Chemistry 251 Laboratory -- Spring 2002
Frontalin Project Home Page

Index

• 2001 Frontalin Synthesis Home Page
• Students working on this project
• Table of Reagents and Amounts Available
• Acetone Enolate (or equivalent) Alkylation
  • Tosylation Suggestions
  • Imine Formation
• [3,3] Sigmatropic Rearrangement
• Formation and Use of Grignard from Methallyl Chloride
• Zn-Cu Couple or Organocuprate

Frontalin is an anti-aggregation pheromone of the Douglas Fir Bark Beetle (a northwest pest), and is manufactured by a company in Canada ("Pherotech") for use against beetle attacks. (They actually called last year and were interested in any potential syntheses we come up with.) For more information on frontalin and its synthesis, I recommend the excellent website authored by Gerard Dupuis and Nicole Berland. One effective strategy for synthesis of frontalin involves treatment of 6-methyl-6-hepten-2-one with MCPBA followed by acid-catalyzed rearrangement of the resulting epoxide.

We will focus our efforts on the following strategies for synthesizing the desired 6-methyl-6-hepten-2-one.

• Acetone Enolate (or equivalent) Alkylation
• [3,3] Sigmatropic Rearrangement
• Zn-Cu Couple or Organocuprate

The table below lists the chemicals that we will have available for this project. If you need something that is not on this list, consult with the mentor for your project. Also note the "Amount/group" column. This is the total amount of material available for each group to use on the project.

Reagent	Source	Amount/group	Location	Comments
p-Toluenesulfonyl chloride	Aldrich Cat. # 24,087-7	15 g	TA room	Water sensitive. Keep sealed.
1,4-Diazabicyclo[2.2.2]octane (DABCO)	Aldrich Cat. # D2,780-2	15 g	TA room
3-methyl-3-butene-1-ol	Aldrich Cat. # 12,940-2	8 mL	TA room
Cyclohexyl amine	Aldrich Cat. # 24,064-8	20 mL	TA room	Toxic
Acetone	Aldrich Cat. #	20 mL	TA room	Use high quality acetone.
3-chloroperoxybenzoic acid (MCPBA)	Aldrich	5 g	Organic Fridge
Methyl vinyl ketone	Aldrich Cat. # 26,954-9	10 mL	Th 208 Freezer	Caution: Highly Toxic. Cancer Suspect Agent. Wear gloves and work in the hood.
3-chloro-2-methylpropene	Aldrich Cat. # 28,197-2	10 mL	TA room	Cancer Suspect Agent. Wear gloves and dispense in the hood.
3-bromo-2-methylpropene	Aldrich Cat. # 34,583-0	2 mL	TA room	Cancer Suspect Agent. Wear gloves and dispense in the hood.
Zinc (granular) 20 mesh	Baker Cat. # 4244-01	10 g	TA Room
Copper(I) Iodide	Aldrich Cat. # 20,544-0	5 g	TA Room	Light sensitive.
Copper(I) Chloride	Aldrich Cat. # 21,294-6	5 g	TA Room	Moisture sensitive. Toxic.
Magnesium turnings	Repackaged from Mallinckrodt	10 g	TA Room
Magnesium turnings (New bottle)	Fisher Cat. #M11-500	10 g	TA Room
Magnesium powder 100 mesh	MCB Cat. # CB1196	10 g	TA Room
Palladium(II) chloride	Acros Cat. # 195200050	0.2 g	TA Room

Acetone Enolate (or equivalent) Alkylation

One approach involves alkylation of an enolate anion (or equivalent). The Bartlett synthesis (see reference above) uses the enolate formed from ethylacetoacetate to react with the tosylate shown in Scheme 1, followed by hydrolysis and decarboxylation to afford the desired 6-methyl-6-hepten-2-one. But this has never has never worked well for us. (It didn't work well for Bartlett either, affording only a 15-30% yield.) To avoid having to do the hydrolysis and decarboxylation we could try to use an acetone enolate equivalent. A paper in Biorganic & Medicinal Chemistry (see reference above) described a synthesis of the needed 6-methyl-6-hepten-2-one using the anion formed by deprotonation of the cyclohexylimine of acetone (Scheme 1). We didn't have much luck last year, but I think it is worth looking at this reaction again.

