl followed by 2x100 mL portions of saturated NaHCO3 solution and 100 mL saturated NaCl solution. The final solution was dried with anhydrous sodium sulfate before being concentrated under reduced pressure to give ~0.7 g product, yield 88%.

Ref:  Tetrahedron Letters 41 (2000) 4011-4014

Title: As you all know, substituted ...
Post by: Lilienthal on July 30, 2004, 12:45:00 AM
As you all know, substituted dihydrobenzofurans can be easily converted into the aromatic benzofurans by treatment with DDQ or Pd/C at higher temperatures.
Title: More benzofuran possibilities
Post by: Kinetic on July 30, 2004, 09:25:00 AM
Thanks for the input everyone! This is getting interesting. ;)

My initial idea was to try to make 5-(2-aminopropyl)-2,3-dihydrobenzofuran which can - as Lili said - be dehydrogenated to the aromatic benzofuranyl analogue. If the one-pot cyclisation and formylation of 2-(2,4-dibromophenoxy)ethyl chloride works as hoped, I think this will still be the easiest way to get both products to test without performing two full syntheses. I still have other plans should the cyclialkylation fail.

phenethyl_man: That is a very nice method using 2-chlorophenylethanol; it's a real shame the precursor alcohol is so expensive. I researched into a simliar method last year, and have a paper detailing the synthesis of 2,3-dihydrobenzofurans from 2-aminophenylethanol via a diazo intermediate. The yield isn't great but the procedure is very easy, if you can get the 2-aminophenylethanol. Here is the paper:

Synthesis of 2,3-dihydrobenzofuran
Jan M. Bakke and Hæge M. Rohdolt
Acta Chem. Scand. Ser. B
, 34(1), 1980, 73-74


Rhodium: I also did some research on a similar method to the one you posted. I would add a slight modification to your method, however. If one were to start with 4-bromophenol, then the end product would of course be 5-bromobenzofuran, greatly assisting formylation in the desired 5-position. Phenol can be selectively brominated in the para-position using a number of reagents. The Bull Chem Soc Jpn paper I referenced in my first post in this thread had a regioselective para-bromination of phenol in 93% yield. Here is the full paper:

Halogenation Using Quaternary Ammonium Polyhalides. IV. Selective Bromination of Phenols by Use of Tetraalkylammonium Tribromides
Shoji Kajigaeshi, Takaaki Kakinami, Tsuyoshi Okamoto, Hiroko Nakamura and Masahiro Fujikawa
Bull. Chem. Soc. Jpn.
, 60, 4187-4189 (1987)

Reaction of phenols with calculated amounts of benzyltrimethylammonium tribromide or tetrabutylammonium tribromide in dichloromethane-methanol for 0.5-1h under mild conditions gave, selectively, the objective mono-, di-, or tribromophenols in good yields.


I also came across some other interesting references for the same para-bromination of phenol. In J. Chem. Soc. Chem. Commun., 23, 1996, 2679-2680, the authors para-brominate phenol in 89% yield using nothing but aqueous HBr in DMSO; I will retrieve this as soon as possible. The same transformation is carried out using elemental bromine in J. Org. Chem., 23, 1958, 280; likewise, bromine is also used in a fantastically ancient Compt Rend article. I can't find reference to any 4-bromophenol in this article, but then again I don't speak French. Here it is anyway:

Synthèse de la résorcine
M. W. Korner
Compt Rend
, 63, 564-566 (1866)


There are a lot more references, but I will finish with this patent, assigned to Dow in 1957:

Halogenation of phenols with cupric halides

Patent US2805263 (

It should be possible to formylate the thus formed 5-bromobenzofuran by forming the Grignard reagent and adding DMF followed by acidic workup. This works for the formation of benzaldehyde from bromobenzene in 88% yield (see

Post 512847 (

(Kinetic: "Grignard formylation article", Novel Discourse)
, above). The formylation certainly works if 5-bromobenzofuran is first treated with tert-butyllithium followed by DMF; for this, the article below (which may be interestig for several other reasons too):

Potent and Selective Non-Benzodioxole-Containing Endothelin-A Receptor Antagonists
Andrew S. Tasker et. al.
J. Med. Chem.
, 40(3), 1997, 322-330.


I think I'd better stop there for now.
Title: Benzofurans from Isovanillin
Post by: demorol on July 30, 2004, 09:55:00 AM
Synthesis of Benzofurans from Isovanillin via C-Propenylation-O-Vinylation and Ring-Closing Metathesis
Tzu-Wei Tsai,ab Eng-Chi Wang, Keng-Shiang Huang, Sie-Rong Li, You-Feng Wang, Yu-Li Lin, and Yung-Hua Chena
Heterocycles 63(8), 1771 - 1781 (2004)


Substituted benzofurans derived from isovanillin were synthesized. 2-Allyl-3-alkoxy-4-methoxyphenol, prepared from isovanillin via the Claisen rearrangement, O-alkylation and Baeyer-Villiger oxidation, were chloroethylated by two-phase reaction to furnish 1-allyl-3-alkoxy-(2-chloroethoxy)-4-methoxybenzenes. The given compounds were treated with potassium tert-butoxide to undergo the isomerization of O-allyl group and dehydrochlorination of 2-chloroethoxy group to efficiently construct the precursors with C-propenyl-O-vinyl function for the ring-closing metathesis (RCM) in one pot. Then, then precursors were subjected to RCM to furnish 4,5-O difunctionalized benzofurans in good over-all yield, respectively.


