Flow Asymmetric Propargylation: Development of Continuous Processes for the Preparation of a Chiral β-Amino Alcohol

Article abstract of DOI:10.1002/anie.201704882

The development of a flow chemistry process for asymmetric propargylation using allene gas as a reagent is reported. The connected continuous process of allene dissolution, lithiation, Li-Zn transmetallation, and asymmetric propargylation provides homopropargyl β-amino alcohol 1 with high regio- and diastereoselectivity in high yield. This flow process enables practical use of an unstable allenyllithium intermediate. The process uses the commercially available and recyclable (1S,2R)-N-pyrrolidinyl norephedrine as a ligand to promote the highly diastereoselective (32:1) propargylation. Judicious selection of mixers based on the chemistry requirement and real-time monitoring of the process using process analytical technology (PAT) enabled stable and scalable flow chemistry runs.

Full text of DOI:10.1002/anie.201704882

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
International Edition Chemie
www.angewandte.org
Accepted Article
Title: Flow Asymmetric Propargylation: Development of Continuous
Processes for a Chiral beta-Amino Alcohol
Authors: Yuan-Qing Fang, Hui Li, Jillian W Sheeran, Andrew M
Clausen, Matthew M Bio, and Scott Bader
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201704882
Angew. Chem. 10.1002/ange.201704882
Link to VoR: http://dx.doi.org/10.1002/anie.201704882
http://dx.doi.org/10.1002/ange.201704882

Angewandte Chemie International Edition
10.1002/anie.201704882
COMMUNICATION
Flow%Asymmetric%Propargylation:%Development%of%Continuous%
Processes%for%a%Chiral%
β
;Amino%Alcohol%
[
a]
[a]
[a]
[a]
[a]
Hui$Li, $Jillian$W.$Sheeran, $Andrew$M.$Clausen, $$Yuan:Qing$Fang,* $$Matthew$M.$Bio* ,$and$Scott$
[
b]
Bader $
Abstract:$ The$ development$ of$ a$ flow$ chemistry$ process$ for$ access$ to$ chemistries$ involving$ highly$ reactive$ or$ short:lived$
asymmetric$propargylation$using$allene$gas$as$a$reagent$is$reported.$ intermediates,$ and$ chemistries$ difficult$ or$ impossible$ to$ scale$
8
The$connected$continuous$process$of$allene$dissolution,$lithiation,$Li:
using$standard$batch$vessels. $$In$addition,$dosing$of$gases$can$
Zn$ transmetallation$ and$ asymmetric$ propargylation$ provides$ be$accurately$achieved$in$flow$by$mass$flow$controllers$(MFC).$
homopropargyl$ β:amino$ alcohol$ 1$ with$ high$ regio:$ and$ Herein$we$report$on$the$development$of$a$practical$continuous$
diastereoselectivity$in$high$yield.$This$flow$process$enables$practical$ flow$ manufacturing$ process$ for$ the$ synthesis$ of$ chiral$
use$of$an$unstable$allenyllithium$intermediate.$The$process$uses$the$ homopropargyl$β:amino$alcohol$1% (Scheme$1)$based$on$allene$
commercially$ available$ and$ recyclable$ (1S,2R):N:pyrrolidinyl:
lithiation$under$non:cryogenic$conditions.$$$
norephedrine$ as$ a$ ligand$ to$ promote$ the$ highly$ diastereoselective$
(
32ꢀ1)$ propargylation.$ Judicious$ selection$ of$ mixers$ based$ on$ the$
