Org Prep Daily

December 7, 2010

Diastereoselective cinnamate reduction, Oppolzer auxiliary

Filed under: procedures — milkshake @ 2:48 pm

6-Methoxy-2H-chromene-3-carboxylic acid 12.56g (60.9 mmol) suspension in anhydrous dichloromethane 100mL was combined with 7.0 mL of neat oxalyl chloride, followed by 4 drops of DMF. The mixture was stirred under gas outlet tube filled with Drierite for 1 day; by this time the gas evolution ceased. The homogenous reaction mixture was evaporated to dryness, the residue was briefly dried on highvac, the resulting solid was crushed with a spatula and re-dried on highvac for 1 day. Y=13.66g (100%) of a yellow solid. [This acyl chloride readily decomposes on storage – it is best kept under high vacuum while protected from a direct sunlight, and used on the same day.]

Oppolzer (+)sultam auxiliary 5.785g (26.87 mmol) solid in a 0.5L flask was flushed with dry Ar, 60% NaH in mineral oil 1.505g (37.6 mmol) was added followed by anhydrous toluene 250mL (gas evolution). The slurry was briefly sonicated for 5 min on a sonicator bath, then stirred at ambient temperature under Ar for 2h45 min. The mix was then cooled to 0 C, a solid acyl chloride 5.983g (26.87 mmol) was added in one portion, the mixture was placed placed on ambient water bath and stirred vigorously for 2 hours under Ar. (There was a delayed gas evolution accompanied by foaming). At the completion of the acylation, the reaction was quenched by addition of silica (50g) followed by hexanes (100mL). After additional 10 minutes, the entire reaction mix was applied onto a column of silica (500g) in hexanes-ethyl acetate 10:1, then rapidly eluted with the 10:1 mix and then with hexanes-ethyl acetate 7:3 mix (3L). (There is a risk of the product crystallizing on the column if the elution is too slow.) A yellow band was collected.  Combined fractions provided upon evaporation and drying on highvac 9.965g (92% th) of the chromene-acylated auxiliary as a yellow fluorescent solid. 1H(CDCl3, 400MHz): 7.31(br s, 1H), 6.80(m, 2H), 6.73(d, 2.4Hz, 1H), 5.03(dd, 13.6Hz, 1.2Hz, 1H), 4.81(dd, 14.0Hz, 1.2Hz, 1H), 4.12(dd, 7.6Hz, 4.8Hz, 1H), 3.77(s, 3H), 3.54(d, 13.6Hz, 1H), 3.43(d, 13.6Hz, 1H), 2.05(m, 1H), 1.93(m, 4H), 1.43(m, 2H), 1.30(s, 3H), 1.02(s, 3H) [Note 1]

The intermediate from the previous step 2.020g (5.00 mmol) in a 300mL RB flask was dissolved in anh THF 50mL under Ar and the solution was cooled to -50 C. L-Selectride 1M solution in THF 6.5 mL was added dropwise with vigorous stirring over 5 min period and the reaction was then maintained at -50C for additional 45 min. The reaction was quenched at -55 C by dropwise addition of 2M sulfuric acid 25 mL, the cooling bath was removed and the reaction mixture was stirred at ambient temperature in an open flask for 2 hours. The reaction mix was then partitioned between ether 120 mL and water 80 mL. The organic phase was separated, washed with water 100mL and saturated sodium bicarbonate 100mL. The aqueous phases were re-extracted with ether 130 mL. The combined organic extracts were dried (MgSO4) and evaporated. The evaporation residue was kept under Ar [Note 2] until it could be purified on a column of silica (120g) in ethyl acetate gradient in hexanes (0 to 30% EtOAc). The obtained column-purified material (1.57g; 98:2 dr by 1H-NMR) was suspended in cyclohexane 60mL, the slurry was refluxed for 10 min and then allowed to sit at ambient temperature overnight. The solid product was collected by flitration, washed with hexanes, dried by suction and on highvac. Y=1.410g (69.5% th) of a diastereomerically pure material. 1H(CDCl3, 400MHz): 6.77(d, 8.9Hz, 1H), 6.69(dd, 8.9Hz, 3.0Hz, 1H), 6.59(d, 2.9Hz, 1H), 4.44(ddd, 10.7Hz, 3.3Hz, 2.0Hz, 1H), 4.04(t, 10.3Hz, 1H), 3.92(t, 6.3Hz, 1H), 3.74(s, 3H), 3.58(m, 1H), 3.54(d, 13.9Hz, 1H), 3.47(d, 13.9Hz, 1H), 3.03(m, 2H), 2.08(m, 2H), 1.90(m, 2H), 1.41(m, 2H0, 1.20(s, 3H), 0.99(s, 3H) [Note 2]

