A method for preparing sodium acrylate-d3, a useful and stable precursor for deuterated acrylic monomers

Article abstract of DOI:10.1002/jlcr.1915

A convenient and economical method for converting propiolic acid to sodium acrylate-d₃ is described. Successive D/H exchange of the alkyne proton of sodium propiolate (prepared from propiolic acid) using D₂O affords sodium propiolate-d having up to 99 at.% D. Sodium propiolate-d can be partially reduced to sodium acrylate-d₃ with 90% conversion and 89% yield using D₂ and the Lindlar catalyst, with control of reaction parameters to maximize conversion while minimizing over-reduction. Copyright 2011 John Wiley & Sons, Ltd. A three-step synthetic route for preparing sodium acrylate-d₃ in high yield and high isotopic purity from propiolic acid is described. Copyright

Full text of DOI:10.1002/jlcr.1915

Research Article
Received 24 May 2011,
Revised 26 June 2011,
Accepted 27 June 2011
Published online 11 August 2011 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.1915
A method for preparing sodium acrylate-d3,
a useful and stable precursor for deuterated
acrylic monomers
Jun Yang, Kunlun Hong, and Peter V. Bonnesen*
A convenient and economical method for converting propiolic acid to sodium acrylate-d
of the alkyne proton of sodium propiolate (prepared from propiolic acid) using D O affords sodium propiolate-d having up
to 99at.% D. Sodium propiolate-d can be partially reduced to sodium acrylate-d with 90% conversion and 89% yield using
3
is described. Successive D/H exchange
2
3
D
2
and the Lindlar catalyst, with control of reaction parameters to maximize conversion while minimizing over-reduction.
Keywords: deuterated acrylic; Lindlar catalyst; propiolate; sodium acrylate; sodium acrylate-d3
Introduction
where sodium acrylate was prepared directly by the partial reduc-
tion of sodium propiolate. In that work, the sodium propiolate
6
Strategic deuteration of macromolecules can be used in combi-
was reduced over Pt black in 0.10M NaOH solution affording a
mixture of sodium acrylate and sodium propionate, although
synthetic details were not provided. As we had some experience
1
2
nation with NMR, infrared analysis, and neutron-based probing
3
methods such as small-angle neutron scattering to obtain rich
structural and dynamic information that may be inaccessible
by other methods. This is particularly important for molecules
that do not crystallize or where understanding of the solution’s
structural and/or dynamic details is required. Acrylic-based ma-
cromolecules are ubiquitous in many materials used in daily life,
and acrylic polymers with novel properties for new applications
are highly sought. It is both important and challenging to eluci-
date the structure–function relationship of these macromolecules.
Techniques such as neutron scattering can provide much insight
into this relationship; however, the generally high cost of key mono-
mers such as acrylic acid-d
for researchers wishing to prepare strategically deuterated
acrylic polymers. Because D C=CD-CO H is an important syn-
thetic precursor to a wide variety of deuterated acrylic mono-
mers of interest, we were motivated to develop a synthetic
7,8
using the Lindlar catalyst (Pd supported on CaCO or BaSO ,
3
4
modified with Pb) to reduce terminal alkynes to alkenes, we first
attempted the partial reduction of sodium propiolate with
hydrogen and the Lindlar catalyst in various aqueous media,
including 0.05M NaOH, but found that essentially no reduction
took place, even after stirring overnight.
As sodium propiolate and sodium acrylate are both somewhat
soluble in methanol, we next attempted the partial reduction in
methanol and found that this did work but that reaction condi-
tions need to be carefully controlled to obtain good conversion
of sodium propiolate to sodium acrylate while minimizing
the production of sodium propionate, the product of ‘over-
reduction’ (Scheme 1). After tuning the reaction using hydrogen,
3 2 2
(D C=CD-CO H) has been an obstacle
2
2
we adapted the reaction to produce sodium acrylate-d
we report a convenient route for preparing sodium acrylate-d
3-D) in high yield and good conversion from propiolic acid (1).
3
. Herein,
3
route that was practical and inexpensive. Although there exists
(
2
a variety of examples for introducing the vinyl-d
group as a substituent using D
we were impressed by the fact that there was no simple way
to directly make D C=CD-CO H/D C=CD-CO D. In addition, even
3
( H
3
-acrylic)
4
,5
2
CO and the Wittig reaction,
2
2
2
2
when stored cold and inhibited, acrylic acid has shelf-life limitations.