chemistry$requirement$and$real:time$monitoring$of$the$process$using$
process$ analytical$ technology$ (PAT)$ enabled$ stable$ and$ scalable$
flow$chemistry$runs.$
Homopropargyl$ β:amino$ alcohols$ are$ useful$ building$ blocks$
for$ the$ construction$ of$ pharmaceuticals$ and$ bioactive$ natural$
products$ like$ (−):rhizopodin,$ pumiliotoxins$ and$ antimicrobial$
$
Scheme%1.$Synthetic$route$to$homopropargyl$alcohol$1$
Although$pumping$allene$THF$solution$could$be$achieved$by$
1
agent$AI:77:B. $Recently,$we$required$a$practical$manufacturing$
pressurizing$the$feeding$bottle,$we$found$allene$could$be$rapidly$
dissolved$as$gas/liquid$slugs$in$THF$in$a$short$FEP$tubing$(≤$0.2$
mL,$1/16”$OD)$at$room$temperature$by$controlling$the$flow$rates$
of$ THF$ and$ MFC:metered$ allene$ gas.$ The$ setup$ was$ further$
equipped$ with$ an$ online$ ATR:IR$ flow:cell$ (Mettler:Toledo$
FlowIR),$which$quantitatively$tracked$the$allene$absorption$peak$
$
2
process$to$access$chiral$homopropargyl$β:amino$alcohol$1. $$We$
envisioned$ a$ straightforward$ approach$ based$ on$ catalyst$ or$
ligand:controlled$asymmetric$propargylation$of$L:alanine:derived$
aldehyde$ 2.$ $ In$ the$ last$ few$ decades,$ tremendous$ progresses$
have$ been$ made$ in$ the$ area$ of$ catalytic$ asymmetric$
propargylation$ of$ aldehydes$ and$ ketones.$ Despite$ these$
advances,$ established$ methods$ require$ low$ reaction$
temperatures$ to$ avoid$ substrate$ epimerization$ when$ applied$ to$
:
1
:1
at$842$cm $(or$1390$cm ).$We$were$able$to$accurately$prepare$
allene$solution$at$variable$concentrations$up$to$2.5$M$with$<$3$
bar$ backpressure,$ much$ higher$ than$ the$ saturation$ point$ of$
allene$in$THF$(~1$M)$under$ambient$condition$(see$Supporting$
Information).$$$$
3,4
carbonyl$substrates$with$α:stereogenic$centers.$ $$$
The$most$commonly$used$propargylation$reagents$are$allenyl$
boronic$ acids$ prepared$ via$ Hg:catalyzed$ magnesiation$ of$
$
5
propargyl$ bromide. $The$ use$ of$ mercury,$ the$ corrosiveness$ of$
propargyl$ bromide,$ and$ the$ pyrophoric$ nature$ of$ the$ allenyl$
boronic$acid$raise$significant$environmental$and$safety$concerns$
when$ using$ these$ reagents$ for$ large:scale$ applications.$ The$
direct$ lithiation$ of$ allene,$ a$ readily$ available$ and$ inexpensive$
starting$material,$is$a$potential$alternative$to$propargyl$bromide$
magnesiation$ if$ a$ practical$ lithiation$ /$ transmetallation$ process$
could$be$developed.$$Although$allenyllithium$has$been$prepared$
6
at$−78$°C, $it$is$generally$considered$as$unstable,$which$limits$its$
application$in$organic$synthesis.$$In$addition,$handling$allene$gas$
in$batch$also$poses$many$practical$challenges.$$$$$$
$
Continuous:flow$ synthesis$ has$ evolved$ into$ important$
technology$ in$ fine$ chemical$ and$ pharmaceutical$ research,$
Figure%1.$Allene$lithiation$in$flow$
Initially,$ the$ flow:prepared$ allene/THF$ solution$ was$ mixed$
together$in$a$simple$tee$with$a$stream$of$n:HexLi$(2.1$M),$a$safer$