This reduced chromane-auxiliary intermediate 1.410g (3.477 mmol) was dissolved in THF 140 mL and the solution was cooled to 0 C. Water 36mL and 50% H2O2 15mL was added, followed by 1M aqueous LiOH 5.0mL (prepared freshly from Aldrich LiOH monohydrate). The reaction mixture was stirred at 0 C for for 20 min, then quenched with 2M H2SO4 1.5mL and warmed to ambient temperature. The reaction mix was partitioned between ether 300 mL and water 150 mL. The organic phase was washed with additional water 200mL, then shaken for 5 min with 1M Na2SO3 200mL (to convert the peroxyacid into a carboxylic acid) . The aqueous phases were sequentially re-extracted with additional ether 300mL.
The combined organic phases containing a mixture of the product and the liberated auxiliary were extracted twice with 3:1 mix of water with conc. aqueous ammonia (2x100mL) and then with water. These combined ammonia extracts were then acidified with 6M HCl (140mL) with cooling on ice bath, the acidified mixture was then extracted 3-times with dichloromethane (3x150mL). The dichloromethane extracts were washed with water (100mL), combined, dried (MgSO4) and evaporated. The residue was re-crystallized from cyclohexane 60mL at reflux (then kept at ambient temperature overnight), the precipitated product was collected by filtration, washed with hexanes (2x10mL) and dried on highvac. Y=682mg of white cotton-like fluffy needles (94% th, (S)-enantiomer, >99% ee) [note 3] 1H(CDCl3, 400MHz): 6.77(d, 8.8Hz, 1H), 6.69(dd, 8.8Hz, 2.8Hz, 1H), 6.63(d, 2.8Hz, 1H), 4.39(m, 1H), 4.16(m, 1H), 3.75(s, 3H), 3.06(m, 3H); 13C(CDCl3, 100MHz): 177.9, 153.7, 148.0, 120.5, 117.4, 114.0, 113.8, 66.2, 55.7, 38.4, 27.3; [alpha]D27= +2.04(c=0.981) Chiral HPLC assay: Chiralpak  AD-RH, 17 to 20% MeCN in water with 0.1% TFA, at 75C @ 0.8 mL/min

Note 1: Using the same procedure, Oppolzer (-) sultam auxiliary 6.238g (29 mmol) with 60% NaH 1.625g (40.6 mmol) and acylchloride 6.967g (31 mmol) provided 11.06g of the opposite enantiomer (94.5% th)

Note 2: The L-Selectride reduction procedure can be difficult to manage on a large scale. Air oxidation of borane species that carried over into the crude product inspite the workup seems to be responsible for variable yields (40-50%) on a larger scale. [A careful workup with perborate would probably solve this problem.]

Note 3:  Using the above procedure, L-Selectride reduction of the acylated intermediate prepared from the (-) auxiliary provided 1.210g (59.5%th) of the reduced intermediate, which was then hydrolyzed as above in 93% yield to provide 580mg of pure (R)-enantiomer (>99% ee)

Note 4: A more direct approach to the optically pure chromane acid, by asymmetric hydrogenation with a Ru catalyst, is described here


  1. Hello Milkshake,
    Did you think of trying an asymmetric hydrogenation on the free acid? This looks like an ideal substrate for it, and is much quicker and more scalable than using a chiral auxiliary.

    Comment by Nick K — December 11, 2010 @ 4:14 pm

  2. Whoops, sorry about my last post. I failed to read to the end of the article.

    Comment by Nick K — December 11, 2010 @ 4:16 pm

    • this auxiliary method is rather laborious but it was useful as a fail-safe approach to make research quantities of the chromane acid – and as the sense of asymmetric induction is so well precedented in the literature that it also provided absolute stereochemistry assignment of the active enantiomer.
      But asymmetric hydrogenation of the free acid was much better way and we used it for scale up

      Comment by milkshake — December 13, 2010 @ 10:12 am

  3. Off topic; Any luck finding a new job?

    Comment by Chemist07 — December 21, 2010 @ 9:35 pm

    • I have no decent job offer yet: There are companies hiring at my level (senior associate MS) for medicinal chemistry and process research groups. I got asked to provide a research summary few times. I hope to hear back from at least some of the companies when the holiday season is over.

      I was visiting a government medicinal chemistry lab here in Maryland just yesterday and they really wanted to have me. But the proposed arrangement was problematic (for reasons mostly related to the extremely limited research budget) and so I turned down that offer already. I am going back to my house; I think it is better to be without a job and concentrate on job hunting rather than to accept a temp position at a place where I could hardly do any productive work.
      Please let me know if you hear about anything suitable.

      Comment by milkshake — December 21, 2010 @ 11:55 pm

  4. I’m just curious – do you know of any good references for basic organic synthetic procedures (i.e. NaBH4 reduction of ketones, Fischer esterification, etc.)? I’ve used Vogel’s “Practical Organic Chemistry” in the past but the reactions are carried out on quite a large scale, so it is sometimes hard to reduce everything by a factor of 10 or 100. Thanks!

    Comment by Ron — January 4, 2011 @ 10:14 pm

    • I don’t know – there are Orgsyn procedures which are free and searchable by the reaction type, they should be generally pretty good and checked for reproducibility. Also whenever I needed something general to do with functional group transformation I would look up a reference in Larock book (now somewhat dated) or in Paquette encyclopedia of reagents. There is also online database of organic reactions EROS, I think that one was culled from Houben Weyl which used to be a very solid German monograph series complete with experimental examples. There is Organic Reactions.

      Scaling down isn’t a problem at all – in fact I trust scale-up procedures published in Orgsyn or Organic Process R&D journal more, simply because procedures performed a on large scale have to be pretty robust.

      I should mention that it is worthwhile to keep some rudimentary folder system for article printouts sorted by the reaction type (that you commonly use) and stuff the selected and particularly useful examples of experimental procedures there over time.

      Comment by milkshake — January 5, 2011 @ 12:13 am

    • Modern Organic Synthesis in the Laboratory, Jie Jack Li et al. 2007, Oxford Press is a really great reference I have found. It is simply a book of standard experimental procedures. Many of them are also large scale examples, but they are always a great starting point.

      Comment by PotStirrer — January 10, 2011 @ 7:19 am

  5. Heres a question- I have a substituted proline tert butyl ester. What I would like to is acylate that alpha position. Initially I tried forming the ester enolate and reacting with a load of acylating agents, such as acyl amides, even things i wouldnt normally try like acid halides and aldehyde followed by oxidation. I have seriously poor yields with mostly starting material recovered. I tried quenching the enolate with CD3CO2D and the NMR shows complete exchange of the proton. The next strategy tried was to try to get the tert butyl ester turned into a weinreb amide which i could hit with a grignard. AlMe3 and MeONHMe.HCl doesnt work, neither does i-PrMgCl and the hydroxylamine. I cant cleave the ester to the acid with TFA or TMSOTf as I have an acid labile protecting group elsewhere on the molecule.
    Any thoughts on how to either get the ester to the acid, acylate the alpha position or turn the ester to a Weinreb amide? I’m stumped….

    Comment by artemisia — February 1, 2011 @ 5:16 pm

  6. If you use LiOH.H2O solution overnight at room temperature you should eventually hydrolyze the tBu ester. (I have even used LiOH solution for cleaving lactams under mild conditions: optically active 4-hydroxy-2-pyrrolidone, 45 C)

    About the acylation: When you do lithiation of a hindered substrate and then react your anion with an electrophile, you often get lots of unconverted starting material back. My understanding is that the role of present diisopropylamine generated from LDA is far from innocuous – it competes with your anion and the side-reaction generates H(+) that quenches your anion.
    The way to minimize this problem is to use Li-TMP (Li-2,2,6,6 tetramethylpiperidine) as a base. (Unlike LDA this reagent has to be generated at 0C and warmed to RT before use at -78C).
    The nature of electrophile also plays important role. From what I heard acyl fluorides are pretty good for acylating carbanions prepared by lithiation but I have not done it myself.
    (The best way to make acyl fluorides from carboxylic acid is TFFH reagent 1 eq, iPr2NEt 2 equivs, DCM RT 20 min, quick workup with a chilled diluted HCl, extraction with ether and wash with brine. It is better to use commercial 4A powdered activated molecular sieves for drying the etheral extract of acyl fluoride). The produced acyl fluoride will contain some tetramethyl urea but that does not interfere. There are other reagents for making acyl fluorides (cyanuric fluoride, Deoxyfluor etc) but I only used TFFH for this purpose.

    Comment by milkshake — February 1, 2011 @ 9:44 pm

  7. Hi Milkshake, any progress in the job search? I really hope you find something decent.

    Reagent question: any experience with cobalt carbonyl, Co2(CO)8? Some sources list it as pyrophoric, others do not. Can it be handled w/o glovebox or Schlenkware?


    Comment by MrOrange — February 14, 2011 @ 1:05 pm

    • I don’t have any experience with metal carbonyls. Dicobalt octacarbonyl is solid so it is less hazardous to handle than others, but still you need to be very careful not to breath the vapors.

      One technique I often use for air sensitive solids is to weight them out into a test tube closed with a septa. (Load it in the hood. The test tube can be first flushed with Ar). You then dump the test tube content into Ar-flushed flask.

      My job situation: I am still looking

      Comment by milkshake — February 14, 2011 @ 4:59 pm

  8. thanks for replying regarding my proline t-Bu ester problems. Unfortunately LiOH saponification of the ester didn’t work, I’m currently trying LiOH/ H2O2 as an alternative. LDA wasn’t a particularly great base for forming the enolate, so I was using s-BuLi and have been trialling LiTMP this week. I am making TFFH tomorrow using your procedure, my supervisor says that the iminium intermediate is quite lively and to be careful when filtering it. he wondered if it was soluble in the acetonitrile you use as a solvent in the halogen/ counterion exchange step? cheers

    Comment by artemisia — February 17, 2011 @ 10:25 am

    • the chloro formamidinium intermediate Me2N-CCl=NMe2(+) Cl(-) is actually quite well-behaved, and it is soluble in MeCN. (KF and KPF6 have only a limited solubility in MeCN but enough for this to work). The chloro formamidinium is a heavy granular solid that settles nicely in toluene. It is very hygroscopic but you can rinse it with toluene under nitrogen blanket with ease, by adding a frited adaptor to one of the flask necks. To form the chloroformamidinium, one needs to add the entire amount of oxaloyl chloride to tetramethyl urea in toluene at room temperature, the mix becomes biphasic as the adduct of the two oils out. Then you gradually heat the flask and the initial adduct decomposes with gas evolution (CO+CO2)and the chloro intermediate precipitates as a coarse salt-like solid. You may want to do the gradual heating on a large oil bath – to ensure that the gas evolution is not too sudden.

      Since the chloro intermediate is very moisture sensitive, you keep it in the flask where it was made – you just add pre-dried KF and KPF6 and acetonitrile to it in the same flask. I found that Burdick and Jackson solvents in 4L bottles were dry enough for the purpose when used from a freshly opened bottle but pre-drying KF and KPF6 was necessary for a good yield.
      And when you are done, check the product purity by 1H and 19F NMR of your TFFH in d6-DMSO and do a melting point also. The product should be stored in the fridge.