General procedure for the preparation of 1-allyl-2-(2-chloroethoxy)benzene (6a-d).

To the solution of allyl phenol (5a-d) (10 mmol) in dichloroethane (15 mL) was added NaOH solution (2.4 M, 15 mL), and TBAB (10% mol) at rt. The reaction mixture was stirred, and heated to the reflux for 8 h. Then, the organic layer was separated, dried under anhydrous MgSO4, and filtered. The filtrate was concentrated to remove the excess dichloroethane in vacuo, and the resulting residue was purified by column chromatography on silical gel (ethyl acetate: n-hexane = 1:12) to give pale yellow to colorless liquids (6a-d) in good yields.

General Procedure for the Preparation of Propenyl vinyloxybenzenes (7a-d)

To a stirred solution of 1-(1-propenyl)-2-chloroethoxybenzenes (6a-d) (5 mmol) in anhydrous THF (30 mL) was added potassium tert-butoxide (0.62 g, 5.5 mmol) at rt, and the reaction mixture was under reflux for 30 min. THF was removed from the reaction mixture in vacuo, and the residue was extracted with ethyl acetate (5×20 mL). The extracted solution was dried from anhydrous MgSO4. After filtration, the filtrate was concentrated in vacuo, and the resulting residue was purified by column chromatography on silica gel (ethyl acetate: n-hexane = 1:20) to give 7a-d, respectively.

General Procedure for the Preparation of Benzofurans (8a-d)

To a stirred solution of 1-(1-propenyl)-2-vinyloxybenzenes (7a-d) (1.15 mmol) in dichloromethane (23 mL) was added Grubbs’ catalyst (50 mg, 5% mmol) at rt, and the reaction mixture was stirred for 8 h. Excess dichloromethane was removed from the reaction mixture in vacuo, and the resulting residue was purified by column chromatography on silica gel (ethyl acetate: n-hexane = 1:50) to give 8a-d.
Title: 2,3-dihydrobenzofurans from phenethyl alcohols
Post by: phenethyl_man on July 30, 2004, 03:32:00 PM

I'm not sure what you mean; phenethyl alcohol is easily and inexpensively obtained or synthesized.  It is used abundantly in the cosmetics industry and I believe it is even present in respectable amounts in many essential oils.

It can be synthesized by the reduction of ethyl phenylacetate, or a friedel-crafts alkylation with ethylene oxide and your desired aromatic.

Once you have your phenethyl alcohol, just dihalogenate it, cyclize the 2,4-dihalophenethyl alcohol using the procedure above; then form the grignard and away you go..

As I said, the drawback here is finding a suitable base to obtain a decent yield.  I think at some point I will attempt this synth using potassium tert-butoxide.

However, depending on what base you use, it may fuck w/the second halogen.  I would only use 1.0 eq of base instead of the recommended 1.25 eq if you plan on keeping the halogen intact.  It's too bad sodium methoxide doesn't cut it as a base; then you could form a precursor to an MMDA-type 2,3-dihydrobenzofuran analogue in one step.

Of course there is a million ways to go about this; basically just halogenate a phenethyl alcohol that has the para position occupied with something you can manipulate later.

Title: Brominations and benzofuranyl DOM analogues
Post by: Kinetic on July 31, 2004, 08:06:00 AM
After seeing the cyclisation of 2-(2-chlorophenyl)ethanol I too thought of dibromination of 2-phenylethanol. I even included it in my post, then deleted it. My idea was essentially the same as yours: selectively dibrominate 2-phenylethanol, cyclise as in your post above (obtaining NaH shouldn't be a problem for me), then form the Grignard to formylate the ring. Dichlorination isn't a viable option here as aryl chlorides are notoriously difficult to prepare Grignard reagents from.

I deleted that part of my post because - to my amazement - the only reference I could find for a regioselective 2,4-dibromination of the similar toluene involved the use of bromine monofluoride ::)  at -78oC. The reference is J. Org. Chem., 53(23), 1988, 5545-5547. All other references appear to give an ugly mix of products.

In Chem. Lett., 32(10), 2003, 932-933, 2,4-dibromotoluene is produced in 6% yield, but appears to be a byproduct from the monobromination with NBS and FeCl3. Maybe using more NBS could produce the desired dibromo compound as the major product. But I don't have access to the journal so I can't comment further.

The authors of J. Indian Chem. Soc., 14, 1937, 157 brominate ethylbenzene with bromine in a mixture of GAA, fuming nitric acid and SO2 to give a mixture of 1-ethyl-4-brombenzene and 1-ethyl-2,4-dibromobenzene. But I don't have any more information on the selectivity.

A final paper which may actually be of use involves the (apparently) selective 2,4-diiodination of ethylbenzene with aqueous iodic acid, iodine, and sulfuric acid. The reference is J. Prakt. Chem., 14, 1961, 24-33. If you have any more references, or ideas, I would very much like to see/hear them. Two heads are better than one.