non:pyrophoric$substitute$for$n:BuLi,$to$achieve$lithiation$below$
7
development$ and$ manufacturing. $Advantages$ such$ as$ rapid$
mixing$ and$ efficient$ heat$ transfer$ of$ flow$ technologies$ allow$
−
20$ °C.$ The$ formation$ of$ allenyllithium$ was$ quantified$ using$
FlowIR$ by$ tracking$ appearance$ of$ its$ characteristic$ absorption$
[
$
a]$
Dr.$H.$Li,$J.$Sheeran,$Dr.$A.$Clausen,$Dr.$Y.$Fang,$Dr.$M.$Bio$
Snapdragon$Chemistry$Inc.$$
:
1
peak$at$1864$cm .$The$IR$signature$also$allowed$us$to$confirm$
the$previously$reported$instability$of$allenyllithium$species,$which$
we$ observed$ to$ have$ a$ decay$ half:life$ of$ about$ 5$ minutes$ at$
$$$$$$$$$$85$Bolton$St.,$Cambridge,$MA$02140$(United$States)$
E:mailꢀ$eric.fang@snapdragonchemistry.com$
$$$$$$$$$$$$$$$$$$$$$$matt.bio@snapdragonchemistry.com$
[b]$ Dr.$S.$Bader$$$
$
$
$
$
$
2
2$°C$(see$Supporting$Information).$$$
Although$ we$ achieved$ flow$ lithiation$ of$ allene$ under$ non:
$$$$$$$$$$Chemical$R&D$
$$$$$$$$$$Pfizer$Worldwide$R&D$
$$$$$$$$$$Eastern$Point$Road,$MS$8118D:4047,$Groton,$CT$06340$
cryogenic$ conditions,$ when$ operating$ the$ lithiation$ with$ ≥$ 0.5$
equiv$ of$ HexLi$ at$ 0$ °C,$ we$ observed$ increase$ of$ system$
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201704882
COMMUNICATION
pressure$ over$ time$ and$ eventual$ clogging$ of$ the$ reactor.$ We$
intentionally$treated$allene$with$a$large$excess$amount$of$HexLi$
in$ batch$ and$ observed$ the$ same$ precipitation$ and$ a$
:
1
characteristic$IR$stretch$at$1900$cm .$$We$also$found$that$the$
precipitation$ could$ be$ dissolved$ by$ adding$ excess$ allene,$ with$
:
1
concurrent$ disappearance$ of$ 1900$ cm $ absorption$ band$ (see$
Supporting$Information).$$These$data$indicate$that$over:lithiation$
is$a$reversible$process$(Scheme$2).$$Under$the$slow$laminar$flow$
mixing$ conditions,$ the$ poor$ mixing$ leads$ to$ regimes$ where$ the$
allene:versus:HexLi$ ratio$ is$ very$ low,$ thus$ resulting$ in$ over$
lithiation$and$eventual$clogging$(Figure$2,$left).$$$
$
Scheme%3.$Transmetallation$of$allenyllithium$prepared$in$flow$
Scheme%2.$Reversible$allene$overlithiation$based$on$FlowIR$data$
We$next$turned$our$attention$to$the$asymmetric$propargylation$
of$ Boc:L:alaninal$ 2.$ $ Our$ initial$ screening$ studies$ found$ that$
previously$ reported$ enantioselective$ methods$ based$ on$
organoboron$ reagents$ gave$ sluggish$ reactivity$ and$ low$
11
diastereomeric$ratio$(dr$typically$<$3ꢀ1). $We$found$allenylindium$
along$ with$ chiral$ aminoalcohol$ ligand$ could$ afford$ the$ desired$
12
product$ with$ a$ dr$ of$ up$ to$ 10ꢀ1. $However,$ it$ requires$
stoichiometric$ amount$ of$ expensive$ indium$ metal$ under$
cryogenic$ condition.$ $ During$ the$ course$ of$ these$ studies,$ we$
found$that$the$chiral$center$in$the$substrate$poses$little$bias$for$
the$formation$of$new$chiral$center$(vide.infra).$$$
$
Among$ the$ organometallic$ reagents,$ we$ were$ attracted$ to$
inexpensive$ and$ readily$ available$ organozinc$ reagents,$
combined$ with$ chiral$ ligands$ (eꢀg.$ BINOLs,$ β:amino$ alcohols),$
which$ has$ proven$ to$ be$ a$ practical$ method$ for$ asymmetric$
Figure% 2.$ leftꢀ$ Slow$ laminar$ flow$ mixing$ led$ to$ clogging$ by$ over:lithiated$
speciesꢁ$rightꢀ$microjet$n:HexLi$flow$coupled$with$dynamic$mixer$led$to$clog:
free$lithiation$
$
13
synthesis.
$Nonetheless,$ little$ is$ known$ for$ asymmetric$
Achieving$rapid$mixing$with$low$flow$rate$using$static$mixers$is$ propargylation$ using$ allenylzinc$ reagents.$ We$ hoped$ that$
a$ significant$ challenge,$ particularly$ when$ faced$ with$ clogging$ transmetallation$ of$ allenyllithium$ to$ zinc$ with$ appropriate$ chiral$
potential.$$To$overcome$this$challenge,$we$designed$a$laboratory$ ligand$would$provide$a$ligand:controlled$asymmetric$addition$to$
dynamic$mixer$with$multiple$1/8”$x$½”$magnetic$stir$bars$housed$ 2$ and$ thus$ deliver$ the$ desired$ diastereomer$ 1.$ One$ of$
9
in$a$1/4“$PTFE$tubing$spaced$out$with$PTFE$balls. $The$HexLi$ challenges$ was$ control$ of$ the$ regioselectivity$ favoring$ desired$
1
4
stream$ was$ directly$ introduced$ into$ the$ high:speed$ spinning:
motioned$stir$bars$with$a$microjet$(see$Supporting$Information).$
homopropargyl$alcohol$1$over$α:allenic$carbinols$4. $$
When$ we$ performed$ propargylation$ screening$ experiments$
2
With$this$device,$we$were$able$to$address$mixing$and$clogging$ upon$allenyllithium$transmetallation,$Et Zn$was$used$as$the$zinc$
15
issue,$enabling$us$to$prepare$allenyl$lithium$reagent$at$≥$0.7$M$ source. $Little$substrate$control$was$observed$in$the$absence$of$
concentration$ (HexLi$ 0.65−0.80$ equiv)$ at$ 0$ °C$ under$ steady$ a$chiral$ligand$(Table$1,$entry$1).$Initial$chiral$auxiliary$screening$
10
state$ for$ multiple$ hours. $$ This$ provided$ us$ a$ versatile$ results$ using$ cinchona$ alkaloid$ were$ promising$ (L1,$ Table$ 1,$
transmetallation$ starting$ material$ for$ preparation$ of$ entry$ 2)ꢁ$ however,$ we$ found$ the$ results$ were$ not$ reproducible$
allenylmagnesium,$ boron,$ and$ zinc$ species$ (Scheme$ 3).$ Using$ when$ we$ scaled$ up$ due$ to$ the$ formation$ of$ a$ more$
this$ technology,$ potassium$ alllenyltrifluoroboronate$ was$ thermodynamically$stable$precipitate$of$allenylzinc$complex$with$
prepared$ on$ multi:gram$ scale,$ avoiding$ use$ of$ Hg$ and$ cinchona$alkaloid$L1.$$We$further$screened$other$synthetic$chiral$
circumvented$the$use$of$pyrophoric$boronic$acid.$$
$
amino$ alcohol$ scaffolds,$ such$ as$ diphenylethylene:based$ L3,$
proline:derived$ L4,$ and$ the$ norephedrine:derived$ L5.$ L5$ was$
found$to$give$reproducibly$high$dr$of$the$desired$product$1%under$
homogeneous$ conditions% even$ under$ various$ stressed$
conditions$(Table$1,$entry$6).$
Although$ (1S,2R):N:pyrrolidinylnorephedrine (L5)$ gave$
promising$diastereoselectivity,$we$noticed$that$the$reaction$was$
plagued$ by$ formation$ of$ substantial$ amounts$ of$ the$ α:allenic$
carbinol$ regioisomer$ 4$ (1ꢀ4$ =$ 1.7ꢀ1).$ $ To$ our$ delight,$ the$
formation$of$wrong$regioisomer$4$could$be$largely$suppressed$to$
<
7
$ 5%$ by$ using$ achiral$ additive$ trifluoroethanol$ (entry$ 1,$ Table$
16
), $$ which$ also$ improved$ diastereoselectivity$ to$ 23ꢀ1$ and$
increased$reaction$rate.$$
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201704882
COMMUNICATION
Table%1.$Chiral$ligand$screening$for$allenylzinc$propargylation$
equiv.),$Et
2
Zn$(1.8$equiv.).$[b]$1ꢀ4$>$20ꢀ1$in$all$cases.$[c]$L5$(1.8$equiv.),$achiral$
additive$ (1.8$ equiv.).$ [d]$ Determined$ by$ GC$ analysis$ of$ unpurified$ mixture$on$
β−Dex$chiral$column.$$
When$ we$ studied$ the$ propargylation$ reaction$ kinetics,$ we$
found$ that$ product$ dr$ increased$ as$ a$ function$ of$ reaction$
17
conversion$ (Figure$ 3). $Consistent$ with$ this$ notion,$ when$ we$
performed$the$asymmetric$propargylation$reaction$in$continuous$
flow$reactor,$we$found$product$dr$was$sensitive$to$the$mode$of$
mixing$ between$ the$ aldehyde$ 2$ and$ the$ allenylzinc$ chiral$
complex.$ $ When$ the$ two$ streams$ were$ mixed$ together$ either$
with$ a$ Tee$ or$ low:volume$ dynamic$ mixer,$ the$ reaction$
diastereoselectivity$ was$ significantly$ lower$ than$ the$ continuous$
18
stirred$tank$reactor$(CSTR)$mixer$(Table$3). $Since$product$dr$is$
a$weight$average$of$the$entire$reaction,$the$latter$capitalized$the$
high$dr$regime$of$the$reaction.$
[
a]
Entry
$
Chiral$
ligand$
Achiral$
d.r.$(1%ꢀ$
Conversion$
1$ꢀ$4$
[
e]
[e]
additive$
3)
$
(1+3)$(%)
$
[b
]
1
2
3
4
5
6
$
$
$
$
$
$
−
−$
−$
−$
−$
−$
−$
1.1ꢀ1$
82$
−$
[
d]
24
L1%
L2%
L3%
L4%
L5%
L5%
5.7ꢀ1
$
79$
9.7ꢀ1$
13ꢀ1$
2
2
1
1
1
1
2
0
8
6
4
2
2.6ꢀ1$
2.3ꢀ1$
3.4ꢀ1$
84$
87$
7.7ꢀ1$
10ꢀ1$
85$
14.5ꢀ1$
23ꢀ1$
56$
1.7$ꢀ$1$
>$20$ꢀ$1$
[
c]
7
$
CF
3
CH
2
OH$
90$
[
a]$ General$ conditionꢀ$2$ (1$ equiv.),$ chiral$ ligand$ (2.88$ equiv.),$ Et
equiv.).$ [b]$ Ligand/additive$ not$ used$ or$ data$ not$ determined.$ [c]$ L5$ (1.8$
equiv.),$ CF CH OH$ (1.8$ equiv.).$ [d]$ data$ on$ 0.4$ mmol$ scale,$ dr$ 1ꢀ1$ was$
2
Zn$ (1.8$
5
0%
60%
70%
80%
90%
100%
Conversion
3
2
$
obtained$on$54$mmol$scale.$$[e]Determined$by$GC$analysis$of$unpurified$
Figure% 3.$ Diastereoselectivity$ as$ a$ function$ of$ reaction$ conversion$ in$
asymmetric$ propargylation$ using$ trifluoroethanol$ as$ an$ additive$ at$ −20$ °C$
mixture$on$β−Dex$chiral$column.$
(
Table$1,$entry$7$conditions)$
Given$the$benefit$of$the$additive,$further$screening$of$a$variety$
of$alcohols$and$phenol$at$room$temperature$found$the$reaction$
could$ be$ influenced$ by$ subtle$ change$ of$ the$ alcohol$ structures$
Table%3.$Impact$of$mixing$mode$on$reaction$dr$
(
Table$ 2,$ entry$ 1:5).$ In$ all$ cases,$ no$ apparent$ formation$ of$ α:
[
a]
Entry
$
Mixing$mode$
d.r.$(1%ꢀ$3)$
Conversion$
1+3)$(%)$
allenic$carbinol$regioisomer$4$was$observed.$We$also$found$that$
(
the$ratio$of$chiral$to$achiral$ligand$could$be$reduced$to$1ꢀ1$while$
[
b]
maintaining$high$efficiency$of$diastereochemical$control$(Table$2,$ 1$
entries$5$vs$6).$Ultimately,$we$identify$that$1:propanol$gave$the$
Tee$mixer $%
11ꢀ1$
16ꢀ1$
26ꢀ1$
84$
84$
90$
2
$
Dynamic$mixer$(τ=0.6$min)%
CSTR$(τ=5$min)%
best$results$of$32ꢀ1$dr$and$the$reaction$completed$in$<$20$min$at$
room$temperature$(Table$2,$entry$7).$
3
$
Table% 2.$ Achiral$ additive$ studies$ for$ propargylation$ using$ allenylzinc;L5$
complex$
[
a]$General$conditionꢀ$2$(1$equiv.),$L5$(2.88$equiv.),$trifluoroethanol$(0.72$
equiv.),$Et Zn$(1.8$equiv.).$[b]$0.05”$inner$diameter$of$thru:hole$
2
[
a][b]
[d]
Entry
$
Achiral$additive$
d.r.$(1%ꢀ$3)
$
Conversion$
With$ the$ optimal$ combination$ of$ ligands$ identified,$ the$
asymmetric$ allenylzinc$ propargylation$ was$ optimized$ under$
continuous:flow$ conditions$ and$ was$ connected$ to$ the$ allene$
lithiation$ in$ flow.$ As$ shown$ in$ Figure$ 4,$ the$ connected$ flow$
process$ started$ with$ dissolution$ of$ allene$ gas$ in$ THF$ under$
pressure$ (20$ psi),$ which$ was$ then$ connected$ to$ n:HexLi$ in$ a$
dynamic$ mixer$ to$ produce$ allenyllithium$ reagent.$ The$ average$
[
d]
(
1+3)$(%) $
1
2
3
4
5
$
CF
3
3
CH
CHOH$
PhOH$
2
OH$
19ꢀ1$
10ꢀ1$
16ꢀ1$
23ꢀ1$
27ꢀ1$
29ꢀ1$
32ꢀ1$
84$
66$
90$
80$
83$
82$
84$
$
(CF
)
2
$
$
$
4:NO
2
∆
T$ (T:Tbath)$ for$ allenyl$ lithiation$ at$ a$ combined$ flow$ rate$ of$ 2.3$
PhCH
2
OH$
mL/min$(2.5$mmol/min)$is$about$9$°C.$
CH
CH
2
2
=CHCH
2
2
OH$
OH$
[
c]
6
7
$
$
=CHCH
[
c]
3 2 2
CH CH CH OH$
[
a]$ General$ conditionꢀ$ 2$ (1$ equiv.),$ L5$ (2.88$ equiv.),$ achiral$ additive$ (0.72$
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201704882
COMMUNICATION
Keywords:$continuous:flow$synthesis•$allene$lithiation$•$
asymmetric$propargylation$•$flow$IR$•$Process$Analytical$
Technology$
1
$
a)$S.$M.$Dalby,$J.$Goodwin:Tindall,$I.$Paterson,$Angewꢀ.Chemꢀ.Intꢀ.Ed.$2013,$
5
2,$6517:6521.$b)$N.$Lin,$L.$E.$Overman,$M.$H.$Rabinowitz,$L.$A.$Robinson,$M.$
L.$ Sharp,$ J.$ Zablocki,$ Jꢀ. Amꢀ. Chemꢀ. Soc.$ 1996,$ 118,$ 9062:9072.$ c)$ A.$ K.$
Ghosh,$A.$Bischoff,$J.$Cappiello,$Eurꢀ.Jꢀ.Orgꢀ.Chem.$2003,$821:832.$d)$M.$Y.$
Pettersson,$D.$S.$Johnson,$C.$Subramanyam,$C.$J.$O’Donnell,$C.$W.$Ende,$M.$
E.$Green,$N.$C.$Patel,$C.$M.$Stiff,$T.$P.$Tran,$G.$W.$Kauffman,$A.$F.$Stepan,$P.$
R.$Verhoest,$US$2014/0088111$A1,$March$27,$2014.$
2
$
Compound$1$has$been$previously$synthesized$in$~34%$overall$yield$over$6$
stepsꢀ$Am$Ende,$C.$W.ꢁ$Green,$M.$E.ꢁ$Johnson,$D.$S.ꢁ$Kauffman,$G.$W.ꢁ$
O'Donnell,$C.$J.ꢁ$Patel,$N.$C.ꢁ$Pettersson,$M.$Y.ꢁ$Stepan,$A.$F.ꢁ$Stiff,$C.$M.ꢁ$
Figure%4.$Continuous:flow$synthesis$of$1
Subramanyam,$C.ꢁ$Tran,$T.$P.ꢁ$Verhoest,$Patrick$R.$$WO$2014045156,$2014
3
$
For$selected$review,$seeꢀ$a)$C.$Ding,$X.$Hou,$Chemꢀ.Rev.$2011,$111,$1914:
The$ chiral$ zinc$ complex$ was$ prepared$ from$ L5,$ 1:propanol$
1
937.$For$recent$example,$seeꢀ$b)$M.$Chen,$W.$R.$Roush,$Jꢀ.Amꢀ.Chemꢀ.Soc.$
and$Et
2
Zn.$The$resulting$zinc$complex$is$then$connected$to$the$
2012,$134,$10947:10952.$c)$T.$Liang,$S.$K.$Woo,$M.$J.$Krische,$Angewꢀ.Chemꢀ.
allenyllithium$ stream$ in$ a$ dynamic$ mixer$ to$ give$ the$ chiral$
allenylzinc$complex.$The$formation$of$chiral$allenylzinc$complex$
was$monitored$by$FlowIR$with$a$characteristic$IR$absorption$at$
Intꢀ.Ed.$2016,$55,$9207:9211.$$
4
$
For$selected$examples,$seeꢀ$a)$D.$R.$Fandrick,$K.$R.$Fandrick,$J.$T.$Reeves,$
Z.$Tan,$W.$Tang,$A.$G.$Capacci,$S.$Rodriguez,$J.$Song,$H.$Lee,$N.$K.$Yee,$C.$
H.$Senanayake,$Jꢀ.Amꢀ.Chemꢀ.Soc.$2010,$132,$7600:7601.$b)$D.$R.$Fandrick,$
J.$T.$Reeves,$J.$M.$Bakonyi,$P.$R.$Nyalapatla,$Z.$Tan,$O.$Niemeier,$D.$Akalay,$
K.$Fandrick,$W.$Wohlleben,$S.$Ollenbeger,$J.$Song,$X.$Sun,$B.$Qu,$N.$Haddad,$
S.$Sanyal,$S.$Shen,$S.$Ma,$D.$Byrne,$A.$Chitroda,$V.$Fuchs,$B.$A.$Narayanan,$
N.$Grinberg,$H.$Lee,$N.$Yee,$M.$Brenner,$C.$Senanayake,$Jꢀ.Orgꢀ.Chem.$2013,$
3592:3615.$c)$M.$Inoue,$M.$Nakada,$Orgꢀ.Lett.% 2004,$6,$2977:2980.$d)$L.$C.$
Hirayama,$ T.$ D.$ Haddad,$ A.$ G.$ Oliver,$ B.$ Singaram,$ Jꢀ. Orgꢀ. Chem.$ 2012,$
:
1
1
905$ cm $ (Figure$ 4).$ The$ allenylzinc$ stream$ and$ a$ solution$ of$
enantiopure$ Boc:L:alaninal$ 2% were$ then$ combined$ in$ a$ 25:mL$
CSTR$ at$ room$ temperature.$ An$ internal$ volume$ of$
$
approximately$ 16$ mL$ was$ maintained$ in$ the$ CSTR$ (2.6$ min$
residence$time).$The$CSTR$outflows$to$a$50:mL$tubular$reactor$
(
1/4"$OD$PTFE$tubing)$also$at$ambient$temperature$(nominal$8$
4
342:4353.$
minutes$residence$time),$providing$a$total$residence$time$of$10.6$
5
19
$H.$Hopf,$I.$Bohm,$J.$Kleinschroth,$Orgꢀ.Synth.$1981,60,$41.$
$
$
minutes$ for$ the$ asymmetric$ propargylation$ step. $The$ flow$
system$steadily$provided$homopropargyl$β:amino$alcohol$1$with$
a$production$rate$of$15$g/h.$In$addition$to$the$high$throughput,$
the$asymmetric$propargylation$provided$1$with$a$dr$of$32ꢀ1$and$a$
yield$of$85%.$$
6
7
Jaffe,$F.$Jꢀ.Organometꢀ.Chemꢀ.1970,$23,$53.$
a)$L.$Malet:Sanz,$F.$Susanne,$Jꢀ.Medꢀ.Chem.$2012,$55,$4062:4098.$b)$D.$R.$
Snead,$ T.$ F.$ Jamison. Angewꢀ. Chemꢀ. Intꢀ. Ed.$ 2012,$ 54,$ 983:987ꢁ$ and$
reference$therein.$c)$D.$E.$Fitzpatrick,$C.$Battilocchio,$S.$V.$Ley$ACSꢀ.Centꢀ.Sci.$
2016,$2,$131:138.$d)$R.$Porta,$M.$Benaglia,$A.$Puglisi,$Orgꢀ.Process.Resꢀ.Dev.%
2016,$ 20,$ 2:25.$ e)$ D.$ Cambie,$ C.$ Bottecchia,$ N.$ J.$ Straathof,$ Volker,$ H.ꢁ$ T.$
Noel,$Chemꢀ.Rev.$2016,$116,$10276:10341.$For$example$of$flow$asymmetric$
synthesis$using$organolithiums,$seeꢀ$f)$Y.$Tomida,$A.$Nagaki,$J.$Yoshida,$ Jꢀ.
It$ is$ noteworthy$ that$ 1.37$ equiv.$ of$ chiral$ zinc$ complex$ was$
used$ in$ the$ flow$ reaction$ while$ 1.8$ equiv$ was$ necessary$ to$
maintain$ high$ diastereoselectivity$ under$ batch$ conditions.$
Moreover,. L5$ can$ be$ recovered$ with$ acid:base$ treatment$ and$
has$been$re:used$as$a$chiral$ligand$for$up$to$eight$cycles$with$no$
change$in$performance.$$
Amꢀ.Chemꢀ.Socꢀ$2011,$133,$3744:3747$
8
$
For$ selected$ review,$ seeꢀ$ a)$ J.$ Yoshida,$ Y.$ Takahashi,$ A.$ Nagaki,$ Chemꢀ.
Commun.$2013,$49,$9896:9904.$For$recent$examples,$seeꢀ$b)$A.$Nagaki,$K.$
Imai,$S.$Ishiuchi,$J.$Yoshida,$Angewꢀ.Chemꢀ.Intꢀ.Ed.$2015,$54,$1914:1918.$c)$A.$
Hafner,$M.$Meisenbach,$J.$Sedelmeier,$Orgꢀ.Lettꢀ$2016,$18,$3630:3633.$d)$A.$
Nagaki,$ K.$ Imai,$ S.$ Ishiuchi,$ J.$ Yoshida,$ 2016,$ Angewꢀ. Chemꢀ. Intꢀ. Edꢀ. 55,$
5327:5331.$ e)$ H.$ Usutani,$ T.$ Nihei,$ C.$ D.$ Papageorgiou,$ D.$ G.$ Cork,$ Orgꢀ.