      PS: with LiOH+H2O2 you want to wash the organic extract of your acid with NaHSO3 to reduce the peroxyacid that you get from peroxolysis. See in the procedure for cleaving off the Oppolzer auxiliary above.

      Comment by milkshake — February 17, 2011 @ 11:39 am

  9. Should change the name of your blog to Org Prep Sometimes 🙂

    Comment by Industry Guy — February 25, 2011 @ 10:46 am

    • yeah, this has been pointed out to me couple times but I am not changing the name. Also, since I currently don’t have a job it is difficult for me to come up with new procedures. (I can go to old notebooks, and find something crappy or mundane – but thats not the point).

      I started Org Prep Daily in 2006 and at that time I was posting one or two procedures a day, from accumulated pile of synthetic procedures that I did with my hands and enjoyed. I knew that I was going to run out of new stuff to post daily at this pace within few months but the purpose was to build a readership and above all, to encourage others to join in and post things here – like in a private version of OrgSyn. The idea was that medchem and methodology research projects generate interesting procedures that do not have IP value (so they don’t need to be kept secret) and on their own these bits of useful information are not publishable in a journal. All too often they are buried in old notebooks and forgotten.

      Since you would probably like to give a copy of your good procedure to a guy working in another lab – or to someone whom you just met at a conference, maybe you would also like to upload your procedure here. Now that self-publishing on the web costs nothing and requires no coding skills, perhaps some technical people would want to share examples of their synthetic expertise occasionally, or discuss some weird stuff that surprised them in their research. I invited a number of good chemists that I knew and gave them admin access; some of them wrote a post or two. Unfortunately this idea did not catch on – maybe most people are less enthusiastic and more busy than I thought, or they are afraid to ask their boss for permission… So, Org Prep Daily still remains a personal chemistry page of a guy who is currently looking for a job.

      Comment by milkshake — February 25, 2011 @ 2:43 pm

  10. I’m here to corroborate Milkshake’s statement. There used to be times in synthetic worlds when one wouldn’t spend a quarter of one’s life secretly worrying about job security. When passion, dedication and enthusiasm in R&D would be what would matter most. And when we’d be paid very damn well for our life passion too! But, alas, times have changed. Good scientists are becoming the victims of the Outsourcing school of thought – that takes careers away from happy and merry individuals in North America, Britain, and many other places else. This, coming from an individual who’s been unemployed from consequences of a Big Pharma closure since last November, and has spent his life the following months attending seminars in nearby universities, going to workshops, attending various ACS Webinars. That was all science I’ve done in the past 4 months.

    The purpose of this diatribe is not only to expose the reality of synthetic careers in 2011, but also to encourage any writer, student, worked, whatever, to submit procedures to this most glorious blog of them all. Believe me, if my procedures weren’t all in electronic format on my former company’s server, well, quite a few of them would be here right now. So please, guys and dolls (to quote Sinatra), act up. Before you regret! After all, isn’t that what science is all about? The open diffusion of ideas to advance the art of whatever we’re studying, in this case, organic synthesis?

    Yours truly,

    Comment by HPCC — February 25, 2011 @ 5:07 pm

  11. Sorry for a unrelated question – I am trying to use Pd/C to reduce two allyl groups substituted on the same nitrogen with hydrogen balloon. The reaction is not very clean as there is competitive de-allylation. Most likely due to steric strain. I am wondering whether there is a way to aviod this side reaction.

    Comment by artemisinin — March 1, 2011 @ 12:24 pm

    • You need to use platinum, preferably 5% Pt on carbon. The hydrogenation will be much faster and you should not have the deallylation problem. Try ethyl acetate or ethanol as a solvent.

      (Pd-C cleaves off allyls with the same mechanism as debenzylation)

      Comment by milkshake — March 1, 2011 @ 1:18 pm

      • Hi Milkshake, thanks for the quick response as usual.

        Can you tell me what is the difference between Pd/C and Pt/C in this case – saturation double bond vs de-allylation?

        Ps, I mentioned steric stain as the de-allylation only cleaves one chain.

        Comment by artemisinin — March 1, 2011 @ 1:31 pm

        • Platinum does not cleave benzyls from a heteroatom and does not transfer allyls. Pt is also much faster than Pd in C=C hydrogention. The risk of aromatic ring saturation is not too great with Pt under H2 balloon. (I have done hydrogenation of 1-allyl-4-nitropyrazole to 1-propyl-4-aminopyrazole over Pt-C and it was quite clean).

          Pd(0) is rather good at transfering unsusbtituted allyls from tertiary amines, in their protonated form. With bis-allyl amines the first allyl goes off much faster. For complete deallylation of amines, the best system uses N,N’-dimethylbarbituric acid as scavenger of allyls, and the used catalyst is often Pd(PPh3)4. If you want to cleave just one allyl of the two, one would use thiosalicylic acid as a allyl scavenger instead. (There is a number of suitable scavengers for deallylations, including formate salts, borohydride and triethylsilane. The problem with using hydrogen-producing scavengers like these for Pd(0) deallylation is that they often effect C=C saturation of allyl into propyl as well.)