If you want to try a MMDA type benzofuran analogue, you might like to read the related articles below. I had the idea of a cyclic DOM or 2C-D analogue (dibromination of 4-methoxyphenol, O-alkylation with ethylene chloride, one-pot cyclisation and formylation with Mg followed by DMF to give the benzaldehyde) but according to the papers it has a 10-fold decrease in potency compared to DOM [Edit: Oops - this proposal will not give the desired benzaldehyde but an isomer, 7-formyl-5-methoxy-2,3-dihydrobenzofuran, whose corresponding amphetamine or PEA has not been made. Maybe they will be of interest; dehydrogenation to the benzofuran will give a hemi bis-benzofuranyl analogue]. Anyway, here are the articles:

Synthesis and Evaluation of 2,3-Dihydrobenzofuran Analogues of the Hallucinogen 1-(2,5-Dimethoxy-4-methylphenyl)-2-aminopropane: Drug Discrimination Studies in Rats
David E. Nichols,* Andrew J. Hoffman, Robert A. Oberlender, and Robert M. Riggs
Journal of Medicial Chemistry
, 29, 302-304, (1986)

Two analogues, 6-(2-aminopropyl)-5-methoxy-2,3-dihydrobenzofuran and 6-(2-aminopropyl)-5-methoxy-2-methyl-2,3-dihydrobenzofuran, of the hallucinogenic agent 1-(2,5-dimethoxy-4-methylphenyl)-2-aminopropane (DOM) were synthesized and tested in the two-lever drug discrimination paradigm. In rats trained to discriminate saline from LSD tartrate (0.08 mg/kg), stimulus generalization occurred to both of the 2,3-dihydrobenzofuran analogues but at doses more than 10-fold higher than for DOM. A possible explanation for this dramatic attenuation of LSD-like activity could involve a highly directional electrophilic binding site on the receptor that cannot accept the orientation of the unshared electron pairs on the heterocyclic oxygen atom in the benzofurans.


2,3-Dihydrobenzofuran Analogues of Hallucinogenic Phenethylamines
David E. Nichols,* Scott E. Snyder, Robert Oberlender, Michael P. Johnson, and Xuemei Huang
Journal of Medicinal Chemistry
, 34, 276-281 (1991)

Two 2,3-dihydrobenzofuran analogues of hallucinogenic amphetamines were prepared and evaluated for activity in the two-lever drug-discrimination paradigm in rats trained to discriminate saline from LSD tartrate (0.08 mg/ kg) and for the ability to displace [125I]-(R)-DOI ([125I]-(R)-1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane) from rat cortical homogenate 5-HT2 receptors. The compounds, 1-(5-methoxy-2,3-dihydrobenzofuran-4-yl)-2-aminopropane (6a) and its 7-brominated analogue 6b, possessed activity comparable to their conformationally flexible counterparts 1-(2,5-dimethoxypheny1)-2-aminopropane (3) and, its 4-bromo derivative DOB (5), respectively. The results suggest that the dihydrofuran ring in 6a and 6b models the active conformation of the 5-methoxy groups in 3 and 5. Free energy of binding, derived from radioligand displacement KA values, indicated that addition of the bromine in either series contributes 2.4-3.2 kcal/mol of binding energy. On the basis of surface area of the bromine atom, this value is 2-3 times higher than would be expected on the basis of hydrophobic binding. Thus, hydrophobicity of the para substituent alone cannot account for the dramatic enhancement of hallucinogenic activity. Although this substituent may play a minor role in orienting the conformation of the 5-methoxy group in derivatives such as 4 and 5, there appears to be some other, as yet unknown, critical receptor interaction.


Title: "super-potent" compound from eugenol?
Post by: phenethyl_man on August 02, 2004, 02:23:00 AM
It's funny, when Nichols published that article about the "super-potent" bis-dihydrofuran analogues of the 2,5-dimethoxy-substituted series of hallucinogenic amphetamines, I got this crazy idea for a novel compound of this sort which might be able to be synthesized from eugenol.  The scheme goes something like this..


It's all very basic chem, but if you want a ref on any of the reactions just say so.  Basically, it goes:  alkylation of eugenol w/an allyl halide, claisen rearrangement, cyclization w/GAA to the mono-2-methyldihydrobenzofuran, elbs persulfate oxidation, and finally, another cyclization to form a 5-methoxy-bis(2-methyldihydrobenzofuran).

I have not even attempted this synth; I am concerned the claisen rearragement or the persulfate oxidation will not work as advertised..

However, if this compound were to somehow be synthesized, or for that matter, the benzofuran or dihydrobenzofuran analogues (minus the 2-methyl group), what would make them even more interesting is the orientation of the ring-oxygen atoms.