Because sodium acrylate is a solid material that is consider-
ably more shelf stable than acrylic acid and because it does
Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak
Ridge, TN 37831-6494, USA
not require cold storage with a radical inhibitor, we were *Correspondence to: Peter V. Bonnesen, Center for Nanophase Materials Sciences,
drawn to developing a synthetic route to sodium acrylate- Oak Ridge National Laboratory, Oak Ridge, TN 37831-6494, USA.
E-mail: bonnesenpv@ornl.gov
d3, which could be conveniently stored, and from which we
could prepare deuterated acrylic acid and other specialty deuter- This manuscript has been authored by UT-Battelle, LLC, under contract no.
ated acrylic monomers on demand.
DE-AC05-00OR22725 with the US Department of Energy. The United States Gov-
ernment retains and the publisher, by accepting the article for publication,
acknowledges that the United States Government retains a non-exclusive,
paid-up, irrevocable, world-wide license to publish or reproduce the published
form of this manuscript, or allow others to do so, for United States Government
Accordingly, we examined methods to synthesize sodium
acrylate directly and wondered if this transformation could be
accomplished by the partial reduction of the alkyne analog,
sodium propiolate. A literature search revealed only one example purposes.
J. Label Compd. Radiopharm 2011, 54 743–748
Published 2011. This article is a US Government work
and is in the public domain in the USA.

J. Yang et al.
NaOH
CH OH
[H], Catalyst
HC
C
CO H
HC
C
CO Na
H C CH CO Na
+
CH CH2 CO Na
3 2
2
2
2
2
3
propiolic acid
1
sodium propiolate
2-H
sodium acrylate
3-H
sodium propionate
4-H
Scheme 1. Conversion of propiolic acid (1) to sodium propiolate (2-H) and reduction to sodium acrylate (3-H) and sodium propionate (4-H).
7,8
which is known to retard the over-reduction reaction. After 1h,
a distribution ratio of 21:75:4 of 2-H:3-H:4-H was observed.
In those first experiments, the reaction flask was placed under
partial vacuum (evacuated and nitrogen refilled three times,
Results and discussion
Survey reactions with hydrogen
Sodium propiolate (2-H) can be conveniently prepared in nearly then evacuated again until the solvent bubbled), and a balloon
quantitative yield by treating propiolic acid with sodium hydro- containing the approximately twofold excess of hydrogen gas
9
xide in methanol; however, sodium propiolate is somewhat was attached to the flask. To better control the reaction and
light sensitive, and the reaction and the product should be car- prevent over-reduction, in subsequent reactions, a gas tight
ried out protected from light. Prior to conducting reactions using syringe was used to deliver close to the stoichiometric quan-
deuterium, a series of survey reactions were performed using tity of hydrogen required. However, it became apparent that
hydrogen gas to determine the optimum reaction conditions an inert blanketing atmosphere was also needed so that the
that would maximize the yield of sodium acrylate (3-H) while reactant hydrogen could be consumed without partial vacuum
minimizing the amount of sodium propiolate (2-H) remaining conditions being created in the flask. Hence, all subsequent
and the amount of sodium propionate (4-H) produced. It was reactions were performed by first placing the reaction flask
previously reported that treatment of sodium propiolate in 0.1 under 1atm of nitrogen, attaching a nitrogen purged but com-
ꢀ
6
M NaOH at 40 C with hydrogen over Pt black could produce pletely deflated balloon to the flask, and then syringing into
sodium acrylate; however, synthetic details are absent, and no the flask a slight molar excess (4–9%) of hydrogen.
information regarding the yield of sodium acrylate is provided.
Quinoline is customarily added to Lindlar-catalyzed alkyne to
On the basis of this work, the partial reduction of sodium propio- alkene reduction reactions at anywhere from 0.05ꢁ (5%) to
8
late using the Lindlar catalyst (200mg per 1.15g of 2-H) was first 10ꢁ (1000%) on the weight of the catalyst to enhance selectiv-
investigated in 0.05M NaOH. However, after 24h of stirring with ity for the alkene by, it is believed, decreasing surface interac-
10
an approximately twofold excess of hydrogen, only unreacted tions of the alkene with the catalyst and hence retarding
-H was observed in the NMR spectrum (Table 1). Switching to reduction of the alkene. As the reaction proceeds and the start-
a 1:1 mixture of 0.05M NaOH:CH OH also resulted in only ing alkyne is reduced to the alkene, the alkyne concentration
2
3
unreacted 2-H after 24h, and no reaction was also observed decreases, and the alkene concentration increases, and competi-
when the solvent was deionized water (2h). However, when tion for hydrogen between the desired alkene product and the
pure CH OH was used as the solvent, in the presence of an starting alkyne increases. Maintaining reaction conditions to
3
approximately twofold excess of hydrogen, complete over- retard reaction between hydrogen and the alkene (here, the
reduction to sodium propionate (4-H) was observed within 1h. desired sodium acrylate) to give the alkane (sodium propionate)
In this case, the balloon was completely deflated, and because becomes increasingly important toward the end of the reaction,
of the excessive consumption of hydrogen, the flask was under when the [alkene]:[alkyne] ratio surpasses 9:1 (90% conversion).
negative pressure. This reaction was then repeated with the addi- However, high concentrations of quinoline also slow the reaction
tion of 1.0g of quinoline (500% on the weight of the catalyst), rate to the extent that conversion efficiency is impaired. As a
a
Table 1. Percent distribution of products as a function of solvent
excess H over
2
Lindlar catalyst
HC
C
CO Na
H C CH CO Na
+
CH CH2 CO Na
3 2
2
2
2
solvent, rt
2
-H
3-H
4-H
b
Solvent
NaOH (0.05M)
Other additive
Reaction time (h)
Distribution 2-H:3-H:4-H
None
None
None
None
24
24
2
1
1
Unreacted 2-H only
Unreacted 2-H only
Unreacted 2-H only
4-H only
1
:1 0.05M NaOH:CH OH
3
Deionized H O
2
CH OH
3
c
CH OH
Quinoline
21:75:4
3
a
All reactions used 1.15g sodium propiolate, 200 mg Acros Lindlar catalyst, and 40 mL solvent, and excess hydrogen with
no blanketing nitrogen atmosphere present.
b
c
13
The percent distribution is ꢂ1% by C NMR.
Quinoline was added at 500 wt.% on the catalyst.
www.jlcr.org
Published 2011. This article is a US Government work
and is in the public domain in the USA.
J. Label Compd. Radiopharm 2011, 54 743–748

J. Yang et al.
result, it was necessary to determine the optimum amount of optimized reaction conditions for converting sodium propiolate,
quinoline to add.
3
2-H, to sodium acrylate-d , 3-D, are shown in Scheme 2. In the
Accordingly, a series of reactions were conducted as that pre- first step, the weakly acidic methine proton of 2-H is exchanged
viously mentioned (using 4–9% excess hydrogen in nitrogen), in with a deuteron by stirring 2-H in D O (at approximately 1.5g
2
which the amount of quinoline was varied (0, 50, 250, or 500wt. D2O per g of 2-H) three times successively. For the first contact,
%
of the catalyst; see the Experimental section). A small amount sodium carbonate was present at 10mM to facilitate the H/D
(
10mg) of hydroquinone was added as an inhibitor, although exchange. This reaction has been performed previously using
11
its presence may be unnecessary. Aliquots of the reaction mix- NaOD, but sodium carbonate offers the advantage of being
ture were retrieved at 1, 2, and 3h for analysis by proton and car- an inexpensive non-deuterated base. The progress of the
bon NMR. After 5h, the reaction mixture was filtered, and the exchange can be monitored by proton and carbon NMR. After
solvent was removed to obtain the mixture of sodium carbox- equilibrium is achieved, the DHO/D O mixture can be recov-
2
ylates 2-H, 3-H, and 4-H. As can be seen in the results shown in ered by vacuum distillation or rotary evaporation, and fresh
Table 2, increasing the concentration of quinoline retards the D2O can be added to the solids that remain (which include
reaction rate, affording larger amounts of unreacted 2-H; how- sodium carbonate; hence, additional sodium carbonate does
ever, the amount of over-reduced product 4-H is very similar in not need to be added). The procedure is repeated for a total of
all cases. It appears that when the amount of hydrogen gas deliv- three contacts. Typically, the sodium propiolate-d was >88% D
ered is only slightly in excess, very little quinoline is required to following the first contact, >96% D following the second con-
retard over-reduction: within the margin of uncertainty, the tact, and ≥99% D following the third contact.
product distribution obtained using no quinoline and 50wt.%
quinoline (on the weight of the catalyst) are essentially the formed using the same reaction conditions as that previously
same, with only approximately 5% 4-H being formed. However, mentioned, except that CH OD was used in place of CH OH
Trial deuteration reactions of 2-D at the 1.16-g scale were per-
3
3
because a small amount of quinoline did not appear to be harm- and that deuterium was syringed into the reaction flask. The
ful to the reaction, subsequent reactions were performed using results of these trials indicate that the deuteration reaction pro-
quinoline at 50wt.%.
ceeded in a manner nearly identical to that encountered with
hydrogen, with percent conversion to 3-D being in the range
of 87–91%. It is unclear at this time whether or not there is
a deuterium isotope effect, as the differences in rates and
percent conversion between 2-H to 3-H and 2-D to 3-D appear
to be within the uncertainty of the product distribution
measurements.
To perform the second step of converting 2-D to 3-D using
deuterium at a more practical (e.g., 10ꢁ) scale than the 1.15-g
scale employed for the survey hydrogenation reactions, a 3-L
capacity SKC Flex Foil gas sample bag was used to contain and
deliver the deuterium gas to the reaction mixture. A practice
hydrogenation reaction of 2-H at a 11.5-g scale using hydrogen
in the Flex Foil bag afforded, after stirring for 24h, material
possessing a 2-H:3-H:4-H molar distribution of 9%:89%:2%
Comparison of Lindlar catalyst from different sources
As the same batch of Lindler catalyst was employed in all the sur-
vey reactions, to test whether the product distribution was
dependent on any particular batch of Lindlar catalyst, the reac-
tion was repeated (using conditions of 50wt.% quinoline) with
batches of Lindlar catalyst from different sources. As shown in
Table 3, the results obtained using Lindlar catalyst from three dif-
ferent sources were very similar, within the margin of uncer-
tainty, indicating that similar performance can be expected
with Lindlar catalysts from various sources.
Deuteration reactions
Having established a set of reaction conditions that could lead to (all valuesꢂ1%). This reaction was then repeated using 11.6g
≥
90% conversion of sodium propiolate to sodium acrylate, we (0.125mol) of 2-D dissolved in 400mL CH OD with 2.0g Lindlar
3
turned our attention to preparing sodium acrylate-d
3
(3-D). The catalyst, 1.0mL quinoline, and 0.1g hydroquinone, in a 1-L reaction
a
Table 2. Percent distribution of products as a function of quinoline concentration
H /N mixture over
2
2
Lindlar catalyst
HC
C
CO Na
H C CH CO Na
+
CH CH2 CO Na
3 2
2
2
2
CH OH/Quinoline, rt
3
2-H
3-H
4-H
No quinoline
50wt.% quinoline
250wt.% quinoline
500wt.% quinoline
b
Reaction time (h)
2-H:3-H:4-H
2-H:3-H:4-H
2-H:3-H:4-H
2-H:3-H:4-H
1
2
3
5
36:60:4
17:78:5
7:90:3
37:59:4
19:77:4
7:90:3
45:51:4
23:73:4
16:80:4
9:89:2
46:49:5
20:76:4
10:87:3
9:88:3
0:95:5
0:96:4
a
13
Wt.% quinoline on Acros Lindlar catalyst; 1atm nitrogen at start; all percent distributions are ꢂ1% by C NMR.
b
Aliquots taken at 1, 2, and 3h; product isolated at 5h (NMR on approximately 100mg isolated product mixture).
J. Label Compd. Radiopharm 2011, 54 743–748
Published 2011. This article is a US Government work
and is in the public domain in the USA.
www.jlcr.org

J. Yang et al.
a
Table 3. Percent distribution of products as a function of Lindlar catalyst source
Acros lot no. A017882001
Aldrich batch no. 05812HJ
Alfa Aesar lot no. K055049
b
Reaction time (h)
2-H:3-H:4-H
2-H:3-H:4-H
2-H:3-H:4-H
1
2
3
5
37:59:4
19:77:4
7:90:3
Not determined
20:76:4
Not determined
21:74:5
13:82:5
7:90:3
10:85:5
3:94:3
0:96:4
a
13
All reaction used 50wt.% quinoline on Lindlar catalyst; 1atm nitrogen at start; all percent distributions are ꢂ1% by C NMR.
b
Aliquots taken at 1, 2, 3h; product isolated at 5h (NMR on approximately 100mg isolated product mixture).
flask under nominally 1atm of nitrogen. The reaction was stirred at adding a small amount of the sodium salt of 3-(trimethylsilyl)-
room temperature for 24h while open to a Flex Foil bag filled to 3 propionic-2,2,3,3-d
L with deuterium gas. The reaction was then filtered to recover the nance for the trimethylsilyl group to zero. Deuterium ( H) NMR
catalyst, the CH OD recovered for reuse by evaporation, and the was acquired by disconnecting the lock channel from the probe,
4
acid (TSP, Aldrich) and by setting the reso-
2
3
2
residue was washed with ether. After drying, 12.01g of a gray solid removing the H reject filter from the X-band cable, replacing the
was recovered that consisted of a 2-D:3-D:4-D molar distribution broad band quarter wave cable with a 65–90MHz quarter wave
of 7%:90%:3% (all valuesꢂ1%) by carbon NMR and a molar distri- cable on the low band pre-amp, and tuning the probe to deuter-
2
bution of 6.3%:90.5%:3.2% (all valuesꢂ0.3%) by deuterium NMR, cor- ium (76.7MHz). For H NMR spectra acquired in D
2
O, a chemical
responding to an 89% yield of sodium acrylate-d
3
. We anticipate the shift of 4.80 for D
2
O was used as the internal reference.
reaction parameters could be further adjusted so that all of 2-D is
consumed, with very little 4-D being formed.
9
Sodium propiolate (2-H)
In a darkened hood, ground sodium hydroxide (8.565g, 0.214mol)
was dissolved in 1L methanol in a 2-L round-bottom flask. The
solution was cooled to 10 C, and propionic acid (1, 15.0g, 0.214
Experimental
ꢀ
General
mol) was added with stirring. The solution was stirred for 1h at
room temperature, following which the solvent was removed
by rotary evaporation with water bath temperature ≤25 C. A
All chemicals were reagent grade and used as received with-
out further purification unless noted otherwise. Deuterium
oxide (99.9at.% D) and deuterium (99.8% D, 99.6% D2+0.4%
HD) were obtained from Cambridge Isotope Laboratories, Inc.
The Lindlar catalysts were obtained as follows: palladium on
calcium carbonate, poisoned with 3.5% lead, 5% Pd (Acros part
no. 195070500, lot no. A017882001); palladium, 5wt.% on cal-
cium carbonate, poisoned with lead (Aldrich part no. 205737,
batch no. 05812HJ); and palladium, 5% on calcium carbonate,
ꢀ
white solid product, which was dried under high vacuum, was
obtained (19.50g, 99%). The compound is light sensitive and
1
13
should be stored in the dark. H NMR (CD OD): d 3.36; C NMR
3
(
CD OD): d 160.8 (C¼O), 81.8 (H-CꢃC), 69.3 (H-CꢃC).
3
General method for small-scale hydrogenation
survey reactions
lead poisoned (Alfa Aesar stock no. 43172, lot no. K05S049). A 250-mL round-bottom flask was charged with sodium propiolate
Flex Foil’ Grab Bags for containing hydrogen or deuterium (1.15g, 12.5mmol), the appropriate Lindlar catalyst (200mg, 5wt.%
were purchased from SKC, Inc. Proton and proton-decoupled Pd(0) on CaCO with 3wt.% of PbCO ), hydroquinone (10mg),
‘
3
3
carbon (without Nuclear Overhauser Effect [NOE] enhance- and a Teflon-coated stir bar. The appropriate solvent or solvent
ment) NMR spectra were obtained on a Varian VNMRS 500 mixture (40mL) was added, and the solution was stirred to dis-
NMR spectrometer, operating at 499.72MHz for proton. For solve the propiolate. The desired amount of quinoline (0, 50,
13
C NMR experiments used to determine product distribution 250, or 500wt.% of the catalyst, depending on the experiment)
ratios, the acquisition was performed with no NOE, a recycle was then added via syringe. The flask was sealed with a rubber
delay of 5s, and a minimum of 8000 transients for aliquots from septum, wrapped with Parafilm, then evacuated and back-filled
reaction mixtures (prepared by diluting approximately 200mL of with N three times by means of a needle attached to a double-
2
2
the reaction mixture with approximately 500mL of D O) or 300 manifold vacuum line. A plastic 5-mL syringe barrel with the
transients for sodium acrylate product isolated from the reaction large end cutoff was fitted with two He-quality latex balloons
13
(
typically 100mg solid dissolved in 700mL D
2
O.) For C NMR (one balloon was inserted inside a second balloon), using rubber
spectra acquired in D
2
O, chemical shifts were determined by bands and parafilm. The balloon assembly was purged with
a)
b,c)
HC
C
CO Na
DC
C
CO Na
D C CD CO Na
2 2
2
2
2
-H
2-D
3-D
Scheme 2. Synthesis of sodium acrylate-d
D/H exchange); (b) D /N gas mixture over 5% Pd–CaCO
ether, and dry (85–90% conversion, ≥95at.% D).
3
(3-D) from sodium propiolate (2-H). Reagents and conditions: (a) successive exchanges 10 mM Na in D
2
CO
3
2
O (99%, 97–99%
2
2
3
with 3% PbCO (Lindlar catalyst) in CH OD containing quinoline; and (c) filter, distill off CH OD, wash residue with
3
3
3
www.jlcr.org
Published 2011. This article is a US Government work
and is in the public domain in the USA.
J. Label Compd. Radiopharm 2011, 54 743–748

J. Yang et al.
nitrogen at least three times. For reactions involving NO nitrogen non-deuterated), 81.8 (s, carbon from residual H-CꢃC), 81.3 (t,
blanketing atmosphere, the reaction flask was evacuated one D-CꢃC, J2CD=7.3Hz), 73.5 (s, carbon from residual H-CꢃC), 73.2
2
more time. The balloon assembly was evacuated, filled with (t, D-CꢃC, JCD=39.0Hz); H NMR (D
2
O): d 3.08 (br s, DCꢃC-CO
2
Na).
approximately 600–800mL of hydrogen gas (>twofold excess),
Sodium acrylate-d
3
(3-D)
and the syringe barrel fitted with a needle that was then inserted
into the septum of the reaction flask. The solution was stirred at
room temperature for up to 24h, with a small amount (approxi-
mately 0.2mL) of solution removed periodically by syringe.
Sodium propiolate-d (2-D, 11.6g, 0.125mol), Lindlar catalyst
2.0g, 5wt.% Pd(0) on CaCO with 3wt.% of PbCO , Acros Organ-
ics lot no. A017882001), hydroquinone (0.1g), quinoline (1.0mL,
(
3 3
Those aliquots were diluted with 0.4mL of D O for analysis by
2
8.5mmol, approximately 0.5ꢁ weight of the catalyst), and
1
13
H and C NMR. For reactions involving nominally 1atm nitrogen
CH OD (400mL, 324g, 99.0% D) were combined in a 1-L two-
3
blanketing atmosphere, the syringe barrel of the evacuated/
deflated balloon assembly was fitted with a needle that was
inserted through the septum of the nitrogen-filled reaction flask.
An amount of hydrogen corresponding to a total of 13.3ꢂ0.3
mmol (the uncertainty being due to the day-to-day variance in
lab temperature and pressure) was syringed into the reaction flask
through the septum, using a 100-mL gas-tight syringe equipped
with a valve, upon which the balloon was observed to inflate.
The solution was stirred at room temperature for 5h. A small
amount (approximately 0.2mL) of solution was removed by syr-
neck round-bottom flask with a Teflon-coated stir bar. One neck
was fitted with a rubber septum; the other neck of the flask was
fitted with a Pyrex inlet adapter with a 90 hose connection with
ꢀ
a short section of Tygon tubing connected to a glass T-connec-
tor. One end of the T was connected to a double-manifold
vacuum line; the remaining end of the T was connected using
Tygon tubing to the valve stem of a 3-L capacity SKC Flex Foil
gas sample bag, containing nominally 3L (0.134mol) of deuter-
ium gas. The valve on the bag was closed, and the Tygon tubing
connecting the bag to the T adapter pinched shut with a tubing
clamp. The tubing connecting the flask to the vacuum line was
not clamped at this time. Using the vacuum line, the flask was
1
13
2
inge and diluted with 0.4mL of D O for H and C NMR at reac-
tion times of 1, 2, and 3h. After 5h of stirring (the balloon was
completely deflated, indicating that the hydrogen had been con-
sumed), the solution was filtered to remove the catalyst and was
2
evacuated and filled with N gas 3 times. The tubing between
the T and the vacuum line was then clamped shut, and deuter-
ium gas was admitted to the reaction flask by opening the clamp
and the valve on the Flex Foil bag. With the valve open, the
solution was stirred at room temperature overnight. At reaction
times of 3, 6, and 8h, 0.2mL aliquots were removed via the
septum-covered neck using a syringe, for NMR analysis (see
preceding discussion). After 24h, the foil bag was completely
flat. The catalyst was filtered off, and the solid was washed
washed (to remove quinoline) with 5mLꢁ3 of CH
3
OH.
The combined washing and reaction solvent was removed by
rotary evaporation, and the solid that remained was suspended
in 50mL of ether and was stirred for 10min. The solid was then
filtered, was washed with additional ether (10mLꢁ3), and was
dried under vacuum to afford generally 1.10 to 1.15g of a gray
1
3
1
solid. C{ H} NMR (no NOE; see previous discussion) was used
to determine the final distribution of sodium propiolate (2-H),
sodium acrylate (3-H), and sodium propionate (4-H).
with CH
3
OD (10mLꢁ3). The combined solution was placed on
a rotary evaporator, and the methanol-D evaporated and recov-
ered. The solid that remained was washed with ether (50mLꢁ3)
and was dried under vacuum, to afford 12.01g of a gray solid.
Sodium propiolate-d (2-D)
1
3
1
C{ H} NMR (no NOE) revealed a molar product distribution
based on the ratio of the peak areas of the carbonyl carbons
In a typical procedure, sodium propiolate (2-H, 13.81g, 0.150mol)
was dissolved under nitrogen in a solution of 10mM sodium car-
bonate in D O (21.04g, 1.05mol D O) in a 200-mL round-bottom
(
from unreacted sodium propiolate-d (2-D, d 162.9), sodium
acrylate-d (3-D, d 178.3), and sodium propionate-d (4-D, over
2
2
3
5
flask with a Teflon-coated stirbar. The solution was stirred at
room temperature for 24h, and at which time, a 650-mL sam-
ple was withdrawn for analysis by proton and carbon NMR.
reduced, d 188.5), as 7:90:3 (all values areꢂ1), resulting in a
2
sodium acrylate-d yield of 89%). H NMR (D O) revealed a very
3
2
13
1
similar ratio at 6.3:90.5:3.2. C{ H} NMR (D O, TSP): d 178.3
2
13
1
C{ H} NMR with no NOE revealed an 88.7:11.3 ratio of the
(
s, C=O), 136.0 (t, D C=CD-, J =24.4Hz), 128.8 (p, D C=CD-,
2 CD 2
D-CC triplet (J2CD=7.3Hz) to the H-CC singlet from the non-
deuterated starting material. The NMR sample was returned
2
JCD=24.4Hz). H NMR (D O): d 6.12 (br s, D DtransC=CDgem-), 6.01
2
cis
(
br s, D DtransC=CDgem-), 5.65 (br s, DcisDtransC=CDgem-), 3.10 (br
cis
to the reaction flask, and the D O was removed and recovered
2
s, DCC-CO Na), 2.10 (br s, CD CD -CO Na), 0.98 (br s, CD CD -
2 3 2 2 3 2
by vacuum distillation. To the solid residue which remained was
13
CO Na). Comparison of the C integrals of the residual signals
2
added 21.04 g (1.05mol) of pure D O. (Additional sodium carbo-
2
due to non-deuterated sodium acrylate with deuterated sodium
nate was not needed, because the initial amount was still
present as part of the solid in the flask). Stirring was resumed
under nitrogen at room temperature for 24h. Analysis of a
acrylate indicated that the sodium acrylate-d was >95at.% D.
3
The recovered CH OD was distilled from 3Å molecular sieves to
3
afford 276g (85% recovery) of clean material suitable for reuse.
1
3
1
6
50-mL sample by C{ H} NMR with no NOE revealed a 96.8:3.2
ratio of deuterated to non-deuterated. The NMR sample was
again returned to the flask, whereupon the existing D O/DHO
was distilled off and replaced with 21.04g of fresh D
Summary
2
2
O. Analysis A convenient and economical method for converting the relatively
by C{ H} NMR with no NOE revealed a 99.0:1.0 ratio of deuter- inexpensive propiolic acid to sodium acrylate-d , D C=CD-CO Na,
ated to non-deuterated. The NMR sample was returned to the a shelf-stable precursor to a wide variety of acrylic-d monomers,
flask, and the D O again recovered by vacuum distillation. A has been developed. The method described results in very good
few milliliters of CH OD was added and then evaporated to isolated yields (up to 89%, at >95% D), containing only small
further dry the salt, which was then dried under high vacuum. amounts of sodium propiolate-d and sodium propionate-d
1
3
1
3
2
2
3
2
3
5
.
13
The yield of the white powder obtained was 13.84g (99%).
H} NMR (D O, TSP): d 163.0 (s, C=O, both deuterated and be conveniently recovered for potential reuse, and the relatively
C
The synthesis uses readily available Lindlar catalyst, which can
1
{
2
J. Label Compd. Radiopharm 2011, 54 743–748
Published 2011. This article is a US Government work
and is in the public domain in the USA.
www.jlcr.org

J. Yang et al.
inexpensive deuterated starting materials D
2
, D
2
O, and CH
3
OD. References
The latter two can also be effectively recovered for potential
reuse. The procedure is amenable to scale up, and with further
optimization, the yield and purity can likely be improved.
Cross-labeling experiments (e.g., reaction of H2 with sodium
[
[
[
1] L. J. Mathias, R. F. Colletti, Macromolecules 1988, 21, 858–860.
2] D. Bucca, B. Gordon, III, Macromolecules 1994, 27, 862–864.
3] D. Hourdet, F. L’alloret, A. Durand, F. Lafuma, R. Audebert, J.-P.
Cotton, Macromolecules 1998, 31, 5323–5335.
[
4] P. Keller, D. L. Thomsen, III, M.-H. Li, Macromolecules 2002, 35,
propiolate-d in CH OD) have indicated that the mechanism of
3
5
81–584.
5] J. Pitlik, C. A. Townsend, J. Labelled Compd. Radiopharm. 1997, 39,
99–1009.
proton and deuteron transfer is complex, with substantial isotopic
scrambling being observed beyond what would be expected from
simple cis and trans addition of H or D across the triple bond.
[
9
2
2
[6] L. D. Volkova, V. Z. Gabdrakipov, O. V. Vyaznikovtseva, D. V. Sokolskii,
Further work to elucidate the mechanism of the partial reduction of
propiolates to acrylates over the Lindlar type catalysts is in progress.
React. Kinet. Catal. Lett. 1985, 29, 345–351.
7] H. Lindlar, R. DuBuis, U.S. Patent 3,715,404, Feb. 6, 1973, and refer-
ences cited.
8] H. Lindlar, R. DuBuis, Organic Syntheses, Wiley, New York, 1973, Col-
lect. Vol. V pp. 880–882.
9] J. D. Jaufmann, C. B. Case, R. B. Sandor, B. M. Foxman, J. Solid State.
Chem. 2000, 152, 99–104.
[
[
[
Acknowledgement
This research was conducted at the Center for Nanophase
Materials Sciences, which is sponsored at the Oak Ridge National
[
10] J. G. Ulan, E. Kuo, W. F. Maier, R. S. Rai, G. Thomas, J. Org. Chem. 1987,
2, 3126–3132.
5
Laboratory by the Office of Basic Energy Sciences, US Department [11] J. E. Katon, N. T. McDevitt, Spectrochim. Acta, 1965, 21,
of Energy.
1717–1724.
www.jlcr.org
Published 2011. This article is a US Government work
and is in the public domain in the USA.
J. Label Compd. Radiopharm 2011, 54 743–748