          Steric and electronic effects: for example para-methoxybenzyl is quite difficult to remove by hydrogenation on Pd, also substituents on allyl often hindesr the Pd-catalyzed allyl transfer reactions. The nature of the heteroatom plays a role: the fastest debenzylation and allyl transfer proceeds with heteroatoms that are good leaving group (carbamates, carbonates, acetate esters, halides). There are effects of the substrate coordination to the catalyst – amines are known to deactivate catalysts like Pd, Pt. Often the hydrogenation of amines and amine-producing substrates (like Cbz-protected amines) are done in presence of sulfuric acid or HCl, or in straight acetic acid (or even in TFA) to make the hydrogenation go faster. On the other hand, by adding extra amine, like NEt3, amine-treated Pd-C can still hydrogenate C=C and cleave halogens off but debenzylation is inhibited.

          Comment by milkshake — March 1, 2011 @ 2:07 pm

  12. Hi Guys,
    i am trying to alkylate an aromatic methyl group (LDA 2.2 eq, -78 °C, THF) on an almost fully substituted benzene ring (resorcinol like), with a weinreb amide 1 eq., and 1.6 eq of aromatic ring.
    The main problem is that the anion at the benzylic position is forming very well, you can follow it as the solution turns deep red in few seconds, then after the addition of the cold solution of the weinreb i have the complete disappearance of the electrophile, and only a disappointing yield of the product, except few times 65% yield. As the weinreb amide is also a michael acceptor i would like to know if is possible to avoid michael addition of the aromatic ring instead of the desired 1,2 addition.Low temperature seems not to help.

    Comment by madforit — March 4, 2011 @ 7:52 am

    • You are able to get a decent yield sometime so there must be an important variable in your reaction conditions that you do not control fully. My guess is that the variable factor is either in the lithiation step or in the way that you combine your Weinreb with the organolithium.

      You mentioned you have conjugated C=C in your Weinreb. This means, your Weinreb amide is quite base-sensitive and has a tendency to crap up in the presence of LDA excess. (Maybe you have acidic proton in gamma position to Weinreb amide.). I think it is wrong to use excess of LDA relative to the aromatic substrate – you don’t want to have free LDA around when you add the Weinreb amide.
      Also, following the the formation/dissapearance of the Li anion by development of color it is not reliable – what if you just formed only 50% yield of your anion and the rest is there still as a free LDA?

      I would suggest that you try to use Li-tetramethylpiperidine inplace of LDA. (Sometimes diisopropylamine from LDA is not that inert and it does bad things). I suggest you use only 1.5 eq. of Li-TMP and 1.6 eq. of your aromatic piece. You need to form the Li-TMP reagent from BuLi and TMP in THF at 0C, then warm it up to RT before use. You probably want to perform the methyl group lithiation at high enough temperature and give this enough time to to make sure that you have reached an equilibrium in your lithiation.

      I would also try to add the solution of the anion by using a canula to the solution of Weinreb at -78C and then warm it up. (Chances are, if your Weinreb is crapping up, unpredictably, that the order of addition makes a difference).

      Also, if you are working on a total synthesis project where the Weinreb piece is labor-intensive, maybe you can practice the reaction conditions on a model system, preferably one having C=C like your material.

      Comment by milkshake — March 4, 2011 @ 11:14 am

  13. Thank you so much Milkshake,
    i totally agree with the variable of the lithium anion formation, seems that is the most difficult step to control.
    I have found quite interesting also, that if you work with a little excess of aromatic core you can get more desired product, since is my first two months in this group, the best results was my 65%.
    I have just tried LiTMP, right now. Tlc shows only trace of the desired product, and also, even if i agree with you that you definitely cannot control a lithiation step with some color formation, this time the red color was orange-yellow.
    I will try to invert the addition, and to perform the lithiation at 0°C, and also i will warm the LiTMP to rt.
    Thank you so much!!

    Comment by madforit — March 4, 2011 @ 11:33 am

    • OK, just don’t use LDA excess. I think you are having enolisation problem with your Weinreb amide, with C=C migration. I wonder if transmetallating the lithium to organomagnesium (by adding anhydrous MgX2 to the anion solution) would help to make the system less basic.

      Comment by milkshake — March 4, 2011 @ 11:45 am

  14. This should be a good idea, i have been told that in this kind of aromatic “addition”, anhydrous LiCl added to the weinreb solution help to boost the yield.i Still wondering why…
    thanks again!

    Comment by madforit — March 7, 2011 @ 7:37 am

    • Li(+) acts as a mild Lewis acid and by coordination activates carbonyl towards addition of nucleophiles. Also with LiCl in aprotic media one gets halogen-bridged binuclear complexes RLi…Cl(-)…Li(+)…COR2 type, this helps to assemble the reacting partners. Finally, the Weinreb amide adduct with organolithium RR’C(OLi)NMeOMe persists as a tetrahedral intermediate and does not collapse into a ketone before quenching the reaction, so you won’t get double addition. And this happens because of chelation to Li(+). (So perhaps adding more Li(+) helps to make this more stable. But I think the first and second explanation is more likely.)

      Comment by milkshake — March 7, 2011 @ 8:40 am

  15. Completely off topic, but I’d appreciate your help.

    So I ran a column on my product in a hexanes/EtOAc mix. I rotavapped it down a couple of times and put it on the highvac line for an hour. It failed to solidify and I got a gummy paste instead. Based on past experience I know these types of compounds should be nice crystalline solids, so I figured something was wrong.