They are oriented in the notoriously difficult to synthesize and largely unexplored 2,3,5 configuration (see Shulgin's commentary in PiHKAL under the TMA-4 entry.)  Since MMDA-4 hasn't even been synthesized, who would know what kind of activity to suspect, however, I'd certainly be willing to bet on the fact that it won't be inactive   ;)

Title: On chlorinations and brominations
Post by: Kinetic on August 03, 2004, 02:23:00 PM
Now that'd be something fun to make from eugenol. ;)

But I fear that the isolated alkene may react with persulfate more favourably than the aromatic ring. The same conditions (persulfate and OH-) required to add a hydroxyl group to the ring can also oxidise an aromatic methyl group to an aldehyde. Maybe you'd end up with a glycol as byproduct, I don't know. I should stress that I have no experience with the reaction, and this worry is only based on the information from the Merck Index entry:

Elbs Persulfate Oxidation (

( If you have any references for this reaction (especially in the presence of an isolated alkene) I would like to see them. :)  For the ring closing steps I have similar refs as Shulgin uses them in his synthesis of the F-series.

Back to the original topic for now: after making the 2-phenoxyethyl chloride I tried the dibromination, but unfortunately my product didn't crystallise, nor did it appear to have taken up the full two equivalents of bromine. Since I need to be sure I have pure 2-phenoxyethyl chloride I have decided to make this by the chlorination of 2-phenoxyethanol. Until I get a crystalline substance in both steps I don't want to proceed with the Grignard. My first idea was to use concentrated HCl under PTC conditions, using the following procedure which has been referenced here several times:

Organic Synthesis in Micellar Media. Oxidation of Alcohols and Their Conversion into Alkyl Chlorides
Branko Jursic
, 1988, 868-871

The use of micelles was investigated for various organic reactions: oxidation of alcohols with sodium hypochlorite in micelles, oxidation of alcohols with hexadecyltrimethylammonium chromate as micelle, and conversion of primary alcohols to 1-chloroalkanes by aqueous hydrogen chloride in the presence of micelles. In all cases, product isolation was simple and satisfactory yields were obtained.


However, as the reaction conditions could cleave the ether, I think I will instead try the following. This procedure also happens to give superior yields:

An Efficient Route to Alkyl Chlorides from Alcohols Using the Complex TCT/DMF
Lidia De Luca, Giampaolo Giacomelli,* and Andrea Porcheddu
Organic Letters
, 2002, 4(4), 553-555

Efficient conversion of alcohols and beta-amino alcohols to the corresponding chlorides (and bromides) can be carried out at room temperature in methylene chloride, using 2,4,6-trichloro[1,3,5]triazine and N,N-dimethyl formamide. This procedure can also be applied to optically active carbinols.


Finally, here is the paper I mentioned earlier: the bromination of arenes using nothing but aqueous HBr in DMSO. I may well try this on benzodioxole; hopefully the dioxole ring will survive:

Novel site-specific one-step bromination of substituted benzenes
Sanjay K. Srivastava, Prem Man Singh Chauhan* and Amiya P Bhaduri
Chem Commun
, 1996, 2679-2680

Regiospecific bromination of benzene derivatives has been carried out with Me2SO-HBr; this method gives excellent yields of 2-bromobenzaldehyde and 2-bromonitrobenzene; strong ortho- and para- directing monosubstituted benzenes give para-bromo derivatives; a general discussion of the mechanism of these reactions is given.


Title: hmm.. I wonder if this procedure could be...
Post by: phenethyl_man on August 05, 2004, 04:01:00 AM
hmm.. I wonder if this procedure could be applied to bromo phenols with a formyl group already in place; for ex. from the easily acquired 5-bromo-vanillin.. would be much easier then the MMDA synth from the same, and prob near as potent.


kinetic;  did you synth dichloroethane?  if so, could you give the procedure you used?

why don't you describe your dibromination as well, maybe we can figure out where it failed..

Title: drug design and its difficulties
Post by: longimanus on August 08, 2004, 10:16:00 AM
 it`s nice to see that someones brain is working. But sometimes another person should help the first one showing him/her his/her mistakes. It`s called criticism and it`s very essential for our work at The Hive.

 What I`m talking about is the compound you suggested in

Post 523206 (

(phenethyl_man: ""super-potent" compound from eugenol?", Novel Discourse)
. We all know that the potency of that compound - 2-(5-methoxy-2,7-dimethyl-2,3,7,8-tetrahydrobenzo[1,2-b;3,4-b']difuran-4-yl)-1-methylethylamine - depends on its interaction with the 5-HTRs. And the ligand-receptor interaction is relevant to some of the compound properties. One of them is the lone pairs orientation.

Acta Pharmaceutica Suecia Suppl. 1985:2 (


...Perhaps most interesting was the finding by Shulgin (personal communication) that the compound where R=H 2-(5-methoxy-2,3-dihydrobenzofuran-6-yl)-1-methylethylamine - l., as hydrochloride, produced no effects in humans when administered in an acute oral dose up to 20 mg.

...if arguments about oxygen lone pair directionality are valid, this would indicate that a potential electrophilic site on the serotonin receptor may have strict requirements for electron pair directionality in substituted phenethylamines. As has been noted by Glennon et al. [

J. Med. Chem. 1980, 23, 294-299 (

(] - l., a similar situation applies to phenethylamines with a 2-methoxy group, that have been also substituted at the 3-position.

 And now lets look at the model of phenethylamine binding to the human 5-HT2A receptor suggested in

Journal of Computer-Aided Molecular Desing, 16: 511-520, 2002 (



or more schematic:


 And after all a comparison between the oxygen lone pair directionality in Nichol`s "super-potent" and your compound (without the "2-methyl"s):


 Different, don`t you think so?
 Possibly lack of interaction with 3.36 and possible interaction with 5.46; possibly lower affinity than DOM or even maybe different pharmacological profile.
 What I`m trying to say is that that compound is not likely to be a very potent 5-HT2A agonist.
 Keep digging.

Title: further speculation..
Post by: phenethyl_man on August 09, 2004, 02:31:00 AM
longimanus:  I appreciate all the information, however, I never stated that this theoretical compound would have a similar pharmacological profile as DOM, nor that it would be as potent; only that it would be something interesting to explore, and for precisely that reason; it is different.

The two compounds I did compare it to, TMA-4 and MMDA-4, one is unexplored and the other Shulgin states to have activity around 80mg.  So it is still more potent than the traditional 3,4,5 orientation of TMA and mescaline.  TMA-2, which has the orientation which Nichol's "super-potent" compound is based upon, is active around 20-40mg.  It would be logical then to assume that the compound I proposed would at least be quite significantly more potent than TMA-4.  After all, every amphetamine with three-MeO groupings is active at some level, except for TMA-3 I believe.

I believe Nichol's point in that paper is that the oxygen lone pair directionality of DOM is what contributes to it's extraordinary potency.  However, if potency was all that mattered, compounds like MDA wouldn't even be worth exploring..

Title: "greater than 80 mg"
Post by: longimanus on August 09, 2004, 07:15:00 AM
Greater than 80 mg doesn`t mean it`s active at this range - it`s the bottom threshold, so TMA-4 should be twice as potent as mescaline. On the other side lets compare TMA-4 with G-3 - it`s dose is 12-18 mg!!! So, we even don`t need that 3- or 4-positioned oxygen, actually it seems to be lowering the activity.

 And now the numbers - if your compound (have you already named it after sth?) binds to the 5-HTR in a manner similar to that of DOM its active dose ahould be (just some scientific quessing ;) ) around 25-100 mg or more. But I`m almost sure that its binding and pharmacological profile will be rather different. That`s why I`d be very interested in the exploring of that cmpd when it`s synthesized. Hope you`ll be the one to do this :) .
Title: I doubt I will be the one to do it, at least...
Post by: phenethyl_man on August 09, 2004, 09:36:00 PM
I doubt I will be the one to do it, at least not in the near future, simply don't have the time..

If I were to try thou.. a good place to start would be to stop after the third step in that synth I proposed.  Right there you have an allylbenzene that would be good for a MMDA analogue.  I would probably use oxymercuration on that, demercuration to the phenyl-2-propanol, oxidation to the 2-propanone and finally reduction to the amine.  I recently used this process for safrole to MDP2P and it was by far the easiest synth from an allylbenzene I have ever performed.  If this compound was inactive, I wouldn't even bother with the latter.

Kinetic is right about the step following that thou, a persulfate oxidation right there would also oxidize the allyl side chain.  Instead, the compound would have to be brominated or iodinated right there, the halogen would go para to the MeO group, and then the halogen could be stripped to a hydroxy using a copper catalyst and NaOH using well known methods (even thou I hate this reaction).  Then the second cyclization could be performed and the benzene could then be formulated or whatever is desired.  Anyone see any pitfalls there?

Of course the good thing about the synth would be the cheap and unlimited supply of the starting material (eugenol).

It's interesting you bring up G-3, an indane.  It is true that the analogue of MDA with no oxygens on the ring (IAP) is more potent in the binding studies I have seen than any with oxygens on there.  I know of no good synthesis for indane though, can anyone chime in here?  The only one I know of is converting phenylpropanoic acid to an acid chloride, cyclization only gives you the indanone, which still needs to be reduced to indane.  A real PITA.  I guess that's why most all the synths for indanyl-phenethylamines I have seen start with the indane already formed.

Title: On indane
Post by: Kinetic on August 10, 2004, 07:13:00 AM
Maybe we should start a new thread for a discussion on indanyl analogues; it's something I'm also very interested in. I actually made some IAP for the first time last week, and have tried it twice. It is most definitely active at 30mg; a distinct mood elevation, but nothing much else. Starting from indane is 'safe' for me as there's essentially no way I can end up with a controlled substance.

I had a similar idea about a MMDA indanyl analogue, which would be the analogue of MMDA-2. I would synthesise it by the 5-bromination of indane (72% yield), methoxylation of this to 5-methoxyindane and formylation to 6-methoxyindan-5-carboxaldehyde, probably using a modified Duff as I don't have any POCl3 handy. Of course, I have no idea if the final product will be active or interesting. If you start a thread on the synthesis of indane or analogues containing the indanyl skeleton I'll add whatever relevant information I can find. I have a couple of references for the synthesis of indane, but at first glance there doesn't seem to be any simple way. A PITA is probably a good description, if it stands for what I think it does.

In response to your earlier questions: I bought my 1,2-dichloroethane as it is so cheap; and the problems with the bromination are at least in part due to the impure starting 2-phenoxyethyl chloride. Once I get this pure (hopefully soon) I should have no trouble with the subsequent bromination following the procedure I used for the bromination of 2-phenoxyethyl bromide. I need to make sure there is as little monobrominated product as possible as the cyclisation won't take place without a bromine atom ortho- to the 2-chloroethoxy group, and - as you will know - the best way to ensure complete dibromination is to start with as pure a product as possible. The result of incomplete bromination would be the contamination of my end product with the illegal 4-ethoxyamphetamine or the probably toxic 4-(2-chloroethoxy)amphetamine, depending on the method I used for the reduction of the nitrostyrene.
Title: MMDA analogue; benzodifuran paper..
Post by: phenethyl_man on August 10, 2004, 08:19:00 AM
kinetic;  ya, unfortunetly I don't even have an account with a chem supplier anymore (they keep asking for a business license, go figure), so buying indane is not a possibility.  Did you use a modified Duff formylation on indane?  What acid did you use?

Right now I'm working on a dihydrobenzofuran MMDA analogue from vanillin.  I happenend to have some 2-butoxyethanol lying around, so I cleaved the ether with HBr and hopefully some dibromoethane is crystallizing out right now.  I plan to react that with vanillin and NaOH to produce vanillin bromoethyl ether; and that should be easily cyclized to 7-methoxy-2,3-dihydrobenzofuran-5-carbaldehyde-- precisely the same molecule as myristicinaldehyde except only one lone oxygen on the furan ring (ortho to the methoxy group.)  Should be quite an interesting compound and so easily synthesized..  maybe I'll name it MDBAP: 1-(7-methoxy-2,3-dihydrobenzofuran-5-yl)-2-aminopropane. ;)

I also have something else very interesting to add to this thread.  I actually found a synthesis for the symmetrical version of tetrahydro-benzo-difuran from the year 1920!  They use resorcinol, form the di-beta-hydroxyethyl ether, and then perform a cyclic dehydration with P2O5 to form the difuran.

I would venture to guess, that using hydroquinone here instead of resorcinol would form the assymetrical version; which would be a precursor for the potent compounds in the Nichol's paper.  Here is the ref: J. Am. Chem. Soc.; 1920; 42(1); 157-165.

Title: Duff
Post by: Kinetic on August 10, 2004, 09:39:00 AM
Did you use a modified Duff formylation on indane?  What acid did you use?

Indeed, and the acid was TFA. See (

The 5-bromination of indane is also covered. ;)

That is a nice idea for the synthesis of MDBAP but are some things you should be aware of, assuming you are planning on a Parham-type cyclialkylation:

Unlike with the less reactive dichloroethane, etherification with dibromoethane may lead to dphenoxyethane formation as well as the desired monoalkylation. From the Synthesis article in

Post 512847 (

(Kinetic: "Grignard formylation article", Novel Discourse)

Bromoethyl ethers 1a and 1b were obtained from phenols 2 in only moderate yields,11 due to formation of diphenoxyethanes.

Further, the bromoethyl ether when treated with Mg is more likely to lead to elimination than the chloroethyl ether:

The ease of halogen-metal exchange has been found to be ArBr> ArOCH2CH2Br> ArOCH2CH2Cl and it is essential that conditions are used, which favour the formation of the organometal derivatives 4 leading to cyclisation and 6, rather than the alkyl-metal derivatives 5, which result in dealkylation and generation of the phenoxides 8. In order to minimise this side-reaction, we reasoned that utilisation of the chloroethyl ethers would allow selective arylbromide-lithium exchange to occur at temperatures higher than -100oC and render the reaction more suitable for large scale plant [as well as most bees].

Finally, of course, the aldehyde function will need protecting, but this could be almost to your advantage, if you were to dry it thoroughly using a Dean-Stark trap and ethylene glycol. The necessary acidic workup would regenerate the carbonyl from the acetal.

The above is based on the assumption you are performing the reaction I suspect: bromination of vanillin ortho- to the OH followed by alkylation with a 1,2-dihaloethane, then cyclisation with Mg. All of these problems should be surmountable with a little work. Maybe you could try and make some 1-bromo-2-chloroethane from your 2-butoxyethanol: it should assist the desired etherification, and the chloroethoxy compound will perform better in the subsequent cyclisation. I look forward to hearing how you get on.

Some references:

For the formation of 4-(2-bromoethoxy)-3-methoxy-benzaldehyde from ethylene bromide and vanillin, see Monatsh. Chem., 88, 1957, 1064-1067, and J. Org. Chem., 19, 1954, 1029-1031. The product melts at 69-70oC from aqueous IPA, and 66-67oC from water.

For the formation of 4-(2-chloroethoxy)-3-methoxy-benzaldehyde from ethylene chloride and vanillin in 85% yield, see  J. Med. Chem., 42(4), 1999, 649-658. The product melts at 60-61oC from ether.

Title: kinetic; check out the paper in the edit to my
Post by: phenethyl_man on August 10, 2004, 10:04:00 AM
kinetic;  check out the paper in the edit to my previous posting; also take a look at this one:  J. Am. Chem. Soc.; 1919; 41(4); 665-670.  I think you will find them quite interesting..

My synthesis of 4-(2-bromoethoxy)-3-methoxy-benzaldehyde from vanillin and DBE seems to have worked well.  I slowly added the DBE to the vanillin dissolved in a stoichiometric methanolic NaOH soln and then I heated it under reflux for one hour.  Judging by the large precipitation of NaBr, something got alkylated.

Title: JACS 1919
Post by: Kinetic on August 10, 2004, 02:07:00 PM
Thanks for the refs; as it happens, I have the earlier article as a PDF already. In fact my carpet still smells of partially polymerised 2-phenoxyethanol from an attempt at the direct cyclisation to dihydrobenzofuran last summer. Unfortunately, nothing was isolated from the carpet the product was spilled on. Fortunately, I still have some 2-phenoxyethanol for my proposed chlorination to 2-phenoxyethyl chloride, using trichloroisocyanuric acid and DMF.

I will get the JACS 1920 article asap.

I almost mentioned the Friedel-Crafts alkylation in my earlier post; is this the method you're using? It certainly is a lot easier. Fingers crossed the O-alkylation has worked as planned, and good luck for the cyclisation, whichever route you're using. Maybe the more activated ring of 4-(2-bromoethoxy)-3-methoxybenzaldehyde will lead to a higher yield than the 30-40% for plain 2,3-dihydrobenzofuran from 2-phenoxyethyl bromide.

Here is the article from JACS 1919 for those who are interested. Although the yield of dihydrobenzofuran is relatively low, the procedure is extremely easy:

Synthesis of Chromanes and Coumarans
R. E. Rindfusz
Journal of the Americal Chemical Society
, 41(4), 665-670, 1919



Coumarane [2,3-dihydrobenzofuran] from beta-Hydroxyethylphenyl ether C6H5O.CH2CH2OH [2-phenoxyethanol]

This ether is prepared in 50% yields from sodium phenolate and ethylene chlorohydrine as described by Bentley, Haworth and Perkin.4 It is a colourless liquid boiling at 134-135o at 18mm. [n]D20 1.534, d22 1.102. 50g of this ether is heated for 5 hours with 5g of zinc chloride. The temperature goes at first to 225o and then slowly drops to 190o. The product may be distilled directly and boils at 88-90o at 18mm. Yield 25%, d24 1.0576,[n]D20 1.543.

Coumarane from p-Bromo-ethyl-phenyl Ether [2-phenoxyethyl bromide]

The beta-bromoethyl phenyl ether is prepared by treating sodium phenolate with an excess of ethylene bromide, following the method of Weddige.1 On treating this with 1/10 of its weight of zinc chloride, the reaction is not so vigorous as in the analogous formation of chromane and two hours' heating is necessary. The product may then be distilled directly in 30-40% yields.
Title: yea.. the JACS 1920 article is part II of the...
Post by: phenethyl_man on August 10, 2004, 02:50:00 PM
yea.. the JACS 1920 article is part II of the 1919 article.  In this one he realizes that the cyclization takes place merely due to dehydration instead of his halogenation hypothesis stated in the first article.  Thus, he uses phosphorus pentoxide as a dehydrating agent instead of zinc chloride and obtains around 50-80% yields as opposed to only a 10-15% yield with ZnCl2.

Here is the general procedure:


Preparation from y-hydroxypropyl-o-tolyl ether and phosphorus
pentoxide. Forty g. of phosphorus pentoxide was suspended in 200 cc.
of dry benzene and 100 g. of the hydroxy ether slowly added with shaking.
After refluxing for a short time, the mixture was poured from the
phosphorus compounds and distilled. The product may be washed with
alkali and with water and redistilled with very little loss. An excellent
grade of material practically all boiling at 114-115 deg at 20 mm. was obtained
in 76% yield.

For coumaranes, just use the hydroxyethyl ether instead.

Title: Well, after one of those nightmarish ...
Post by: phenethyl_man on August 11, 2004, 09:49:00 PM
Well, after one of those nightmarish post-reaction workups, I finally recovered the bromoethyl ether from the reaction mixture in good yield.  It is very soluble in water; Et2O would hardly even touch the product in the aqueous..

Title: "very soluble in water"?
Post by: longimanus on August 12, 2004, 10:52:00 AM
OK, phenethyl_man, lets make it clear - you`re talking about 4-(2-bromoethoxy)-3-methoxybenzaldehyde, aren`t you?
 The strange thing here is that the estimated logP for that compound is 2,36 but now you`re telling us that your product is "very soluble in water"! I`m not sure if the real logP is the same and if someone has any info, please, share it.
 Just informatively:
logP(vanillin)= 1,26
logP(4-(2-hydroxyethoxy)-3-methoxybenzaldehyde)= 1,01(estimated); 0,58(experimental) - maybe your product :(
Title: You guessed it; not the expected product..
Post by: phenethyl_man on August 12, 2004, 02:40:00 PM
You guessed it; not the expected product.. just another failure to add to the list.

In the first reaction, the HBr cleaved the ether of butyl cellosolve leaving the bromohydrin, however the hydroxy group was not replaced by bromine as expected (merely adding some H2SO4 to catalyze this reaction almost surely would have overcome this problem.)

Thus, the subsequent alkylation left me with 4-(2-hydroxyethoxy)-3-methoxybenzaldehyde..  and if only I didn't have that damn formyl group to worry about, then I could oxidize this to a phenoxyacetaldehyde and condense/cyclize with HOAc/ZnCl2 to get the unsaturated furan, but oh well..

I giveth up for now..

Title: Failed cyclialkylation
Post by: Kinetic on September 10, 2004, 10:30:00 PM
I tried and failed in my attempt to cyclise and formylate 2-(2,4-dibromophenoxy)ethyl chloride to 2,3-dihydrobenzofuran-5-carboxaldehyde in one-pot.

Bearing in mind the results azole recently posted (

Post 522253 (

(azole: "3-Br-4-(2-chloroethoxy)BA failed to cyclize", Novel Discourse)
), it seems the reaction may not be possible to do without the use of BuLi. The cyclisation of 2-(2-bromophenoxy)ethyl chloride with Mg is clearly possible (

Post 510718 (

(Kinetic: "2,3-Dihydrobenzofurans without BuLi", Novel Discourse)
) but the extra bromide on the ring may interfere in this case. It could well be that the Parham cyclisation only works with 2-(2,4-dibromophenoxy)ethyl chloride because it selectively metalates the position ortho- to the stabilising oxygen first. If Mg is not selective in this way, then the para- position may well react first instead. This could interfere with the subsequent Grignard formation and Wurtz-type coupling which is necessary to form the dihydrobenzofuran ring.

After the above failures, I am leaning more towards the synthesis of the dehydrogenated aromatic analogue 5-(2-aminopropyl)benzofuran. I will post a proposal for the synthesis of the precursor aldehyde soon: via para-bromination of phenoxyacetic acid, followed by cyclisation (either via the acid chloride followed by Lewis acid, or simply with polyphosphoric acid), then reduction of the formed benzofuranone and dehydration to 5-bromobenzofuran. The dehydration will happen readily as the product is aromatic, and it should be possible to react the product with Mg followed by DMF to give the benzaldehyde.

Below is the synthesis of 2-(2,4-dibromophenoxy)ethyl chloride. This will hopefully be useful for anyone who wants to make 2,3-dihydrobenzofuran-5-carboxaldehyde from it using BuLi (see the very interesting Tet Lett article posted above in

Post 499729 (

(Kinetic: "An interesting possiblilty", Novel Discourse)
for this), but also because the two steps are similar to the etherification of hydroquinone with 1,2-dichloroethane followed by dibromination to give 1,4-bis(2-chloroethoxy)2,5-dibromobenzene; this of course is an intermediate in the synthesis of Nichols' super-potent DOX analogues.

2-Phenoxyethyl chloride

23.5g phenol (250mmol)
30.0g NaOH (750mmol)
5g aliquat 336 (12.5mmol, 5mol%)
2.5g sodium metabisulfite (12.5mmol, 5mol%)
200ml 1,2-dichloroethane
200ml water

A biphasic solution of the above was vigorously stirred at reflux for 12 hours, then allowed to cool and acidified with a small amount of concentrated HCl. The phases were separated, the organic layer washed with 2x100ml water and then dried over MgSO4. Removal of the solvent left a liquid which was distilled at 93-107oC (using a water pump) to give the title product as a clear, colourless liquid.

Yield: 28.2g (180mmol, 72%)

Notes: The product is a solid in the fridge but melts rather quickly on warming to room temperature. According to the literature the product should melt at around 25oC, but the product is pure enough to use in the next step.
The product smells identical to an authentic sample of 2-phenoxyethyl bromide.

2-(2,4-Dibromophenoxy)ethyl chloride

31.3g 2-phenoxyethyl chloride (200mmol)
70.3g bromine (440mmol)
60.0g anhydrous zinc chloride (440mmol)
Acetic acid

A solution of the 2-phenoxyethyl chloride and zinc chloride in 150ml acetic acid was cooled to around 5oC. The flask was covered in aluminium foil to exclude the contents from light and, with ice-bath cooling, a solution of bromine in 25ml acetic acid was added over 1 1/4 hours. The flask was then removed from the ice-bath and stirring was continued for a further 3 hours. The clear orange solution was then added to 800ml water, and the pale semi-liquid which precipitated was extracted with 2x60ml DCM. The combined orange extracts were washed in succession with 100ml water, 100ml 0.5M sodium metabisulfite solution, 2x100ml 1M potassium carbonate solution and 100ml brine. After drying over MgSO4 and removing the solvent, a very pale, colourless oil remained. This quickly set to sparkling crystals weighing 49.5g (crude yield: 157mmol, 79%). Recrystallisation from 50ml methanol provided the title product as absolutely clear, sparkling plates.

Yield: 43.3g (138mmol, 69%)

Notes: The two equivalents of zinc chloride appear to be necessary, as when a catalytic amount was used, the yield was lower and the product refused to crystallise at room temperature. As this product had a similar smell to 4-bromoanisole (aniseed), this was probably due to incomplete bromination.
The sodium metabisulfite removes all of the remaining orange (bromine) colour.
The first potassium carbonate wash takes on a very slight yellow tint, and the second does not change colour. Phenolic byproducts therefore appear to be negligible.
Using more methanol for the recrystallisation leads to a large amount of the product being held up in solution, even on standing in the freezer.