4
$\begingroup$

The question might seem kind of trivial, although it is even harder than I've thought. How can we synthesise branched alkanes from linear one? For example, I want to obtain 3,4,7,8-tetramethyldecane from n-decane. I've already found Stille coupling, but I guess I'm searching for easier method. I'm just a high school student, studying chemistry by myself as well.

$\endgroup$
  • 1
    $\begingroup$ You have to functionalise first, introduce a substituent that you can manipulate e.g. halide $\endgroup$ – Waylander Jul 10 '18 at 15:11
  • $\begingroup$ Yeah, that's what I know, but what methylating agents should I use? $\endgroup$ – Radi Jul 10 '18 at 15:14
  • $\begingroup$ The easiest way is to make the Grignard from the halide then react with MeI $\endgroup$ – Waylander Jul 10 '18 at 15:16
  • 1
    $\begingroup$ There is an odd photochemical route to many hard-to-get branched alkanes described by Crabtree. It is unclear how much it is used (chemistry.stackexchange.com/q/84777/81). $\endgroup$ – matt_black Jul 10 '18 at 16:25
  • 2
    $\begingroup$ If this is a homework problem, it is an exercise in futility. No knowledgable chemist would consider n-decane as a starting material to obtain your tetradecane, stereochemistry notwithstanding. A better question is to prepare the tetradecane from commercially available compounds --possibly with some restrictions-- with no regard for stereochemistry. $\endgroup$ – user55119 Jul 10 '18 at 16:46
4
$\begingroup$

@Radi: Critical changes to original post! I will guide you through a reasonable approach. You have to realize that you have not specified the stereochemistry of 11 -- of which there are 16 possible stereoisomers (8 racemates) -- but I will ignore the issue. You will recognize the symmetry in 3,4,7,8-tetramethyldecane (11), so we will look for a coupling reaction of two 7-carbon entities. (E)-2-Buten-1-ol (crotyl alcohol) (4) and triethylorthopropionate 5 are commercially available. The reaction of 4 and 5 is a Johnson-type Claisen rearrangement (Google it!) which produces racemic esters 7 via transition state 6. Reduction of the double bond of unsaturated esters 7 leads to saturated esters 8. Reduction of the ester group to an alcohol followed by conversion to a bromide gives structure 9. The Wurtz reaction (9 --> 11) is rarely used but here is an opportunity.

Alternatively, esters 8 under the same conditions of the Wurtz reaction produces the α-hydroxyketones 10 (acyloin condensation). While the Clemmensen reduction (10 --> 11) of aliphatic ketones does not work as well as with aromatic ketones, under these conditions you would lose both oxygens. If the ketone is produced (structure 10 less the hydroxyl group), a Wolff-Kishner reduction would lead to the tetradecanes (plural) 11.

Read about the these name reactions which will be of more value to you than your original proposal. Good luck! enter image description here

$\endgroup$
  • $\begingroup$ Thank you for guiding me through really interesing synthesis, but the question was if there is a general method (simpler that Stille's reaction) that allows us to obtain branched alkanes from linear. Synthesis of 3,4,7,8-tetramethyldecane from n-decane was just an example since they both have 10-carbon chain. $\endgroup$ – Radi Jul 11 '18 at 12:39
  • $\begingroup$ No, there are no general methods. Waylander suggested a coupling -- like your thought of the Stille -- between an organometallic and halide. The organometallic would be better as the methyl unit. To halogenate (Cl or Br) n-decane at the 3,4,7,8 positions, the yield calculated is 0.02% of all tetrahaloisomers. And this is optimistic amid a myriad of other polyhalogenated compounds. Absolutely impractical. $\endgroup$ – user55119 Jul 11 '18 at 20:07
0
$\begingroup$

This problem has two parts: One is industrial chemistry. You typically halogenate alkanes, which gives you a wild mixture, and then fractionate and purify the products. You want to make a lot of different chemicals, so you can use most of the products, and you try to fine-tune the process so the composition matches your customers' demands, more or less.

Now you sell these n- and sec-haloalkanes (and other relatively simple, functionalised compounds) to a chemical laboratory. They now make some fancy metal-organic coupling reaction(s) like Stille (after perhaps making alcohols or ketones out of your haloalkanes) with some specific pair of compounds, possibly a following reduction step, and get a defined, branched alkane. A straightforward example is the https://en.wikipedia.org/wiki/McMurry_reaction .

There are a lot more possibilities for both parts, but the guy making the end product will always buy pre-products that each already contain sizeable parts of the structure he aims for. To plan such a sequence of reaction for a certain target is called retrosynthesis.

OR

If you don't want some specific compound, but just more branching, https://en.wikipedia.org/wiki/Catalytic_reforming would be your horse. You also get some aromats then, or cracking, depending on hydrogen pressure. This is also an industrial process, hard to and sort of pointless to do in the lab unless you are developing catalysts. I've visited a lab doing that, and it looked like a shrunk-down refinery with a roof and three outside walls, not at all like a chemistry lab.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.