    NMR tells me I have ~20% hexanes still stuck in there. I have absolutely no idea how this happened, seeing as all the EtOAc is gone. Anyway, do you have good ideas for removing that annoying residual hexanes?

    Comment by UG4 — March 11, 2011 @ 10:41 pm

    • it is not hexane but something very greasy. This kind of thing happens easily. The possibilities include: 1) stopcock grease (either leaking from your glassware joints or from your rotovap) 2) mineral oil or paraffin wax (from reagents like NaH that come coated with mineral oil) 3) bad-grade solvents that contain nonvolatile residue (you can check it by evaporating blank solvents) 4) contaminated silica that soaked up plasticizers from vinyl plastics (those are typically a long-chain phthalates so you should see some extra aromatic protons and OCH2 methylenes too)

      Comment by milkshake — March 12, 2011 @ 12:12 am

  16. I just double checked the NMR, and oh my god you’re absolutely correct. It’s not hexanes but grease, f###king grease!

    I’m finishing off a scope expansion and this is THE most important example. Thanks to the person before me I had just enough substrate for one reaction. Now this’ll set me back for another week. FML

    Comment by UG4 — March 12, 2011 @ 12:57 am

    • if it is grease and your product is crystalline, maybe your product is not very soluble in hexanes. I wonder if you can suspend your stuff in hexane with sonication, and collect the precipitated solids by filtration. You will lose some but the drop in yield should not be so bad.. If this fails you can always re-purify on the column, and give it a good first elution with straight hexanes (or maybe with hexane containing few % of ether) so as to flush the grease out, before you apply the hex-EtOAc mix to elute the product. Sometimes a strong band of grease can pull the product with it to the front and the way to avoid the problem is to start easy

      Comment by milkshake — March 12, 2011 @ 2:46 am

      • Thanks a lot milkshake. I’ll go ahead and tell you my product is an indene derivative, which means it’s massively greasy and likes to dissolve in hexanes for the most part. But for sure I’ll give your ideas a try, and if worst comes to worst, could always make more substrate.

        Comment by UG4 — March 12, 2011 @ 2:06 pm

  17. Dear MilkShake

    I am doing a free radical side chain bromination of toluene derivative. Are there instantaneous terminators available to terminate the reaction.
    Beacuse starting material seems to be more in the bulk of the reaction mass after complete workup than in the aliquot (aliquot work up is similar to bulk mass) which is analyzed by LC prior to complete work up. Guess there could be some disproportionation between benzylic bromide derivative and the corresponding dibromo impurity. If we have to fix this, i need to terminate the radicals before doing the work up. Your thoughts on this issue is appreciated.
    take care

    Comment by pash — March 15, 2011 @ 5:50 am

  18. I don’t believe that there is a disproportionation of benzyl bromide going on.

    I suppose you are using some sort of aqueous reducing agent at the end, to consume the excess of unreacted bromine. I wonder if the debromination is happening during the workup. (This would make sense – when you are quenching a small sample, the contact with the reducing agent will be much shorter than when you are working up the entire reaction mix.)
    If it turns out that the reducing agent used in the workup is the source of your problem, maybe you could modify your workup and use something else. Also, I noticed that people occasionally make the mistake of using dithionite (also known as “hydrosulfite”) Na2S2O4 in place of NaHSO3 (“bisulfite”). Dithionite is unsuitable for workup because it is an exceedingly strong reducing agent.

    Can you just concentrate your reaction mixture under reduced pressure on a Teflon pump, to pull off Br2 and HBr? (Maybe with some chlorbenzene added as a chaser)

    Comment by milkshake — March 15, 2011 @ 8:01 am

  19. Hello Milkshake,

    Any chance of posting a new prep? I’m getting withdrawal symptoms! On another note, what’s your opinion about the future of Med Chem? I ask because I’m seriously considering doing something else.

    Comment by Nick K — March 22, 2011 @ 1:16 pm

    • maybe you should change the field while you still can. It depends on your personal situation too – how important is for you to work in a synthetic lab, how long you can afford to be unemployed, and how well you can adjust working for an employer who does not appreciate what you do

      Comment by milkshake — March 22, 2011 @ 4:11 pm

  20. L-Selectride is not the same as LiEt3BH.

    Comment by Eman — March 26, 2011 @ 1:53 pm

    • You are right – LiBHEt3 is called Superhydride. I used L-Selectride. Sorry for the error, I corrected the scheme and the text. Thank you for noticing this.

      Comment by milkshake — March 26, 2011 @ 5:02 pm

  21. Hi Milkshake,
    Not a chemical query, and sorry for breaking away from the chemistry stream. I am wondering if you or any of your readers have any experience with aligning the UV lamp inside a beckman HPLC? I have bought a UV lamp on e-bay and fixed it inside the heat sink of the detector, however, the baseline is very noisy (over 100 times the usual). If i make this work, i can save about 1200$ dollars. Any tips, suggestions?
    Thanks, sks

    Comment by sks — April 7, 2011 @ 3:50 pm

    • I am not an expert, my guess is that not enough light gets into detector. Are you sure they sold you a new lamp? I could imagine someone saving and re-packing (by accident – or on purpose) a degraded lamp, then selling it to you on e-bay… It is hard to tell by its appearance. There could be misalignment problem or something wrong with the detector or dirty flow cell but I think a bad lamp is more likely than the rest.