Step 1: Tosylation

1. The procedure listed above talks about washing with t-butyl methyl ether, but this is probably not necessary. Try filtering the reaction mixture through a Buchner funnel, washing the solid with a small amount of dichloromethane, and then extracting the filtrate as described above.
   
   
2. The solid that is formed in the reaction is probably DABCO hydrochloride.
   
   
3. Because of limitations in the amount of reagents, you should not use more than 15 g of DABCO or TsCl in your procedure. Scale the literature procedure accordingly.
   
   
4. Here is the procedure reported by Julianna Taylor and Christy Mather, two students in Spring 1998.
   
   
   In a 250 ml round bottom flask, DABCO (15.04g, 0.13 mol), dichloromethane (60 mL), and 3-methyl-3-buten-1-ol (6.07 ml, 0.6 mol) were mixed and cooled in an ice bath. At this point p-toluenesulfonyl chloride (17.17 g, 0.9 mol) was added slowly. [Other students report adding the alcohol to the mixture containing the TsCl and DABCO. I'm not sure if one method is superior to the other.] The mixture was stirred in the ice bath for one hour at which point it was removed from the ice and allowed to stir 35 more minutes. The reaction mixture was filtered through a Buchner funnel and rinsed with ethyl acetate which began to dissolve the filtrate. [I assume they mean that it began to dissolve the filter cake. I'm not sure why they didn't rinse with dichloromethane.] In a large separatory funnel, the filtrate was added to dichloromethane (40 ml). The organic phase was washed with 2 M HCl (3 x 20 ml), 12% NaHCO₃ (120 ml), and H₂O (120 ml). The product was a brownish yellow liquid, % yield 92.9 (13.38 g); ¹H-NMR (CDCl₃ 7.80 (dd, J=8.4,1.4 Hz, 2H), 7.35 (d, J=7.7 Hz, 2H), 4.79 (s, 1H), 4.68 (s, 1H), 4.08-4.16 (m, 2H) [looks like a triplet to me], 2.46 (s, 3H), 2.35 (t, J=6.6 Hz, 2H), 1.66 (s, 3H).

Step 2: Formation of Imine -- Cyclohexylamine + Acetone

[3,3] Sigmatropic Rearrangement

Another strategy involves attack of the Grignard reagent formed from 2-methyl-3-chloropropene on methyl vinyl ketone producing an alcohol that undergoes a palladium catalyzed [3,3] sigmatropic rearrangement to afford the needed 6-methyl-6-hepten-2-one (Scheme 2).

1. Brian and Lindsey found a detailed paper on the formation of the Grignard reagent: Baker, K.V.; Brown, J.M.; Hughes, N.; Skarnulis, A.J.; Sexton, A. Mechanical Activation of Magnesium Turnings for the Preparation of Reactive Grignard Reagents. J. Org. Chem. 1991, 56, 698-703.



2. 3/28/02 -- Nikki and Dan tried using Mg powder to make the Grignard reagent. But it didn't seem to work as evidenced by the fact that they isolated only starting materials after the reaction. (In fact we saw no evidence of Grignard formation.) We will wait and see how it goes with Hattie and Kanani tonight. We may want to try to get the Grignard forming by warming gently or by adding a small crystal of iodine.



3. 4/3/02 -- Hattie and Kanani were also unable to get the Grignard reagent to form after stirring Mg ribbon overnight as suggested in the paper on "Mechanical Activation of Magnesium Turnings...." mentioned in Note 1 above. (It didn't look like the stirring in the round bottom flask did a very good job of producing Mg powder.) We even tried adding a small crystal of iodine without success.



4. 4/3/02 -- Nikki and Dan found an article in Organic Syntheses (Collective Volume VI, pp 845-852) on preparing highly reactive Magnesium by reaction of MgCl₂ with potassium, but it looks like it would be fairly difficult.



5. 4/3/02 -- This week we will try preparing the Grignard reaction again, but we will try some of the following tricks for getting Grignard reagents to form:
• Dry glassware and magnesium in the oven overnight.
• Crush Mg in a mortar and pestle immediately before use.
• After the glassware has been assembled and charged with Mg, add a small crystal of iodine and then heat the flask as you would when flame drying. This will sublime the iodine, creating a purple haze that will coat the flask after it cools.


6. 4/4/02 -- Dan and Nikki were able to get the Grignard to form using the above tricks. After addition of a little bit of the methallyl chloride to the Mg the solution in the flask turned cloudy, and eventually a lot of precipitate was formed making the contents about the consistency of a slurpy. They filtered the mixture through a sintered glass funnel under nitrogen, and then added the methyl vinyl ketone to this solution. Their yield was not great, but it looked like the correct stuff by NMR. They will do some GC/MS to verify. We can now try to optimize this procedure.



7. 4/4/02 -- Kanani and Hattie tried conditions similar to Dan and Nikki's but didn't see much evidence of reaction. They tried sonicating and adding a little iodododecane to get the Grignard to react. After all the methallyl chloride was added (after about 2 hours) the reaction mixture was a bit cloudy. The mixture was refluxed for about an hour at which point a precipitate, similar to Dan and Nikki's, had formed. After rereading the paper by Baker et. al. (see above) it seems that the precipitate is probably MgCl₂ formed by homocoupling, thus this is an unwanted side reaction.



8. 4/7/02 -- Dan and Nikki did a GC/MS on their product. They started at 100 degrees C for 3 minutes and then ramped to 180 degrees at 10 deg/min. They had a peak (the largest non-solvent peak) at 3.4 minutes that seems to be the desired product: weak parent at m/z 126, and a base peak at 71. Also contains M-18 and M-15 peaks as well as significant peaks at 43 and 55. We might want to try a lower temperature (80 degrees?) just so that we can see starting materials.



9. 4/7/02 -- The NMR for 3,5-dimethyl-1,5-hexadien-3-ol was given in the Perez et. al article (see above): H-NMR (CDCl₃) 1.30 (s, 3H), 1.78 (3H, s), 1.86 (1H, s), 2.30 (2H, s), 4.77 (1H, d, J=1.3 Hz), 4.92 (1H, d, J=1.3 Hz), 5.04 (1H, dd, J=10.7, 1.3), 5.23 (1H, dd, J=17.3, 1.3 Hz), 5.96 (1H, dd, J=10.7, 17.3 Hz). FT-IR (neat): 3445, 3075, 2933, 1710, 1643, 1107, 920, 777 cm^-1.



10. 4/7/02 -- The article by Perez et al. (see above) also notes that 2-methyl-1-propenyl magnesium chloride (the Grignard reagent we are trying to prepare) was "prepared by Grignard reaction between 3-chloro-2-methyl-1-propene (6.78 g, 7.4 mL, 75 mmol) and activated magnesium powder (3.40 g, 0.14 mol) in THF at 0 degrees C according to Oppolzer et al." We should order this reference: Oppolzer, W.; Kundig, E.P.; Bishop, P.M.; Perret, C. Tetrahedron Lett. 1982, 3901.



11. 4/16/02 -- We looked at the Oppolzer reference referred to above, and the procedure they use is beyond our capabilities. (They heat the Mg in an aluminum crucible at 0.01 torr and evaporate into a rotating vessel containing THF at -110 degrees!)



12. 4/16/02 -- I got a different, unopened, container of Mg turnings. They are much shinier than the old bottle that we were using, so I am hoping that this will help in our efforts to get the Grignard reagent to form.



13. 4/16/02 -- Brian and Lindsey have used the new Mg and are trying the "dry-stir" method again. They flame dried their flasks and Mg under vacuum. In one flask they used a regular "octagonal" stir bar (3/4") and the Mg turnings. In the other flask we used a large (3/4") football shaped stir bar, and also added some Mg powder that we hope will help to grind up the Mg pieces. They are currently stirring under Nitrogen.



14. 4/16/02 -- A few other ideas to try: Use the 3-bromo-2-methylpropene instead of the chloro, at least to try to start the Grignard. Use an alkyl or aryl halide that more readily forms a Grignard to help start the reaction, then switch to the chloro after some fresh Mg surface has been exposed. I was thinking of phenyl bromide. I still think that sonication might be promising.



15. 4/16/02 -- As mentioned in a 4/7/02 note, Nikki and Dan seemed to get some product. They then did some TLC and Flash Chromatography work to purify it. Here are their notes:
    Here is the information regarding our procedure for TLC and purification. Our initial TLC, before flash chromatography, was performed with a 1:1 ether:pentane eluent. A glass-backed silica gel plate was used and stained with a ceric solution. There were two major spots: 1) Rf = 0.60 2) Rf = 0.170. After the crude TLC, purification was performed using flash chromatography. Three eluents were used. The first eluent was a 50:50 ether:pentane mixture (100 mL). The second eluent used was a 75:25 ether:pentane mixture (100 mL). The last eluent used was straight ether. 20 fractions were collected. The same plates and stain were used for TLC. The following Rf values were obtained for the fractions listed. 1) fractions 6-7: Rf = 0.736 2) fractions 8-12: Rf = 0.358; fractions 13-15: Rf = 0.366 3) fractions 16-18 = 0.293. A GC-MS was done for the three groups above prior to rotovapping, and the product appeared in the combined 6th and 7th fractions. The GC initial temp was 80 C and after three minutes, the temp raised 10 degrees/min, until it reached 140 C. Those combined fractions (6 & 7) were rotovapped for a concentrated product (0.04 g) and an H-NMR was obtained. We plan on rotovapping the other two combined fraction solutions as well so that we can obtain an NMR to see what else we have.


16. 4/16/02 -- The dienol that is produced in this reaction is relatively volatile (bp = 146-147 at atmospheric, and 46-47 at 10 mm) so be careful not to over rotovap.

Zn-Cu Couple or Organocuprate

The Zn-Cu couple mediated coupling of methallyl bromide and methyl vinyl ketone is reported to produce the desired 6-methyl-6-hepten-2-one in 80% yield. Alternatively, it might be possible to use the cuprate formed from the Grignard reagent produced from 3-bromo- or 3-chloro-2-methylpropene to do a conjugate addition into methyl vinyl ketone.

To the zinc-copper couple prepared as above, was added 3-buten-2-one (1.0 mmole, 0.070 g) and 1-bromo-2-methylpropene [This must be named incorrectly, the structure in the paper is that of 3-bromo-2-methylpropene.] (1.5 mmol, 0.203 g) and the mixture was sonicated for 45 min. It was quenched with saturated brine and worked-up as usual to furnish pure 6-methyl-6-hepten-2-one after chromatography over silica gel and eluting with pet. ether: ethyl acetate (9:1) (0.114 g, 80%). NMR and IR data provided.

1. 3/27/02 -- Brian and Lindsey tried the Zn-Cu couple procedure, but didn't isolate any of the desired product. The Zn definitely turned black after sonicating for a few minutes with the CuI, but it may be that the Zinc that we are using (20 mesh) is not fine enough, or maybe the reaction is slower than we expected. They will try it again, but follow the reaction by GC/MS. If that doesn't work, we'll order some new Zn powder.



2. 4/4/02 -- Still no luck. But The Wizard has some Zn powder, so they'll try that next time.



3. 4/15/02 -- Brian and Lindsey tried three times and never saw any product. They've decided to abandon this approach and will try working on the Grignard strategy.