In$ conclusion,$ we$ have$ demonstrated$ continuous$ flow$ as$ an$
enabling$ technology$ for$ the$ practical$ preparation$ of$ highly$
reactive$ allenyllithium$ from$ allene$ gas$ under$ non:cryogenic$
conditions.$ Further,$ the$ allenyllithium$ was$ connected$ to$
transmetallation$ and$ carbonyl$ addition$ stages$ to$ deliver$ a$
connected$continuous$reactor$train$producing$the$amino$alcohol$
Process.Resꢀ.Devꢀ,$2017,$21,$669–673,$and$references$therein.$$
9
$
This$is$a$significantly$modified$design$based$on$previously$published$work$by$
Merck$scientisttsꢀ$$S.$J.$Dolman,$J.$L.$Nyrop,$J.$T.$Kuethe,$Jꢀ.Orgꢀ.Chemꢀ$2011,$
76,$993–996$
10
$
Since$the$dynamic$mixer$is$housed$in$poorly$thermal$conducting$PTFE,$heat$
1
$ with$ high$ yield,$ diastereo:$ and$ regioselectivity.$ Application$ of$
transfer$capability$is$low.$We$found$at$a$highest$3.5$mL/min$(3.75$mmol/min),$
close$to$adiabatic$temperature$rise$(ATR)$of$21$°C$was$reached.$$In$such$case,$
a$bath$temperature$of$<−20$°C$is$recommended$for$allene$lithiation$in$order$to$
on:line$ process$ analytical$ technologies$ such$ as$ FlowIR$ and$
temperature$ profiling$ allowed$ us$ to$ monitor$ in$ real:time$ the$
formation$ of$ reactive$ intermediates$ and$ track$ the$ overall$
performance$of$the$reactor$train.$
k11eep$mixer$at$<0$°C.$
2$
Ikeda,$N.ꢁ$Arai,$I.ꢁ$Yamamoto,$H.$Jꢀ.Amꢀ.Chemꢀ.Soc.$1986,$108,$483$
$T.$D.$Haddad,$L.$C.$Hirayama,$J.$J.$Buckley,$B.$Singaram,$Jꢀ.Orgꢀ.Chemꢀ$
012,$77,$889−898$
1
2
13
$
For$selected$review,$seeꢀ$a)$L.$Pu,$Accꢀ.Chemꢀ.Res.$2014,$47,$1523:1535.$
For$ selected$ example,$ seeꢀ$ b)$ L.$ Tan,$ C.$ Chen,$ R.$ D.$ Tillyer,$ E.$ J.$ J.$
Grabowski,$P.$J.$Reider,$Angewꢀ.Chemꢀ.Intꢀ.Ed.$1999,$38,$711:713.$c)$S.$Cui,$
S.$D.$Walker,$J.$C.$S.$Woo,$C.$J.$Borths,$H.$Mukherjee,$M.$J.$Chen,$M.$M.$Faul,$
Jꢀ.Amꢀ.Chemꢀ.Soc.$2010,$132,$436:437.$
Acknowledgements%%
We$thank$Prof.$Timothy$F.$Jamison$(Massachusetts$Institute$of$
Technology)$ and$ Aaron$ Beeler$ (Boston$ University)$ for$ helpful$
discussions.$$We$thanks$Dr.$Jade$Nelson$(Pfizer)$and$Dr.$Adam$
Brown$ (Pfizer)$ for$ valuable$discussions$ on$ the$ manuscript.$ We$
also$thank$Zaiput$Flow$Technologies$for$the$use$of$liquid$/$liquid$
separators$and$back$pressure$regulators.$$
14
$(a)$E.$J.$Corey,$C.:M.$Yu,$D.:H.$Lee,$Jꢀ.Amꢀ.Chemꢀ.Socꢀ$1990,$112,$879:
81ꢁ$ (b)$ C.$ A.$ Osborne,$ $ T.$ B.$ D.$ Endean,$ E.$ R.$ Jarvo$ Orgꢀ. Lettꢀ$ 2015,$ 17,$
340−5343$
8
5
15
$
Other$ zinc$ sources$ such$ as$ ZnBr
2
$ was$ also$ evaluated$ but$ gave$ inferior$
Zn$as$zinc$sources.$
Griffiths,$G.$J.ꢁ$Warm,$A.$Orgꢀ.Process.Resꢀ.Devꢀ,$2016,$20,$803–813$
diastereoselectivities$compared$to$the$ones$using$Et
$
2
16
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201704882
COMMUNICATION
17
$
This$indicates$that$propargyl$alcohol$1$formed$in$the$reaction$may$have$a$
positive$impact$on$the$reaction$diastereoselectivity,$perhaps$even$functioning$
as$a$ligand$on$the$allenylzinc$species.$To$test$this$hypothesis$we$used$purified$
1
$(with$d.r.$19.2ꢀ1)$as$an$additive$and$pre:mixed$it$with$chiral$Zn$complex.$The$
added$ aminoalcohol$ 1$ showed$ no$ improvement$ in$ reaction$ selectivity$ as$
compared$to$the$normal$procedure.$$
18
$
The$CSTR$experiment$result$are$more$close$to$batch$results$since$in$batch$
experiments,$starting$material$aldehyde$2$was$dropwise$added$over$>$1$min,$
and$ upon$ completion$ of$ aldehyde$ addition,$ high$ conversions$ are$ typically$
obtained$at$rt.$
19
$
Due$ to$ excess$ allene$ off:gas$ upon$ warmed$ to$ room$ temperature,$ actual$
residence$time$is$shorter$than$8$min$in$the$tubular$reactor.
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201704882
COMMUNICATION
COMMUNICATION$
Hui$Li,$Jillian$Sheeran,$Andrew$M.$
Clausen,$Yuan7Qing$Fang,*$Matthew$
M.$Bio*$and$Scott$Bader$
Page$No.$–$Page$No.$
Flow.Asymmetric.Propargylation:.
Development.of.Continuous.Processes.
β
for.a.Chiral. QAmino.Alcohol
$
This article is protected by copyright. All rights reserved.