      I have no experience with Beckman. With Agilent HPLC a UV lamp at the end of it lifetime will have diminished luminosity especially at the shorter end of spectra, the weak detector reading shows up as high background noise and occasional step-like baseline jump. Agilent software has lamp troubleshooting routine that allows you to check the lamp health – maybe Beckman has some similar diagnostics tool as well.

      Comment by milkshake — April 8, 2011 @ 2:13 am

  22. Thanks Milkshake! Couldn’t find the diagnostic tools for lamps on the BC, however, the baseline is getting better with the adjustmentss on the position of the lamp.
    cheers, sks

    Comment by sks — April 11, 2011 @ 3:26 pm

  23. Hi Milkshake, i am trying to prepare a series of compounds (esters and amide) of a substituted benzoic acid. Most of the amines are very unreactive but the reaction with the acid chloride seems to behave quite well, at least from TLC analysis (no starting acid in the mixture).
    After standard quench (NaHCO3) and column i can only isolate little quantity of a spot that is a bad mixture of compounds, nothing else..
    Do you know any method to react acid with poorly reactive nitrogen amines (2-aminobenzothiazole, aminopyrimidine etc)?also amide coupling agents didn’t work..

    Comment by madforit — May 7, 2011 @ 7:20 am

    • First make sure that you actually formed the acyl chloride. I don’t know what kind of substitution your benzoic acid carries, maybe you have a problem right there. Take proton NMR of your acyl chloride and compare it with the starting acid (you should see downfield shift of Ar-H protons that are close to the carbonyl).

      Next, I would use iPr2NEt as a base, and add few mol% of DMAP as a catalyst at RT and run the coupling with your aniline in CH2Cl2 or chloroform overnight.

      Comment by milkshake — May 7, 2011 @ 1:58 pm

  24. I am using Hunig’s base 2 eq. with 1.8 eq of amine, acid chloride is forming very well, i check the formation quenching few drops of the solution in dry methanol and run TLC in DCM-EtOH-Formic Acid mixture to see if there’s starting acid left.
    is kind of strange ’cause with normal aniline i got almost quantitative yield of the product, but with hetreoaromatic amines is a mess, and this is quite sad since is a med.chem project..
    Thanks Milkshake!!

    Comment by madforit — May 8, 2011 @ 6:20 am

    • please have you run LC/MS to identify what your sideproducts are? I have seen this kind of problem with very electron-deficient heteroaromatic amines once before: I actually got double N-acylation product. (I suppose the first acylation was slow and the monoacylated product was acidic enough to be deprotonated by iPr2NEt and thus it got acylated the second time.) I remember that I had to use two equivalents of acyl chloride/base and I had to treat the obtained doubly acylated product with concentrated methylamine solution to bring about deacylation to the desired monoacyl compound.

      Maybe you can try less basic conditions, with your acylchloride plus pyridine. If this does not work I would try silver cyanide (4 eq,) as a base in dimethylacetamide. You can also try HATU as activation agent with a base.

      Comment by milkshake — May 8, 2011 @ 1:00 pm

  25. Already tried HATU in DMF with Hunig’s….same story, and for pyridine bad reaction even with aniline, i’ll run LC.MS this is a good idea.
    What d’you think about i-Buchloroformate and strong base to try to deprotonate the amine?bad idea?
    p.s there’s no special substitution around the ring, protected phenol and nitro group
    Thanks a lot as always

    Comment by madforit — May 8, 2011 @ 1:50 pm

    • I would not use iBu-chloroformate, in my experience Et-chloroformate usually gives cleaner results. The base has to be N-methylmorpholine. The typical conditions are: the acid + Me-morpholine 1.1 eq. at -20 in DCM or THF, chloroformate 1 eq is added, warmed up to 0C, amine is added and stirred overnight at room temp.

      The problem with these chloroformate-derived mixed anhydrides is that they are not very stable. And they generate 1 equivalent of alcohol which may interfere with less reactive anilines. You can also use pivaloyl chloride in place of chloroformate, with NE3 as a base – it produces a less reactive mixed anyhydride but the anhydride is quite stable so you can then heat it with an excess of DMAP and the aniline.

      How expensive is your benzoic acid – is it commercial? (If it is cheap enough I would definitely try to make a symmetric anhydride with DCC in MeCN, isolate the anhydride in pure form and acylate in the presence of 4-pyrrolidinopyridine as a base with heating)

      Comment by milkshake — May 8, 2011 @ 2:30 pm

  26. This sound new to me, is possible to isolate the DCC derivative of an acid?
    My benzoic acid takes 3 steps, i would like to have more infos on this method, do you have any reference?
    many thanks

    Comment by madforit — May 8, 2011 @ 2:46 pm

    • when you take 2 equivalents of a carboxylic acid in a dry aprotic solvent and add 1 equivalent of DCC and stirr it overnight at room temperature, you get a symmetric anhydride and dicyclohexylurea precipitates out. You want to use some solvent where the urea is poorly soluble so that you can remove it by filtration because it would interfere with coupling to less reactive anilines. MeCN is quite good solvent for the purpose. When you have a nice isolated symmetric anhydride you can use DMAP or better yet 4-pyrrolidinopyridine as a base and heat the symmetric anhydride with poorly reactive anilines.

      But maybe before you go to all this trouble you could try the acylchloride with 4 equivalents of AgCN in dimethylacetamide (tetramethyl urea and DMEU also work), at 60-80C. The reaction is heterogennous throughout and needs to be protected from light so wrap the flask in aluminum foil. This is surprisingly powerful but a mild method and it proceeds under non-basic conditions.

      Comment by milkshake — May 8, 2011 @ 2:53 pm

  27. ok Milkshake, this is very interesting thank you..
    i’ll try the AgCN method tomorrow, i have to dissolve the acid chloride in DMAc, then add AgCN + aniline and heat?!

    Comment by madforit — May 8, 2011 @ 3:07 pm

    • I would make a slurry of the amine with AgCN (4 equivs) in DMAc, add acyl chloride last and gradually heat up. (You filter off silver salts at the end). This system also works in solvents like toluene but in my experience it tends to be cleaner in DMAc.

      Comment by milkshake — May 8, 2011 @ 3:54 pm

  28. Hi milkshake,
    which solvent you would use for the reaction of symmetric anhydride with poorly reactive anilines?and also DMAP should be catalytic or stechiometric?

    Comment by madforit — May 10, 2011 @ 4:48 pm

    • DMAP should be stoechiometric as a base (it is quite cheap) but in this case 4-pyrrolidinopyridine works even faster than DMAP. For solvent I would try acetonitrile if the reactants are soluble in it, and use a relatively small volume of solvent (like 3-5 mL of MeCN /g of substrate). If reactants do not dissolve in MeCN I would try dimethylacetamide but MeCN would be my first choice, since you can form the anhydride in it conveniently with help of DCC. Other less polar solvents like toluene or THF are also useful for this reaction (if the material is soluble in it) but I would not recommend dichloromethane or dichloroethane (because they react quite fast with DMAP at reflux).

      Comment by milkshake — May 10, 2011 @ 5:21 pm

  29. Thanks a lot, i’ll give it a try for sure.
    i’ve prepared the anhydride today, and the reaction seems ok even if my compound was not dissolving so much in CH3CN.
    I dont have 4-pyrrolidinopyridine i’ll go with DMAP.

    Comment by madforit — May 10, 2011 @ 5:29 pm

  30. Hi Milkshake,
    is there any particular method to quench the AgCN reaction with the acid chloride?
    Washing with bleach all the glassware-balance etc i think that should be enough?!
    Thank you so much.

    Comment by madforit — May 16, 2011 @ 10:49 am

    • So did it work?

      The typical workup was like this: remove the Ag-containing solids by filtration, evaporate the filtrates on highvac to pull off as much DMAc ac possible, then take up the residue into EtOAc and wash it with aq. bicarbonate. Column at the end.

      You can rinse the used glassware with diluted ammonia to get the silver off and it should be enough. (More difficult silver deposits can be cleaned with nitric acid.) There is some cyanide but if you work in the hood, on a small scale, and treat the waste accordingly, it should not present a problem.

      Comment by milkshake — May 16, 2011 @ 2:53 pm

  31. I still have to check, tomorrow morning i’ll run some TLC and LC-MS,
    the nucleophile is really unreactive, i don’t know how people manage to form amide bond with heteroaromatic nitrogen atom?!
    thank you for the advice, this med-chem is painful..

    Comment by madforit — May 16, 2011 @ 5:51 pm

    • there is an alternative route to difficult benzanilides – adding aryl lithiums to aryl isocyanates.

      Comment by milkshake — May 16, 2011 @ 6:54 pm

  32. Hi Milkshake,

    Just curious if you have ever worked with MTO (methyltrioxorhenium) It’s supposed to be a colorless solid but what I found on the shelf is a black powder?

    Comment by Blue — May 21, 2011 @ 5:57 pm

    • I never worked with MTO but I remember that the decomposition is a problem. It should not be black. I would recommend to buy a new bottle from Strem. You can also try to re-purify the decomposed MTO stuff by vacuum sublimation but I think the vacuum has to be really good, in low militorr range.

      Comment by milkshake — May 21, 2011 @ 7:55 pm

  33. Dear MS,

    I am trying to find an optimal solvent system for crystallizing amino nitrile from Strecker reaction. The pure amino nitriles (mixture of diasteromers) are low melting solids (mp : 55-60 °C) and I need to try using chiral acids but I am worried that both diasteromers can form solids and I wont be able to isolate only one diasteromer. Use of too strong acid can also result in retro-Strecker reaction.

    Please suggest me few chiral acids which can form high melting solid with amino nitriles.

    Please help me with the relevant literature if you have any.



    Comment by marto — July 21, 2013 @ 2:02 am

    • what do you mean by “diasteromers”? do you mean enantiomers? How many chiral centers your molecule has? As for resolving alpha aminonitriles from Strecker, it is not done too often because of the stability problem you mentioned, and because these aminonitriles are weakly basic and do not form salts easily. If you want some resolving agent, it has to be a strong acid, so (speaking within range of practical resolving agents), it has to be camphorsulphonic acid or bromocamphorsulphonic acid. Sorry I cannot be of more help. Do a literature search/

      Comment by milkshake — July 22, 2013 @ 11:18 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 )

Google photo

You are commenting using your Google 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: