Org Prep Daily

February 16, 2010

4-(2′-bromoethyl)-3-chloro-2-inolinone from isochromane

Filed under: procedures — milkshake @ 11:44 am

1) 2-(2′-bromoethyl)-benzaldehyde:

A solution of isochroman 25.30g (188.5 mmol) in acetic acid 100mL was cooled on ice-water bath. 33% HBr in acetic acid 3.0mL was added followed by a “tip of spatula” amount of solid AIBN, followed by neat bromine 10.0 mL over 10 min (carefully!). The mixture was stirred on ambient water bath for 1 day. 33% HBr in acetic acid 50mL was then added and the mix was refluxed on a 90C oil bath for 30 min. Acetic acid and HBr was then removed carefully by vacuum distillation first at 50 Torr from 90C bath, and then at 15 Torr/90C. The obtained residue was then distilled at 0.6 Torr from 90 -140C oil bath. The main fraction distilled  at 105C/0.6 Torr. Y=34.35g (85.5% th) of a pale blue oily liquid that is somewhat impure (about 15% impurity) and it is best used immediately for the next step.

2) 2-(2′-bromoethyl)-beta-nitrostyrene:

A mixture of methanol 100mL and 25% wt methanolic sodium methoxide solution 1.0mL (Aldrich) was cooled on ice bath and nitromethane 9.0mL was added, followed by neat crude 2-(2′-bromoethyl)-benzaldehyde 25.89g (121.5 mmol) from the previous step. (The flask containing the aldehyde was washed with a portion of the reaction mixture and the washings were added back to the reaction mixture). The reaction mix was stirred on ice for 10 min, then additional 25% wt sodium methoxide in methanol 33mL (=30g) was added over 10 min and the stirring on ice was continued for 90 min.  Meanwhile 600mL of 6M HCl (conc. HCl + water 1:1 by volume)  in a large beaker was chilled on ice bath. The reaction mix was poured into cold stirred 6M HCl in a thin stream and the resulting sludge was stirred on ice for 30 min. The solids were collected by filtration using a large glass Buchner funnel, washed thoroughly with ice-cold water, then rinsed with a small volume of cold hexane (about 20mL) on the Buchner funnel, dried by suction and on highvac. The crude product was re-dissolved in hot benzene 30mL and the solution was gradually diluted with cyclohexane 250mL at reflux and the solution was then allowed to crystallize overnight. The crystalline precipitate was collected by filtration, washed with hexane and dried on highvac. This provided 13.88g of pure nitrostyrene product. Additional 5.673g of pure product was obtained by concentrating the supernatants and purifying the residue on a column of silica (100g) in 3:1 hexane-ethyl acetate. (The zone of the purified product on silica has a tendency to crystallize on he column so it is advisable to avoid using Combiflash and work reasonably fast. ) Combined Y= 19.553g (63% th)


Anhydrous FeCl3 30g was suspended in anhydrous dichloromethane 450mL under Ar and cooled to -18C on ice/salt slush. Acetyl chloride 9.0mL was added dropwise and the mixture was stirred for 15 min. The nitrostyrene from the previous step, 15.77g (61.57 mmol) dissolved in anh. dichloromethane 50mL was then slowly added through a canula over 20 min at -18C. After additional 45 min at -18C the mix was placed on ice bath and stirred at 0C for 4 hours 30 min. The reaction mix was  then poured onto crushed ice 300g in a large beaker and the mix was stirred for 45 min. The obtained biphasic mix was filtered through a 150mL glass medium-porosity Buchner funnel, the obtained precipitate was thoroughly washed with water and then with small volume of chilled dichloromethane (2x10mL)  to provide 4.498g of pure product as a off-white powder. The combined filtrates were separated, the organic phases were washed with half-saturated NaCl aq. solution. The aqueous phases were re-extracted with dichloromethane (3x300mL). The combined organic extracts were dried (MgSO4) and concentrated down to about 100mL volume, then placed to a fridge (0C) overnight. The precipitated second crop was collected by filtration, washed with a small volume of chilled dichloromethane and dried. This provided 4.139g of pure product. A third crop 1.250g was obtained by applying the supernatants on a column of silica (80g) and eluting with EtOAc gradient in DCM, 0 to 20% EtOAc, UV detection @261 nm. The combined Y=9.887g (56.5% th)

1H(d6-DMSO, 400MHz): 10.787(s, 1H), 7.275(t, 7.8Hz, 1H), 6.968(d, 7.8Hz, 1H), 6.778(d, 7.7Hz, 1H), 5.708(s, 1H), 3.755(m, 2H), 3.217(m, 2H)


  1. Cool stuff. Almost seems like alchemy or black magic to me, and I’m a chemist…

    Comment by Kai — February 16, 2010 @ 6:01 pm

  2. I was also impressed – this comes from a process research paper, on a dopamine agonist scale-up, from 80s, unfortunately I don’t have the exact reference at hand. No explanation was given for the nitrostyrene-to-oxindole cyclization and the selective removal of chlorine with hypophosphite. Then they commented that the bromoethyl group is not very reactive and has tendency to eliminate to vinyl group when they attempted do displace it with di-n-propylamine, so they re-did the whole scheme with benzoyl-oxo instead of bromine and tosylated at the end.

    Comment by milkshake — February 16, 2010 @ 6:20 pm

  3. All this neat chemistry is unfortunately completely beyond a lowly undergrad like, though I still enjoy all the chem tidbits whenever your blog is updated. For some odd reason I’ve taken a liking to seeing all this… how shall we call it, “non-traditional” chemistry in the sense that it’s definitely not something you’d see in textbooks. The subtleties and endless variations of synthetic chemistry never ceases to amaze me.

    On that note, I hope you don’t mind answering a few questions. I’m currently trying to make activated esters of coumarins and it’s been giving me ginormous trouble. I’ve tried making the O-Su ester with DCC but the reaction is incredibly messy. DCU precipitation is incomplete, I see quite a bit of the anhydride forming, worst part is the O-Su ester is very sensitive and decomposes on silica, so I can’t even monitor the reaction much less column it (tried once just for kicks, O-Su ester completely dies, DCU and leftover DCC just streak like no tomorrow).

    So now I’m going to try my luck with p-nitrophenyl esters. I want to minimize N-acylurea formation and according to a paper (Macromolecules, 1990, 23, 65-70) using stoichiometric DMAP and p-TsOH does just that. A few things I’d like to ask:
    1. Is stoichiometric DMAP and p-TsOH really necessary? I want to maximize yield obviously but don’t want to complicate workup/purification unnecessarily.
    2. How stable are p-NP esters? Will they survive a 1M HCl/bicarb wash without significantly hydrolyzing?
    3. Any tips for removing as much DCU and DCC as possible? My product has to be fairly clean since I need to quantify it for the next reaction which requires very precise stoichiometry.

    Sorry for the long post, and thanks in advance.

    Comment by Student — February 18, 2010 @ 10:11 pm

  4. I would not use DMAP (with TsOH). DCC activation works well in acetonitrile or benzene (don’t use DCM, DMF or chloroform – the urea is too soluble in it). The right procedure is to pre-mix the acid with the p-nitrophenol in MeCN and throw in 1 equiv of DCC as a solid, stir overnight, filter, wash the solids with MeCN (or EtOAc or benzene), evaporate the filtrates, re-crystallize the residue from something like benzene-cyclohexane mix. There is a similar procedure for making Mosher acid pentafluorophenyl ester: Pfp esters are more reactive than p-nitrophenyl esters, and they are reasonably stable on silica, do not produce colorful sideproducts and have lovely 1H and 19F-NMR spectra . So give a try to pentafluorophenol as well.

    Comment by milkshake — February 18, 2010 @ 10:39 pm

  5. Hey thanks for the very prompt response milkshake, much appreciated. I should’ve mentioned earlier but the problem is the coumarin has very limited solubility in pretty much anything except DCM, so I’m kinda stuck with using DCM at the moment…

    I just thought of something though, wouldn’t it be simpler and more efficient to just use the chloroformate instead of the phenol to make the ester? There’s a JOC paper way back in 1985 (p 560-565) that describes just that. Apparently it doesn’t work too well with hindered acids but I’m thinking benzoic acid kinda looks like half of the coumarin-3-carboxylic acid so I want to give that a shot, workup and purification should be much easier if this works.

    Oh right, p-nitrophenyl esters are silica stable as well I assume? Also it won’t kill me to have traces of moisture in my reaction?

    Comment by Student — February 18, 2010 @ 10:51 pm

    • I would not do chloroformate activation – I think you would be better off with making acyl chloride in that case.

      If you have to run your DCC coupling in DCM then go ahead – but the next day filter, evaporate, re-dissolve the residue in EtOAc (or MeCN) and filter again, to get the remaining urea out.

      Comment by milkshake — February 18, 2010 @ 10:59 pm

  6. Or a good trick is to run the reaction as a suspension in MeCN; chances are your reactants will be 1% soluble in it, and the product as well, thereby driving the equilibrium towards products formation. It’s always amazing to start a reaction with nothing soluble at all, and 1 hour later, get that magic golden solution! 😀

    Comment by HPCC — February 19, 2010 @ 12:44 pm

  7. happy birthday, milkshake 🙂

    Comment by psi*psi — February 21, 2010 @ 2:54 pm

    • Thank you Jes, really. I wish I was somewhere completely else. If there is time – and if some mundane things like a job search do not interfere – maybe I will go to see Rhenium in Flagstaff this May because I have bunch of friends in Tucson and I miss them (and Arizona desert too) and the two places are within a driving distance, so I could combine the visit.

      Comment by milkshake — February 21, 2010 @ 5:12 pm

      • mmm. I bet May is a nice time for you, too, to get out of the humidity 🙂

        Comment by psi*psi — February 22, 2010 @ 1:37 am

      • Yay! The offer is still very much there, I have just been swamped of late. There is a potential NSF inorganic workshop in mid-May in Santa Fe and then I’m off home for a month, but early May will be a perfect time.

        Comment by Rhenium — February 25, 2010 @ 2:22 pm

      • thank you. I should know soon what my job situation will be like, lets hope that I will end up swamped too

        Comment by milkshake — February 25, 2010 @ 3:00 pm

  8. Milkshake,
    This reminded me of the synthesis of ropinerole stage 2. I can no longer get at the data on this and the Publications are behind paywalls. Ropinerole stage 1 is the synthesis of the nitrostyrene, stage 2 the Ferric chloride cyclisation, the quality of the ferric chloride is critical for yield and reaction time, the latter varying from ~5 hrs to oner 24 hrs

    Comment by andrew — February 26, 2010 @ 8:09 am

    • it is actually from a ropinirole process paper. Eventually they run into bromide elimination problem with dipropylamine as a nucleophile (n-Pr2NH is a somewhat hindered amine) so they re-did the synthesis with benzoyloxy and tosylated in the end, and turns out tosyl does not eliminate nearly as much as bromo. But making the tosyl analog is steppier – first they open the ring of isochromane with PhCOCl and ZnCl2, then they do Sommelet oxidation on the obtained benzyl chloride, etc.

      Comment by milkshake — February 26, 2010 @ 9:20 am

    • I forgot to mention: after I did the FeCl3-AcCl cyclization and got my product in not-so-great yield, I was curious where the remaining mass balance went so I columned the supernatants and isolated the main sideproduct. I did not manage to figure out what was it but it had the alpha chlorophenacetic portion, bromoethyl and AcONH in the molecule. And it was oily – definitely not an oxindole. It looked like non-cyclized product on NMR

      Comment by milkshake — February 26, 2010 @ 9:25 am

  9. hey milkshake,

    i am a ghost reader for while for your blog. I m deeply sorry for what you are going through. But remember one thing ‘ In this great jungle, mighty lion sleeps happy every night. Great lion is not. So, i strongly suggest that you should try to be might lion not great lion for a while’. Also, Everything happens for our good. Peace.

    Comment by parit — March 6, 2010 @ 2:19 pm

    • Thank you for the well wishes but I would rather be an elephant – one that comes to a village at night to soak from their barrels of rice beer

      Comment by milkshake — March 6, 2010 @ 7:24 pm

  10. Hey Milkshake,
    Have been glad to hear from you more often of late. Happy belated b’day. That is some funky chemistry you posted with a interesting rearrangement. Some mechanisms on this would be interesting. Also, in reference to that student post about DCC, surely using a different activating reagent, even EDC would be the best option. Although, of late biphasic mixtures or suspended reactions have taken my interest as well. Sometimes even water sensitive reactions being done in biphasic reactions (H2O/DCM or Toluene) have proven successful.

    Comment by OrganicOverdose — March 10, 2010 @ 12:28 am

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Blog at

%d bloggers like this: